Smoking cessation is important in patients with tuberculosis because it can reduce the high rates of treatment failure and mortality. We aimed to assess the effectiveness and safety of cystine as a smoking cessation aid in patients with tuberculosis in Bangladesh and Pakistan.

We did a randomised, double-blind, placebo-controlled, trial at 32 health centres in Bangladesh and Pakistan. Eligible patients were adults (aged >18 years in Bangladesh; aged >15 years in Pakistan) with pulmonary tuberculosis diagnosed in the previous 4 weeks, who smoked tobacco on a daily basis and were willing to stop smoking. Patients were randomly assigned (1:1) to receive behavioural support plus either oral cytisine (9 mg on day 0, which was gradually reduced to 1·5 mg by day 25) or placebo for 25 days. Randomisation was done using pregenerated block randomisation lists, stratified by trial sites. Investigators, clinicians, and patients were masked to treatment allocation. The primary outcome was continuous abstinence at 6 months, defined as self-report (of not having used more than five cigarettes, bidis, a water pipe, or smokeless tobacco products since the quit date), confirmed biochemically by a breath carbon monoxide reading of less than 10 parts per million. Primary and safety analysis were done in the intention-to-treat population. This trial is registered with the International Standard Randomised Clinical Trial Registry, ISRCTN43811467, and enrolment is complete.

Between June 6, 2017, and April 30, 2018, 2472 patients (1527 patients from Bangladesh; 945 patients from Pakistan) were enrolled and randomly assigned to receive cytisine (n=1239) or placebo (n=1233). At 6 months, 401 (32·4%) participants in the cytisine group and 366 (29·7%) participants in the placebo group had achieved continuous abstinence (risk difference 2·68%, 95% CI −0·96 to 6·33; relative risk 1·09, 95% CI 0·97 to 1·23, p=0·114). 53 (4·3%) of 1239 participants in the cytisine group and 46 (3·7%) of 1233 participants in the placebo group reported serious adverse events (94 events in the cytisine group and 90 events in the placebo group), which included 91 deaths (49 in the cytisine group and 42 in the placebo group). None of the adverse events were attributed to the study medication.

Our findings do not support the addition of cytisine to brief behavioural support for the treatment of tobacco dependence in patients with tuberculosis.

European Union Horizon 2020 and Health Data Research UK.

For the Bengali and Urdu translations of the abstract see Supplementary Materials section.