Comparison of mean normal prothrombin time (PT) with PT of fresh normal pooled plasma or of a lyophilized control plasma (R82A) as denominator to express PT results: collaborative study of the International Federation of Clinical Chemistry. IFCC Working Group Standardization of Coagulation Tests.

The mean normal prothrombin time (MNPT) is currently recommended as the denominator term in the expression of PT ratio or International Normalized Ratio (INR) values. The PT of lyophilized normal control plasmas might also be used in calculating PT ratios, but the overall accuracy of this approach and its dependence on reagents and endpoint detectors have not been evaluated in detail. In an IFCC collaborative study involving 15 expert laboratories and 58 PT systems, the PT ratios of 30 apparently healthy subjects were expressed with the use of the MNPT, the PT of fresh normal pooled plasma (FNPP) obtained from the same apparently healthy subjects, or the PT of plasma R82A--a lyophilized normal pooled plasma prepared by the Verband der Deutschen Geräte-Hersteller for in-house calibration of a large amount of control plasma--as the denominator term. The total imprecision of the PT of plasma R82A averaged 2.16%. Mean PT ratios did not differ from 1.00 (mean 1.00, range 1.00-1.01) with the use of the MNPT as the denominator term. Mean PT ratios were > 1.00 with the FNPP-PT as the denominator term (1.02, 0.96-1.05), and differed according to endpoint detectors (P = 0.024). Mean PT ratios with plasma R82A-PT as the denominator term averaged 0.98 (range 0.91-1.06) with plain thromboplastins (n = 11), 1.02 (0.98-1.06) with combined thromboplastins (n = 3), and 0.93 (0.87-0.97) with recombinant thromboplastins (n = 2), but they differed according to the brand of plain or recombinant reagents (P = 0.00001), the endpoint detector (P < 0.0025), and the plasma citrate concentration (P < 0.0025). These findings underline the differences in the PT of lyophilized plasma R82A and the MNPT and PT of FNPP obtained from the same individuals and support the recommendation that the system-specific MNPT should be used as the ideal index of the normal PT in the calculation of INR values.