Skip to main content

Table 1 Overview of the four diagnostic scenarios to diagnose drug-resistant tuberculosis that are compared in the model

From: The potential of a multiplex high-throughput molecular assay for early detection of first and second line tuberculosis drug resistance mutations to improve infection control and reduce costs: a decision analytical modeling study

   Base case Deployment of high-throughput MRD assay
Scenario   A. MRD assay following culture B. Improved analytical sensitivity C. Improved clinical accuracy
Resistance test for first-line drugs
 Assay Smear+ LiPA1 LiPA1 rapid MRD LiPA1
  Smear- Xpert Xpert rapid MRD Xpert
 Specimen   directly on sputum directly on sputum directly on sputum directly on sputum
 Accuracy: sensitivity; specificity rifampicin 99.0; 99.0 % as Base case as Base case 99.8; 99.8 %
isoniazid 96.0; 100 % [28] 99.2; 100 %
(Optimizeda)
Resistance test for second-line drugs
 Assay   DST on LJ rapid MRD rapid MRD rapid MRD
 Specimen   cultured isolate cultured isolate directly on sputum cultured isolate
 Accuracy: sensitivity; specificity fluoroquinolones 100 % (definition) 83.1; 97.7 % as A 96.6; 99.5 %
second-line injectable drugs 79.5; 95.8 % [9] 95.9; 99.2 %
(Optimizeda)
Treatment regimen individualization
 First-line regimen   Standard regimen Standard regimen Standard regimen Standard regimen
 Second-line regimen   Empirical at treatment initiation, individualized after DST result (2+ months) Empirical at treatment initiation, individualized after culture + MRD result (2+ weeks) Individualized from treatment initiation Empirical at treatment initiation, individualized after culture + MRD result (2+ weeks)
  1. Main assumptions: The sensitivity and specificity of molecular tests to detect Rifampicin and INH resistance are the same for all molecular tests (LiPA, MLPA, Xpert MTB/RIF) and are taken as the values of LiPA [29]
  2. MRD Molecular Resistance Detection, DR drug resistance, LiPA1 Line Probe Assay for first-line TB drugs, Xpert Xpert MTB/RIF assay
  3. 2+ =2 or more
  4. aOptimized assumes 80 % less false negatives and 80 % less false positives