Prevalence of drug-resistant pulmonary tuberculosis in India: systematic review and meta-analysis

Abstract

Background: Drug-resistant pulmonary tuberculosis (DR-TB) is a significant public health issue that considerably deters the ongoing TB control efforts in India. The purpose of this review was to investigate the prevalence of DR-TB and understand the regional variation in resistance pattern across India from 1995 to 2015, based on a large body of published epidemiological studies.

Methods: A systematic review of published studies reporting prevalence of DR-TB from biomedical databases (PubMed and IndMed) was conducted. Meta-analysis was performed using random effects model and the pooled prevalence estimate (95% confidence interval [CI]) of DR-TB, multidrug resistant (MDR-) TB, pre-extensively drug-resistant (pre-XDR) TB and XDR-TB were calculated across two study periods (decade 1: 1995 to 2005; decade 2: 2006 to 2015), countrywide and in different regions. Heterogeneity in this meta-analysis was assessed using I² statistic.

Results: A total of 75 of 635 screened studies that fulfilled the inclusion criteria were selected. Over 40% of 45,076 isolates suspected for resistance to any first-line anti-TB drugs tested positive. Comparative analysis revealed a worsening trend in DR-TB between the two study decades (decade 1: 37.7% [95% CI = 29.0; 46.4], n = 25 vs decade 2: 46.1% [95% CI = 39.0; 53.2], n = 36). The pooled estimate of MDR-TB resistance was higher in previously treated patients (decade 1: 29.8% [95% CI = 20.7; 39.0], n = 13; decade 2: 35.8% [95% CI = 29.2; 42.4], n = 24) as compared with the newly diagnosed cases (decade 1: 4.1% [95% CI = 2.7; 5.6], n = 13; decade 2: 5.6% [95% CI = 3.8; 7.4], n = 17). Overall, studies from Western states of India reported highest prevalence of DR-TB (57.8% [95% CI = 37.4; 78.2], n = 6) and MDR-TB (39.9% [95% CI = 21.7; 58.0], n = 6) during decade 2. Prevalence of pre-XDR TB was 7.9% (95% CI = 4.4; 11.4, n = 5) with resistance to fluoroquinolone (66.3% [95% CI = 58.2; 74.4], n = 5) being the highest. The prevalence of XDR-TB was 1.9% (95% CI = 1.2; 2.6, n = 14) over the 20-year period.

Conclusion: The alarming increase in the trend of anti-TB drug resistance in India warrants the need for a structured nationwide surveillance to assist the National TB Control Program in strengthening treatment strategies for improved outcomes.

Keywords: Drug-resistant tuberculosis; India; Prevalence.

Fig. 1

Flow diagram for selection of studies

Fig. 2

Forest plot of prevalence of DR-TB and MDR-TB. (a) Decade 1995–2005 (DR-TB) (b) Decade 2006–2015 (DR-TB) (c) Decade 1995–2005 (MDR-TB) (d) Decade 2006–2015 (MDR-TB). Abbreviations: DR-TB, drug resistant tuberculosis; MDR-TB, multidrug resistant tuberculosis

Fig. 3

Subgroup analysis – prevalence of DR-TB and MDR-TB. (a) Decade 1995–2005 (Region-wise, DR-TB) (b) Decade 2006–2015 (Region-wise, DR-TB) (c) Decade 1995–2005 (Region-wise, MDR-TB) (d) Decade 2006–2015 (Region-wise, MDR-TB). Abbreviations: CI, confidence interval; DR-TB, drug resistant tuberculosis; ES, estimate; MDR-TB, multidrug resistant tuberculosis; n, number of studies. Notes: Negative I² was set to zero. Any missing data means that studies conducted in that region did not present results eligible for inclusion in this analysis

Fig. 4

Subgroup analysis- prevalence of MDR-TB among previously treated and newly diagnosed patients. (a) Decade: 1995 to 2005 (previously treated patients) (b) Decade: 2006 to 2015 (previously treated patients) (c) Decade: 1995 to 2005 (newly diagnosed patients) (d) Decade: 2006 to 2015 (newly diagnosed patients). Abbreviations: CI, confidence interval; ES, estimate; MDR-TB, multidrug resistant tuberculosis; n, number of studies Notes: Negative I² was set to zero. Any missing data means that studies conducted in that region did not present results eligible for inclusion in this analysis. Figure 4b and 4d: Countrywide prevalence includes 1 study from Central_East region (not presented individually)

Fig. 5

Subgroup analysis- Countrywide prevalence of Pre-XDR and XDR-TB. Abbreviations: CI, confidence interval; ES, estimate; FQ, Fluoroquinolone; Inj, aminoglycoside injectable; XDR-TB, extensively drug-resistant TB; n, number of studies. Notes: Negative I² was set to zero. Any missing data means that studies conducted in that region did not present results eligible for inclusion in this analysis

Fig. 6

Subgroup analysis- prevalence of mono-drug resistance. (a) Decade 1995-2005 (North India) (b) Decade 2006-2015 (North India) (c) Decade 1995-2005 (South India) (d) Decade 2006-2015 (South India) (e) Decade 1995-2005 (West India) (f) Decade 2006-2015 (West India) (g) Decade 1995-2005 (Central & East India) (h) Decade 2006-2015 (Central & East India) Abbreviations: CI, confidence interval; EMB, ethambutol; ES, estimate; INH, isoniazid; MDR-TB, multidrug resistant tuberculosis; n, number of studies; RMP, rifampicin; SM, streptomycin. Notes: Negative I² was set to zero. Any missing data means that studies conducted in that region did not present results eligible for inclusion in this analysis