Delivering the bronchodilator tiotropium with the Respimat inhaler did not increase the risk of death in chronic obstructive pulmonary disease, compared with the HandiHaler delivery system, according to a large study.
A randomized trial of more than 17,000 patients with COPD found that tiotropium Respimat 2.5 mcg and 5 mcg were both noninferior to tiotropium HandiHaler 18 mcg with respect to mortality, Dr. Robert A. Wise and his colleagues reported in the Sept. 8 issue of the New England Journal of Medicine and presented simultaneously at the European Respiratory Society Annual Congress (N. Eng. J. Med. 2013 [doi:10.1056/NEJMoa1303342]).
The 5-mcg Respimat dose also was as effective as the HandiHaler at preventing a first exacerbation of COPD, wrote Dr. Wise of Johns Hopkins Medical Center, Baltimore, and his colleagues.
Boehringer Ingelheim, which makes the Respimat inhaler, sponsored the study.
The 5-mcg Respimat dose and the 18-mcg HandiHaler dose have been shown to be pharmacokinetically equivalent. However, "concern about the safety of tiotropium Respimat was expressed when a post hoc pooled analysis of three 1-year trials and one 6-month placebo-controlled trial showed that [the 5-mcg dose] was associated with excess mortality in the planned treatment period (relative risk, 1.33), particularly among patients with known cardiac-rhythm disorders," Dr. Wise and the coauthors said. Subsequent meta-analyses appeared to confirm the finding.
The researchers designed the TIOSPIR (Tiotropium Safety and Performance in Respimat) trial to investigate the possible increased mortality risk in a large number of patients – many with preexisting cardiac disease. TIOSPIR included 1,825 patients with cardiac arrhythmias and 3,152 with ischemic heart disease, coronary artery disease, or heart failure.
The study comprised 17,135 patients with COPD, who were randomized to once-daily tiotropium at 2.5 mcg or 5 mcg delivered by Respimat, or 18 mcg delivered by HandiHaler. The study was designed to continue until at least 1,266 deaths had occurred. The mean follow-up was 2.3 years.
Average age of the patients was 65 years; 71% were male. About a third (38%) were current smokers. The mean forced expiratory volume in 1 second was 48% of predicted value.
Cardiac disease was not uncommon: 11% had a history of arrhythmia; 6% a prior heart attack; 2% a prior stroke; and 15% ischemic heart disease or coronary artery disease. Most (62%) were taking a long-acting beta2-agonist, and 68% used inhaled glucocorticoids.
The primary endpoint was the risk of death from any cause. Death occurred in 7.7% of the 2.5-mcg Respimat group, 7.4% of the 5-mcg Respimat group, and 7.7% of the HandiHaler group. There was no significant difference in the risk of death between the HandiHaler and either the 2.5-mcg or 5-mcg Respimat dose (hazard ratios, 1.00 and 0.96, respectively).
Cause of death was judged to be cardiovascular in 2% of cases from each group, and respiratory in about 2.5% of cases in each group.
"In particular, there was no increased risk of death among the 1,221 patients with a history of cardiac arrhythmia in the Respimat 5-mcg group as compared with the HandiHaler group (10.6% and 12.9%, respectively)," the authors noted, resulting in a nonsignificant hazard ratio of 0.81.
The risk of first COPD exacerbation was the secondary endpoint. In that measure, all of the treatments performed similarly. The hazard ratio for Respimat 5 mcg vs. HandiHaler was 0.98; exacerbations occurred in 48% of the Respimat 5-mcg group and 49% of the HandiHaler group, with median time to exacerbation of 756 and 719 days, respectively.
The investigators also performed a spirometry substudy. That showed that Respimat 5 mcg was noninferior to the HandiHaler treatment; the 2.5-mcg dose, however, was not as effective as the HandiHaler.
About a third of patients in each group experienced a serious adverse event. Most were respiratory in nature (about 17% in each group), followed by cardiovascular events (4% in each group).
Dr. Wise reported that he receives consulting fees from the company. Of the 11 coauthors, 6 are Boehringer Ingelheim employees and 4 reported consulting fees, research support, or other financial relationships with the company.