The ‘Bangladesh regimen’, which is a 9-month multidrug-resistant tuberculosis (MDR-TB) treatment regimen that contains high-dose (maximum 800 mg once daily) gatifloxacin, clofazimine, ethambutol and pyrazinamide given throughout supplemented by high-dose isoniazid, prothionamide and kanamycin in the initial 4–6 months, has demonstrated safety and efficacy among second-line treatment-naïve patients.1 Gatifloxacin has been withdrawn from the market owing to reports of associated dysglycaemia. A prospective observational study of a modified Bangladesh regimen comprising standard-dose (400 mg once daily) moxifloxacin instead of high-dose gatifloxacin has demonstrated high treatment success rates among second-line treatment-naïve MDR-TB patients.2 STREAM Stage 1, a randomized controlled trial, has evaluated a modified Bangladesh regimen that contains high-dose (maximum 800 mg once daily) moxifloxacin instead of high-dose gatifloxacin in comparison with the longer 20-month fluoroquinolone-based MDR-TB treatment regimen recommended by the World Health Organization. Preliminary findings of the STREAM Stage 1 trial suggested that both regimens were similar by efficacy and the rates of severe adverse events, but with a non-significantly higher frequency of new-onset QT prolongation on electrocardiography.3 With publication of the final results, some national MDR-TB programmes may consider using shorter MDR-TB treatment regimens that contain high-dose rather than standard-dose moxifloxacin. We submit here some views that might help optimize management of the adverse reactions to high-dose moxifloxacin under programmatic settings. Standard-dose moxifloxacin can incur a wide range of adverse reactions (see Table 1, based on https://www.drugs.com/sfx/moxifloxacin-side-effects.html). Other important toxicities pertain to fungal superinfections and drug–drug interactions. Intuitively, as an extensive spectrum of adverse reactions to standard-dose moxifloxacin has been reported among non-TB patients treated for 1–2 weeks, one can reasonably predict similar, if not more frequent or severe, reactions to high-dose moxifloxacin among TB patients treated for 9 months. The current focus of many clinical trials on serious adverse drug reactions may not adequately address treatment tolerance, which substantially affects treatment adherence and outcomes in programmatic settings. Thus, many adverse drug reactions, which are frequently dose-dependent and aggravated by co-morbidities (especially those with organ dysfunction and ageing4, 5), are often insufficiently evaluated in clinical trial settings but highly pertinent at both patient care and programme implementation levels when a very large patient population is involved. The need of clinical vigilance and diligent monitoring for all potential toxicities associated with high-dose moxifloxacin is imperative, especially for cardiotoxicity, which necessitates periodic electrocardiographic assessment throughout the treatment for all patients, with possibly 24-h Holter monitoring for selected cases to comprehensively detect potentially dangerous arrhythmia. More research is apparently required to further inform clinical decision for the optimal management of moxifloxacin-induced cardiotoxicity, and to unravel a possible association between cardiac arrhythmia and disturbance of cardiac metabolism in the context of drug-induced cardiotoxicity.6 The use of a shorter MDR-TB treatment regimen containing high-dose moxifloxacin has potential programmatic advantages, especially pertaining to anticipated adherence and curtailment of drug resistance, but intensified clinical and related monitoring possibly has resource implication that might offset cost-effectiveness in programmatic settings. W.W.Y. was consultant to Otsuka Pharmaceutical Company until July 2016, and member of the Independent Data and Safety Monitoring Committee of the STREAM trial until September 2018.