Diethylene Glycol Poisoning

Environmental Pathograph 25 Show in embeddings browser

Diethylene glycol poisoning is an acute toxic nephropathy with delayed neurological sequelae, caused by ingestion of diethylene glycol - a sweet, viscous industrial solvent and humectant. Almost every large outbreak has been a pharmaceutical mass poisoning in which diethylene glycol was substituted, as a cheaper adulterant, for pharmaceutical grade glycerin or propylene glycol in a liquid medicine. Diethylene glycol itself is not the toxicant. It is oxidised by alcohol and aldehyde dehydrogenase to 2-hydroxyethoxyacetic acid and then to diglycolic acid, and it is diglycolic acid - a close structural mimic of the citric acid cycle dicarboxylate succinate - that is carried into renal proximal tubular cells by the transporters that normally reclaim succinate, is poorly effluxed once inside, and inhibits mitochondrial complex II. The resulting ATP depletion produces proximal tubular necrosis rather than apoptosis, and the swollen necrotic epithelium obstructs the tubular lumen, giving the anuric acute kidney injury that defines the syndrome. Survivors of the renal phase may develop a distinct delayed neurological syndrome of cranial neuropathies - bilateral facial palsy is the classic finding - bulbar palsy, and a sensorimotor polyneuropathy. The cited sources do not agree on how delayed: one case series places the cranial neuropathies in the second week after ingestion, while a rat study's review of the human literature reports onset at two to seven days.

Ask OpenScientist

Ask a research question about Diethylene Glycol Poisoning. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

11
Pathophys.
17
Phenotypes
4
Hypotheses
2
Gaps
25
Pathograph
4
Medical Actions
5
Models
2
References
1
Deep Research
◈

Mechanistic Hypotheses

4
Diglycolic acid is the proximate renal toxicant
dga_proximal_tubular_necrosis CANONICAL
The renal lesion is produced by diglycolic acid rather than by diethylene glycol or by 2-hydroxyethoxyacetic acid. This is the best evidenced part of the mechanism: in human proximal tubule cells the parent compound and the first metabolite produce no cell death at all, and in rats the toxicity appears only at doses at which diglycolic acid accumulates in kidney tissue.
Pathograph links
2-Hydroxyethoxyacetic Acid Formation -> Diglycolic Acid Formation DIRECT
Diglycolic Acid Formation -> Dicarboxylate Transporter-Mediated Renal Uptake DIRECT
Dicarboxylate Transporter-Mediated Renal Uptake -> Intracellular Diglycolic Acid Trapping DIRECT
Intracellular Diglycolic Acid Trapping -> Renal Mitochondrial Complex II Inhibition DIRECT
Renal Mitochondrial Complex II Inhibition -> Proximal Tubular ATP Depletion and Necrosis DIRECT
Proximal Tubular ATP Depletion and Necrosis -> Tubular Luminal Obstruction DIRECT
Diglycolic acid is directly neurotoxic
dga_direct_neurotoxicity EMERGING
The delayed neuropathy reflects the same complex II inhibition acting on neurons, demonstrated in a neuroblastoma line. The evidence is in vitro and uses millimolar concentrations, and no human nerve tissue measurement supports it yet.
Pathograph links
Diglycolic Acid Formation -> Neuronal Mitochondrial Complex II Inhibition INDIRECT UNKNOWN INTERMEDIATES The neuronal arm is drawn with unknown intermediates on purpose, and carries two hypothesis groups because two accounts share this edge and are not exclusive. What is demonstrated is that diglycolic acid inhibits complex II in a neuronal cell line, and that brain diglycolic acid is markedly higher in rats that develop kidney injury than in rats that do not; how it reaches human cranial nerve is not established.
Neuronal Mitochondrial Complex II Inhibition -> Facial palsy INDIRECT UNKNOWN INTERMEDIATES
Neuronal Mitochondrial Complex II Inhibition -> Sensorimotor neuropathy INDIRECT UNKNOWN INTERMEDIATES
2-hydroxyethoxyacetic acid is the neurotoxic moiety
heaa_neurotoxicity ALTERNATIVE
A review of delayed neurological sequelae across the toxic alcohols attributes the cranial neuropathies of diethylene glycol to 2-hydroxyethoxyacetic acid rather than to diglycolic acid. The source states this as what is "thought to be" the case rather than as a demonstrated result, and the entry carries it with that hedge intact.
Pathograph links
2-Hydroxyethoxyacetic Acid Formation -> Neuronal Mitochondrial Complex II Inhibition INDIRECT UNKNOWN INTERMEDIATES The alternative account of the delayed neuropathy, drawn so the competing hypothesis has an edge to attach to rather than existing only as prose. A review names this metabolite as the causal moiety in the cranial neuropathies; no experimental result supports it.
The neuropathy is conditional on renal failure
aki_dependent_metabolite_retention ALTERNATIVE
Delayed neurotoxicity appears only in those who sustain renal injury, in both a human case series and two rat studies, where brain diglycolic acid was markedly higher in animals that developed acute kidney injury than in those that did not. On this account the nerve injury is not an independent arm but a consequence of losing the clearance route for the metabolite. It is not exclusive of direct neurotoxicity - it supplies the exposure that direct toxicity would then act on.
Pathograph links
Diglycolic Acid Formation -> Neuronal Mitochondrial Complex II Inhibition INDIRECT UNKNOWN INTERMEDIATES The neuronal arm is drawn with unknown intermediates on purpose, and carries two hypothesis groups because two accounts share this edge and are not exclusive. What is demonstrated is that diglycolic acid inhibits complex II in a neuronal cell line, and that brain diglycolic acid is markedly higher in rats that develop kidney injury than in rats that do not; how it reaches human cranial nerve is not established.
?

Discussions and Knowledge Gaps

2
Is the delayed cranial and peripheral neuropathy of diethylene glycol poisoning caused by diglycolic acid acting directly on nerve, by 2-hydroxyethoxyacetic acid, or by retention of metabolite that only happens once the kidney has failed?
KNOWLEDGE GAP delayed_neuropathy_mechanism
Attached to
pathophysiology#Neuronal Mitochondrial Complex II Inhibition phenotypes#Facial palsy
Three accounts are live and this entry carries all three as hypothesis groups. The direct-toxicity account has the only positive mechanistic data, and it is an in vitro neuroblastoma result at millimolar concentrations. The 2-hydroxyethoxyacetic acid account comes from a review that states it as what is thought to be the case. The renal-dependence account has the most human support - a case series in which only the patients with established renal impairment went on to develop cranial neuropathy - and is echoed in two rat studies where brain metabolite accumulated only in animals with kidney injury. The three are not mutually exclusive: renal failure could supply the exposure that direct toxicity then acts on. What no source settles is why the facial nerve in particular.
Show evidence (4 references)
PMID:15902788 SUPPORT PRIMARY RESULT Human Clinical
"Established renal impairment may predict subsequent delayed neurologic toxicity."
The human observation behind the renal-dependence account, stated by its authors with "may" rather than as a settled relation.
PMID:21192754 SUPPORT REVIEW SYNTHESIS Other
"2-hydroxyethoxyacetic acid is thought to be the causal moiety in cranial neuropathies resulting from diethylene glycol toxicity."
The competing attribution, carried with its hedge. Note it names the other metabolite - not the one this entry's renal arm blames.
PMID:21192754 SUPPORT REVIEW SYNTHESIS Other
"Further research is required to explain why the facial nerve is the cranial nerve most commonly involved and why the basal ganglia are predisposed to injury."
The gap stated by a source rather than inferred by this entry.
+ 1 more reference
If diethylene glycol is not metabolised to oxalate, why do calcium oxalate crystals appear in the kidneys of some human cases?
KNOWLEDGE GAP oxalate_in_human_deg_cases
Attached to
pathophysiology#Proximal Tubular ATP Depletion and Necrosis
The rat metabolic study found no urinary oxalate or glycolate and concluded that toxicity does not proceed through conversion to ethylene glycol. A human autopsy case nonetheless describes calcium oxalate crystals in both kidneys. Candidate explanations the cited sources do not distinguish between: ethylene glycol generated endogenously as a minor metabolite of diethylene glycol - the rat study detected small amounts of it in urine even while finding no oxalate - ethylene glycol present as a contaminant alongside diethylene glycol in the ingested product, a species difference in that minor route, or oxalate of dietary or endogenous origin deposited in already necrotic tubules. An earlier draft of this rationale offered only the product-impurity explanation, which misread the quoted sentence: the ethylene glycol it reports was measured in the urine of dosed rats, so it is a metabolite, not an impurity. The entry records the conflict because the tidy version of the claim - that this poisoning never produces oxalate - is contradicted by a source it cites, and because the diagnostic advice that follows from it matters at the bedside.
Show evidence (3 references)
PMID:20530232 SUPPORT PRIMARY RESULT Model Organism
"Small amounts of ethylene glycol (EG), but not oxalate or glycolate, were observed in the urine."
The metabolic result: the oxalate route that drives ethylene glycol nephrotoxicity is not detectably engaged here.
PMID:28085681 REFUTE PRIMARY RESULT Human Clinical
"Both kidneys illustrate tubular necrosis with scattered tubular deposition of polarizable calcium oxalate crystals."
REFUTE against the general claim that this poisoning does not deposit oxalate. It does not refute the diglycolic acid mechanism, which the same kidneys illustrate as tubular necrosis - the object is the oxalate claim alone.
PMID:25170933 SUPPORT PRIMARY RESULT Human Clinical
"However, the difference in metabolism and the absence of oxalate crystalluria or oxalate deposition in sections of tissue examined14,15 favor that toxicity may be through different mechanisms."
A human series pointing the other way from the autopsy above - no oxalate crystalluria and none in the tissue examined - which is what makes this a genuine two-sided question rather than one anomalous report against a rat study. The inline "14,15" is the source's own citation markers, quoted as they appear.
⚙

Pathophysiology

11
Systemic Diethylene Glycol Exposure
Ingested diethylene glycol is absorbed rapidly and distributes to the kidney and brain ahead of other tissues, delivering the parent compound to the liver where it is bioactivated.
Downstream
Alcohol Dehydrogenase Oxidation of Diethylene Glycol
Show evidence (1 reference)
PMID:8135655 SUPPORT PRIMARY RESULT Model Organism
"14C-DEG was rapidly distributed from the blood into the organs and tissues in the order kidneys > brain > spleen > liver > muscle > fat, i.e. the same order as the blood flow."
Sources the distribution claim in this node's description, including the ordering and the reason for it. The two organs heading the list are the two this disease damages.
Alcohol Dehydrogenase Oxidation of Diethylene Glycol
Alcohol dehydrogenase oxidises diethylene glycol to 2-hydroxyethoxyacetaldehyde. This is the single step the antidote blocks, and it is separated from the aldehyde dehydrogenase step below precisely so that the antidote's target can be named exactly.
alcohol dehydrogenase (NAD+) activity GO:0004022 Gene Ontology (GO) Relation: this pathophysiological event involves this molecular function This pathophysiological event involves alcohol dehydrogenase (NAD+) activity (GO:0004022). GO:0004022 is a molecular function from the Gene Ontology.
Downstream
2-Hydroxyethoxyacetic Acid Formation
Show evidence (1 reference)
PMID:20530232 SUPPORT PRIMARY RESULT Model Organism
"These results indicate that the mechanism for the target organ toxicity results from metabolites of DEG, and not DEG itself nor formation of EG from DEG, and that fomepizole may be a useful antidote for treating DEG poisoning."
The causal keystone of this entry. It establishes that bioactivation is required, and rules out the alternative that diethylene glycol is toxic by conversion to ethylene glycol.
2-Hydroxyethoxyacetic Acid Formation
Aldehyde dehydrogenase oxidises the aldehyde to 2-hydroxyethoxyacetic acid, the quantitatively dominant urinary metabolite. It is not the renal toxicant - it produces no death in human proximal tubule cells - but it is an organic acid and so contributes to the acidosis, and one review proposes it rather than diglycolic acid as the neurotoxic moiety.
2-hydroxyethoxyacetic acid is the neurotoxic moiety Diglycolic acid is the proximate renal toxicant
aldehyde dehydrogenase (NAD+) activity GO:0004029 Gene Ontology (GO) Relation: this pathophysiological event involves this molecular function This pathophysiological event involves aldehyde dehydrogenase (NAD+) activity (GO:0004029). GO:0004029 is a molecular function from the Gene Ontology.
Downstream
Diglycolic Acid Formation Diglycolic acid is the proximate renal toxicant
Organic Acid Accumulation
Neuronal Mitochondrial Complex II Inhibition The alternative account of the delayed neuropathy, drawn so the competing hypothesis has an edge to attach to rather than existing only as prose. A review names this metabolite as the causal moiety in the cranial neuropathies; no experimental result supports it. 2-hydroxyethoxyacetic acid is the neurotoxic moiety
Show evidence (1 reference)
PMID:20530232 SUPPORT PRIMARY RESULT Model Organism
"After low and high doses of DEG, 2-hydroxyethoxyacetic acid (HEAA) was the primary metabolite in the urine, with only minor amounts of urinary diglycolic acid (DGA)."
Establishes this metabolite as the dominant one by quantity, which is the asymmetry that makes the toxicity of the minor metabolite surprising and worth modelling separately.
Diglycolic Acid Formation
Diglycolic acid is a four-carbon dicarboxylic acid with an ether oxygen, close enough in shape to succinate that the kidney handles it as though it were a citric acid cycle intermediate. That mistaken identity is the reason the lesion is renal and proximal.
Diglycolic acid is directly neurotoxic Diglycolic acid is the proximate renal toxicant The neuropathy is conditional on renal failure
Animal models
High-dose diethylene glycol gavage in the rat Rat recapitulates
Downstream
Dicarboxylate Transporter-Mediated Renal Uptake Diglycolic acid is the proximate renal toxicant
Neuronal Mitochondrial Complex II Inhibition The neuronal arm is drawn with unknown intermediates on purpose, and carries two hypothesis groups because two accounts share this edge and are not exclusive. What is demonstrated is that diglycolic acid inhibits complex II in a neuronal cell line, and that brain diglycolic acid is markedly higher in rats that develop kidney injury than in rats that do not; how it reaches human cranial nerve is not established. Diglycolic acid is directly neurotoxic The neuropathy is conditional on renal failure
Show evidence (1 reference)
PMID:21856646 SUPPORT PRIMARY RESULT In Vitro
"DGA is structurally similar to citric acid cycle intermediates that are taken up by specific transporters in kidney cells."
States the structural mimicry this node rests on, and does so in the paper that identified diglycolic acid as the toxic metabolite.
Dicarboxylate Transporter-Mediated Renal Uptake
Sodium-dicarboxylate cotransporters on the proximal tubular cell, whose normal job is reclaiming filtered succinate, carry diglycolic acid into the cell.
Diglycolic acid is the proximate renal toxicant
renal proximal tubular epithelial cell CL:0002306 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves renal proximal tubular epithelial cell, annotated with epithelial cell of proximal tubule (CL:0002306). CL:0002306 is a cell type from the Cell Ontology.
SLC13A2 hgnc:10917 HUGO Gene Nomenclature Committee (hgnc) Relation: this pathophysiological event involves this gene This pathophysiological event involves SLC13A2 (hgnc:10917). hgnc:10917 is a gene from the HUGO Gene Nomenclature Committee.
dicarboxylic acid transport GO:0006835 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves dicarboxylic acid transport (GO:0006835). GO:0006835 is a biological process from the Gene Ontology.
proximal tubule UBERON:0004134 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in proximal tubule (UBERON:0004134). UBERON:0004134 is an anatomical location from the Uberon multi-species anatomy ontology.
Downstream
Intracellular Diglycolic Acid Trapping Diglycolic acid is the proximate renal toxicant
Show evidence (2 references)
PMID:36939119 SUPPORT PRIMARY RESULT Model Organism
"Significantly higher expression of sodium-dependent dicarboxylate transporter-1 mRNA was present in the kidneys of rats with acute kidney injury and diglycolic acid accumulation compared to those in rats that had no diglycolic acid in their kidneys and no acute kidney injury."
Makes this node's transporter the locus of individual susceptibility, not merely the route of entry: identically dosed rats diverged, and the ones that were injured were the ones expressing more transporter. It is the best current explanation for why a given dose kills some people and not others.
PMID:36087010 SUPPORT PRIMARY RESULT In Vitro
"Pretreatment with the NaDC-1 inhibitor N-(p-amylcinnamoyl)anthranilic acid (ACA) markedly reduced apical uptakes of both succinate and DGA."
Inhibiting the succinate transporter reduces diglycolic acid uptake, which is what makes this a transporter-mediated step rather than passive entry.
Intracellular Diglycolic Acid Trapping
Once inside, diglycolic acid is not effluxed. The cell has a route in and effectively no route out, so concentration builds - in rats kidney concentrations reach far above blood levels - which is why a metabolite circulating at modest concentration produces a lesion confined to this one cell type.
Diglycolic acid is the proximate renal toxicant
Downstream
Renal Mitochondrial Complex II Inhibition Diglycolic acid is the proximate renal toxicant
Show evidence (1 reference)
PMID:36958672 SUPPORT PRIMARY RESULT In Vitro
"This study concluded that DGA is poorly transported out of cells and that stimulation of OAT transporters is not a viable target for reducing DGA accumulation in cells."
Establishes the one-way character of the step, and reports the negative result that the obvious efflux route cannot be recruited to reverse it.
Renal Mitochondrial Complex II Inhibition
Accumulated diglycolic acid inhibits succinate dehydrogenase, the complex II of the respiratory chain, and depolarises the mitochondrion. The selectivity is striking - complexes I, III and IV are untouched - and is what one would predict from a succinate mimic.
Diglycolic acid is the proximate renal toxicant
mitochondrial depolarization GO:0051882 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves increased mitochondrial depolarization (GO:0051882). GO:0051882 is a biological process from the Gene Ontology. ↑ INCREASED
succinate dehydrogenase activity GO:0000104 Gene Ontology (GO) Relation: this pathophysiological event involves this molecular function This pathophysiological event involves decreased succinate dehydrogenase activity (GO:0000104). GO:0000104 is a molecular function from the Gene Ontology. ↓ DECREASED
Downstream
Proximal Tubular ATP Depletion and Necrosis Diglycolic acid is the proximate renal toxicant
Show evidence (2 references)
PMID:27002734 SUPPORT PRIMARY RESULT In Vitro
"DGA inhibited Complex II activity without altering Complex I, III or IV activities."
The specificity claim this node makes, stated with the negative controls that give it force.
PMID:27002734 SUPPORT PRIMARY RESULT In Vitro
"DGA decreased the mitochondrial membrane potential in HPT cells."
The depolarisation half of the node, measured in human proximal tubule cells rather than isolated mitochondria.
Proximal Tubular ATP Depletion and Necrosis
Loss of oxidative ATP production in a cell type that depends on it kills by necrosis rather than apoptosis - the caspase machinery is not engaged, and the cells simply run out of energy.
Diglycolic acid is the proximate renal toxicant
renal proximal tubular epithelial cell CL:0002306 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves renal proximal tubular epithelial cell, annotated with epithelial cell of proximal tubule (CL:0002306). CL:0002306 is a cell type from the Cell Ontology.
ATP biosynthetic process GO:0006754 Gene Ontology (GO) Relation: this pathophysiological event involves this biological process This pathophysiological event involves decreased ATP biosynthetic process (GO:0006754). GO:0006754 is a biological process from the Gene Ontology. ↓ DECREASED
Experimental models
Primary human proximal tubule cells PRIMARY CELL CULTURE Reproduces the cell death and its mode in the human target cell type.
Animal models
Diglycolic acid gavage in the rat Rat recapitulates Reproduces the renal injury from the metabolite alone. Retargeted from the transporter node during review: what this study measures is tissue accumulation and organ injury, not transporter-mediated uptake.
Downstream
Tubular Luminal Obstruction Diglycolic acid is the proximate renal toxicant
Show evidence (2 references)
PMID:21856646 SUPPORT PRIMARY RESULT In Vitro
"DGA induced necrosis, not apoptosis, as corroborated by severe depletion of cellular adenosine triphosphate levels."
Both halves of this node in one sentence: the mode of death and the energetic failure behind it.
PMID:21856646 SUPPORT PRIMARY RESULT In Vitro
"DEG and 2-HEAA did not produce any cell death. DGA produced dose-dependent necrosis at concentrations above 25 mmol/l."
The negative control that makes the whole canonical arm work - neither the parent compound nor the first metabolite kills these cells.
Tubular Luminal Obstruction
The swollen and vacuolated necrotic epithelium occludes the tubular lumen. Urine flow falls for a mechanical reason on top of the cellular one, which is why the renal failure of this poisoning is characteristically anuric rather than merely oliguric.
kidney UBERON:0002113 Uberon multi-species anatomy ontology (UBERON) Relation: this pathophysiological event occurs in this anatomical location This pathophysiological event occurs in kidney (UBERON:0002113). UBERON:0002113 is an anatomical location from the Uberon multi-species anatomy ontology.
Downstream
Acute kidney injury
Anuria
Show evidence (1 reference)
PMID:37186212 SUPPORT REVIEW SYNTHESIS Other
"Marked vacuolization and edema of epithelial cells obstruct the lumen, reducing urine flow and, consequently, resulting in anuria and uremia."
States the obstruction step and its clinical consequence directly. Graded OTHER and REVIEW_SYNTHESIS because the source is a narrative overview summarising renal biopsy morphology rather than reporting it.
Organic Acid Accumulation
2-hydroxyethoxyacetic acid and diglycolic acid are unmeasured anions, and their accumulation produces the high anion gap metabolic acidosis of the early phase. This arm is separable from the renal arm: the antidote prevents both, and the acidosis appears before renal failure is established.
Downstream
Metabolic acidosis
Show evidence (1 reference)
PMID:20530232 SUPPORT PRIMARY RESULT Model Organism
"Rats treated with high-dose DEG had metabolic acidosis, increased BUN and creatinine, and marked kidney necrosis, noted by histopathology."
Reports the acidosis alongside the renal lesion in the same dosed animals, which is what this node and its position in the chain assert.
Neuronal Mitochondrial Complex II Inhibition
The same complex II inhibition demonstrated in renal cells occurs in a neuronal cell line, with mitochondrial depolarisation, ATP depletion and inflammatory cytokine release. This is the proposed substrate of the delayed neurological syndrome, and it is an in vitro result rather than an observation in nerve.
Diglycolic acid is directly neurotoxic
neuron CL:0000540 Cell Ontology (CL) Relation: this pathophysiological event involves this cell type This pathophysiological event involves neuron (CL:0000540). CL:0000540 is a cell type from the Cell Ontology.
succinate dehydrogenase activity GO:0000104 Gene Ontology (GO) Relation: this pathophysiological event involves this molecular function This pathophysiological event involves decreased succinate dehydrogenase activity (GO:0000104). GO:0000104 is a molecular function from the Gene Ontology. ↓ DECREASED
Experimental models
SH-SY5Y neuroblastoma line CELL LINE Demonstrates complex II inhibition, depolarisation and ATP depletion in a neuronal context.
Animal models
Subacute diethylene glycol dosing with motor testing in the rat Rat partially recapitulates Produces measurable motor impairment and brain metabolite accumulation, but not the cranial neuropathy that defines the human delayed syndrome.
Downstream
Facial palsy Diglycolic acid is directly neurotoxic
Sensorimotor neuropathy Diglycolic acid is directly neurotoxic
Show evidence (1 reference)
PMID:36754214 SUPPORT PRIMARY RESULT In Vitro
"These studies demonstrate that the DGA-induced mechanism of SH-SY5Y cell death involves complex II inhibition leading to mitochondrial depolarization, and subsequent ATP depletion with accompanying inflammatory cytokine release."
The whole node in one sentence, and explicitly in a neuronal line rather than in the kidney cells where the mechanism was first shown.
⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.
Pathograph: causal mechanism network for Diethylene Glycol Poisoning Interactive directed graph showing how pathophysiology mechanisms, phenotypes, genetic factors and variants, experimental models, environmental triggers, and treatments relate through causal and linked edges.
●

Phenotypes

17
Blood 1
Gastrointestinal hemorrhage HP:0002239 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is gastrointestinal bleeding, annotated with Gastrointestinal hemorrhage (HP:0002239). HP:0002239 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:21062849 SUPPORT REVIEW SYNTHESIS Other
"These children have presented with gastrointestinal bleeding, seizures and liver failure as well as renal failure."
Names gastrointestinal bleeding among the presenting features across the paediatric epidemics.
Cardiovascular 1
Hypertension HP:0000822 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Hypertension (HP:0000822). HP:0000822 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:22246572 SUPPORT PRIMARY RESULT Human Clinical
"Compared to children who had not ingested it, they were significantly (p < 0.05) younger (11.95 vs. 31 months), more were anuric (98.3 vs. 74.6%), hypertensive (84 vs. 52%), had severe metabolic acidosis (46.7 vs. 20.5%), and died (96.6 vs. 71.2%)."
Hypertension in 84% of exposed children against 52% of unexposed children with acute kidney injury, so the figure is a contrast rather than a bare prevalence.
Digestive 2
Gastrointestinal prodrome Nausea and vomiting HP:0002017 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is nausea, vomiting and abdominal pain, annotated with Nausea and vomiting (HP:0002017), qualified as temporality acute. HP:0002017 is a phenotype from the Human Phenotype Ontology.
Temporal: ACUTE
Modelled as one phenotype rather than four because the cited sentence names the phase collectively ("gastrointestinal symptoms") and does not itemise nausea, vomiting, abdominal pain and diarrhoea separately. Splitting it into four would assert four findings where the source reports one. `HP:0002017` Nausea and vomiting is bound as the closest single term; `preferred_term` carries the fuller picture.
Show evidence (1 reference)
PMID:37186212 SUPPORT REVIEW SYNTHESIS Other
"Patients may present with gastrointestinal symptoms and anion gap metabolic acidosis, followed by renal failure, and, later, encephalopathy and neuropathy."
Names the gastrointestinal presentation as the first of the three phases. The same sentence is cited on Metabolic acidosis for the other half of that phase.
Hepatic failure HP:0001399 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is liver failure, annotated with Hepatic failure (HP:0001399). HP:0001399 is a phenotype from the Human Phenotype Ontology.
Show evidence (2 references)
PMID:28074668 SUPPORT PRIMARY RESULT Model Organism
"DGA concentrations in the kidneys and liver correlated with renal and hepatic injury, respectively."
The same metabolite accumulation drives injury in both organs, which is why the hepatic involvement is carried here rather than treated as incidental.
PMID:21062849 SUPPORT REVIEW SYNTHESIS Other
"These children have presented with gastrointestinal bleeding, seizures and liver failure as well as renal failure."
Names liver failure among the presenting features in children.
Eye 1
Optic neuropathy HP:0001138 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is optic nerve injury, annotated with Optic neuropathy (HP:0001138). HP:0001138 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:21192754 SUPPORT REVIEW SYNTHESIS Other
"Diethylene glycol ingestion has been associated with the development of optic nerve injury, cranial nerve deficits, quadraparesis and peripheral neuropathy."
Names optic nerve injury among the delayed sequelae.
Genitourinary 2
Acute kidney injury HP:0001919 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Acute kidney injury (HP:0001919), qualified as temporality acute. HP:0001919 is a phenotype from the Human Phenotype Ontology.
Temporal: ACUTE
Show evidence (1 reference)
PMID:37186212 SUPPORT REVIEW SYNTHESIS Other
"The kidney injury in DEG poisoning is secondary to proximal tubular necrosis caused by DGA."
Names the phenotype and attributes it to the mechanism this entry models.
Anuria HP:0100519 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Anuria (HP:0100519), qualified as temporality acute. HP:0100519 is a phenotype from the Human Phenotype Ontology.
Temporal: ACUTE
Show evidence (1 reference)
PMID:15902788 SUPPORT PRIMARY RESULT Human Clinical
"Initially three patients, ingesting the largest quantities of DEG, presented more than 24 h postingestion with severe metabolic acidemia (pH range 6.8-7.1) and anuric acute renal failure requiring HD."
Anuric renal failure observed directly, and tied to the largest ingested quantities.
Head and Neck 1
Facial palsy FREQUENT Bilateral facial palsy HP:0430025 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is bilateral facial palsy (HP:0430025). HP:0430025 is a phenotype from the Human Phenotype Ontology.
Frequency from the Panama series as summarised by a later case series ("more than half"), which is why the frequency evidence below is graded REVIEW_SYNTHESIS. An earlier draft cited `PMID:21192754` here. That paper names cranial nerve deficits for diethylene glycol without saying which, and the one sentence in it that does name the seventh nerve attributes it to **ethylene** glycol - the neighbouring toxin this entry's top `notes` exists to keep separate. It was the wrong citation for this phenotype and was replaced rather than left to look adequate.
Show evidence (2 references)
PMID:33475432 SUPPORT BACKGROUND Human Clinical
"we were not able to observe bilateral facial paresis or “facial drooping” in rats, a specific neurological effect seen in many DEG human cases."
Names bilateral facial paresis as a specific effect of this poisoning in humans. HUMAN_CLINICAL with quote_role BACKGROUND because the citing paper is a rat study and this sentence reports the human picture it could not reproduce - grading it MODEL_ORGANISM would assert that rats showed the sign, which is the opposite of what the sentence says.
PMID:25170933 SUPPORT REVIEW SYNTHESIS Human Clinical
"More than half of the patients had lower motor neuron facial palsy."
The frequency band. "More than half" places it in FREQUENT without supporting a point estimate, so none is given.
Limbs 1
Limb muscle weakness FREQUENT HP:0003690 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is limb weakness, annotated with Limb muscle weakness (HP:0003690). HP:0003690 is a phenotype from the Human Phenotype Ontology.
Bound to `HP:0003690` Limb muscle weakness, not `HP:0003324` Generalized muscle weakness. Those two are siblings under `HP:0001324` Muscle weakness, not a general and a specific form of one thing: HPO defines the first as weakness "of the muscles of the arms and legs" and the second as "affecting both distal and proximal musculature". The source says limb weakness, so the distribution is the arms and legs. An earlier draft of this phenotype bound `HP:0003324`, which asserted a different distribution from its own snippet. Frequency is from the Panama 2006 series as reported by a later case series, not by that series itself, so the evidence is graded REVIEW_SYNTHESIS - the citing paper is summarising Sosa et al rather than counting its own patients.
Show evidence (1 reference)
PMID:25170933 SUPPORT REVIEW SYNTHESIS Human Clinical
"The most common neurological finding was limb weakness in 31 (78%) of whom almost all had reduced or absent deep tendon reflexes."
Both the rank and the figure behind the FREQUENT band, and the areflexia that accompanies it.
Metabolism 2
Metabolic acidosis HP:0001942 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Metabolic acidosis (HP:0001942), qualified as temporality acute. HP:0001942 is a phenotype from the Human Phenotype Ontology.
Temporal: ACUTE
Show evidence (1 reference)
PMID:37186212 SUPPORT REVIEW SYNTHESIS Other
"Patients may present with gastrointestinal symptoms and anion gap metabolic acidosis, followed by renal failure, and, later, encephalopathy and neuropathy."
Places the acidosis in the first of the three phases, which is why this phenotype hangs off the organic acid node rather than off renal failure.
Cerebral edema HP:0002181 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Cerebral edema (HP:0002181). HP:0002181 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:15902788 SUPPORT PRIMARY RESULT Human Clinical
"One patient died with cerebral oedema and a progressive encephalopathy."
Cerebral oedema as a fatal outcome of the delayed phase in this series.
Musculoskeletal 1
Bulbar palsy HP:0001283 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Bulbar palsy (HP:0001283). HP:0001283 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:15902788 SUPPORT PRIMARY RESULT Human Clinical
"All three remained dialysis-dependent and developed significant cranial neuropathies with bulbar palsy in the second week postingestion."
Bulbar palsy observed directly, with the second-week timing that makes this the delayed rather than the acute syndrome.
Nervous System 5
Autonomic dysfunction FREQUENT Abnormal autonomic nervous system physiology HP:0012332 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is signs suggestive of autonomic dysfunction, annotated with Abnormal autonomic nervous system physiology (HP:0012332). HP:0012332 is a phenotype from the Human Phenotype Ontology.
Bound to `HP:0012332`, the functional child, rather than its parent `HP:0002270` Abnormality of the autonomic nervous system, because the claim is about dysfunction rather than about structure. The deep research report offered `HP:0012647` for autonomic dysfunction; that identifier is Abnormal inflammatory response and was not used. Frequency is from the Panama 2006 series as reported by a later case series, not by that series itself, hence REVIEW_SYNTHESIS.
Show evidence (1 reference)
PMID:25170933 SUPPORT REVIEW SYNTHESIS Human Clinical
"Sixty three percent developed signs suggestive of autonomic dysfunction."
The figure behind the FREQUENT band. Quoted in the source's own wording, which spells the number and says "signs suggestive of" rather than asserting the diagnosis.
Sensorimotor neuropathy HP:0007141 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is sensorimotor polyneuropathy, annotated with Sensorimotor neuropathy (HP:0007141). HP:0007141 is a phenotype from the Human Phenotype Ontology.
No `frequency` is recorded. The strongest available figure - severe sensorimotor axonal neuropathy in 19 of 21 - is conditional on having limb weakness and on having been tested, so it is a frequency within a selected subgroup rather than among poisoned patients. Entering it as a disease-level band would overstate it, so the numerator and denominator are carried in the evidence instead.
Show evidence (2 references)
PMID:25170933 SUPPORT REVIEW SYNTHESIS Human Clinical
"Nerve conduction on 21 patients with limb weakness showed evidence of severe sensorimotor axonal neuropathy in 19."
Electrophysiological confirmation in a cohort, with its denominator stated - the 21 are those with limb weakness who were tested, not the whole series.
PMID:15824363 SUPPORT PRIMARY RESULT Human Clinical
"These data indicate that DEG intoxication may cause a primary acute axonal sensorimotor polyneuropathy with demyelinating physiology during recovery."
Characterises the neuropathy electrophysiologically, and is the source of this phenotype's description distinguishing the acute from the recovery picture.
Encephalopathy HP:0001298 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Encephalopathy (HP:0001298). HP:0001298 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:15824363 SUPPORT PRIMARY RESULT Human Clinical
"The authors report a 24-year-old man who developed encephalopathy and rapid quadriplegia following ingestion of a solution containing diethylene glycol (DEG)."
Encephalopathy observed in a documented single ingestion.
Tetraparesis OCCASIONAL HP:0002273 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is quadriparesis, annotated with Tetraparesis (HP:0002273). HP:0002273 is a phenotype from the Human Phenotype Ontology.
Frequency is from the Panama 2006 series as reported by a later case series, not by that series itself, so the evidence is graded REVIEW_SYNTHESIS - the citing paper is summarising Sosa et al rather than counting its own patients.
Show evidence (2 references)
PMID:21192754 SUPPORT REVIEW SYNTHESIS Other
"Diethylene glycol ingestion has been associated with the development of optic nerve injury, cranial nerve deficits, quadraparesis and peripheral neuropathy."
Names quadraparesis among the delayed sequelae.
PMID:25170933 SUPPORT REVIEW SYNTHESIS Human Clinical
"Ten patients (22%) developed progressive flaccid paralysis culminating in quadriparesis."
The figure behind the OCCASIONAL band, and the reason this phenotype is distinguished from the commoner limb weakness rather than merged with it.
Seizure HP:0001250 Human Phenotype Ontology (HP) Relation: this clinical feature is this phenotype This clinical feature is Seizure (HP:0001250). HP:0001250 is a phenotype from the Human Phenotype Ontology.
Show evidence (1 reference)
PMID:21062849 SUPPORT REVIEW SYNTHESIS Other
"These children have presented with gastrointestinal bleeding, seizures and liver failure as well as renal failure."
Names seizures among the presenting features in children.
💊

Medical Actions

4
Fomepizole
Action: PharmacotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Pharmacotherapy (NCIT:C15986). NCIT:C15986 is a clinical intervention from the NCI Thesaurus. NCIT:C15986
Agent: fomepizole CHEBI:5141 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses fomepizole (CHEBI:5141). CHEBI:5141 is a therapeutic agent from Chemical Entities of Biological Interest.
Platform: Small molecule
The antidote, and unusually for a toxidrome its rationale is fully worked out: it inhibits alcohol dehydrogenase, so the parent compound is never converted to the acids that do the damage. In the rat model it prevents the metabolic acidosis and the renal and hepatic injury outright. Its value is therefore greatest before metabolism has occurred, and it cannot undo a lesion that diglycolic acid has already produced.
Mechanism Target:
INHIBITS Alcohol Dehydrogenase Oxidation of Diethylene Glycol — Blocks the first oxidation step specifically, which is why the bioactivation chain is modelled as two nodes rather than one - the antidote acts on the alcohol dehydrogenase step alone, and everything downstream of it is prevented as a consequence rather than as a second target.
Show evidence (1 reference)
PMID:20530232 SUPPORT PRIMARY RESULT Model Organism
"Treatment with fomepizole blocked the formation of HEAA and DGA and the development of metabolic acidosis and the kidney and liver toxicity."
Supports this edge precisely: the intervention is shown to act on metabolite formation, and the downstream toxicity is prevented.
Show evidence (1 reference)
PMID:10793034 SUPPORT INDIRECT PRIMARY RESULT Human Clinical
"We report a case of DEG ingestion in a 17-month-old girl who was managed with activated charcoal, fomepizole (a recently available alcohol dehydrogenase inhibitor), and hemodialysis (HD)."
The only human evidence in this entry for the antidote, and it is a single case in which three interventions were given together, so it is INDIRECT: it establishes that fomepizole has been used in this poisoning, not that it was what worked.
Ethanol
Action: PharmacotherapyNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Pharmacotherapy (NCIT:C15986). NCIT:C15986 is a clinical intervention from the NCI Thesaurus. NCIT:C15986
Agent: ethanol CHEBI:16236 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses ethanol (CHEBI:16236). CHEBI:16236 is a therapeutic agent from Chemical Entities of Biological Interest.
Platform: Small molecule
The alternative alcohol dehydrogenase inhibitor, and the one that matters most in practice: fomepizole is expensive and often unavailable in the settings where these outbreaks happen. It occupies the same enzyme as fomepizole and so blocks the same step, but the evidence for it in this poisoning is weaker - the source cited here says inhibition "should be" therapeutic rather than reporting that it was.
Mechanism Target:
INHIBITS Alcohol Dehydrogenase Oxidation of Diethylene Glycol — Competes for the same enzyme fomepizole inhibits, and so is drawn to the same node.
Show evidence (1 reference)
PMID:36939119 SUPPORT INDIRECT BACKGROUND Other
"It is also likely that inhibition of alcohol dehydrogenase, either with ethanol or fomepizole, should be therapeutic."
INDIRECT and BACKGROUND, and deliberately so: this is the authors' framing sentence proposing that the mechanism implies a therapy, not a result. It is the strongest statement about ethanol in this entry's cached sources, and the treatment description says as much rather than implying a demonstrated benefit.
Supportive correction of the metabolic acidosis
Action: Supportive CareNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Supportive Care (NCIT:C15747). NCIT:C15747 is a clinical intervention from the NCI Thesaurus. NCIT:C15747
Agent: sodium bicarbonate CHEBI:32139 Chemical Entities of Biological Interest (CHEBI) Relation: this treatment uses this therapeutic agent This treatment uses sodium bicarbonate, annotated with sodium hydrogencarbonate (CHEBI:32139). CHEBI:32139 is a therapeutic agent from Chemical Entities of Biological Interest.
Platform: Other
Bicarbonate against the acidosis, alongside general supportive care and gastric decontamination. This is what most patients in the outbreaks actually received, since the antidote and reliable dialysis were absent.
Show evidence (2 references)
PMID:36939119 SUPPORT BACKGROUND Other
"Current therapy centers on supportive care such as bicarbonate to combat the acidosis and the use of extracorporeal measures to remove DEG and its metabolites (such removal is implied, although it has not really been shown experimentally) and to fix acid-base abnormalities"
Describes current practice, and is carried with the authors' own parenthesis intact - they note that extracorporeal removal of the toxin is assumed rather than demonstrated, which qualifies the Hemodialysis entry as well as this one.
PMID:37186212 SUPPORT REVIEW SYNTHESIS Other
"The management includes supportive care, gastric decontamination, correction of acid-base disorders, and hemodialysis."
The management list this treatment entry covers the first three elements of.
Hemodialysis
Action: HemodialysisNCI Thesaurus (NCIT) Relation: this treatment is this clinical intervention This treatment is Hemodialysis (NCIT:C15248). NCIT:C15248 is a clinical intervention from the NCI Thesaurus. NCIT:C15248
Platform: Other
Clears the parent compound - measured across a dialysis session in a treated child - and becomes renal replacement once tubular necrosis is established. A human case series reports acidaemia and renal impairment responding to emergent haemodialysis. Its availability matters: the paediatric epidemics in resource-constrained settings record deterioration despite peritoneal dialysis, and their authors call for haemodialysis facilities to be provided. An earlier draft asserted that haemodialysis availability was the main determinant of survival in the outbreaks and that peritoneal dialysis clears these solutes far less efficiently; neither is stated by a cited source and both were removed.
Mechanism Target:
INHIBITS Systemic Diethylene Glycol Exposure — Removes the parent compound from the circulation, lowering the systemic burden that everything downstream depends on. This is the only treatment here that acts on the exposure node itself rather than on a metabolic step.
Show evidence (1 reference)
PMID:10793034 SUPPORT PRIMARY RESULT Human Clinical
"Pre-HD and post-HD DEG levels support clearance of DEG with HD."
Measured clearance of the parent compound across a dialysis session, which is exactly what this edge asserts. Note the qualification carried on the supportive-care entry: one cited source describes extracorporeal removal as implied rather than experimentally shown, and this case measures levels in a single patient.
Show evidence (3 references)
PMID:15902788 SUPPORT PRIMARY RESULT Human Clinical
"We report a series of seven patients presenting with epidemic DEG poisoning from a correctional facility with varying degrees of metabolic acidemia and acute renal impairment responding to emergent hemodialysis (HD)."
The affirmative human evidence: acidaemia and renal impairment responding to emergent haemodialysis.
PMID:10793034 SUPPORT PRIMARY RESULT Human Clinical
"Pre-HD and post-HD DEG levels support clearance of DEG with HD."
Measured clearance of the parent compound across a dialysis session, which is the specific claim that haemodialysis removes the toxin rather than only replacing renal function.
PMID:22246572 SUPPORT INDIRECT PRIMARY RESULT Human Clinical
"They developed increasing metabolic acidosis and multiorgan dysfunction despite peritoneal dialysis."
INDIRECT, and kept for a different point than the two items above: the modality available in that outbreak did not arrest the syndrome, which is why haemodialysis rather than dialysis in general is named here.
🌍

Environmental Factors

1
Ingestion of a liquid medicine adulterated with diethylene glycol
exposure to diethylene glycol ECTO:9002156 Environmental Conditions, Treatments and Exposures Ontology (ECTO) Relation: this environmental factor is this exposure This environmental factor is exposure to diethylene glycol, annotated with exposure to glycol (ECTO:9002156). ECTO:9002156 is an exposure from the Environmental Conditions, Treatments and Exposures Ontology.
ECTO has no exposure class for diethylene glycol, so the genus term is bound and `preferred_term` carries the specificity. The searches run through the OLS `ecto` index were "diethylene glycol", "exposure to diethylene glycol", "diethyleneglycol", "diethylene", "oxydiethanol" and "2-(2-hydroxyethoxy)ethanol". The only ECTO hit among them is `ECTO:9001432` exposure to diethylenetriamine, a different chemical; `CHEBI:46807` diethylene glycol exists but ECTO has not generated an exposure class from it. Two near-misses were rejected on positive grounds rather than overlooked. `ECTO:9001111` exposure to ethylene glycol is exact for the wrong toxin - binding it would assert the confusion this entry's top `notes` exists to prevent. `ECTO:9002161` exposure to glycol ether was rejected on narrower grounds than an earlier draft claimed. That draft said the term names the monoalkyl ethers of ethylene glycol, whose target organs are testis and bone marrow; that is industrial-hygiene usage, not the ontology's definition, and ChEBI defines a glycol ether structurally as a hydroxyether - which diethylene glycol is. The term is therefore arguably correct and simply less apt than the bound one: `exposure to glycol` is the reading a curator would expect for this compound, and `glycol ether` invites the industrial-hygiene reading and its wrong target organs.
The dominant exposure route and the one behind every large outbreak. Diethylene glycol is substituted for pharmaceutical grade glycerin or propylene glycol in a syrup or elixir, usually as a cheaper adulterant introduced upstream in the excipient supply chain rather than by the finished-product manufacturer.
Show evidence (1 reference)
PMID:18949211 SUPPORT PRIMARY RESULT Human Clinical
"After controlling for pre-existing hypertension and renal disease and the use of angiotensin-converting enzyme inhibitors, a significant association was found between ingestion of prescription cough syrup and illness onset (adjusted odds ratio: 31.0, 95% confidence interval: 6.93-138)."
The epidemiological link between the medicine and the illness, adjusted for the obvious renal confounders.
Mechanism Target:
TRIGGERS Systemic Diethylene Glycol Exposure — The contaminated medicine is the entire source of the systemic burden; there is no endogenous diethylene glycol.
Show evidence (1 reference)
PMID:18949211 SUPPORT PRIMARY RESULT Human Clinical
"Laboratory analyses confirmed the presence of diethylene glycol (DEG) in biological samples from case patients, 8% DEG contamination in cough syrup samples and 22% contamination in the glycerin used to prepare the cough syrup."
Traces the exposure from patient tissue back through the finished syrup to the contaminated glycerin, which is exactly the route this edge asserts.
🔬

Diagnosis

2
Serum diethylene glycol and metabolite quantification
The definitive test, and rarely available where the outbreaks happen. Chromatographic measurement of the parent compound and its acid metabolites is what confirmed the exposure route in the Panama investigation, working from patient samples back to the contaminated excipient.
chromatographic quantification of serum diethylene glycol and its metabolites NCIT:C18020 NCI Thesaurus (NCIT)
Bound to the generic procedure term because the specific one is out of range. `NCIT:C111203` Gas Chromatography Mass Spectrometry names the technique exactly but is not reachable from `NCIT:C25218` Clinical Intervention or Procedure, the root of the `TreatmentActionTerm` enum that `diagnosis_term` uses, so `just validate-terms` rejects it - the same class of failure as binding a device term in a clinical-action slot. `preferred_term` carries the technique and the analytes.
Show evidence (1 reference)
PMID:18949211 SUPPORT PRIMARY RESULT Human Clinical
"Laboratory analyses confirmed the presence of diethylene glycol (DEG) in biological samples from case patients, 8% DEG contamination in cough syrup samples and 22% contamination in the glycerin used to prepare the cough syrup."
Laboratory confirmation in patient samples and in the product, which is what this test establishes.
Clinical recognition without toxicological confirmation
The diagnosis that is actually made in most of the world. Where assay is unavailable, the combination of a drug history, a worsening metabolic acidosis and multiorgan deterioration that does not respond to dialysis is what raises suspicion - and the authors who set out this approach are explicit that identifying the syndrome this way is difficult.
Show evidence (3 references)
PMID:22246572 SUPPORT PRIMARY RESULT Human Clinical
"Detailed drug history, increasing metabolic acidosis, and multiorgan deterioration despite peritoneal dialysis should arouse suspicion."
The clinical triad this diagnosis rests on, from a study designed around the absence of toxicology facilities.
PMID:33475432 SUPPORT INDIRECT BACKGROUND Other
"Blood work shows increased BUN and creatinine levels and osmolal gap along with a metabolic acidosis, but such results could allude to a spectrum of illnesses"
Why routine bloods cannot make this diagnosis: the abnormalities are real but non-specific. INDIRECT and BACKGROUND because the citing paper is a rat study and this is its framing of the human diagnostic problem.
PMID:10793034 SUPPORT INDIRECT BACKGROUND Human Clinical
"Patients presenting less than 12 hours after DEG ingestion may not show metabolic acidosis, whereas those presenting later may show florid metabolic acidosis."
INDIRECT and a caution rather than a criterion: the acidosis that anchors clinical recognition is time-dependent, so an early presentation can look reassuring. Graded BACKGROUND because the sentence states the established kinetic picture the case report is framed against.
📊

Prevalence

3
Worldwide paediatric mass poisonings reported up to 2010
Cases In Literature Ultra Rare
A cumulative death toll across epidemics rather than a rate in a defined population at risk, which is the only form this disease's occurrence takes - there is no background incidence between outbreaks.
Show evidence (1 reference)
PMID:21062849 SUPPORT REVIEW SYNTHESIS Other
"More than 300 children worldwide have died from DEG poisoning."
The cumulative paediatric figure as of this review, carried with its date attached rather than as a current total.
The Gambia, 2022 outbreak
Cases In Literature Ultra Rare
Included because the entry's other records stop in 2010 and this is not a historical disease: the 2022 wave, traced to syrup medications from a single manufacturer, reached The Gambia, Indonesia and Uzbekistan.
Show evidence (1 reference)
PMID:36862590 SUPPORT PRIMARY RESULT Human Clinical
"The preliminary investigation suggested that various contaminated syrup-based children's medications contributed to the AKI outbreak."
The outbreak investigation's own finding, stated with the hedge its authors used - this was a preliminary attribution at publication.
Panama, 2006 outbreak
Cases In Literature Ultra Rare
Adults rather than children, with a median age of 68 - unusual among these outbreaks and a consequence of the vehicle being a prescription cough syrup rather than a paediatric preparation.
Show evidence (1 reference)
PMID:18949211 SUPPORT PRIMARY RESULT Human Clinical
"Twelve (57%) of 21 patients had died of the illness."
The case fatality among the initially identified patients that prompted the outbreak investigation.
🧫

Experimental Models

2
Primary human proximal tubule cells PRIMARY_CELL_CULTURE
Primary cultures from normal human renal cortex, exposed to the parent compound and each metabolite separately. The design is what identifies the toxicant, because it includes the negative arms.
Publication
PMID:21856646
Pathograph links
Proximal Tubular ATP Depletion and Necrosis recapitulates fidelity: moderate observes: cellular Reproduces the cell death and its mode in the human target cell type.
Limitations: Cultured cells lack the filtration and concentration that deliver the metabolite to this cell in vivo, so the millimolar concentrations used are applied directly rather than reached physiologically.
Show evidence (1 reference)
PMID:21856646 SUPPORT PRIMARY RESULT In Vitro
"These studies demonstrate that DGA is the toxic metabolite responsible for DEG-induced proximal tubular necrosis and suggest a possible transporter-mediated uptake of DGA leading to toxic accumulation and cellular dysfunction."
The paper's own statement of what the model established, in the human cell type the disease destroys.
SH-SY5Y neuroblastoma line CELL_LINE
A neuronal line used to ask whether the renal mechanism also operates in nerve. It is the only direct evidence for the emerging neurotoxicity hypothesis.
Publication
PMID:36754214
Pathograph links
Neuronal Mitochondrial Complex II Inhibition recapitulates fidelity: low observes: molecular Demonstrates complex II inhibition, depolarisation and ATP depletion in a neuronal context.
Limitations: A neuroblastoma line is not a cranial motor neuron, and the effective concentrations are millimolar. It shows the mechanism is available to neurons; it does not show it is what injures the facial nerve in poisoned people.
Divergences from the human mechanism
population mismatch qualifying
SH-SY5Y is a neuroblastoma-derived line with an immature catecholaminergic phenotype. The human lesion is in cranial and peripheral motor and sensory nerve, which this line does not model.
Show evidence (1 reference)
PMID:36754214 SUPPORT PRIMARY RESULT In Vitro
"Complex II activity was significantly decreased at all DGA concentrations tested, with 12.5 mM DGA causing 80% inhibition and 25 and 50 mM DGA causing 97 and 100% inhibition, respectively."
The dose-response for the inhibition, with the concentrations stated so a reader can judge the physiological distance.
🐁

Animal Models

3
Diglycolic acid gavage in the rat
The metabolite given directly, bypassing diethylene glycol entirely. This is the in vivo counterpart of the cell-culture result and is what makes diglycolic acid sufficient rather than merely correlated.
Species
Rat
Genotype
Wild-type Wistar
Publication
PMID:28074668
Pathograph links
Proximal Tubular ATP Depletion and Necrosis recapitulates fidelity: moderate observes: organism Reproduces the renal injury from the metabolite alone. Retargeted from the transporter node during review: what this study measures is tissue accumulation and organ injury, not transporter-mediated uptake.
Limitations: The rat requires a far larger diethylene glycol dose than a human to reach the same lesion, so the dose-response is not transferable, and this design administers the metabolite rather than testing whether ingestion produces it.
Show evidence (1 reference)
PMID:28074668 SUPPORT PRIMARY RESULT Model Organism
"DGA concentrations in the kidneys and liver correlated with renal and hepatic injury, respectively."
Ties tissue accumulation of the metabolite to the organ injury, which is the claim this link makes.
High-dose diethylene glycol gavage in the rat
A dose-response design across two strains that establishes a steep threshold: below it, neither metabolite accumulation nor toxicity occurs.
Species
Rat
Genotype
Wild-type Wistar and Fischer-344
Publication
PMID:25545985
Pathograph links
Diglycolic Acid Formation recapitulates fidelity: moderate observes: organism
Limitations: The threshold sits far above the human toxic dose, so the model establishes the necessity of metabolite accumulation without transferring the dose at which it happens.
Show evidence (1 reference)
PMID:25545985 SUPPORT PRIMARY RESULT Model Organism
"Kidney diglycolic acid (DGA), the presumed nephrotoxic metabolite of DEG, was markedly elevated in both rat strains administered 10g/kg DEG, but no DGA was present at 2 or 5g/kg, asserting its necessary role in DEG-induced toxicity."
Metabolite accumulation and toxicity appear together and are absent together, which is the necessity claim.
Subacute diethylene glycol dosing with motor testing in the rat
The model behind the renal-dependence hypothesis. Only a minority of dosed animals developed kidney injury, and brain metabolite accumulation and motor deficits tracked that minority rather than the dose.
Species
Rat
Genotype
Wild-type Wistar-Han
Publication
PMID:33475432
Pathograph links
Neuronal Mitochondrial Complex II Inhibition partially recapitulates fidelity: low observes: organism Produces measurable motor impairment and brain metabolite accumulation, but not the cranial neuropathy that defines the human delayed syndrome.
Limitations: Grip strength and locomotor activity are not facial palsy or bulbar palsy, and the paper says so itself - its authors record that they had no animal test for the facial drooping seen in human cases. The model shows that nerve is affected without showing the syndrome.
Show evidence (1 reference)
PMID:33475432 SUPPORT PRIMARY RESULT Model Organism
"Renal and brain DGA accumulation was markedly increased in animals that developed AKI compared to animals without AKI."
The observation the renal-dependence hypothesis rests on: brain exposure to the metabolite segregates with kidney injury.
{ }

Source YAML

click to show 
name: Diethylene Glycol Poisoning
creation_date: "2026-09-19T14:00:00Z"
category: Environmental
categories:
- Toxic Exposure Disorder
- Pharmaceutical Contamination
- Toxic Alcohol Poisoning
synonyms:
- DEG poisoning
- diethylene glycol toxicity
- diethylene glycol intoxication
- diglycol poisoning
description: >-
  Diethylene glycol poisoning is an acute toxic nephropathy with delayed
  neurological sequelae, caused by ingestion of diethylene glycol - a sweet,
  viscous industrial solvent and humectant. Almost every large outbreak has
  been a pharmaceutical mass poisoning in which diethylene glycol was
  substituted, as a cheaper adulterant, for pharmaceutical grade glycerin or
  propylene glycol in a liquid medicine. Diethylene glycol itself is not the
  toxicant. It is oxidised by alcohol and aldehyde dehydrogenase to
  2-hydroxyethoxyacetic acid and then to diglycolic acid, and it is diglycolic
  acid - a close structural mimic of the citric acid cycle dicarboxylate
  succinate - that is carried into renal proximal tubular cells by the
  transporters that normally reclaim succinate, is poorly effluxed once
  inside, and inhibits mitochondrial complex II. The resulting ATP depletion
  produces proximal tubular necrosis rather than apoptosis, and the swollen
  necrotic epithelium obstructs the tubular lumen, giving the anuric acute
  kidney injury that defines the syndrome. Survivors of the renal phase may
  develop a distinct delayed neurological syndrome of cranial neuropathies -
  bilateral facial palsy is the classic finding - bulbar palsy, and a
  sensorimotor polyneuropathy. The cited sources do not agree on how
  delayed: one case series places the cranial neuropathies in the second
  week after ingestion, while a rat study's review of the human literature
  reports onset at two to seven days.
notes: >-
  **This is not ethylene glycol poisoning with a longer name, and the
  difference is mechanistic.** `kb/disorders/Ethylene_Glycol_Poisoning.yaml`
  models a chain that runs through glycolic acid to oxalic acid and calcium
  oxalate crystal deposition. Diethylene glycol does not run through that
  chain: in the rat model that established the metabolic route the urinary
  metabolites were 2-hydroxyethoxyacetic acid and diglycolic acid, with
  small amounts of ethylene glycol but no detectable oxalate or glycolate,
  and the authors concluded that toxicity does not proceed through
  conversion to ethylene glycol (`PMID:20530232`). The practical consequence is diagnostic: a clinician who looks for the
  calcium oxalate crystalluria that anchors ethylene glycol poisoning will
  not reliably find it. An earlier draft of this note added that the osmolal
  and anion gap behave differently, which was unsourced and half wrong -
  both poisonings produce a high anion gap acidosis, as this entry's own
  `Organic Acid Accumulation` node says.

  **That said, this entry does not claim diethylene glycol produces no
  oxalate at all, because one cited source contradicts it.** A fatal human
  case reports autopsy kidneys with "scattered tubular deposition of
  polarizable calcium oxalate crystals" (`PMID:28085681`). The
  `oxalate_in_human_deg_cases` discussion records the tension rather than
  resolving it in favour of the tidier statement.

  **No MONDO term exists for this disease.** `disease_term` is therefore
  omitted rather than bound to a neighbour. The searches run were
  `runoak -i ols:mondo search "diethylene glycol"`, which returns nothing,
  and `"glycol poisoning"`, which returns only `MONDO:0017861` ethylene
  glycol poisoning and `MONDO:1013081` ethylene glycol poisoning, non-human
  animal. Binding either would assert this entry is about the other toxin -
  precisely the confusion the first note exists to prevent. Thirty-four
  disorder entries already carry no `disease_term`, including
  `Aconitine_Poisoning` and `Arsenic_Poisoning`.

  **Mortality figures are reported with their outbreaks attached, not
  pooled.** The 1937 Elixir Sulfanilamide disaster killed 105
  (`PMID:7856995`); the 2006 Panama outbreak killed 12 of the first 21
  identified patients (`PMID:18949211`); the 2008 Nigerian teething syrup
  outbreak reports 96.6% mortality among exposed children against 71.2% in
  unexposed children with acute kidney injury (`PMID:22246572`); and a 2010
  review puts the cumulative paediatric toll at more than 300
  (`PMID:21062849`). These count different populations over different
  periods with different access to dialysis, and averaging them would
  produce a number that describes nothing.

  **Most phenotypes here are not wired into the pathograph.** The rule this
  entry follows is that an edge is drawn only where a cited source
  attributes the feature to the mechanism, not wherever a mechanism and a
  feature are both plausible. `just list-disconnected-phenotypes` reports
  the current state; a count written here would be wrong the moment a
  phenotype is added.
mechanistic_hypotheses:
- hypothesis_group_id: dga_proximal_tubular_necrosis
  hypothesis_label: Diglycolic acid is the proximate renal toxicant
  status: CANONICAL
  description: >-
    The renal lesion is produced by diglycolic acid rather than by diethylene
    glycol or by 2-hydroxyethoxyacetic acid. This is the best evidenced part
    of the mechanism: in human proximal tubule cells the parent compound and
    the first metabolite produce no cell death at all, and in rats the
    toxicity appears only at doses at which diglycolic acid accumulates in
    kidney tissue.
- hypothesis_group_id: dga_direct_neurotoxicity
  hypothesis_label: Diglycolic acid is directly neurotoxic
  status: EMERGING
  description: >-
    The delayed neuropathy reflects the same complex II inhibition acting on
    neurons, demonstrated in a neuroblastoma line. The evidence is in vitro
    and uses millimolar concentrations, and no human nerve tissue measurement
    supports it yet.
- hypothesis_group_id: heaa_neurotoxicity
  hypothesis_label: 2-hydroxyethoxyacetic acid is the neurotoxic moiety
  status: ALTERNATIVE
  description: >-
    A review of delayed neurological sequelae across the toxic alcohols
    attributes the cranial neuropathies of diethylene glycol to
    2-hydroxyethoxyacetic acid rather than to diglycolic acid. The source
    states this as what is "thought to be" the case rather than as a
    demonstrated result, and the entry carries it with that hedge intact.
- hypothesis_group_id: aki_dependent_metabolite_retention
  hypothesis_label: The neuropathy is conditional on renal failure
  status: ALTERNATIVE
  description: >-
    Delayed neurotoxicity appears only in those who sustain renal injury, in
    both a human case series and two rat studies, where brain diglycolic acid
    was markedly higher in animals that developed acute kidney injury than
    in those that did not. On this account
    the nerve injury is not an independent arm but a consequence of losing
    the clearance route for the metabolite. It is not exclusive of direct
    neurotoxicity - it supplies the exposure that direct toxicity would then
    act on.
pathophysiology:
- name: Systemic Diethylene Glycol Exposure
  biological_scale: ORGANISM
  description: >-
    Ingested diethylene glycol is absorbed rapidly and distributes to the
    kidney and brain ahead of other tissues, delivering the parent compound
    to the liver where it is bioactivated.
  chemical_entities:
  - preferred_term: diethylene glycol
    term:
      id: CHEBI:46807
      label: diethylene glycol
    modifier: INCREASED
  evidence:
  - reference: PMID:8135655
    reference_title: Toxicokinetics of diethylene glycol (DEG) in the rat.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "14C-DEG was rapidly distributed from the blood into the organs and tissues in the order kidneys > brain > spleen > liver > muscle > fat, i.e. the same order as the blood flow."
    explanation: >-
      Sources the distribution claim in this node's description, including
      the ordering and the reason for it. The two organs heading the list are
      the two this disease damages.
  downstream:
  - target: Alcohol Dehydrogenase Oxidation of Diethylene Glycol
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:20530232
      reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "After low and high doses of DEG, 2-hydroxyethoxyacetic acid (HEAA) was the primary metabolite in the urine, with only minor amounts of urinary diglycolic acid (DGA)."
      explanation: >-
        Establishes that the ingested compound is metabolised rather than
        excreted unchanged, and names the two acids this edge produces.
- name: Alcohol Dehydrogenase Oxidation of Diethylene Glycol
  biological_scale: MOLECULAR
  description: >-
    Alcohol dehydrogenase oxidises diethylene glycol to
    2-hydroxyethoxyacetaldehyde. This is the single step the antidote blocks,
    and it is separated from the aldehyde dehydrogenase step below precisely
    so that the antidote's target can be named exactly.
  molecular_functions:
  - preferred_term: alcohol dehydrogenase (NAD+) activity
    term:
      id: GO:0004022
      label: alcohol dehydrogenase (NAD+) activity
  evidence:
  - reference: PMID:20530232
    reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "These results indicate that the mechanism for the target organ toxicity results from metabolites of DEG, and not DEG itself nor formation of EG from DEG, and that fomepizole may be a useful antidote for treating DEG poisoning."
    explanation: >-
      The causal keystone of this entry. It establishes that bioactivation is
      required, and rules out the alternative that diethylene glycol is
      toxic by conversion to ethylene glycol.
  downstream:
  - target: 2-Hydroxyethoxyacetic Acid Formation
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:20530232
      reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Treatment with fomepizole blocked the formation of HEAA and DGA and the development of metabolic acidosis and the kidney and liver toxicity."
      explanation: >-
        Supports this edge specifically: inhibiting the alcohol dehydrogenase
        step prevents both acid metabolites downstream of it from being
        formed at all.
- name: 2-Hydroxyethoxyacetic Acid Formation
  biological_scale: MOLECULAR
  description: >-
    Aldehyde dehydrogenase oxidises the aldehyde to 2-hydroxyethoxyacetic
    acid, the quantitatively dominant urinary metabolite. It is not the renal
    toxicant - it produces no death in human proximal tubule cells - but it
    is an organic acid and so contributes to the acidosis, and one review
    proposes it rather than diglycolic acid as the neurotoxic moiety.
  molecular_functions:
  - preferred_term: aldehyde dehydrogenase (NAD+) activity
    term:
      id: GO:0004029
      label: aldehyde dehydrogenase (NAD+) activity
  notes: >-
    No CHEBI term is bound. Searching the OLS `chebi` index for
    "2-hydroxyethoxyacetic acid" and for "hydroxyethoxyacetic acid" returns
    nothing, so there is no identifier to bind rather than a better one that
    was missed.
  evidence:
  - reference: PMID:20530232
    reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "After low and high doses of DEG, 2-hydroxyethoxyacetic acid (HEAA) was the primary metabolite in the urine, with only minor amounts of urinary diglycolic acid (DGA)."
    explanation: >-
      Establishes this metabolite as the dominant one by quantity, which is
      the asymmetry that makes the toxicity of the minor metabolite
      surprising and worth modelling separately.
  downstream:
  - target: Diglycolic Acid Formation
    causal_link_type: DIRECT
    hypothesis_groups:
    - dga_proximal_tubular_necrosis
  - target: Organic Acid Accumulation
    causal_link_type: DIRECT
  - target: Neuronal Mitochondrial Complex II Inhibition
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - heaa_neurotoxicity
    description: >-
      The alternative account of the delayed neuropathy, drawn so the
      competing hypothesis has an edge to attach to rather than existing only
      as prose. A review names this metabolite as the causal moiety in the
      cranial neuropathies; no experimental result supports it.
- name: Diglycolic Acid Formation
  biological_scale: MOLECULAR
  description: >-
    Diglycolic acid is a four-carbon dicarboxylic acid with an ether oxygen,
    close enough in shape to succinate that the kidney handles it as though
    it were a citric acid cycle intermediate. That mistaken identity is the
    reason the lesion is renal and proximal.
  chemical_entities:
  - preferred_term: diglycolic acid
    term:
      id: CHEBI:169983
      label: Diglycolic acid
    modifier: INCREASED
  evidence:
  - reference: PMID:21856646
    reference_title: Diglycolic acid is the nephrotoxic metabolite in diethylene glycol poisoning inducing necrosis in human proximal tubule cells in vitro.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "DGA is structurally similar to citric acid cycle intermediates that are taken up by specific transporters in kidney cells."
    explanation: >-
      States the structural mimicry this node rests on, and does so in the
      paper that identified diglycolic acid as the toxic metabolite.
  downstream:
  - target: Dicarboxylate Transporter-Mediated Renal Uptake
    causal_link_type: DIRECT
    hypothesis_groups:
    - dga_proximal_tubular_necrosis
    evidence:
    - reference: PMID:28074668
      reference_title: In-vivo evidence of nephrotoxicity and altered hepatic function in rats following administration of diglycolic acid, a metabolite of diethylene glycol.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "DGA accumulated significantly in both kidney and liver tissue only at 300 mg DGA/kg."
      explanation: >-
        Diglycolic acid given directly to rats accumulates in kidney tissue,
        which is the step this edge asserts.
  - target: Neuronal Mitochondrial Complex II Inhibition
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - dga_direct_neurotoxicity
    - aki_dependent_metabolite_retention
    description: >-
      The neuronal arm is drawn with unknown intermediates on purpose, and
      carries two hypothesis groups because two accounts share this edge and
      are not exclusive. What is demonstrated is that diglycolic acid
      inhibits complex II in a neuronal cell line, and that brain diglycolic
      acid is markedly higher in rats that develop kidney injury than in
      rats that do not; how it reaches human cranial nerve is not
      established.
    evidence:
    - reference: PMID:33475432
      reference_title: Neurotoxic effects of nephrotoxic compound diethylene glycol.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Renal and brain DGA accumulation was markedly increased in animals that developed AKI compared to animals without AKI."
      explanation: >-
        Supports the renal-dependence half of this edge: brain exposure to
        the metabolite segregates with kidney injury rather than with dose.
    - reference: PMID:34278906
      reference_title: Diethylene glycol produces nephrotoxic and neurotoxic effects in female rats.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Of 12 animals treated with DEG, 3 developed AKI as confirmed by increased BUN and creatinine concentrations."
      explanation: >-
        The second of the two rat studies this entry's hypothesis prose
        refers to, and the source of the striking observation that
        identically dosed animals diverge - only a quarter developed kidney
        injury, which is what the transporter-expression finding later
        explains.
- name: Dicarboxylate Transporter-Mediated Renal Uptake
  biological_scale: CELLULAR
  description: >-
    Sodium-dicarboxylate cotransporters on the proximal tubular cell, whose
    normal job is reclaiming filtered succinate, carry diglycolic acid into
    the cell.
  biological_processes:
  - preferred_term: dicarboxylic acid transport
    term:
      id: GO:0006835
      label: dicarboxylic acid transport
  cell_types:
  - preferred_term: renal proximal tubular epithelial cell
    term:
      id: CL:0002306
      label: epithelial cell of proximal tubule
  locations:
  - preferred_term: proximal tubule
    term:
      id: UBERON:0004134
      label: proximal tubule
  genes:
  - preferred_term: SLC13A2
    term:
      id: hgnc:10917
      label: SLC13A2
  evidence:
  - reference: PMID:36939119
    reference_title: Variable sensitivity to diethylene glycol poisoning is related to differences in the uptake transporter for the toxic metabolite diglycolic acid.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "Significantly higher expression of sodium-dependent dicarboxylate transporter-1 mRNA was present in the kidneys of rats with acute kidney injury and diglycolic acid accumulation compared to those in rats that had no diglycolic acid in their kidneys and no acute kidney injury."
    explanation: >-
      Makes this node's transporter the locus of individual susceptibility,
      not merely the route of entry: identically dosed rats diverged, and the
      ones that were injured were the ones expressing more transporter. It is
      the best current explanation for why a given dose kills some people and
      not others.
  - reference: PMID:36087010
    reference_title: Role of Plasma Membrane Dicarboxylate Transporters in the Uptake and Toxicity of Diglycolic Acid, a Metabolite of Diethylene Glycol, in Human Proximal Tubule Cells.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "Pretreatment with the NaDC-1 inhibitor N-(p-amylcinnamoyl)anthranilic acid (ACA) markedly reduced apical uptakes of both succinate and DGA."
    explanation: >-
      Inhibiting the succinate transporter reduces diglycolic acid uptake,
      which is what makes this a transporter-mediated step rather than
      passive entry.
  downstream:
  - target: Intracellular Diglycolic Acid Trapping
    causal_link_type: DIRECT
    hypothesis_groups:
    - dga_proximal_tubular_necrosis
- name: Intracellular Diglycolic Acid Trapping
  biological_scale: CELLULAR
  description: >-
    Once inside, diglycolic acid is not effluxed. The cell has a route in and
    effectively no route out, so concentration builds - in rats kidney
    concentrations reach far above blood levels - which is why a metabolite
    circulating at modest concentration produces a lesion confined to this
    one cell type.
  evidence:
  - reference: PMID:36958672
    reference_title: Lack of efflux of diglycolic acid from proximal tubule cells leads to its accumulation and to toxicity of diethylene glycol.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "This study concluded that DGA is poorly transported out of cells and that stimulation of OAT transporters is not a viable target for reducing DGA accumulation in cells."
    explanation: >-
      Establishes the one-way character of the step, and reports the negative
      result that the obvious efflux route cannot be recruited to reverse it.
  downstream:
  - target: Renal Mitochondrial Complex II Inhibition
    causal_link_type: DIRECT
    hypothesis_groups:
    - dga_proximal_tubular_necrosis
- name: Renal Mitochondrial Complex II Inhibition
  biological_scale: MOLECULAR
  description: >-
    Accumulated diglycolic acid inhibits succinate dehydrogenase, the complex
    II of the respiratory chain, and depolarises the mitochondrion. The
    selectivity is striking - complexes I, III and IV are untouched - and is
    what one would predict from a succinate mimic.
  molecular_functions:
  - preferred_term: succinate dehydrogenase activity
    term:
      id: GO:0000104
      label: succinate dehydrogenase activity
    modifier: DECREASED
  biological_processes:
  - preferred_term: mitochondrial depolarization
    term:
      id: GO:0051882
      label: mitochondrial depolarization
    modifier: INCREASED
  evidence:
  - reference: PMID:27002734
    reference_title: Diglycolic acid, the toxic metabolite of diethylene glycol, chelates calcium and produces renal mitochondrial dysfunction in vitro.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "DGA inhibited Complex II activity without altering Complex I, III or IV activities."
    explanation: >-
      The specificity claim this node makes, stated with the negative
      controls that give it force.
  - reference: PMID:27002734
    reference_title: Diglycolic acid, the toxic metabolite of diethylene glycol, chelates calcium and produces renal mitochondrial dysfunction in vitro.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "DGA decreased the mitochondrial membrane potential in HPT cells."
    explanation: >-
      The depolarisation half of the node, measured in human proximal tubule
      cells rather than isolated mitochondria.
  downstream:
  - target: Proximal Tubular ATP Depletion and Necrosis
    causal_link_type: DIRECT
    hypothesis_groups:
    - dga_proximal_tubular_necrosis
- name: Proximal Tubular ATP Depletion and Necrosis
  biological_scale: CELLULAR
  description: >-
    Loss of oxidative ATP production in a cell type that depends on it kills
    by necrosis rather than apoptosis - the caspase machinery is not engaged,
    and the cells simply run out of energy.
  biological_processes:
  - preferred_term: ATP biosynthetic process
    term:
      id: GO:0006754
      label: ATP biosynthetic process
    modifier: DECREASED
  cell_types:
  - preferred_term: renal proximal tubular epithelial cell
    term:
      id: CL:0002306
      label: epithelial cell of proximal tubule
  evidence:
  - reference: PMID:21856646
    reference_title: Diglycolic acid is the nephrotoxic metabolite in diethylene glycol poisoning inducing necrosis in human proximal tubule cells in vitro.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "DGA induced necrosis, not apoptosis, as corroborated by severe depletion of cellular adenosine triphosphate levels."
    explanation: >-
      Both halves of this node in one sentence: the mode of death and the
      energetic failure behind it.
  - reference: PMID:21856646
    reference_title: Diglycolic acid is the nephrotoxic metabolite in diethylene glycol poisoning inducing necrosis in human proximal tubule cells in vitro.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "DEG and 2-HEAA did not produce any cell death. DGA produced dose-dependent necrosis at concentrations above 25 mmol/l."
    explanation: >-
      The negative control that makes the whole canonical arm work - neither
      the parent compound nor the first metabolite kills these cells.
  downstream:
  - target: Tubular Luminal Obstruction
    causal_link_type: DIRECT
    hypothesis_groups:
    - dga_proximal_tubular_necrosis
- name: Tubular Luminal Obstruction
  biological_scale: TISSUE
  description: >-
    The swollen and vacuolated necrotic epithelium occludes the tubular
    lumen. Urine flow falls for a mechanical reason on top of the cellular
    one, which is why the renal failure of this poisoning is characteristically
    anuric rather than merely oliguric.
  locations:
  - preferred_term: kidney
    term:
      id: UBERON:0002113
      label: kidney
  evidence:
  - reference: PMID:37186212
    reference_title: "Renal toxicity caused by diethylene glycol: an overview."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Marked vacuolization and edema of epithelial cells obstruct the lumen, reducing urine flow and, consequently, resulting in anuria and uremia."
    explanation: >-
      States the obstruction step and its clinical consequence directly.
      Graded OTHER and REVIEW_SYNTHESIS because the source is a narrative
      overview summarising renal biopsy morphology rather than reporting it.
  downstream:
  - target: Acute kidney injury
    causal_link_type: DIRECT
  - target: Anuria
    causal_link_type: DIRECT
- name: Organic Acid Accumulation
  biological_scale: ORGANISM
  description: >-
    2-hydroxyethoxyacetic acid and diglycolic acid are unmeasured anions, and
    their accumulation produces the high anion gap metabolic acidosis of the
    early phase. This arm is separable from the renal arm: the antidote
    prevents both, and the acidosis appears before renal failure is
    established.
  evidence:
  - reference: PMID:20530232
    reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "Rats treated with high-dose DEG had metabolic acidosis, increased BUN and creatinine, and marked kidney necrosis, noted by histopathology."
    explanation: >-
      Reports the acidosis alongside the renal lesion in the same dosed
      animals, which is what this node and its position in the chain assert.
  downstream:
  - target: Metabolic acidosis
    causal_link_type: DIRECT
- name: Neuronal Mitochondrial Complex II Inhibition
  biological_scale: MOLECULAR
  description: >-
    The same complex II inhibition demonstrated in renal cells occurs in a
    neuronal cell line, with mitochondrial depolarisation, ATP depletion and
    inflammatory cytokine release. This is the proposed substrate of the
    delayed neurological syndrome, and it is an in vitro result rather than
    an observation in nerve.
  molecular_functions:
  - preferred_term: succinate dehydrogenase activity
    term:
      id: GO:0000104
      label: succinate dehydrogenase activity
    modifier: DECREASED
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  evidence:
  - reference: PMID:36754214
    reference_title: Diglycolic acid inhibits succinate dehydrogenase activity, depletes mitochondrial membrane potential, and induces inflammation in an SH-SY5Y neuroblastoma model of neurotoxicity in vitro.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: PRIMARY_RESULT
    snippet: "These studies demonstrate that the DGA-induced mechanism of SH-SY5Y cell death involves complex II inhibition leading to mitochondrial depolarization, and subsequent ATP depletion with accompanying inflammatory cytokine release."
    explanation: >-
      The whole node in one sentence, and explicitly in a neuronal line
      rather than in the kidney cells where the mechanism was first shown.
  downstream:
  - target: Facial palsy
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - dga_direct_neurotoxicity
  - target: Sensorimotor neuropathy
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - dga_direct_neurotoxicity
phenotypes:
- category: Gastrointestinal
  name: Gastrointestinal prodrome
  description: >-
    The first phase, within hours of ingestion: nausea, vomiting and abdominal
    pain, usually with an anion gap metabolic acidosis. Non-specific on its
    own, which is much of why the syndrome is recognised late.
  phenotype_term:
    preferred_term: nausea, vomiting and abdominal pain
    term:
      id: HP:0002017
      label: Nausea and vomiting
    temporality: ACUTE
  notes: >-
    Modelled as one phenotype rather than four because the cited sentence
    names the phase collectively ("gastrointestinal symptoms") and does not
    itemise nausea, vomiting, abdominal pain and diarrhoea separately.
    Splitting it into four would assert four findings where the source
    reports one. `HP:0002017` Nausea and vomiting is bound as the closest
    single term; `preferred_term` carries the fuller picture.
  evidence:
  - reference: PMID:37186212
    reference_title: "Renal toxicity caused by diethylene glycol: an overview."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Patients may present with gastrointestinal symptoms and anion gap metabolic acidosis, followed by renal failure, and, later, encephalopathy and neuropathy."
    explanation: >-
      Names the gastrointestinal presentation as the first of the three
      phases. The same sentence is cited on Metabolic acidosis for the other
      half of that phase.
- category: Renal
  name: Acute kidney injury
  description: >-
    The defining feature and the usual cause of death. Histologically an acute
    tubular necrosis centred on the proximal tubule.
  phenotype_term:
    preferred_term: Acute kidney injury
    term:
      id: HP:0001919
      label: Acute kidney injury
    temporality: ACUTE
  evidence:
  - reference: PMID:37186212
    reference_title: "Renal toxicity caused by diethylene glycol: an overview."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "The kidney injury in DEG poisoning is secondary to proximal tubular necrosis caused by DGA."
    explanation: >-
      Names the phenotype and attributes it to the mechanism this entry
      models.
- category: Renal
  name: Anuria
  description: >-
    Characteristically anuric rather than oliguric, which is what the luminal
    obstruction predicts.
  phenotype_term:
    preferred_term: Anuria
    term:
      id: HP:0100519
      label: Anuria
    temporality: ACUTE
  evidence:
  - reference: PMID:15902788
    reference_title: Delayed neurologic sequelae resulting from epidemic diethylene glycol poisoning.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Initially three patients, ingesting the largest quantities of DEG, presented more than 24 h postingestion with severe metabolic acidemia (pH range 6.8-7.1) and anuric acute renal failure requiring HD."
    explanation: >-
      Anuric renal failure observed directly, and tied to the largest
      ingested quantities.
- category: Metabolic
  name: Metabolic acidosis
  description: >-
    A high anion gap acidosis from the accumulating organic acid metabolites,
    appearing in the first phase before renal failure is established.
  phenotype_term:
    preferred_term: Metabolic acidosis
    term:
      id: HP:0001942
      label: Metabolic acidosis
    temporality: ACUTE
  evidence:
  - reference: PMID:37186212
    reference_title: "Renal toxicity caused by diethylene glycol: an overview."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Patients may present with gastrointestinal symptoms and anion gap metabolic acidosis, followed by renal failure, and, later, encephalopathy and neuropathy."
    explanation: >-
      Places the acidosis in the first of the three phases, which is why this
      phenotype hangs off the organic acid node rather than off renal failure.
- category: Neurologic
  name: Facial palsy
  description: >-
    Bilateral facial palsy is the classic sign of the delayed phase. The
    facial nerve's particular vulnerability is unexplained.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: bilateral facial palsy
    term:
      id: HP:0430025
      label: Bilateral facial palsy
  notes: >-
    Frequency from the Panama series as summarised by a later case series
    ("more than half"), which is why the frequency evidence below is graded
    REVIEW_SYNTHESIS.

    An earlier draft cited `PMID:21192754` here. That paper names cranial
    nerve deficits for diethylene glycol without saying which, and the one
    sentence in it that does name the seventh nerve attributes it to
    **ethylene** glycol - the neighbouring toxin this entry's top `notes`
    exists to keep separate. It was the wrong citation for this phenotype
    and was replaced rather than left to look adequate.
  evidence:
  - reference: PMID:33475432
    reference_title: Neurotoxic effects of nephrotoxic compound diethylene glycol.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: BACKGROUND
    snippet: "we were not able to observe bilateral facial paresis or “facial drooping” in rats, a specific neurological effect seen in many DEG human cases."
    explanation: >-
      Names bilateral facial paresis as a specific effect of this poisoning
      in humans. HUMAN_CLINICAL with quote_role BACKGROUND because the citing
      paper is a rat study and this sentence reports the human picture it
      could not reproduce - grading it MODEL_ORGANISM would assert that rats
      showed the sign, which is the opposite of what the sentence says.
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "More than half of the patients had lower motor neuron facial palsy."
    explanation: >-
      The frequency band. "More than half" places it in FREQUENT without
      supporting a point estimate, so none is given.
- category: Neurologic
  name: Limb muscle weakness
  description: >-
    The commonest neurological finding, and in the Panama series almost all
    of those affected had reduced or absent deep tendon reflexes alongside
    it.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: limb weakness
    term:
      id: HP:0003690
      label: Limb muscle weakness
  notes: >-
    Bound to `HP:0003690` Limb muscle weakness, not `HP:0003324` Generalized
    muscle weakness. Those two are siblings under `HP:0001324` Muscle
    weakness, not a general and a specific form of one thing: HPO defines
    the first as weakness "of the muscles of the arms and legs" and the
    second as "affecting both distal and proximal musculature". The source
    says limb weakness, so the distribution is the arms and legs. An earlier
    draft of this phenotype bound `HP:0003324`, which asserted a different
    distribution from its own snippet.

    Frequency is from the Panama 2006 series as reported by a later case
    series, not by that series itself, so the evidence is graded
    REVIEW_SYNTHESIS - the citing paper is summarising Sosa et al rather than
    counting its own patients.
  evidence:
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "The most common neurological finding was limb weakness in 31 (78%) of whom almost all had reduced or absent deep tendon reflexes."
    explanation: >-
      Both the rank and the figure behind the FREQUENT band, and the
      areflexia that accompanies it.
- category: Neurologic
  name: Autonomic dysfunction
  description: >-
    Signs suggestive of autonomic involvement were reported in most of the
    Panama cohort. The source records signs rather than a confirmed
    diagnosis, and this phenotype keeps that hedge.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: signs suggestive of autonomic dysfunction
    term:
      id: HP:0012332
      label: Abnormal autonomic nervous system physiology
  notes: >-
    Bound to `HP:0012332`, the functional child, rather than its parent
    `HP:0002270` Abnormality of the autonomic nervous system, because the
    claim is about dysfunction rather than about structure. The deep
    research report offered `HP:0012647` for autonomic dysfunction; that
    identifier is Abnormal inflammatory response and was not used.

    Frequency is from the Panama 2006 series as reported by a later case
    series, not by that series itself, hence REVIEW_SYNTHESIS.
  evidence:
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Sixty three percent developed signs suggestive of autonomic dysfunction."
    explanation: >-
      The figure behind the FREQUENT band. Quoted in the source's own
      wording, which spells the number and says "signs suggestive of" rather
      than asserting the diagnosis.
- category: Neurologic
  name: Bulbar palsy
  phenotype_term:
    preferred_term: Bulbar palsy
    term:
      id: HP:0001283
      label: Bulbar palsy
  evidence:
  - reference: PMID:15902788
    reference_title: Delayed neurologic sequelae resulting from epidemic diethylene glycol poisoning.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "All three remained dialysis-dependent and developed significant cranial neuropathies with bulbar palsy in the second week postingestion."
    explanation: >-
      Bulbar palsy observed directly, with the second-week timing that makes
      this the delayed rather than the acute syndrome.
- category: Neurologic
  name: Sensorimotor neuropathy
  description: >-
    Electrophysiologically an axonal polyneuropathy in the acute phase, with
    demyelinating features appearing during recovery.
  phenotype_term:
    preferred_term: sensorimotor polyneuropathy
    term:
      id: HP:0007141
      label: Sensorimotor neuropathy
  notes: >-
    No `frequency` is recorded. The strongest available figure - severe
    sensorimotor axonal neuropathy in 19 of 21 - is conditional on having
    limb weakness and on having been tested, so it is a frequency within a
    selected subgroup rather than among poisoned patients. Entering it as a
    disease-level band would overstate it, so the numerator and denominator
    are carried in the evidence instead.
  evidence:
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Nerve conduction on 21 patients with limb weakness showed evidence of severe sensorimotor axonal neuropathy in 19."
    explanation: >-
      Electrophysiological confirmation in a cohort, with its denominator
      stated - the 21 are those with limb weakness who were tested, not the
      whole series.
  - reference: PMID:15824363
    reference_title: Encephalopathy and peripheral neuropathy following diethylene glycol ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "These data indicate that DEG intoxication may cause a primary acute axonal sensorimotor polyneuropathy with demyelinating physiology during recovery."
    explanation: >-
      Characterises the neuropathy electrophysiologically, and is the source
      of this phenotype's description distinguishing the acute from the
      recovery picture.
- category: Neurologic
  name: Encephalopathy
  phenotype_term:
    preferred_term: Encephalopathy
    term:
      id: HP:0001298
      label: Encephalopathy
  evidence:
  - reference: PMID:15824363
    reference_title: Encephalopathy and peripheral neuropathy following diethylene glycol ingestion.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The authors report a 24-year-old man who developed encephalopathy and rapid quadriplegia following ingestion of a solution containing diethylene glycol (DEG)."
    explanation: Encephalopathy observed in a documented single ingestion.
- category: Neurologic
  name: Cerebral edema
  phenotype_term:
    preferred_term: Cerebral edema
    term:
      id: HP:0002181
      label: Cerebral edema
  evidence:
  - reference: PMID:15902788
    reference_title: Delayed neurologic sequelae resulting from epidemic diethylene glycol poisoning.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "One patient died with cerebral oedema and a progressive encephalopathy."
    explanation: >-
      Cerebral oedema as a fatal outcome of the delayed phase in this series.
- category: Neurologic
  name: Tetraparesis
  description: >-
    The severe end of the motor picture - progressive flaccid paralysis
    culminating in quadriparesis - and markedly less common than the limb
    weakness it develops out of.
  frequency: OCCASIONAL
  phenotype_term:
    preferred_term: quadriparesis
    term:
      id: HP:0002273
      label: Tetraparesis
  notes: >-
    Frequency is from the Panama 2006 series as reported by a later case
    series, not by that series itself, so the evidence is graded
    REVIEW_SYNTHESIS - the citing paper is summarising Sosa et al rather than
    counting its own patients.
  evidence:
  - reference: PMID:21192754
    reference_title: Delayed neurological sequelae from ethylene glycol, diethylene glycol and methanol poisonings.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Diethylene glycol ingestion has been associated with the development of optic nerve injury, cranial nerve deficits, quadraparesis and peripheral neuropathy."
    explanation: Names quadraparesis among the delayed sequelae.
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Ten patients (22%) developed progressive flaccid paralysis culminating in quadriparesis."
    explanation: >-
      The figure behind the OCCASIONAL band, and the reason this phenotype is
      distinguished from the commoner limb weakness rather than merged with
      it.
- category: Ophthalmologic
  name: Optic neuropathy
  phenotype_term:
    preferred_term: optic nerve injury
    term:
      id: HP:0001138
      label: Optic neuropathy
  evidence:
  - reference: PMID:21192754
    reference_title: Delayed neurological sequelae from ethylene glycol, diethylene glycol and methanol poisonings.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Diethylene glycol ingestion has been associated with the development of optic nerve injury, cranial nerve deficits, quadraparesis and peripheral neuropathy."
    explanation: Names optic nerve injury among the delayed sequelae.
- category: Cardiovascular
  name: Hypertension
  phenotype_term:
    preferred_term: Hypertension
    term:
      id: HP:0000822
      label: Hypertension
  evidence:
  - reference: PMID:22246572
    reference_title: Diagnosing renal failure due to diethylene glycol in children in a resource-constrained setting.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Compared to children who had not ingested it, they were significantly (p < 0.05) younger (11.95 vs. 31 months), more were anuric (98.3 vs. 74.6%), hypertensive (84 vs. 52%), had severe metabolic acidosis (46.7 vs. 20.5%), and died (96.6 vs. 71.2%)."
    explanation: >-
      Hypertension in 84% of exposed children against 52% of unexposed
      children with acute kidney injury, so the figure is a contrast rather
      than a bare prevalence.
- category: Hepatic
  name: Hepatic failure
  description: >-
    A secondary target. Both the rat studies that establish the renal lesion
    report liver injury alongside it, and the paediatric epidemics record
    liver failure among the presenting features.
  phenotype_term:
    preferred_term: liver failure
    term:
      id: HP:0001399
      label: Hepatic failure
  evidence:
  - reference: PMID:28074668
    reference_title: "In-vivo evidence of nephrotoxicity and altered hepatic function in rats following administration of diglycolic acid, a metabolite of diethylene glycol."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "DGA concentrations in the kidneys and liver correlated with renal and hepatic injury, respectively."
    explanation: >-
      The same metabolite accumulation drives injury in both organs, which is
      why the hepatic involvement is carried here rather than treated as
      incidental.
  - reference: PMID:21062849
    reference_title: Epidemics of acute renal failure in children (diethylene glycol toxicity).
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "These children have presented with gastrointestinal bleeding, seizures and liver failure as well as renal failure."
    explanation: Names liver failure among the presenting features in children.
- category: Neurologic
  name: Seizure
  phenotype_term:
    preferred_term: Seizure
    term:
      id: HP:0001250
      label: Seizure
  evidence:
  - reference: PMID:21062849
    reference_title: Epidemics of acute renal failure in children (diethylene glycol toxicity).
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "These children have presented with gastrointestinal bleeding, seizures and liver failure as well as renal failure."
    explanation: Names seizures among the presenting features in children.
- category: Gastrointestinal
  name: Gastrointestinal hemorrhage
  phenotype_term:
    preferred_term: gastrointestinal bleeding
    term:
      id: HP:0002239
      label: Gastrointestinal hemorrhage
  evidence:
  - reference: PMID:21062849
    reference_title: Epidemics of acute renal failure in children (diethylene glycol toxicity).
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "These children have presented with gastrointestinal bleeding, seizures and liver failure as well as renal failure."
    explanation: >-
      Names gastrointestinal bleeding among the presenting features across
      the paediatric epidemics.
environmental:
- name: Ingestion of a liquid medicine adulterated with diethylene glycol
  description: >-
    The dominant exposure route and the one behind every large outbreak.
    Diethylene glycol is substituted for pharmaceutical grade glycerin or
    propylene glycol in a syrup or elixir, usually as a cheaper adulterant
    introduced upstream in the excipient supply chain rather than by the
    finished-product manufacturer.
  exposure_term:
    preferred_term: exposure to diethylene glycol
    term:
      id: ECTO:9002156
      label: exposure to glycol
  influences_mechanisms:
  - target: Systemic Diethylene Glycol Exposure
    environmental_effect: TRIGGERS
    causal_link_type: DIRECT
    description: >-
      The contaminated medicine is the entire source of the systemic burden;
      there is no endogenous diethylene glycol.
    evidence:
    - reference: PMID:18949211
      reference_title: "Outbreak of acute renal failure in Panama in 2006: a case-control study."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Laboratory analyses confirmed the presence of diethylene glycol (DEG) in biological samples from case patients, 8% DEG contamination in cough syrup samples and 22% contamination in the glycerin used to prepare the cough syrup."
      explanation: >-
        Traces the exposure from patient tissue back through the finished
        syrup to the contaminated glycerin, which is exactly the route this
        edge asserts.
  evidence:
  - reference: PMID:18949211
    reference_title: "Outbreak of acute renal failure in Panama in 2006: a case-control study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "After controlling for pre-existing hypertension and renal disease and the use of angiotensin-converting enzyme inhibitors, a significant association was found between ingestion of prescription cough syrup and illness onset (adjusted odds ratio: 31.0, 95% confidence interval: 6.93-138)."
    explanation: >-
      The epidemiological link between the medicine and the illness, adjusted
      for the obvious renal confounders.
  notes: >-
    ECTO has no exposure class for diethylene glycol, so the genus term is
    bound and `preferred_term` carries the specificity. The searches run
    through the OLS `ecto` index were "diethylene glycol", "exposure to
    diethylene glycol", "diethyleneglycol", "diethylene", "oxydiethanol" and
    "2-(2-hydroxyethoxy)ethanol". The only ECTO hit among them is
    `ECTO:9001432` exposure to diethylenetriamine, a different chemical;
    `CHEBI:46807` diethylene glycol exists but ECTO has not generated an
    exposure class from it.

    Two near-misses were rejected on positive grounds rather than
    overlooked. `ECTO:9001111` exposure to ethylene glycol is exact for the
    wrong toxin - binding it would assert the confusion this entry's top
    `notes` exists to prevent. `ECTO:9002161` exposure to glycol ether was rejected on narrower grounds
    than an earlier draft claimed. That draft said the term names the
    monoalkyl ethers of ethylene glycol, whose target organs are testis and
    bone marrow; that is industrial-hygiene usage, not the ontology's
    definition, and ChEBI defines a glycol ether structurally as a
    hydroxyether - which diethylene glycol is. The term is therefore
    arguably correct and simply less apt than the bound one: `exposure to
    glycol` is the reading a curator would expect for this compound, and
    `glycol ether` invites the industrial-hygiene reading and its wrong
    target organs.
treatments:
- name: Fomepizole
  description: >-
    The antidote, and unusually for a toxidrome its rationale is fully worked
    out: it inhibits alcohol dehydrogenase, so the parent compound is never
    converted to the acids that do the damage. In the rat model it prevents
    the metabolic acidosis and the renal and hepatic injury outright. Its
    value is therefore greatest before metabolism has occurred, and it cannot
    undo a lesion that diglycolic acid has already produced.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: fomepizole
      term:
        id: CHEBI:5141
        label: fomepizole
  target_mechanisms:
  - target: Alcohol Dehydrogenase Oxidation of Diethylene Glycol
    treatment_effect: INHIBITS
    description: >-
      Blocks the first oxidation step specifically, which is why the
      bioactivation chain is modelled as two nodes rather than one - the
      antidote acts on the alcohol dehydrogenase step alone, and everything
      downstream of it is prevented as a consequence rather than as a second
      target.
    evidence:
    - reference: PMID:20530232
      reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Treatment with fomepizole blocked the formation of HEAA and DGA and the development of metabolic acidosis and the kidney and liver toxicity."
      explanation: >-
        Supports this edge precisely: the intervention is shown to act on
        metabolite formation, and the downstream toxicity is prevented.
  evidence:
  - reference: PMID:10793034
    reference_title: Childhood diethylene glycol poisoning treated with alcohol dehydrogenase inhibitor fomepizole and hemodialysis.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We report a case of DEG ingestion in a 17-month-old girl who was managed with activated charcoal, fomepizole (a recently available alcohol dehydrogenase inhibitor), and hemodialysis (HD)."
    explanation: >-
      The only human evidence in this entry for the antidote, and it is a
      single case in which three interventions were given together, so it is
      INDIRECT: it establishes that fomepizole has been used in this
      poisoning, not that it was what worked.
- name: Ethanol
  description: >-
    The alternative alcohol dehydrogenase inhibitor, and the one that matters
    most in practice: fomepizole is expensive and often unavailable in the
    settings where these outbreaks happen. It occupies the same enzyme as
    fomepizole and so blocks the same step, but the evidence for it in this
    poisoning is weaker - the source cited here says inhibition "should be"
    therapeutic rather than reporting that it was.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: ethanol
      term:
        id: CHEBI:16236
        label: ethanol
  target_mechanisms:
  - target: Alcohol Dehydrogenase Oxidation of Diethylene Glycol
    treatment_effect: INHIBITS
    description: >-
      Competes for the same enzyme fomepizole inhibits, and so is drawn to
      the same node.
    evidence:
    - reference: PMID:36939119
      reference_title: Variable sensitivity to diethylene glycol poisoning is related to differences in the uptake transporter for the toxic metabolite diglycolic acid.
      supports: SUPPORT
      directness: INDIRECT
      evidence_source: OTHER
      quote_role: BACKGROUND
      snippet: "It is also likely that inhibition of alcohol dehydrogenase, either with ethanol or fomepizole, should be therapeutic."
      explanation: >-
        INDIRECT and BACKGROUND, and deliberately so: this is the authors'
        framing sentence proposing that the mechanism implies a therapy, not
        a result. It is the strongest statement about ethanol in this
        entry's cached sources, and the treatment description says as much
        rather than implying a demonstrated benefit.
- name: Supportive correction of the metabolic acidosis
  description: >-
    Bicarbonate against the acidosis, alongside general supportive care and
    gastric decontamination. This is what most patients in the outbreaks
    actually received, since the antidote and reliable dialysis were absent.
  therapeutic_modality: OTHER
  treatment_term:
    preferred_term: Supportive Care
    term:
      id: NCIT:C15747
      label: Supportive Care
    therapeutic_agent:
    - preferred_term: sodium bicarbonate
      term:
        id: CHEBI:32139
        label: sodium hydrogencarbonate
  evidence:
  - reference: PMID:36939119
    reference_title: Variable sensitivity to diethylene glycol poisoning is related to differences in the uptake transporter for the toxic metabolite diglycolic acid.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: BACKGROUND
    snippet: "Current therapy centers on supportive care such as bicarbonate to combat the acidosis and the use of extracorporeal measures to remove DEG and its metabolites (such removal is implied, although it has not really been shown experimentally) and to fix acid-base abnormalities"
    explanation: >-
      Describes current practice, and is carried with the authors' own
      parenthesis intact - they note that extracorporeal removal of the
      toxin is assumed rather than demonstrated, which qualifies the
      Hemodialysis entry as well as this one.
  - reference: PMID:37186212
    reference_title: "Renal toxicity caused by diethylene glycol: an overview."
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "The management includes supportive care, gastric decontamination, correction of acid-base disorders, and hemodialysis."
    explanation: >-
      The management list this treatment entry covers the first three
      elements of.
- name: Hemodialysis
  description: >-
    Clears the parent compound - measured across a dialysis session in a treated
    child - and becomes renal replacement once tubular necrosis is
    established. A human case series reports acidaemia and renal impairment
    responding to emergent haemodialysis. Its availability matters: the
    paediatric epidemics in resource-constrained settings record
    deterioration despite peritoneal dialysis, and their authors call for
    haemodialysis facilities to be provided. An earlier draft asserted that
    haemodialysis availability was the main determinant of survival in the
    outbreaks and that peritoneal dialysis clears these solutes far less
    efficiently; neither is stated by a cited source and both were removed.
  therapeutic_modality: OTHER
  treatment_term:
    preferred_term: Hemodialysis
    term:
      id: NCIT:C15248
      label: Hemodialysis
  target_mechanisms:
  - target: Systemic Diethylene Glycol Exposure
    treatment_effect: INHIBITS
    description: >-
      Removes the parent compound from the circulation, lowering the
      systemic burden that everything downstream depends on. This is the
      only treatment here that acts on the exposure node itself rather than
      on a metabolic step.
    evidence:
    - reference: PMID:10793034
      reference_title: Childhood diethylene glycol poisoning treated with alcohol dehydrogenase inhibitor fomepizole and hemodialysis.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Pre-HD and post-HD DEG levels support clearance of DEG with HD."
      explanation: >-
        Measured clearance of the parent compound across a dialysis session,
        which is exactly what this edge asserts. Note the qualification
        carried on the supportive-care entry: one cited source describes
        extracorporeal removal as implied rather than experimentally shown,
        and this case measures levels in a single patient.
  evidence:
  - reference: PMID:15902788
    reference_title: Delayed neurologic sequelae resulting from epidemic diethylene glycol poisoning.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "We report a series of seven patients presenting with epidemic DEG poisoning from a correctional facility with varying degrees of metabolic acidemia and acute renal impairment responding to emergent hemodialysis (HD)."
    explanation: >-
      The affirmative human evidence: acidaemia and renal impairment
      responding to emergent haemodialysis.
  - reference: PMID:10793034
    reference_title: Childhood diethylene glycol poisoning treated with alcohol dehydrogenase inhibitor fomepizole and hemodialysis.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Pre-HD and post-HD DEG levels support clearance of DEG with HD."
    explanation: >-
      Measured clearance of the parent compound across a dialysis session,
      which is the specific claim that haemodialysis removes the toxin rather
      than only replacing renal function.
  - reference: PMID:22246572
    reference_title: Diagnosing renal failure due to diethylene glycol in children in a resource-constrained setting.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "They developed increasing metabolic acidosis and multiorgan dysfunction despite peritoneal dialysis."
    explanation: >-
      INDIRECT, and kept for a different point than the two items above: the
      modality available in that outbreak did not arrest the syndrome, which
      is why haemodialysis rather than dialysis in general is named here.
  notes: >-
    `NCIT:C15248` Hemodialysis is used rather than a broader procedure term.
    The deep research report offered `NCIT:C15571` for hemodialysis; that
    identifier is Differentiation Therapy, and was not used. The query run
    was a search of the OLS `ncit` index for "Hemodialysis", which returns
    `NCIT:C15248` as the exact match.
diagnosis:
- name: Serum diethylene glycol and metabolite quantification
  description: >-
    The definitive test, and rarely available where the outbreaks happen.
    Chromatographic measurement of the parent compound and its acid
    metabolites is what confirmed the exposure route in the Panama
    investigation, working from patient samples back to the contaminated
    excipient.
  diagnosis_term:
    preferred_term: chromatographic quantification of serum diethylene glycol and its metabolites
    term:
      id: NCIT:C18020
      label: Diagnostic Procedure
  notes: >-
    Bound to the generic procedure term because the specific one is out of
    range. `NCIT:C111203` Gas Chromatography Mass Spectrometry names the
    technique exactly but is not reachable from `NCIT:C25218` Clinical
    Intervention or Procedure, the root of the `TreatmentActionTerm` enum
    that `diagnosis_term` uses, so `just validate-terms` rejects it - the
    same class of failure as binding a device term in a clinical-action
    slot. `preferred_term` carries the technique and the analytes.
  evidence:
  - reference: PMID:18949211
    reference_title: "Outbreak of acute renal failure in Panama in 2006: a case-control study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Laboratory analyses confirmed the presence of diethylene glycol (DEG) in biological samples from case patients, 8% DEG contamination in cough syrup samples and 22% contamination in the glycerin used to prepare the cough syrup."
    explanation: >-
      Laboratory confirmation in patient samples and in the product, which is
      what this test establishes.
- name: Clinical recognition without toxicological confirmation
  description: >-
    The diagnosis that is actually made in most of the world. Where assay is
    unavailable, the combination of a drug history, a worsening metabolic
    acidosis and multiorgan deterioration that does not respond to dialysis
    is what raises suspicion - and the authors who set out this approach are
    explicit that identifying the syndrome this way is difficult.
  evidence:
  - reference: PMID:22246572
    reference_title: Diagnosing renal failure due to diethylene glycol in children in a resource-constrained setting.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Detailed drug history, increasing metabolic acidosis, and multiorgan deterioration despite peritoneal dialysis should arouse suspicion."
    explanation: >-
      The clinical triad this diagnosis rests on, from a study designed
      around the absence of toxicology facilities.
  - reference: PMID:33475432
    reference_title: Neurotoxic effects of nephrotoxic compound diethylene glycol.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: OTHER
    quote_role: BACKGROUND
    snippet: "Blood work shows increased BUN and creatinine levels and osmolal gap along with a metabolic acidosis, but such results could allude to a spectrum of illnesses"
    explanation: >-
      Why routine bloods cannot make this diagnosis: the abnormalities are
      real but non-specific. INDIRECT and BACKGROUND because the citing
      paper is a rat study and this is its framing of the human diagnostic
      problem.
  - reference: PMID:10793034
    reference_title: Childhood diethylene glycol poisoning treated with alcohol dehydrogenase inhibitor fomepizole and hemodialysis.
    supports: SUPPORT
    directness: INDIRECT
    evidence_source: HUMAN_CLINICAL
    quote_role: BACKGROUND
    snippet: "Patients presenting less than 12 hours after DEG ingestion may not show metabolic acidosis, whereas those presenting later may show florid metabolic acidosis."
    explanation: >-
      INDIRECT and a caution rather than a criterion: the acidosis that
      anchors clinical recognition is time-dependent, so an early
      presentation can look reassuring. Graded BACKGROUND because the
      sentence states the established kinetic picture the case report is
      framed against.
prevalence:
- population: Worldwide paediatric mass poisonings reported up to 2010
  measure_type: CASES_IN_LITERATURE
  prevalence_class: ULTRA_RARE
  notes: >-
    A cumulative death toll across epidemics rather than a rate in a defined
    population at risk, which is the only form this disease's occurrence
    takes - there is no background incidence between outbreaks.
  evidence:
  - reference: PMID:21062849
    reference_title: Epidemics of acute renal failure in children (diethylene glycol toxicity).
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "More than 300 children worldwide have died from DEG poisoning."
    explanation: >-
      The cumulative paediatric figure as of this review, carried with its
      date attached rather than as a current total.
- population: The Gambia, 2022 outbreak
  measure_type: CASES_IN_LITERATURE
  prevalence_class: ULTRA_RARE
  notes: >-
    Included because the entry's other records stop in 2010 and this is not a
    historical disease: the 2022 wave, traced to syrup medications from a
    single manufacturer, reached The Gambia, Indonesia and Uzbekistan.
  evidence:
  - reference: PMID:36862590
    reference_title: "Acute Kidney Injury Among Children Likely Associated with Diethylene Glycol-Contaminated Medications - The Gambia, June-September 2022."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The preliminary investigation suggested that various contaminated syrup-based children's medications contributed to the AKI outbreak."
    explanation: >-
      The outbreak investigation's own finding, stated with the hedge its
      authors used - this was a preliminary attribution at publication.
- population: Panama, 2006 outbreak
  measure_type: CASES_IN_LITERATURE
  prevalence_class: ULTRA_RARE
  notes: >-
    Adults rather than children, with a median age of 68 - unusual among
    these outbreaks and a consequence of the vehicle being a prescription
    cough syrup rather than a paediatric preparation.
  evidence:
  - reference: PMID:18949211
    reference_title: "Outbreak of acute renal failure in Panama in 2006: a case-control study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Twelve (57%) of 21 patients had died of the illness."
    explanation: >-
      The case fatality among the initially identified patients that prompted
      the outbreak investigation.
animal_models:
- name: Diglycolic acid gavage in the rat
  species: Rat
  genotype: Wild-type Wistar
  publication: PMID:28074668
  description: >-
    The metabolite given directly, bypassing diethylene glycol entirely. This
    is the in vivo counterpart of the cell-culture result and is what makes
    diglycolic acid sufficient rather than merely correlated.
  modeled_mechanisms:
  - target: Proximal Tubular ATP Depletion and Necrosis
    relationship: RECAPITULATES
    fidelity: MODERATE
    model_scale: ORGANISM
    description: >-
      Reproduces the renal injury from the metabolite alone. Retargeted from
      the transporter node during review: what this study measures is tissue
      accumulation and organ injury, not transporter-mediated uptake.
    limitations: >-
      The rat requires a far larger diethylene glycol dose than a human to
      reach the same lesion, so the dose-response is not transferable, and
      this design administers the metabolite rather than testing whether
      ingestion produces it.
    evidence:
    - reference: PMID:28074668
      reference_title: In-vivo evidence of nephrotoxicity and altered hepatic function in rats following administration of diglycolic acid, a metabolite of diethylene glycol.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "DGA concentrations in the kidneys and liver correlated with renal and hepatic injury, respectively."
      explanation: >-
        Ties tissue accumulation of the metabolite to the organ injury, which
        is the claim this link makes.
- name: High-dose diethylene glycol gavage in the rat
  species: Rat
  genotype: Wild-type Wistar and Fischer-344
  publication: PMID:25545985
  description: >-
    A dose-response design across two strains that establishes a steep
    threshold: below it, neither metabolite accumulation nor toxicity occurs.
  modeled_mechanisms:
  - target: Diglycolic Acid Formation
    relationship: RECAPITULATES
    fidelity: MODERATE
    model_scale: ORGANISM
    limitations: >-
      The threshold sits far above the human toxic dose, so the model
      establishes the necessity of metabolite accumulation without
      transferring the dose at which it happens.
    evidence:
    - reference: PMID:25545985
      reference_title: Diethylene glycol-induced toxicities show marked threshold dose response in rats.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Kidney diglycolic acid (DGA), the presumed nephrotoxic metabolite of DEG, was markedly elevated in both rat strains administered 10g/kg DEG, but no DGA was present at 2 or 5g/kg, asserting its necessary role in DEG-induced toxicity."
      explanation: >-
        Metabolite accumulation and toxicity appear together and are absent
        together, which is the necessity claim.
- name: Subacute diethylene glycol dosing with motor testing in the rat
  species: Rat
  genotype: Wild-type Wistar-Han
  publication: PMID:33475432
  description: >-
    The model behind the renal-dependence hypothesis. Only a minority of
    dosed animals developed kidney injury, and brain metabolite accumulation
    and motor deficits tracked that minority rather than the dose.
  modeled_mechanisms:
  - target: Neuronal Mitochondrial Complex II Inhibition
    relationship: PARTIALLY_RECAPITULATES
    fidelity: LOW
    model_scale: ORGANISM
    description: >-
      Produces measurable motor impairment and brain metabolite accumulation,
      but not the cranial neuropathy that defines the human delayed syndrome.
    limitations: >-
      Grip strength and locomotor activity are not facial palsy or bulbar palsy,
      and the paper says so itself - its authors record that they had no
      animal test for the facial drooping seen in human cases. The model
      shows that nerve is affected without showing the syndrome.
    evidence:
    - reference: PMID:33475432
      reference_title: Neurotoxic effects of nephrotoxic compound diethylene glycol.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Renal and brain DGA accumulation was markedly increased in animals that developed AKI compared to animals without AKI."
      explanation: >-
        The observation the renal-dependence hypothesis rests on: brain
        exposure to the metabolite segregates with kidney injury.
experimental_models:
- name: Primary human proximal tubule cells
  experimental_model_type: PRIMARY_CELL_CULTURE
  publication: PMID:21856646
  description: >-
    Primary cultures from normal human renal cortex, exposed to the parent
    compound and each metabolite separately. The design is what identifies
    the toxicant, because it includes the negative arms.
  modeled_mechanisms:
  - target: Proximal Tubular ATP Depletion and Necrosis
    relationship: RECAPITULATES
    fidelity: MODERATE
    model_scale: CELLULAR
    description: >-
      Reproduces the cell death and its mode in the human target cell type.
    limitations: >-
      Cultured cells lack the filtration and concentration that deliver the
      metabolite to this cell in vivo, so the millimolar concentrations used
      are applied directly rather than reached physiologically.
    evidence:
    - reference: PMID:21856646
      reference_title: Diglycolic acid is the nephrotoxic metabolite in diethylene glycol poisoning inducing necrosis in human proximal tubule cells in vitro.
      supports: SUPPORT
      evidence_source: IN_VITRO
      quote_role: PRIMARY_RESULT
      snippet: "These studies demonstrate that DGA is the toxic metabolite responsible for DEG-induced proximal tubular necrosis and suggest a possible transporter-mediated uptake of DGA leading to toxic accumulation and cellular dysfunction."
      explanation: >-
        The paper's own statement of what the model established, in the human
        cell type the disease destroys.
- name: SH-SY5Y neuroblastoma line
  experimental_model_type: CELL_LINE
  publication: PMID:36754214
  description: >-
    A neuronal line used to ask whether the renal mechanism also operates in
    nerve. It is the only direct evidence for the emerging neurotoxicity
    hypothesis.
  modeled_mechanisms:
  - target: Neuronal Mitochondrial Complex II Inhibition
    relationship: RECAPITULATES
    fidelity: LOW
    model_scale: MOLECULAR
    description: >-
      Demonstrates complex II inhibition, depolarisation and ATP depletion in
      a neuronal context.
    limitations: >-
      A neuroblastoma line is not a cranial motor neuron, and the effective
      concentrations are millimolar. It shows the mechanism is available to
      neurons; it does not show it is what injures the facial nerve in
      poisoned people.
    divergences:
    - divergence_type: POPULATION_MISMATCH
      materiality: QUALIFYING
      description: >-
        SH-SY5Y is a neuroblastoma-derived line with an immature
        catecholaminergic phenotype. The human lesion is in cranial and
        peripheral motor and sensory nerve, which this line does not model.
    evidence:
    - reference: PMID:36754214
      reference_title: Diglycolic acid inhibits succinate dehydrogenase activity, depletes mitochondrial membrane potential, and induces inflammation in an SH-SY5Y neuroblastoma model of neurotoxicity in vitro.
      supports: SUPPORT
      evidence_source: IN_VITRO
      quote_role: PRIMARY_RESULT
      snippet: "Complex II activity was significantly decreased at all DGA concentrations tested, with 12.5 mM DGA causing 80% inhibition and 25 and 50 mM DGA causing 97 and 100% inhibition, respectively."
      explanation: >-
        The dose-response for the inhibition, with the concentrations stated
        so a reader can judge the physiological distance.
discussions:
- kind: KNOWLEDGE_GAP
  discussion_id: delayed_neuropathy_mechanism
  prompt: >-
    Is the delayed cranial and peripheral neuropathy of diethylene glycol
    poisoning caused by diglycolic acid acting directly on nerve, by
    2-hydroxyethoxyacetic acid, or by retention of metabolite that only
    happens once the kidney has failed?
  attaches_to:
  - pathophysiology#Neuronal Mitochondrial Complex II Inhibition
  - phenotypes#Facial palsy
  rationale: >-
    Three accounts are live and this entry carries all three as hypothesis
    groups. The direct-toxicity account has the only positive mechanistic
    data, and it is an in vitro neuroblastoma result at millimolar
    concentrations. The 2-hydroxyethoxyacetic acid account comes from a
    review that states it as what is thought to be the case. The
    renal-dependence account has the most human support - a case series in
    which only the patients with established renal impairment went on to
    develop cranial neuropathy - and is echoed in two rat studies where brain
    metabolite accumulated only in animals with kidney injury. The three are
    not mutually exclusive: renal failure could supply the exposure that
    direct toxicity then acts on. What no source settles is why the facial
    nerve in particular.
  evidence:
  - reference: PMID:15902788
    reference_title: Delayed neurologic sequelae resulting from epidemic diethylene glycol poisoning.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Established renal impairment may predict subsequent delayed neurologic toxicity."
    explanation: >-
      The human observation behind the renal-dependence account, stated by
      its authors with "may" rather than as a settled relation.
  - reference: PMID:21192754
    reference_title: Delayed neurological sequelae from ethylene glycol, diethylene glycol and methanol poisonings.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "2-hydroxyethoxyacetic acid is thought to be the causal moiety in cranial neuropathies resulting from diethylene glycol toxicity."
    explanation: >-
      The competing attribution, carried with its hedge. Note it names the
      other metabolite - not the one this entry's renal arm blames.
  - reference: PMID:21192754
    reference_title: Delayed neurological sequelae from ethylene glycol, diethylene glycol and methanol poisonings.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "Further research is required to explain why the facial nerve is the cranial nerve most commonly involved and why the basal ganglia are predisposed to injury."
    explanation: >-
      The gap stated by a source rather than inferred by this entry.
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The neurological course and toxicity is less well characterized than its renal counterpart."
    explanation: >-
      A second human series stating the asymmetry this discussion exists
      because of - the renal arm is settled and the neurological one is not.
- kind: KNOWLEDGE_GAP
  discussion_id: oxalate_in_human_deg_cases
  prompt: >-
    If diethylene glycol is not metabolised to oxalate, why do calcium
    oxalate crystals appear in the kidneys of some human cases?
  attaches_to:
  - pathophysiology#Proximal Tubular ATP Depletion and Necrosis
  rationale: >-
    The rat metabolic study found no urinary oxalate or glycolate and
    concluded that toxicity does not proceed through conversion to ethylene
    glycol. A human autopsy case nonetheless describes calcium oxalate
    crystals in both kidneys. Candidate explanations the cited sources do not distinguish between:
    ethylene glycol generated endogenously as a minor metabolite of
    diethylene glycol - the rat study detected small amounts of it in urine
    even while finding no oxalate - ethylene glycol present as a contaminant
    alongside diethylene glycol in the ingested product, a species
    difference in that minor route, or oxalate of dietary or endogenous
    origin deposited in already necrotic tubules. An earlier draft of this
    rationale offered only the product-impurity explanation, which misread
    the quoted sentence: the ethylene glycol it reports was measured in the
    urine of dosed rats, so it is a metabolite, not an impurity. The entry records the conflict because the tidy version of the
    claim - that this poisoning never produces oxalate - is contradicted by
    a source it cites, and because the diagnostic advice that follows from
    it matters at the bedside.
  evidence:
  - reference: PMID:20530232
    reference_title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "Small amounts of ethylene glycol (EG), but not oxalate or glycolate, were observed in the urine."
    explanation: >-
      The metabolic result: the oxalate route that drives ethylene glycol
      nephrotoxicity is not detectably engaged here.
  - reference: PMID:28085681
    reference_title: Basal Ganglion Hemorrhage as Delayed Complication of Diethylene Glycol Ingestion.
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Both kidneys illustrate tubular necrosis with scattered tubular deposition of polarizable calcium oxalate crystals."
    explanation: >-
      REFUTE against the general claim that this poisoning does not deposit
      oxalate. It does not refute the diglycolic acid mechanism, which the
      same kidneys illustrate as tubular necrosis - the object is the oxalate
      claim alone.
  - reference: PMID:25170933
    reference_title: "Neurological manifestation of recreational fatal and near-fatal diethylene glycol poisonings: case series and review of literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "However, the difference in metabolism and the absence of oxalate crystalluria or oxalate deposition in sections of tissue examined14,15 favor that toxicity may be through different mechanisms."
    explanation: >-
      A human series pointing the other way from the autopsy above - no
      oxalate crystalluria and none in the tissue examined - which is what
      makes this a genuine two-sided question rather than one anomalous
      report against a rat study. The inline "14,15" is the source's own
      citation markers, quoted as they appear.
references:
- reference: PMID:21856646
  title: "Diglycolic acid is the nephrotoxic metabolite in diethylene glycol poisoning inducing necrosis in human proximal tubule cells in vitro."
- reference: PMID:20530232
  title: Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
📚

References & Deep Research

References

2
PMID:21856646
Diglycolic acid is the nephrotoxic metabolite in diethylene glycol poisoning inducing necrosis in human proximal tubule cells in vitro.
No top-level findings curated for this source.
PMID:20530232
Inhibition of metabolism of diethylene glycol prevents target organ toxicity in rats.
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)

Review round 2: limb weakness rebinding, autonomic dysfunction, hemodialysis edge · 2026-09-19T15:43:47Z · View source

Review round 2 on PR #12297, answered in one push. Four findings, all taken. Two blocking, both correctly identified as my errors: 1. Limb muscle weakness was bound to HP:0003324 Generalized muscle weakness while its preferred_term, description, snippet and explanation all said limb. Those two terms are siblings under HP:0001324, not a general and a specific form of one concept: HPO defines HP:0003690 as weakness "of the muscles of the arms and legs" and HP:0003324 as "affecting both distal and proximal musculature". Rebound to HP:0003690 and renamed to match. Worth recording how this happened, because the gates cannot catch it. The round-1 reviewer supplied HP:0003324 as a suggestion. I verified it existed and that its label was canonical, which is what validate-terms checks, and did not check that it was the right term for the claim. That is the same self-consistent-but-wrong shape CLAUDE.md documents for gene bindings, applied to a phenotype, and it is the second time this session a CURIE reached a commit because the existence check was mistaken for a correctness check. A CURIE from a reviewer is a lead on exactly the same footing as a CURIE from a deep-research report. 2. Autonomic dysfunction was absent entirely, at 63% in the Panama cohort, with the sentence sitting in the same paragraph that supplied three frequencies added in round 1. Added, bound to HP:0012332 Abnormal autonomic nervous system physiology - the functional child rather than its structural parent HP:0002270, since the claim is about dysfunction. The parent/child relationship was checked against OLS rather than assumed. Frequency FREQUENT; the snippet keeps the source's "signs suggestive of" hedge and its spelled-out "Sixty three percent". The review's diagnosis of why it dropped out is right: the research report offered HP:0012647 for autonomic dysfunction, that identifier is Abnormal inflammatory response, and refusing it was correct. What was missing was rebinding rather than dropping the phenotype. Two optional, both taken: 3. Hemodialysis now carries target_mechanisms against Systemic Diethylene Glycol Exposure. It was the only treatment outside the pathograph, and its own cited snippet - measured pre- and post-dialysis levels - is a claim about removing the toxin. The edge description carries the qualification from the supportive-care entry, that one cited source describes extracorporeal removal as implied rather than experimentally shown. 4. Supportive correction of the metabolic acidosis now carries CHEBI:32139 sodium hydrogencarbonate as a therapeutic_agent, so the agent is queryable rather than named only in prose. The review also recorded that two of its round-1 asks were unfounded - the osmolal-gap figure is sourceable only from an ethylene glycol paper, and the 68% facial palsy figure appears in no cached source - and that the two refusals in round 1 (no frequency on sensorimotor neuropathy, GI prodrome as one phenotype) were correct. No change needed for either. Validation: just validate-disorders clean with 70/70 snippets verified (was 68/68); just validate-terms passed; check-entity-refs, check-causal-targets, check-duplicate-keys, check-enum-values, check-qualifier-terms all clean. Connectivity 5/17 - the ratio keeps falling because each round adds phenotypes the entry cannot causally explain, which is the honest outcome.

Review round 1: GI prodrome, phenotype frequencies, ethanol and supportive care · 2026-09-19T15:21:21Z · View source

Review round 1 on PR #12297, answered in one push. Two blocking findings, both content-completeness gaps against the deep research artifact, both closeable from references already cached: 1. The phase-1 gastrointestinal prodrome had no phenotype, although the entry's own three-phase framing rests on it and the supporting sentence (PMID:37186212) was already quoted in the file for the acidosis half of the same phase. Added as one phenotype rather than four: the source names the phase collectively as "gastrointestinal symptoms" and does not itemise nausea, vomiting, abdominal pain and diarrhoea, so splitting it would assert four findings where one is reported. Bound HP:0002017 Nausea and vomiting as the closest single term with the reasoning in notes. 2. No phenotype carried a frequency. Added FREQUENT to limb weakness (78%) and bilateral facial palsy ("more than half"), and OCCASIONAL to tetraparesis (22%), all from PMID:25170933. Every one of those figures is that paper reporting Sosa et al's Panama cohort rather than counting its own patients, so the evidence is graded REVIEW_SYNTHESIS and each phenotype's notes says so. Also added Generalized muscle weakness (HP:0003324) as a distinct, commoner phenotype than the tetraparesis it progresses to. Deliberately no frequency on Sensorimotor neuropathy. The available figure - severe sensorimotor axonal neuropathy in 19 of 21 - is conditional on having limb weakness and on having been tested, so it is a frequency within a selected subgroup. Entering ~90% as a disease-level band would overstate it; the numerator and denominator are carried in the evidence instead. Three optional findings, all taken: 3. Added Ethanol as the alternative alcohol dehydrogenase inhibitor, targeting the same node as fomepizole, and a Supportive correction of the metabolic acidosis entry. Both from PMID:36939119, already cited here. The ethanol evidence is graded INDIRECT and BACKGROUND because the sentence is the authors proposing that the mechanism implies a therapy ("should be therapeutic"), not a result; the treatment description says so rather than implying demonstrated benefit. The bicarbonate snippet is quoted with the authors' parenthesis intact, since they note that extracorporeal removal of the toxin is assumed rather than shown - which qualifies the Hemodialysis entry too. 4. Added a diagnostic caveat about routine bloods, but NOT the one the review proposed. The review's claim was that the osmolar gap is less sensitive for diethylene glycol than for methanol or ethylene glycol. The only cached source making an osmolal-gap-sensitivity claim is PMID:33425517, which is about ETHYLENE glycol - citing it here would be the cross-toxin error this entry exists to prevent. What a diethylene glycol source does say (PMID:33475432) is that the osmolal gap is increased but non-specific, and that is what was added. 5. Added the opposing human data point to the oxalate discussion: PMID:25170933 reports absence of oxalate crystalluria and of oxalate deposition in tissue examined, pointing the other way from the PMID:28085681 autopsy. This turns a one-sided refutation into the two-sided open question the discussion claims to be. The snippet includes the source's own inline citation markers ("examined14,15") because that is how the sentence appears in the cached text. Verified rather than accepted: the review's frequency figures were checked against the cache before use. 78%, 22% and "more than half" are exact; "63%" appears as "Sixty three percent" and "19 of 21" as "in 19", so the snippets use the source's wording. The review also cited facial palsy at 68%, which appears nowhere in the cached sources and was not used. Validation after the round: just validate-disorders clean with 68/68 snippets verified (was 58/58); just validate-terms passed; check-entity-refs, check-causal-targets, check-duplicate-keys, check-enum-values and check-qualifier-terms all clean; every reference_title still byte-identical to its cache frontmatter. Connectivity 5/16 - the ratio fell because this round added two phenotypes that nothing in the entry causally explains, which is the honest outcome rather than a regression.

Create: Diethylene Glycol Poisoning (no MONDO term) · 2026-09-19T14:21:58Z · View source

Create: Diethylene Glycol Poisoning, a toxin-exposure entry with no MONDO term. Deep research: `just research-disorder claude_code` (336 s, 20 web searches, 62 citations). The run produced no inline validation block, so reference and term validation were retro-fitted with `just validate-research-reference` and `just validate-research-terms`: 31/31 references resolved, 0 unresolved, 1 off topic (and only 31 of the 62 citations were checkable at all, the rest being web pages rather than PMIDs or DOIs). No CURIE was taken from the report. That was load-bearing: term validation plus my own checks found seven of its suggestions naming different concepts. HP:0025314, offered for "CSF protein elevation", is Choroidal nevus. NCIT:C1421, offered for fomepizole, is Oxygen Radical. NCIT:C15571, offered for hemodialysis, is Differentiation Therapy. CL:1001016, offered as proximal convoluted tubule epithelial cell, is loop of Henle ascending limb - in an entry whose whole mechanism is proximal-tubule-specific. HP:0100519 and HP:0100518 were offered as Oliguria and Anuria and are Anuria and Dysuria. HP:0012647, offered as Autonomic dysfunction, is Abnormal inflammatory response. Every binding in this entry came from a live OLS lookup performed while writing the line. Why this entry rather than a subtype of the existing Ethylene_Glycol_Poisoning entry: the mechanisms diverge. Ethylene glycol runs through glycolic to oxalic acid and calcium oxalate deposition; diethylene glycol is oxidised to diglycolic acid, a succinate mimic taken into proximal tubular cells by the dicarboxylate transporters, not effluxed, and inhibiting complex II specifically. The evidence has clean negative controls - the parent compound and the first metabolite produce no cell death at all - and fomepizole blocks metabolite formation and every downstream toxicity, which makes bioactivation necessary rather than incidental. Four mechanistic_hypotheses, all wired to edges: the canonical renal arm, and three competing accounts of the delayed neuropathy (direct diglycolic acid neurotoxicity, 2-hydroxyethoxyacetic acid as the causal moiety, and retention conditional on renal failure). No disease_term. MONDO has no diethylene glycol concept; the searches are recorded in notes, and binding MONDO:0017861 ethylene glycol poisoning would assert the confusion this entry exists to prevent. A red-team subagent reviewed against dismech-pr-review before any PR and returned 14 findings. All were acted on. Four were blocking: 1. Three of four hypothesis groups were declared but no causal edge opted into them, including the CANONICAL one. All four are now wired. 2. The bioactivation node bundled three sequential reactions, which made it impossible to express that fomepizole blocks only the first, and left 2-hydroxyethoxyacetic acid with no node - so the hypothesis naming it as the neurotoxic moiety had no edge to attach to. Split into an alcohol dehydrogenase node and a 2-HEAA node. Retargeting fomepizole to the split node was then missed and caught by check-causal-targets, which is the rename-dangling case that gate exists for. 3. The Facial palsy phenotype cited a paper that names cranial nerve deficits without saying which - and whose only seventh-nerve sentence attributes it to ETHYLENE glycol, the exact confusion this entry warns about. Replaced with a sentence naming bilateral facial paresis in human DEG cases, and rebound from HP:0010628 to the more specific HP:0430025. 4. Twenty-four cache files were committed uncited. Four high-value ones are now used: the only human fomepizole case (PMID:10793034), the toxicokinetics study sourcing the tissue distribution claim (PMID:8135655), the NaDC-1 expression study explaining variable susceptibility (PMID:36939119), and a human neurological case series (PMID:25170933). The rest are cited by the committed research report and so are not orphans. Unsourced prose the review caught and this round fixed: a tissue distribution claim now cited; an "osmolal and anion gap behave differently" sentence that was unsourced and half wrong, since both poisonings give a high anion gap; three hemodialysis claims including "the main determinant of survival"; and a "one to three weeks" delay figure contradicted by a cited source saying two to seven days. The oxalate discussion also mischaracterised its own quoted source, offering product impurity as an explanation for ethylene glycol that the quote shows was measured in rat urine and is therefore a metabolite. Also added on review: hepatic failure and seizure phenotypes, both from snippets already quoted elsewhere in the file; affirmative human evidence for hemodialysis, which previously had only a negative observation about peritoneal dialysis; an SLC13A2 gene descriptor on the transporter node; a diagnosis section, since the entry's own notes argue the distinction from ethylene glycol is diagnostic; and corrections to experimental_model_type, two biological_scale tags and an animal-model link that pointed at a node the study does not measure. One process failure worth recording: I wrote HP:0002239 from memory rather than from a lookup. It happened to be correct, which is luck and not process - the rule is that a CURIE is read from a source in the same step it is written, and a correct guess is not evidence the rule was followed. I also mistyped a snippet ("was given between" for "was found between") and caught it before validation. Validation: just validate-disorders clean with 56/56 snippets verified; just validate-terms passed; check-entity-refs, check-causal-targets, check-duplicate-keys, check-enum-values, check-qualifier-terms, check-snippet-grading, check-title-snippets, check-snippet-length, check-folded-hyphens, check-environmental-evidence, check-case-collisions, check-reference-titles and check-not4curation all clean. check-genereviews returns NO_CHAPTER, correct for a non-Mendelian acquired poisoning. Connectivity 5/14.

Claude Code ▸
Diethylene Glycol Poisoning: Comprehensive Research Report
claude-haiku-4-5-20251001, claude-sonnet-5 62 citations 2026-09-19T13:48:26.864089

Diethylene Glycol Poisoning: Comprehensive Research Report

1. Disease Information

Overview. Diethylene glycol (DEG) poisoning is an acute (and, in survivors, subacute-to-chronic) toxic syndrome caused by ingestion of diethylene glycol — an industrial solvent, humectant, and antifreeze/brake-fluid component chemically similar to, but distinct from, ethylene glycol. DEG is not itself intended for human consumption; poisoning occurs almost exclusively through unintentional exposure (most catastrophically, as an adulterant substituting for pharmaceutical-grade glycerin or propylene glycol in liquid medicines) or, less commonly, deliberate ingestion. The clinical syndrome evolves through a predictable sequence: an early gastrointestinal/inebriation phase, a renal failure phase (acute tubular necrosis), and — in survivors, roughly 1–3 weeks later — a delayed neurologic phase with cranial and peripheral neuropathies and encephalopathy (Saudi J Kidney Dis Transplant, 2016; emDocs ToxCard).

Chemical identity. - IUPAC name: 2,2′-oxydiethanol; CAS 111-46-6 - ChEBI: CHEBI:46807 (diethylene glycol); classified as a hydroxyether/diol (ZFIN CHEBI:46807) - PubChem CID: 8117; KEGG: C14689 - Formula: C4H10O3, MW 106.12 g/mol

Key identifiers for the disease entity: - No dedicated OMIM entry exists (DEG poisoning is an acquired toxic/environmental disease, not Mendelian). - Orphanet lists the closely related entity "Ethylene glycol poisoning" (ORPHA:31826) but has no distinct DEG entry (Orphanet); DEG poisoning is generally grouped under toxic-alcohol/glycol poisoning nosology alongside ethylene glycol. - ICD-10-CM: T52.8X1– (Toxic effect of other organic solvents, accidental) is used in practice for DEG, since DEG has no dedicated subcode distinct from the ethylene-glycol code T51.8 family; MeSH: "Ethylene Glycols" (DEG is indexed as a related glycol; no separate MeSH heading exists for DEG poisoning specifically). - MONDO ID: not established as a distinct MONDO term at the time of this report; DEG poisoning would most naturally be curated as a toxic/environmental poisoning entity analogous to the existing ethylene glycol poisoning framework.

Synonyms/alternative names: DEG; diglycol; 2,2′-dihydroxydiethyl ether; bis(2-hydroxyethyl) ether; digol; 3-oxa-1,5-pentanediol; "DEG poisoning"; in the epidemiologic literature also referred to under the umbrella terms "diethylene glycol mass poisoning," "toxic syrup poisoning," and (colloquially in the 2022–2025 outbreaks) "contaminated cough syrup deaths."

Data provenance. Nearly all clinical knowledge of human DEG poisoning derives from aggregated outbreak/disease-level resources — case series and outbreak investigations published by health ministries, WHO, and the US CDC (MMWR) — rather than individual longitudinal EHR cohorts, because poisoning events are epidemic, geographically clustered in low- and middle-income countries, and typically investigated retrospectively. The single largest evidence base comes from outbreak investigations in Haiti (1995–96), Bangladesh/Nigeria (1990), Panama (2006), and The Gambia/Indonesia/Uzbekistan (2022) and the 2025 India (Madhya Pradesh) cluster (CDC MMWR Haiti; MMWR Gambia 2023). Mechanistic detail is drawn from rodent studies and in vitro human proximal-tubule-cell experiments.

2. Etiology

Disease causal factor

DEG poisoning is exclusively an environmental/toxic exposure disease — there is no genetic or infectious cause. The causal agent is oral (occasionally intravenous, in adulterated injectable products) ingestion of diethylene glycol, either as the pure industrial chemical or, far more commonly at the population level, as an unlabeled contaminant of pharmaceutical excipients (glycerin, propylene glycol, sorbitol solution, or polyethylene glycol) or, in some cases, of unregulated recreational/counterfeit alcohol products (Clin Toxicol PMC4616334).

Risk factors

Environmental/exposure risk factors (dominant): - Consumption of DEG/ethylene-glycol–contaminated liquid pharmaceuticals — the mechanism behind every major epidemic (Haiti 1995–96, Bangladesh/Nigeria 1990, Panama 2006, India-manufactured syrups exported to The Gambia/Indonesia/Uzbekistan 2022, and the Coldrif/Sresan Pharmaceuticals cluster in Madhya Pradesh, India, 2025) (2025 India cough syrup crisis, Wikipedia; CNN, Oct 2025). - Occupational/industrial exposure to DEG-containing solvents, antifreeze, and brake fluid. - Ingestion of counterfeit/adulterated alcoholic beverages. - Recreational or self-harm ingestion of DEG-containing consumer products (correctional-facility case clusters have been reported: Clin Toxicol PMC4616334). - Age (young children): the overwhelming majority of deaths in every mass-poisoning event have been in children under 5, likely reflecting both higher exposure per kg body weight from syrup dosing and possibly greater physiologic vulnerability. - Concentration/dose of contaminant: the 2025 Coldrif product reportedly contained DEG at levels "hundreds of times above" the pharmacopeial limit of 0.1% (up to ~45% DEG reported) (CPSP). - Number of contaminated medicines ingested — in The Gambia outbreak, exposure to a single contaminated medicine (rather than ≥2) was associated with markedly higher AKI risk (aOR = 4.21 for ≥2 vs 1 medicine, direction as reported), a counterintuitive dose-related finding under investigation (PLOS Global Public Health, case-case-control analysis).

Genetic risk/susceptibility factors: No confirmed human genetic risk variant has been established. However, mechanistic and rodent data strongly implicate variability in expression/activity of: - Alcohol dehydrogenase (ADH) / aldehyde dehydrogenase (ALDH) — the enzymes that bioactivate DEG into its toxic metabolites; individuals or populations with higher ADH/ALDH activity would be predicted to generate more toxic metabolite, analogous to established ADH1B/ALDH2 polymorphism effects on ethanol/methanol/ethylene-glycol metabolism (inferred by analogy; not yet directly demonstrated for DEG in humans). - NaDC-1 (SLC13A2), the renal proximal-tubule sodium-dicarboxylate transporter responsible for diglycolic acid (DGA) uptake: a 2023 rat study found that only ~60% of rats developed AKI after identical DEG/DGA dosing, and that AKI occurrence correlated with 1.6–2-fold higher renal NaDC-1 mRNA expression, proposing that inter-individual variability in this transporter may explain the variable human sensitivity documented across historical outbreaks (PMC10263375).

Protective factors

  • Older age: each additional year of age reduced AKI risk in exposed children (adjusted OR = 0.58 per year) in the Gambia case-case-control analysis (PLOS GPH).
  • Multivitamin supplementation: associated with markedly reduced AKI risk (aOR = 0.24) in the same analysis — the leading environmental/nutritional protective factor identified to date, though the biological mechanism (general nutritional status vs. a specific micronutrient effect on ADH/ALDH activity or renal transporter expression) is not established.
  • Genetic/pharmacologic protective factor (mechanistic, not epidemiologic): inhibition of ADH — pharmacologically achieved with fomepizole or ethanol — prevents bioactivation of DEG to its toxic metabolites and is the basis of antidotal therapy; a rat study demonstrated that inhibiting DEG metabolism (via ADH blockade) prevents target-organ toxicity entirely, confirming that the parent compound itself is not the proximate toxicant (Toxicol Sci 117:25, PMID:20530232).
  • Low NaDC-1 transporter expression (inferred protective factor from the rat model above).

Gene–environment interactions

The clearest documented gene(enzyme)–environment interaction is pharmacologic: ADH/ALDH activity level determines how much toxic metabolite is generated from a given DEG dose, and NaDC-1 transporter expression level determines how much of the toxic metabolite (DGA) accumulates in target renal tubule cells once formed. Both interactions have been demonstrated in animal/in vitro systems and are the rationale for antidotal ADH blockade in humans; direct human genotype-outcome data (e.g., linking ADH1B or SLC13A2 variants to clinical severity) have not yet been published from the outbreak cohorts.

3. Phenotypes

DEG poisoning phenotypes cluster into three temporally distinct clinical phases (Saudi J Kidney Dis Transplant 2016; Int Urol Nephrol 2023, PMID:37186212):

Phase 1 (hours, gastrointestinal/inebriation) — onset within hours of ingestion

Phenotype Suggested HP term
Nausea HP:0002018
Vomiting HP:0002013
Abdominal pain HP:0002027
Diarrhea HP:0002014
Inebriation/CNS depression HP:0007270 (Cerebral palsy - not exact); more precisely "Lethargy" HP:0001254 or "Ataxia" HP:0001251 depending on presentation
Metabolic acidosis (high or normal anion gap) HP:0001942

Phase 2 (1–3 days post-exposure, renal/systemic) — hallmark: acute renal failure

Phenotype Suggested HP term Notes
Acute kidney injury / acute tubular necrosis HP:0001919 (Acute kidney injury) Biopsy hallmark: proximal tubular necrosis with epithelial vacuolization/edema
Oliguria progressing to anuria HP:0100519 (Oliguria) / HP:0100518 (Anuria) Obstructive vacuolization reduces urine flow
Refractory metabolic acidosis HP:0001942
Hyperkalemia HP:0002153
Hypertension HP:0000822
Cardiac dysrhythmia / tachycardia HP:0001662 / HP:0001649
Pulmonary edema HP:0100598
Acute pancreatitis HP:0001733
Hepatitis/elevated transaminases HP:0002240 (Hepatomegaly) / HP:0002910 (Elevated hepatic transaminase)
Fever HP:0001945 Especially in pediatric AKI outbreaks (The Gambia)

Phase 3 (5–20 days post-exposure, delayed neurologic) — occurs mainly in survivors of AKI

Phenotype Suggested HP term Frequency (from case series)
Encephalopathy HP:0001298 Present in severe delayed cases; can progress to cerebral edema and death
Facial nerve palsy/diplegia (classic finding) HP:0010628 (Facial palsy) 68% in one case series of neurologic survivors (PMC4616334)
Multiple cranial neuropathies / bulbar palsy HP:0007311 or HP:0002015 (Bulbar palsy) Described as "classic" in delayed syndrome
Limb weakness (flaccid paralysis) HP:0003324 (Generalized muscle weakness) / HP:0002078 (Motor delay - not exact) 77% limb weakness among neurologic cases
Sensorimotor peripheral neuropathy HP:0007141 Demonstrated in 90% of those tested by nerve conduction studies
Areflexia HP:0001284
CSF albuminocytologic dissociation (Guillain-Barré–like) — (laboratory finding, no direct HP phenotype term; closest: HP:0025314 CSF protein elevation)
Optic neuritis HP:0100653 Reported in a subset
Seizures HP:0001250 Reported in severe cases
Coma HP:0001259 In fatal/near-fatal cases
Autonomic dysfunction (delayed) HP:0012647 (Autonomic dysfunction) Case reports of delayed autonomic neuropathy

Age of onset: Phase 1 begins within hours; Phase 2 (renal) at 1–3 days; Phase 3 (neurologic) at 5–20 days ("delayed" progressive neurologic syndrome), with some series describing onset in the second week post-ingestion.

Severity/progression: Highly dose-dependent and variable between individuals (see NaDC-1 discussion above). In the 2022 Gambia outbreak, AKI carried ~80% case-fatality among affected children, while ~30% of documented DEG-exposed children did not develop AKI at all (MMWR; PMC9997663). Among survivors of renal failure who develop the delayed neurologic syndrome, motor signs may improve over 4–6 months but dialysis-dependence can persist (Clin Toxicol PMC4616334). A 2026 latent-class analysis of the Gambia outbreak formally stratified clinical phenotype severity classes (Pediatric Nephrology, in press).

Quality of life impact: Survivors with residual chronic kidney disease and/or peripheral neuropathy face long-term disability; specific EQ-5D/SF-36 data for DEG poisoning survivors were not identified in the literature — this is a recognized gap, most outbreak follow-up has focused on mortality and short-term renal recovery rather than standardized QoL instruments.

4. Genetic/Molecular Information

DEG poisoning is not a genetic disease — there are no causal genes, pathogenic variants, or chromosomal abnormalities to report in the classic Mendelian sense. The "genetic/molecular" content relevant here concerns the enzymes and transporters governing individual susceptibility:

  • ADH (alcohol dehydrogenase) — HGNC gene family ADH1A/ADH1B/ADH1C etc. — catalyzes the first bioactivation step (DEG → 2-hydroxyethoxyacetaldehyde). No DEG-specific human polymorphism data published; extrapolated from the general toxic-alcohol metabolism literature.
  • ALDH (aldehyde dehydrogenase), e.g., ALDH2 — catalyzes the second step (aldehyde → 2-hydroxyethoxyacetic acid, 2-HEAA). Same caveat as above.
  • SLC13A2 (NaDC-1) — sodium-dicarboxylate cotransporter on the renal proximal tubule apical membrane; mediates cellular uptake of diglycolic acid (DGA), the nephrotoxic metabolite, via molecular mimicry of the Krebs-cycle intermediate succinate. Rat data show renal NaDC-1 mRNA expression correlates with AKI susceptibility (PMC10263375; Toxicol Sci 190:1, PMID:36087010).
  • OAT1/OAT3 (SLC22A6/SLC22A8) — basolateral organic anion transporters. Contrary to initial hypotheses, a 2023 study found OAT1/OAT3 stimulation does not promote efflux of DGA from proximal tubule cells — DGA accumulates intracellularly regardless, ruling out OAT-targeted efflux as a therapeutic strategy (PMID:36958672).

Functional consequence: the pathway is a toxic gain of a novel metabolite-mediated cytotoxic function rather than a loss-of-function/gain-of-function genetic mechanism — DGA depletes cellular ATP and chelates calcium, producing mitochondrial dysfunction and necrosis (Clin Toxicol 54:6, PMID:27002734).

Epigenetics, somatic variants, chromosomal abnormalities: Not applicable/not reported — no epigenetic or chromosomal mechanism has been described for this acquired toxic disease.

5. Environmental Information

Environmental (chemical/toxin) factors — the primary etiologic category

  • Pharmaceutical adulteration is the dominant environmental cause at the population level: DEG or ethylene glycol substituted (deliberately, by fraud, or accidentally through mislabeled industrial-grade chemical supply chains) for pharmaceutical glycerin, propylene glycol, sorbitol solution, or PEG excipients in liquid medicines (cough syrups, acetaminophen elixirs) (CDC/PMC10081810; WHO news).
  • Industrial/occupational exposure: DEG is a widely used solvent in antifreeze, brake fluid, printing inks, plasticizers, and as a humectant.
  • Counterfeit alcohol: DEG has been used to adulterate bootleg spirits in several countries.
  • Consumer product ingestion (accidental pediatric exposure to household antifreeze-type products, and recreational ingestion in adults, including reported correctional-facility clusters).

Lifestyle factors

Not directly disease-modifying in the classic sense (e.g., diet/exercise), though nutritional status (multivitamin use) modified outcome risk in the Gambia cohort (see Section 2).

Infectious agents

Not applicable — DEG poisoning is a pure chemical toxicity, not an infectious disease. (Note: differential diagnosis in pediatric outbreaks initially considered infectious causes of AKI before the toxic etiology was confirmed.)

6. Mechanism / Pathophysiology

Ordered causal chain

  1. Ingestion of diethylene glycol (via contaminated pharmaceutical excipient, industrial solvent, or counterfeit alcohol) → rapid, near-complete gastrointestinal absorption (rat data: absorption rate constants 2.95–4.24 h⁻¹) and distribution to tissues in order kidney > brain > spleen > liver > muscle > fat, tracking blood flow (toxicokinetics study, PMID:8135655) → leads to systemic exposure of the liver and kidney to unmetabolized DEG.
  2. Hepatic oxidation by alcohol dehydrogenase (ADH) converts DEG to 2-hydroxyethoxyacetaldehyde → leads to generation of a reactive aldehyde intermediate (Toxicol Sci 117:25).
  3. Oxidation by aldehyde dehydrogenase (ALDH) converts the aldehyde to 2-hydroxyethoxyacetic acid (2-HEAA) → results in the first major circulating toxic metabolite, which is renally eliminated and itself contributes to neurotoxicity and, per some models, general systemic toxicity.
  4. Further/alternate oxidation generates diglycolic acid (DGA), structurally analogous to the Krebs-cycle dicarboxylate succinate → leads to DGA being a substrate for the sodium-dicarboxylate cotransporter NaDC-1 (SLC13A2) on the apical membrane of renal proximal tubule cells, driving preferential renal accumulation (PMC10263375; mechanistic uptake study PMID:36087010).
  5. Intracellular DGA is poorly effluxed (OAT1/OAT3 stimulation does not rescue efflux) → results in progressive intracellular DGA accumulation in proximal tubule cells (PMID:36958672).
  6. Accumulated DGA chelates calcium and produces mitochondrial dysfunction, with severe depletion of cellular ATP → leads to proximal tubule cell necrosis (this step is directly demonstrated in vitro in human proximal tubule cells: Toxicol Sci 124:35, PMID:21856646; Clin Toxicol 54:6, PMID:27002734).
  7. Tubular epithelial necrosis, vacuolization, and edema obstruct the tubular lumen → results in reduced urine flow, progressing to oliguria/anuria (acute tubular necrosis) → causes acute kidney injury, uremia, and the systemic Phase 2 clinical syndrome (metabolic acidosis, hyperkalemia, pulmonary edema).
  8. Concurrently, hepatocyte and pancreatic acinar cell exposure to circulating metabolites (mechanism less well defined than the renal pathway) → contributes to hepatitis (transaminase elevation) and pancreatitis observed in Phase 2.
  9. In patients who survive the acute renal insult, and particularly those with more severe renal impairment, a delayed process — hypothesized to involve either persistent low-level metabolite exposure, immune-mediated peripheral nerve injury (supported by CSF albuminocytologic dissociation resembling Guillain-Barré syndrome), or direct neurotoxicity of 2-HEAA/DGA on peripheral and cranial nerve myelin/axons — leads to, 5–20 days post-ingestion, a delayed progressive neurologic syndrome: encephalopathy plus multiple cranial neuropathies (classically bilateral facial palsy) and sensorimotor peripheral neuropathy (Neurology, PMID:15824363; Clin Toxicol, PMID:15902788).
  10. Established renal impairment predicts subsequent delayed neurotoxicity, suggesting the renal and neurologic injury pathways are linked (possibly via reduced clearance of neurotoxic metabolites in AKI, prolonging systemic exposure) (PMC4616334).
  11. Severe cases progress to cerebral edema and death, typically from a combination of refractory metabolic derangement, cardiovascular collapse, and/or progressive encephalopathy.

Key branch point: it is the metabolites (2-HEAA and DGA), not DEG itself, that drive target-organ toxicity — demonstrated by the fact that pharmacological ADH inhibition (fomepizole/ethanol) blocks metabolite formation and prevents target-organ toxicity entirely in rat models (PMID:20530232). This is the central mechanistic rationale for antidotal therapy in humans.

Molecular pathways / cellular processes

  • ADH/ALDH-mediated oxidative metabolism (xenobiotic biotransformation pathway).
  • Sodium-dicarboxylate cotransport (NaDC-1/SLC13A2) — physiologically a succinate/dicarboxylate reabsorption pathway hijacked by the structurally mimetic toxin DGA.
  • Mitochondrial dysfunction / ATP depletion — GO:0001836 (release of cytochrome c from mitochondria) and related apoptosis/necrosis pathways are plausible GO annotations; the literature specifically emphasizes necrosis (not classic apoptosis) as the dominant proximal tubule cell death mode (GO:0070265, necrotic cell death).
  • Calcium chelation by DGA, disrupting calcium-dependent mitochondrial and cellular signaling.

Cell types and tissue involvement (suggested CL/UBERON terms)

  • Renal proximal tubule epithelial cell — CL:1001016 (kidney proximal convoluted tubule epithelial cell) — primary target, site of NaDC-1-mediated DGA uptake and necrosis.
  • Hepatocyte — CL:0000182 — site of ADH/ALDH-mediated bioactivation and secondary hepatotoxicity.
  • Peripheral/cranial motor and sensory neurons — CL:0000540 (neuron), with axonal/myelin involvement — site of delayed neurotoxicity (facial nerve, other cranial nerves, peripheral nerves).
  • Pancreatic acinar cell — CL:0002064 — site of DEG-associated pancreatitis.
  • Anatomical sites (UBERON): kidney (UBERON:0002113), renal proximal tubule (UBERON:0004134 or similar), liver (UBERON:0002107), pancreas (UBERON:0001264), facial nerve (UBERON:0001654), peripheral nervous system (UBERON:0000010), brain/cerebral cortex for encephalopathy (UBERON:0000955).

Biochemical abnormalities

  • Increased anion-gap metabolic acidosis (accumulation of 2-HEAA and DGA as unmeasured anions).
  • Osmolar gap may be minimally elevated or normal — DEG's larger molecular weight compared to methanol/ethylene glycol makes the osmolar gap a less sensitive diagnostic marker than for other toxic alcohols (EMCrit IBCC; PMC7785502).
  • Hyperkalemia, elevated BUN/creatinine, elevated hepatic transaminases, lipase/amylase elevation in pancreatitis.

Chemical entities (CHEBI)

  • Diethylene glycol — CHEBI:46807 (parent compound)
  • 2-hydroxyethoxyacetaldehyde — intermediate metabolite (no widely indexed CHEBI ID identified in this search)
  • 2-hydroxyethoxyacetic acid (2-HEAA) — toxic metabolite
  • Diglycolic acid (DGA) — the principal nephrotoxic metabolite

7. Anatomical Structures Affected

Organ level: - Primary target: kidney — proximal tubules specifically, producing acute tubular necrosis and AKI. - Secondary: liver (hepatitis, transaminase elevation), pancreas (acute pancreatitis), peripheral and cranial nervous system (delayed neuropathy/palsies), central nervous system (encephalopathy, seizures, coma, cerebral edema), cardiovascular system (dysrhythmias, hypotension/hypertension), respiratory system (pulmonary edema). - Body systems: renal, hepatobiliary, gastrointestinal (pancreas), nervous (both central and peripheral), cardiovascular, respiratory.

Tissue/cell level: - Renal proximal tubule epithelium (specific target of DGA-mediated necrosis). - Hepatocytes. - Peripheral nerve axons/Schwann cells (myelin), cranial nerve motor fibers (especially facial nerve — bilateral facial diplegia is the classic finding). - Pancreatic acinar tissue.

Subcellular level: - Mitochondria (ATP depletion, calcium dysregulation — GO Cellular Component: GO:0005739 mitochondrion). - Renal apical/basolateral membrane transporters (NaDC-1, OAT1/OAT3) — GO:0005886 plasma membrane.

Localization/lateralization: - Renal injury is typically bilateral/diffuse (both kidneys). - Cranial neuropathy is characteristically bilateral (bilateral facial palsy is described as the "classic" finding in delayed DEG neurotoxicity), distinguishing it from many other causes of facial palsy which are typically unilateral.

8. Temporal Development

Onset: Acute, occurring in previously healthy individuals following a discrete toxic ingestion event; no congenital or age-restricted susceptibility beyond the empirical predominance of severe outcomes in young children in outbreak settings.

Onset pattern and staging (well-defined three-phase natural history): 1. Phase 1 (hours): Acute GI/inebriation phase. 2. Phase 2 (1–3 days): Renal failure phase — hallmark of the disease; may include cardiac, pulmonary, hepatic, and pancreatic involvement. 3. Phase 3 (5–20 days, up to ~2 weeks classically): Delayed neurologic phase in survivors.

Progression rate: Can be rapid and fulminant (death within days in severe pediatric cases) or, in survivors of the renal phase, slowly evolving over weeks (neurologic phase) with further slow partial recovery over 4–6 months.

Disease course pattern: Monophasic in fatal cases; in survivors, a triphasic pattern (acute–renal–delayed neurologic) with potential for chronic sequelae (dialysis-dependence, residual neuropathy).

Duration: Acute poisoning is self-limited in mild cases with supportive care; severe cases can result in chronic kidney disease requiring long-term dialysis and persistent neurologic deficits.

Remission patterns: Neurologic signs in survivors have been reported to improve over 4–6 months, though renal dialysis-dependence may persist beyond that window (PMC4616334).

Critical intervention window: Early presentation (within ~12 hours) may precede the development of overt metabolic acidosis, meaning early cases can be missed on initial labs — antidotal treatment initiated before extensive metabolite generation (i.e., before ADH has converted much of the ingested DEG) is mechanistically the most effective window, paralleling the treatment logic for ethylene glycol/methanol.

9. Inheritance and Population

Not a genetic disease — there is no inheritance pattern, penetrance, expressivity, anticipation, mosaicism, or carrier frequency to report; this is an acquired environmental/toxic disease.

Epidemiology (prevalence/incidence): DEG poisoning has no baseline "prevalence" in the sense of an endemic disease; it occurs in discrete, source-attributable outbreaks. Documented major outbreaks and their scale:

Outbreak Year Location Deaths (approx.) Source Citation
Sulfanilamide elixir 1937 USA >100 Sulfanilamide elixir vehicle (historical; foundational to US drug regulation)
Paracetamol syrup 1990 Nigeria (Jos) 47 children Contaminated glycerin PMID:1280035
Acetaminophen elixir 1990 Bangladesh dozens DEG-contaminated syrup ScienceDirect overview
Acetaminophen syrup 1995–96 Haiti ≥76 children (most confirmed deaths; near-total mortality in Haiti-based cases) DEG-contaminated glycerin PMID:8769471; CDC MMWR
Cough syrup 2006 Panama 219 DEG mislabeled as glycerin from Chinese manufacturer via European trader ScienceDirect
Nigeria (paracetamol) 2008–2009 Nigeria dozens (fatal poisoning cluster) Contaminated acetaminophen PMID:20010509
Pediatric cough/cold syrups 2022 The Gambia 66 deaths of 82 AKI cases (80% case-fatality among AKI cases) Maiden Pharmaceuticals (India) exported syrups MMWR/CDC
Cough syrup 2022 Indonesia >200 children DEG/EG-contaminated syrup Chemistry World
Cough syrup 2023 Uzbekistan 19 children DEG/EG-contaminated syrup WHO
Coldrif and related syrups 2025 India (Madhya Pradesh/Rajasthan) ≥24 children (19 in Madhya Pradesh alone) Sresan Pharmaceuticals; DEG up to ~45% of product CNN; Wikipedia

Cumulatively, WHO issued six global Medical Product Alerts between October 2022 and 2023 for DEG/EG-contaminated OTC medicines, with over 300 pediatric deaths across the Gambia, Indonesia, and Uzbekistan outbreaks alone (WHO).

Population demographics: - Age: Overwhelmingly young children (median age 19 months in the Gambia outbreak, range 5 months–7 years) in the pharmaceutical-contamination outbreaks; adult cases occur predominantly via occupational, recreational, or self-harm ingestion. - Sex ratio: No strong sex predilection has been reported specifically for DEG poisoning; exposure follows medication administration patterns rather than biological sex differences. - Geographic distribution: Disproportionately affects low- and middle-income countries with weaker pharmaceutical regulatory/testing infrastructure — India (as a manufacturing source implicated in multiple outbreaks), The Gambia, Indonesia, Uzbekistan, Panama, Haiti, Bangladesh, Nigeria. This reflects supply-chain and regulatory failure rather than any biological geographic susceptibility. - Founder effects/consanguinity/ethnic susceptibility: Not applicable (non-genetic disease).

10. Diagnostics

Clinical/laboratory tests: - Basic metabolic panel: elevated anion gap, metabolic acidosis, elevated BUN/creatinine (AKI), hyperkalemia. - Osmolar gap: may be measured but is a less sensitive marker for DEG than for methanol/ethylene glycol due to DEG's higher molecular weight; a normal osmolar gap does not exclude DEG poisoning (EMCrit; PMC7785502). - Liver function tests: elevated transaminases. - Lipase/amylase: elevated in pancreatitis. - Urinalysis: may show tubular injury markers; oliguria/anuria clinically. - Direct DEG/metabolite quantification: gas chromatography (GC) or GC-mass spectrometry can measure serum DEG, 2-HEAA, and DGA directly and is the definitive diagnostic and forensic/outbreak-investigation tool; also used for pre-/post-hemodialysis clearance monitoring (WHO analytical methods, Dec 2023). - Renal biopsy/histopathology: acute tubular necrosis with epithelial vacuolization; documented in the Panama 2006 outbreak investigation as a defining histopathologic feature (ScienceDirect Panama). - Nerve conduction studies/EMG: in the delayed neurologic phase, demonstrates severe sensorimotor peripheral neuropathy (90% of tested cases in one series). - CSF analysis: albuminocytologic dissociation in the delayed neurologic (Guillain-Barré–mimicking) presentation.

Genetic testing: Not applicable — DEG poisoning is not diagnosed via genetic testing. (Research-stage transporter/enzyme genotyping — e.g., SLC13A2/NaDC-1 expression — remains investigational and is not part of clinical diagnostic practice.)

Clinical/diagnostic criteria: No formal DSM/ICD diagnostic criteria exist beyond history of exposure plus the clinical/laboratory triad of metabolic acidosis, AKI, and (if present) delayed neuropathy. Differential diagnosis includes methanol and ethylene glycol poisoning (toxic alcohol syndromes), sepsis-associated AKI, and Guillain-Barré syndrome (for the delayed neurologic phase) — the latter distinguished by exposure history and the temporal association with preceding AKI.

Screening: Population-level "screening" for DEG poisoning takes the form of regulatory testing of pharmaceutical excipients rather than individual patient screening: - USP-NF finished-dosage-form testing methods for DEG/EG (USP-NF, Feb 2025) - FDA Guidance for Industry (May 2023) mandating DEG/EG testing of glycerin, propylene glycol, maltitol solution, hydrogenated starch hydrolysate, sorbitol solution, and other high-risk components (Federal Register, May 2023; FDA guidance PDF) - WHO-developed two-tier analytical screening approach (TLC screening → GC confirmation) for the International Pharmacopoeia (December 2023) (WHO)

11. Outcome/Prognosis

Mortality: Case-fatality is highly dependent on dose and access to care. In the 2022 Gambia outbreak, 80% of children who developed AKI died (66 of 82 cases) (MMWR). In the 1995–96 Haiti outbreak, of the 76 affected children who remained in Haiti (i.e., were not evacuated for advanced care), only one is known to have survived (PMID:8769471) — underscoring how outcome is strongly modified by access to dialysis and antidote. Overall case-fatality across historical mass-poisoning events has generally exceeded 50% without early antidotal/dialysis treatment, and can approach near-total mortality in resource-limited settings without dialysis access.

Morbidity in survivors: - Chronic kidney disease / ongoing dialysis dependence, particularly in those who develop the delayed neurologic syndrome. - Persistent peripheral/cranial neuropathy, though partial improvement over 4–6 months has been documented in some survivors. - Long-term neurodevelopmental impact in young children who survive AKI is plausible but not well characterized in formal follow-up cohorts — a recognized evidence gap.

Complications: Secondary infections (from prolonged dialysis access), cerebral edema (in fulminant encephalopathy, can be directly fatal), chronic renal failure, permanent neurologic deficits.

Prognostic factors: - Time to treatment — earlier fomepizole/ethanol administration and earlier hemodialysis initiation both improve survival and reduce long-term damage (tandfonline, PMID review). - Degree of established renal impairment — predicts subsequent delayed neurotoxicity, i.e., more severe AKI predicts a higher likelihood of the delayed neurologic syndrome (PMC4616334). - Dose/concentration ingested. - Age — younger children fare worse in outbreak data, though this partly reflects dosing/exposure patterns rather than purely biological vulnerability. - Nutritional status (multivitamin use, as noted) — protective in the Gambia cohort.

12. Treatment

Pharmacotherapy (antidotal, primary treatment): - Fomepizole (4-methylpyrazole) — competitive ADH inhibitor; blocks DEG bioactivation to toxic metabolites. This is now the preferred first-line antidote given superior safety profile (fewer medication errors/adverse events) compared to ethanol (Univ. Utah Poison Control). - Dosing: loading dose 15 mg/kg IV, then 10 mg/kg every 12 hours for 4 doses; after 48 hours, dose increases to 15 mg/kg every 12 hours (reflecting fomepizole's autoinduction of its own CYP450-mediated metabolism); dosing/interval must be adjusted during hemodialysis (fomepizole is dialyzable) — continuous or supplemented dosing during HD sessions is required. - NCIT term: NCIT:C1421 (Fomepizole) or generic Pharmacotherapy NCIT:C15986 with therapeutic_agent bound to fomepizole. - Ethanol — alternative ADH inhibitor where fomepizole is unavailable; target serum concentration 100–125 mg/dL; requires more intensive monitoring (risk of hypoglycemia, CNS depression, medication errors) than fomepizole (NEJM fomepizole trial, PMID reference). - Activated charcoal — considered for early GI decontamination if presentation is within the appropriate window post-ingestion (limited evidence specific to DEG, extrapolated from general toxicology practice).

Extracorporeal treatment (renal replacement): - Hemodialysis (intermittent, IHD) — preferred modality; clears both DEG and its toxic metabolites (2-HEAA, DGA), and is essential in patients with significant acidosis, AKI, or large ingestions. Pre-/post-dialysis DEG level monitoring supports effective clearance (EXTRIP workgroup guidance for the ethylene glycol series, extrapolated to DEG). - CRRT — recommended when IHD is unavailable. - Extraction Corporeal Treatments in Poisoning (EXTRIP) criteria for the structurally related ethylene glycol (used analogously for DEG in practice, since no DEG-specific EXTRIP guideline exists): initiate ECTR for severe acidosis (anion gap >27 mmol/L), AKI, coma/seizures, or when standard antidote alone is insufficient; discontinue once anion gap normalizes. - NCIT term: NCIT:C15571 (Hemodialysis) or broader NCIT:C15329 (Surgical/Interventional Procedure category is not correct — hemodialysis maps under NCIT renal replacement therapy terms).

Supportive care: - Correction of metabolic acidosis (sodium bicarbonate as needed), electrolyte management (hyperkalemia), fluid management, blood pressure support, mechanical ventilation for pulmonary edema/respiratory failure, nutritional/pancreatitis supportive care. - Rehabilitation (physical/occupational therapy) for survivors with peripheral neuropathy/motor deficits (NCIT:C15302 Physical Therapy; NCIT:C15315 Rehabilitation).

Experimental/investigational: - NaDC-1 (SLC13A2) inhibition — proposed as a potential future therapeutic target based on rat transporter data, to block cellular DGA uptake in the kidney and thereby prevent nephrotoxicity independent of ADH blockade timing; not yet in human trials (PMC10263375). - No DEG-specific registered clinical trials (NCT) were identified; management is derived by analogy from the ethylene glycol/methanol toxic-alcohol treatment literature and outbreak case-series experience.

Treatment outcomes: Fomepizole is associated with reduced medication errors and adverse events relative to ethanol; combined antidote + hemodialysis strategy, initiated early, is associated with markedly improved survival compared with supportive care alone or delayed presentation (as starkly illustrated by the Haiti vs. evacuated-patient outcome contrast).

Treatment algorithm (general approach): 1. Recognize elevated anion gap metabolic acidosis with history/suspicion of glycol exposure. 2. Initiate fomepizole (or ethanol if unavailable) as soon as poisoning is suspected — do not wait for confirmatory GC testing. 3. Assess for hemodialysis indications (severe acidosis, AKI, large ingestion, clinical deterioration); initiate IHD/CRRT as indicated. 4. Provide supportive/organ-specific care (renal, hepatic, pancreatic, cardiovascular, respiratory). 5. Monitor for delayed neurologic syndrome in the 1–3 weeks following recovery from the acute renal phase; institute neurologic supportive care/rehabilitation as needed.

13. Prevention

DEG poisoning is almost entirely preventable, and prevention is overwhelmingly a regulatory/pharmacovigilance matter rather than a clinical/medical one, since the disease results from a supply-chain contamination failure rather than an intrinsic host risk process.

Primary prevention (dominant strategy): - Pharmacopoeial testing mandates: revised USP-NF excipient monographs now require DEG/EG limit testing for glycerin, propylene glycol, sorbitol solution, maltitol solution, hydrogenated starch hydrolysate, and PEG (USP-NF notice, 2025; FDA guidance, 2023). - FDA Guidance for Industry (May 2023): mandates testing of high-risk drug components for DEG/EG before use in finished pharmaceutical products (Federal Register). - WHO global alerts and analytical methods: six WHO Medical Product Alerts (2022–2023) and a validated two-tier TLC-screening/GC-confirmation testing protocol developed for the International Pharmacopoeia, intended for use by national regulatory laboratories, especially in LMICs with limited GC capacity (WHO, Dec 2023). - Supply-chain/manufacturer oversight: stricter certificate-of-analysis verification for each batch of glycerin/propylene glycol/sorbitol, rather than relying on supplier certification alone — a repeated root cause across Panama (2006), Haiti (1996), and the 2022–2025 outbreaks was acceptance of falsified or unverified certificates of analysis from chemical intermediaries. - Regulatory license revocation/enforcement: e.g., suspension of Sresan Pharmaceuticals' manufacturing license following the 2025 Madhya Pradesh cluster.

Secondary prevention (early detection/outbreak response): - Rapid epidemiologic investigation and product recall once clusters of unexplained pediatric AKI are recognized (the pattern-recognition approach that identified the causative product in each of the historical outbreaks). - Sentinel surveillance for pediatric AKI clusters as an early-warning signal for potential DEG/EG contamination events.

Tertiary prevention: Early antidotal treatment and dialysis access (see Treatment section) to prevent complications once exposure has occurred.

Public health interventions: - WHO Medical Product Alerts distributed globally to national regulators upon detection of a contaminated product. - International information-sharing and rapid recall coordination between manufacturing-country and importing-country regulators (a persistent weak point, since several outbreaks involved products exported from one country and causing deaths in another before recall).

Counseling/behavioral: Public and prescriber education discouraging use of unregulated/informally sourced liquid medicines, particularly in outbreak-affected regions during active recalls.

Immunization/prophylactic medication: Not applicable (non-infectious disease).

14. Other Species / Natural Disease

  • Taxonomy of affected species: Primarily a human poisoning syndrome via pharmaceutical/industrial exposure; among animals, dogs and cats are the most commonly reported veterinary victims of glycol poisoning, though the veterinary literature overwhelmingly concerns ethylene glycol (antifreeze) rather than DEG specifically, reflecting differential product exposure patterns (antifreeze is predominantly ethylene glycol-based) (NCBI Taxon: Canis lupus familiaris NCBITaxon:9615; Felis catus NCBITaxon:9685).
  • Natural/accidental veterinary disease: DEG-specific veterinary poisoning case reports exist but are far less common in the literature than ethylene glycol veterinary toxicosis; when it occurs, the pathophysiology (ADH/ALDH bioactivation, renal DGA accumulation, proximal tubular necrosis) is believed to be conserved across mammals, consistent with the rat experimental model data forming the bulk of the mechanistic evidence base.
  • Comparative pathology: The rat is the dominant experimental model and recapitulates the核心 renal pathophysiology (proximal tubular necrosis via DGA/NaDC-1) and, in more recent studies, neurotoxic effects as well (Clin Toxicol 60:3, PMID female rat neurotoxicity study; Clin Toxicol 59:9, neurotoxic effects).
  • Evolutionary conservation: ADH/ALDH-mediated glycol bioactivation and dicarboxylate-transporter-mediated renal toxicant uptake are conserved biochemical pathways across mammals, underpinning the translational validity of the rat model for human mechanism (though not necessarily for exact dose thresholds — rat LD50 for DEG is far higher on a per-kg basis, 20–25 g/kg, than doses causing severe toxicity in human children, reflecting toxicokinetic/toxicodynamic differences and the smaller absolute margin of safety in small children receiving syrup-dosed medication).
  • Zoonotic potential: Not applicable — this is a direct chemical toxicity, not a transmissible infectious disease.

15. Model Organisms

Primary model: Rat (Rattus norvegicus) — by far the best-characterized animal model for DEG toxicity. - Strains used: Wistar, Fischer-344, and Wistar-Han rats (threshold dose-response study, PMID:25545985; PMC5237385). - Model type: Chemically induced (acute or repeated-dose oral gavage), not genetic. - Key findings: - LD50 in male Wistar rats: 20–25 g/kg body weight. - Sharp threshold dose-response: at 2 and 5 g/kg DEG, no kidney diglycolic acid (DGA) accumulation or renal/hepatic toxicity occurred in either Wistar or Fischer-344 strains; at 10 g/kg, DGA accumulated markedly and toxicity was observed — establishing that toxicity requires metabolite accumulation past a threshold, not simple dose-linear injury. - Toxicokinetics: rapid, near-complete GI absorption; tissue distribution kidney > brain > spleen > liver > muscle > fat (PMID:8135655). - Nephrotoxicity model recapitulates human findings well: hydropic degeneration of renal tubules, anuria, azotemia, death at 2–7 days post-dosing in high-dose groups. - Neurotoxicity model (newer, 2021–2022): repeated-dose studies in female rats demonstrated both nephrotoxic and neurotoxic effects of DEG, extending the model beyond the classic acute nephrotoxicity paradigm to capture the delayed neurologic phenotype seen in human survivors (Clin Toxicol 60:3; Clin Toxicol 59:9). - Variable individual susceptibility model (2023): identical DEG/DGA dosing produced AKI in only ~60% of rats; susceptibility correlated with renal NaDC-1 (SLC13A2) mRNA expression level, providing a candidate mechanistic explanation for the well-documented variable human susceptibility across outbreaks (PMC10263375). - Model limitations: Rat LD50 (20–25 g/kg) is far higher per unit body weight than doses causing fatal human pediatric poisoning, and rats do not spontaneously develop the classic human delayed cranial-neuropathy/facial-diplegia syndrome in most protocols — the neurotoxicity models capture peripheral neurotoxic endpoints but have not been shown to fully recapitulate the bilateral facial palsy/bulbar syndrome that is the human hallmark. Translational fidelity for the acute nephrotoxicity mechanism (DGA/NaDC-1) is considered high; fidelity for the full human delayed neurologic syndrome is lower/still being established.

In vitro / cellular models: - Human proximal tubule cells (HPT cells) and RPTEC/TERT1 cell line — used extensively to establish DGA as the direct nephrotoxic metabolite, to characterize NaDC-1-mediated apical uptake, and to demonstrate that OAT1/OAT3-mediated basolateral efflux does not rescue cells from DGA accumulation (PMID:21856646; PMID:36087010; PMID:36958672). These in vitro human-cell models are actually a higher-fidelity human-relevant system for the core nephrotoxic mechanism than the whole-animal rat model, since they directly use human renal tubule cells.

Other model systems: No zebrafish, Drosophila, C. elegans, or iPSC-derived model systems specific to DEG toxicity were identified in this search — this is a notable gap relative to the depth of rat and human-cell-line data, and could represent an opportunity area for future model development (e.g., an iPSC-derived proximal tubule organoid model to further probe individual human variability in NaDC-1 expression/DGA sensitivity).

Summary Table: Suggested Ontology Bindings for KB Curation

Concept Suggested term
Disease/chemical entity CHEBI:46807 (diethylene glycol)
Related disease entity (nosological neighbor) ORPHA:31826 (Ethylene glycol poisoning) — no dedicated DEG MONDO/Orphanet term identified
Causal gene/enzyme family ADH1A/ADH1B/ADH1C (hgnc:...), ALDH2 (hgnc:404) — mechanistic, not causal-variant
Key transporter SLC13A2/NaDC-1 (hgnc:11035)
Key phenotype (renal) HP:0001919 Acute kidney injury; HP:0100518 Anuria
Key phenotype (neuro) HP:0010628 Facial palsy; HP:0007141 Peripheral neuropathy; HP:0001298 Encephalopathy
Biological process GO:0006066 alcohol metabolic process; renal dicarboxylate transport
Cell type CL:1001016 kidney proximal tubule epithelial cell
Anatomical site UBERON:0002113 kidney
Treatment (antidote) NCIT — Fomepizole/therapeutic_agent under NCIT:C15986 Pharmacotherapy
Treatment (renal replacement) Hemodialysis / CRRT

Notable Evidence Gaps

  1. No human genotype-outcome study has directly tested ADH1B/ALDH2 or SLC13A2 variant effects on DEG poisoning severity, despite strong mechanistic rat/in-vitro rationale.
  2. No standardized quality-of-life outcome data for DEG poisoning survivors (chronic kidney disease burden, residual neuropathy disability) using EQ-5D/SF-36 or pediatric equivalents.
  3. No DEG-specific EXTRIP consensus guideline exists; clinical extracorporeal-treatment practice is extrapolated from the ethylene glycol EXTRIP recommendations.
  4. No dedicated MONDO/OMIM identifier for DEG poisoning as a distinct entity from ethylene glycol poisoning was identified — a candidate curation gap for the disorder ontology.
  5. Long-term neurodevelopmental follow-up of young pediatric survivors (the population most affected in the 2022–2025 outbreaks) is not yet published in the peer-reviewed literature.

Sources

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Outcome Count
References checked 31
Resolved 31
Unresolved (possible confabulation) 0
Unverifiable 0
References weighed for topical relevance 31
On topic 18
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:

  • PMC:PMC11841886 (3 mentions) - Global and national actions to prevent trade in substandard and adulterated medicines.
  • shared terms: none

Weighed against this report's own most characteristic terms: deg, poisoning, renal, glycol, outbreak, aki, kidney, toxic, dga, nadc-1, disease, delayed, metabolite, exposure, peripheral, human, acute, ethylene, proximal, adh.

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

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Outcome Count
Terms checked 59
Resolved 57
Unresolved (possible confabulation) 0
Obsolete 1
Unverifiable 1
Terms whose name was checked 28
Terms named correctly 17
Terms named as a different term 7
Terms whose name is worth a second look 4

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

  • CHEBI:46807 (8 mentions) - the report calls it "parent compound", "diethylene glycol"; CHEBI calls it diethylene glycol
  • HP:0012647 (1 mention) - the report calls it "Autonomic dysfunction"; HP calls it Abnormal inflammatory response
  • CL:0000182 (1 mention) - the report calls it "site of ADH/ALDH-mediated bioactivation and secondary hepatotoxicity"; CL calls it hepatocyte
  • CL:0002064 (1 mention) - the report calls it "site of DEG-associated pancreatitis"; CL calls it pancreatic acinar cell
  • NCIT:C1421 (1 mention) - the report calls it "Fomepizole"; NCIT calls it Oxygen Radical
  • NCIT:C15571 (1 mention) - the report calls it "Hemodialysis"; NCIT calls it Differentiation Therapy
  • NCIT:C15329 (1 mention) - the report calls it "Surgical/Interventional Procedure category is not correct — hemodialysis maps under NCIT renal replacement therapy terms"; NCIT calls it Surgical Procedure

Obsolete terms

These terms are real but deprecated. Citing one is not a fabrication; it does mean the report is naming something the ontology has retired:

  • GO:0070265 (obsolete necrotic cell death) (1 mention)

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

  • HP:0001942 (2 mentions) - the report calls it "Metabolic acidosis (high or normal anion gap)", "Refractory metabolic acidosis"; HP calls it Metabolic acidosis
  • HP:0001733 (1 mention) - the report calls it "Acute pancreatitis"; HP calls it Pancreatitis
  • HP:0007141 (2 mentions) - the report calls it "Sensorimotor peripheral neuropathy"; HP calls it Sensorimotor neuropathy, and lists "Sensorimotor peripheral neuropathy" among its other names
  • CL:1001016 (2 mentions) - the report calls it "kidney proximal convoluted tubule epithelial cell"; CL calls it kidney loop of Henle ascending limb epithelial cell

Terms named inconsistently

The report gives these identifiers more than one name of its own:

  • CHEBI:46807 - called "parent compound", "diethylene glycol"
  • HP:0001942 - called "Metabolic acidosis (high or normal anion gap)", "Refractory metabolic acidosis"

Prefixes with no resolver

Terms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: ORPHA.