Reference change value and measurement uncertainty of CRP, ferritin, procalcitonin, troponin I, and BNP tests in COVID-19 monitoring: How much change is significant?

Abstract

Objectives: This study was designed to determine the point of clinically significant change in consecutive measurements of high concentrations of C-reactive protein (CRP), ferritin, procalcitonin, troponin I, and brain natriuretic peptide (BNP) in patients diagnosed with COVID-19. Methods: The reference change value (RCV) for each analyte was estimated using the following formula: RCV=21/ 2×Z×(CVa 2+CVi 2)1/2. The combined RCV was calculated using the mean CV value determined using different concentrations. Measurement uncertainty (MU) was estimated according to NordTest guideline. The root-mean-square (RMS) bias and nominal uncertainty values (uCref) were calculated. Standard uncertainty (ubias) and combined standard uncertainty (u) were calculated using the formula RMSbias²+uCref² and √(uRW²+ubias²)/2, respectively. Expanded uncertainty (U) was obtained by multiplying u by a coverage factor. Results: The lowest RCV values of CRP, ferritin, procalcitonin, BNP, and troponin I tests were 94.57%, 37.16%, 40.55%, 42.01%, and 28.99%, respectively. The combined RCV value was 94.6%, 38.31%, 41.57%, 42.35%, and 29.36% for CRP, ferritin, procalcitonin, troponin I, and BNP, respectively. The U of the CRP test was between 4.70%, and 5.49%. The lowest MU values of ferritin, troponin I, and BNP tests were 48.45%, 48.24%, and 53%, respectively. Conclusion: Providing the RCV and MU values with the test results in laboratory reports of these important biomarkers, which are frequently seen in high concentrations in COVID-19 patients, would offer valuable information for more accurate clinical interpretation.