Smoking is associated with a significantly increased risk of recurrence and mortality during treatment among patients with tuberculosis (TB), according to study findings published in Chest.

Evidence on mortality related to smoking among patients with TB has been inconclusive. Researchers therefore conducted a systematic review and meta-analysis to determine whether tobacco use is associated with risk of mortality and/or recurrence/relapse in patients with TB, reviewing literature in the MEDLINE, EMBASE, and CINAHL databases from inception to November 22, 2021. Epidemiologic studies that included patients with TB and measured the effect of ever, current, or past tobacco use on TB treatment outcomes for drug-susceptible and drug-resistant TB were eligible.

The primary outcomes were TB recurrence/relapse and mortality during treatment. The meta-analyses used the number of individuals exposed, unexposed, and events observed as reported in the individual studies to calculate risk ratios (RRs) and 95% confidence intervals (95% CI).

A total of 15 studies were included in the meta-analysis for TB recurrence/relapse and 13 for the meta-analysis on mortality during treatment. The international studies were published from 1998 to 2021.

[O]ur findings show increased risk of TB recurrence/relapse and mortality during treatment with smoking compared to never/nonsmoking.

For TB recurrence/relapse, 15 studies reported data to be pooled in at least 1 of the 3 meta-analyses: 5 for ever smoking, 13 for current smoking, and 3 for past smoking; notably, some studies reported on more than 1 exposure. The risk for TB recurrence/relapse was increased among those who had ever smoked (pooled RR, 1.78; 95% CI, 1.31-2.43, I2 = 85%), those who currently smoked (RR, 1.95; 95% CI, 1.59-2.40; I2 = 72%), and those who had previously smoked (RR, 1.84; 95% CI, 1.21-2.80; I2 = 74%), compared with those who had never smoked.

Regarding mortality during treatment, 4 of 13 studies reported estimates for ever smoking vs never smoking and 9 for current smoking vs nonsmoking. An increased risk for mortality during treatment was associated with ever smoking (RR, 1.55; 95% CI, 1.32-1.81; I2 = 0%) and current smoking (RR, 1.51; 95% CI, 1.09-2.10; I2 = 87%), compared with never/nonsmoking.

All-cause mortality was assessed in 11 studies and TB mortality in 11 studies. For all-cause mortality, excluding 2 studies that did not provide risk estimates and 2 that found no association with current smoking, the others reported an increased risk with smoking vs nonsmoking. Regarding TB mortality, all studies found a higher risk with smoking and 1 with smokeless tobacco use.

A total of 24 studies reported risk estimates for smoking and default (ie, loss to follow-up after the most recent course of TB treatment); all except 1 of these studies found an increased risk of default with smoking compared with never/nonsmoking.

The researchers noted that confounders such as age and alcohol use were not adequately accounted for, and that the method of data reporting for covariates did not allow for their use in meta-regression. In addition, high heterogeneity was observed, and the Grading of Recommendations, Assessment, Development and Evaluations assessments for all meta-analyses were either very low or low.

“Taken together, our findings show increased risk of TB recurrence/relapse and mortality during treatment with smoking compared to never/nonsmoking,” the study authors concluded. “Smoking is also a clear risk factor for other unfavorable TB treatment outcomes, as documented in earlier reviews.”

Source link