External quality assurance of fibrinogen assays using normal plasma: Results of the 2008 college of american pathologists proficiency testing program in coagulation

Context. - Proper diagnosis and therapy of fibrinogen deficiency requires high-quality fibrinogen assays. Objective. - To assess the interlaboratory bias, precision, and grading of fibrinogen assays used by laboratories participating in the United States College of American Pathologists proficiency testing program in coagulation. Design. - Two identical vials of normal plasma were sent to more than 3500 laboratories. Participants measured fibrinogen levels using local methods. Results. - Fifty different fibrinogen methods were evaluated. All-method bias was 8.3% (range of method-specific biases, 0.0%-27.0%) and all-method coefficient of variation was 7.7%(range ofmethod-specific coefficients of variation, 0.7%-25.8%). After controlling for reagent/instrument type, mean fibrinogen levels were 11.6% higher for prothrombin time-based reagents compared to Clauss (P , .001), and coefficient of variation was 46% lower for mechanical endpoint instruments compared to photo-optical. Most testing events (97.4%) could be reliably graded as pass or fail using a target range of ±20% from the method mean (total pass rate, 98.8%). Total fail rate was 3.0-fold lower for mechanical instruments compared to photo-optical (0.5% versus 1.5%, P = .001). Nonetheless many photo-optical methods had very high precision and very low fail rates. Conclusions. - Fibrinogen assays showed highly variable methodology and performance characteristics. Bias, precision, and grading were affected by the type of reagent or instrument used.