Improved Sensitivity and Discriminative Power of Factor FVIII Assay By Applying Turbidimetric Clotting Curve Analysis

Background: The need for sensitive detection of coagulation factor between the levels of 0.0 and1.0% is now growing continuously. The most popular method of measuring factor VIII (FVIII) activity is one stage clotting time (activated partial thromboplastin time, APTT)-based assay. From the first and the second derivatives of the original turbidimetric curve, the velocity and acceleration of clot formation can be followed and parameters like maximum velocity (peak1) and acceleration (peak2) of clot formation can also be derived. We examined the limit of detection of FVIII activities measured based on clotting time, peak1 and peak2, following the recommendations from Clinical and Laboratory Standards Institute guideline.

Method: We performed APTT with sample/reagent volume and incubation time modified to be identical to those adopted for factor assay on factor deficient plasma as blank and plasmas prepared to have variable FVIII activities from 0.1 to 1.6%. Peak1 and Peak2 were also measured to determine the limit of blank (LoB) and limit of detection (LoD) corresponding to each parameter. Also, by modifying the method of LoB and LoD determination, we determined lower limit of 1.0% (lower Lo1) corresponding to each parameter.

Results: The mean clotting time of blank sample (FVIII 0.0%) was 131.4 ± 1.90 seconds (mean ± SD). Thus, 95% of clotting times measured from a blank sample (FVIII 0.0%) are longer than 128.3 seconds, which was determined to be the LoB for clotting time. The pooled standard deviation (SD) of clotting times measured from low FVIII samples was 2.29 seconds and 124.49 seconds which was the theoretically minimum value that was statistically different from LoB was set as the clotting time corresponding to LoD of FVIII. The LoD clotting time was between the clotting times of FVIII 0.2% and FVIII 0.4% sample and 0.4% was safely determined to be LoD of FVIII. This implied that plasma sample with at least 0.4% FVIII level was guaranteed to be measured higher than blank (FVIII 0.0%) sample. The mean peak1 height for blank sample was 21.4 ± 0.68 and LoB peak height was determined to be 22.53. The pooled SD for peak1 height was 0.68 and peak1 height of LoD was calculated to be 23.63. Because the mean peak1 height for 0.2% sample was 24.27, the LoD FVIII activity could be safely determined to be 0.2%. Thus, by applying peak1 as primary measure to estimate FVIII activity, the sensitivity of FVIII assay was increased with lower LoD of 0.2% compared with clotting time based assay. For peak2 height, LoB and LoD peak2 height were 14.86 and 18.59 respectively and The LoD could be set at FVIII 0.4%. Next, we determined lower Lo1, which meant FVIII level that was guaranteed to be measured significantly lower than 1.0% sample. For clotting time, lower Lo1 was FVIII 0.2% and for peak1 and 2 FVIII, was 0.4%. These results implied that by conventional clotting time based FVIII assay FVIII activity between 0.0 and 1.0% could not be measured credibly. FVIII should be at least 0.4% to be ever detected but ironically FVIII should be less than 0.2% to be assuredly measured lower than 1.0%. However, with peak1 there was an interval of FVIII value that could be assured to be measured higher than 0.0% but lower than 1.0%.

Conclusion: We concluded that the maximum clotting velocity derived from turbidimetric curve analysis can be applied to measure FVIII activity between 0.0 and 1.0% credibly.