Improved performance of Xpert MTB/RIF assay on sputum sediment samples obtained from presumptive pulmonary tuberculosis cases at Kibong'oto infectious diseases hospital in Tanzania

BMC Infect Dis. 2017 Dec 29;17(1):808. doi: 10.1186/s12879-017-2931-6.

Abstract

Background: The introduction of Xpert MTB/RIF assay (Xpert) has significantly improved diagnosis of Tuberculosis (TB) in resource limited human immunodeficiency virus (HIV) endemic settings. We aimed to modify the Xpert protocol to improve the detection of Mycobacterium tuberculosis (MTB).

Methods: This cross sectional study was conducted among presumptive pulmonary tuberculosis (PTB) patients at Kibong'oto Infectious Diseases Hospital between August and November 2015. Each patient consented to provide 2 samples of raw sputa. One-sputum sample was sedimented using the Petroff's method and divided into two portions. One portion of sediment was inoculated on Lowenstein-Jensen culture media and observed for any growth for up to 8 weeks. Both, raw sputum and the portions of sediments were tested separately using Xpert with a sample reagent ratio of 1:2. Mean age of patients, prevalence of MTB, Xpert sensitivity, specificity, positive and negative predictive value were calculated. An incremental sensitivity was determined. Pearson chi-square and either an independent T or Mann-Whitney U-test were used to compared categorical and continuous variables respectively. A p- value of ≤0.05 was considered significant.

Results: Of the 270 presumptive PTB cases, 262 were eligible for analysis. Eight (3%) were excluded due to contaminated culture. Patients' mean age was 42.9 (±SD 15.1) years of which 173 (66%) were female. The overall prevalence of PTB was 112 (43%), of which the Xpert detected 105 (40%) in sediments and 98 (37%) in raw sputa as compared to culture which detected 85 (32%) cases of PTB. Sensitivity, specificity, positive and negative predictive values of Xpert on sputum sediments were 92%, 85%, 74% and 96% respectively. Overall, the incremental sensitivity of Xpert on sediment over raw sputum was 6%. In HIV infected Presumptive PTB, the incremental sensitivity was 12%.

Conclusion: Lowering the sample reagent to sediment dilution ratio increases sensitivity of Xpert on MTB detection among presumptive PTB cases, especially in HIV infected individuals.

Keywords: Kibong’oto; Presumptive-tuberculosis; Sputum-sediments; Tanzania; Tuberculosis; Xpert® MTB/RIF.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • AIDS-Related Opportunistic Infections / microbiology
  • Adult
  • Cross-Sectional Studies
  • Female
  • Humans
  • Male
  • Middle Aged
  • Molecular Diagnostic Techniques
  • Mycobacterium tuberculosis / genetics
  • Mycobacterium tuberculosis / pathogenicity
  • Prevalence
  • Sensitivity and Specificity
  • Sputum / microbiology*
  • Tanzania / epidemiology
  • Tuberculosis, Pulmonary / diagnosis*
  • Tuberculosis, Pulmonary / epidemiology