📘

Definitions

1

CDC 1989 surveillance case definition

The definition used to count the epidemic: debilitating generalised myalgia with an absolute eosinophil count at or above 1.0 x 10^9 cells per litre, and no infection or neoplasm accounting for the findings. It was built for outbreak surveillance rather than for diagnosis, and every frequency in this entry is conditioned on it.

CASE_DEFINITION

Surveillance criteria

Minimum required: 2

Core clinical characteristics

debilitating generalised myalgia Severe enough to limit usual activity.
absolute eosinophil count at or above 1.0 x 10^9 cells per litre The numeric threshold the definition sets.

Exclusion criteria

infection accounting for the findings Any infection that could explain the myalgia and the eosinophilia. Deliberately unbound: the exclusion is a class of alternative diagnosis rather than a single concept, and no cited sentence names a term for it.
neoplasm accounting for the findings Any neoplasm that could explain the findings, including the clonal eosinophilias. Unbound for the same reason.

Show evidence (1 reference)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"A case is defined by debilitating myalgias and absolute eosinophilia greater than or equal to 1.0 x 10(9) cells/L."

The definition in the surveillance report's own words. `minimum_required` is 2 because both features are required, not one of two.

Notes: `minimum_required: 2` records that the definition is conjunctive. The exclusion clause is carried as free text because the quoted sentence states the two positive requirements only; the exclusion comes from the definition as described elsewhere in the same literature and is not separately cited here. The `case_definition_validity` discussion is where the consequences of this definition for every frequency in the entry are argued.

◈

Mechanistic Hypotheses

3

A contaminant drives a type 2 immune response that recruits and activates eosinophils

contaminant_driven_type2_immunity CANONICAL

Evidence balance 2 support

On this account the impurity acts as an immunogen or adjuvant. Susceptible hosts mount a type 2 cytokine response, eosinophils are expanded, recruited and degranulated in tissue, and the fibrosis follows the inflammation. The strongest support is the cytokine work on peak E and the eosinophil granule proteins recoverable from patients.

Pathograph links

Systemic Exposure to L-Tryptophan Process Contaminants -> Type 2 Cytokine Response to Peak E INDIRECT UNKNOWN INTERMEDIATES How peak E reaches a cytokine-producing cell is not described by any cited source. What is measured is the cytokine output of mononuclear cells exposed to it, in subjects who did not have EMS.

Systemic Exposure to L-Tryptophan Process Contaminants -> Eosinophil Chemotaxis and Endothelial Adhesion DIRECT

Type 2 Cytokine Response to Peak E -> Eosinophil Expansion with Tissue Degranulation DIRECT

Eosinophil Chemotaxis and Endothelial Adhesion -> Eosinophil Expansion with Tissue Degranulation DIRECT

Eosinophil Expansion with Tissue Degranulation -> Inflammatory Infiltration of Fascia, Perimysium and Dermis INDIRECT UNKNOWN INTERMEDIATES Recorded as unknown intermediates because the tissue the eosinophil products are deposited in contains almost no eosinophils. Whatever connects the two is not described.

Eosinophil Expansion with Tissue Degranulation -> Pulmonary Eosinophilic and Lymphocytic Inflammation DIRECT Unlike the fascia, the lung does show the eosinophil: lavage fluid from affected patients carries them in excess. This is the one compartment where the cell and the lesion are found in the same place.

Show evidence (2 references)

PMID:1727618 SUPPORT REVIEW SYNTHESIS Human Clinical

"Eosinophil activation and the release of major basic protein and other eosinophil-derived toxic proteins into the extracellular space is a striking feature in the eosinophilia-myalgia syndrome and implicates eosinophils or their products in the pathogenesis."

States the eosinophil-centred reading of the pathogenesis that this hypothesis group asserts.

PMID:11703359 SUPPORT INDIRECT PRIMARY RESULT In Vitro

"six of the seven FSS patients reacting with peak E produced IL-5 and/or IL-10"

INDIRECT: the type 2 cytokine response to peak E is measured, and it is measured in mononuclear cells from patients with functional somatic syndromes rather than from anyone with EMS.

The contaminant acts directly on fibroblasts

direct_fibroblast_activation ALTERNATIVE

Evidence balance 2 support

On this account the fibrosis does not require the immune response to reach it. Peak E stimulates dermal fibroblast DNA synthesis, procollagen transcription and gel contraction on its own, and L-tryptophan itself and the other major contaminant do not. The two hypotheses are not exclusive and the pathograph carries both edges.

Pathograph links

Systemic Exposure to L-Tryptophan Process Contaminants -> Fibroblast Activation and Type I Collagen Overproduction DIRECT The edge that makes the alternative hypothesis a separate route rather than a restatement: peak E raises procollagen transcription in dermal fibroblasts with no immune cell in the dish.

Show evidence (2 references)

PMID:7593596 SUPPORT PRIMARY RESULT In Vitro

"We now report that peak E, a dimer of L-tryptophan, is a potent stimulus for human dermal fibroblast DNA and collagen synthesis."

The direct fibroblast effect this hypothesis is named for.

PMID:7593596 SUPPORT PRIMARY RESULT In Vitro

"No increase in procollagen mRNA levels was found after the addition of another major L-tryptophan contaminant, peak 5, or with L-tryptophan itself."

The specificity control, and the reason the effect is attributed to peak E rather than to L-tryptophan or to contamination in general.

A T cell response against extracellular matrix sustains the chronic disease

t_cell_matrix_autoimmunity ALTERNATIVE

Evidence balance 1 support

On this account the chronic phase is not residual damage but an ongoing cell-mediated process. Fascia and perimysium contain CD8+ cells and macrophages with almost no intact eosinophils, most fibroblasts are activated and a third aberrantly express HLA-DR, and a repeat biopsy a year later showed less inflammation and more activated fibroblasts.

Pathograph links

Systemic Exposure to L-Tryptophan Process Contaminants -> T Cell Mediated Immunity Against Extracellular Matrix INDIRECT UNKNOWN INTERMEDIATES The contaminant reaches the inflammatory cells as well as the eosinophils. The review that states this names no intermediate step, so the link type records none.

T Cell Mediated Immunity Against Extracellular Matrix -> Inflammatory Infiltration of Fascia, Perimysium and Dermis DIRECT

T Cell Mediated Immunity Against Extracellular Matrix -> Pulmonary Eosinophilic and Lymphocytic Inflammation DIRECT The same CD8+ compartment found in fascia and perimysium is recoverable from the lung.

T Cell Mediated Immunity Against Extracellular Matrix -> Fibroblast Activation and Type I Collagen Overproduction DIRECT

Show evidence (1 reference)

PMID:8100551 SUPPORT PRIMARY RESULT Human Clinical

"We conclude that in EMS there is a T-cell-mediated process against components of the extracellular matrix, including fibroblasts, in the fascia and the perimysium that persists even years after the drug is discontinued."

The authors' own statement of this hypothesis, from their own immunocytochemistry.

?

Discussions and Knowledge Gaps

6

Which constituent of the implicated L-tryptophan caused eosinophilia-myalgia syndrome, and by what molecular route does it start the disease?

KNOWLEDGE GAP unidentified_etiologic_agent

Attached to

pathophysiology#Systemic Exposure to L-Tryptophan Process Contaminants pathophysiology#Type 2 Cytokine Response to Peak E

The exposure is not in doubt and the agent is. Six compounds were associated with case lots. EBT has the most work behind it and is the only one with animal and fibroblast data, and its epidemiological association does not survive stratification by date of manufacture. Nothing traces a molecular route from any candidate compound to the first lesion, which is why the edge out of the exposure node carries unknown intermediates.

Show evidence (2 references)

PMID:8895185 SUPPORT REVIEW SYNTHESIS Other

"The precise cause of EMS remains unknown."

States the gap, in the closing sentence of a review of every animal study in the disease. Graded OTHER rather than MODEL_ORGANISM because the sentence is a statement about the human disease's etiology, not an animal result; the same reference's two other snippets in this entry are animal findings and stay MODEL_ORGANISM.

PMID:37453474 SUPPORT REVIEW SYNTHESIS Other

"Many in vitro and in vivo studies have been conducted to assess the putative correlation between impurities in L-tryptophan preparations and EMS, but no clear and convincing conclusions have been drawn so far."

The same gap thirty years later, after the experimental work the earlier review was asking for.

Why does EBT produce the fibrosis of this disease in mice and rats without producing the eosinophilia that defines it, and what host factor is missing?

HUMAN MODEL MISMATCH animal_models_lack_eosinophilia

Attached to

animal_models#Mouse animal_models#Rat pathophysiology#Eosinophil Expansion with Tissue Degranulation

This is a mismatch and not an absence of evidence. Both animal models reproduce the fascial and perimysial lesion, and neither produces the blood eosinophilia, the rash or the weakness. So the compound can make the fibrosis without making the disease, and the eosinophil arm of the human pathograph has no animal counterpart. A critical review of the whole animal literature went further and doubted the models could be reproduced at all, which is why every fidelity in this entry is MODERATE or lower.

Show evidence (3 references)

PMID:8895185 SUPPORT REVIEW SYNTHESIS Model Organism

"However, a critical review of all the animal studies casts doubt on the validity of these assertions."

The reviewers' verdict on the claim that these are models of the syndrome, which is the mismatch this discussion records.

PMID:8895185 SUPPORT REVIEW SYNTHESIS Model Organism

"These studies could generally not be reproduced and had methodologic flaws that limited extrapolation of results."

The stated basis of that verdict.

PMID:7991132 SUPPORT PRIMARY RESULT Model Organism

"No rash or weakness occurred in either group."

The concrete instance of the mismatch: the tissue lesion without the illness, in the study whose own conclusion is that it replicates an important feature of the human disease.

What produces the incapacitating myalgia of this disease, given that muscle shows minimal myofiber atrophy, regeneration or necrosis?

KNOWLEDGE GAP myalgia_without_myofiber_injury

Attached to

phenotypes#Myalgia pathophysiology#Inflammatory Infiltration of Fascia, Perimysium and Dermis

Myalgia is one of the two features that define the syndrome, and the muscle pathology does not account for it. Eleven biopsied cases showed minimal myofiber damage despite severe myalgia in almost all of them, and the inflammatory infiltrate sits in perimysium, epimysium and fascia rather than in the fibres. No cited source asserts a mechanism, so `Myalgia` has no incoming causal edge in this entry. Toxic oil syndrome records the same gap for the same reason.

Show evidence (2 references)

PMID:1563745 SUPPORT PRIMARY RESULT Human Clinical

"Minimal myofiber atrophy, regeneration, or necrosis was seen despite the clinical history of severe myalgias in almost all patients."

The authors state the dissociation themselves, in the same sentence as the clinical severity.

PMID:8895179 SUPPORT REVIEW SYNTHESIS Human Clinical

"The pathophysiology of the chronic symptoms is poorly understood but may involve ischemia, neuropathy, and metabolic abnormalities."

The three candidate routes named at review level, none of which any cited source ties to the myalgia.

Is the abnormal tryptophan metabolism reported in EMS patients a step in the disease, or a consequence of the inflammation?

KNOWLEDGE GAP tryptophan_metabolism_abnormality

Attached to

pathophysiology#Systemic Exposure to L-Tryptophan Process Contaminants pathophysiology#Type 2 Cytokine Response to Peak E

Patients show abnormal handling of tryptophan, which a contemporary review reads as increased activity of indoleamine 2,3-dioxygenase, the rate-limiting enzyme of the kynurenine pathway. IDO is induced by inflammatory cytokines, so the finding sits on both sides of the question and no cited source settles which. Nothing is wired into the pathograph for it, and that is the point of recording the gap: an entry about a tryptophan-derived contaminant has an obvious temptation to draw an arrow here.

Show evidence (1 reference)

PMID:8423409 SUPPORT REVIEW SYNTHESIS Human Clinical

"Evidence of abnormal L-tryptophan metabolism has been described in patients with EMS, and most likely reflects increased activity of indoleamine 2,3-dioxygenase, the rate-limiting enzyme of tryptophan metabolism."

The finding and the proposed enzyme, at review level. "Most likely reflects" is the review's own hedge and is why this is a gap rather than a node.

Does the 1989 surveillance case definition identify the disease, or does it select a severe subset and make its two required features unmeasurable?

CONTROVERSY case_definition_validity

Attached to

phenotypes#Myalgia phenotypes#Peripheral eosinophilia

Every frequency in this literature is conditioned on a definition that was built for outbreak surveillance and never validated. Myalgia and eosinophilia are reported at 100% because the definition requires them, and milder illness without one of them falls outside the counted population. This is why the two definitional phenotypes here carry no `frequency` at all and every other frequency names its cohort.

Show evidence (1 reference)

PMID:8895176 SUPPORT REVIEW SYNTHESIS Human Clinical

"The widely disseminated surveillance case definition of the eosinophilia-myalgia syndrome (EMS) recommended by the Centers for Disease Control in 1989 has never been validated by an appropriate challenge and has commonly been used for unintended purposes."

States both halves of the problem: never validated, and used for things it was not built for.

Is the illness that followed contaminated L-5-hydroxytryptophan the same disease as the 1989 L-tryptophan epidemic?

OPEN QUESTION five_htp_related_illness

Attached to

pathophysiology#Systemic Exposure to L-Tryptophan Process Contaminants

One family became ill after taking L-5-HTP that contained an impurity absent from comparison samples. One member met criteria for the syndrome and two others had eosinophilia that resolved when the product was replaced. An NIH observational study was built around such cases. Whether this is the same disease reached through a different supplement is unsettled, and this entry stays scoped to the L-tryptophan epidemic rather than deciding it.

Show evidence (3 references)

PMID:7699627 SUPPORT PRIMARY RESULT Human Clinical

"The L-5-HTP used by the family contained an impurity not present in samples from the other patient group."

The chemical finding that makes the parallel worth recording.

PMID:7699627 SUPPORT PRIMARY RESULT Human Clinical

"One member of the family had EMS, and 2 others had eosinophilia."

The clinical extent, and its limit. One case meeting criteria is why the title of that paper says "related disorder".

clinicaltrials:NCT00001918 SUPPORT BACKGROUND Other

"More recently, similar impurities have been detected in batches of a similar dietary supplement called L-5-hydroxytryptophan."

An NIH study description recording that the same class of impurity turned up in 5-HTP. Graded OTHER because a trial registration document is not study evidence.

⚙

Pathophysiology

14

Ingestion of Contaminated Manufactured L-Tryptophan

Oral L-tryptophan sold over the counter as a sleep and mood supplement. Case exposure traced to a single manufacturer whose fermentation process had changed, and the epidemic ended when the product was recalled.

Downstream

Systemic Exposure to L-Tryptophan Process Contaminants The manufacturing conditions, not the amino acid, are what the case-control study associated with illness.

Show evidence (3 references)

PMID:2370887 SUPPORT PRIMARY RESULT Human Clinical

"29 of 30 case patients (97 percent) and 21 of 35 controls (60 percent) had consumed tryptophan manufactured by a single company (odds ratio, 19.3; 95 percent confidence interval, 2.5 to 844.9; P less than 0.001)"

The single-manufacturer association with its own odds ratio and interval, which is why this node names manufactured product rather than tryptophan.

PMID:2370887 SUPPORT PRIMARY RESULT Human Clinical

"This company used a fermentation process involving Bacillus amyloliquefaciens to manufacture tryptophan."

Names the production route behind the implicated lots.

PMID:8895184 SUPPORT REVIEW SYNTHESIS Human Clinical

"There was a dose-response effect, with risk of illness increasing as a function of the amount of tryptophan consumed."

The dose-response element of the causal argument, which an association with a manufacturer alone would not supply.

Systemic Exposure to L-Tryptophan Process Contaminants

Case lots carried more than one trace impurity. EBT, first reported as an HPLC peak and then characterised as a tryptophan dimer, is the compound with the most experimental work behind it and is carried here as the worked candidate rather than as the established agent.

A contaminant drives a type 2 immune response that recruits and activates eosinophils A T cell response against extracellular matrix sustains the chronic disease The contaminant acts directly on fibroblasts

response to xenobiotic stimulus GO:0009410

Downstream

Type 2 Cytokine Response to Peak E How peak E reaches a cytokine-producing cell is not described by any cited source. What is measured is the cytokine output of mononuclear cells exposed to it, in subjects who did not have EMS.

Eosinophil Chemotaxis and Endothelial Adhesion

Fibroblast Activation and Type I Collagen Overproduction The edge that makes the alternative hypothesis a separate route rather than a restatement: peak E raises procollagen transcription in dermal fibroblasts with no immune cell in the dish.

T Cell Mediated Immunity Against Extracellular Matrix The contaminant reaches the inflammatory cells as well as the eosinophils. The review that states this names no intermediate step, so the link type records none.

Show evidence (4 references)

PMID:2270484 SUPPORT PRIMARY RESULT Other

"Spectral and chemical studies now demonstrate that peak E is 1,1'-ethylidenebis[tryptophan]."

Establishes the chemical identity of the compound this node binds. Graded OTHER because it is an analytical chemistry result on product samples rather than a study of people, animals or cells.

PMID:2270484 SUPPORT PRIMARY RESULT Other

"This novel amino acid may be the etiological agent responsible for EMS, or it may be a marker of a still unidentified causal agent."

The authors' own hedge in the paper that named the compound, and the reason this node is titled for the contaminants collectively.

PMID:8356958 REFUTE PRIMARY RESULT Human Clinical

"higher EBT levels were still associated with a lot's case status, but the association lacked statistical significance (p = 0.120, odds ratio = 1.56, 95% confidence interval 0.758-3.23)"

REFUTE against EBT alone as the agent. Stratifying lots by date of manufacture removes the significance of the EBT association, which is the single strongest reason this node does not name one compound.

+ 1 more reference

Hepatic Conversion of PAP to 3-(Phenylamino)alanine

PAA was isolated from L-tryptophan associated with EMS, and liver tissue converts the toxic oil syndrome marker PAP into it. Bioactivation of PAA by human liver microsomes then yields 4-aminophenol through a quinoneimine, which is the same metabolite PAP yields. This node is the chemical bridge between the two epidemics.

response to xenobiotic stimulus GO:0009410

Show evidence (3 references)

PMID:8555405 SUPPORT PRIMARY RESULT In Vitro

"A related aniline derivative, 3-(phenylamino)-L-alanine (PAA), was recently isolated from L-tryptophan associated with the onset of EMS."

Establishes that PAA was present in the implicated product, which is what puts this node in an EMS entry at all.

PMID:8555405 SUPPORT PRIMARY RESULT In Vitro

"Here, we demonstrate the biotransformation of PAP into PAA by both rat hepatocytes and human liver tissue."

The measured conversion, in rat hepatocytes and human liver tissue.

PMID:17892268 SUPPORT PRIMARY RESULT In Vitro

"These findings establish that EMS and TOS are linked by a common toxic metabolite (4-aminophenol)"

Extends the link from a shared compound to a shared reactive metabolite, which is the claim this node makes.

Type 2 Cytokine Response to Peak E

Mononuclear cells exposed to peak E release IL-5 and IL-10. IL-5 is the eosinophil growth and survival cytokine, and reviews of the syndrome name it and TGF-beta as the important mediators.

A contaminant drives a type 2 immune response that recruits and activates eosinophils

interleukin-5 production GO:0032634 ↑ INCREASED

Downstream

Eosinophil Expansion with Tissue Degranulation

Show evidence (2 references)

PMID:8423409 SUPPORT REVIEW SYNTHESIS Human Clinical

"Pathologic observations and experimental studies indicate that eosinophils, mononuclear inflammatory cells, and fibroblasts are potential effector cells, and interleukin-5 and transforming growth factor-beta are important mediators in the pathogenesis of the syndrome."

Names IL-5 as a mediator of the syndrome, which is the cytokine this node is about. The word "potential" is the authors' own and is why the node's outgoing edge carries the hypothesis group.

PMID:7797795 REFUTE PRIMARY RESULT In Vitro

"this response is caused by endotoxin contamination of the L-tryptophan products and not by a specific L-tryptophan contaminant"

REFUTE, and the most useful negative result in the entry. An earlier GM-CSF response to implicated L-tryptophan turned out to be driven by endotoxin in the product, so a cytokine response measured in this assay system is not by itself evidence about the EMS contaminant.

Eosinophil Chemotaxis and Endothelial Adhesion

The contaminants make eosinophils migrate and make them stick to endothelium. Adhesion to stimulated human umbilical vein endothelial cells is blocked by antibody to ICAM-1 and not by antibody to VCAM-1, and the eosinophils' own integrin expression does not change, so the effect is on the endothelial side.

A contaminant drives a type 2 immune response that recruits and activates eosinophils

eosinophil CL:0000771 endothelial cell CL:0000115

eosinophil migration GO:0072677 ↑ INCREASED

Downstream

Eosinophil Expansion with Tissue Degranulation

Show evidence (2 references)

PMID:8906111 SUPPORT PRIMARY RESULT In Vitro

"Purified eosinophils adhered to peak-E- or peak-5-stimulated human umbilical vein endothelial cells, and this adherence was inhibited by the presence of antibody to intercellular adhesion molecule-1"

The adhesion step and the adhesion molecule it runs through.

PMID:8906111 SUPPORT PRIMARY RESULT In Vitro

"Human peripheral blood mononuclear cells (PBMCs) produced eosinophil survival-enhancing activity when cultivated with peak-E, but not with medium alone, peak-5 or control tryptophan."

Survival enhancement specific to peak E among the conditions tested, which is how a migration effect becomes an accumulation.

Eosinophil Expansion with Tissue Degranulation

Blood eosinophilia with extracellular deposition of eosinophil granule proteins in tissue, and raised major basic protein and eosinophil-derived neurotoxin in serum and urine. The count is the case definition; the granule proteins are the evidence of activity.

A contaminant drives a type 2 immune response that recruits and activates eosinophils

eosinophil CL:0000771

eosinophil degranulation GO:0043308 ↑ INCREASED eosinophil activation GO:0043307 ↑ INCREASED

Downstream

Inflammatory Infiltration of Fascia, Perimysium and Dermis Recorded as unknown intermediates because the tissue the eosinophil products are deposited in contains almost no eosinophils. Whatever connects the two is not described.

Pulmonary Eosinophilic and Lymphocytic Inflammation Unlike the fascia, the lung does show the eosinophil: lavage fluid from affected patients carries them in excess. This is the one compartment where the cell and the lesion are found in the same place.

Peripheral eosinophilia

Show evidence (3 references)

PMID:2314421 SUPPORT PRIMARY RESULT Human Clinical

"All three patients had elevated serum and urinary levels of this protein and eosinophil-derived neurotoxin, indicative of eosinophil degranulation."

Degranulation measured in the patients themselves, in the paper that first reported the association.

PMID:2314421 SUPPORT PRIMARY RESULT Human Clinical

"revealed eosinophilic infiltration, as well as the extracellular deposition of eosinophil-granule major basic protein"

The tissue deposition of granule protein, which is the part of this node that a blood count does not establish.

PMID:1563745 REFUTE PRIMARY RESULT Human Clinical

"Minimal tissue eosinophilia was seen despite the extent of blood eosinophilia."

REFUTE against tissue eosinophil accumulation as the lesion. The blood count and the tissue count come apart, which is why the downstream edge is typed INDIRECT_UNKNOWN_INTERMEDIATES.

T Cell Mediated Immunity Against Extracellular Matrix

Fascia and perimysium carry CD8+ cells, CD4+ cells and macrophages, scattered and perivascular, with MHC class I expression on muscle fibres near vessels and fascicles. Most fibroblasts are activated and a minority express HLA-DR, which the authors read as the fibroblast being the target.

A T cell response against extracellular matrix sustains the chronic disease

CD8-positive, alpha-beta T cell CL:0000625 macrophage CL:0000235

T cell mediated immunity GO:0002456 ↑ INCREASED

Downstream

Inflammatory Infiltration of Fascia, Perimysium and Dermis

Pulmonary Eosinophilic and Lymphocytic Inflammation The same CD8+ compartment found in fascia and perimysium is recoverable from the lung.

Fibroblast Activation and Type I Collagen Overproduction

Show evidence (2 references)

PMID:8100551 SUPPORT PRIMARY RESULT Human Clinical

"We found inflammatory cells consisting of CD8+ cells (45% +/- 8.9%), T4 cells (36% +/- 10.1%), and macrophages (19% +/- 12%), scattered or perivascularly in the fascia, the perimysium, and the endomysial septae."

The cell populations and their distribution, counted.

PMID:8100551 SUPPORT PRIMARY RESULT Human Clinical

"Only rare granulated or degranulating eosinophils were noted."

Why this node exists beside the eosinophil node rather than downstream of it. In the tissue the eosinophil is almost absent and the T cell is not.

Inflammatory Infiltration of Fascia, Perimysium and Dermis

The earliest changes sit in the septa between subcutaneous fat lobules and in the deep dermis or fascia, infiltrated with lymphocytes and histiocytes alongside reactive mesenchymal cells. Skeletal muscle shows perimysial, epimysial and fascial infiltrate. Dermal and fascial vessel walls thicken and endothelium swells without overt vasculitis.

macrophage CL:0000235 mast cell CL:0000097

inflammatory response GO:0006954 ↑ INCREASED

Animal models

EBT-treated Lewis rat Rat partially recapitulates Reproduces the fascial and perimysial infiltrate, including the sparse perineurial component, from the candidate compound.

Downstream

Microangiopathy of Small Vessels A review of the pathophysiology reads the cellular immune response as leading to a microangiopathy. The step is asserted at review level and the intermediates are not named.

Fibroblast Activation and Type I Collagen Overproduction

Perineural Inflammation and Axonal Injury

Show evidence (4 references)

PMID:1563745 SUPPORT PRIMARY RESULT Human Clinical

"Skeletal muscle biopsies showed a perimysial, epimysial, and/or fascial inflammatory infiltrate of lymphocytes and distinctive reactive mesenchymal cells with some eosinophils."

The muscle-compartment distribution of the infiltrate, from 11 biopsied cases.

PMID:1563745 SUPPORT PRIMARY RESULT Human Clinical

"Vessel walls in the dermis and fascia showed thickening and endothelial swelling, but no overt vasculitis was noted."

The vascular change, and its limit. This is the reason no vasculitis node appears in this entry, which is one of the differences from toxic oil syndrome.

PMID:8895179 SUPPORT REVIEW SYNTHESIS Human Clinical

"Such responses are most marked within the dermis, subcutis, fascia, and connective tissue in and around muscles, nerves, and other tissues."

The tissue distribution this node is named for, at review level.

+ 1 more reference

Pulmonary Eosinophilic and Lymphocytic Inflammation

Lavage fluid from affected lungs carries excess eosinophils in some patients and excess lymphocytes in others, with a raised CD8+ proportion, and it also carries activity that stimulates fibroblast proliferation. That activity falls to normal on corticosteroid treatment. High-resolution CT is abnormal in most patients examined, at times with a normal plain radiograph. It is the same three-cell story the entry models in fascia and dermis, in the compartment where most of the reported deaths had complications.

eosinophil CL:0000771 CD8-positive, alpha-beta T cell CL:0000625 fibroblast CL:0000057

inflammatory response GO:0006954 ↑ INCREASED fibroblast proliferation GO:0048144 ↑ INCREASED

Downstream

Dyspnea

Decreased DLCO

Show evidence (3 references)

PMID:1582284 SUPPORT PRIMARY RESULT Human Clinical

"Serial measurements of fibroblast proliferation-stimulating-activity in samples of BAL fluid obtained from serial examinations in two patients exhibited heightened pretreatment activity that returned to the normal range following corticosteroid therapy."

The fibroblast-activating arm of this node, measured in lung fluid, and the one measurement anywhere in the entry that falls back to normal under treatment. It is why `GO:0048144` is bound here as well as on the dermal fibroblast node.

PMID:1582284 SUPPORT PRIMARY RESULT Human Clinical

"In four patients, HRCT scanning of the chest was abnormal."

Imaging evidence that the lesion is in the lung parenchyma in most of the six patients examined.

PMID:8295183 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"Of the 36 patients who died, 33 (92%) had neuromuscular sequelae, 29 (81%) had pulmonary complications, and 23 (64%) had cardiac manifestations."

INDIRECT, and carried for scale rather than for mechanism: pulmonary complications in 81% of the deaths reported to surveillance is why this node exists at all. The sentence counts complications and does not describe the lesion.

Microangiopathy of Small Vessels

Small-vessel change is described in the pathophysiology review as the consequence of the cellular immune response, and postmortem hearts show focal fibromuscular dysplasia narrowing arteries alongside endarteritis and panarteritis. Whether the same lesion accounts for the chronic ischaemic symptoms is unsettled.

endothelial cell CL:0000115

Downstream

Cardiac Arterial and Conduction System Fibrosis

Show evidence (1 reference)

PMID:8895179 SUPPORT INDIRECT REVIEW SYNTHESIS Human Clinical

"The pathophysiology of the chronic symptoms is poorly understood but may involve ischemia, neuropathy, and metabolic abnormalities."

INDIRECT and quoted for what it concedes. Ischaemia is named as a possible contributor to the chronic symptoms, and the same sentence says the chronic pathophysiology is poorly understood.

Fibroblast Activation and Type I Collagen Overproduction

Fibroblasts cultured from affected skin produce more collagen and carry higher alpha 1(I) procollagen mRNA than matched normal cells, and the COL1A1 promoter drives more reporter activity in them. The phenotype is transcriptional and it persists into the chronic stage, years after the supplement was stopped. TGF-beta 1 transcripts are abundant in fibroblasts of affected fascia and absent from fibroblasts in the adjacent dermis of the same specimen.

fibroblast CL:0000057

collagen biosynthetic process GO:0032964 ↑ INCREASED transforming growth factor beta receptor signaling pathway GO:0007179 ↑ INCREASED fibroblast proliferation GO:0048144 ↑ INCREASED

Animal models

EBT-treated C57BL/6 mouse Mouse recapitulates The collagen programme and its TGF-beta 1 association are reproduced in mouse skin, which is the human in vitro finding shown in vivo.

Downstream

Fascial and Dermal Fibrosis

Show evidence (6 references)

PMID:8230009 SUPPORT PRIMARY RESULT In Vitro

"Cell lines derived from the affected skin from patients with EMS exhibited greater collagen production and higher steady state levels of alpha 1(I) procollagen mRNA compared with fibroblasts from age and sex matched healthy individuals."

The overproduction phenotype against matched controls.

PMID:9076948 SUPPORT PRIMARY RESULT In Vitro

"The study shows increased CAT activity driven by a 174 bp fragment of COL1A1 in transiently transfected skin fibroblasts from patients with EMS even in the chronic stage of disease."

Locates the effect in the COL1A1 promoter and shows it persists into chronic disease, which is what makes this node an ongoing state rather than a memory of the acute illness.

PMID:1702819 SUPPORT PRIMARY RESULT Human Clinical

"In situ hybridizations of affected fascia with a human sequence-specific TGF-beta 1 cDNA demonstrated numerous fibroblasts displaying positive hybridization signals indicative of high levels of transcripts for this cytokine."

The TGF-beta 1 binding on this node, measured in situ in affected fascia rather than inferred from the fibrosis.

+ 3 more references

Fascial and Dermal Fibrosis

Dermal thickening with homogenised collagen bundles replacing fat and adnexa, indistinguishable from scleroderma or morphea, and fascial thickening. This is the lesion behind the chronic sclerodermatous skin disease and the fasciitis.

extracellular matrix organization GO:0030198 ↑ INCREASED

Animal models

EBT-treated C57BL/6 mouse Mouse partially recapitulates Reproduces the tissue lesion in the right compartments from the candidate compound alone.

Downstream

Sclerodermatous skin induration

Eosinophilic fasciitis

Show evidence (1 reference)

PMID:1727618 SUPPORT REVIEW SYNTHESIS Human Clinical

"Mononuclear cell activation and infiltration of various affected tissues as well as fibrosis of the integument and of the connective tissue components of blood vessels, nerves, and muscles are additional frequent findings."

The distribution of the fibrosis across integument, vessel, nerve and muscle connective tissue, which is what this node claims.

Perineural Inflammation and Axonal Injury

Perivascular infiltrates and fibrosis in the connective tissue of nerve, presenting clinically as a sensorimotor polyneuropathy that in some patients has prominent demyelinating features on electrodiagnostic study. Neuropathy is the one physical finding that did not improve over two years of follow-up.

Downstream

Peripheral neuropathy

Demyelinating sensorimotor polyneuropathy

Muscle weakness The weakness that mattered clinically was the weakness of the neuromuscular disease, and it is the step surveillance data traced onward to respiratory failure.

Show evidence (3 references)

PMID:1323757 SUPPORT PRIMARY RESULT Human Clinical

"We describe 3 patients with EMS who presented with a severe demyelinating sensorimotor polyneuropathy."

The clinical form of the nerve lesion this node describes.

PMID:8295183 SUPPORT PRIMARY RESULT Human Clinical

"The most commonly observed disease process leading to death was progressive polyneuropathy and myopathy (24 of the 36 reported deaths) which produced complications of pneumonia and sepsis or respiratory failure due to weakness"

Establishes that this node, and not the fibrosis, is the route to most of the deaths reported to surveillance.

PMID:8285738 SUPPORT PRIMARY RESULT Human Clinical

"Prominent staining of transforming growth factor-beta was also present in the perimysial connective tissue of five (63%) of eight EMS muscle biopsy specimens and one sural nerve biopsy specimen."

The only molecular observation in a peripheral nerve specimen anywhere in the cited literature, and it puts the same growth factor in the nerve that drives the fibrosis elsewhere. One nerve, so it is a single observation rather than a series.

Cardiac Arterial and Conduction System Fibrosis

In three hearts examined after death from the disease, arterial narrowing and arteritis were widespread, neuritis and ganglionitis were present throughout the heart including the conduction system, and dense fibrosis had replaced all sinus node tissue.

Downstream

Sudden cardiac death The authors argue the lesions supply an anatomic substrate for electrical instability. A substrate is not a demonstrated mechanism of death, and the link is typed accordingly.

Show evidence (1 reference)

PMID:2058857 SUPPORT PRIMARY RESULT Human Clinical

"Within the sinus node, areas of dense fibrosis replaced all nodal tissue."

The conduction-system lesion this node is named for.

⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.

●

Phenotypes

25

Blood 1

Peripheral eosinophilia Increased total eosinophil count HP:0001880

Show evidence (2 references)

PMID:2314421 SUPPORT PRIMARY RESULT Human Clinical

"All three patients, who were women 37 to 44 years of age, had severe muscle pain, muscle weakness, mouth ulcers, and striking eosinophilia"

The eosinophilia in the index series. As with myalgia, no `frequency` is recorded because the case definition requires it.

PMID:21702023 SUPPORT PRIMARY RESULT Human Clinical

"Although treatment with prednisone and mycophenolate resulted in modest improvement in proximal muscle strength and myalgia along with prompt resolution of peripheral blood eosinophilia and ground glass opacifications seen on CT scans, skin induration and neuropathy progressed."

The count and the disease coming apart, in one patient and one sentence: the eosinophilia resolved promptly while the skin and nerve disease got worse. It is the source for this record's claim that a normal count does not establish remission.

Cardiovascular 2

Pulmonary arterial hypertension HP:0002092

Show evidence (1 reference)

PMID:1582284 SUPPORT PRIMARY RESULT Human Clinical

"Another remains markedly dyspneic with pulmonary hypertension."

A single patient in a six-patient series, which is why no frequency is recorded. Toxic oil syndrome's pulmonary hypertension is far better characterised and is not imported here.

Sudden cardiac death HP:0001645

Show evidence (2 references)

PMID:8295183 SUPPORT PRIMARY RESULT Human Clinical

"cardiomyopathy was the underlying cause of death for 4 patients, primary pulmonary disease for 3, sudden death attributed to arrhythmia for 2, stroke for 2, and septic complications of therapy for one"

The specific claim this record makes, in the source's own words: two of the 36 reported deaths were sudden and attributed to arrhythmia. Frequency is deliberately absent, because these are proportions of deaths rather than of cases.

PMID:8295183 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"Of the 36 patients who died, 33 (92%) had neuromuscular sequelae, 29 (81%) had pulmonary complications, and 23 (64%) had cardiac manifestations."

INDIRECT, and kept for context rather than for the claim: cardiac involvement of some kind in 64% of the deaths is the background against which two sudden deaths sit.

Digestive 1

Dysphagia FREQUENT HP:0002015

Self-reported, with the same limits as the xerostomia record. No cited source reports a swallowing study, and no edge is drawn to it: whether the dysphagia is myopathic, neuropathic or fibrotic is not addressed anywhere in this literature.

Show evidence (1 reference)

PMID:7857025 SUPPORT PRIMARY RESULT Human Clinical

"Among the 28 various head and neck manifestations studied, 70% of EMS patients complained of generalized muscle spasms, 66% xerostomia, 62% dyspnea, and 56% dysphagia."

56% for dysphagia, from the same survey and the same sentence.

Head and Neck 2

Xerostomia FREQUENT HP:0000217

Self-reported on a questionnaire, with no objective measure of salivary flow anywhere in the cited literature, and the respondents were reached through patient contact rather than sampled. The frequency is therefore a share of respondents. It is recorded because dryness is otherwise absent from this entry and is prominent in the sibling toxic oil syndrome.

Show evidence (1 reference)

PMID:7857025 SUPPORT PRIMARY RESULT Human Clinical

"Among the 28 various head and neck manifestations studied, 70% of EMS patients complained of generalized muscle spasms, 66% xerostomia, 62% dyspnea, and 56% dysphagia."

66% for xerostomia, which is the frequency this record follows. The same sentence carries the dysphagia figure used below and corroborates the dyspnoea figure taken from surveillance.

Oral ulcer HP:0000155

Show evidence (1 reference)

PMID:2314421 SUPPORT PRIMARY RESULT Human Clinical

"All three patients, who were women 37 to 44 years of age, had severe muscle pain, muscle weakness, mouth ulcers, and striking eosinophilia"

Three of three in the index series, so no frequency. No later cohort cited here reports the feature.

Immune 2

Skin rash FREQUENT HP:0000988

Show evidence (2 references)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

Rash in 60% of surveillance-reported cases.

PMID:2273104 SUPPORT PRIMARY RESULT Human Clinical

"Overall, cutaneous manifestations developed in 26 patients (87%)."

The broader skin-involvement figure from prospective follow-up. The recorded `frequency` follows the surveillance rash figure rather than this one, because 87% counts every cutaneous manifestation including the sclerodermatous ones curated separately.

Eosinophilic fasciitis FREQUENT HP:0045029

Show evidence (2 references)

PMID:2273104 SUPPORT PRIMARY RESULT Human Clinical

"Clinical and/or biopsy evidence of eosinophilic fasciitis was seen in nine patients (30%)."

30% of 30 prospectively followed patients, which is FREQUENT.

PMID:2369429 SUPPORT PRIMARY RESULT Human Clinical

"We describe 4 patients with EMS, 3 of whom presented with eosinophilic fasciitis, and we review the epidemiology of EMS reported in Maryland"

Fasciitis as the presenting picture in three of four reported cases, from an independent series.

Integument 2

Sclerodermatous skin induration FREQUENT HP:0100324

Show evidence (2 references)

PMID:2273104 SUPPORT PRIMARY RESULT Human Clinical

"The most characteristic abnormality noted was the spectrum of sclerodermatous disease in 15 patients (50%) often after a subacute stage of peripheral or truncal edema."

50% in prospective follow-up of 30 patients, which is FREQUENT.

PMID:1673503 SUPPORT PRIMARY RESULT Human Clinical

"The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."

54% in an independent cohort, agreeing with the 50% above. Two cohorts is why this frequency is stated with more confidence than most here.

Alopecia FREQUENT HP:0001596

Show evidence (1 reference)

PMID:2273104 SUPPORT PRIMARY RESULT Human Clinical

"Alopecia, frequently a late sequela, developed in 11 (37%)."

37% of 30 patients, and its late timing, from the same series.

Metabolism 4

Peripheral edema FREQUENT HP:0012398

Show evidence (2 references)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

Peripheral oedema in 59% of surveillance-reported cases.

PMID:2273104 SUPPORT PRIMARY RESULT Human Clinical

"The most characteristic abnormality noted was the spectrum of sclerodermatous disease in 15 patients (50%) often after a subacute stage of peripheral or truncal edema."

Places the oedema in sequence before the skin induration.

Fever HP:0001945

Show evidence (1 reference)

PMID:2314421 SUPPORT PRIMARY RESULT Human Clinical

"Other manifestations included fever, abdominal pain, dyspnea, skin rash, and elevated serum concentrations of aminotransferase and aldolase."

Lists fever among the acute manifestations of the index three cases. No frequency is recorded: the published frequencies for fever in this disease come from cohorts this entry does not cite.

Elevated circulating aldolase concentration FREQUENT HP:0012544

Show evidence (2 references)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

46% in surveillance, which is the figure the frequency band follows.

PMID:2001136 SUPPORT PRIMARY RESULT Human Clinical

"Aldolase levels were abnormal in all patients tested."

Abnormal in every tested patient of a clinical series, which is higher than the surveillance figure. The band follows the surveillance number because the series does not say how many were tested.

Elevated circulating hepatic transaminase concentration FREQUENT HP:0002910

Show evidence (2 references)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

43% in surveillance.

PMID:2001136 SUPPORT PRIMARY RESULT Human Clinical

"Fourteen (88%) of the 16 patients tested had mild liver function abnormalities."

88% of those tested in a clinical series, and the word "mild", which the surveillance figure does not supply.

Musculoskeletal 2

Muscle weakness FREQUENT HP:0001324

Show evidence (3 references)

PMID:8624177 SUPPORT PRIMARY RESULT Human Clinical

"survivors with definite EMS continued to report excess morbidity for 6 major EMS symptoms (myalgia, arthralgia, weakness, rash, alopecia, and sclerodermiform skin changes)"

Weakness among six symptoms with excess morbidity in survivors against exposed-but-not-ill controls. It is not the source of the frequency band: this sentence reports persisting excess rather than a share of cases.

PMID:1520057 SUPPORT PRIMARY RESULT Human Clinical

"At 12 months, 41 patients (77%) continued to report fatigue, 36 (68%) weakness, and 34 (64%) myalgias"

68% at twelve months, which is the FREQUENT band and the figure this record follows. The cohort is population-based rather than referral, which is why it is preferred to the higher referral-cohort numbers.

PMID:8129767 SUPPORT PRIMARY RESULT Human Clinical

"Fatigue (91%), muscle cramping (75%), myalgia (70%), paresthesias with objectively demonstrated hypesthesias (62%), articular symptoms (54%), scleroderma-like skin changes (44%), and proximal muscle weakness (40%) are the most common features of chronic EMS."

Proximal weakness in 40% at a mean of three years, in a 57-patient prospective cohort. The same sentence carries the frequencies used on `Fatigue`, `Paresthesia` and `Proximal myopathy`.

Proximal myopathy FREQUENT HP:0003198

Show evidence (1 reference)

PMID:1673503 SUPPORT PRIMARY RESULT Human Clinical

"The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."

36% in a 32-patient prospectively followed cohort, which is FREQUENT. `preferred_term` is narrower than the bound HPO term, which has no proximal-myopathy child.

Nervous System 4

Peripheral neuropathy FREQUENT HP:0009830

Course: PROGRESSIVE

Show evidence (4 references)

PMID:1673503 SUPPORT PRIMARY RESULT Human Clinical

"The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."

61% in a followed cohort, which is FREQUENT.

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"Neuropathy or neuritis, resulting in paralysis and death in some patients, was seen in 27%, and chest roentgenogram abnormalities were noted in 21% of those tested."

The surveillance figure, which is lower than the cohort one, and the severity at its extreme. The band follows the cohort figure because the surveillance count depends on what reporting physicians recorded.

PMID:7741371 SUPPORT PRIMARY RESULT Human Clinical

"Nearly all physical findings were also reported to have improved or resolved in most patients; only peripheral neuropathy was unchanged."

The sentence that separates this phenotype from every other in the entry: at two years it was the one physical finding that had not improved. Read exactly, it supports a failure to remit and not the `clinical_course` qualifier, which is carried by the surveillance item below instead. An earlier draft cited this sentence for PROGRESSIVE, which inverts it.

+ 1 more reference

Demyelinating sensorimotor polyneuropathy Demyelinating peripheral neuropathy HP:0007108

Show evidence (2 references)

PMID:1323757 SUPPORT PRIMARY RESULT Human Clinical

"We describe 3 patients with EMS who presented with a severe demyelinating sensorimotor polyneuropathy."

A three-patient report, so no frequency is recorded. Carried separately from `Peripheral neuropathy` because the electrophysiology and the outcome are different.

PMID:1323757 SUPPORT PRIMARY RESULT Human Clinical

"Despite plasmapheresis; corticosteroids; and, in 1 patient, cyclophosphamide, 2 patients died and the remaining patient experienced minimal recovery."

The outcome, and the reason this form is separated out.

Paresthesia FREQUENT HP:0003401

Show evidence (2 references)

PMID:8129767 SUPPORT PRIMARY RESULT Human Clinical

"Fatigue (91%), muscle cramping (75%), myalgia (70%), paresthesias with objectively demonstrated hypesthesias (62%), articular symptoms (54%), scleroderma-like skin changes (44%), and proximal muscle weakness (40%) are the most common features of chronic EMS."

62% with objectively demonstrated hypesthesia, which is the frequency this record follows. The objective confirmation is what separates this from a purely reported symptom.

PMID:1520057 SUPPORT PRIMARY RESULT Human Clinical

"Symptoms with later onset included paresthesias, muscle cramps, extremity weakness, and alopecia."

The timing: paresthesia belongs to the later illness rather than to the acute phase, which is the sequence this record's description asserts.

Cognitive impairment VERY_FREQUENT HP:0100543

Show evidence (4 references)

PMID:7741371 SUPPORT PRIMARY RESULT Human Clinical

"Cognitive changes were reported to be worse in 32% of patients."

Not the source of the frequency band. 32% is the share reported worse at follow-up rather than the share affected, and the two are different quantities.

PMID:8129767 SUPPORT PRIMARY RESULT Human Clinical

"New findings identified among this cohort include cognitive symptoms in 86% of the study group, tremor, and myoclonus."

86% in a 57-patient prospective cohort, which is the VERY_FREQUENT band and the figure this record follows. It is a single referral cohort, which the description says. The tremor and myoclonus in the same sentence are real findings and are not curated as phenotypes here, because nothing in the cited literature quantifies or localises them.

PMID:9802492 SUPPORT PRIMARY RESULT Human Clinical

"Neurologic findings that were increased in CNS-EMS included minor depression (100%), amnesia (88%), and intermittent confusion (38%)"

Frequencies within a series selected for CNS abnormality, which is why they are not read as disease frequencies either.

+ 1 more reference

Respiratory 2

Dyspnea FREQUENT HP:0002094

Show evidence (1 reference)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

Cough or dyspnoea in 59%. The surveillance category pools the two symptoms, which is why the frequency sits on this record and not on a separate cough record.

Decreased DLCO HP:0045051

Show evidence (2 references)

PMID:1582284 SUPPORT PRIMARY RESULT Human Clinical

"The diffusing capacity for carbon monoxide was decreased in five patients tested."

The measurement. Note the source gives no denominator for the testing, so this record says five of the six reported rather than five of five tested. No `frequency` is given either way: a six-patient series selected for dyspnoea cannot supply a disease frequency.

PMID:1582284 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"Two patients undergoing BAL exhibited increased eosinophils in the lavage fluid; a third had elevated lymphocytes."

INDIRECT: lavage eosinophilia indicates the lung compartment is involved by the same process, which is an inference from the cell counts to the cause of the gas transfer defect. The same sentence gives the denominator this record uses: a third patient also had lavage, so it is two of three rather than two of two.

Constitutional 3

Myalgia HP:0003326

Show evidence (2 references)

PMID:2001136 SUPPORT PRIMARY RESULT Human Clinical

"All cases involved eosinophilia (eosinophil count, greater than or equal to 2.0 x 10(9)/L) and incapacitating myalgias."

The severity in a defined case series. No `frequency` is recorded here because the definition requires the feature - see the entry `notes:`.

PMID:1673503 SUPPORT PRIMARY RESULT Human Clinical

"The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."

Cited here for the chronic form: episodic myalgia in 64% of a followed cohort, which is a different claim from the definitional acute myalgia.

Arthralgia FREQUENT HP:0002829

Show evidence (1 reference)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

73% in national surveillance, which is FREQUENT on the HPO band. The same sentence carries five other frequencies used elsewhere in this entry.

Fatigue FREQUENT HP:0012378

Show evidence (3 references)

PMID:1563745 SUPPORT INDIRECT BACKGROUND Human Clinical

"Arthralgias, edema of the extremities, morbilliform rashes, skin induration, weakness, fatigue, and respiratory weakness may be present as well."

BACKGROUND because this is the pathology paper's framing sentence describing the established clinical picture rather than its own finding, and INDIRECT because "may be present" is a statement about the syndrome rather than an observation in the 11 cases.

PMID:1520057 SUPPORT PRIMARY RESULT Human Clinical

"At 12 months, 41 patients (77%) continued to report fatigue, 36 (68%) weakness, and 34 (64%) myalgias"

77% at twelve months in a population-based cohort, which is the figure the FREQUENT band follows. The referral cohort below reports 91%, which would be VERY_FREQUENT; the lower population-based figure is preferred because the cohort was not selected by referral.

PMID:8129767 SUPPORT PRIMARY RESULT Human Clinical

"Fatigue (91%), muscle cramping (75%), myalgia (70%), paresthesias with objectively demonstrated hypesthesias (62%), articular symptoms (54%), scleroderma-like skin changes (44%), and proximal muscle weakness (40%) are the most common features of chronic EMS."

91% at a mean of three years, the highest figure for any feature in this disease. It is recorded rather than used for the band, for the reason given on the item above.

🧬

Genetic Associations

2

HLA-DRB1 alleles as susceptibility markers

Gene: HLA-DRB1 hgnc:4948 relationship_type: RISK_FACTOR

Show evidence (4 references)

PMID:19790128 SUPPORT PRIMARY RESULT Human Clinical

"higher LT dose (odds ratio [OR] 1.4, 95% confidence interval [95% CI] 1.1-1.8), age >45 years (OR 3.0, 95% CI 1.0-8.8), and HLA-DRB1*03 (OR 3.9, 95% CI 1.2-15.2), DRB1*04 (OR 3.9, 95% CI 1.1-16.4), and DQA1*0601 (OR 13.7, 95% CI 1.3-1.8) were risk factors for the development of EMS"

The positive associations with their own intervals, alongside dose and age, which is the point of the study and the reason this record is typed RISK_FACTOR rather than CAUSATIVE.

PMID:19790128 SUPPORT PRIMARY RESULT Human Clinical

"Similar risk and protective factors were seen for developing EMS following ingestion of implicated LT, except that DRB1*03 was not a risk factor and DQA1*0201 was an additional protective factor."

The implicated-lot analysis, which is the one this entry's scope actually needs, and its two departures from the all-sources result. It supports the record while narrowing it: among users of the lots that caused the epidemic, DRB1*03 was not a risk allele.

PMID:1673503 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"HLA-class II typing revealed a non-significant trend towards an association with HLA-DR4."

INDIRECT, and the weakest item on this record. An earlier and smaller cohort found the direction and not the significance, which is a failure to confirm rather than a contradiction: a non-significant trend the same way does not cut against the association. An earlier draft graded this REFUTE, which read a null result as a negative one.

+ 1 more reference

HLA-DQA1 alleles as protective markers

Gene: HLA-DQA1 hgnc:4942 relationship_type: PROTECTIVE

Show evidence (1 reference)

PMID:19790128 SUPPORT PRIMARY RESULT Human Clinical

"whereas DRB1*07 (OR 0.12, 95% CI 0.02-0.48) and DQA1*0501 (OR 0.23, 95% CI 0.05-0.85) were protective"

The protective associations with their intervals. The sentence names a DRB1 allele too, which is why the DRB1 record above is typed for risk on a different sentence rather than this one.

💊

Medical Actions

5

Withdrawal of the Implicated Supplement

Action: discontinuation of the causative supplement Ontology label: Pharmacotherapy Discontinuation NCIT:C128535

Stopping the L-tryptophan product. It is the only intervention that addresses the cause, and in the index series it was followed by improvement while patients remained symptomatic months later. At population level the recall ended the epidemic.

Show evidence (1 reference)

PMID:2314421 SUPPORT PRIMARY RESULT Human Clinical

"The discontinuation of tryptophan and the initiation of glucocorticoid treatment resulted in improvement, but all three women were still symptomatic three to five months later."

Improvement after withdrawal, and its limit. The same sentence bundles withdrawal with glucocorticoid, so neither can be credited alone from it.

Systemic Glucocorticoid Therapy

Action: Pharmacotherapy NCIT:C15986

Agent: prednisone CHEBI:8382

Platform: Small molecule

Prednisone in the acute phase. It was reported helpful in 79% of the patients who received it, and it reduces muscle pain and the eosinophil count. It does not prevent the chronic disease.

Show evidence (3 references)

PMID:7741371 SUPPORT PRIMARY RESULT Human Clinical

"Prednisone was reported to be helpful in 79% of patients who received it during the acute phase of the syndrome."

The acute-phase benefit, from physician-completed review of 205 patients. It is a report of helpfulness rather than a controlled comparison.

PMID:2001136 SUPPORT PRIMARY RESULT Human Clinical

"Seven patients were treated with prednisone, and six showed improvement in muscle pain and a decrease in eosinophilia."

Six of seven improving on two named measures, in an independent series.

PMID:1673503 REFUTE PRIMARY RESULT Human Clinical

"Early therapy with corticosteroids did not seem to prevent the development of chronic manifestations."

REFUTE against corticosteroids as disease-modifying. There is deliberately no `target_mechanisms` edge on this treatment: the acute benefit is on symptoms and the eosinophil count, and the one cohort that looked for prevention of chronic disease did not find it.

Steroid-Sparing Immunosuppression

Action: Immunosuppressive Therapy NCIT:C15261

Agent: mycophenolate mofetil CHEBI:8764 methotrexate CHEBI:44185

Platform: Small molecule

Mycophenolate, methotrexate and anakinra have been used in the one post-epidemic case reported in detail. Prednisone with mycophenolate cleared the eosinophilia and the pulmonary ground-glass change and improved strength and myalgia modestly, and the skin induration and neuropathy went on worsening through it. Methotrexate and then anakinra added nothing.

Show evidence (2 references)

PMID:21702023 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"Although treatment with prednisone and mycophenolate resulted in modest improvement in proximal muscle strength and myalgia along with prompt resolution of peripheral blood eosinophilia and ground glass opacifications seen on CT scans, skin induration and neuropathy progressed."

The partial benefit, in one patient, and INDIRECT because the regimen bundles mycophenolate with prednisone so neither can be credited alone. The same sentence is carried as REFUTE below against the fibrosis and the neuropathy, which it says got worse regardless.

PMID:21702023 REFUTE PRIMARY RESULT Human Clinical

"Addition of methotrexate (20 mg weekly for 5 months) followed by daily injections of anakinra for 3 months failed to yield further improvement in myalgia, neuropathy, or skin induration."

REFUTE against methotrexate and anakinra in this disease. It is one patient and an uncontrolled sequence, which is why `notes` says what this record does not establish.

Supportive and Symptomatic Care

Action: Supportive Care NCIT:C15747

Analgesia, respiratory management and management of the chronic complications. It is the mainstay because no other treatment was found consistently beneficial, and because most symptoms improved with time whatever was given.

Show evidence (1 reference)

PMID:7741371 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"No other treatment was reported to be consistently beneficial."

INDIRECT, and quoted for what it concedes: with prednisone the only agent reported to help and only in the acute phase, supportive care is what remains. It is not a study of supportive care.

Rehabilitation

Action: Rehabilitation NCIT:C15315

Platform: Behavioral / lifestyle

For the myopathy, the neuropathic disability and the contractures of the chronic phase.

Show evidence (1 reference)

PMID:1673503 SUPPORT INDIRECT PRIMARY RESULT Human Clinical

"The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."

INDIRECT: the source establishes the disabling sequelae that rehabilitation addresses, not the efficacy of rehabilitation in this disease, which no cited source reports.

🌍

Environmental Factors

1

Ingestion of manufactured L-tryptophan supplement containing process impurities

The bound term names a contaminant in an ingested product rather than tryptophan, and that is deliberate. ECTO does have a tryptophan exposure class, `ECTO:9002895 exposure to tryptophan`, returned by `runoak -i ols:ecto search "tryptophan"` - it is rejected on two grounds. Ordinary dietary tryptophan and uncontaminated manufactured tryptophan are not implicated in this disease, so binding the amino acid would assert what the epidemiology rules out; and that CURIE is not resolvable in the committed ECTO build (`runoak -i sqlite:obo:ecto info ECTO:9002895` returns no label), so it is not a member of the `ExposureTerm` enum and does not validate. The only other candidate, `ECTO:0070245 exposure to dietary supplement, vitamin(s) and fatty acids via ingestion` from `runoak -i ols:ecto search "dietary supplement"`, names a route and loses the contaminant. `preferred_term` carries the full concept.

Over-the-counter L-tryptophan taken as a sleep or mood supplement, from lots produced by one manufacturer after a change of fermentation strain and a reduction in the carbon purification step. Withdrawal of the product from the United States market ended the epidemic.

Show evidence (2 references)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"Since the recall of over-the-counter preparations of tryptophan in November 1989, the number of new cases of this potentially fatal disorder has fallen dramatically."

Removing the exposure removed the disease, which is the strongest single statement about this exposure in the surveillance literature.

PMID:21702023 SUPPORT BACKGROUND Human Clinical

"Following the ban by the Food and Drug Administration (FDA) on the sale of L-tryptophan, the incidence of EMS declined rapidly."

BACKGROUND: this is the post-epidemic case report's framing of the established history rather than its own result, and it is cited here because the same paper reports a case occurring after that decline.

Mechanism Target:

TRIGGERS Ingestion of Contaminated Manufactured L-Tryptophan — The exposure is the disease's cause on every published account. It was essentially universal among cases and rare among controls, and incidence fell away when the product was recalled.

Show evidence (1 reference)

PMID:8895184 SUPPORT REVIEW SYNTHESIS Human Clinical

"In case-control studies of EMS, LT exposure was essentially universal among cases but rare among controls."

States the exposure-disease link this edge asserts, across the case-control literature rather than in one study.

🔬

Diagnosis

3

High-Resolution CT of the Chest

Abnormal in four of the six patients in the one pulmonary series, and in one of them the plain chest radiograph was normal. So a normal radiograph does not exclude the lung involvement, which matters in a disease where most of the reported deaths had pulmonary complications.

Show evidence (1 reference)

PMID:1582284 SUPPORT PRIMARY RESULT Human Clinical

"One of these patients showed no abnormality on routine chest roentgenogram."

The specific claim this record makes: HRCT found disease that the plain film missed. One patient, in a six-patient series.

Electrodiagnostic Studies

Nerve conduction study and electromyography. In the severe form they show multifocal conduction block, slowed and temporally dispersed motor responses and prolonged or absent F-responses, which is what identifies the neuropathy as demyelinating rather than axonal.

Show evidence (1 reference)

PMID:1323757 SUPPORT PRIMARY RESULT Human Clinical

"Electrodiagnostic studies revealed multifocal conduction block, slowing and temporal dispersion of motor responses, and prolonged or absent F-responses."

The findings themselves, from the three-patient report of the demyelinating form. They are what the `Demyelinating sensorimotor polyneuropathy` phenotype rests on.

Muscle Biopsy

Shows eosinophilic perimyositis or interstitial inflammation. Note what it does not show: the infiltrate sits in perimysium, epimysium and fascia, and the muscle fibres themselves are close to intact, so a biopsy read for myofiber damage will look unremarkable beside the patient's pain.

This section carries only what a cited abstract states. The wider diagnostic workup the deep-research report describes, which includes MRI of symptomatic muscle and fascia, serial eosinophil counts and exclusion of parasitic infection, is general clinical practice that no reference cached here attributes to a study of this disease, so it is not recorded as though it were. The case definition itself is a `definitions:` entry rather than a diagnostic test.

Show evidence (2 references)

PMID:2001136 SUPPORT PRIMARY RESULT Human Clinical

"Muscle biopsies were done in five patients; four showed eosinophilic perimyositis, and one had interstitial inflammation."

What the biopsy found, in the five patients of a 21-patient series who had one. Four of five is a count within a small series and not a sensitivity.

PMID:1563745 SUPPORT PRIMARY RESULT Human Clinical

"The earliest apparent histologic changes were observed at the septa between subcutaneous fat lobules and in the deep dermis or fascia."

Where to look, and why a sampling decision matters here: the earliest change is septal and fascial, so a biopsy that stops above the fascia can miss it. This is the reason the literature recommends full-thickness sampling, although no cited sentence states that recommendation directly.

📈

Progression

3

Acute phase

Subacute onset of incapacitating generalised myalgia with blood eosinophilia, and with rash, peripheral oedema, fever, arthralgia, cough or dyspnoea, raised aldolase and abnormal liver tests. Eosinophil granule proteins are detectable in serum, urine and tissue.

Show evidence (1 reference)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."

The acute clinical picture with frequencies, from national surveillance.

Chronic fibrotic and neuromuscular phase

Sclerodermatous skin thickening, sensorimotor polyneuropathy, proximal myopathy and episodic myalgia, persisting after the eosinophilia has resolved. Nearly all findings improved over eighteen to twenty-four months except peripheral neuropathy, and cognitive change was reported worse in about a third.

Show evidence (2 references)

PMID:7741371 SUPPORT PRIMARY RESULT Human Clinical

"After 18 to 24 months, all symptoms except cognitive changes were reported to have improved in most patients."

The direction of the chronic phase and its one named exception.

PMID:1673503 SUPPORT PRIMARY RESULT Human Clinical

"The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."

The four sequelae that disable, with frequencies, in a followed cohort.

Late outcome

Excess mortality concentrated early. All-cause mortality over follow-up was 19% in definite cases against 3% in exposed-but-not-ill users, and two thirds of the deaths in definite cases happened within eighteen months of onset. Survivors kept reporting excess morbidity in six major symptoms, with severity falling over time.

Show evidence (3 references)

PMID:8624177 SUPPORT PRIMARY RESULT Human Clinical

"During the follow-up interval, mortality from all causes was 19% in those with definite EMS, 7% in possible EMS, and 3% in those who were not ill."

The mortality gradient across definite, possible and not-ill members of one tryptophan-exposed cohort, which is the comparison that makes the figure interpretable.

PMID:8624177 SUPPORT PRIMARY RESULT Human Clinical

"Six deaths (66%) among the definite EMS case patients occurred during the 18 months immediately after symptom onset."

The timing of the excess mortality, which is the prognostic point.

PMID:8295183 SUPPORT PRIMARY RESULT Human Clinical

"Older age and involvement of more than one organ system suggest a particularly poor prognosis"

The two prognostic factors surveillance data identified.

📊

Prevalence

1

United States, 1989-1990 epidemic

Cases In Literature Not yet documented

A point-source epidemic that ended with a product recall, so a population rate would mislead and no numeric slot is filled. `prevalence_class` is NOT_YET_DOCUMENTED because no rate has been documented, not because the epidemic went uncounted. Published death tolls differ with the surveillance cut-off: 27 by July 1990 and 36 by August 1991. Those are the same quantity at two dates and are reported with their dates rather than reconciled.

Show evidence (2 references)

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"As of July 10, 1990, a total of 1531 cases had been reported nationwide, including 27 deaths."

The case count and death toll with the surveillance date attached.

PMID:2398610 SUPPORT PRIMARY RESULT Human Clinical

"the highest rates of reported illness are concentrated in the western states, 68% are non-Hispanic white women aged 35 years and older"

The demographic concentration of reported cases, which tracked supplement use rather than any inherited factor.

⚖️

Clinical Burden

High

Mortality concentrated early and ran through the neuromuscular disease. Over follow-up of a tryptophan-exposed cohort, all-cause mortality was 19% in definite cases against 3% in exposed users who did not fall ill, and two thirds of those deaths came within eighteen months of onset. Of 36 deaths reported to national surveillance, 92% had neuromuscular sequelae and the commonest fatal process was progressive polyneuropathy and myopathy leading to pneumonia, sepsis or respiratory failure. Chronic morbidity in survivors is the larger burden: survivors of definite disease kept reporting excess myalgia, arthralgia, weakness, rash, alopecia and sclerodermiform skin change against exposed controls, with severity diminishing over time.

Show evidence (2 references)

PMID:8295183 SUPPORT PRIMARY RESULT Human Clinical

"Of the 36 patients who died, 33 (92%) had neuromuscular sequelae, 29 (81%) had pulmonary complications, and 23 (64%) had cardiac manifestations."

The organ involvement among the deaths, counted.

PMID:8624177 SUPPORT PRIMARY RESULT Human Clinical

"survivors with definite EMS continued to report excess morbidity for 6 major EMS symptoms (myalgia, arthralgia, weakness, rash, alopecia, and sclerodermiform skin changes)"

The six persisting symptoms, measured against exposed users who did not fall ill rather than against the general population.

📊

Related Datasets

1

Post-epidemic eosinophilia myalgia syndrome associated with L-Tryptophan geo:GSE26934

Expression profiling by array of lesional skin from the 2011 post-epidemic case, against comparison samples. This is the source of the TGF-beta and IL-4 signalling signature reported in PMID:21702023, and it is the only EMS-specific dataset in GEO.

human MICROARRAY n=6

Conditions: lesional skin from L-tryptophan-associated eosinophilia-myalgia syndrome

PMID:21702023

Six samples from a single post-epidemic case and its comparisons, so the series is a case-level profiling experiment rather than a cohort. The signature it carries is curated on `Fibroblast Activation and Type I Collagen Overproduction` as a `PMID:21702023` evidence item; an earlier draft of this note said so before that item existed, which made the note a false statement about the entry. The evidence item quotes the publication rather than this GEO record because the GEO summary states the disease and the exposure and not the result.

Show evidence (1 reference)

GEO:GSE26934 SUPPORT BACKGROUND Other

"The EMS epidemic in 1989 was linked to L-tryptophan consumption originating from a single source."

The GEO record's own summary establishes the series is about this disease and this exposure, which is the relevance check an accession resolving does not supply.

🔬

Clinical Trials

1

NCT00001918 NOT_APPLICABLE COMPLETED

An NIH observational study evaluating patients who developed eosinophilia-myalgia syndrome after taking L-5-hydroxytryptophan, with clinical, neurologic, psychiatric, imaging and laboratory assessment and chemical analysis of the patients' own supplement samples. Not a treatment trial.

Target Phenotypes: Myalgia HP:0003326 Increased total eosinophil count HP:0001880

Show evidence (1 reference)

clinicaltrials:NCT00001918 SUPPORT Other

"This study is designed to learn more about EMS that develops in patients taking L-5-hydroxytryptophan."

The study's own statement of purpose. Graded OTHER because a registration record is a description of a study rather than evidence from one.

🐁

Animal Models

2

EBT-treated C57BL/6 mouse

Daily intraperitoneal EBT in female C57BL/6 mice produces inflammation and fibrosis of dermis and subcutis including fascia and perimysium, with increased and apparently degranulating mast cells, and with type I, III and VI collagen gene expression rising alongside dermal TGF-beta 1. It is the closest any animal comes to the human fibrosis.

Species

Mouse

Genotype

wild type

Publication

PMID:8163652

Pathograph links

Fascial and Dermal Fibrosis partially recapitulates fidelity: moderate observes: tissue Reproduces the tissue lesion in the right compartments from the candidate compound alone.

Limitations: Intraperitoneal rather than oral, which is not the route the disease took, and pure EBT rather than the contaminant mixture people ate. The mice do not develop the blood eosinophilia that defines the human syndrome, so the model reproduces the fibrosis without the disease. A review of every animal study in this disease concluded the models generally could not be reproduced.

Show evidence (2 references)

PMID:8163652 SUPPORT PRIMARY RESULT Model Organism

"We report the development of inflammation and fibrosis affecting the dermis and subcutis, including the fascia and perimyseal tissues, after the daily intraperitoneal administration of EBT to female C57BL/6 mice."

The measured lesion and its tissue distribution, which is what makes this link to the fibrosis node rather than to the whole disease.

+ 1 more reference

Fibroblast Activation and Type I Collagen Overproduction recapitulates fidelity: moderate observes: molecular The collagen programme and its TGF-beta 1 association are reproduced in mouse skin, which is the human in vitro finding shown in vivo.

Limitations: Expression measured in whole dermis and subcutis rather than in isolated fibroblasts, so the cell of origin is inferred from location.

Show evidence (1 reference)

PMID:8863345 SUPPORT PRIMARY RESULT Model Organism

"Increased procollagen gene expression was accompanied by evidence of enhanced TGF-beta 1 expression in the dermis and subcutis."

Puts the collagen and TGF-beta 1 findings together in the same tissue, which is what the human node asserts.

Show evidence (1 reference)

PMID:8163652 SUPPORT PRIMARY RESULT Model Organism

"Such changes are accompanied by increased numbers of mast cells, many of which appear to be degranulating."

The mast cell finding, and the degranulation, which the mouse shows more clearly than any human specimen. Its human counterpart is now curated on `Inflammatory Infiltration of Fascia, Perimysium and Dermis` from `PMID:2273104`, which reports numerous mast cells in some patients. An earlier draft of this explanation said the finding had no human counterpart in the entry, which was true of the entry and not of the literature.

EBT-treated Lewis rat

Lewis rats given case-associated L-tryptophan or synthetic EBT develop myofascial thickening, and a separate study found perimysial and fascial thickening with lymphocytes, macrophages and sparse eosinophils plus sparse perineurial infiltrate. Control L-tryptophan produced a mild but significant thickening of its own.

Species

Rat

Genotype

wild type

Publication

PMID:8450062

Pathograph links

Inflammatory Infiltration of Fascia, Perimysium and Dermis partially recapitulates fidelity: low observes: tissue Reproduces the fascial and perimysial infiltrate, including the sparse perineurial component, from the candidate compound.

Limitations: No rash and no weakness in either group, so the clinical syndrome is absent. Control L-tryptophan also thickened the myofascia significantly against vehicle, which weakens the attribution to the contaminant, and the authors say their own results do not exclude other impurities. The second study found the changes in two of three treated rats.

Show evidence (2 references)

PMID:7991132 SUPPORT PRIMARY RESULT Model Organism

"In two rats, perimysium and fascia were abnormally thickened and infiltrated with lymphocytes, macrophages, and sparse eosinophils; two rats had sparse perineurial inflammatory cells."

The infiltrate and its compartments, matching the human node this link targets, with the animal counts stated.

+ 1 more reference

Show evidence (2 references)

PMID:8450062 SUPPORT PRIMARY RESULT Model Organism

"All animals treated for 6 wk with case-associated L-TRP or EBT developed significant myofascial thickening, compared with animals in the vehicle control and control L-TRP groups."

The primary result, with its comparison groups.

PMID:8450062 SUPPORT INDIRECT PRIMARY RESULT Model Organism

"they do not rule out the possibility that other impurities in the EMS-case-associated L-TRP may also contribute to some of the features of EMS"

INDIRECT, and the authors' own limit on what their positive result licenses. It is the reason the exposure node names contaminants in the plural.

{ }

Source YAML

click to show

name: Eosinophilia-Myalgia Syndrome
creation_date: "2026-09-25T12:00:00Z"
category: Environmental
categories:
- Toxic Exposure Disorder
- Dietary Supplement Contamination Disorder
- Acquired Sclerodermiform Syndrome
synonyms:
- eosinophilia-myalgia syndrome
- EMS
- L-tryptophan-associated eosinophilia-myalgia syndrome
- tryptophan-associated eosinophilia-myalgia syndrome
description: >-
  Eosinophilia-myalgia syndrome is an acquired multisystem inflammatory and
  fibrosing disease caused by eating manufactured L-tryptophan that carried
  trace process impurities. It appeared as a point-source epidemic in the
  United States in the second half of 1989, and 1531 cases with 27 deaths had
  been reported to national surveillance by July 1990. New cases fell away
  after the product was recalled that November.

  The presenting illness is incapacitating generalised myalgia with a blood
  eosinophil count above 1.0 x 10^9 cells per litre, and those two features are
  the surveillance case definition rather than two findings among many.
  Arthralgia, rash, cough or dyspnoea, peripheral oedema, a raised aldolase and
  abnormal liver tests fill out the acute picture. Eosinophil granule proteins
  are deposited in tissue and recoverable from serum and urine, which is the
  evidence that the eosinophil is doing something rather than merely being
  counted.

  The chronic disease is a fibrosis. Biopsies show inflammatory infiltration of
  the septa, fascia and perimysium, dermal thickening with homogenised collagen
  indistinguishable from scleroderma, and cultured lesional fibroblasts that
  overproduce type I collagen and keep doing so years later. Sclerodermatous
  skin thickening, sensorimotor polyneuropathy, proximal myopathy and episodic
  myalgia are the sequelae that disable. Mortality concentrates in the first
  eighteen months and runs through progressive polyneuropathy and myopathy
  rather than through the fibrosis.

  The etiologic agent has never been settled. Six compounds were associated
  with case lots, and 1,1'-ethylidenebis[tryptophan], first named peak E, is
  the one with the most experimental work behind it. It activates fibroblasts
  and produces fascial fibrosis in mice and rats. It also failed to reach
  significance in the one study that stratified lots by date of manufacture,
  and no animal reproduces the human syndrome.
disease_term:
  preferred_term: eosinophilia-myalgia syndrome
  term:
    id: MONDO:0004941
    label: eosinophilia-myalgia syndrome
notes: >-
  **The etiologic agent is not established and this entry does not assert one.**
  The pathograph names the L-tryptophan process contaminants collectively and
  carries EBT as the worked candidate, because EBT is the compound with in
  vitro, murine and rat data behind it. `PMID:8356958` is curated as REFUTE
  against EBT as the sole agent: once case and non-case lots are stratified by
  time of manufacture the EBT association loses significance, and the authors
  raise a distinct compound as a live possibility. Both readings are in the
  entry, and the `unidentified_etiologic_agent` discussion states what is
  missing.

  **Myalgia and peripheral eosinophilia deliberately carry no `frequency`.**
  Both are reported at 100% in every series, and that figure is circular: the
  CDC surveillance definition requires them, so a cohort assembled under it
  cannot report anything else. `PMID:8895176` records that the definition was
  never validated and was routinely used for purposes it was not built for.
  Every other phenotype's frequency comes from a named cohort and says which.

  **The severity of the myalgia is not explained by the muscle pathology, and
  no edge asserts that it is.** `PMID:1563745` found minimal myofiber atrophy,
  regeneration or necrosis in 11 cases despite severe myalgia in almost all of
  them, and `PMID:8100551` found only rare degranulating eosinophils in fascia
  and perimysium that were nonetheless full of CD8+ cells and macrophages. The
  `myalgia_without_myofiber_injury` discussion carries the gap. This is the same
  gap the `Toxic_Oil_Syndrome` entry records for its own myalgia, and the two
  diseases are routinely discussed together for that reason.

  **Relation to toxic oil syndrome.** The two are separate epidemics with
  separate vehicles, and they are linked by chemistry rather than by exposure:
  PAP from the Spanish oil and PAA from the implicated L-tryptophan converge on
  a shared metabolite. That link is curated here as its own pathophysiology
  node with its own evidence (`PMID:8555405`, `PMID:17892268`). No toxic oil
  syndrome clinical finding is imported as an EMS finding.

  **A related illness followed L-5-hydroxytryptophan**, and it is not curated
  as EMS. `PMID:7699627` reports one family where one member met criteria for
  EMS and two had eosinophilia, with an impurity present in their 5-HTP and
  absent from comparison samples, and the NIH study `NCT00001918` was built
  around such cases. The `five_htp_related_illness` discussion records it; the
  disease entry stays scoped to the L-tryptophan epidemic.

  **Most phenotypes here are deliberately unwired, and connectivity is
  correspondingly low.** Run `just list-disconnected-phenotypes
  kb/disorders/Eosinophilia-Myalgia_Syndrome.yaml` for the current set rather
  than trusting a count written here. The rule is the one the rest of this entry
  follows: an edge goes in only where a cited source makes the causal link. The
  constitutional features (fever, fatigue, oral ulcer) have no named lesion in
  any cited source; the raised aldolase and transaminases are readouts nothing
  here explains, and are attached observationally through `reports_on` rather
  than causally; `Proximal myopathy` and `Myalgia` sit on the muscle-pathology
  gap the `myalgia_without_myofiber_injury` discussion states; the cognitive
  findings have two competing readings in the one paper that imaged them; and
  `Xerostomia` and `Dysphagia` are self-reported survey figures with no cited
  study of salivary flow or swallowing. An edge added to raise that figure would
  be worse than the gap.

  **The pulmonary phenotypes are wired, with one exception that is not.**
  `Pulmonary Eosinophilic and Lymphocytic Inflammation` was added in review after
  the lung turned out to be the one organ system with phenotypes and no node,
  which was an omission rather than a scoping decision: 81% of the deaths
  reported to surveillance had pulmonary complications. `Dyspnea` and `Decreased
  DLCO` hang off it. `Pulmonary arterial hypertension` does not, and that one is
  deliberate: it rests on a single patient in a six-patient series, and no cited
  source connects the inflammation to vascular remodelling.

  **Three features the deep-research report reports are not curated, for one
  reason.** Weight loss at 50%, and the Maryland series' own frequencies, come
  from full-text pages that no cached reference here holds, so there is no
  quotable sentence for them. Paresthesia and xerostomia were in the same
  position until a further PubMed pass found cohort abstracts that state them,
  and both are now curated. Weight loss is still uncited and is therefore still
  absent, rather than carried on a number with no source behind it.

  **There is no `histopathology:` or `diagnosis:` section, deliberately.** The
  histology of this disease is dense and well reported, and it is carried as
  pathophysiology nodes because in EMS the histology *is* the mechanism: the
  septal and fascial infiltrate, the homogenised dermal collagen, the
  conduction-system fibrosis. Splitting it into a parallel descriptive section
  would state each finding twice and put the causal reading in only one of them.
  The case definition, which is the part of the diagnostic picture that carries
  real weight here, is curated as a `definitions:` entry instead, and the
  `case_definition_validity` discussion argues its consequences.

  No GeneReviews chapter exists and none is expected. This is an acquired
  point-source intoxication with no Mendelian basis; `just check-genereviews`
  returns NO_CHAPTER for both Bookshelf collections.
mechanistic_hypotheses:
- hypothesis_group_id: contaminant_driven_type2_immunity
  hypothesis_label: A contaminant drives a type 2 immune response that recruits and activates eosinophils
  status: CANONICAL
  description: >-
    On this account the impurity acts as an immunogen or adjuvant. Susceptible
    hosts mount a type 2 cytokine response, eosinophils are expanded, recruited
    and degranulated in tissue, and the fibrosis follows the inflammation. The
    strongest support is the cytokine work on peak E and the eosinophil granule
    proteins recoverable from patients.
  evidence:
  - reference: PMID:1727618
    reference_title: The cause and pathogenesis of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Eosinophil activation and the release of major basic protein and other eosinophil-derived toxic proteins into the extracellular space is a striking feature in the eosinophilia-myalgia syndrome and implicates eosinophils or their products in the pathogenesis."
    explanation: >-
      States the eosinophil-centred reading of the pathogenesis that this
      hypothesis group asserts.
  - reference: PMID:11703359
    reference_title: L-tryptophan contaminant 'peak E' induces the release of IL-5 and IL-10 by peripheral blood mononuclear cells from patients with functional somatic syndromes.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "six of the seven FSS patients reacting with peak E produced IL-5 and/or IL-10"
    explanation: >-
      INDIRECT: the type 2 cytokine response to peak E is measured, and it is
      measured in mononuclear cells from patients with functional somatic
      syndromes rather than from anyone with EMS.
- hypothesis_group_id: direct_fibroblast_activation
  hypothesis_label: The contaminant acts directly on fibroblasts
  status: ALTERNATIVE
  description: >-
    On this account the fibrosis does not require the immune response to reach
    it. Peak E stimulates dermal fibroblast DNA synthesis, procollagen
    transcription and gel contraction on its own, and L-tryptophan itself and
    the other major contaminant do not. The two hypotheses are not exclusive and
    the pathograph carries both edges.
  evidence:
  - reference: PMID:7593596
    reference_title: "Enhanced collagen synthesis and transcription by peak E, a contaminant of L-tryptophan preparations associated with the eosinophilia myalgia syndrome epidemic."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "We now report that peak E, a dimer of L-tryptophan, is a potent stimulus for human dermal fibroblast DNA and collagen synthesis."
    explanation: The direct fibroblast effect this hypothesis is named for.
  - reference: PMID:7593596
    reference_title: "Enhanced collagen synthesis and transcription by peak E, a contaminant of L-tryptophan preparations associated with the eosinophilia myalgia syndrome epidemic."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "No increase in procollagen mRNA levels was found after the addition of another major L-tryptophan contaminant, peak 5, or with L-tryptophan itself."
    explanation: >-
      The specificity control, and the reason the effect is attributed to peak E
      rather than to L-tryptophan or to contamination in general.
- hypothesis_group_id: t_cell_matrix_autoimmunity
  hypothesis_label: A T cell response against extracellular matrix sustains the chronic disease
  status: ALTERNATIVE
  description: >-
    On this account the chronic phase is not residual damage but an ongoing
    cell-mediated process. Fascia and perimysium contain CD8+ cells and
    macrophages with almost no intact eosinophils, most fibroblasts are
    activated and a third aberrantly express HLA-DR, and a repeat biopsy a year
    later showed less inflammation and more activated fibroblasts.
  evidence:
  - reference: PMID:8100551
    reference_title: Immune-mediated mechanisms and immune activation of fibroblasts in the pathogenesis of eosinophilia-myalgia syndrome induced by L-tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We conclude that in EMS there is a T-cell-mediated process against components of the extracellular matrix, including fibroblasts, in the fascia and the perimysium that persists even years after the drug is discontinued."
    explanation: The authors' own statement of this hypothesis, from their own immunocytochemistry.
pathophysiology:
- name: Ingestion of Contaminated Manufactured L-Tryptophan
  biological_scale: ORGANISM
  description: >-
    Oral L-tryptophan sold over the counter as a sleep and mood supplement. Case
    exposure traced to a single manufacturer whose fermentation process had
    changed, and the epidemic ended when the product was recalled.
  chemical_entities:
  - preferred_term: L-tryptophan
    term:
      id: CHEBI:16828
      label: L-tryptophan
  downstream:
  - target: Systemic Exposure to L-Tryptophan Process Contaminants
    causal_link_type: DIRECT
    description: >-
      The manufacturing conditions, not the amino acid, are what the
      case-control study associated with illness.
    evidence:
    - reference: PMID:2370887
      reference_title: An investigation of the cause of the eosinophilia-myalgia syndrome associated with tryptophan use.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "The outbreak of the eosinophilia-myalgia syndrome in 1989 resulted from the ingestion of a chemical constituent that was associated with specific tryptophan-manufacturing conditions at one company."
      explanation: >-
        Supports this specific step - that what was ingested carried a
        constituent tied to how it was made - rather than either node alone.
  evidence:
  - reference: PMID:2370887
    reference_title: An investigation of the cause of the eosinophilia-myalgia syndrome associated with tryptophan use.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "29 of 30 case patients (97 percent) and 21 of 35 controls (60 percent) had consumed tryptophan manufactured by a single company (odds ratio, 19.3; 95 percent confidence interval, 2.5 to 844.9; P less than 0.001)"
    explanation: >-
      The single-manufacturer association with its own odds ratio and interval,
      which is why this node names manufactured product rather than tryptophan.
  - reference: PMID:2370887
    reference_title: An investigation of the cause of the eosinophilia-myalgia syndrome associated with tryptophan use.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "This company used a fermentation process involving Bacillus amyloliquefaciens to manufacture tryptophan."
    explanation: Names the production route behind the implicated lots.
  - reference: PMID:8895184
    reference_title: Tryptophan produced by Showa Denko and epidemic eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "There was a dose-response effect, with risk of illness increasing as a function of the amount of tryptophan consumed."
    explanation: >-
      The dose-response element of the causal argument, which an association
      with a manufacturer alone would not supply.
- name: Systemic Exposure to L-Tryptophan Process Contaminants
  biological_scale: ORGANISM
  description: >-
    Case lots carried more than one trace impurity. EBT, first reported as an
    HPLC peak and then characterised as a tryptophan dimer, is the compound with
    the most experimental work behind it and is carried here as the worked
    candidate rather than as the established agent.
  chemical_entities:
  - preferred_term: 1,1'-ethylidenebis[tryptophan]
    term:
      id: CHEBI:172675
      label: 1,1'-Ethylidenebistryptophan
  biological_processes:
  - preferred_term: response to xenobiotic stimulus
    term:
      id: GO:0009410
      label: response to xenobiotic stimulus
  downstream:
  - target: Type 2 Cytokine Response to Peak E
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - contaminant_driven_type2_immunity
    description: >-
      How peak E reaches a cytokine-producing cell is not described by any cited
      source. What is measured is the cytokine output of mononuclear cells
      exposed to it, in subjects who did not have EMS.
    evidence:
    - reference: PMID:11703359
      reference_title: L-tryptophan contaminant 'peak E' induces the release of IL-5 and IL-10 by peripheral blood mononuclear cells from patients with functional somatic syndromes.
      supports: SUPPORT
      directness: INDIRECT
      evidence_source: IN_VITRO
      quote_role: PRIMARY_RESULT
      snippet: "PBMC from seven of the 12 FSS patients, but only three of the 24 controls, produced cytokines after incubation with peak E (P < 0.05)."
      explanation: >-
        Supports the edge from contaminant exposure to a cytokine response, and
        INDIRECT because the responding cells came from patients with functional
        somatic syndromes rather than from EMS patients.
  - target: Eosinophil Chemotaxis and Endothelial Adhesion
    causal_link_type: DIRECT
    hypothesis_groups:
    - contaminant_driven_type2_immunity
    evidence:
    - reference: PMID:8906111
      reference_title: "Effect of L-tryptophan products on function of human eosinophils: investigation of the causal mechanisms of eosinophilia myalgia syndrome associated with L-tryptophan products."
      supports: SUPPORT
      evidence_source: IN_VITRO
      quote_role: PRIMARY_RESULT
      snippet: "Contaminants in the L-tryptophan products, known as peak-E and peak-5, at a concentration of 1-10 micrograms/ml had the ability to elicit chemokinetic migration of eosinophils."
      explanation: >-
        The contaminants act on eosinophil migration directly in culture, which
        is exactly this edge.
  - target: Fibroblast Activation and Type I Collagen Overproduction
    causal_link_type: DIRECT
    hypothesis_groups:
    - direct_fibroblast_activation
    description: >-
      The edge that makes the alternative hypothesis a separate route rather
      than a restatement: peak E raises procollagen transcription in dermal
      fibroblasts with no immune cell in the dish.
    evidence:
    - reference: PMID:7593596
      reference_title: "Enhanced collagen synthesis and transcription by peak E, a contaminant of L-tryptophan preparations associated with the eosinophilia myalgia syndrome epidemic."
      supports: SUPPORT
      evidence_source: IN_VITRO
      quote_role: PRIMARY_RESULT
      snippet: "peak E (0.5 to 100 microM) caused a progressive, more than threefold increase in alpha 1(I) procollagen mRNA levels and collagenous protein"
      explanation: >-
        The measured transcriptional and protein response, in a monolayer
        culture, which is the direct link this edge asserts.
  - target: T Cell Mediated Immunity Against Extracellular Matrix
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - t_cell_matrix_autoimmunity
    description: >-
      The contaminant reaches the inflammatory cells as well as the eosinophils.
      The review that states this names no intermediate step, so the link type
      records none.
    evidence:
    - reference: PMID:1727618
      reference_title: The cause and pathogenesis of the eosinophilia-myalgia syndrome.
      supports: SUPPORT
      directness: INDIRECT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Peak E or other, as yet unidentified, contaminants may trigger activation of eosinophils and inflammatory cells and increase biosynthesis of connective tissue components"
      explanation: >-
        The one cited sentence that runs from the contaminant to the
        inflammatory cells rather than to the eosinophil alone, which is what
        this edge asserts. INDIRECT because "may trigger" is the review's own
        hedge.
  evidence:
  - reference: PMID:2270484
    reference_title: "Characterization of \"peak E,\" a novel amino acid associated with eosinophilia-myalgia syndrome."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: PRIMARY_RESULT
    snippet: "Spectral and chemical studies now demonstrate that peak E is 1,1'-ethylidenebis[tryptophan]."
    explanation: >-
      Establishes the chemical identity of the compound this node binds. Graded
      OTHER because it is an analytical chemistry result on product samples
      rather than a study of people, animals or cells.
  - reference: PMID:2270484
    reference_title: "Characterization of \"peak E,\" a novel amino acid associated with eosinophilia-myalgia syndrome."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: PRIMARY_RESULT
    snippet: "This novel amino acid may be the etiological agent responsible for EMS, or it may be a marker of a still unidentified causal agent."
    explanation: >-
      The authors' own hedge in the paper that named the compound, and the
      reason this node is titled for the contaminants collectively.
  - reference: PMID:8356958
    reference_title: "Tryptophan contaminants associated with eosinophilia-myalgia syndrome. The Eosinophilia-Myalgia Studies of Oregon, New York and New Mexico."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "higher EBT levels were still associated with a lot's case status, but the association lacked statistical significance (p = 0.120, odds ratio = 1.56, 95% confidence interval 0.758-3.23)"
    explanation: >-
      REFUTE against EBT alone as the agent. Stratifying lots by date of
      manufacture removes the significance of the EBT association, which is the
      single strongest reason this node does not name one compound.
  - reference: PMID:8356958
    reference_title: "Tryptophan contaminants associated with eosinophilia-myalgia syndrome. The Eosinophilia-Myalgia Studies of Oregon, New York and New Mexico."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "they raise the possibility that other chemical contaminants in manufactured tryptophan modify the effects of EBT or that the causal agent of EMS is an entirely distinct compound"
    explanation: >-
      The two live alternatives the same authors name, both of which this node
      leaves open.
- name: Hepatic Conversion of PAP to 3-(Phenylamino)alanine
  biological_scale: MOLECULAR
  description: >-
    PAA was isolated from L-tryptophan associated with EMS, and liver tissue
    converts the toxic oil syndrome marker PAP into it. Bioactivation of PAA by
    human liver microsomes then yields 4-aminophenol through a quinoneimine,
    which is the same metabolite PAP yields. This node is the chemical bridge
    between the two epidemics.
  biological_processes:
  - preferred_term: response to xenobiotic stimulus
    term:
      id: GO:0009410
      label: response to xenobiotic stimulus
  evidence:
  - reference: PMID:8555405
    reference_title: "Biotransformation of 3-(phenylamino)-1,2-propanediol to 3-(phenylamino)alanine: a chemical link between toxic oil syndrome and eosinophilia-myalgia syndrome."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "A related aniline derivative, 3-(phenylamino)-L-alanine (PAA), was recently isolated from L-tryptophan associated with the onset of EMS."
    explanation: >-
      Establishes that PAA was present in the implicated product, which is what
      puts this node in an EMS entry at all.
  - reference: PMID:8555405
    reference_title: "Biotransformation of 3-(phenylamino)-1,2-propanediol to 3-(phenylamino)alanine: a chemical link between toxic oil syndrome and eosinophilia-myalgia syndrome."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "Here, we demonstrate the biotransformation of PAP into PAA by both rat hepatocytes and human liver tissue."
    explanation: The measured conversion, in rat hepatocytes and human liver tissue.
  - reference: PMID:17892268
    reference_title: "Generation of quinoneimine intermediates in the bioactivation of 3-(N-phenylamino)alanine (PAA) by human liver microsomes: a potential link between eosinophilia-myalgia syndrome and toxic oil syndrome."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "These findings establish that EMS and TOS are linked by a common toxic metabolite (4-aminophenol)"
    explanation: >-
      Extends the link from a shared compound to a shared reactive metabolite,
      which is the claim this node makes.
  notes: >-
    This node is deliberately isolated, with no edge in and no edge out. Nothing
    in the cited literature traces a route from 4-aminophenol, or from PAA, to
    any lesion in either disease, so there is no downstream edge. An earlier
    draft drew an incoming edge from the exposure node typed UNKNOWN, whose own
    description then disclaimed the causation the arrow asserts: in EMS the PAA
    was already present in the product rather than made from PAP, so the
    conversion is the toxic oil syndrome route to the same molecule and not a
    step in this disease. That edge is removed. The chemical bridge is stated
    here and in the `unidentified_etiologic_agent` discussion instead.
- name: Type 2 Cytokine Response to Peak E
  biological_scale: MOLECULAR
  description: >-
    Mononuclear cells exposed to peak E release IL-5 and IL-10. IL-5 is the
    eosinophil growth and survival cytokine, and reviews of the syndrome name it
    and TGF-beta as the important mediators.
  biological_processes:
  - preferred_term: interleukin-5 production
    modifier: INCREASED
    term:
      id: GO:0032634
      label: interleukin-5 production
  downstream:
  - target: Eosinophil Expansion with Tissue Degranulation
    causal_link_type: DIRECT
    hypothesis_groups:
    - contaminant_driven_type2_immunity
  evidence:
  - reference: PMID:8423409
    reference_title: "L-tryptophan and the eosinophilia-myalgia syndrome: current understanding of the etiology and pathogenesis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Pathologic observations and experimental studies indicate that eosinophils, mononuclear inflammatory cells, and fibroblasts are potential effector cells, and interleukin-5 and transforming growth factor-beta are important mediators in the pathogenesis of the syndrome."
    explanation: >-
      Names IL-5 as a mediator of the syndrome, which is the cytokine this node
      is about. The word "potential" is the authors' own and is why the node's
      outgoing edge carries the hypothesis group.
  - reference: PMID:7797795
    reference_title: "Eosinophil-active cytokine from mononuclear cells cultured with L-tryptophan products: an unexpected consequence of endotoxin contamination."
    supports: REFUTE
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "this response is caused by endotoxin contamination of the L-tryptophan products and not by a specific L-tryptophan contaminant"
    explanation: >-
      REFUTE, and the most useful negative result in the entry. An earlier
      GM-CSF response to implicated L-tryptophan turned out to be driven by
      endotoxin in the product, so a cytokine response measured in this assay
      system is not by itself evidence about the EMS contaminant.
  notes: >-
    One process is bound, and the history is worth recording because the first
    two attempts were both wrong in ways no validator could see. The first bound
    `GO:0032604` granulocyte macrophage colony-stimulating factor production
    under a `preferred_term` of interleukin-5 production, and defended the
    mismatch on the ground that the node had to carry the GM-CSF finding for the
    REFUTE item to attach to. That reasoning was wrong twice: an evidence item
    attaches to the node rather than to a process descriptor, and
    `biological_processes` is multivalued. The second kept `GO:0032604` as a
    second descriptor beside the correct `GO:0032634`, which fixed the mismatch
    and left a process the node is not about: GM-CSF is not a type 2 cytokine,
    and the paper that measured it attributed the response to endotoxin rather
    than to peak E. So it is scoped out by both halves of this node's name and is
    now dropped. The REFUTE evidence item stays on the node, where it always
    attached, which is the whole point.
- name: Eosinophil Chemotaxis and Endothelial Adhesion
  biological_scale: CELLULAR
  description: >-
    The contaminants make eosinophils migrate and make them stick to
    endothelium. Adhesion to stimulated human umbilical vein endothelial cells
    is blocked by antibody to ICAM-1 and not by antibody to VCAM-1, and the
    eosinophils' own integrin expression does not change, so the effect is on
    the endothelial side.
  cell_types:
  - preferred_term: eosinophil
    term:
      id: CL:0000771
      label: eosinophil
  - preferred_term: endothelial cell
    term:
      id: CL:0000115
      label: endothelial cell
  biological_processes:
  - preferred_term: eosinophil migration
    modifier: INCREASED
    term:
      id: GO:0072677
      label: eosinophil migration
  downstream:
  - target: Eosinophil Expansion with Tissue Degranulation
    causal_link_type: DIRECT
    hypothesis_groups:
    - contaminant_driven_type2_immunity
  evidence:
  - reference: PMID:8906111
    reference_title: "Effect of L-tryptophan products on function of human eosinophils: investigation of the causal mechanisms of eosinophilia myalgia syndrome associated with L-tryptophan products."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "Purified eosinophils adhered to peak-E- or peak-5-stimulated human umbilical vein endothelial cells, and this adherence was inhibited by the presence of antibody to intercellular adhesion molecule-1"
    explanation: The adhesion step and the adhesion molecule it runs through.
  - reference: PMID:8906111
    reference_title: "Effect of L-tryptophan products on function of human eosinophils: investigation of the causal mechanisms of eosinophilia myalgia syndrome associated with L-tryptophan products."
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "Human peripheral blood mononuclear cells (PBMCs) produced eosinophil survival-enhancing activity when cultivated with peak-E, but not with medium alone, peak-5 or control tryptophan."
    explanation: >-
      Survival enhancement specific to peak E among the conditions tested, which
      is how a migration effect becomes an accumulation.
- name: Eosinophil Expansion with Tissue Degranulation
  biological_scale: CELLULAR
  description: >-
    Blood eosinophilia with extracellular deposition of eosinophil granule
    proteins in tissue, and raised major basic protein and eosinophil-derived
    neurotoxin in serum and urine. The count is the case definition; the granule
    proteins are the evidence of activity.
  cell_types:
  - preferred_term: eosinophil
    term:
      id: CL:0000771
      label: eosinophil
  biological_processes:
  - preferred_term: eosinophil degranulation
    modifier: INCREASED
    term:
      id: GO:0043308
      label: eosinophil degranulation
  - preferred_term: eosinophil activation
    modifier: INCREASED
    term:
      id: GO:0043307
      label: eosinophil activation
  downstream:
  - target: Inflammatory Infiltration of Fascia, Perimysium and Dermis
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - contaminant_driven_type2_immunity
    description: >-
      Recorded as unknown intermediates because the tissue the eosinophil
      products are deposited in contains almost no eosinophils. Whatever
      connects the two is not described.
  - target: Pulmonary Eosinophilic and Lymphocytic Inflammation
    causal_link_type: DIRECT
    hypothesis_groups:
    - contaminant_driven_type2_immunity
    description: >-
      Unlike the fascia, the lung does show the eosinophil: lavage fluid from
      affected patients carries them in excess. This is the one compartment
      where the cell and the lesion are found in the same place.
    evidence:
    - reference: PMID:1582284
      reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Two patients undergoing BAL exhibited increased eosinophils in the lavage fluid; a third had elevated lymphocytes."
      explanation: >-
        Eosinophils recovered from the lung itself, which is what this edge
        asserts. The third patient's lymphocytosis is why the target node names
        both cell populations.
  - target: Peripheral eosinophilia
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:2314421
    reference_title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "All three patients had elevated serum and urinary levels of this protein and eosinophil-derived neurotoxin, indicative of eosinophil degranulation."
    explanation: >-
      Degranulation measured in the patients themselves, in the paper that first
      reported the association.
  - reference: PMID:2314421
    reference_title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "revealed eosinophilic infiltration, as well as the extracellular deposition of eosinophil-granule major basic protein"
    explanation: >-
      The tissue deposition of granule protein, which is the part of this node
      that a blood count does not establish.
  - reference: PMID:1563745
    reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Minimal tissue eosinophilia was seen despite the extent of blood eosinophilia."
    explanation: >-
      REFUTE against tissue eosinophil accumulation as the lesion. The blood
      count and the tissue count come apart, which is why the downstream edge is
      typed INDIRECT_UNKNOWN_INTERMEDIATES.
- name: T Cell Mediated Immunity Against Extracellular Matrix
  biological_scale: CELLULAR
  description: >-
    Fascia and perimysium carry CD8+ cells, CD4+ cells and macrophages,
    scattered and perivascular, with MHC class I expression on muscle fibres
    near vessels and fascicles. Most fibroblasts are activated and a minority
    express HLA-DR, which the authors read as the fibroblast being the target.
  cell_types:
  - preferred_term: CD8-positive, alpha-beta T cell
    term:
      id: CL:0000625
      label: CD8-positive, alpha-beta T cell
  - preferred_term: macrophage
    term:
      id: CL:0000235
      label: macrophage
  biological_processes:
  - preferred_term: T cell mediated immunity
    modifier: INCREASED
    term:
      id: GO:0002456
      label: T cell mediated immunity
  downstream:
  - target: Inflammatory Infiltration of Fascia, Perimysium and Dermis
    causal_link_type: DIRECT
    hypothesis_groups:
    - t_cell_matrix_autoimmunity
  - target: Pulmonary Eosinophilic and Lymphocytic Inflammation
    causal_link_type: DIRECT
    hypothesis_groups:
    - t_cell_matrix_autoimmunity
    description: >-
      The same CD8+ compartment found in fascia and perimysium is recoverable
      from the lung.
    evidence:
    - reference: PMID:1582284
      reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "In these two patients, increased proportions of T-suppressor/cytolytic (CD8+) cells were observed in the BAL fluid."
      explanation: >-
        Puts the T cell population this node describes inside the lung, which is
        the step this edge asserts. Two patients, so it is an observation rather
        than a series.
  - target: Fibroblast Activation and Type I Collagen Overproduction
    causal_link_type: DIRECT
    hypothesis_groups:
    - t_cell_matrix_autoimmunity
    evidence:
    - reference: PMID:8100551
      reference_title: Immune-mediated mechanisms and immune activation of fibroblasts in the pathogenesis of eosinophilia-myalgia syndrome induced by L-tryptophan.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Up to 70% of the fibroblasts in EMS were activated and up to 30% of them expressed HLA-DR antigen."
      explanation: >-
        Supports this edge specifically: the fibroblasts in the infiltrated
        tissue are activated and aberrantly antigen-presenting, which is how a T
        cell process reaches the collagen programme.
  evidence:
  - reference: PMID:8100551
    reference_title: Immune-mediated mechanisms and immune activation of fibroblasts in the pathogenesis of eosinophilia-myalgia syndrome induced by L-tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We found inflammatory cells consisting of CD8+ cells (45% +/- 8.9%), T4 cells (36% +/- 10.1%), and macrophages (19% +/- 12%), scattered or perivascularly in the fascia, the perimysium, and the endomysial septae."
    explanation: The cell populations and their distribution, counted.
  - reference: PMID:8100551
    reference_title: Immune-mediated mechanisms and immune activation of fibroblasts in the pathogenesis of eosinophilia-myalgia syndrome induced by L-tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Only rare granulated or degranulating eosinophils were noted."
    explanation: >-
      Why this node exists beside the eosinophil node rather than downstream of
      it. In the tissue the eosinophil is almost absent and the T cell is not.
- name: Inflammatory Infiltration of Fascia, Perimysium and Dermis
  biological_scale: TISSUE
  description: >-
    The earliest changes sit in the septa between subcutaneous fat lobules and
    in the deep dermis or fascia, infiltrated with lymphocytes and histiocytes
    alongside reactive mesenchymal cells. Skeletal muscle shows perimysial,
    epimysial and fascial infiltrate. Dermal and fascial vessel walls thicken
    and endothelium swells without overt vasculitis.
  cell_types:
  - preferred_term: macrophage
    term:
      id: CL:0000235
      label: macrophage
  - preferred_term: mast cell
    term:
      id: CL:0000097
      label: mast cell
  biological_processes:
  - preferred_term: inflammatory response
    modifier: INCREASED
    term:
      id: GO:0006954
      label: inflammatory response
  downstream:
  - target: Microangiopathy of Small Vessels
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      A review of the pathophysiology reads the cellular immune response as
      leading to a microangiopathy. The step is asserted at review level and the
      intermediates are not named.
    evidence:
    - reference: PMID:8895179
      reference_title: Pathophysiology of the eosinophilia-myalgia syndrome.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "The cellular immune response seems to lead to a microangiopathy and release of cytokines that can induce eosinophilia and fibrosis."
      explanation: >-
        The one cited source that makes this causal step, and it makes it with
        "seems to", which is why the link type is not DIRECT.
  - target: Fibroblast Activation and Type I Collagen Overproduction
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
  - target: Perineural Inflammation and Axonal Injury
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:1323757
      reference_title: Demyelinating polyneuropathy in eosinophilia-myalgia syndrome.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Pathology revealed patchy perivascular infiltrates and fibrosis in the connective tissue of muscle and nerve."
      explanation: >-
        Puts the infiltrate and the fibrosis in nerve connective tissue in the
        same specimens, which is the tissue step this edge asserts.
  evidence:
  - reference: PMID:1563745
    reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Skeletal muscle biopsies showed a perimysial, epimysial, and/or fascial inflammatory infiltrate of lymphocytes and distinctive reactive mesenchymal cells with some eosinophils."
    explanation: The muscle-compartment distribution of the infiltrate, from 11 biopsied cases.
  - reference: PMID:1563745
    reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Vessel walls in the dermis and fascia showed thickening and endothelial swelling, but no overt vasculitis was noted."
    explanation: >-
      The vascular change, and its limit. This is the reason no vasculitis node
      appears in this entry, which is one of the differences from toxic oil
      syndrome.
  - reference: PMID:8895179
    reference_title: Pathophysiology of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Such responses are most marked within the dermis, subcutis, fascia, and connective tissue in and around muscles, nerves, and other tissues."
    explanation: The tissue distribution this node is named for, at review level.
  - reference: PMID:2273104
    reference_title: "Cutaneous manifestations of the L-tryptophan-associated eosinophilia-myalgia syndrome: a spectrum of sclerodermatous skin disease."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "deposition of glycosaminoglycans in the dermis, and, in some patients, numerous mast cells"
    explanation: >-
      The human counterpart of the mast cell accumulation the murine model
      reports, from prospective histology. It is why `CL:0000097` is bound here.
- name: Pulmonary Eosinophilic and Lymphocytic Inflammation
  biological_scale: TISSUE
  description: >-
    Lavage fluid from affected lungs carries excess eosinophils in some patients
    and excess lymphocytes in others, with a raised CD8+ proportion, and it also
    carries activity that stimulates fibroblast proliferation. That activity
    falls to normal on corticosteroid treatment. High-resolution CT is abnormal
    in most patients examined, at times with a normal plain radiograph. It is the
    same three-cell story the entry models in fascia and dermis, in the
    compartment where most of the reported deaths had complications.
  cell_types:
  - preferred_term: eosinophil
    term:
      id: CL:0000771
      label: eosinophil
  - preferred_term: CD8-positive, alpha-beta T cell
    term:
      id: CL:0000625
      label: CD8-positive, alpha-beta T cell
  - preferred_term: fibroblast
    term:
      id: CL:0000057
      label: fibroblast
  biological_processes:
  - preferred_term: inflammatory response
    modifier: INCREASED
    term:
      id: GO:0006954
      label: inflammatory response
  - preferred_term: fibroblast proliferation
    modifier: INCREASED
    term:
      id: GO:0048144
      label: fibroblast proliferation
  downstream:
  - target: Dyspnea
    causal_link_type: DIRECT
  - target: Decreased DLCO
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:1582284
      reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
      supports: SUPPORT
      directness: INDIRECT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Exercise testing with arterial blood gas sampling in three patients was consistent with pulmonary vascular or parenchymal disease."
      explanation: >-
        INDIRECT, and the closest any cited source comes to tying the gas
        transfer defect to the lung lesion: physiology consistent with vascular
        or parenchymal disease in the same patients whose lavage was abnormal.
        No source states the causal step outright.
  evidence:
  - reference: PMID:1582284
    reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Serial measurements of fibroblast proliferation-stimulating-activity in samples of BAL fluid obtained from serial examinations in two patients exhibited heightened pretreatment activity that returned to the normal range following corticosteroid therapy."
    explanation: >-
      The fibroblast-activating arm of this node, measured in lung fluid, and the
      one measurement anywhere in the entry that falls back to normal under
      treatment. It is why `GO:0048144` is bound here as well as on the dermal
      fibroblast node.
  - reference: PMID:1582284
    reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "In four patients, HRCT scanning of the chest was abnormal."
    explanation: >-
      Imaging evidence that the lesion is in the lung parenchyma in most of the
      six patients examined.
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Of the 36 patients who died, 33 (92%) had neuromuscular sequelae, 29 (81%) had pulmonary complications, and 23 (64%) had cardiac manifestations."
    explanation: >-
      INDIRECT, and carried for scale rather than for mechanism: pulmonary
      complications in 81% of the deaths reported to surveillance is why this
      node exists at all. The sentence counts complications and does not describe
      the lesion.
  notes: >-
    The whole node rests on one six-patient series selected for dyspnoea, so
    every figure in it is a count within that series rather than a disease
    frequency. `Pulmonary arterial hypertension` is deliberately not wired to
    this node: one patient in that series had it, and no cited source connects
    the inflammation to the vascular remodelling. Toxic oil syndrome's pulmonary
    hypertension is far better characterised, and it is not imported here.
- name: Microangiopathy of Small Vessels
  biological_scale: TISSUE
  description: >-
    Small-vessel change is described in the pathophysiology review as the
    consequence of the cellular immune response, and postmortem hearts show
    focal fibromuscular dysplasia narrowing arteries alongside endarteritis and
    panarteritis. Whether the same lesion accounts for the chronic ischaemic
    symptoms is unsettled.
  cell_types:
  - preferred_term: endothelial cell
    term:
      id: CL:0000115
      label: endothelial cell
  downstream:
  - target: Cardiac Arterial and Conduction System Fibrosis
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:2058857
      reference_title: Postmortem studies of the heart in three fatal cases of the eosinophilia-myalgia syndrome.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Arterial abnormalities were numerous and primarily of two types: focal fibromuscular dysplasia causing moderate to severe narrowing, as well as endarteritis and panarteritis."
      explanation: >-
        The arterial lesion, in the hearts of three patients who died of the
        disease, which is the step from small-vessel disease to cardiac
        pathology.
  evidence:
  - reference: PMID:8895179
    reference_title: Pathophysiology of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "The pathophysiology of the chronic symptoms is poorly understood but may involve ischemia, neuropathy, and metabolic abnormalities."
    explanation: >-
      INDIRECT and quoted for what it concedes. Ischaemia is named as a possible
      contributor to the chronic symptoms, and the same sentence says the
      chronic pathophysiology is poorly understood.
- name: Fibroblast Activation and Type I Collagen Overproduction
  biological_scale: CELLULAR
  description: >-
    Fibroblasts cultured from affected skin produce more collagen and carry
    higher alpha 1(I) procollagen mRNA than matched normal cells, and the COL1A1
    promoter drives more reporter activity in them. The phenotype is
    transcriptional and it persists into the chronic stage, years after the
    supplement was stopped. TGF-beta 1 transcripts are abundant in fibroblasts
    of affected fascia and absent from fibroblasts in the adjacent dermis of the
    same specimen.
  cell_types:
  - preferred_term: fibroblast
    term:
      id: CL:0000057
      label: fibroblast
  biological_processes:
  - preferred_term: collagen biosynthetic process
    modifier: INCREASED
    term:
      id: GO:0032964
      label: collagen biosynthetic process
  - preferred_term: transforming growth factor beta receptor signaling pathway
    modifier: INCREASED
    term:
      id: GO:0007179
      label: transforming growth factor beta receptor signaling pathway
  - preferred_term: fibroblast proliferation
    modifier: INCREASED
    term:
      id: GO:0048144
      label: fibroblast proliferation
  downstream:
  - target: Fascial and Dermal Fibrosis
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:8230009
    reference_title: Increased type I collagen gene expression in L-tryptophan associated eosinophilia-myalgia syndrome skin fibroblasts.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "Cell lines derived from the affected skin from patients with EMS exhibited greater collagen production and higher steady state levels of alpha 1(I) procollagen mRNA compared with fibroblasts from age and sex matched healthy individuals."
    explanation: The overproduction phenotype against matched controls.
  - reference: PMID:9076948
    reference_title: Increased activity of the alpha 1(I) procollagen promoter in skin fibroblasts from patients with chronic eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "The study shows increased CAT activity driven by a 174 bp fragment of COL1A1 in transiently transfected skin fibroblasts from patients with EMS even in the chronic stage of disease."
    explanation: >-
      Locates the effect in the COL1A1 promoter and shows it persists into
      chronic disease, which is what makes this node an ongoing state rather
      than a memory of the acute illness.
  - reference: PMID:1702819
    reference_title: Elevated expression of the genes for transforming growth factor-beta 1 and type VI collagen in diffuse fasciitis associated with the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "In situ hybridizations of affected fascia with a human sequence-specific TGF-beta 1 cDNA demonstrated numerous fibroblasts displaying positive hybridization signals indicative of high levels of transcripts for this cytokine."
    explanation: >-
      The TGF-beta 1 binding on this node, measured in situ in affected fascia
      rather than inferred from the fibrosis.
  - reference: PMID:1702819
    reference_title: Elevated expression of the genes for transforming growth factor-beta 1 and type VI collagen in diffuse fasciitis associated with the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "These findings suggest that TGF-beta 1 may play an important role in the development of the connective tissue alterations present in EMS-associated diffuse fasciitis."
    explanation: >-
      INDIRECT: the authors' inference from their own expression data to a role
      in the connective tissue change, which is a step beyond the measurement.
  - reference: PMID:21702023
    reference_title: Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We report the clinical, histopathologic, and immunogenetic features of a new case of L-tryptophan-associated EMS, along with evidence of activated transforming growth factor β and interleukin-4 signaling in the lesional skin."
    explanation: >-
      The only cited source that measures activated TGF-beta *signalling*
      rather than ligand expression, by expression profiling of lesional skin,
      and so the item that grounds the `GO:0007179` receptor-signalling binding
      on this node. The two 1991 in-situ items above measure transcripts and
      epitope deposition, which is a weaker claim.
  - reference: PMID:8285738
    reference_title: "Fibrogenic growth factors in the eosinophilia-myalgia syndrome and the toxic oil syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The presence of TGF-beta and platelet-derived growth factorAA in the periappendageal dermis was significantly more prevalent in EMS than toxic oil syndrome (57% vs 0%)."
    explanation: >-
      Immunohistochemistry putting TGF-beta in the dermis of EMS skin, and
      naming a second growth factor, PDGF-AA, that this entry does not otherwise
      carry. The contrast with toxic oil syndrome is the authors' own and is the
      reason they conclude the two diseases' fibrosis may run on different
      growth factors.
- name: Fascial and Dermal Fibrosis
  biological_scale: TISSUE
  description: >-
    Dermal thickening with homogenised collagen bundles replacing fat and
    adnexa, indistinguishable from scleroderma or morphea, and fascial
    thickening. This is the lesion behind the chronic sclerodermatous skin
    disease and the fasciitis.
  biological_processes:
  - preferred_term: extracellular matrix organization
    modifier: INCREASED
    term:
      id: GO:0030198
      label: extracellular matrix organization
  downstream:
  - target: Sclerodermatous skin induration
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:1563745
      reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Dermal thickening and homogenization of collagen bundles occurred with replacement of fat and adnexa (changes indistinguishable from scleroderma or morphea)."
      explanation: >-
        The histology that is the clinical induration, which is exactly this
        edge from lesion to sign.
  - target: Eosinophilic fasciitis
    causal_link_type: DIRECT
  evidence:
  - reference: PMID:1727618
    reference_title: The cause and pathogenesis of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Mononuclear cell activation and infiltration of various affected tissues as well as fibrosis of the integument and of the connective tissue components of blood vessels, nerves, and muscles are additional frequent findings."
    explanation: >-
      The distribution of the fibrosis across integument, vessel, nerve and
      muscle connective tissue, which is what this node claims.
- name: Perineural Inflammation and Axonal Injury
  biological_scale: TISSUE
  description: >-
    Perivascular infiltrates and fibrosis in the connective tissue of nerve,
    presenting clinically as a sensorimotor polyneuropathy that in some patients
    has prominent demyelinating features on electrodiagnostic study. Neuropathy
    is the one physical finding that did not improve over two years of
    follow-up.
  downstream:
  - target: Peripheral neuropathy
    causal_link_type: DIRECT
  - target: Demyelinating sensorimotor polyneuropathy
    causal_link_type: DIRECT
  - target: Muscle weakness
    causal_link_type: DIRECT
    description: >-
      The weakness that mattered clinically was the weakness of the
      neuromuscular disease, and it is the step surveillance data traced onward
      to respiratory failure.
    evidence:
    - reference: PMID:8295183
      reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "The most commonly observed disease process leading to death was progressive polyneuropathy and myopathy (24 of the 36 reported deaths) which produced complications of pneumonia and sepsis or respiratory failure due to weakness"
      explanation: >-
        Names weakness as the consequence of the neuropathy, which is what this
        edge asserts. The sentence credits the myopathy for it as well, and this
        entry leaves `Proximal myopathy` unwired because the muscle pathology
        does not support an edge to it - see the
        `myalgia_without_myofiber_injury` discussion.
  evidence:
  - reference: PMID:1323757
    reference_title: Demyelinating polyneuropathy in eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We describe 3 patients with EMS who presented with a severe demyelinating sensorimotor polyneuropathy."
    explanation: The clinical form of the nerve lesion this node describes.
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most commonly observed disease process leading to death was progressive polyneuropathy and myopathy (24 of the 36 reported deaths) which produced complications of pneumonia and sepsis or respiratory failure due to weakness"
    explanation: >-
      Establishes that this node, and not the fibrosis, is the route to most of
      the deaths reported to surveillance.
  - reference: PMID:8285738
    reference_title: "Fibrogenic growth factors in the eosinophilia-myalgia syndrome and the toxic oil syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Prominent staining of transforming growth factor-beta was also present in the perimysial connective tissue of five (63%) of eight EMS muscle biopsy specimens and one sural nerve biopsy specimen."
    explanation: >-
      The only molecular observation in a peripheral nerve specimen anywhere in
      the cited literature, and it puts the same growth factor in the nerve that
      drives the fibrosis elsewhere. One nerve, so it is a single observation
      rather than a series.
- name: Cardiac Arterial and Conduction System Fibrosis
  biological_scale: TISSUE
  description: >-
    In three hearts examined after death from the disease, arterial narrowing
    and arteritis were widespread, neuritis and ganglionitis were present
    throughout the heart including the conduction system, and dense fibrosis had
    replaced all sinus node tissue.
  downstream:
  - target: Sudden cardiac death
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      The authors argue the lesions supply an anatomic substrate for electrical
      instability. A substrate is not a demonstrated mechanism of death, and the
      link is typed accordingly.
    evidence:
    - reference: PMID:2058857
      reference_title: Postmortem studies of the heart in three fatal cases of the eosinophilia-myalgia syndrome.
      supports: SUPPORT
      directness: INDIRECT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "The pathologic lesions present in the coronary arteries, neural structures, and conduction system of the heart in patients who died from the eosinophilia-myalgia syndrome provide a suitable anatomic substrate for substantial cardiac electrical instability, including the occurrence of sudden death."
      explanation: >-
        INDIRECT: the paper offers the lesions as a substrate for sudden death
        rather than reporting that they caused it.
  evidence:
  - reference: PMID:2058857
    reference_title: Postmortem studies of the heart in three fatal cases of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Within the sinus node, areas of dense fibrosis replaced all nodal tissue."
    explanation: The conduction-system lesion this node is named for.
phenotypes:
- category: Musculoskeletal
  name: Myalgia
  description: >-
    Severe generalised muscle pain, incapacitating and limiting usual activity.
    It is one of the two features the surveillance case definition requires, so
    it is reported in every case by construction.
  phenotype_term:
    preferred_term: Myalgia
    term:
      id: HP:0003326
      label: Myalgia
  evidence:
  - reference: PMID:2001136
    reference_title: Eosinophilia-myalgia syndrome. A clinical case series of 21 patients. New Mexico Eosinophilia-Myalgia Syndrome Study Group.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "All cases involved eosinophilia (eosinophil count, greater than or equal to 2.0 x 10(9)/L) and incapacitating myalgias."
    explanation: >-
      The severity in a defined case series. No `frequency` is recorded here
      because the definition requires the feature - see the entry `notes:`.
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."
    explanation: >-
      Cited here for the chronic form: episodic myalgia in 64% of a followed
      cohort, which is a different claim from the definitional acute myalgia.
- category: Hematologic
  name: Peripheral eosinophilia
  description: >-
    An absolute blood eosinophil count at or above 1.0 x 10^9 cells per litre in
    the surveillance definition, and at or above 2.0 x 10^9 in some series. It
    can resolve while the chronic disease continues, so a normal count does not
    establish remission.
  phenotype_term:
    preferred_term: Increased total eosinophil count
    term:
      id: HP:0001880
      label: Increased total eosinophil count
  evidence:
  - reference: PMID:2314421
    reference_title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "All three patients, who were women 37 to 44 years of age, had severe muscle pain, muscle weakness, mouth ulcers, and striking eosinophilia"
    explanation: >-
      The eosinophilia in the index series. As with myalgia, no `frequency` is
      recorded because the case definition requires it.
  - reference: PMID:21702023
    reference_title: Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Although treatment with prednisone and mycophenolate resulted in modest improvement in proximal muscle strength and myalgia along with prompt resolution of peripheral blood eosinophilia and ground glass opacifications seen on CT scans, skin induration and neuropathy progressed."
    explanation: >-
      The count and the disease coming apart, in one patient and one sentence:
      the eosinophilia resolved promptly while the skin and nerve disease got
      worse. It is the source for this record's claim that a normal count does
      not establish remission.
- category: Musculoskeletal
  name: Arthralgia
  frequency: FREQUENT
  description: Joint pain, the most frequent feature after the two definitional ones.
  phenotype_term:
    preferred_term: Arthralgia
    term:
      id: HP:0002829
      label: Arthralgia
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: >-
      73% in national surveillance, which is FREQUENT on the HPO band. The same
      sentence carries five other frequencies used elsewhere in this entry.
- category: Dermatologic
  name: Skin rash
  frequency: FREQUENT
  description: >-
    Early lesions are nonspecific and mostly an erythematous macular eruption on
    trunk and extremities. Cutaneous involvement of some kind developed in 87%
    of one prospectively followed series.
  phenotype_term:
    preferred_term: Skin rash
    term:
      id: HP:0000988
      label: Skin rash
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: Rash in 60% of surveillance-reported cases.
  - reference: PMID:2273104
    reference_title: "Cutaneous manifestations of the L-tryptophan-associated eosinophilia-myalgia syndrome: a spectrum of sclerodermatous skin disease."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Overall, cutaneous manifestations developed in 26 patients (87%)."
    explanation: >-
      The broader skin-involvement figure from prospective follow-up. The
      recorded `frequency` follows the surveillance rash figure rather than this
      one, because 87% counts every cutaneous manifestation including the
      sclerodermatous ones curated separately.
- category: Cardiovascular
  name: Peripheral edema
  frequency: FREQUENT
  description: >-
    Peripheral or truncal oedema, often the subacute stage that precedes the
    sclerodermatous skin change.
  phenotype_term:
    preferred_term: Peripheral edema
    term:
      id: HP:0012398
      label: Peripheral edema
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: Peripheral oedema in 59% of surveillance-reported cases.
  - reference: PMID:2273104
    reference_title: "Cutaneous manifestations of the L-tryptophan-associated eosinophilia-myalgia syndrome: a spectrum of sclerodermatous skin disease."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most characteristic abnormality noted was the spectrum of sclerodermatous disease in 15 patients (50%) often after a subacute stage of peripheral or truncal edema."
    explanation: Places the oedema in sequence before the skin induration.
- category: Respiratory
  name: Dyspnea
  frequency: FREQUENT
  description: >-
    Breathlessness, with or without cough. Diffusing capacity was reduced in
    five of the six patients in one pulmonary series, and bronchoalveolar lavage
    showed raised eosinophils in two of the three who underwent it.
  phenotype_term:
    preferred_term: Dyspnea
    term:
      id: HP:0002094
      label: Dyspnea
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: >-
      Cough or dyspnoea in 59%. The surveillance category pools the two
      symptoms, which is why the frequency sits on this record and not on a
      separate cough record.
- category: Respiratory
  name: Decreased DLCO
  description: >-
    Reduced carbon monoxide diffusing capacity in five of the six patients
    reported in a pulmonary series. Exercise testing in three was consistent
    with pulmonary vascular or parenchymal disease.
  phenotype_term:
    preferred_term: Decreased DLCO
    term:
      id: HP:0045051
      label: Decreased DLCO
  evidence:
  - reference: PMID:1582284
    reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The diffusing capacity for carbon monoxide was decreased in five patients tested."
    explanation: >-
      The measurement. Note the source gives no denominator for the testing, so
      this record says five of the six reported rather than five of five tested.
      No `frequency` is given either way: a six-patient series selected for
      dyspnoea cannot supply a disease frequency.
  - reference: PMID:1582284
    reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Two patients undergoing BAL exhibited increased eosinophils in the lavage fluid; a third had elevated lymphocytes."
    explanation: >-
      INDIRECT: lavage eosinophilia indicates the lung compartment is involved
      by the same process, which is an inference from the cell counts to the
      cause of the gas transfer defect. The same sentence gives the denominator
      this record uses: a third patient also had lavage, so it is two of three
      rather than two of two.
- category: Cardiovascular
  name: Pulmonary arterial hypertension
  description: >-
    Reported in the pulmonary series as a persisting outcome in one patient
    alongside a death from respiratory failure despite immunosuppression.
  phenotype_term:
    preferred_term: Pulmonary arterial hypertension
    term:
      id: HP:0002092
      label: Pulmonary arterial hypertension
  evidence:
  - reference: PMID:1582284
    reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Another remains markedly dyspneic with pulmonary hypertension."
    explanation: >-
      A single patient in a six-patient series, which is why no frequency is
      recorded. Toxic oil syndrome's pulmonary hypertension is far better
      characterised and is not imported here.
- category: Constitutional
  name: Fever
  description: Fever in the acute illness.
  phenotype_term:
    preferred_term: Fever
    term:
      id: HP:0001945
      label: Fever
  evidence:
  - reference: PMID:2314421
    reference_title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Other manifestations included fever, abdominal pain, dyspnea, skin rash, and elevated serum concentrations of aminotransferase and aldolase."
    explanation: >-
      Lists fever among the acute manifestations of the index three cases. No
      frequency is recorded: the published frequencies for fever in this disease
      come from cohorts this entry does not cite.
- category: Musculoskeletal
  name: Muscle weakness
  frequency: FREQUENT
  description: >-
    Weakness, in 68% of a population-based cohort still reporting it a year
    after onset, and proximal weakness in 40% of a referral cohort at three
    years. It also carries persisting excess against exposed controls who did
    not fall ill.
  phenotype_term:
    preferred_term: Muscle weakness
    term:
      id: HP:0001324
      label: Muscle weakness
  evidence:
  - reference: PMID:8624177
    reference_title: The natural history of eosinophilia-myalgia syndrome in a tryptophan-exposed cohort in South Carolina.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "survivors with definite EMS continued to report excess morbidity for 6 major EMS symptoms (myalgia, arthralgia, weakness, rash, alopecia, and sclerodermiform skin changes)"
    explanation: >-
      Weakness among six symptoms with excess morbidity in survivors against
      exposed-but-not-ill controls. It is not the source of the frequency band:
      this sentence reports persisting excess rather than a share of cases.
  - reference: PMID:1520057
    reference_title: Eosinophilia-myalgia syndrome. Natural history in a population-based cohort.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "At 12 months, 41 patients (77%) continued to report fatigue, 36 (68%) weakness, and 34 (64%) myalgias"
    explanation: >-
      68% at twelve months, which is the FREQUENT band and the figure this record
      follows. The cohort is population-based rather than referral, which is why
      it is preferred to the higher referral-cohort numbers.
  - reference: PMID:8129767
    reference_title: Chronicity of the eosinophilia-myalgia syndrome. A reassessment after three years.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Fatigue (91%), muscle cramping (75%), myalgia (70%), paresthesias with objectively demonstrated hypesthesias (62%), articular symptoms (54%), scleroderma-like skin changes (44%), and proximal muscle weakness (40%) are the most common features of chronic EMS."
    explanation: >-
      Proximal weakness in 40% at a mean of three years, in a 57-patient
      prospective cohort. The same sentence carries the frequencies used on
      `Fatigue`, `Paresthesia` and `Proximal myopathy`.
- category: Musculoskeletal
  name: Proximal myopathy
  frequency: FREQUENT
  description: >-
    Proximal myopathy in 36% of a followed cohort. It sits oddly beside the
    muscle histology, which shows minimal myofiber damage.
  phenotype_term:
    preferred_term: Proximal myopathy
    term:
      id: HP:0003198
      label: Myopathy
  evidence:
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."
    explanation: >-
      36% in a 32-patient prospectively followed cohort, which is FREQUENT.
      `preferred_term` is narrower than the bound HPO term, which has no
      proximal-myopathy child.
- category: Dermatologic
  name: Sclerodermatous skin induration
  frequency: FREQUENT
  description: >-
    A spectrum of sclerodermatous disease: diffuse, limited or localised
    scleroderma, sometimes with scleromyxedema-like mucinous papules. It follows
    a subacute oedematous stage and is the most characteristic chronic skin
    finding.
  phenotype_term:
    preferred_term: Sclerodermatous skin induration
    term:
      id: HP:0100324
      label: Scleroderma
  evidence:
  - reference: PMID:2273104
    reference_title: "Cutaneous manifestations of the L-tryptophan-associated eosinophilia-myalgia syndrome: a spectrum of sclerodermatous skin disease."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most characteristic abnormality noted was the spectrum of sclerodermatous disease in 15 patients (50%) often after a subacute stage of peripheral or truncal edema."
    explanation: 50% in prospective follow-up of 30 patients, which is FREQUENT.
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."
    explanation: >-
      54% in an independent cohort, agreeing with the 50% above. Two cohorts is
      why this frequency is stated with more confidence than most here.
- category: Dermatologic
  name: Eosinophilic fasciitis
  frequency: FREQUENT
  description: >-
    Clinical or biopsy evidence of eosinophilic fasciitis, in 30% of a
    prospective series. Three of four patients in the Maryland series presented
    with it, which is how the disease was first met in some centres.
  phenotype_term:
    preferred_term: Eosinophilic fasciitis
    term:
      id: HP:0045029
      label: Eosinophilic fasciitis
  evidence:
  - reference: PMID:2273104
    reference_title: "Cutaneous manifestations of the L-tryptophan-associated eosinophilia-myalgia syndrome: a spectrum of sclerodermatous skin disease."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Clinical and/or biopsy evidence of eosinophilic fasciitis was seen in nine patients (30%)."
    explanation: 30% of 30 prospectively followed patients, which is FREQUENT.
  - reference: PMID:2369429
    reference_title: Eosinophilia-myalgia syndrome due to L-tryptophan ingestion. Report of four cases and review of the Maryland experience.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We describe 4 patients with EMS, 3 of whom presented with eosinophilic fasciitis, and we review the epidemiology of EMS reported in Maryland"
    explanation: >-
      Fasciitis as the presenting picture in three of four reported cases, from
      an independent series.
- category: Dermatologic
  name: Alopecia
  frequency: FREQUENT
  description: Hair loss, described as frequently a late sequela.
  phenotype_term:
    preferred_term: Alopecia
    term:
      id: HP:0001596
      label: Alopecia
  evidence:
  - reference: PMID:2273104
    reference_title: "Cutaneous manifestations of the L-tryptophan-associated eosinophilia-myalgia syndrome: a spectrum of sclerodermatous skin disease."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Alopecia, frequently a late sequela, developed in 11 (37%)."
    explanation: 37% of 30 patients, and its late timing, from the same series.
- category: Neurologic
  name: Peripheral neuropathy
  frequency: FREQUENT
  description: >-
    Sensorimotor polyneuropathy, in 61% of one followed cohort and reported as
    neuropathy or neuritis in 27% of surveillance cases. It is the physical
    finding that did not improve over two years, and it leads the disease
    processes behind reported deaths.
  phenotype_term:
    preferred_term: Peripheral neuropathy
    term:
      id: HP:0009830
      label: Peripheral neuropathy
    clinical_course: PROGRESSIVE
  evidence:
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."
    explanation: 61% in a followed cohort, which is FREQUENT.
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Neuropathy or neuritis, resulting in paralysis and death in some patients, was seen in 27%, and chest roentgenogram abnormalities were noted in 21% of those tested."
    explanation: >-
      The surveillance figure, which is lower than the cohort one, and the
      severity at its extreme. The band follows the cohort figure because the
      surveillance count depends on what reporting physicians recorded.
  - reference: PMID:7741371
    reference_title: "The eosinophilia-myalgia syndrome: status of 205 patients and results of treatment 2 years after onset."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Nearly all physical findings were also reported to have improved or resolved in most patients; only peripheral neuropathy was unchanged."
    explanation: >-
      The sentence that separates this phenotype from every other in the entry:
      at two years it was the one physical finding that had not improved. Read
      exactly, it supports a failure to remit and not the `clinical_course`
      qualifier, which is carried by the surveillance item below instead. An
      earlier draft cited this sentence for PROGRESSIVE, which inverts it.
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most commonly observed disease process leading to death was progressive polyneuropathy and myopathy (24 of the 36 reported deaths) which produced complications of pneumonia and sepsis or respiratory failure due to weakness"
    explanation: >-
      The word this record's `clinical_course: PROGRESSIVE` rests on, from
      surveillance data on the deaths. The two items together are the honest
      picture: the neuropathy progressed in those it killed and was merely
      unchanged in the survivors followed for two years.
- category: Neurologic
  name: Demyelinating sensorimotor polyneuropathy
  description: >-
    A severe form with multifocal conduction block, slowing and temporal
    dispersion of motor responses and prolonged or absent F-responses. Two of
    three reported patients died despite plasmapheresis, corticosteroids and in
    one case cyclophosphamide.
  phenotype_term:
    preferred_term: Demyelinating peripheral neuropathy
    term:
      id: HP:0007108
      label: Demyelinating peripheral neuropathy
  evidence:
  - reference: PMID:1323757
    reference_title: Demyelinating polyneuropathy in eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We describe 3 patients with EMS who presented with a severe demyelinating sensorimotor polyneuropathy."
    explanation: >-
      A three-patient report, so no frequency is recorded. Carried separately
      from `Peripheral neuropathy` because the electrophysiology and the outcome
      are different.
  - reference: PMID:1323757
    reference_title: Demyelinating polyneuropathy in eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Despite plasmapheresis; corticosteroids; and, in 1 patient, cyclophosphamide, 2 patients died and the remaining patient experienced minimal recovery."
    explanation: The outcome, and the reason this form is separated out.
- category: Neurologic
  name: Paresthesia
  frequency: FREQUENT
  description: >-
    Abnormal sensation, with objectively demonstrated hypesthesia in 62% of a
    prospective cohort at three years. It is a later-onset feature rather than
    part of the presenting illness.
  phenotype_term:
    preferred_term: Paresthesia
    term:
      id: HP:0003401
      label: Paresthesia
  evidence:
  - reference: PMID:8129767
    reference_title: Chronicity of the eosinophilia-myalgia syndrome. A reassessment after three years.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Fatigue (91%), muscle cramping (75%), myalgia (70%), paresthesias with objectively demonstrated hypesthesias (62%), articular symptoms (54%), scleroderma-like skin changes (44%), and proximal muscle weakness (40%) are the most common features of chronic EMS."
    explanation: >-
      62% with objectively demonstrated hypesthesia, which is the frequency this
      record follows. The objective confirmation is what separates this from a
      purely reported symptom.
  - reference: PMID:1520057
    reference_title: Eosinophilia-myalgia syndrome. Natural history in a population-based cohort.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Symptoms with later onset included paresthesias, muscle cramps, extremity weakness, and alopecia."
    explanation: >-
      The timing: paresthesia belongs to the later illness rather than to the
      acute phase, which is the sequence this record's description asserts.
- category: Gastrointestinal
  name: Xerostomia
  frequency: FREQUENT
  description: >-
    Dry mouth, reported by 66% of patients in a questionnaire survey across 33
    states. It is the entry's only mucosal-secretory feature, and toxic oil
    syndrome's sicca syndrome is its counterpart in the sibling disease.
  phenotype_term:
    preferred_term: Xerostomia
    term:
      id: HP:0000217
      label: Xerostomia
  evidence:
  - reference: PMID:7857025
    reference_title: Head and neck manifestations of eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Among the 28 various head and neck manifestations studied, 70% of EMS patients complained of generalized muscle spasms, 66% xerostomia, 62% dyspnea, and 56% dysphagia."
    explanation: >-
      66% for xerostomia, which is the frequency this record follows. The same
      sentence carries the dysphagia figure used below and corroborates the
      dyspnoea figure taken from surveillance.
  notes: >-
    Self-reported on a questionnaire, with no objective measure of salivary flow
    anywhere in the cited literature, and the respondents were reached through
    patient contact rather than sampled. The frequency is therefore a share of
    respondents. It is recorded because dryness is otherwise absent from this
    entry and is prominent in the sibling toxic oil syndrome.
- category: Gastrointestinal
  name: Dysphagia
  frequency: FREQUENT
  description: >-
    Difficulty swallowing, reported by 56% of respondents in the same
    head-and-neck survey, which raises the question of aspiration in a disease
    that also weakens respiratory muscle.
  phenotype_term:
    preferred_term: Dysphagia
    term:
      id: HP:0002015
      label: Dysphagia
  evidence:
  - reference: PMID:7857025
    reference_title: Head and neck manifestations of eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Among the 28 various head and neck manifestations studied, 70% of EMS patients complained of generalized muscle spasms, 66% xerostomia, 62% dyspnea, and 56% dysphagia."
    explanation: 56% for dysphagia, from the same survey and the same sentence.
  notes: >-
    Self-reported, with the same limits as the xerostomia record. No cited source
    reports a swallowing study, and no edge is drawn to it: whether the dysphagia
    is myopathic, neuropathic or fibrotic is not addressed anywhere in this
    literature.
- category: Laboratory
  name: Elevated circulating aldolase concentration
  frequency: FREQUENT
  description: >-
    Raised serum aldolase, in 46% of surveillance cases and abnormal in every
    patient tested in a 21-patient series.
  phenotype_term:
    preferred_term: Elevated circulating aldolase concentration
    term:
      id: HP:0012544
      label: Elevated circulating aldolase concentration
  reports_on:
  - target: Inflammatory Infiltration of Fascia, Perimysium and Dermis
    relationship: READOUT_OF
    direction: POSITIVE
    endpoint_context: DIAGNOSTIC
    interpretation: >-
      Aldolase is released from muscle, so a raised level reports on the muscle
      compartment being involved. It is an observational link and not a causal
      edge: no cited source explains the enzyme rise, and the muscle histology
      in this disease shows little myofiber damage.
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: 46% in surveillance, which is the figure the frequency band follows.
  - reference: PMID:2001136
    reference_title: Eosinophilia-myalgia syndrome. A clinical case series of 21 patients. New Mexico Eosinophilia-Myalgia Syndrome Study Group.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Aldolase levels were abnormal in all patients tested."
    explanation: >-
      Abnormal in every tested patient of a clinical series, which is higher
      than the surveillance figure. The band follows the surveillance number
      because the series does not say how many were tested.
- category: Laboratory
  name: Elevated circulating hepatic transaminase concentration
  frequency: FREQUENT
  description: Abnormal liver tests, in 43% of surveillance cases and mild where described.
  phenotype_term:
    preferred_term: Elevated circulating hepatic transaminase concentration
    term:
      id: HP:0002910
      label: Elevated circulating hepatic transaminase concentration
  reports_on:
  - target: Systemic Exposure to L-Tryptophan Process Contaminants
    relationship: READOUT_OF
    direction: POSITIVE
    endpoint_context: DIAGNOSTIC
    interpretation: >-
      Raised transaminases report that the exposure reached the liver, which is
      where the aniline-derivative chemistry of this disease is studied.
      Observational only: no cited source describes a hepatic lesion, and the
      abnormality is reported as mild.
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: 43% in surveillance.
  - reference: PMID:2001136
    reference_title: Eosinophilia-myalgia syndrome. A clinical case series of 21 patients. New Mexico Eosinophilia-Myalgia Syndrome Study Group.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Fourteen (88%) of the 16 patients tested had mild liver function abnormalities."
    explanation: >-
      88% of those tested in a clinical series, and the word "mild", which the
      surveillance figure does not supply.
- category: Gastrointestinal
  name: Oral ulcer
  description: Mouth ulcers, reported in all three patients of the index series.
  phenotype_term:
    preferred_term: Oral ulcer
    term:
      id: HP:0000155
      label: Oral ulcer
  evidence:
  - reference: PMID:2314421
    reference_title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "All three patients, who were women 37 to 44 years of age, had severe muscle pain, muscle weakness, mouth ulcers, and striking eosinophilia"
    explanation: >-
      Three of three in the index series, so no frequency. No later cohort cited
      here reports the feature.
- category: Neurologic
  name: Cognitive impairment
  frequency: VERY_FREQUENT
  description: >-
    Memory and attention complaints, present in 86% of a 57-patient prospective
    cohort at a mean of three years, where the authors report it as a new
    manifestation of the chronic disease. It is also the exception to the general
    improvement at two years: reported worse in 32% of 205 patients. In a series
    selected for CNS abnormality, amnesia was present in 88% and intermittent
    confusion in 38%.
  phenotype_term:
    preferred_term: Cognitive impairment
    term:
      id: HP:0100543
      label: Cognitive impairment
  evidence:
  - reference: PMID:7741371
    reference_title: "The eosinophilia-myalgia syndrome: status of 205 patients and results of treatment 2 years after onset."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Cognitive changes were reported to be worse in 32% of patients."
    explanation: >-
      Not the source of the frequency band. 32% is the share reported worse at
      follow-up rather than the share affected, and the two are different
      quantities.
  - reference: PMID:8129767
    reference_title: Chronicity of the eosinophilia-myalgia syndrome. A reassessment after three years.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "New findings identified among this cohort include cognitive symptoms in 86% of the study group, tremor, and myoclonus."
    explanation: >-
      86% in a 57-patient prospective cohort, which is the VERY_FREQUENT band and
      the figure this record follows. It is a single referral cohort, which the
      description says. The tremor and myoclonus in the same sentence are real
      findings and are not curated as phenotypes here, because nothing in the
      cited literature quantifies or localises them.
  - reference: PMID:9802492
    reference_title: "Neurologic, MR imaging, and MR spectroscopic findings in eosinophilia myalgia syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Neurologic findings that were increased in CNS-EMS included minor depression (100%), amnesia (88%), and intermittent confusion (38%)"
    explanation: >-
      Frequencies within a series selected for CNS abnormality, which is why
      they are not read as disease frequencies either.
  - reference: PMID:9802492
    reference_title: "Neurologic, MR imaging, and MR spectroscopic findings in eosinophilia myalgia syndrome."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The pattern of cerebral lesions and neurometabolites is consistent with widespread inflammatory cerebrovascular disease."
    explanation: >-
      INDIRECT, and the only mechanistic reading of the cognitive findings in
      the cited literature. It is not wired into the pathograph: the same paper
      reports a second spectral pattern consistent instead with a metabolic
      encephalopathy, and says the abnormalities should be interpreted
      cautiously.
- category: Constitutional
  name: Fatigue
  frequency: FREQUENT
  description: >-
    Fatigue, and the commonest persisting complaint in both cohorts that
    followed patients for a year or more: 77% at twelve months in a
    population-based cohort and 91% at three years in a referral cohort. It is
    the feature most patients are left with.
  phenotype_term:
    preferred_term: Fatigue
    term:
      id: HP:0012378
      label: Fatigue
  evidence:
  - reference: PMID:1563745
    reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: BACKGROUND
    snippet: "Arthralgias, edema of the extremities, morbilliform rashes, skin induration, weakness, fatigue, and respiratory weakness may be present as well."
    explanation: >-
      BACKGROUND because this is the pathology paper's framing sentence
      describing the established clinical picture rather than its own finding,
      and INDIRECT because "may be present" is a statement about the syndrome
      rather than an observation in the 11 cases.
  - reference: PMID:1520057
    reference_title: Eosinophilia-myalgia syndrome. Natural history in a population-based cohort.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "At 12 months, 41 patients (77%) continued to report fatigue, 36 (68%) weakness, and 34 (64%) myalgias"
    explanation: >-
      77% at twelve months in a population-based cohort, which is the figure the
      FREQUENT band follows. The referral cohort below reports 91%, which would
      be VERY_FREQUENT; the lower population-based figure is preferred because
      the cohort was not selected by referral.
  - reference: PMID:8129767
    reference_title: Chronicity of the eosinophilia-myalgia syndrome. A reassessment after three years.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Fatigue (91%), muscle cramping (75%), myalgia (70%), paresthesias with objectively demonstrated hypesthesias (62%), articular symptoms (54%), scleroderma-like skin changes (44%), and proximal muscle weakness (40%) are the most common features of chronic EMS."
    explanation: >-
      91% at a mean of three years, the highest figure for any feature in this
      disease. It is recorded rather than used for the band, for the reason given
      on the item above.
- category: Cardiovascular
  name: Sudden cardiac death
  description: >-
    Sudden death attributed to arrhythmia in two of 36 deaths reported to
    surveillance, against a background of arterial, neural and conduction system
    lesions found in every heart examined after death.
  phenotype_term:
    preferred_term: Sudden cardiac death
    term:
      id: HP:0001645
      label: Sudden cardiac death
  evidence:
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "cardiomyopathy was the underlying cause of death for 4 patients, primary pulmonary disease for 3, sudden death attributed to arrhythmia for 2, stroke for 2, and septic complications of therapy for one"
    explanation: >-
      The specific claim this record makes, in the source's own words: two of
      the 36 reported deaths were sudden and attributed to arrhythmia. Frequency
      is deliberately absent, because these are proportions of deaths rather
      than of cases.
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Of the 36 patients who died, 33 (92%) had neuromuscular sequelae, 29 (81%) had pulmonary complications, and 23 (64%) had cardiac manifestations."
    explanation: >-
      INDIRECT, and kept for context rather than for the claim: cardiac
      involvement of some kind in 64% of the deaths is the background against
      which two sudden deaths sit.
diagnosis:
- name: High-Resolution CT of the Chest
  description: >-
    Abnormal in four of the six patients in the one pulmonary series, and in one
    of them the plain chest radiograph was normal. So a normal radiograph does
    not exclude the lung involvement, which matters in a disease where most of
    the reported deaths had pulmonary complications.
  evidence:
  - reference: PMID:1582284
    reference_title: Pulmonary manifestations of the eosinophilia-myalgia syndrome associated with tryptophan ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "One of these patients showed no abnormality on routine chest roentgenogram."
    explanation: >-
      The specific claim this record makes: HRCT found disease that the plain
      film missed. One patient, in a six-patient series.
- name: Electrodiagnostic Studies
  description: >-
    Nerve conduction study and electromyography. In the severe form they show
    multifocal conduction block, slowed and temporally dispersed motor responses
    and prolonged or absent F-responses, which is what identifies the neuropathy
    as demyelinating rather than axonal.
  evidence:
  - reference: PMID:1323757
    reference_title: Demyelinating polyneuropathy in eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Electrodiagnostic studies revealed multifocal conduction block, slowing and temporal dispersion of motor responses, and prolonged or absent F-responses."
    explanation: >-
      The findings themselves, from the three-patient report of the demyelinating
      form. They are what the `Demyelinating sensorimotor polyneuropathy`
      phenotype rests on.
- name: Muscle Biopsy
  description: >-
    Shows eosinophilic perimyositis or interstitial inflammation. Note what it
    does not show: the infiltrate sits in perimysium, epimysium and fascia, and
    the muscle fibres themselves are close to intact, so a biopsy read for
    myofiber damage will look unremarkable beside the patient's pain.
  evidence:
  - reference: PMID:2001136
    reference_title: Eosinophilia-myalgia syndrome. A clinical case series of 21 patients. New Mexico Eosinophilia-Myalgia Syndrome Study Group.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Muscle biopsies were done in five patients; four showed eosinophilic perimyositis, and one had interstitial inflammation."
    explanation: >-
      What the biopsy found, in the five patients of a 21-patient series who had
      one. Four of five is a count within a small series and not a sensitivity.
  - reference: PMID:1563745
    reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The earliest apparent histologic changes were observed at the septa between subcutaneous fat lobules and in the deep dermis or fascia."
    explanation: >-
      Where to look, and why a sampling decision matters here: the earliest
      change is septal and fascial, so a biopsy that stops above the fascia can
      miss it. This is the reason the literature recommends full-thickness
      sampling, although no cited sentence states that recommendation directly.
  notes: >-
    This section carries only what a cited abstract states. The wider diagnostic
    workup the deep-research report describes, which includes MRI of symptomatic
    muscle and fascia, serial eosinophil counts and exclusion of parasitic
    infection, is general clinical practice that no reference cached here
    attributes to a study of this disease, so it is not recorded as though it
    were. The case definition itself is a `definitions:` entry rather than a
    diagnostic test.
definitions:
- name: CDC 1989 surveillance case definition
  definition_type: CASE_DEFINITION
  derivation_basis: ESTABLISHED_CRITERIA
  description: >-
    The definition used to count the epidemic: debilitating generalised myalgia
    with an absolute eosinophil count at or above 1.0 x 10^9 cells per litre,
    and no infection or neoplasm accounting for the findings. It was built for
    outbreak surveillance rather than for diagnosis, and every frequency in this
    entry is conditioned on it.
  validation_status:
    status: UNVALIDATED
    rationale: >-
      Never validated by an appropriate challenge, on the account of the working
      group that reviewed it, and routinely used for purposes it was not built
      for. A revised set of criteria was proposed in the same paper and is not
      curated here.
    evidence:
    - reference: PMID:8895176
      reference_title: Criteria for the definition of the eosinophilia-myalgia syndrome.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "The widely disseminated surveillance case definition of the eosinophilia-myalgia syndrome (EMS) recommended by the Centers for Disease Control in 1989 has never been validated by an appropriate challenge and has commonly been used for unintended purposes."
      explanation: The validation status of this definition, stated directly.
  criteria_sets:
  - name: Surveillance criteria
    minimum_required: 2
    core_clinical_characteristics:
    - preferred_term: debilitating generalised myalgia
      term:
        id: HP:0003326
        label: Myalgia
      description: Severe enough to limit usual activity.
    - preferred_term: absolute eosinophil count at or above 1.0 x 10^9 cells per litre
      term:
        id: HP:0001880
        label: Increased total eosinophil count
      description: The numeric threshold the definition sets.
    exclusion_criteria:
    - preferred_term: infection accounting for the findings
      description: >-
        Any infection that could explain the myalgia and the eosinophilia.
        Deliberately unbound: the exclusion is a class of alternative diagnosis
        rather than a single concept, and no cited sentence names a term for it.
    - preferred_term: neoplasm accounting for the findings
      description: >-
        Any neoplasm that could explain the findings, including the clonal
        eosinophilias. Unbound for the same reason.
    evidence:
    - reference: PMID:2398610
      reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "A case is defined by debilitating myalgias and absolute eosinophilia greater than or equal to 1.0 x 10(9) cells/L."
      explanation: >-
        The definition in the surveillance report's own words. `minimum_required`
        is 2 because both features are required, not one of two.
  notes: >-
    `minimum_required: 2` records that the definition is conjunctive. The
    exclusion clause is carried as free text because the quoted sentence states
    the two positive requirements only; the exclusion comes from the definition
    as described elsewhere in the same literature and is not separately cited
    here. The `case_definition_validity` discussion is where the consequences of
    this definition for every frequency in the entry are argued.
environmental:
- name: Ingestion of manufactured L-tryptophan supplement containing process impurities
  description: >-
    Over-the-counter L-tryptophan taken as a sleep or mood supplement, from lots
    produced by one manufacturer after a change of fermentation strain and a
    reduction in the carbon purification step. Withdrawal of the product from
    the United States market ended the epidemic.
  exposure_term:
    preferred_term: exposure to a manufactured dietary supplement carrying process impurities
    term:
      id: ECTO:9002126
      label: exposure to environmental food contaminant
  influences_mechanisms:
  - target: Ingestion of Contaminated Manufactured L-Tryptophan
    environmental_effect: TRIGGERS
    causal_link_type: DIRECT
    description: >-
      The exposure is the disease's cause on every published account. It was
      essentially universal among cases and rare among controls, and incidence
      fell away when the product was recalled.
    evidence:
    - reference: PMID:8895184
      reference_title: Tryptophan produced by Showa Denko and epidemic eosinophilia-myalgia syndrome.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "In case-control studies of EMS, LT exposure was essentially universal among cases but rare among controls."
      explanation: >-
        States the exposure-disease link this edge asserts, across the
        case-control literature rather than in one study.
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Since the recall of over-the-counter preparations of tryptophan in November 1989, the number of new cases of this potentially fatal disorder has fallen dramatically."
    explanation: >-
      Removing the exposure removed the disease, which is the strongest single
      statement about this exposure in the surveillance literature.
  - reference: PMID:21702023
    reference_title: Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: BACKGROUND
    snippet: "Following the ban by the Food and Drug Administration (FDA) on the sale of L-tryptophan, the incidence of EMS declined rapidly."
    explanation: >-
      BACKGROUND: this is the post-epidemic case report's framing of the
      established history rather than its own result, and it is cited here
      because the same paper reports a case occurring after that decline.
  notes: >-
    The bound term names a contaminant in an ingested product rather than
    tryptophan, and that is deliberate. ECTO does have a tryptophan exposure
    class, `ECTO:9002895 exposure to tryptophan`, returned by `runoak -i
    ols:ecto search "tryptophan"` - it is rejected on two grounds. Ordinary
    dietary tryptophan and uncontaminated manufactured tryptophan are not
    implicated in this disease, so binding the amino acid would assert what the
    epidemiology rules out; and that CURIE is not resolvable in the committed
    ECTO build (`runoak -i sqlite:obo:ecto info ECTO:9002895` returns no label),
    so it is not a member of the `ExposureTerm` enum and does not validate. The
    only other candidate, `ECTO:0070245 exposure to dietary supplement,
    vitamin(s) and fatty acids via ingestion` from `runoak -i ols:ecto search
    "dietary supplement"`, names a route and loses the contaminant.
    `preferred_term` carries the full concept.
genetic:
- name: HLA-DRB1 alleles as susceptibility markers
  relationship_type: RISK_FACTOR
  gene_term:
    preferred_term: HLA-DRB1
    term:
      id: hgnc:4948
      label: HLA-DRB1
  notes: >-
    Not a causal gene and not heritable disease. In a cohort of 94 documented
    L-tryptophan users, DRB1*03 and DRB1*04 were associated with developing the
    syndrome after exposure, with wide confidence intervals in a small
    retrospective sample. That pair is the all-sources analysis, and it does not
    carry over whole: restricted to users of the implicated lots, which is the
    exposure this entry is scoped to, DRB1*03 was not a risk factor and DRB1*04
    remained one. The qualification is curated as its own evidence item below,
    and an earlier draft of this note omitted it. An earlier 32-patient cohort
    found a non-significant trend towards DR4.
  evidence:
  - reference: PMID:19790128
    reference_title: Immunogenetic risk and protective factors for the development of L-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "higher LT dose (odds ratio [OR] 1.4, 95% confidence interval [95% CI] 1.1-1.8), age >45 years (OR 3.0, 95% CI 1.0-8.8), and HLA-DRB1*03 (OR 3.9, 95% CI 1.2-15.2), DRB1*04 (OR 3.9, 95% CI 1.1-16.4), and DQA1*0601 (OR 13.7, 95% CI 1.3-1.8) were risk factors for the development of EMS"
    explanation: >-
      The positive associations with their own intervals, alongside dose and
      age, which is the point of the study and the reason this record is typed
      RISK_FACTOR rather than CAUSATIVE.
  - reference: PMID:19790128
    reference_title: Immunogenetic risk and protective factors for the development of L-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Similar risk and protective factors were seen for developing EMS following ingestion of implicated LT, except that DRB1*03 was not a risk factor and DQA1*0201 was an additional protective factor."
    explanation: >-
      The implicated-lot analysis, which is the one this entry's scope actually
      needs, and its two departures from the all-sources result. It supports the
      record while narrowing it: among users of the lots that caused the
      epidemic, DRB1*03 was not a risk allele.
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "HLA-class II typing revealed a non-significant trend towards an association with HLA-DR4."
    explanation: >-
      INDIRECT, and the weakest item on this record. An earlier and smaller
      cohort found the direction and not the significance, which is a failure to
      confirm rather than a contradiction: a non-significant trend the same way
      does not cut against the association. An earlier draft graded this REFUTE,
      which read a null result as a negative one.
  - reference: PMID:19790128
    reference_title: Immunogenetic risk and protective factors for the development of L-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "In addition to the xenobiotic dose and subject age, polymorphisms in immune response genes may underlie the development of certain xenobiotic-induced immune-mediated disorders"
    explanation: >-
      INDIRECT: the authors' generalisation from their own associations to a
      class of xenobiotic-induced disease, which is the reading that makes this
      an immune-response-gene record rather than an HLA curiosity.
- name: HLA-DQA1 alleles as protective markers
  relationship_type: PROTECTIVE
  gene_term:
    preferred_term: HLA-DQA1
    term:
      id: hgnc:4942
      label: HLA-DQA1
  notes: >-
    In the same 94-user cohort, DQA1*0501 was protective overall and DQA1*0201
    additionally protective among users of implicated lots, while DQA1*0601 was
    a risk allele. One gene therefore carries alleles in both directions, which
    is why this record is separate from the DRB1 one rather than folded into it.
  evidence:
  - reference: PMID:19790128
    reference_title: Immunogenetic risk and protective factors for the development of L-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "whereas DRB1*07 (OR 0.12, 95% CI 0.02-0.48) and DQA1*0501 (OR 0.23, 95% CI 0.05-0.85) were protective"
    explanation: >-
      The protective associations with their intervals. The sentence names a
      DRB1 allele too, which is why the DRB1 record above is typed for risk on a
      different sentence rather than this one.
treatments:
- name: Withdrawal of the Implicated Supplement
  description: >-
    Stopping the L-tryptophan product. It is the only intervention that
    addresses the cause, and in the index series it was followed by improvement
    while patients remained symptomatic months later. At population level the
    recall ended the epidemic.
  treatment_term:
    preferred_term: discontinuation of the causative supplement
    term:
      id: NCIT:C128535
      label: Pharmacotherapy Discontinuation
  evidence:
  - reference: PMID:2314421
    reference_title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The discontinuation of tryptophan and the initiation of glucocorticoid treatment resulted in improvement, but all three women were still symptomatic three to five months later."
    explanation: >-
      Improvement after withdrawal, and its limit. The same sentence bundles
      withdrawal with glucocorticoid, so neither can be credited alone from it.
  notes: >-
    `NCIT:C128535` Pharmacotherapy Discontinuation is defined as "The
    discontinuation of treatment with a drug", and it is reachable from the
    `TreatmentTerm` root: `C128535` is-a `C21090` Pharmacologic Management is-a
    `C49236` Therapeutic Procedure is-a `C25218` Clinical Intervention or
    Procedure. `preferred_term` names the supplement rather than a drug, which
    is the one respect in which the binding is broader than the action.

    An earlier draft bound `NCIT:C15747` Supportive Care here and justified it
    with a note claiming no reachable NCIT term for stopping a causative agent
    exists. That claim was false, and it was reached by searching the committed
    `cache/ncit/terms.csv`, which holds only CURIEs already bound elsewhere in
    `kb/`. An empty result there is silence rather than a negative answer, which
    is the same error CLAUDE.md records for `ECTO:9001524`. The term was found by
    asking the ontology: `runoak -i sqlite:obo:ncit search "l~discontinuation"`.
- name: Systemic Glucocorticoid Therapy
  description: >-
    Prednisone in the acute phase. It was reported helpful in 79% of the
    patients who received it, and it reduces muscle pain and the eosinophil
    count. It does not prevent the chronic disease.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: prednisone
      term:
        id: CHEBI:8382
        label: prednisone
  evidence:
  - reference: PMID:7741371
    reference_title: "The eosinophilia-myalgia syndrome: status of 205 patients and results of treatment 2 years after onset."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Prednisone was reported to be helpful in 79% of patients who received it during the acute phase of the syndrome."
    explanation: >-
      The acute-phase benefit, from physician-completed review of 205 patients.
      It is a report of helpfulness rather than a controlled comparison.
  - reference: PMID:2001136
    reference_title: Eosinophilia-myalgia syndrome. A clinical case series of 21 patients. New Mexico Eosinophilia-Myalgia Syndrome Study Group.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Seven patients were treated with prednisone, and six showed improvement in muscle pain and a decrease in eosinophilia."
    explanation: >-
      Six of seven improving on two named measures, in an independent series.
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Early therapy with corticosteroids did not seem to prevent the development of chronic manifestations."
    explanation: >-
      REFUTE against corticosteroids as disease-modifying. There is deliberately
      no `target_mechanisms` edge on this treatment: the acute benefit is on
      symptoms and the eosinophil count, and the one cohort that looked for
      prevention of chronic disease did not find it.
- name: Steroid-Sparing Immunosuppression
  description: >-
    Mycophenolate, methotrexate and anakinra have been used in the one
    post-epidemic case reported in detail. Prednisone with mycophenolate cleared
    the eosinophilia and the pulmonary ground-glass change and improved strength
    and myalgia modestly, and the skin induration and neuropathy went on
    worsening through it. Methotrexate and then anakinra added nothing.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Immunosuppressive Therapy
    term:
      id: NCIT:C15261
      label: Immunosuppressive Therapy
    therapeutic_agent:
    - preferred_term: mycophenolate mofetil
      term:
        id: CHEBI:8764
        label: mycophenolate mofetil
    - preferred_term: methotrexate
      term:
        id: CHEBI:44185
        label: methotrexate
  evidence:
  - reference: PMID:21702023
    reference_title: Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Although treatment with prednisone and mycophenolate resulted in modest improvement in proximal muscle strength and myalgia along with prompt resolution of peripheral blood eosinophilia and ground glass opacifications seen on CT scans, skin induration and neuropathy progressed."
    explanation: >-
      The partial benefit, in one patient, and INDIRECT because the regimen
      bundles mycophenolate with prednisone so neither can be credited alone.
      The same sentence is carried as REFUTE below against the fibrosis and the
      neuropathy, which it says got worse regardless.
  - reference: PMID:21702023
    reference_title: Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan.
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Addition of methotrexate (20 mg weekly for 5 months) followed by daily injections of anakinra for 3 months failed to yield further improvement in myalgia, neuropathy, or skin induration."
    explanation: >-
      REFUTE against methotrexate and anakinra in this disease. It is one
      patient and an uncontrolled sequence, which is why `notes` says what this
      record does not establish.
  notes: >-
    One patient. Everything here comes from a single 2011 case report, so the
    record is an account of what was tried rather than evidence of efficacy, and
    there is deliberately no `target_mechanisms` edge: nothing here shows any of
    these agents acting on a node of the pathograph. Read the REFUTE item the
    same way, as a failed sequence in one person rather than a demonstration
    that the drugs do not work.
- name: Supportive and Symptomatic Care
  description: >-
    Analgesia, respiratory management and management of the chronic
    complications. It is the mainstay because no other treatment was found
    consistently beneficial, and because most symptoms improved with time
    whatever was given.
  treatment_term:
    preferred_term: Supportive Care
    term:
      id: NCIT:C15747
      label: Supportive Care
  evidence:
  - reference: PMID:7741371
    reference_title: "The eosinophilia-myalgia syndrome: status of 205 patients and results of treatment 2 years after onset."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "No other treatment was reported to be consistently beneficial."
    explanation: >-
      INDIRECT, and quoted for what it concedes: with prednisone the only agent
      reported to help and only in the acute phase, supportive care is what
      remains. It is not a study of supportive care.
- name: Rehabilitation
  description: >-
    For the myopathy, the neuropathic disability and the contractures of the
    chronic phase.
  therapeutic_modality: BEHAVIORAL
  treatment_term:
    preferred_term: Rehabilitation
    term:
      id: NCIT:C15315
      label: Rehabilitation
  evidence:
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."
    explanation: >-
      INDIRECT: the source establishes the disabling sequelae that
      rehabilitation addresses, not the efficacy of rehabilitation in this
      disease, which no cited source reports.
prevalence:
- population: United States, 1989-1990 epidemic
  measure_type: CASES_IN_LITERATURE
  prevalence_class: NOT_YET_DOCUMENTED
  notes: >-
    A point-source epidemic that ended with a product recall, so a population
    rate would mislead and no numeric slot is filled. `prevalence_class` is
    NOT_YET_DOCUMENTED because no rate has been documented, not because the
    epidemic went uncounted. Published death tolls differ with the surveillance
    cut-off: 27 by July 1990 and 36 by August 1991. Those are the same quantity
    at two dates and are reported with their dates rather than reconciled.
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "As of July 10, 1990, a total of 1531 cases had been reported nationwide, including 27 deaths."
    explanation: The case count and death toll with the surveillance date attached.
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "the highest rates of reported illness are concentrated in the western states, 68% are non-Hispanic white women aged 35 years and older"
    explanation: >-
      The demographic concentration of reported cases, which tracked supplement
      use rather than any inherited factor.
progression:
- phase: Acute phase
  notes: >-
    Subacute onset of incapacitating generalised myalgia with blood
    eosinophilia, and with rash, peripheral oedema, fever, arthralgia, cough or
    dyspnoea, raised aldolase and abnormal liver tests. Eosinophil granule
    proteins are detectable in serum, urine and tissue.
  evidence:
  - reference: PMID:2398610
    reference_title: Eosinophilia-myalgia syndrome. Results of national surveillance.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The most frequent features include arthralgia (73%), rash (60%), cough or dyspnea (59%), peripheral edema (59%), elevated aldolase level (46%), and elevations in the results of liver function tests (43%)."
    explanation: The acute clinical picture with frequencies, from national surveillance.
- phase: Chronic fibrotic and neuromuscular phase
  notes: >-
    Sclerodermatous skin thickening, sensorimotor polyneuropathy, proximal
    myopathy and episodic myalgia, persisting after the eosinophilia has
    resolved. Nearly all findings improved over eighteen to twenty-four months
    except peripheral neuropathy, and cognitive change was reported worse in
    about a third.
  evidence:
  - reference: PMID:7741371
    reference_title: "The eosinophilia-myalgia syndrome: status of 205 patients and results of treatment 2 years after onset."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "After 18 to 24 months, all symptoms except cognitive changes were reported to have improved in most patients."
    explanation: The direction of the chronic phase and its one named exception.
  - reference: PMID:1673503
    reference_title: Clinical follow-up and immunogenetic studies of 32 patients with eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The chronic sequelae most often associated with long-term disability are sclerodermatous skin thickening (54%), sensorimotor polyneuropathy (61%), proximal myopathy (36%), and severe episodic myalgias (64%)."
    explanation: The four sequelae that disable, with frequencies, in a followed cohort.
- phase: Late outcome
  notes: >-
    Excess mortality concentrated early. All-cause mortality over follow-up was
    19% in definite cases against 3% in exposed-but-not-ill users, and two
    thirds of the deaths in definite cases happened within eighteen months of
    onset. Survivors kept reporting excess morbidity in six major symptoms, with
    severity falling over time.
  evidence:
  - reference: PMID:8624177
    reference_title: The natural history of eosinophilia-myalgia syndrome in a tryptophan-exposed cohort in South Carolina.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "During the follow-up interval, mortality from all causes was 19% in those with definite EMS, 7% in possible EMS, and 3% in those who were not ill."
    explanation: >-
      The mortality gradient across definite, possible and not-ill members of
      one tryptophan-exposed cohort, which is the comparison that makes the
      figure interpretable.
  - reference: PMID:8624177
    reference_title: The natural history of eosinophilia-myalgia syndrome in a tryptophan-exposed cohort in South Carolina.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Six deaths (66%) among the definite EMS case patients occurred during the 18 months immediately after symptom onset."
    explanation: The timing of the excess mortality, which is the prognostic point.
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Older age and involvement of more than one organ system suggest a particularly poor prognosis"
    explanation: The two prognostic factors surveillance data identified.
animal_models:
- name: EBT-treated C57BL/6 mouse
  species: Mouse
  genotype: wild type
  publication: PMID:8163652
  description: >-
    Daily intraperitoneal EBT in female C57BL/6 mice produces inflammation and
    fibrosis of dermis and subcutis including fascia and perimysium, with
    increased and apparently degranulating mast cells, and with type I, III and
    VI collagen gene expression rising alongside dermal TGF-beta 1. It is the
    closest any animal comes to the human fibrosis.
  modeled_mechanisms:
  - target: Fascial and Dermal Fibrosis
    relationship: PARTIALLY_RECAPITULATES
    fidelity: MODERATE
    model_scale: TISSUE
    description: >-
      Reproduces the tissue lesion in the right compartments from the candidate
      compound alone.
    limitations: >-
      Intraperitoneal rather than oral, which is not the route the disease took,
      and pure EBT rather than the contaminant mixture people ate. The mice do
      not develop the blood eosinophilia that defines the human syndrome, so the
      model reproduces the fibrosis without the disease. A review of every
      animal study in this disease concluded the models generally could not be
      reproduced.
    evidence:
    - reference: PMID:8163652
      reference_title: "A murine model of the eosinophilia-myalgia syndrome induced by 1,1'-ethylidenebis (L-tryptophan)."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "We report the development of inflammation and fibrosis affecting the dermis and subcutis, including the fascia and perimyseal tissues, after the daily intraperitoneal administration of EBT to female C57BL/6 mice."
      explanation: >-
        The measured lesion and its tissue distribution, which is what makes
        this link to the fibrosis node rather than to the whole disease.
    - reference: PMID:8163652
      reference_title: "A murine model of the eosinophilia-myalgia syndrome induced by 1,1'-ethylidenebis (L-tryptophan)."
      supports: SUPPORT
      directness: INDIRECT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "This murine model suggests that EBT may have been one of the mediators of EMS and should facilitate studies of the pathogenesis of EMS."
      explanation: >-
        INDIRECT, and the authors' own hedge: "one of the mediators", which is
        the level of claim this model supports.
  - target: Fibroblast Activation and Type I Collagen Overproduction
    relationship: RECAPITULATES
    fidelity: MODERATE
    model_scale: MOLECULAR
    description: >-
      The collagen programme and its TGF-beta 1 association are reproduced in
      mouse skin, which is the human in vitro finding shown in vivo.
    limitations: >-
      Expression measured in whole dermis and subcutis rather than in isolated
      fibroblasts, so the cell of origin is inferred from location.
    evidence:
    - reference: PMID:8863345
      reference_title: A contaminant of L-tryptophan enhances expression of dermal collagen in a murine model of eosinophilia myalgia syndrome.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Increased procollagen gene expression was accompanied by evidence of enhanced TGF-beta 1 expression in the dermis and subcutis."
      explanation: >-
        Puts the collagen and TGF-beta 1 findings together in the same tissue,
        which is what the human node asserts.
  evidence:
  - reference: PMID:8163652
    reference_title: "A murine model of the eosinophilia-myalgia syndrome induced by 1,1'-ethylidenebis (L-tryptophan)."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "Such changes are accompanied by increased numbers of mast cells, many of which appear to be degranulating."
    explanation: >-
      The mast cell finding, and the degranulation, which the mouse shows more
      clearly than any human specimen. Its human counterpart is now curated on
      `Inflammatory Infiltration of Fascia, Perimysium and Dermis` from
      `PMID:2273104`, which reports numerous mast cells in some patients. An
      earlier draft of this explanation said the finding had no human
      counterpart in the entry, which was true of the entry and not of the
      literature.
- name: EBT-treated Lewis rat
  species: Rat
  genotype: wild type
  publication: PMID:8450062
  description: >-
    Lewis rats given case-associated L-tryptophan or synthetic EBT develop
    myofascial thickening, and a separate study found perimysial and fascial
    thickening with lymphocytes, macrophages and sparse eosinophils plus sparse
    perineurial infiltrate. Control L-tryptophan produced a mild but significant
    thickening of its own.
  modeled_mechanisms:
  - target: Inflammatory Infiltration of Fascia, Perimysium and Dermis
    relationship: PARTIALLY_RECAPITULATES
    fidelity: LOW
    model_scale: TISSUE
    description: >-
      Reproduces the fascial and perimysial infiltrate, including the sparse
      perineurial component, from the candidate compound.
    limitations: >-
      No rash and no weakness in either group, so the clinical syndrome is
      absent. Control L-tryptophan also thickened the myofascia significantly
      against vehicle, which weakens the attribution to the contaminant, and the
      authors say their own results do not exclude other impurities. The second
      study found the changes in two of three treated rats.
    evidence:
    - reference: PMID:7991132
      reference_title: "1,1'-Ethylidenebis[tryptophan] induces pathologic alterations in muscle similar to those observed in the eosinophilia-myalgia syndrome."
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "In two rats, perimysium and fascia were abnormally thickened and infiltrated with lymphocytes, macrophages, and sparse eosinophils; two rats had sparse perineurial inflammatory cells."
      explanation: >-
        The infiltrate and its compartments, matching the human node this link
        targets, with the animal counts stated.
    - reference: PMID:7991132
      reference_title: "1,1'-Ethylidenebis[tryptophan] induces pathologic alterations in muscle similar to those observed in the eosinophilia-myalgia syndrome."
      supports: REFUTE
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "No rash or weakness occurred in either group."
      explanation: >-
        REFUTE against this being a model of the disease. The tissue lesion
        appears and the illness does not, which is why fidelity is LOW.
  evidence:
  - reference: PMID:8450062
    reference_title: "Pathological and immunological effects of ingesting L-tryptophan and 1,1'-ethylidenebis (L-tryptophan) in Lewis rats."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "All animals treated for 6 wk with case-associated L-TRP or EBT developed significant myofascial thickening, compared with animals in the vehicle control and control L-TRP groups."
    explanation: The primary result, with its comparison groups.
  - reference: PMID:8450062
    reference_title: "Pathological and immunological effects of ingesting L-tryptophan and 1,1'-ethylidenebis (L-tryptophan) in Lewis rats."
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "they do not rule out the possibility that other impurities in the EMS-case-associated L-TRP may also contribute to some of the features of EMS"
    explanation: >-
      INDIRECT, and the authors' own limit on what their positive result
      licenses. It is the reason the exposure node names contaminants in the
      plural.
discussions:
- kind: KNOWLEDGE_GAP
  discussion_id: unidentified_etiologic_agent
  prompt: >-
    Which constituent of the implicated L-tryptophan caused eosinophilia-myalgia
    syndrome, and by what molecular route does it start the disease?
  attaches_to:
  - pathophysiology#Systemic Exposure to L-Tryptophan Process Contaminants
  - pathophysiology#Type 2 Cytokine Response to Peak E
  rationale: >-
    The exposure is not in doubt and the agent is. Six compounds were associated
    with case lots. EBT has the most work behind it and is the only one with
    animal and fibroblast data, and its epidemiological association does not
    survive stratification by date of manufacture. Nothing traces a molecular
    route from any candidate compound to the first lesion, which is why the edge
    out of the exposure node carries unknown intermediates.
  evidence:
  - reference: PMID:8895185
    reference_title: Animal models of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "The precise cause of EMS remains unknown."
    explanation: >-
      States the gap, in the closing sentence of a review of every animal study
      in the disease. Graded OTHER rather than MODEL_ORGANISM because the
      sentence is a statement about the human disease's etiology, not an animal
      result; the same reference's two other snippets in this entry are animal
      findings and stay MODEL_ORGANISM.
  - reference: PMID:37453474
    reference_title: Safety concerns regarding impurities in L-Tryptophan associated with eosinophilia myalgia syndrome.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Many in vitro and in vivo studies have been conducted to assess the putative correlation between impurities in L-tryptophan preparations and EMS, but no clear and convincing conclusions have been drawn so far."
    explanation: >-
      The same gap thirty years later, after the experimental work the earlier
      review was asking for.
- kind: HUMAN_MODEL_MISMATCH
  discussion_id: animal_models_lack_eosinophilia
  prompt: >-
    Why does EBT produce the fibrosis of this disease in mice and rats without
    producing the eosinophilia that defines it, and what host factor is missing?
  attaches_to:
  - animal_models#Mouse
  - animal_models#Rat
  - pathophysiology#Eosinophil Expansion with Tissue Degranulation
  rationale: >-
    This is a mismatch and not an absence of evidence. Both animal models
    reproduce the fascial and perimysial lesion, and neither produces the blood
    eosinophilia, the rash or the weakness. So the compound can make the
    fibrosis without making the disease, and the eosinophil arm of the human
    pathograph has no animal counterpart. A critical review of the whole animal
    literature went further and doubted the models could be reproduced at all,
    which is why every fidelity in this entry is MODERATE or lower.
  evidence:
  - reference: PMID:8895185
    reference_title: Animal models of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: REVIEW_SYNTHESIS
    snippet: "However, a critical review of all the animal studies casts doubt on the validity of these assertions."
    explanation: >-
      The reviewers' verdict on the claim that these are models of the syndrome,
      which is the mismatch this discussion records.
  - reference: PMID:8895185
    reference_title: Animal models of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: REVIEW_SYNTHESIS
    snippet: "These studies could generally not be reproduced and had methodologic flaws that limited extrapolation of results."
    explanation: The stated basis of that verdict.
  - reference: PMID:7991132
    reference_title: "1,1'-Ethylidenebis[tryptophan] induces pathologic alterations in muscle similar to those observed in the eosinophilia-myalgia syndrome."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "No rash or weakness occurred in either group."
    explanation: >-
      The concrete instance of the mismatch: the tissue lesion without the
      illness, in the study whose own conclusion is that it replicates an
      important feature of the human disease.
- kind: KNOWLEDGE_GAP
  discussion_id: myalgia_without_myofiber_injury
  prompt: >-
    What produces the incapacitating myalgia of this disease, given that muscle
    shows minimal myofiber atrophy, regeneration or necrosis?
  attaches_to:
  - phenotypes#Myalgia
  - pathophysiology#Inflammatory Infiltration of Fascia, Perimysium and Dermis
  rationale: >-
    Myalgia is one of the two features that define the syndrome, and the muscle
    pathology does not account for it. Eleven biopsied cases showed minimal
    myofiber damage despite severe myalgia in almost all of them, and the
    inflammatory infiltrate sits in perimysium, epimysium and fascia rather than
    in the fibres. No cited source asserts a mechanism, so `Myalgia` has no
    incoming causal edge in this entry. Toxic oil syndrome records the same gap
    for the same reason.
  evidence:
  - reference: PMID:1563745
    reference_title: "Pathologic manifestations of the eosinophilia myalgia syndrome: analysis of 11 cases."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Minimal myofiber atrophy, regeneration, or necrosis was seen despite the clinical history of severe myalgias in almost all patients."
    explanation: >-
      The authors state the dissociation themselves, in the same sentence as the
      clinical severity.
  - reference: PMID:8895179
    reference_title: Pathophysiology of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "The pathophysiology of the chronic symptoms is poorly understood but may involve ischemia, neuropathy, and metabolic abnormalities."
    explanation: >-
      The three candidate routes named at review level, none of which any cited
      source ties to the myalgia.
- kind: KNOWLEDGE_GAP
  discussion_id: tryptophan_metabolism_abnormality
  prompt: >-
    Is the abnormal tryptophan metabolism reported in EMS patients a step in the
    disease, or a consequence of the inflammation?
  attaches_to:
  - pathophysiology#Systemic Exposure to L-Tryptophan Process Contaminants
  - pathophysiology#Type 2 Cytokine Response to Peak E
  rationale: >-
    Patients show abnormal handling of tryptophan, which a contemporary review
    reads as increased activity of indoleamine 2,3-dioxygenase, the rate-limiting
    enzyme of the kynurenine pathway. IDO is induced by inflammatory cytokines,
    so the finding sits on both sides of the question and no cited source settles
    which. Nothing is wired into the pathograph for it, and that is the point of
    recording the gap: an entry about a tryptophan-derived contaminant has an
    obvious temptation to draw an arrow here.
  evidence:
  - reference: PMID:8423409
    reference_title: "L-tryptophan and the eosinophilia-myalgia syndrome: current understanding of the etiology and pathogenesis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Evidence of abnormal L-tryptophan metabolism has been described in patients with EMS, and most likely reflects increased activity of indoleamine 2,3-dioxygenase, the rate-limiting enzyme of tryptophan metabolism."
    explanation: >-
      The finding and the proposed enzyme, at review level. "Most likely
      reflects" is the review's own hedge and is why this is a gap rather than a
      node.
- kind: CONTROVERSY
  discussion_id: case_definition_validity
  prompt: >-
    Does the 1989 surveillance case definition identify the disease, or does it
    select a severe subset and make its two required features unmeasurable?
  attaches_to:
  - phenotypes#Myalgia
  - phenotypes#Peripheral eosinophilia
  rationale: >-
    Every frequency in this literature is conditioned on a definition that was
    built for outbreak surveillance and never validated. Myalgia and
    eosinophilia are reported at 100% because the definition requires them, and
    milder illness without one of them falls outside the counted population.
    This is why the two definitional phenotypes here carry no `frequency` at all
    and every other frequency names its cohort.
  evidence:
  - reference: PMID:8895176
    reference_title: Criteria for the definition of the eosinophilia-myalgia syndrome.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "The widely disseminated surveillance case definition of the eosinophilia-myalgia syndrome (EMS) recommended by the Centers for Disease Control in 1989 has never been validated by an appropriate challenge and has commonly been used for unintended purposes."
    explanation: >-
      States both halves of the problem: never validated, and used for things it
      was not built for.
- kind: OPEN_QUESTION
  discussion_id: five_htp_related_illness
  prompt: >-
    Is the illness that followed contaminated L-5-hydroxytryptophan the same
    disease as the 1989 L-tryptophan epidemic?
  attaches_to:
  - pathophysiology#Systemic Exposure to L-Tryptophan Process Contaminants
  rationale: >-
    One family became ill after taking L-5-HTP that contained an impurity absent
    from comparison samples. One member met criteria for the syndrome and two
    others had eosinophilia that resolved when the product was replaced. An NIH
    observational study was built around such cases. Whether this is the same
    disease reached through a different supplement is unsettled, and this entry
    stays scoped to the L-tryptophan epidemic rather than deciding it.
  evidence:
  - reference: PMID:7699627
    reference_title: An eosinophilia-myalgia syndrome related disorder associated with exposure to L-5-hydroxytryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The L-5-HTP used by the family contained an impurity not present in samples from the other patient group."
    explanation: The chemical finding that makes the parallel worth recording.
  - reference: PMID:7699627
    reference_title: An eosinophilia-myalgia syndrome related disorder associated with exposure to L-5-hydroxytryptophan.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "One member of the family had EMS, and 2 others had eosinophilia."
    explanation: >-
      The clinical extent, and its limit. One case meeting criteria is why the
      title of that paper says "related disorder".
  - reference: clinicaltrials:NCT00001918
    reference_title: "L-5-Hydroxy-Tryptophan-Related Eosinophilia-Myalgia Syndrome (EMS): Clinical Patient Evaluation"
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: BACKGROUND
    snippet: "More recently, similar impurities have been detected in batches of a similar dietary supplement called L-5-hydroxytryptophan."
    explanation: >-
      An NIH study description recording that the same class of impurity turned
      up in 5-HTP. Graded OTHER because a trial registration document is not
      study evidence.
clinical_trials:
- name: NCT00001918
  phase: NOT_APPLICABLE
  status: COMPLETED
  description: >-
    An NIH observational study evaluating patients who developed
    eosinophilia-myalgia syndrome after taking L-5-hydroxytryptophan, with
    clinical, neurologic, psychiatric, imaging and laboratory assessment and
    chemical analysis of the patients' own supplement samples. Not a treatment
    trial.
  target_phenotypes:
  - preferred_term: Myalgia
    term:
      id: HP:0003326
      label: Myalgia
  - preferred_term: Increased total eosinophil count
    term:
      id: HP:0001880
      label: Increased total eosinophil count
  evidence:
  - reference: clinicaltrials:NCT00001918
    reference_title: "L-5-Hydroxy-Tryptophan-Related Eosinophilia-Myalgia Syndrome (EMS): Clinical Patient Evaluation"
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "This study is designed to learn more about EMS that develops in patients taking L-5-hydroxytryptophan."
    explanation: >-
      The study's own statement of purpose. Graded OTHER because a registration
      record is a description of a study rather than evidence from one.
clinical_burden:
  burden_level: HIGH
  rationale: >-
    Mortality concentrated early and ran through the neuromuscular disease. Over
    follow-up of a tryptophan-exposed cohort, all-cause mortality was 19% in
    definite cases against 3% in exposed users who did not fall ill, and two
    thirds of those deaths came within eighteen months of onset. Of 36 deaths
    reported to national surveillance, 92% had neuromuscular sequelae and the
    commonest fatal process was progressive polyneuropathy and myopathy leading
    to pneumonia, sepsis or respiratory failure. Chronic morbidity in survivors
    is the larger burden: survivors of definite disease kept reporting excess
    myalgia, arthralgia, weakness, rash, alopecia and sclerodermiform skin
    change against exposed controls, with severity diminishing over time.
  evidence:
  - reference: PMID:8295183
    reference_title: "Eosinophilia-myalgia syndrome: mortality data from the US national surveillance system."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Of the 36 patients who died, 33 (92%) had neuromuscular sequelae, 29 (81%) had pulmonary complications, and 23 (64%) had cardiac manifestations."
    explanation: The organ involvement among the deaths, counted.
  - reference: PMID:8624177
    reference_title: The natural history of eosinophilia-myalgia syndrome in a tryptophan-exposed cohort in South Carolina.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "survivors with definite EMS continued to report excess morbidity for 6 major EMS symptoms (myalgia, arthralgia, weakness, rash, alopecia, and sclerodermiform skin changes)"
    explanation: >-
      The six persisting symptoms, measured against exposed users who did not
      fall ill rather than against the general population.
datasets:
- accession: geo:GSE26934
  title: Post-epidemic eosinophilia myalgia syndrome associated with L-Tryptophan
  description: >-
    Expression profiling by array of lesional skin from the 2011 post-epidemic
    case, against comparison samples. This is the source of the TGF-beta and
    IL-4 signalling signature reported in PMID:21702023, and it is the only
    EMS-specific dataset in GEO.
  organism:
    preferred_term: human
    term:
      id: NCBITaxon:9606
      label: Homo sapiens
  data_type: MICROARRAY
  sample_count: 6
  conditions:
  - lesional skin from L-tryptophan-associated eosinophilia-myalgia syndrome
  publication: PMID:21702023
  evidence:
  - reference: GEO:GSE26934
    reference_title: Post-epidemic eosinophilia myalgia syndrome associated with L-Tryptophan
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: BACKGROUND
    snippet: "The EMS epidemic in 1989 was linked to L-tryptophan consumption originating from a single source."
    explanation: >-
      The GEO record's own summary establishes the series is about this disease
      and this exposure, which is the relevance check an accession resolving
      does not supply.
  notes: >-
    Six samples from a single post-epidemic case and its comparisons, so the
    series is a case-level profiling experiment rather than a cohort. The
    signature it carries is curated on `Fibroblast Activation and Type I
    Collagen Overproduction` as a `PMID:21702023` evidence item; an earlier
    draft of this note said so before that item existed, which made the note a
    false statement about the entry. The evidence item quotes the publication
    rather than this GEO record because the GEO summary states the disease and
    the exposure and not the result.

references:
- reference: PMID:2314421
  title: Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.
- reference: PMID:2370887
  title: An investigation of the cause of the eosinophilia-myalgia syndrome associated with tryptophan use.
- reference: PMID:2398610
  title: Eosinophilia-myalgia syndrome. Results of national surveillance.
- reference: PMID:8895185
  title: Animal models of the eosinophilia-myalgia syndrome.
- reference: PMID:37453474
  title: Safety concerns regarding impurities in L-Tryptophan associated with eosinophilia myalgia syndrome.

📚

References & Deep Research

References

5

PMID:2314421

Association of the eosinophilia-myalgia syndrome with the ingestion of tryptophan.

No top-level findings curated for this source.

PMID:2370887

An investigation of the cause of the eosinophilia-myalgia syndrome associated with tryptophan use.

No top-level findings curated for this source.

PMID:2398610

Eosinophilia-myalgia syndrome. Results of national surveillance.

No top-level findings curated for this source.

PMID:8895185

Animal models of the eosinophilia-myalgia syndrome.

No top-level findings curated for this source.

PMID:37453474

Safety concerns regarding impurities in L-Tryptophan associated with eosinophilia myalgia syndrome.

No top-level findings curated for this source.

Deep Research

1

Deep research results are used as seeds for research; they do not undergo the same validation as the main records and may contain errors. How we use deep research.

Evaluations and curation notes (3)

Record notes

**The etiologic agent is not established and this entry does not assert one.** The pathograph names the L-tryptophan process contaminants collectively and carries EBT as the worked candidate, because EBT is the compound with in vitro, murine and rat data behind it. `PMID:8356958` is curated as REFUTE against EBT as the sole agent: once case and non-case lots are stratified by time of manufacture the EBT association loses significance, and the authors raise a distinct compound as a live possibility. Both readings are in the entry, and the `unidentified_etiologic_agent` discussion states what is missing. **Myalgia and peripheral eosinophilia deliberately carry no `frequency`.** Both are reported at 100% in every series, and that figure is circular: the CDC surveillance definition requires them, so a cohort assembled under it cannot report anything else. `PMID:8895176` records that the definition was never validated and was routinely used for purposes it was not built for. Every other phenotype's frequency comes from a named cohort and says which. **The severity of the myalgia is not explained by the muscle pathology, and no edge asserts that it is.** `PMID:1563745` found minimal myofiber atrophy, regeneration or necrosis in 11 cases despite severe myalgia in almost all of them, and `PMID:8100551` found only rare degranulating eosinophils in fascia and perimysium that were nonetheless full of CD8+ cells and macrophages. The `myalgia_without_myofiber_injury` discussion carries the gap. This is the same gap the `Toxic_Oil_Syndrome` entry records for its own myalgia, and the two diseases are routinely discussed together for that reason. **Relation to toxic oil syndrome.** The two are separate epidemics with separate vehicles, and they are linked by chemistry rather than by exposure: PAP from the Spanish oil and PAA from the implicated L-tryptophan converge on a shared metabolite. That link is curated here as its own pathophysiology node with its own evidence (`PMID:8555405`, `PMID:17892268`). No toxic oil syndrome clinical finding is imported as an EMS finding. **A related illness followed L-5-hydroxytryptophan**, and it is not curated as EMS. `PMID:7699627` reports one family where one member met criteria for EMS and two had eosinophilia, with an impurity present in their 5-HTP and absent from comparison samples, and the NIH study `NCT00001918` was built around such cases. The `five_htp_related_illness` discussion records it; the disease entry stays scoped to the L-tryptophan epidemic. **Most phenotypes here are deliberately unwired, and connectivity is correspondingly low.** Run `just list-disconnected-phenotypes kb/disorders/Eosinophilia-Myalgia_Syndrome.yaml` for the current set rather than trusting a count written here. The rule is the one the rest of this entry follows: an edge goes in only where a cited source makes the causal link. The constitutional features (fever, fatigue, oral ulcer) have no named lesion in any cited source; the raised aldolase and transaminases are readouts nothing here explains, and are attached observationally through `reports_on` rather than causally; `Proximal myopathy` and `Myalgia` sit on the muscle-pathology gap the `myalgia_without_myofiber_injury` discussion states; the cognitive findings have two competing readings in the one paper that imaged them; and `Xerostomia` and `Dysphagia` are self-reported survey figures with no cited study of salivary flow or swallowing. An edge added to raise that figure would be worse than the gap. **The pulmonary phenotypes are wired, with one exception that is not.** `Pulmonary Eosinophilic and Lymphocytic Inflammation` was added in review after the lung turned out to be the one organ system with phenotypes and no node, which was an omission rather than a scoping decision: 81% of the deaths reported to surveillance had pulmonary complications. `Dyspnea` and `Decreased DLCO` hang off it. `Pulmonary arterial hypertension` does not, and that one is deliberate: it rests on a single patient in a six-patient series, and no cited source connects the inflammation to vascular remodelling. **Three features the deep-research report reports are not curated, for one reason.** Weight loss at 50%, and the Maryland series' own frequencies, come from full-text pages that no cached reference here holds, so there is no quotable sentence for them. Paresthesia and xerostomia were in the same position until a further PubMed pass found cohort abstracts that state them, and both are now curated. Weight loss is still uncited and is therefore still absent, rather than carried on a number with no source behind it. **There is no `histopathology:` or `diagnosis:` section, deliberately.** The histology of this disease is dense and well reported, and it is carried as pathophysiology nodes because in EMS the histology *is* the mechanism: the septal and fascial infiltrate, the homogenised dermal collagen, the conduction-system fibrosis. Splitting it into a parallel descriptive section would state each finding twice and put the causal reading in only one of them. The case definition, which is the part of the diagnostic picture that carries real weight here, is curated as a `definitions:` entry instead, and the `case_definition_validity` discussion argues its consequences. No GeneReviews chapter exists and none is expected. This is an acquired point-source intoxication with no Mendelian basis; `just check-genereviews` returns NO_CHAPTER for both Bookshelf collections.

Review round 1: wire the pulmonary branch (PR #12853) · 2026-09-25T20:44:24Z · View source

Addresses the ai4c-reviewer CHANGES_REQUESTED review on PR #12853. One blocking finding and four suggestions; all are answered in a single push. Blocking: the lung had phenotypes and no pathophysiology node. Dyspnea, Decreased DLCO and Pulmonary arterial hypertension had no incoming edge, and the entry notes listed which gaps were deliberate without listing those three. The reviewer was right that this read as an omission rather than a scoping decision, and it was one: the entry's own clinical_burden quotes 81% of the 36 reported deaths as having pulmonary complications. Added Pulmonary Eosinophilic and Lymphocytic Inflammation, a TISSUE node downstream of both Eosinophil Expansion with Tissue Degranulation and T Cell Mediated Immunity Against Extracellular Matrix, and upstream of Dyspnea and Decreased DLCO. It is grounded in four quotes from PMID:1582284, which was already cached and already cited: BAL eosinophils in two patients and lymphocytes in a third, raised CD8+ proportion in BAL, fibroblast proliferation-stimulating activity in BAL that fell to normal on corticosteroid, and abnormal HRCT in four patients. GO:0048144 fibroblast proliferation is bound here as well as on the dermal node, because the lavage measurement is the one figure anywhere in the entry that reverses under treatment. Pulmonary arterial hypertension is deliberately left unwired: it rests on one patient in a six-patient series and no cited source connects the inflammation to vascular remodelling. The entry notes now say all of this, including why the one exception is an exception. Suggestion 2 accepted, and it reverses part of the pre-PR self-review. GO:0032604 granulocyte macrophage colony-stimulating factor production was dropped from the Type 2 Cytokine Response to Peak E node. The reviewer's argument is the entry's own argument turned around: GM-CSF is not a type 2 cytokine, PMID:7797795 attributes the response to endotoxin rather than to peak E, and the REFUTE evidence item attaches to the node regardless, which was the reason the descriptor was never needed. The node now binds GO:0032634 alone and its notes record all three attempts, because the first two were each wrong in ways no validator could see. Suggestion 1 accepted after a PubMed pass. The reviewer noted that weight loss, paresthesia and xerostomia appear in the deep-research report with frequencies and in no cached abstract, and asked for one search. Three cohort abstracts state them: PMID:1520057 (Oregon population-based, 53 interviewed), PMID:8129767 (57 patients prospective, mean 36 months) and PMID:7857025 (head and neck questionnaire across 33 states). Added Paresthesia (HP:0003401, FREQUENT, 62% with objectively demonstrated hypesthesia), Xerostomia (HP:0000217, FREQUENT, 66%) and Dysphagia (HP:0002015, FREQUENT, 56%). The same abstracts finally put sourced frequencies on three records that had none: Fatigue FREQUENT, Muscle weakness FREQUENT and Cognitive impairment VERY_FREQUENT. Fatigue takes the population-based 77% rather than the referral cohort's 91%, which would have crossed into VERY_FREQUENT, and the explanation says so. Weight loss stays absent: its 50% figure is still unquotable from anything cached, and the notes record that rather than carrying a number with no source. Suggestion 4 accepted in narrowed form. Added a diagnosis section with three entries limited to what a cached abstract states: high-resolution CT (abnormal in four of six, with one normal plain radiograph, so a normal film does not exclude), electrodiagnostic studies (the multifocal conduction block and prolonged F-responses that make the neuropathy demyelinating), and muscle biopsy (eosinophilic perimyositis in four of five biopsied, and the septal and fascial location of the earliest change, which is why sampling depth matters). Its notes record that the wider workup the report describes, including muscle MRI and parasite exclusion, is general practice no cached reference attributes to a study of this disease, so it is not written down as though it were. Suggestion 3 accepted: two errors in the first history record for this entry were corrected before merge, as the reviewer asked. It said 'five discussions' where there are six, and it stated the NCIT:C15747 binding in the present tense in one paragraph while a later paragraph described correcting it to NCIT:C128535. The node count in the same sentence was also stale and is now 13 plus the deliberately isolated PAP node. Validation: 'just validate-disorders' passes with 159/159 snippets verified, 165 titles checked and 0 issues. Green on check-entity-refs, check-causal-targets, check-duplicate-keys, check-enum-values, check-qualifier-terms, check-folded-hyphens, check-snippet-length, check-title-snippets, check-snippet-grading, check-environmental-evidence, check-reference-titles, check-coarse-phenotypes, check-case-collisions, check-term-cache-integrity and check-cache-order. Three new reference cache files committed (PMID_1520057, PMID_8129767, PMID_7857025), all cited. Causal connectivity rises from 7/22 to 9/25 phenotypes.

Create: Eosinophilia-Myalgia Syndrome · 2026-09-25T18:44:28Z · View source

Created kb/disorders/Eosinophilia-Myalgia_Syndrome.yaml (MONDO:0004941) de novo. Deep research: Edison Scientific falcon, via 'just research-disorder falcon Eosinophilia-Myalgia_Syndrome'. The first attempt failed at authentication because the connection to api.platform.edisonscientific.com timed out before the key was judged; the second run of the same command succeeded. The report is committed at research/Eosinophilia-Myalgia_Syndrome-deep-research-falcon.md with its citations sidecar and one artifact. The run exited 3 because term validation could not reach EBI OLS (read timeout resolving MeSH:D016603), so no frontmatter validation block was written; both sections were retro-fitted afterwards with 'just validate-research-reference' and 'just validate-research-terms'. The reference retro-fit reports 9/9 resolved, 0 unresolved, and flags DOI:10.2903/j.efsa.2024.8707 (an EFSA feed-additive opinion) as possibly off topic; that reference is not cited in the entry. The term retro-fit failed again on an OLS timeout, so the report carries no Term Validation section and every CURIE in this entry was instead looked up against the committed caches or via runoak at the point of writing. Falcon cites by author-year key with DOI links and no PMIDs, so its nine sources were mapped DOI to PMID through esearch and fetched. The report's coverage is thin by design (7 papers plus one trial record), so it was used for framing and the entry's evidence base was built out by PubMed search: 44 distinct references are cited, 42 of whose cache files are new here; PMID:8555405 and PMID:8285738 were already committed for Toxic_Oil_Syndrome. Eighteen further fetches, including the eight DOI records the report's reference validation resolved, went uncited and were pruned rather than committed. Pathophysiology is a connected chain of 13 nodes from ingestion of contaminated manufactured L-tryptophan through contaminant exposure, a type 2 cytokine response, eosinophil chemotaxis and degranulation, a T cell process against extracellular matrix, tissue infiltration, pulmonary inflammation, microangiopathy, fibroblast activation, fascial and dermal fibrosis, perineural injury and cardiac conduction-system fibrosis, plus a fourteenth node for the shared PAP metabolite that is deliberately isolated. Three mechanistic_hypotheses groups (contaminant-driven type 2 immunity as CANONICAL, direct fibroblast activation and T cell matrix autoimmunity as ALTERNATIVE) and six discussions record what is unsettled. Deliberate non-assertions, each recorded in the entry: the etiologic agent is not named (PMID:8356958 is carried as REFUTE against EBT as sole agent); myalgia and peripheral eosinophilia carry no frequency because the CDC surveillance definition requires them and PMID:8895176 records that the definition was never validated; myalgia has no incoming causal edge because PMID:1563745 found minimal myofiber damage despite severe myalgia; the PAP-to-PAA node has no downstream edge because nothing traces a route from the shared metabolite to any lesion; and 15 of 22 phenotypes are unwired for the same reason, which is stated in the entry notes rather than papered over with edges. One binding is a compromise and says so in place: environmental[0].exposure_term binds ECTO:9002126 (exposure to environmental food contaminant) rather than ECTO:9002895 (exposure to tryptophan), because the latter names the amino acid, which the epidemiology rules out as the cause, and it also carries no label in the committed ECTO build so it is not in the ExposureTerm enum. The note names the searches run. The withdrawal treatment binds NCIT:C128535 Pharmacotherapy Discontinuation; the paragraph below records that an earlier draft of it was bound to NCIT:C15747 Supportive Care on a false justification, and that the rebinding happened before this record was committed. Datasets: 'just discover-datasets' returned only GENE_ONLY hits on HLA-DRB1, none about this disease, so GEO was searched directly; geo:GSE26934 is the one EMS-specific series and is curated with its cache file and a quoted summary. Validation: 'just validate-disorders' passes with 140/140 snippets verified, 146 titles checked and 0 issues. 'just verify-datasets' OK 1/1. Green on check-entity-refs, check-causal-targets, check-duplicate-keys, check-enum-values, check-qualifier-terms, check-folded-hyphens, check-snippet-length, check-title-snippets, check-snippet-grading, check-environmental-evidence, check-reference-titles, check-case-collisions, check-coarse-phenotypes and check-delivery-system. 'just check-genereviews --online' returns NO_CHAPTER for both Bookshelf collections, which is what the entry notes claim. 7 of 22 phenotypes are causally connected, which is stated in the entry notes rather than raised by adding edges. A self-review round was run before the PR, with a subagent applying the dismech-pr-review skill against the uncommitted entry, and its blocking findings were fixed in this same commit. Two of them were false justification notes of the exact shape CLAUDE.md warns about, both written by me and both refuted by re-running the search the note named. The treatment for supplement withdrawal claimed no reachable NCIT term for stopping a causative agent exists; 'runoak -i sqlite:obo:ncit search "l~discontinuation"' returns NCIT:C128535 Pharmacotherapy Discontinuation, which is-a C21090 is-a C49236 is-a C25218, so it is in the TreatmentTerm enum, and the treatment is now bound to it. The false note had searched only the committed cache/ncit/terms.csv, where an absent row is silence rather than a negative answer. A dataset note said the TGF-beta signature was curated on the fibroblast node through PMID:21702023 when that reference was nowhere on that node; the evidence item now exists and the note records the correction. The ECTO note survived re-running: ECTO:9002895 and ECTO:0070245 both resolve to no label in the committed build, and a search for supplement exposure returns only substance-specific classes. Four further blocking fixes. The Type 2 Cytokine node bound GO:0032604 granulocyte macrophage colony-stimulating factor production under a preferred_term of interleukin-5 production, defended by a note arguing the node had to carry the GM-CSF finding for an evidence item to attach to; an evidence item attaches to the node, not to a process descriptor, and biological_processes is multivalued, so GO:0032634 is now bound alongside it. The Peripheral neuropathy clinical_course PROGRESSIVE was cited against a snippet reading 'only peripheral neuropathy was unchanged'; the surveillance 'progressive polyneuropathy' item now carries the qualifier and the older explanation says what its sentence supports. Two denominators in Decreased DLCO and Dyspnea were asserted beyond the source and are corrected ('five of the six reported', 'two of the three who underwent lavage'). Sudden cardiac death now quotes the sentence naming the two arrhythmic deaths rather than the generic cardiac-involvement figure. Non-blocking items taken in the same push, per CLAUDE.md's one-push rule: a sourced upstream edge from the contaminant exposure to the T cell node, which was the entry's second unexplained root; deletion of the UNKNOWN edge into the PAP metabolite node, whose own description disclaimed the causation the arrow asserted, leaving that node deliberately isolated; PMID:8285738 added for TGF-beta in perimysial connective tissue and one sural nerve, which is the only molecular observation in a peripheral nerve anywhere in the cited literature; a mast cell binding and human counterpart from PMID:2273104; a Steroid-Sparing Immunosuppression treatment from the 2011 case, carrying the methotrexate and anakinra failure as REFUTE; a definitions entry for the CDC 1989 surveillance case definition with validation_status UNVALIDATED; a tryptophan_metabolism_abnormality knowledge gap for IDO; reports_on readout links for the two laboratory phenotypes; the HLA-DRB1 implicated-lot qualification, where DRB1*03 was not a risk factor; and three grading corrections. The DR4 trend item was regraded from REFUTE to SUPPORT with directness INDIRECT, because a non-significant trend in the same direction is a failure to confirm rather than a contradiction. Three reviewer findings were declined. There is no differentials slot on the Disease class and no MeSH slot under mappings, so those two suggestions have nowhere to go. No histopathology or diagnosis section was added, and the entry notes now record why: in this disease the histology is the mechanism and is carried as pathophysiology nodes, so a parallel descriptive section would state each finding twice.

Falcon ▸

Eosinophilia–Myalgia Syndrome: Disease Characteristics Research Report

Edison Scientific Literature 35 citations 2026-09-25T12:54:34.067990

Eosinophilia–Myalgia Syndrome: Disease Characteristics Research Report

Executive summary

Eosinophilia–myalgia syndrome (EMS) is a rare, acquired, noninfectious multisystem inflammatory and fibrosing disease. Its defining presentation is subacute, disabling generalized myalgia with peripheral eosinophilia, often followed by edema, rash, fasciitis, scleroderma-like skin induration, myopathy, and peripheral neuropathy. The major epidemic began in 1989 and was strongly associated with manufactured L-tryptophan supplements—particularly contaminated lots from one producer—not with an inherited mutation. Approximately 1,500 cases were officially reported; mortality estimates vary with surveillance date and source from 27 to about 40 deaths. Current cases are exceptionally rare, but recurrence remains possible with contaminated L-tryptophan or 5-hydroxytryptophan products. (varga1993ltryptophanandthe pages 1-2, allen2011postepidemiceosinophiliamyalgiasyndrome pages 1-2, NCT00001918 chunk 1, silver1994amurinemodel pages 1-3)

The evidence base is unusual: most EMS-specific clinical and mechanistic knowledge comes from the 1989–1990 epidemic and subsequent 1990s studies. Research published in 2023–2024 has concentrated on impurity detection and product safety rather than new patients or therapies. There is no validated disease-specific biomarker, causal gene, approved EMS-specific drug, randomized therapeutic trial, or contemporary incidence estimate. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6, NCT00001918 chunk 1, lee2023simultaneousdeterminationof pages 1-2, bampidis2024safetyandefficacy pages 1-2)

Knowledge-base field	Eosinophilia-Myalgia Syndrome summary	Ontology suggestions
Definition/category	Rare, environmentally acquired multisystem inflammatory and fibrosing disorder characterized by severe generalized myalgia, peripheral eosinophilia, and later fasciitis, skin induration, myopathy, or neuropathy; principally associated with contaminated manufactured L-tryptophan.	MONDO:0004941; MeSH:D016603; category: environmental/toxic exposure disease
Identifiers	MONDO and MeSH identifiers are established above. No disease-specific OMIM entry is expected because EMS is not Mendelian; Orphanet and dedicated ICD-10/ICD-11 identifiers were not confirmed and should not be inferred.	MONDO:0004941; MeSH:D016603
Trigger	Oral L-tryptophan supplements, especially 1989 lots from one manufacturer containing more than 60 impurities. Epidemiologically associated candidates include EBT (“peak E”), PAA, IMT, PIC, HIT, and AAA, but no single contaminant has been proved to be solely causal.	L-tryptophan—CHEBI:16828; EBT/PAA disease-exposure CHEBI identifiers: not confirmed (varga1993ltryptophanandthe pages 4-5, lee2023simultaneousdeterminationof pages 1-2)
Diagnostic core	Historical CDC surveillance definition: blood eosinophils >1,000/mm³, generalized myalgia severe enough to limit usual activity, and exclusion of infection or neoplasm. The definition was intended for surveillance, not as a stand-alone clinical diagnostic standard; revised 2001 criteria reportedly achieved 97% specificity.	Eosinophilia—HP:0001880; Myalgia—HP:0003326 (silver1994amurinemodel pages 1-3, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)
Major acute phenotypes	Myalgia 100%; rash 71%; edema 52%; fever 41%; arthralgia 35%; respiratory manifestations 32% in a summarized outbreak series. A Maryland series reported joint pain 53%, rash and extremity edema 47% each, dyspnea 33%, and fever 27%. Frequencies vary by cohort and ascertainment.	HP:0003326 Myalgia; HP:0000988 Skin rash; HP:0000969 Edema; HP:0001945 Fever; HP:0002829 Arthralgia; HP:0002094 Dyspnea (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, varga1993ltryptophanandthe pages 2-3)
Chronic phenotypes	Weight loss 50%, muscle weakness 44%, paresthesia 42%, scleroderma-like skin induration 42%, xerostomia 36%, and alopecia 33%; persistent painful neuropathy, myopathy, fasciitis, contractures, and disability may occur despite normalization of eosinophilia.	HP:0001824 Weight loss; HP:0001324 Muscle weakness; HP:0003401 Paresthesia; HP:0000958 Dry skin/xerosis only if clinically appropriate—not equivalent to xerostomia; HP:0001596 Alopecia; peripheral neuropathy term/ID should be ontology-validated (varga1993ltryptophanandthe pages 2-3, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)
Epidemiology	The 1989 epidemic produced more than 1,500 reported US cases; reported deaths vary by surveillance date/source from 27 to approximately 40. Cases were predominantly middle-aged women; one Maryland series was 93% female with median age 51 years. Incidence is now extremely low and no reliable current prevalence is available.	Epidemiologic annotations; no inheritance ontology applicable (varga1993ltryptophanandthe pages 1-2, roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, NCT00001918 chunk 1, silver1994amurinemodel pages 1-3)
Genetic susceptibility/protection	No causal gene or pathogenic variant is established. Among L-tryptophan users, HLA-DRB103, HLA-DRB104, and HLA-DQA10601 were susceptibility markers; DRB104 had OR 3.93 (95% CI 1.08–16.37). DRB107 and DQA10501 were protective; among users of implicated lots, the DRB107–DQA10201 haplotype had OR 0.11 (95% CI 0.02–0.77). Results derive from a small retrospective cohort and are not diagnostic.	HLA gene/allele annotations; inheritance: not applicable; ClinVar/ACMG pathogenic-variant classification: not applicable (okada2009immunogeneticriskand pages 2-3, okada2009immunogeneticriskand pages 3-5)
Mechanism	Exposure induces an incompletely defined type-2 inflammatory response involving IL-5/IL-4, eosinophil expansion and tissue degranulation, followed by TGF-β-associated fibroblast activation, extracellular-matrix/collagen deposition, fascial and dermal fibrosis, myocyte injury, and peripheral nerve damage. Kynurenine/quinolinic-acid changes may contribute but remain uncertain.	GO:0043308 Eosinophil activation; GO:0030198 Extracellular matrix organization; GO:0032964 Collagen biosynthetic process; CL:0000771 Eosinophil; CL:0000057 Fibroblast; CL:0000097 Mast cell (varga1993ltryptophanandthe pages 3-4, allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, barth2001ltryptophancontaminant‘peak pages 1-2)
Anatomy	Primary sites are skeletal muscle, fascia/perimysium, dermis, subcutis, and peripheral nerves; lungs are commonly involved, while severe disease may affect the heart and other organs. Distribution is generally bilateral/systemic rather than unilateral.	UBERON:0001134 Skeletal muscle organ; UBERON:0002097 Skin of body; fascia/peripheral nerve/lung UBERON identifiers should be ontology-validated before ingestion (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, lee2023simultaneousdeterminationof pages 1-2)
Diagnosis	Exposure history plus CBC with differential and exclusion of parasitic/infectious, allergic, drug-induced, autoimmune, clonal, and malignant eosinophilia. Evaluate organ injury using CK/chemistry, pulmonary and cardiac testing, MRI of muscle/fascia, EMG/nerve-conduction studies, and full-thickness skin–fascia–muscle biopsy when needed. No validated genetic, proteomic, or metabolomic diagnostic test exists.	HPO terms above; LOINC/SNOMED codes should be mapped at test level; genetic testing: not routinely indicated (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, NCT00001918 chunk 1)
Treatment	Immediately discontinue the implicated supplement. Systemic glucocorticoids often reduce acute inflammation and eosinophilia but have inconsistent effects on chronic fibrosis and neuropathy. Mycophenolate, methotrexate, hydroxychloroquine, chlorambucil, or anakinra have only anecdotal/low-quality evidence; pain control and physical/occupational rehabilitation are important. No EMS-specific approved therapy or genotype-guided treatment exists.	NCIT suggestions: Corticosteroid Therapy, Immunosuppressive Therapy, Physical Therapy, Occupational Therapy; exact NCIT identifiers require validation (varga1993ltryptophanandthe pages 2-3, roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)
Prevention	Avoid suspect or unregulated L-tryptophan/5-HTP products; enforce manufacturing controls, lot traceability, impurity testing, adverse-event surveillance, and rapid product withdrawal. A 2023 LC–MS/MS method measured selected impurities with detection limits below 11.2 μg/kg and quantification limits below 35.7 μg/kg in meat matrices. No vaccine, genetic screening, or chemoprophylaxis applies.	CHEBI:16828 L-tryptophan; environmental-exposure and product-quality annotations (varga1993ltryptophanandthe pages 1-2, lee2023simultaneousdeterminationof pages 1-2)
Clinical study	NCT00001918, “L-5-Hydroxy-Tryptophan-Related EMS: Clinical Patient Evaluation,” was a completed NIH/NIMH observational study (planned enrollment 20; July 1999–August 2000), not a therapeutic trial. It evaluated clinical, neurologic, psychiatric, imaging, laboratory, and supplement-impurity findings.	ClinicalTrials.gov:NCT00001918; NCIT: Observational Study—exact code should be validated (NCT00001918 chunk 1)
Models	Induced female C57BL/6 mouse model: daily intraperitoneal EBT caused dermal/subcutaneous inflammation, mast-cell accumulation, progressive fascial fibrosis, and transient myocyte necrosis. After six weeks, fascia measured 168±22 μm versus 43±5 μm with saline, but mice lacked peripheral eosinophilia and did not reproduce the complete human syndrome. Lewis-rat models also show myofascial inflammation/fibrosis. No confirmed naturally occurring veterinary EMS or transmissible/zoonotic form is known.	NCBI Taxon:10090 Mus musculus; NCBI Taxon:10116 Rattus norvegicus; CL:0000097 Mast cell; GO:0030198 Extracellular matrix organization (silver1994amurinemodel pages 1-3, silver1994amurinemodel pages 3-4, silver1994amurinemodel pages 4-6)

Table: Compact disease-level summary of eosinophilia-myalgia syndrome, including epidemiology, clinical features, exposure biology, genetics, management, prevention, and experimental models. Ontology mappings are limited to identifiers that can be assigned with reasonable confidence, with uncertain mappings explicitly flagged.

1. Disease information

Definition and identifiers

The historical US CDC surveillance definition required: (1) peripheral blood eosinophils greater than 1,000/mm³, (2) generalized myalgia sufficiently severe to limit usual activities, and (3) no infection or neoplasm explaining the findings. This definition was designed for outbreak surveillance, not as a stand-alone clinical diagnostic standard. Revised criteria published in 2001 reportedly achieved 97% specificity. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6, silver1994amurinemodel pages 1-3)

Key identifiers and names are:

MONDO: MONDO:0004941.
MeSH: D016603, Eosinophilia-Myalgia Syndrome, confirmed in the ClinicalTrials.gov record. (NCT00001918 chunk 1)
OMIM: no disease-specific Mendelian entry is expected; EMS is acquired rather than a monogenic disorder.
Orphanet: a dedicated identifier was not confirmed in the retrieved evidence and should not be inferred.
ICD-10/ICD-11: no specific code was confirmed; coding generally requires syndrome manifestations or an adverse-effect/toxic-exposure code appropriate to the jurisdiction.
Synonyms: eosinophilia-myalgia syndrome; L-tryptophan-associated eosinophilia-myalgia syndrome; L-tryptophan-associated EMS; tryptophan-associated EMS; historically, “L-tryptophan-associated neuromyopathy.”

The evidence summarized here is predominantly aggregated disease-level evidence from surveillance cohorts, case series, epidemiology, tissue studies, and experimental models. Individual-patient evidence is used only where post-epidemic cases illuminate recurrence, diagnosis, or molecular pathology; no EHR-derived population dataset was identified.

2. Etiology, risks, protective factors, and gene–environment interaction

Primary causal factor

The strongest causal evidence is for ingestion of manufactured L-tryptophan containing process-related impurities. Epidemic lots came predominantly from one Japanese manufacturer after production changes that included a modified bacterial production strain and reduced purification. More than 60 impurities were subsequently detected. Six compounds were epidemiologically associated with affected lots: 3-phenylaminoalanine (PAA), 1,1′-ethylidenebis-L-tryptophan (EBT; “peak E”), 2-(3-indolylmethyl)-L-tryptophan (IMT), PIC, HIT, and AAA. Nevertheless, no single impurity has been proved to be the sole etiologic agent; EBT may be causal, contributory, or a marker of the responsible manufacturing conditions. (varga1993ltryptophanandthe pages 4-5, lee2023simultaneousdeterminationof pages 1-2, silver1994amurinemodel pages 1-3)

A useful authoritative formulation is the 1994 animal-model abstract: “EBT may have been one of the mediators of EMS.” This appropriately reflects the continuing uncertainty. (silver1994amurinemodel pages 1-3)

Exposure and demographic risks

Reported supplement doses ranged from 10 mg to 15–16 g/day, with a typical median near 1.5 g/day. In the Maryland series, median exposure before onset was six months (range 1.5–11 months), although a later post-epidemic patient developed symptoms within three weeks at 1,500 mg/day. Higher dose was associated with disease: mean intake was 4,160.7 mg/day in affected users versus 2,898.7 mg/day in unaffected users, OR 1.35 (95% CI 1.05–1.79, as modeled in that study). Age ≥45 years also increased risk, OR 3.01 (95% CI 1.03–8.75). (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, allen2011postepidemiceosinophiliamyalgiasyndrome pages 1-2, okada2009immunogeneticriskand pages 2-3)

Women predominated markedly in surveillance cohorts—14/15 Maryland patients (93%) were women—but immunogenetic analysis did not identify sex as an independent risk factor after accounting for exposure and other variables. This may partly reflect patterns of supplement use. Smoking, alcohol, occupation, exercise, family history, and ordinary dietary tryptophan have not been established as EMS risk factors. There is no infectious cause. (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, okada2009immunogeneticriskand pages 2-3)

Genetic susceptibility and protection

No gene mutation causes EMS. A retrospective study of 94 unrelated White L-tryptophan users found exposure-modifying HLA associations:

HLA-DRB1*03: OR 3.89 (95% CI 1.15–15.20).
HLA-DRB1*04: OR 3.93 (95% CI 1.08–16.37); among users of implicated lots, OR 7.52 (95% CI 1.30–86.05).
HLA-DQA1*0601: strong association, but with very wide confidence intervals because it was absent from unaffected exposed subjects.
Protective associations: DRB107, DQA10501, and—among users of implicated lots—DQA10201. The inferred DRB107–DQA1*0201 haplotype occurred in 13.0% of EMS cases versus 57.1% of unaffected exposed users, OR 0.11 (95% CI 0.02–0.77). (okada2009immunogeneticriskand pages 2-3, okada2009immunogeneticriskand pages 3-5)

These are susceptibility markers, not pathogenic variants. The cohort was small, several confidence intervals were very wide, and replication is limited. Accordingly, ClinVar/ACMG variant classification, penetrance, carrier frequency, anticipation, mosaicism, founder effects, and consanguinity are not applicable. No validated modifier gene, protective coding variant, chromosomal abnormality, or EMS-specific epigenetic change is known.

The best-supported gene–environment model is therefore: contaminated supplement exposure is necessary in most epidemic cases, while HLA-mediated antigen presentation and age/dose modify whether exposed persons develop disease. (okada2009immunogeneticriskand pages 1-2, okada2009immunogeneticriskand pages 3-5)

3. Phenotypes

EMS is typically adult-onset and heterogeneous. Frequencies vary because the CDC definition selected severe cases and studies sampled different disease stages.

Acute/subacute phenotypes

Severe generalized myalgia: 100%; usually subacute and disabling. Suggested HPO: HP:0003326.
Peripheral eosinophilia: 100% by surveillance definition; often transient. HPO: HP:0001880.
Rash: 71% in one summarized series; 47% in Maryland. HPO: HP:0000988.
Peripheral/extremity edema: 52% and 47%, respectively. HPO: HP:0000969.
Fever: 41% or 27%. HPO: HP:0001945.
Arthralgia/joint pain: 35% or 53%. HPO: HP:0002829.
Respiratory symptoms/dyspnea: 32–33%, ranging from cough or breathlessness to pneumonitis. HPO: HP:0002094 for dyspnea.
Fatigue and stiffness: common but not consistently quantified.
Laboratory findings: absolute eosinophils exceed 1,000/mm³ under the original definition; one Maryland cohort also reported lymphocytosis in 93%. (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, varga1993ltryptophanandthe pages 2-3)

Later and chronic phenotypes

Reported later manifestations include weight loss (50%), muscle weakness (44%), paresthesia (42%), scleroderma-like induration (42%), xerostomia (36%), and alopecia (33%). Suggested HPO terms include HP:0001824 weight loss, HP:0001324 muscle weakness, HP:0003401 paresthesia, and HP:0001596 alopecia. Fasciitis, painful sensorimotor neuropathy, muscle atrophy, contractures, restricted mobility, and cognitive symptoms may persist. (varga1993ltryptophanandthe pages 2-3, roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4)

The quality-of-life burden can be profound: documented consequences include inability to perform usual activities, wheelchair dependence, chronic pain, weakness, sensory loss, and contractures. No EMS-specific EQ-5D, SF-36, or PROMIS reference dataset was identified. (roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4)

4. Genetic and molecular information

There are no causal genes, pathogenic germline or somatic variants, structural variants, or chromosomal abnormalities. WES, WGS, gene panels, CMA, karyotyping, FISH, mitochondrial testing, and repeat-expansion testing have no established diagnostic role.

Human lesional skin profiling in a post-epidemic case identified 343 differentially expressed genes at FDR 0.64%, including collagen and extracellular-matrix genes and signatures involving TGF-β and IL-4/IL-13 signaling. This is a downstream disease-state signature, not evidence of genomic causation. Earlier tissue work also reported increased TGF-β1 and collagen expression. No validated EMS methylome, proteomic classifier, metabolomic diagnostic signature, single-cell atlas, spatial-transcriptomic dataset, multi-omics study, or CRISPR screen was found. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, allen2011postepidemiceosinophiliamyalgiasyndrome pages 6-7)

5. Environmental, lifestyle, and infectious information

The relevant environment is an ingested manufactured chemical mixture. L-tryptophan itself is CHEBI:16828; EBT and PAA require identifier validation before knowledge-base ingestion. Ordinary food-derived tryptophan has not been shown to cause the epidemic syndrome. A published cashew-associated case is insufficient to establish normal foods as a general risk.

No reproducible association exists with smoking, alcohol, exercise, radiation, ambient pollution, or occupation. There is no bacterial, viral, fungal, or parasitic etiologic agent; however, parasitic infection must be excluded during diagnosis. In five Maryland patients tested, Trichinella serology was negative. (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5)

6. Mechanism and pathophysiology

Ordered causal chain

Ingestion of contaminated manufactured L-tryptophan or possibly 5-HTP leads to systemic exposure to EBT/PAA and other process impurities. The exposure–disease association is demonstrated; attribution to one molecule remains unresolved. (varga1993ltryptophanandthe pages 4-5, NCT00001918 chunk 1, lee2023simultaneousdeterminationof pages 1-2)
Impurity exposure leads to activation of susceptible immune cells and a type-2 cytokine response. EBT induced IL-5 and/or IL-10 in PBMC from 6/7 responding functional-somatic-syndrome subjects; this is in-vitro evidence, not direct proof in EMS patients. (barth2001ltryptophancontaminant‘peak pages 1-2)
IL-5-rich inflammation leads to eosinophil expansion, recruitment, and tissue degranulation. Human lesions contain eosinophil-derived neurotoxin and major basic protein even when intact eosinophils are sparse. Suggested GO: eosinophil activation (GO:0043308); CL: eosinophil (CL:0000771). (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5)
Eosinophil products and Th2 cytokines lead to inflammatory injury of fascia, perimysium, dermis, muscle, lung, and peripheral nerves. This step is biologically supported but the relative contribution of eosinophils versus lymphocytes, monocytes, and mast cells remains inferred. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, silver1994amurinemodel pages 1-3)
Inflammation activates TGF-β/IL-4-associated fibroblast programs, resulting in collagen and extracellular-matrix deposition. Suggested GO: extracellular-matrix organization (GO:0030198) and collagen biosynthetic process (GO:0032964); CL: fibroblast (CL:0000057). (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5)
Fibrosis and myocyte injury result in fascial thickening, skin induration, painful myopathy, weakness, and contractures. (roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4, silver1994amurinemodel pages 4-6)
Branch A: perineural inflammation and eosinophil-mediated toxicity lead to chronic peripheral neuropathy and paresthesia. Quinolinic-acid neurotoxicity is plausible but unproved. (varga1993ltryptophanandthe pages 3-4, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)
Branch B: pulmonary inflammation leads to dyspnea, ground-glass change, pneumonitis, and occasionally life-threatening organ injury. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)

Metabolic and cellular detail

Active disease has been associated with increased kynurenine and cerebrospinal-fluid quinolinic acid, suggesting cytokine-driven indoleamine-2,3-dioxygenase activation. Because EMS preparations did not directly increase IDO in normal mononuclear cells and the expected quinolinate lesion differs from EMS neuropathy, this is more likely secondary or contributory than the initiating defect. No enzyme deficiency, receptor mutation, protein aggregation disorder, or primary mitochondrial lesion is established. (varga1993ltryptophanandthe pages 3-4)

In the EBT mouse model, mast cells increased two- to nearly four-fold in affected skin layers and appeared to degranulate, suggesting a possible amplifying role. Relevant CL terms are mast cell (CL:0000097), lymphocyte, monocyte, eosinophil, fibroblast, skeletal-muscle cell, and peripheral neuron. (silver1994amurinemodel pages 3-4, silver1994amurinemodel pages 4-6)

7. Anatomy

Primary sites are bilateral/systemic skeletal muscle, fascia and perimysium, dermis, subcutis, and peripheral nerves. Suggested mappings include skeletal muscle organ UBERON:0001134 and skin of body UBERON:0002097; fascia, peripheral nerve, lung, and myocardium identifiers should be ontology-validated before ingestion.

Lung involvement includes pneumonitis, ground-glass opacity, and restrictive or obstructive physiology. Cardiac and other visceral involvement can occur in severe disease but is less common. Histology shows dermal/fascial collagen and mucopolysaccharide accumulation, mononuclear and variably eosinophilic infiltrates, eosinophil-protein deposition, perimysial inflammation, myofiber atrophy/necrosis, and fibrosis. There is no characteristic lateralization. (varga1993ltryptophanandthe pages 1-2, allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4)

At the subcellular level, the evidence concerns extracellular eosinophil granule proteins and extracellular matrix rather than a consistently abnormal organelle. Relevant GO cellular components include extracellular matrix and collagen-containing extracellular matrix.

8. Temporal development

EMS is usually adult-onset and acute-to-subacute. During the epidemic, most onset dates fell between July 1989 and February 1990. Exposure may precede disease by weeks to months; onset can also occur shortly after discontinuation. (varga1993ltryptophanandthe pages 1-2, roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, allen2011postepidemiceosinophiliamyalgiasyndrome pages 1-2)

A practical staging model is:

Early inflammatory phase: rapidly progressive myalgia, edema, rash, fever, and marked eosinophilia.
Evolving tissue phase: weakness, fasciitis, pulmonary disease, and neuropathic symptoms.
Chronic fibrotic/neuromuscular phase: eosinophilia may resolve while skin induration, neuropathy, myopathy, contractures, and disability persist.

Spontaneous or treatment-associated improvement occurs, but chronic disease is common. The most important intervention window is early recognition and immediate withdrawal of the exposure, before fixed fibrosis and nerve damage develop. Controlled evidence defining this window is unavailable. (varga1993ltryptophanandthe pages 1-2, roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)

9. Inheritance and population epidemiology

EMS has no Mendelian inheritance pattern. Penetrance applies only informally to exposed populations and depends on dose, product lot, age, and immunogenetic susceptibility.

Official surveillance documented at least 1,543 cases by June 1991 and 27 deaths; later sources cite more than 1,500 cases and 38–40 deaths. These differences reflect surveillance date, case definition, and probable under-ascertainment. No reliable current prevalence or annual incidence per 100,000 exists. (varga1993ltryptophanandthe pages 1-2, NCT00001918 chunk 1, silver1994amurinemodel pages 1-3)

The epidemic was concentrated in the United States but cases occurred internationally. Adults, especially middle-aged women, dominated reported cohorts. There is no established ethnicity-specific causal variant or geographic founder effect. Geographic variation tracked supplement distribution and regulatory/manufacturing conditions rather than inherited ancestry. (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5, varga1993ltryptophanandthe pages 4-5)

10. Diagnostics

Clinical approach

Diagnosis rests on the clinical syndrome, supplement history, eosinophil count, organ assessment, and exclusion of alternatives. Recommended evaluation includes:

CBC with differential and serial absolute eosinophil counts.
CK, aldolase, metabolic panel, liver tests, inflammatory markers, urinalysis, and cardiac biomarkers as indicated.
Parasite testing guided by travel/exposure; medication and supplement reconciliation.
ECG/echocardiography and pulmonary function testing; chest radiography or CT for cardiopulmonary symptoms.
MRI of symptomatic muscle and fascia; muscle/fascial edema can support active disease.
EMG and nerve-conduction studies for myopathy or neuropathy.
Full-thickness skin-to-fascia/muscle biopsy when diagnosis remains uncertain. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4, NCT00001918 chunk 1)

There is no validated circulating EMS biomarker beyond nonspecific eosinophilia, and eosinophil normalization does not prove resolution. Chemical analysis of retained supplement lots by LC-MS/MS can support exposure investigation but cannot exclude EMS when the consumed material is unavailable or when relevant contaminants are unknown. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, lee2023simultaneousdeterminationof pages 1-2)

Differential diagnosis

Important alternatives include eosinophilic fasciitis, hypereosinophilic syndrome and clonal eosinophilia, eosinophilic granulomatosis with polyangiitis, parasitic infection, drug-induced eosinophilia/DRESS, eosinophilic myositis, systemic sclerosis, inflammatory myopathy, toxic-oil syndrome, malignancy, and eosinophilic pneumonia. EMS is distinguished by the exposure history, severe generalized myalgia, epidemic/toxic context, and combined fascial, neuromuscular, cutaneous, and pulmonary phenotype. Genetic testing is reserved for an alternative suspected clonal or inherited eosinophilic disorder, not EMS itself.

There is no recommended population, newborn, carrier, prenatal, or cascade screening.

11. Outcome and prognosis

Acute EMS can be fatal, but disease-specific 5- or 10-year survival estimates are unavailable. Historical surveillance mortality was approximately 2% using 27–40 deaths among roughly 1,500 reported cases, although both numerator and denominator are uncertain. (varga1993ltryptophanandthe pages 1-2, NCT00001918 chunk 1)

Morbidity is more prominent than mortality. Eosinophilia, fever, edema, rash, and pulmonary inflammation may improve, whereas fixed fibrosis, weakness, neuropathic pain, sensory loss, and contractures can persist for years. In the 2011 post-epidemic case, prednisone plus mycophenolate resolved eosinophilia and pulmonary ground-glass changes and modestly improved strength/myalgia, but skin induration and neuropathy progressed. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)

Probable adverse prognostic factors include severe early organ involvement, neuropathy, established fibrosis, delayed withdrawal, and incomplete response to anti-inflammatory therapy. No validated prognostic score or molecular prognostic biomarker exists.

12. Treatment

Immediately stop L-tryptophan, 5-HTP, and any suspect supplement. This is the essential intervention. In one case, withdrawal was followed within four weeks by falling eosinophils and remission of respiratory symptoms and myalgia. (roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4)
Systemic glucocorticoids are commonly used for severe inflammatory, pulmonary, fascial, or neurologic disease. They often suppress eosinophilia and acute inflammation but have inconsistent effects on established fibrosis, myopathy, and neuropathy; historical comparisons did not show clearly superior overall outcomes. Suggested NCIT concept: corticosteroid therapy. (varga1993ltryptophanandthe pages 1-2, varga1993ltryptophanandthe pages 2-3)
Steroid-sparing immunosuppression may be individualized. Mycophenolate produced partial benefit in one modern case; methotrexate and anakinra did not prevent progression. Hydroxychloroquine and chlorambucil have only anecdotal historical use. There are no reliable response rates. (roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)
Supportive care includes analgesia, neuropathic-pain treatment, physical and occupational therapy, contracture prevention, pulmonary/cardiac management, nutrition, and psychological support. Suggested NCIT concepts include physical therapy, occupational therapy, pain management, and immunosuppressive therapy; exact identifiers should be validated.

No gene, RNA, cell, surgical, or approved targeted biologic therapy exists. Anti-IL-5 drugs are mechanistically interesting but cannot be recommended specifically for EMS without clinical evidence. No pharmacogenomic guidance or personalized genotype-directed treatment exists.

Clinical trial status: NCT00001918, “L-5-Hydroxy-Tryptophan-Related EMS: Clinical Patient Evaluation,” was a completed NIH/NIMH observational study with planned enrollment of 20, running July 1999–August 2000. It was not a treatment trial and evaluated clinical, neurologic, psychiatric, imaging, laboratory, and supplement-chemistry findings. URL: https://clinicaltrials.gov/study/NCT00001918. (NCT00001918 chunk 1)

13. Prevention

Primary prevention is the principal public-health strategy: avoid unregulated or suspect L-tryptophan/5-HTP products; use validated fermentation and purification processes; test for known and unknown impurities; maintain lot traceability; and rapidly recall implicated products. The sharp fall in incidence after the FDA recall strongly supports exposure removal. (varga1993ltryptophanandthe pages 1-2)

Recent implementation is analytical rather than clinical. A study published online 2 January 2023 developed a five-minute LC-MS/MS assay for selected tryptophan impurities in meat matrices, with method detection limits below 11.2 μg/kg and quantification limits below 35.7 μg/kg. DOI/URL: https://doi.org/10.1007/s00726-022-03215-8. (lee2023simultaneousdeterminationof pages 1-2)

An EFSA opinion adopted 12 March 2024 concluded that ≥98% L-tryptophan made with a specified non-genetically-modified E. coli strain was safe for non-ruminant feed use and presented no consumer or environmental concern under assessed conditions. This is product-specific regulatory evidence—not proof that every supplement is risk-free. DOI/URL: https://doi.org/10.2903/j.efsa.2024.8707. (bampidis2024safetyandefficacy pages 1-2)

Secondary prevention consists of rapid recognition, supplement cessation, adverse-event reporting, retained-lot testing, and early organ assessment. Tertiary prevention addresses fibrosis, contractures, falls, chronic neuropathy, and cardiopulmonary complications through rehabilitation and specialist monitoring. Vaccination, genetic screening, reproductive counseling, and chemoprophylaxis are not applicable.

14. Other species and natural disease

No confirmed naturally occurring EMS equivalent in companion animals, livestock, or wildlife was identified. EMS is not infectious, transmissible, or zoonotic. There is no breed predisposition, orthologous causal gene, or cross-species transmission concern.

Recent swine, poultry, and EFSA studies concern the toxicologic safety of tryptophan impurities and feed additives, not spontaneous veterinary EMS. Their relevance is regulatory and comparative-toxicologic rather than evidence of natural animal disease. (lee2023simultaneousdeterminationof pages 1-2, bampidis2024safetyandefficacy pages 1-2)

15. Model organisms

Mouse model

Female C57BL/6 mice given daily intraperitoneal EBT developed dermal and subcutaneous inflammation, mast-cell accumulation, progressive fascial/perimysial fibrosis, and transient myofiber necrosis. At six weeks, fascia thickness was 168±22 μm versus 43±5 μm after saline and 69±7 μm after L-tryptophan alone (P<0.01). EBT-associated fibrosis was evident by day 6 and increased by day 21. (silver1994amurinemodel pages 1-3, silver1994amurinemodel pages 3-4, silver1994amurinemodel pages 4-6)

The model reproduces inflammatory fibrosis but not the full human syndrome: mice lacked significant peripheral eosinophilia, gross weakness, and characteristic multisystem disease. Intraperitoneal pure EBT also differs from oral exposure to a complex contaminant mixture. NCBI Taxon: 10090, Mus musculus. (silver1994amurinemodel pages 3-4)

Rat and in-vitro models

Female Lewis rats exposed to implicated tryptophan lots or synthetic EBT developed myofascial inflammation/fibrosis, but findings varied across experiments. NCBI Taxon: 10116, Rattus norvegicus. Human PBMC cultures provide a complementary mechanistic model: 7/12 subjects with functional somatic syndromes responded to EBT versus 3/24 controls (P<0.05), with IL-5 and/or IL-10 in six of seven responders. This suggests host-dependent type-2 reactivity but was not performed directly in an EMS cohort and cannot establish clinical susceptibility. (varga1993ltryptophanandthe pages 4-5, barth2001ltryptophancontaminant‘peak pages 1-2)

Evidence appraisal and current research gaps

The exposure–disease relationship is compelling because of temporal clustering, lot/manufacturer association, dose effects, biological plausibility, and disappearance after recall. Expert interpretation should nevertheless separate that strong inference from the weaker claim that EBT alone caused every case. Animal findings, multiple correlated impurities, occasional cases without demonstrable EBT, and absent retained product all preserve etiologic uncertainty. (varga1993ltryptophanandthe pages 4-5, allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5, allen2011postepidemiceosinophiliamyalgiasyndrome pages 6-7)

Major unmet needs are validated modern diagnostic criteria, prospective natural-history data, contaminant-independent biomarkers, replication of HLA associations, contemporary exposure surveillance, and controlled studies of anti-fibrotic or eosinophil-directed therapy. Because EMS is now extremely rare, international case registries and standardized biobanking are more feasible than conventional randomized trials.

Selected primary and authoritative sources

Varga J, Jimenez SA, Uitto J. L-tryptophan and the eosinophilia-myalgia syndrome. January 1993. DOI: https://doi.org/10.1038/jid.1993.31. (varga1993ltryptophanandthe pages 1-2)
Roubenoff R et al. Eosinophilia-myalgia syndrome due to L-tryptophan ingestion. July 1990. DOI: https://doi.org/10.1002/art.1780330703. (roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5)
Silver RM et al. A murine model…induced by EBT. April 1994. DOI: https://doi.org/10.1172/JCI117125. The abstract states that EBT caused “inflammation and fibrosis affecting the dermis and subcutis, including the fascia and perimyseal tissues.” (silver1994amurinemodel pages 1-3)
Barth H et al. L-tryptophan contaminant ‘peak E’ induces release of IL-5 and IL-10. November 2001. DOI: https://doi.org/10.1046/j.1365-2249.2001.01559.x. (barth2001ltryptophancontaminant‘peak pages 1-2)
Okada S et al. Immunogenetic risk and protective factors…. 15 October 2009. DOI: https://doi.org/10.1002/art.24460. (okada2009immunogeneticriskand pages 1-2, okada2009immunogeneticriskand pages 3-5)
Allen JA et al. Post-epidemic eosinophilia-myalgia syndrome associated with L-tryptophan. November 2011. DOI: https://doi.org/10.1002/art.30514. (allen2011postepidemiceosinophiliamyalgiasyndrome pages 1-2, allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6)
NIH/NIMH ClinicalTrials.gov record NCT00001918. Background references include PMID 1690352, 1727200, and 1969024. (NCT00001918 chunk 1)
Lee DH et al. Simultaneous determination of L-tryptophan impurities in meat products. 2 January 2023. DOI: https://doi.org/10.1007/s00726-022-03215-8. (lee2023simultaneousdeterminationof pages 1-2)
EFSA FEEDAP Panel. Safety and efficacy of an L-tryptophan feed additive. Adopted 12 March 2024. DOI: https://doi.org/10.2903/j.efsa.2024.8707. (bampidis2024safetyandefficacy pages 1-2)

References

(varga1993ltryptophanandthe pages 1-2): John Varga, Sergio A. Jimenez, and Jouni Uitto. L-tryptophan and the eosinophilia-myalgia syndrome: current understanding of the etiology and pathogenesis. The Journal of investigative dermatology, 100 1:97S-105S, Jan 1993. URL: https://doi.org/10.1038/jid.1993.31, doi:10.1038/jid.1993.31. This article has 54 citations.
(allen2011postepidemiceosinophiliamyalgiasyndrome pages 1-2): Jeffrey A. Allen, Alicia Peterson, Robert Sufit, Monique E. Hinchcliff, J. Matthew Mahoney, Tammara A. Wood, Frederick W. Miller, Michael L. Whitfield, and John Varga. Post-epidemic eosinophilia-myalgia syndrome associated with l-tryptophan. Arthritis and rheumatism, 63 11:3633-9, Nov 2011. URL: https://doi.org/10.1002/art.30514, doi:10.1002/art.30514. This article has 99 citations.
(NCT00001918 chunk 1): L-5-HTP-Related EMS. National Institute of Mental Health (NIMH). 1999. ClinicalTrials.gov Identifier: NCT00001918
(silver1994amurinemodel pages 1-3): R. Silver, A. Ludwicka, M. Hampton, T. Ohba, S. A. Bingel, T. Smith, R. Harley, J. Maize, and Melvyn P. Heyes. A murine model of the eosinophilia-myalgia syndrome induced by 1,1'-ethylidenebis (l-tryptophan). Journal of Clinical Investigation, 93:1473-1480, Apr 1994. URL: https://doi.org/10.1172/jci117125, doi:10.1172/jci117125. This article has 41 citations and is from a highest quality peer-reviewed journal.
(allen2011postepidemiceosinophiliamyalgiasyndrome pages 5-6): Jeffrey A. Allen, Alicia Peterson, Robert Sufit, Monique E. Hinchcliff, J. Matthew Mahoney, Tammara A. Wood, Frederick W. Miller, Michael L. Whitfield, and John Varga. Post-epidemic eosinophilia-myalgia syndrome associated with l-tryptophan. Arthritis and rheumatism, 63 11:3633-9, Nov 2011. URL: https://doi.org/10.1002/art.30514, doi:10.1002/art.30514. This article has 99 citations.
(lee2023simultaneousdeterminationof pages 1-2): Doo-Hee Lee, Yang Hee Kim, Mina Baek, In Kyung Heo, and Yonguk Shin. Simultaneous determination of l-tryptophan impurities in meat products. Amino Acids, 55:173-182, Jan 2023. URL: https://doi.org/10.1007/s00726-022-03215-8, doi:10.1007/s00726-022-03215-8. This article has 6 citations and is from a peer-reviewed journal.
(bampidis2024safetyandefficacy pages 1-2): Vasileios Bampidis, Giovanna Azimonti, Maria de Lourdes Bastos, Henrik Christensen, Mojca Durjava, Birgit Dusemund, Maryline Kouba, Marta López‐Alonso, Secundino López Puente, Francesca Marcon, Baltasar Mayo, Alena Pechová, Mariana Petkova, Fernando Ramos, Roberto Edoardo Villa, Ruud Woutersen, Lieve Herman, Montserrat Anguita, Matteo Lorenzo Innocenti, Jordi Tarrés‐Call, and Elisa Pettenati. Safety and efficacy of a feed additive consisting of l‐tryptophan (produced with escherichia coli cgmcc 7.460) for all animal species (kempex holland b.v.). EFSA Journal, Apr 2024. URL: https://doi.org/10.2903/j.efsa.2024.8707, doi:10.2903/j.efsa.2024.8707. This article has 0 citations and is from a peer-reviewed journal.
(varga1993ltryptophanandthe pages 4-5): John Varga, Sergio A. Jimenez, and Jouni Uitto. L-tryptophan and the eosinophilia-myalgia syndrome: current understanding of the etiology and pathogenesis. The Journal of investigative dermatology, 100 1:97S-105S, Jan 1993. URL: https://doi.org/10.1038/jid.1993.31, doi:10.1038/jid.1993.31. This article has 54 citations.
(roubenoff1990eosinophiliamyalgiasyndromedue pages 4-5): Ronenn Roubenoff, Timothy Coté, Rosemarie Watson, Michael L. Levin, and Marc C. Hochberg. Eosinophilia-myalgia syndrome due to l-tryptophan ingestion. report of four cases and review of the maryland experience. Arthritis and rheumatism, 33 7:930-8, Jul 1990. URL: https://doi.org/10.1002/art.1780330703, doi:10.1002/art.1780330703. This article has 24 citations.
(varga1993ltryptophanandthe pages 2-3): John Varga, Sergio A. Jimenez, and Jouni Uitto. L-tryptophan and the eosinophilia-myalgia syndrome: current understanding of the etiology and pathogenesis. The Journal of investigative dermatology, 100 1:97S-105S, Jan 1993. URL: https://doi.org/10.1038/jid.1993.31, doi:10.1038/jid.1993.31. This article has 54 citations.
(okada2009immunogeneticriskand pages 2-3): Satoshi Okada, Mary L. Kamb, Janardan P. Pandey, Rossanne M. Philen, Lori A. Love, and Frederick W. Miller. Immunogenetic risk and protective factors for the development of l-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms. Arthritis and rheumatism, 61 10:1305-11, Oct 2009. URL: https://doi.org/10.1002/art.24460, doi:10.1002/art.24460. This article has 19 citations.
(okada2009immunogeneticriskand pages 3-5): Satoshi Okada, Mary L. Kamb, Janardan P. Pandey, Rossanne M. Philen, Lori A. Love, and Frederick W. Miller. Immunogenetic risk and protective factors for the development of l-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms. Arthritis and rheumatism, 61 10:1305-11, Oct 2009. URL: https://doi.org/10.1002/art.24460, doi:10.1002/art.24460. This article has 19 citations.
(varga1993ltryptophanandthe pages 3-4): John Varga, Sergio A. Jimenez, and Jouni Uitto. L-tryptophan and the eosinophilia-myalgia syndrome: current understanding of the etiology and pathogenesis. The Journal of investigative dermatology, 100 1:97S-105S, Jan 1993. URL: https://doi.org/10.1038/jid.1993.31, doi:10.1038/jid.1993.31. This article has 54 citations.
(allen2011postepidemiceosinophiliamyalgiasyndrome pages 2-5): Jeffrey A. Allen, Alicia Peterson, Robert Sufit, Monique E. Hinchcliff, J. Matthew Mahoney, Tammara A. Wood, Frederick W. Miller, Michael L. Whitfield, and John Varga. Post-epidemic eosinophilia-myalgia syndrome associated with l-tryptophan. Arthritis and rheumatism, 63 11:3633-9, Nov 2011. URL: https://doi.org/10.1002/art.30514, doi:10.1002/art.30514. This article has 99 citations.
(barth2001ltryptophancontaminant‘peak pages 1-2): H Barth, R Klein, and P A Berg. L-tryptophan contaminant ‘peak e’ induces the release of il-5 and il-10 by peripheral blood mononuclear cells from patients with functional somatic syndromes. Clinical and Experimental Immunology, 126:187-192, Nov 2001. URL: https://doi.org/10.1046/j.1365-2249.2001.01559.x, doi:10.1046/j.1365-2249.2001.01559.x. This article has 12 citations and is from a peer-reviewed journal.
(roubenoff1990eosinophiliamyalgiasyndromedue pages 3-4): Ronenn Roubenoff, Timothy Coté, Rosemarie Watson, Michael L. Levin, and Marc C. Hochberg. Eosinophilia-myalgia syndrome due to l-tryptophan ingestion. report of four cases and review of the maryland experience. Arthritis and rheumatism, 33 7:930-8, Jul 1990. URL: https://doi.org/10.1002/art.1780330703, doi:10.1002/art.1780330703. This article has 24 citations.
(silver1994amurinemodel pages 3-4): R. Silver, A. Ludwicka, M. Hampton, T. Ohba, S. A. Bingel, T. Smith, R. Harley, J. Maize, and Melvyn P. Heyes. A murine model of the eosinophilia-myalgia syndrome induced by 1,1'-ethylidenebis (l-tryptophan). Journal of Clinical Investigation, 93:1473-1480, Apr 1994. URL: https://doi.org/10.1172/jci117125, doi:10.1172/jci117125. This article has 41 citations and is from a highest quality peer-reviewed journal.
(silver1994amurinemodel pages 4-6): R. Silver, A. Ludwicka, M. Hampton, T. Ohba, S. A. Bingel, T. Smith, R. Harley, J. Maize, and Melvyn P. Heyes. A murine model of the eosinophilia-myalgia syndrome induced by 1,1'-ethylidenebis (l-tryptophan). Journal of Clinical Investigation, 93:1473-1480, Apr 1994. URL: https://doi.org/10.1172/jci117125, doi:10.1172/jci117125. This article has 41 citations and is from a highest quality peer-reviewed journal.
(okada2009immunogeneticriskand pages 1-2): Satoshi Okada, Mary L. Kamb, Janardan P. Pandey, Rossanne M. Philen, Lori A. Love, and Frederick W. Miller. Immunogenetic risk and protective factors for the development of l-tryptophan-associated eosinophilia-myalgia syndrome and associated symptoms. Arthritis and rheumatism, 61 10:1305-11, Oct 2009. URL: https://doi.org/10.1002/art.24460, doi:10.1002/art.24460. This article has 19 citations.
(allen2011postepidemiceosinophiliamyalgiasyndrome pages 6-7): Jeffrey A. Allen, Alicia Peterson, Robert Sufit, Monique E. Hinchcliff, J. Matthew Mahoney, Tammara A. Wood, Frederick W. Miller, Michael L. Whitfield, and John Varga. Post-epidemic eosinophilia-myalgia syndrome associated with l-tryptophan. Arthritis and rheumatism, 63 11:3633-9, Nov 2011. URL: https://doi.org/10.1002/art.30514, doi:10.1002/art.30514. This article has 99 citations.

Artifacts

Edison artifact artifact-00

Reference Validation

Checked with linkml-reference-validator 0.3.0rc1.

Outcome	Count
References checked	9
Resolved	9
Unresolved (possible confabulation)	0
Unverifiable	0
References weighed for topical relevance	9
On topic	3
Off topic	1

References that may not be about this subject

These identifiers resolve, so they are not fabrications, but the records they resolve to share almost none of this report's vocabulary. That is a clue and not a verdict - a paper can be relevant in ways its title and abstract do not spell out - so read them before deciding:

DOI:10.2903/j.efsa.2024.8707 (4 mentions) - Safety and efficacy of a feed additive consisting of l‐tryptophan (produced with Escherichia coli CGMCC 7.460) for all animal species (Kempex Holland B.V.)
shared terms: exposure

Weighed against this report's own most characteristic terms: ems, disease, model, l-tryptophan, exposure, syndrome, fibrosis, eosinophilia, ebt, include, neuropathy, clinical, eosinophil, nct00001918, peripheral, myalgia, chunk, inflammation, severe, validated.

All extracted references resolved successfully. Resolving is not the same as being relevant, though - see the references listed above as possibly off topic.

Ask OpenScientist

OpenScientist Report

Definitions

Mechanistic Hypotheses

Discussions and Knowledge Gaps

Pathophysiology

Pathograph

Phenotypes

Genetic Associations

Medical Actions

Environmental Factors

Diagnosis

Progression

Prevalence

Clinical Burden

Related Datasets

Clinical Trials

Animal Models

Source YAML

References & Deep Research

References

Deep Research

Eosinophilia–Myalgia Syndrome: Disease Characteristics Research Report

Executive summary

1. Disease information

Definition and identifiers

2. Etiology, risks, protective factors, and gene–environment interaction

Primary causal factor

Exposure and demographic risks

Genetic susceptibility and protection

3. Phenotypes

Acute/subacute phenotypes

Later and chronic phenotypes

4. Genetic and molecular information

5. Environmental, lifestyle, and infectious information

6. Mechanism and pathophysiology

Ordered causal chain

Metabolic and cellular detail

7. Anatomy

8. Temporal development

9. Inheritance and population epidemiology

10. Diagnostics

Clinical approach

Differential diagnosis

11. Outcome and prognosis

12. Treatment

13. Prevention

14. Other species and natural disease

15. Model organisms

Mouse model

Rat and in-vitro models

Evidence appraisal and current research gaps

Selected primary and authoritative sources

Artifacts

Reference Validation

References that may not be about this subject