Lactate Gap Cracks Rare Antifreeze Poisoning Case in Regional ICU Without Fomepizole

A 62-year-old man found unresponsive in his home in regional Queensland survived one of the most severe metabolic crises in clinical toxicology, and the story of how he survived is now being told in a case report published in Clinical Case Reports. He arrived at a small regional hospital in a coma with a blood pH of 6.88, a level at which most biochemical reactions in the body begin to fail. His kidneys were shutting down, his ionized calcium had collapsed, and his point-of-care blood gas analyzer reported a lactate reading above 31 mmol/L, the upper limit of the machine itself. Yet within hours, clinicians discovered that the most alarming number on his chart was not real. The true diagnosis was ethylene glycol poisoning, the toxic alcohol found in antifreeze, and the key that unlocked it was a striking discrepancy between two lactate measurements.

The clinical picture on arrival was bewildering. The man had a history of heavy alcohol use, no witnesses to any ingestion, and no antifreeze or toxic substance found at the scene. He had vomited and aspirated before paramedics could secure his airway. The combination of coma, extreme high-anion gap metabolic acidosis, acute kidney injury, and profound hypocalcemia with an ionized calcium of 0.90 mmol/L raised the possibility of ethylene glycol toxicity early, because oxalate crystals formed during the metabolism of ethylene glycol bind calcium in the blood. But the differential diagnosis remained broad. Septic and cardiogenic shock were considered and excluded by bedside echocardiography, which showed normal biventricular function with hyperdynamic, vasodilated hemodynamics. A swollen left leg prompted concern about vascular obstruction or compartment syndrome, but pulses were intact and compartment pressures were normal. Creatine kinase of 1600 U/L was far too low to explain the acidosis through rhabdomyolysis, and salicylate and paracetamol levels were unremarkable.

The decisive moment came at 19:00, when the formal laboratory result returned. The point-of-care arterial blood gas had shown a lactate of 28.0 mmol/L at the peripheral hospital, later exceeding 31.0 mmol/L in the intensive care unit. The contemporaneous laboratory serum lactate was 0.8 mmol/L, entirely normal. This discordance, a lactate gap of more than 30 mmol/L, is a recognized artifact of assay chemistry rather than a true laboratory error. Point-of-care and blood gas analyzers measure lactate using a lactate oxidase electrode, which also oxidizes glycolate and glyoxylate, the toxic metabolites of ethylene glycol, and therefore overreads dramatically. The central laboratory platform uses a method specific for L-lactate that is not meaningfully cross-reactive with glycolate. The interference is analyzer-dependent, and the magnitude of the gap correlates with glycolate concentration without being equal to it, making it a qualitative clue rather than a quantitative measure.

Equally instructive was what the biochemistry did not show. The measured osmolality was 344 mOsm/kg against a calculated value of 339 mOsm/kg, giving an osmolar gap of just 5 mOsm/kg, essentially normal. Under conventional teaching, a normal osmolar gap argues against toxic alcohol poisoning. But in delayed presentations, that logic fails. As ethylene glycol is metabolized into glycolate and oxalate, the parent compound that generates the osmolar gap is consumed, and the anion gap rises in its place. The anion gap of 28 to 31 mEq/L alongside a near-normal osmolar gap is the signature of an intermediate-to-late presentation, a pattern that is well described but under-appreciated, particularly at peripheral hospitals where toxic alcohol poisoning is rare and confirmatory assays are unavailable in real time.

Confirmation, when it eventually came, took 58 hours. Ethylene glycol was added retrospectively to a stored specimen collected at presentation, and the state reference laboratory returned a level of 5.8 mmol/L, a result that played no role in acute management. The patient, who remained delirious for days and later denied ingesting antifreeze, never provided a definitive ingestion history. Management therefore proceeded entirely on clinical and surrogate biochemical grounds, a reality that the report’s authors argue regional clinicians must be prepared to face. Ethylene glycol is present in many industrial and household products beyond antifreeze, so the absence of a witnessed ingestion should not reduce suspicion when the biochemical pattern fits.

With the diagnosis established, the practical constraints of regional medicine took over. Fomepizole, the preferred alcohol dehydrogenase inhibitor, was not held on site and could not be accessed within the required timeframe, so ethanol, prepared in advance by pharmacy, became the antidote of necessity. Continuous veno-venous hemodiafiltration was commenced at 19:30, one hour after ICU admission, at an effluent dose of 40 mL/kg/h, substantially above the standard 25 to 30 mL/kg/h prescription, to maximize clearance of glycolate, oxalate, and residual ethylene glycol. The choice of anticoagulation was itself a calculated decision. Citrate anticoagulation, the guideline-preferred modality for continuous renal replacement therapy, was explicitly avoided because citrate further chelates ionized calcium in the extracorporeal circuit, a dangerous proposition in a patient whose ionized calcium could not be sustained above 0.90 mmol/L despite aggressive supplementation. Unfractionated heparin was used instead, though filter clotting still occurred, consistent with a hypercoagulable milieu reflected by fibrinogen of 8.7 g/L and elevated FIBTEM amplitudes on rotational thromboelastometry.

The extracorporeal therapy worked quickly. Ethylene glycol fell from 5.8 mmol/L to 2.5 mmol/L four hours into treatment and became undetectable by approximately 14 hours, an apparent on-treatment half-life of roughly 4 to 6 hours, though ethanol blockade and residual metabolism confound any pure calculation of dialytic clearance. The chemistry favors removal: ethylene glycol weighs about 62 daltons, glycolate about 76, and oxalate about 90, all with volumes of distribution near total body water and negligible protein binding, properties that make them efficiently dialyzable. The EXTRIP workgroup recommends extracorporeal treatment when coma, acute kidney injury, or an anion gap above 27 mmol/L is present, all of which applied here, and endorses continuous therapy as the alternative where intermittent hemodialysis, the preferred modality, is unavailable, as was the case for a ventilated patient in this regional unit.

The ethanol therapy, however, carried its own consequences. To maintain a therapeutic serum concentration, clinicians infused a 37% ethanol solution through the nasogastric tube at 125 mL/h, roughly 3 liters over 24 hours. A therapeutic level of 26.6 mmol/L was confirmed on Day 2, within the target range of 20 to 30 mmol/L. But after 24 hours the patient developed progressive abdominal distension attributable to the cumulative enteral volume, compounded by preexisting aspiration injury and ileus. The distension may have worsened respiratory mechanics, increased ventilatory pressures, and contributed to hypoxemia in an already compromised respiratory system. By that point the metabolic acidosis had substantially resolved on hemodiafiltration and the anion gap was normalizing, so ethanol was ceased, balancing diminishing benefit against ongoing respiratory risk. The authors note that the resolution of acidosis occurred on extracorporeal therapy, not as a consequence of ethanol, and raise the question of whether the risk-benefit balance of ethanol is acceptable at all in late presentations when high-dose clearance is already running.

The outcome was ultimately favorable. The patient required five days of mechanical ventilation, with weaning complicated by alcohol withdrawal delirium requiring dexmedetomidine and benzodiazepines. He was extubated on Day 5, left the ICU on Day 7, and was discharged from hospital on Day 21, with creatinine recovering from a peak of 390 μmol/L to 90 μmol/L at discharge. The report draws three practical lessons: a normal osmolar gap does not exclude ethylene glycol poisoning in severe unexplained high-anion gap metabolic acidosis; the lactate gap, identified simply by comparing point-of-care and laboratory lactate values, is a valuable diagnostic tool where confirmatory assays are unavailable; and fomepizole, with its predictable pharmacokinetics, fixed 12-hourly dosing, and freedom from the sedation, withdrawal, and massive enteral volumes of ethanol, deserves improved stocking and rapid-access pathways in regional and rural Australian hospitals. A recent New South Wales cohort found fomepizole could have been advantageous in 64% of ethanol-treated patients, including avoiding ICU admission in eight cases. Survival from a pH of 6.88 without the preferred antidote is a testament to systematic clinical reasoning, but also a reminder of the gaps that persist in regional toxicology.

Subject of Research: Diagnosis and management of late-presentation ethylene glycol poisoning using the lactate gap and extracorporeal therapy without fomepizole

Article Title: Ethylene Glycol Poisoning Without Fomepizole in a Regional Australian Intensive Care Unit: The Lactate Gap as the Diagnostic Key and the Consequences of Nasogastric Ethanol Therapy

Article References: Athans, J., & Maan, P. (2026). Ethylene Glycol Poisoning Without Fomepizole in a Regional Australian Intensive Care Unit: The Lactate Gap as the Diagnostic Key and the Consequences of Nasogastric Ethanol Therapy. Clinical Case Reports, 14(10), Article e73698. https://doi.org/10.1002/ccr3.73698

Image Credits: AI Generated

DOI: 10.1002/ccr3.73698

Keywords: ethylene glycol poisoning, lactate gap, fomepizole, ethanol therapy, metabolic acidosis, CVVHDF, toxic alcohol, hypocalcemia, osmolar gap, regional ICU, acute kidney injury, clinical toxicology