Comparison of variability of different methods of measuring fibrinolytic activity

Abstract

The Northwick Park Heart Study has drawn attention to the variability, especially within-person, of the dilute blood clot lysis time (DBCLT). However, this method has considerable practical advantages in large-scale surveys. The variability of DBCLT has therefore been compared with that of the euglobulin lysis time (ELT) and the euglobulin fibrin plate area (EFPA). Each measurement has been made 10 times over a 9 month period in each of 12 healthy people; fibrinogen levels have also been measured. As expected, highly significant correlations (r) have been obtained between the three methods: DBCLT and ELT, + 0.63; DBCLT and EFPA, - 0.47; ELT and EFPA, - 0.54 (P<0.001 in each case). The between-person variance, distinguishing one person from another, accounts for 60% of the total variance in DBCLT, 35% in ELT and 10% in EFPA. For fibrinogen, the value is nearly 90%. The rest of the variance in each case is accounted for by biological within-person variation and laboratory error. Thus, while DBCLT is a variable method of determining an individual’s characteristic level of fibrinolytic activity, it is probably less so than other commonly used methods. The results of this experiment suggest that a previous estimate of the within-person variability of DBCLT (Meade and North, 1977, British Medical Bulletin, 33, 283) may have been too high.