© 2005 American Thoracic Society
Testosterone Supplementation during Respiratory RehabilitationTo the Editor:Casaburi and colleagues recently reported a randomized controlled trial exploring the effect of testosterone supplementation in men with COPD with or without additional resistance training during respiratory rehabilitation (1). While conducting a systematic review that included this question (2), we noted a few issues for which readers would require clarification.
A "screening" criterion was serum testosterone of Important baseline imbalances in FEV1, peak work rate, and constant work rate existed between patient groups. These problems are likely to originate from the small sample sizes that limit the likelihood of successful randomization. It is inappropriate to test baseline imbalances for statistical significance in a randomized controlled trial (3). Statistical adjustment for baseline imbalances in clinical trials is in order when randomization fails to achieve balanced groups (4). Because randomization did not result in similar distributions for important variables in this trial, the authors should have corrected statistically for these baseline imbalances. Casaburi and coworkers state they are the first to report on the combined effect of androgenic steroid supplementation and rehabilitation in patients with COPD. We would like to alert readers to the trial by Creutzberg and colleagues (5) that assessed the supplemental benefit of nandrolone during respiratory rehabilitation. The authors found no significant differences in improvement between groups in terms of muscle function, exercise capacity, and health status. The latter study included an outcome that is important to patients.
a Horten Centre, University of Zurich, Zurich, Switzerland FOOTNOTES Conflict of Interest Statement: M.A.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter; H.J.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this letter. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
400 ng/dl. However, the mean serum testosterone level was 408 ng/dl in patients randomized to testosterone + resistance training and between 277 and 302 ng/dl in the other three groups. In addition, standard deviations varied greatly between the four groups (from 89 to 154). It is also stated that 44% of all patients had levels above 400 ng/dl at baseline. Thus it seems that the "screening" and actual inclusion criteria were not identical. An explanation why this was not the case and what the actual inclusion criteria were could help to answer our concerns. Another explanation for differences between the testosterone levels at screening and baseline would be that testosterone measurements had poor reproducibility. In that case, repeated testosterone levels may have prevented enrolment of patients for whom this intervention was not intended. The high baseline levels in one group and the lower levels in the other groups also complicate drawing conclusions about patient profiles to which the results apply.