Minding the osmol gap: a sentinel event and subsequent laboratory investigation

Many hospitals are unable to determine toxic alcohol concentrations in a clinically meaningful time frame. Thus, clinicians use surrogate markers when evaluating potentially poisoned patients.

A patient presented after an intentional antifreeze (ethylene glycol) ingestion with an osmol gap of -10.6 that remained stable one hour later. Further investigation revealed that the serum osmolality was calculated and not measured. The true osmol gap was 16.4, which correlated to a measured ethylene glycol concentration of 808 mg/L (80.8 mg/dL, 13.0 mmol/L).

A telephone survey of hospital laboratories in our catchment area was performed to investigate the potential for similar events.

Thirty-eight (47 percent) hospitals responded. No laboratories were able to test for toxic alcohols. One hospital (2.6 percent) reported routinely calculating osmolality based on chemistries, while two hospitals (5.3 percent) reported scenarios in which this might occur. Thirty-five (92.1 percent) hospitals could directly measure osmolality. Two hospitals (5.3 percent) were reliant on outside laboratories for osmolality measurement.

The 47 percent response rate and one geographic area are significant limitations.

Over 10 percent of hospitals that responded could have significant difficulty assessing patients with toxic alcohol ingestion.

Until the standard of rapidly obtaining toxic alcohol concentrations is broadly implemented, we recommend that policies and procedures be put in place to minimize errors associated with the determination of the osmol gap.