Tiotropium in Chronic Obstructive Pulmonary Disease

To the Editor:

The Understanding Potential Long-Term Impacts on Function with Tiotropium (UPLIFT) study by Tashkin et al. (Oct. 9 issue)1 was expected to provide data on the effectiveness and safety of tiotropium in 5993 patients with chronic obstructive pulmonary disease (COPD). However, since the authors provide no information on the proportion of patients who started (or stopped) taking combined inhaled corticosteroids or long-acting beta agonists during follow-up, we do not know whether such a regimen may have confounded the study's findings. The benefit attributed to tiotropium might partly reflect concomitant use of other drugs. Also, it would be helpful if the authors could provide a case definition of moderate-to-severe renal failure so that clinicians can understand what types of patients were excluded from the study on this basis.

Claudio Pedone, M.D.
Raffaele Antonelli Incalzi, M.D.
Università Campus Biomedico, 00128 Rome, Italy
c.pedone@unicampus.it

To the Editor:

The UPLIFT study group reports that tiotropium was associated with improvements in lung function, quality of life, and COPD exacerbations during 4 years of follow-up, but the drug did not significantly reduce the rate of decline in the forced expiratory volume in 1 second (FEV1). Although the two study groups were balanced with respect to the criteria of the Global Initiative for Chronic Obstructive Lung Disease (GOLD), the percentage of patients with COPD who were in a formal pulmonary rehabilitation program was not specified. Since pulmonary rehabilitation can decrease the rate of COPD exacerbations, improve quality-of-life scores, and reduce the rate of hospitalization,1,2 an analysis of patients receiving this treatment should be reported.

En-Ting Chang, M.D.
Buddhist Tzu Chi General Hospital, Hualien 970, Taiwan
evan19760202@yahoo.com.tw

To the Editor:

Tashkin et al. report that “tiotropium was associated with improvements in lung function, quality of life, and exacerbations,” but the data they report do not clearly support these conclusions. Their study protocol called for the discontinuation of inhaled anticholinergic drugs, which were being used by more than 44% of patients at baseline. It is possible that discontinuation of anticholinergic therapy accounted for some or even all the differences between the placebo group and the tiotropium group. Beyond producing potentially misleading results, this trial design may not have adequately protected patients in the study from harm.1

To clarify these issues, the results of two subgroup analyses would be useful. First, patients who were not taking inhaled anticholinergic drugs at baseline should be compared with those who were receiving such therapy for all major outcomes. Second, the authors should demonstrate the safety of their study design by comparing the clinical outcomes of patients in the placebo group who discontinued anticholinergic therapy with those who did not discontinue such therapy.

Alec B. O'Connor, M.D., M.P.H.
University of Rochester School of Medicine and Dentistry, Rochester, NY 14642
alec_oconnor@urmc.rochester.edu

To the Editor:

The protocol for the UPLIFT study1 allowed for the use of any medications, including short-acting anticholinergic drugs, for the treatment of acute exacerbations of COPD, whereas the report by Tashkin et al. states that the use of such drugs was not permitted during the trial. It would be helpful if the authors could clarify this point.

Additional information about adverse cardiovascular events is also needed, since it is not clear why sudden death from cardiac causes, death from an unknown cause, and stroke were not considered to be deaths from cardiovascular causes.2 In keeping with the intention-to-treat approach, the authors should account for about 5% of the mortality data that appear to be missing from the 4-year data and 25% of such data missing from the 30-day post-treatment follow-up.

Another concern is that the increased rate of death from lung cancer in the tiotropium group was listed under “Respiratory (other)” in the Supplementary Appendix (as stated by the authors in an online forum),2 and data concerning lung cancer were excluded from the table listing adverse events in the article. It would be helpful if the authors could provide a clear delineation of lung-cancer deaths. Finally, since urinary retention is a common, often serious adverse event associated with the use of anticholinergic drugs,3 providing data on this outcome would be useful.

Sonal Singh, M.D., M.P.H.
Curt D. Furberg, M.D., Ph.D.
Wake Forest University School of Medicine, Winston-Salem, NC 27157
sosingh@wfubmc.edu

Yoon K. Loke, M.B., B.S., M.D.
University of East Anglia, Norwich NR4 7TJ, United Kingdom

To the Editor:

In the UPLIFT trial, there was a reduced incidence of cardiovascular events in the tiotropium group, as compared with the placebo group (3.56 vs. 4.21 per 100 patient-years; relative risk, 0.84; 95% confidence interval [CI], 0.73 to 0.98). In contrast, in a meta-analysis of five trials involving 7267 patients in which various anticholinergic agents were used and at least 6 months of observation were required, Singh et al.1 reported that there was an increased rate of death and complications from cardiovascular causes, as compared with placebo (2.9% vs. 1.8%; relative risk, 1.73; 95% CI, 1.27 to 2.36). The results of all five trials were consistent.

We find only two possible reasons for this discrepancy in findings. First, it is possible that the selection of studies for meta-analysis excluded trials in which cardiovascular events were not reported (two thirds of them), which could cause a selection bias, given that authors who didn't find any difference in cardiovascular events may have been less prone to report this outcome. Second, the length of observation in the UPLIFT trial was 4 years, whereas the meta-analysis involved studies with a shorter follow-up. The evidence of tiotropium's cardiovascular effects — whether beneficial or damaging — should be considered inconclusive until these inconsistencies are resolved.

Jesús F. Sierra-Sánchez, Pharm.D.
Emilio J. Alegre-del Rey, Pharm.D.
Isidoro Cobo, M.D.
Hospital Universitario Puerto Real, 11510 Puerto Real, Spain
que_que3@hotmail.com

Author/Editor Response

In response to the letter by Pedone and Incalzi: the use of pulmonary medication was balanced between the two study groups at all visits. The presence of moderate-to-severe renal failure at baseline was evaluated by investigators with respect to the patient's medical background, since there was no specific protocol definition.

In response to the letter by Chang: pulmonary rehabilitation during the study was infrequently reported and was balanced between the two study groups. In response to O'Connor: subgroup analyses according to the use of anticholinergic drugs at baseline showed that tiotropium was more effective than placebo, both in patients who received anticholinergic drugs and in those who did not receive such drugs before randomization.

In response to the comments by Singh et al.: the protocol states that patients were not permitted to use inhaled anticholinergic drugs as part of the study regimen but were permitted to use any therapy that was deemed to be medically necessary in case of a life-threatening exacerbation. Regarding vital-status information, we made rigorous attempts to obtain follow-up data, which included the use of national registries, resubmission to health authorities that initially denied permission, and repeated contact with sites. Since this collection was not part of the initial protocol, consent and ethics-committee approval were required. Some patients did not provide consent, some were lost to follow-up, and several regulatory authorities and ethics committees restricted access to information. Despite the challenges, we obtained survival data on 95% of patients for up to 4 years.

Regarding deaths from lung cancer, there were numerically more cases in the tiotropium group; however, the incidence rates with adjustment for exposure time were similar in the two study groups. Urinary retention occurred more often in the tiotropium group than in the placebo group (0.34 vs. 0.21 case per 100 patient-years at risk; rate ratio, 1.65; 95% CI, 0.92 to 2.93). However, there was only one serious adverse event of this type in each study group.

Both Singh et al. and Sierra-Sánchez et al. comment on adverse cardiovascular events that were reported in our study. We agree with Sierra-Sánchez et al. regarding significant limitations to the approach used by Singh et al., which also included incomplete case ascertainment and combining trials with different study groups, as well as an inability to adjust for differential exposure to study drugs.1 We used the Medical Dictionary for Drug Regulatory Affairs, which lists the terms sudden death, sudden cardiac death, cardiac death, and death (unknown cause) under the organ class of general disorders. The combination of the cardiac and vascular organ classes, the terms denoting stroke and myocardial infarction, and the aforementioned terms results in a reduction in the risk of any cardiovascular event (assuming that death of unknown cause was cardiovascular) in the tiotropium group (relative risk, 0.81; 95% CI, 0.68 to 0.97) and in the risk of fatal events (relative risk, 0.80; 95% CI, 0.64 to 1.02). A similar reduction in risk was observed in the tiotropium group when we examined all fatal cardiovascular events, including the vital-status follow-up.

Donald P. Tashkin, M.D.
University of California, Los Angeles, Los Angeles, CA 90095-1690
dtashkin@mednet.ucla.edu

for the UPLIFT Joint Advisory Committee

DOWNLOAD PDF FULL

SHARECASH (U.S recommended)
ZIDDU (worldwide)

Mirror 1
Mirror 2