Prediction of carrier status in Duchenne dystrophy by creatine kinase measurement

Serum creatine kinase (CK) was measured in 515 healthy white women and 28 obligate carriers for Duchenne muscular dystrophy. There was substantial overlap between the control and carrier populations. To analyze the impact of the degree of overlap, the predictive value of a CK result was determined by (1) using sensitivity and specificity analysis, which assumes dichotomization into a positive or negative result based on a particular cut-off value; and (2) using likelihood ratio analysis, which evaluates an individual result based on the continuum observed for control and carrier populations. There was no clinically important difference whether an observed 57% or a hypothetical 33% overlap between control and carrier results was used. Because of the substantial overlap, the CK test utility is limited to those suspected carriers whose results fall above the healthy population interval. A low CK result does not provide sufficient assurance of noncarrier status.