How good are clinical chemistry laboratories at analysing ethylene glycol?

The results of an external proficiency test of clinical chemistry laboratories in Sweden when the target analyte was ethylene glycol (EG) are presented. Specimens of plasma were spiked with EG (10% w/v) to give assigned concentrations ranging from 5 to 50 mmol/L. Over a period of 6 years, two control specimens of plasma were sent for analysis on 21 occasions to between 14 and 20 participating laboratories as a declared proficiency trial. The analytical precision between and within laboratories was determined by spiking the plasma specimens with the same concentration of EG so that the results reported back could be considered a duplicate determination. On one occasion propylene glycol (PG) was substituted for EG without informing the participants. The standard deviation (SD) within laboratories expressed as the coefficient of variation (CV) was 4.5% compared with 11.4% between laboratories. Results reported by laboratories using gas chromatography (GC) were in good agreement with those when an enzymatic method was used. The between-laboratory SD increased with concentration of EG in the specimen and at a mean concentration of 18 mmol/L, the pooled SD was 4.11 mmol/L (CV = 23%). Four laboratories reported finding EG in plasma when PG was the diol present; three laboratories used an enzymatic method and one used GC. Clinical laboratories that provide a toxicology service should regularly participate in external quality assurance schemes that include low-molecular-weight alcohols such as EG. Efforts should be made to standardize the analytical methods used for toxicological analysis.