Reply to Lee and Newton

Clinical Infectious Diseases, Dec 2012

Jarvis, Joseph N., Bradley, John, Meintjes, Graeme, Harrison, Thomas S., Kozel, Thomas R.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

Reply to Lee and Newton

CID Joseph N. Jarvis 0 4 John Bradley 1 4 Graeme Meintjes 2 4 Thomas S. Harrison 0 4 Thomas R. Kozel 3 4 0 Research Centre for Infection and Immunity, Division of Clinical Sciences, St George's University of London 1 Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine , United Kingdom 2 Institute of Infectious Diseases and Molecular Medicine, Department of Medicine, University of Cape Town , South Africa 3 Department of Microbiology and Immunology, University of Nevada School of Medicine , Reno , USA 4 for Infection and Immunity, Division of Clinical Sciences, St George's University of London , Cranmer Terrace, London SW17 0RE , UK ( CORRESPONDENCE - Reply to Lee and Newton TO THE EDITOR?We thank Lee and Newton for taking an interest in our article evaluating a novel point-of-care cryptococcal antigen (CRAG) test [ 1 ]; however, we disagree with their critique of the statistical methods we chose to employ [ 2 ]. To be clear about our analytic objectives, the initial part of our analysis compared the levels of CRAG (measured with a sensitive sandwich enzyme-linked immunosorbent assay [ELISA]) in different biological samples?serum, plasma, and urine? using Spearman rank order correlation. Levels of CRAG were closely correlated and very similar in serum and plasma, and were >20-fold lower in urine, as stated in the Abstract and the Results section. The second part of the analysis was performed to examine the utility of the novel point-of-care CRAG lateral flow assay. The primary use of this test is to distinguish antigen-positive from antigen-negative samples in a dichotomous fashion. To further investigate the sensitivity of the novel assay and explore its usefulness in quantitating antigen load, we also assessed assay reactivity in serial dilutions of serum, plasma, and urine. For this to be a clinically relevant measure, one would want an increase in the dilutional titer to reflect an increase in the antigen concentration. To ascertain this, we compared the titer results obtained with our novel point-of-care CRAG test with results obtained from the highly sensitive sandwich ELISA. It is not important or clinically relevant for these 2 values to be numerically identical; indeed, it would be most unusual for this to be the case when comparing a highly sensitive and technical customized laboratory-based assay with a lateral flow dipstick assay. What is important is that high values of ELISA titer correspond to high values of lateral flow assay titer. Lee and Newton suggest that we should have used the Bland-Altman analysis [ 2, 3 ]. The Bland-Altman analysis is typically used when assessing a novel diagnostic against an established standard, as a means of assessing whether the 2 methods of measurement give the same numerical answer. In our analysis we were neither assessing the novel diagnostic against an established diagnostic test, nor expecting (or requiring) the tests to give an identical numerical answer. Thus we believe the BlandAltman analysis was not appropriate in this case. Our study has clearly demonstrated that the point-of-care lateral flow assay provides clinically relevant results, and, for the reasons given above, we believe that the statistical methods used in our paper are valid. It would be disappointing for a discussion of statistics to get in the way of a clear message that has the potential to greatly reduce the huge burden of cryptococcal-related mortality in the developing world. Note Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. 1. Jarvis JN , Percival A , Bauman S , et al. Evaluation of a novel point of care cryptococcal antigen (CRAG) test on serum, plasma and urine from patients with HIV-associated cryptococcal meningitis . Clin Infect Dis 2011 ; 53 : 1019 - 23 . 2. Lee S , Newton PN . Use of the correlation coefficient to compare a point-of-care antigen test against a quantitative sandwich enzymelinked immunosorbent assay for the detection of cryptococcal meningitis . Clin Infect Dis 2012 ; 55 : 1744 - 5 . 3. Bland JM , Altman DG . Statistical methods for assessing agreement between two methods of clinical measurement . Lancet 1986 ; 1 : 307 - 10 . Clinical Infectious Diseases 2012 ; 55 ( 12 ): 1745 - 6 ? The Author 2012. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals. DOI: 10 .1093/cid/cis781

This is a preview of a remote PDF:

Jarvis, Joseph N., Bradley, John, Meintjes, Graeme, Harrison, Thomas S., Kozel, Thomas R.. Reply to Lee and Newton, Clinical Infectious Diseases, 2012, 1745-1746, DOI: 10.1093/cid/cis781