Open AccessPredictive Accuracy and Densitometric Analysis of Point-of-Care Immunoassay for Adenoviral Conjunctivitis
Author(s) -
Spencer Johnson,
Jennifer Harthan,
Tammy Than,
Mary Migneco,
Ellen Shorter,
Meredith Whiteside,
Christina Morettin,
Christian K. Olson,
Crystal Rosemann,
Mathew Margolis,
Leonard W. Haertter,
Julia Huecker,
Bojana Rodic-Polic,
Richard S. Buller,
Gregory A. Storch,
Mae O. Gordon,
Andrew T. E. Hartwick
Publication year - 2021
Publication title -
translational vision science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.508
H-Index - 21
ISSN - 2164-2591
DOI - 10.1167/tvst.10.9.30
Subject(s) - medicine , immunoassay , densitometry , confidence interval , point of care testing , point of care , receiver operating characteristic , predictive value , predictive value of tests , titer , gastroenterology , pathology , nuclear medicine , immunology , antibody
Purpose Accurate diagnosis of adenoviral conjunctivitis (Ad-Cs) is important for timely and appropriate patient management to reduce disease transmission. This study assessed the diagnostic accuracy of a commercially available point-of-care adenovirus immunoassay and determined whether its predictive accuracy is influenced by signal intensities of test result bands. Methods Point-of-care immunoassay (AdenoPlus) testing and quantitative polymerase chain reaction (qPCR) testing was performed on conjunctival swab samples obtained from eyes of 186 eligible adult participants with presumed infectious conjunctivitis and symptoms of ≤4 days. Masked observers assessed signal intensities of the immunoassay test and control bands using densitometry. Results Ad-Cs was confirmed by qPCR in 28 of the 56 eyes that tested positive on the AdenoPlus, a 50% positive predictive value (95% confidence interval [CI] = 36.9, 63.1). No adenovirus was detected by qPCR in 128 of 130 eyes that tested negative on AdenoPlus, a 98.5% negative predictive value (CI = 96.3, 100). Sensitivity and specificity were 93% (CI = 84.4, 100) and 82% (CI = 76.0, 88.1), respectively. Viral titers significantly correlated with ratio of test band signal intensities (R 2 = 0.32, P = 0.002). Higher positive predictive value was associated with higher densitometry ratios (receiver operating characteristic [ROC] area = 0.71; 95% CI = 0.59, 0.83). Conclusions Densitometric analyses suggest that the diagnostic accuracy of AdenoPlus is influenced by the signal intensity of the test result bands. Visual comparison of the test band intensities by clinicians could reduce the false positive rate of point-of-care immunoassays and aid in the diagnosis of viral infections. Translational Relevance Ratiometric densitometry of point-of-care immunoassays could aid clinicians’ decision making in diagnosing infectious diseases, including Ad-Cs.