Effect of Thromboplastins and Instrumentation on the Prothrombin Time Test: Implications for National Standards
The independent effects that different thromboplastins and types of methods have on prothrombin time were studied by analyzing data from a 1977 proficiency testing survey of 2735 laboratories. Each laboratory received three plasma samples with prothrombin times of approximately 12, 16 and 24 seconds which represented normal, partially anticoagulated, and fully anticoagulated plasma respectively. Combinations of 12 different instruments and 12 different thromboplastins were used by the laboratories in evaluating the samples. As determined by a two-way analysis of variance, both the type of thromboplastin and the type of instrument have similar effects upon the prothrombin time. When the mean prothrombin time corrected for independent effects of thromboplastin and method are plotted against the overall mean prothrombin times of each of the three types of plasma samples, the linear relationship between the two is obvious. These relationships can used to convert to a common scale a prothrombin time obtained with any thromboplastin method combination and provides a means for standardizing the prothrombin time test. Retrospective, analysis of data obtained from individual proficiency testing surveys in 1976, 1977, and 1978 suggests that the relationships remained constant during those years. These data suggest that the relationship of any thromboplastin to a thromboplastin standard should be determined independently of the method effect.