IRUS Total

Insights into the tuberculosis diagnostic pathway

File Description SizeFormat 
Martin-L-2014-MD(Res)-Thesis.pdfThesis11.85 MBAdobe PDFView/Open
Title: Insights into the tuberculosis diagnostic pathway
Authors: Martin, Laura
Item Type: Thesis or dissertation
Abstract: Background: Early identification of patients with drug-resistant tuberculosis (DR-TB) is important. Existing first and second-line drug (SLD) susceptibility testing (DST) strategies can be slow, unreliable and inaccessible. Methods and Findings: SLD DST on the MODS platform (MODS+): 97 drug-naïve and on-treatment patients provided sputum samples for analysis by MODS+ for SLDs. Agreement of MODS+ results with proportions method was >80% for fluoroquinolones and second-line injectables but suboptimal for ethambutol, pyrazinamide, streptomycin, cycloserine, ciprofloxacin and ethionamide. Performance of MODS kit for TB and multi-drug resistant TB (MDR-TB) diagnosis: Sputum samples (N=2446) were cultured in parallel by standard methods and MODS kit. Compared to conventional MODS, MODS kit sensitivity and specificity for TB detection was 99.3% (95%CI:98.3-99.8%) and 99.0% (95%CI:97.8-99.6%) respectively, for detection of rifampicin resistance 98.9% (95%CI:94.2-100.0%) and 99.0% (95%CI:97.8-99.6%) respectively and for detection of isoniazid resistance 96.0% (95%CI:90.8-98.7%) and 99.5% (95%CI:98.4-99.9%) respectively. Can changes in culture time to detection of TB (TTD-TB) predict MDR-TB? 95 patients provided sequential sputum samples before and during treatment for culture by MODS. TTD-TB increased during treatment but TTD-TB distribution at each time point did not differ between MDR and drug-susceptible strains (p=0.4, during eight weeks first-line treatment). Should TB DSTs be rationed to “high risk” TB suspects? Clinico-epidemiological data from TB suspects (N=1545) underwent multivariate analysis for DR-TB predictors. 17 selective testing strategies were based on significant predictors and international guidelines. The proportions of DR-TB patients captured by each strategy were calculated. 147 patients had culture-positive TB (DR-TB, n=31). If only those with a MDR-TB risk factor were tested 53·1% of DR-TB patients would not have undergone DST. Most strategies performed similarly poorly or tested 80% of the population to detect 90% of DR-TB patients. Conclusion: The MODS+ assay and MODS kit are promising diagnostic tools, likely to aid important universal access to early TB DST.
Content Version: Open Access
Issue Date: Feb-2014
Date Awarded: Jun-2014
URI: http://hdl.handle.net/10044/1/33125
DOI: https://doi.org/10.25560/33125
Supervisor: Friedland, Jon
Moore, David
Sponsor/Funder: PATH
Wellcome Trust (London, England)
Department: Department of Medicine
Publisher: Imperial College London
Qualification Level: Doctoral
Qualification Name: Doctor of Medicine (Research) MD (Res)
Appears in Collections:Medicine PhD theses

Unless otherwise indicated, items in Spiral are protected by copyright and are licensed under a Creative Commons Attribution NonCommercial NoDerivatives License.

Creative Commons