FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |   Search Term  Advanced Search

Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe

Previous Volume 334:733-735 March 14, 1996 Number 11 Next

Add to Personal Archive Add to Citation Manager Notify a Friend E-mail When Cited Related Article by Liberman, U. A. Related Article by Liberman, U. A.

To the Editor: The study by Liberman et al. (Nov. 30 issue)¹ on the effects of alendronate on bone mineral density and the incidence of fractures in postmenopausal women with osteoporosis is important, but in our view not as exciting as it first seems. The authors pooled the results for women and, with respect to fractures, for doses of alendronate in two multicenter studies — an approach that at first seems valid. However, as shown in Table 2 of the article, the incidence of new vertebral fractures was lower in the U.S. study than in the multinational study, both in the placebo and the treated groups, suggesting that the women in the two studies may have differed. The authors state that the risk of new vertebral fracture was decreased in the alendronate-treated women, as compared with the placebo group, in both studies and in women stratified according to age or the presence or absence of a previous vertebral fracture. In fact, the incidence of fractures was reduced only in the elderly women, who made up a minority of the patients (relative risk, 0.3; 95 percent confidence interval, 0.1 to 0.8; P = 0.02). The difference in the incidence of vertebral fractures between the treated and placebo groups was not statistically significant in either the U.S. study or the multinational study; only by pooling the data was significance (P = 0.03) reached.

Hip fractures are the most important consequence of osteoporosis. It is noteworthy that there were only three hip fractures in the placebo group and one in the alendronate group during the three-year study. How do we prove that bisphosphonates are cost-effective drugs for the prevention of these fractures?

The evidence that bisphosphonates should be widely used in women with postmenopausal osteoporosis is meager, and we think it is too early for drug committees in local hospitals with shrinking budgets to recommend them, even if the drugs have been approved by regulatory authorities.

Hans Liedholm, M.D., Ph.D.
Agneta Björck Linné, M.Pharm.
Malmö University Hospital
S-205 02 Malmö, Sweden

References

1. Liberman UA, Weiss SR, Bröll J, et al. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. N Engl J Med 1995;333:1437-1443. [Free Full Text]

Hugh McGrath, Jr., M.D.
Louisiana State University Medical Center
New Orleans, LA 70112-2822

To the Editor: Liedholm and Linné question the validity of pooling the two studies for the analysis of vertebral fractures. We disagree. As clearly stated, the a priori data-analysis plan called for combining both studies and all alendronate-dose groups for analyses of fracture-related end points. This approach was valid because it was planned, because the design of the two studies was identical, and because the planned analysis was of relative (not absolute) risk stratified across studies, and there was no evidence that the relative risk differed between studies.

The women in each study and dose group did not differ in age, the number of years since menopause, base-line bone mineral density, or base-line prevalence of fractures. To use the small number of events in the subgroups defined according to study and age group (presented for completeness) to suggest that the groups differed seems an inappropriate conclusion drawn post hoc. Although elderly women had a greater absolute decrease in fractures, both age groups had similarly reduced relative risks of vertebral fractures. We believe that the finding that the number of vertebral fractures was consistently decreased with alendronate therapy in each subgroup supports the overall results.

These studies were too small to analyze the efficacy of alendronate on specific nonvertebral fractures. However, the reduction in fractures of the hip or pelvis in patients taking alendronate is encouraging. A recent pooled analysis of the nonvertebral-fracture results in these two studies combined with results of three similar studies demonstrated a 30 percent (P<0.05) reduction in nonvertebral fractures in women treated with alendronate.¹ Hip fracture occurred in 6 of 1012 women taking alendronate, as compared with 11 of 590 women receiving placebo.

Dr. McGrath expresses concern that the decrease in appendicular fractures was primarily due to a reduction in wrist fractures and that alendronate may not protect bones affected by gravity and stress. This concern seems unfounded, since there were reductions in the number of fractures of the hip, pelvis, and leg. Fractures of the ankle, foot, and toe were pooled for ease of presentation; the respective rates in the alendronate and placebo groups were 0.3 percent and 0.3 percent, 0.8 percent and 0.3 percent, and 1.3 percent and 1.0 percent.

Figure 3 of our article contains an error. The numbers of patients at risk at months 0, 12, and 24 and the survival curves are incorrect, although the values in the text are correct. Figure 1 below shows the correct numbers and survival curves; the curves differ little from the original. The discussion of the nonvertebral-fracture results in the text and Table 3 is accurate, as is the conclusion that treatment with alendronate resulted in a 21 percent reduction in the number of women who had a new nonvertebral fracture. We regret the error.

View larger version (3K): [in this window] [in a new window]   Figure 1. Cumulative Proportion of Women without Nonvertebral Fractures. This is the corrected version of Figure 3. The data on the women in three alendronate groups were pooled for this analysis, and the results shown were calculated by the life-table method.

 

Uri Liberman, M.D., Ph.D.
Beilinson Medical Center
49100 Petah-Tikva, Israel

David B. Karpf, M.D.
Thomas Capizzi, Ph.D.
Merck Research Laboratories
Rahway, NJ 07065-0914

References

1. Shapiro DR, Thompson DE, Karpf DB, et al. Effect of oral alendronate on nonvertebral fractures: a meta-analysis. J Bone Miner Res 1995;10:Suppl:S176-S176.abstract 

Add to Personal Archive Add to Citation Manager Notify a Friend E-mail When Cited Related Article by Liberman, U. A. Related Article by Liberman, U. A.

HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  TERMS OF USE  |  HELP  |  beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved.