High-sensitive detection and quantitation of thyroid-stimulating hormone (TSH) from capillary/fingerstick and venepuncture whole-blood using fluorescence-based rapid lateral flow immunoassay (LFIA)

Background: In the last decade, point of care testing (POCT) such as lateral flow immunoassays (LFIA) were developed for rapid TSH measurement. Most of these TSH-LFIAs are designed for qualitative measurements (i.e., if TSH values > 5, or >15 IU/L) and as screening tests for primary hypothyroidism in children and adults. Serum or plasma, but not venepuncture whole-blood or fingerstick/capillary, are usually used to quantify TSH accurately. Studies on performance evaluation of TSH-LFIAs POCT using venepuncture or fingerstick whole-blood are limited. Addi-tionally, limited studies evaluated the performance and validity of TSH-LFIAs POCT compared to valid and reliable reference methods. To our knowledge, this is the first study to evaluate three different blood withdrawal techniques for evaluating POCT of TSH.Aim: We aim to evaluate the performance of a new fluorescence-based LFIA and its FinecareTM fluorescent reader for quantitative measurement of TSH from a fingerstick, venepuncture whole-blood, and serum. Methods: 102 fingerstick, venepuncture whole-blood, and serum samples (with normal and abnormal TSH values) were analyzed by FinecareTM Rapid Quantitative LFIA test and Roche cobas (R) e 601 as a reference test.Results: Using serum, when compared to cobas (R) e 601 reference method, FinecareTM showed high sensitivity [90.5 % (69.6-98.8)] and specificity [96.3 % (89.6-99.2)] for diagnosis of thyroid abnormalities (<0.35 or >4.5 mIU/L). The actual test values (mIU/L) of FinecareTM showed excellent agreement (Cohen's Kappa = 0.85) and strong correlation (r = 0.93, p < 0.0001) with cobas (R) e 601. Using venepuncture whole-blood samples, FinecareTM showed similar results to serum with high sensitivity [95.2 % (76.2-99.9)], specificity [97.5 % (91.4-99.7)], excellent agreement (Cohen's Kappa = 0.91), and very strong correlation (r = 0.95, p < 0.0001) with cobas (R) e 601. These results suggest that FinecareTM can be used for quantitative measurement of TSH using serum or venepuncture whole-blood. These key performance indicators were slightly decreased when fingerstick whole-blood samples were used: sensitivity [85.7 %(63.7-97)], specificity [90.0 %,(81.5-96)], good agreement (Cohen's Kappa = 0.7) and very strong correlation (r = 0.9, p < 0.0001) with cobas (R) e 601. A subgroup analysis of abnormal TSH samples revealed a strong and significant correlation between the reference, FinecareTM whole-blood (r = 0.692; p = 0.0015), and fingerstick test FinecareTM (r = 0.66; p = 0.0025). A very strong correlation was also observed between cobas (R) e 601 serum and FinecareTM serum (r = 0.88; p < 0.0001). Conclusion: In comparison to the reference assay, our study demonstrates that FinecareTM exhibits high sensitivity, specificity, agreement, and a strong correlation. These findings provide evidence that FinecareTM is a reliable, valid, and accurate point-of-care test for TSH screening and quantitative measurement, especially in non-or small laboratory settings.