Three surveys of the measurement and interpretation of creatine kinase (CK; EC 22.214.171.124) isoenzyme 2 (CK-MB) were conducted in Ontario, Canada, in 1989. Of the clinical laboratories participating, 66% used immunological methods and 24% used electrophoretic methods. Although reference ranges and interpretative routines varied widely, 95% of the laboratories reported correct interpretations for 10 of the 15 vials tested. The only major problems occurred with samples with very low total CK activity. Within-survey duplicate results compared well, and 89% of the laboratories had consistent between-survey results, even for specimens with low total CK activity. Errors were proportional to the frequency of use of the different analytical methods. The lyophilized testing material gave higher results with methods for measuring the mass of CK-2, suggesting that the material contained inactive but immunologically intact CK-2. The surveys indicate that laboratories should review their protocols for measuring CK-2 when only a single sample from the patient is available.