Call for worldwide withdrawal of tiotropium Respimat mist inhaler

Ten years ago serious concerns were raised about a link between inhaled anticholinergics and an increased risk of serious cardiovascular events and death in chronic obstructive pulmonary disease (COPD).1 Since then our understanding has grown about the effects of individual formulations of anticholinergic drugs and about their effects in patients with COPD and comorbidities. We believe that evidence of increased mortality from cardiovascular disease and all cause mortality with the tiotropium Respimat mist inhaler is now so strong that this inhaler should be withdrawn from the market.

In 2002, a randomised placebo controlled trial showed that regular use of ipratropium by metered dose inhaler increased the risk of death from cardiovascular disease (relative risk 2.57, 95% confidence interval 1.12 to 6.62).1 In 2008, a systematic review and meta-analysis of 17 randomised controlled trials of ipratropium and tiotropium (13 645 patients) reported an increased risk of the primary composite endpoint of cardiovascular death, myocardial infarction, or stroke (1.60, 1.22 to 2.10).2 It also reported increased risks of myocardial infarction (1.52, 1.04 to 2.22) and cardiovascular death (1.92, 1.23 to 3.00) but not stroke (1.46, 0.81 to 2.62).2 The relative risk for the primary composite endpoint was 1.70 (1.19 to 2.42) for ipratropium and 1.49 (0.98 to 2.26) for tiotropium.

Later in 2008, reassurance was provided with publication of the industry sponsored randomised placebo controlled UPLIFT (Understanding Potential Long-term Impacts on Function with Tiotropium) trial of 5993 patients. This trial found no increased risk of mortality (hazard ratio 0.89, 0.79 to 1.02) or myocardial infarction (relative risk 0.73, 0.53 to 1.00) with tiotropium (Spiriva Handihaler).3 However, patients with coexisting illnesses were excluded, such as those with moderate to severe renal impairment (which doubles plasma tiotropium concentrations4), a recent history of myocardial infarction, unstable or life threatening cardiac arrhythmia, or admission for heart failure.3 5 These exclusions limited the generalisability of the safety findings and of the overall results to unselected patients with COPD, many of whom also have cardiovascular or renal disease.6 The reduction in COPD exacerbations and modest improvement in lung function and quality of life contributed to the favourable risk-benefit profile of tiotropium reported by this study.3

However, in 2011 a systematic review and meta-analysis of randomised placebo controlled trials of tiotropium delivered by the new Respimat mist inhaler device reported significantly increased risks of all cause mortality (relative risk 1.52, 1.06 to 2.16) and cardiovascular death (2.05, 1.06 to 3.99).7 The findings raised the possibility of a dose-response effect on all cause mortality, with relative risks of 1.46 (1.01 to 2.10) and 2.15 (1.03 to 4.51) with the 5 µg and 10 µg preparations, respectively.7 It was not possible to determine whether the risk of death was associated with duration of treatment because there were too few studies. Importantly, in the largest Respimat study, risk of death from cardiac disease was most notably increased in patients with known cardiac disease or rhythm disorders (relative risks 4.03 (1.15 to 14.13) and 8.61 (1.10 to 67.2), respectively).4 5

In 2012, a Cochrane review of tiotropium versus placebo for COPD confirmed the significant increase in mortality for tiotropium Respimat (Peto odds ratio 1.47, 1.04 to 2.08).8 Subgroup analysis found a significant difference between the studies using the Handihaler and Respimat inhalers (test for subgroup differences P=0.01). This differential risk may result from greater systemic exposure with the Respimat device than with the Handihaler.9 10

Consistent with the Cochrane review,8 an independent systematic review and mixed treatment comparison meta-analysis of randomised controlled trials of drugs used in COPD confirmed that tiotropium Respimat increased the risk of death compared with placebo (odds ratio 1.51, 1.06 to 2.19) and with other inhaled drugs, including tiotropium Handihaler (1.65, 1.13 to 2.43), long acting β agonists (1.63, 1.10 to 2.44), and a combination of corticosteroids and β agonists (1.90, 1.28 to 2.86).11 The risk was greatest for death from cardiovascular disease, in patients with severe COPD, and at a higher daily dose. The generalisability of these findings is supported by a recent analysis from a Dutch general practice database, which showed that the use of tiotropium Respimat in clinical practice increased the risk of death (hazard ratio 1.52, 1.28 to 1.87), an association which remained after adjustment (1.33, 1.07 to 1.65).12

Clearly, more placebo controlled safety studies of anticholinergic inhalers in patients with comorbid conditions are needed. But, for now, the case for withdrawing tiotropium Respimat is compelling. We see no justification to expose patients to a drug for which one excess death can be expected for every 121 patients treated with the 5 µg dose for 12 months,13 when a preparation with similar efficacy and less harm is available. The proposed benefits of the Respimat device—that it is simple to use and that the delivered dose is independent of respiratory effort—are unlikely to be clinically relevant when the Respimat and Handihaler devices have similar efficacy.14 Although tiotropium Respimat is not approved for use in the United States, it remains available in the United Kingdom and at least 54 other countries. Warnings are not enough to protect patients; withdrawal would align all regulatory authorities with the current position of the US Food and Drug Administration. Thus we call for the immediate worldwide withdrawal of the tiotropium Respimat mist inhaler.