Sonoclot coagulation analysis:: A study of test variability

Objective: To test the reproducibility of Sonoclot coagulation analysis (SCA; Sienco Inc, Morrison, CO). The authors wished to determine if the mix/release of the preloaded celite activator in standard Sono-cuvettes could be responsible for the high variation coefficients for SCA parameters with citrated whole blood and if citrated whole blood is optimal for SCA. Design: A prospective trial. Setting: A large academic teaching medical center. Participants: Eight healthy volunteers. Interventions: Repeated blood sampling was performed through indwelling radial artery catheters. Seven different Sonoclot analyzers were used to test seven different types of analysis procedures in the volunteers, involving activators of different types and amounts and in different forms, and the use of citrated or native whole blood. Measurements and Main Results: Two-way and one-way ANOVA, variance, variance analysis, and Tukey's test were used to evaluate differences in SCA methods and volunteer influence. A high variance, with SDs up to 200% of the median values of the SCA parameters with recalcified citrated blood and the standard Sono-cuvette, was confirmed. SCA with native blood and/or the use of other types of preloaded activators, ie, kaolin, significantly (p < 0.05) reduced this variance. Repeated SCAs further reduced the variance to 10% to 35% of the variance for a single analysis (standard cuvette and native blood). Conclusion: Improvement of the activator in the Sono-cuvette, use of native whole blood, and repeated Sonoclot analyses reduced the previously reported high variability of this instrument. Copyright (C) 1999 by W.B. Saunders Company.