Accuracy of screening for hemolysis in plasma samples using a commercial urine dipstick

Abstract Objective. Identification and quantification of hemolysis in serum or plasma samples is an important requirement in laboratory diagnostics, however this is not always possible automatically and visual sample inspection is not sufficiently accurate. We have planned this study to determine whether a commercially available urine dipstick with a hemoglobin-reactive pad could be used for this purpose. Methods: Sixty-five routine plasma samples, whose hemolysis index was previously determined on Roche Cobas 8000, were assayed with a commercial urine dipstick to obtain semi-quantitative data on plasma hemoglobin. Plasma was diluted 1:1000 in water to enter the range of hemoglobin measurement of the dipstick, and 10 µl of this dilution were applied to the hemoglobin pad. Results were visually interpreted within 60 sec by comparing the pad color with that on the dipsticks box label. Results: The sample size consisted of 40 non-hemolyzed (hemolysis index≤0.3 g/L) and 25 hemolyzed (hemolysis index>0.3 g/L) plasma samples. Spearman’s correlation between Cobas hemolysis index and dipstick hemoglobin concentration was r=0.96 (95%CI, 0.93-0.97; p<0.001). The concordance of hemolysis detection was 95.4%, with 1.00 sensitivity, 0.93 specificity, 1.00 negative predictive value and 0.89 positive predictive value compared to the reference hemolysis index measurement on Cobas. The cumulative agreement between Cobas hemolysis index and the various plasma hemoglobin thresholds obtained with the dipstick was 75.4%. Conclusions: We have demonstrated here that plasma hemoglobin assessment with commercially available urine dipsticks may generate semi-quantitative test results accurate enough to influence decision making regarding sample quality and its suitability for testing.