8 July 2011
A meta-analysis of randomised controlled trials found that compared with placebo, tiotropium mist inhaler (Spiriva Respimat▼) 5 micrograms/day (the recommended UK dose) was associated with a 46% relative increased risk of mortality (absolute risk increase [ARI] 0.8%) in people with COPD. The number needed to harm (NNH) in patients taking 5 micrograms/day for one year was 121 (95% confidence interval [CI] 51 to 5556) for one additional death to occur, although the CI was wide so there is considerable uncertainty around this estimate. Earlier evidence included in previous MHRA guidance found no such increased risk with the Spiriva Handihaler.
Level of evidence
Level 1 (good quality patient-oriented evidence) according to the SORT criteria.
Health professionals looking after people with COPD should continue to follow NICE guidance. A long-acting bronchodilator (either a long-acting anticholinergic [tiotropium] or a long-acting beta agonist [LABA]) should be offered for people who experience exacerbations or persistent breathlessness despite use of a short-acting bronchodilator. NICE does not give preference to either tiotropium or LABA.
When decisions are made around which long-acting bronchodilator to use, choice in individual patients should take account of their response to a therapeutic trial, potential side-effects, patient preference (e.g. suitability of different inhaler devices, individual tolerability), and cost. These new safety data on tiotropium Respimat should also feature in discussions with patients, although the evidence is uncertain. Health professionals should follow current MHRA advice on tiotropium Respimat. This reminds prescribers to use tiotropium Respimat with caution in patients with known cardiac rhythm disorders. For both tiotropium Respimat and HandiHaler, prescribers should not exceed the recommended once-daily dose (two puffs of 2.5 micrograms for Respimat, one 18 microgram capsule for HandiHaler). The MHRA continues to review the cardiovascular (CV) safety of all inhaled anticholinergics and any suspected adverse reactions to tiotropium Respimat (and HandiHaler) should be reported via the Yellow Card scheme.
What is the background to this?
Two formulations of tiotropium are available as the brand name Spiriva: a capsule containing 18 micrograms tiotropium delivered via a HandiHaler taken once daily; and a soft-mist Respimat inhaler that delivers 2∙5 micrograms tiotropium per actuation taken as two puffs once a day at the same time of the day.
The CV safety concerns associated with inhaled anticholinergics (ipratropium, and to a lesser extent tiotropium) have been discussed previously (see MeReC Rapid reviews 1214, 223 and 205). The evidence is conflicting, but the large UPLIFT study and a report from the tiotropium trial database were reassuring about the CV safety of tiotropium. The MHRA has advised that because of the uncertainty, it is difficult to draw firm conclusions on the risk associated with inhaled anticholinergics and further analyses are needed.
In the November 2010 edition of Drug Safety Update, the MHRA highlighted that recent analyses had found that tiotropium Respimat was associated with a non-significant increase in all-cause mortality, compared with placebo. In contrast, the HandiHaler device was associated with a decrease in all-cause mortality, compared with placebo. The underlying reasons for the apparent difference in the risk of all-cause mortality between devices are unclear, and may be due to chance. Tiotropium Respimat has not been approved by the FDA for use in the US, and its use in Scotland is restricted to patients who have poor manual dexterity and difficulty using the HandiHaler device.
In view of the safety concerns, the authors conducted a systematic review and meta-analysis of randomised controlled trials (RCTs) on the risk of mortality associated with tiotropium delivered by the mist inhaler (Spiriva Respimat), compared with placebo in people with COPD.
What does this study claim?
Overall, tiotropium Respimat (5 or 10 micrograms/day) was associated with a statistically significantly increased risk of mortality, compared with placebo (5 RCTs, 2.44% vs. 1.66%, relative risk [RR] 1.52, 95%CI 1.06 to 2.16, P=0.02, ARI 0.78%) over 3 to 12 months. There was also a statistically significantly increased risk of mortality in studies that compared only the 5 micrograms/day dose (the recommended UK dose) with placebo (5 RCTs, 2.43% vs. 1.66%, RR 1.46, 95%CI 1.01 to 2.10, P=0.04, ARI 0.77%), and when only longer-term (12 month) studies of 5 or 10 micrograms/day were considered (3 RCTs, 2.62% vs. 1.77%, RR 1.50, 95%CI 1.05 to 2.15, P=0.03, ARI 0.85%). In the small number of participants taking 10 micrograms/day of tiotropium Respimat, compared with placebo, the RR for the risk of mortality was 2.15 (95%CI 1.03 to 4.51, P=0.04).
This new meta-analysis adds to concerns highlighted by the MHRA about a possible safety signal of an increased risk of mortality with the tiotropium Respimat device. However, the authors and the MHRA have recognised that the increased death rate may be a chance finding. The meta-analysis was well conducted but there were some limitations, mainly relating to the quality of the data (e.g. different populations, different doses, low event rates). The trials were also not set up or powered to detect actual differences in CV events. However, all sensitivity analyses produced similar results and the risk of bias was considered low.
This new study does not answer the important question of whether tiotropium Respimat carries a particular risk compared with the HandiHaler device. A 2-year head to head study between tiotropium Respimat and HandiHaler is ongoing to help clarify the situation. It has been suggested that patients receiving Respimat may potentially be exposed to higher concentrations of tiotropium. This study also raises the possibility of a dose-response effect, although this is not definitive due to a lack of data. Furthermore, a once-daily dose of 10 micrograms/day is unlicensed and not usual recommended clinical practice. The MHRA advises that doses above 5 micrograms/day should not be used.
Although a relative increase in the risk of death of around 50% with tiotropium Respimat, compared with placebo appears alarming, as the accompanying editorial points out, this can be misleading and requires further interpretation. As the baseline risk of death was low, the 46% relative increased risk with the 5 micrograms/day dose represents an absolute increased risk of less than 1%, compared with placebo. More information on relative and absolute risk can be found in the NPC e-learning materials on evidence-informed decision making.
When explaining risks and benefits to patients in ways that are meaningful to them, visual images e.g. Cates plots may be helpful, as have been used in many NPC Patient Decision Aids (PDAs). See Figure 1 below for the Cates plot representing data from this meta-analysis. In trials of tiotropium Respimat 5 micrograms/day that lasted for one year, 18 deaths/1,000 patients/year (the red faces) occurred in patients treated with placebo, and an additional 8 deaths/1,000 patients/year (red faces with cross) occurred in those treated with tiotropium Respimat. This translates to an NNH of 121 (95%CI 51 to 5556) over one year — for every 121 patients given 5 micrograms/day of Respimat for a year, one person will die who wouldn’t have died had they all been given placebo. However, the confidence interval is very wide so there is considerable uncertainty around this estimate.
Figure 1. Cates plot for one-year mortality data for tiotropium Respimat 5 micrograms/day. From: Cates CJ. Editorial. Safety of tiotropium. BMJ 2011;342:d2970
It is likely that the choice of long-acting bronchodilator in individual patients with COPD will take account of their response to a therapeutic trial, potential side-effects, patient preference (e.g. suitability of different inhaler devices, individual tolerability), and cost. Tiotropium Respimat is more expensive (£440.08/year) than either tiotropium HandiHaler (refill £386.93/year) or any LABA alone preparations (range £144.08/year to £426.49/year). Prescribers should weigh these new safety data in discussions with patients, and if patients have a strong preference for the Respimat device, the Cates plot may help inform these discussions.
Singh S, et al. Mortality associated with tiotropium mist inhaler in patients with chronic obstructive pulmonary disease: systematic review and meta-analysis of randomised controlled trials. BMJ 2011;342:d3215
Systematic review and meta-analysis of 5 parallel group RCTs. Only studies providing numerical data on mortality were included. All trials were double blind with adequate allocation concealment. Evidence of statistical heterogeneity for mortality was lacking among studies.
6522 patients with COPD.
Intervention and comparison
Tiotropium Respimat (n=3,686) vs. placebo (n=2,836) for at least 30 days (majority of patients came from trials lasting 12 months). Almost 80% of patients receiving tiotropium were using the 5 microgram/day dose (n=2,839).
Primary outcome was mortality from any cause. Post-hoc secondary endpoint was deaths from CV causes (myocardial infarction, stroke, cardiac death and sudden death). Relative risks were estimated using fixed-effect meta-analysis.
See study details. Sensitivity analyses produced similar results.
For the secondary endpoint, risk of CV death — RR 2.05 (95%CI 1.06 to 3.99), P=0.03.
Further information can be found on NHS Evidence and in the NPC e-learning materials on COPD. Watch out for the new less than 60 minutes e-learning events on COPD on the NPC website.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you