Serum myoglobin (Mb) was assayed by immuno-nephelemetry or immuno-turbidimetry together with creatinine kinase activity (CK) by spectrophotometry in 290 consecutive patients admitted to hospital between January 1st and September 30th 1992 in three cardiology departments for chest pain suggesting myocardial infarction (MI). The measurements were made at admission (T0) and 90 minutes later (T90). On admission, patients were classified as certain MI (N = 62), possible MI (N = 107) or definitely not MI (N = 121) and, on discharge, as certain MI (N = 144) or definitely not MI (N = 146). At T0, for a threshold value of 90 mcg/l, Mb assay had a sensitivity of 49.3% and a specificity of 95.2%, a positive predictive value of 91.8% and a negative predictive value of 65.6%. Increasing the threshold of positivity to 130 mcg/l was accompanied by a significant loss of sensitivity (38.2%) without any change in the other parameters. At T90, for a threshold value of 90 mcg/l, Mb assay had a sensitivity of 81.7%, a specificity of 92%, a positive predictive value of 89.5% and a negative predictive value of 85.8%. The increase in sensitivity between T0 and T90 made Mb assay very useful for correctly classifying the initial false negative results (20/28: 71.4%) and for diagnosing the possible MI (27/32: 84.4%). Decreasing the threshold of positivity to 80 or even to 70 mcg/l did not improve the diagnostic value of this test. The sensitivity of Mb assay was significantly higher than that of CK at T0 (49.3% vs 26.4%: p < 0.0001) and at T90 (81.7% vs 48.9%: p < 0.0001).(ABSTRACT TRUNCATED AT 250 WORDS)