|
Correction
to
Israel et al.,
|
| |||||||||||||||||||||||||||||||||||||||
Comparisons between patients with mild asthma and those with persistent asthma who are receiving high doses of inhaled glucocorticoids must include a careful evaluation of base-line characteristics.2 Table 2 of the article shows that the 28 women who did not use inhaled glucocorticoids weighed less than the 42 women who required more than eight puffs of inhaled glucocorticoids per day (mean [±SD], 140±20 vs. 154±40 lb), had nearly twice the level of physical activity (98±54 vs. 55±71 metabolic hours per week), had a lower incidence of past or current use of inhaled glucocorticoids (14±36 percent vs. 62±49 percent), and were less likely to have a history of oral-glucocorticoid use (36± 49 percent vs. 79±42 percent). All of these base-line differences appear to be statistically significant. It is as if we compared the bones of a busload of women soccer players with those of a busload of sedentary women.
A relative lack of gravitational exercise can obviously contribute to bone loss, as shown most clearly in astronauts returning from zero gravity. Because the presence of persistent asthma limits one's ability to exercise, the resulting inactivity and other changes in variables reflecting the severity of asthma (e.g., weight, prednisone use, and airway inflammation) invalidate any reliable analysis of the effects of inhaled glucocorticoids on bone loss in groups that were so dissimilar at base line in the absence of a randomized scheme of treatment allocation.
Edward Kerwin, M.D.
Clinical Research Institute
Medford, OR 97504
References
Israel et al. found that pulmonary function was similar among the three groups and inferred that there was no confounding related to differences in the severity of asthma. Since treatment was not randomly assigned, the high-dose group most likely had more severe asthma. Despite having similar pulmonary function, more patients in the high-dose group than in the other groups were excluded because they had received more than 30 days of oral or parenteral glucocorticoid therapy. Inhaled glucocorticoids can suppress the symptoms of bronchoconstriction, but they do not cure the disease. Their effects on the natural history of asthma are not clearly understood.2 Complications may thus occur independently of the level of bronchoconstriction.
The bone loss associated with the use of oral glucocorticoids is principally trabecular, with a greater loss in the lumbar spine and less of a loss in the proximal femur. The spine is associated with the largest increases in the risk of fracture.3 The pattern of effect on bone density at the spine and hip reported by Israel et al. does not support the hypothesis that inhaled glucocorticoids influence bone in a fashion similar to that of oral glucocorticoids.
We agree that patients using inhaled glucocorticoids have an increased risk of fracture. The potential role of asthma in increasing this risk should not be underestimated.
Tjeerd-Pieter van Staa, M.D., Ph.D.
University of Southampton
Southampton SO16 6YD, United Kingdom
Bert Leufkens, Ph.D.
Utrecht University
3508 TB Utrecht, the Netherlands
Cyrus Cooper, D.M.
University of Southampton
Southampton SO16 6YD, United Kingdom
cc{at}mrc.soton.ac.uk
References
That there is a normal decline in bone mineral density with age also calls into question the data from the study's bone densitometers. Data from the femoral neck and lumbar spine do not correspond to the expected base-line loss of 0.7 percent per year.1 Such measuring error calls into question the small changes in density that Israel et al. report as statistically significant. More analysis of the control group and more data are necessary to understand the consequences of this widespread treatment approach.
James L. Glazer, M.D.
Maine–Dartmouth Family Practice Residency
Augusta, ME 04330
jamesglazer{at}hotmail.com
References
To the Editor: In response to Dr. Kerwin: the "busloads" of women we compared were well matched. There were no statistically significant differences among the groups in weight or the level of physical activity. The apparent difference in the level of physical activity was due to a typographical error in Table 2. The mean level of physical activity in the group of women who did not take inhaled glucocorticoids was 48 metabolic hours per week, not 98. In addition, analyses that also adjusted for weight and level of physical activity did not affect our quantitative conclusions about the dose-related loss in bone density at the hip and trochanter.
Naturally, our groups differed with respect to the use of inhaled glucocorticoids. This was the independent variable used to assemble the groups. We also expected the incidence of a history of oral-glucocorticoid use before the study to differ among the groups. However, the data obtained during the study were not confounded by the use of oral glucocorticoids, which was prospectively monitored; we performed an a priori analysis that was restricted to patients who did not receive oral glucocorticoids during the study. Furthermore, data from van Staa et al.,1 among others, suggest that the presence of a history of glucocorticoid use before the study was unlikely to affect our outcome, since there is a rapid offset of the effects of oral glucocorticoids on bone density once therapy is stopped.
Since we did not examine any patients without asthma, we cannot confirm the observation of van Staa et al. regarding bronchodilator users and controls. However, when van Staa and colleagues compared users of high-dose inhaled glucocorticoids with those who used bronchodilators alone (an analysis similar to ours), their findings were remarkably similar to ours.2 They observed an increased rate of hip fracture with the use of high-dose inhaled glucocorticoids. The rate was not a function of the underlying population, since it declined toward base line once the treatment was discontinued. Furthermore, there was an increased rate of hip fracture and not of spinal fracture. Why inhaled glucocorticoids produce a pattern of accelerating bone loss that differs from that reported with oral glucocorticoids is unclear.
Dr. Glazer misunderstands our analysis. Patients who did not use inhaled glucocorticoids were very much part of the analysis (as indicated by the points superimposed on the ordinate in each panel of Figure 2 of our article). In fact, the yearly decline in bone density per puff of inhaled glucocorticoid that we report is the supplementary decline, which would occur in addition to any physiologic change in bone density that would be occurring in the group that was not using inhaled glucocorticoids. We used a very precise technique for measuring bone mass — dual x-ray absorptiometry — and the results were interpreted by one observer. However, as we noted in the article, on the basis of the results of dietary screening, patients received supplemental calcium, vitamin D, or both. This supplementation may have influenced the yearly rate of bone loss in our subjects, including the rate in the group that did not use inhaled glucocorticoids. Nonetheless, we found that inhaled glucocorticoids were associated with a dose-related decrease in bone density that was superimposed on any positive effect that may have resulted from dietary supplementation.
Elliot Israel, M.D.
Brigham and Women's Hospital
Boston, MA 02115
eisrael{at}partners.org
Garrett M. Fitzmaurice, Sc.D.
Harvard School of Public Health
Boston, MA 02115
Meryl S. LeBoff, M.D.
Brigham and Women's Hospital
Boston, MA 02115
References
| |||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
