Current issues of ACP Journal Club are published in Annals of Internal Medicine


Alendronate safely reduced fractures in postmenopausal women with low bone mass density

ACP J Club. 1997 May-Jun;126:73. doi:10.7326/ACPJC-1997-126-3-073

Source Citation

Black DM, Cummings SR, Karpf DB, et al., for the Fracture Intervention Trial Research Group. Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet. 1996 Dec 7;348:1535-41.



To determine whether alendronate would reduce the frequency of vertebral and nonvertebral fractures in postmenopausal women with low bone mass density and ≥ 1 vertebral fracture at baseline.


Randomized, double-blind, placebo-controlled trial (RCT) with 36-month follow-up.


United States.


2027 postmenopausal women (mean age 71 y, 97% white) aged 55 to 81 years who had femoral-neck bone mass density ≤ 0.68 g/cm2 were included. Exclusion criteria were peptic ulcer; dyspepsia requiring daily treatment; abnormal thyroid, parathyroid, or renal function; major medical problems; severe malabsorption syndrome; uncontrolled hypertension; unstable angina; recent myocardial infarction; or use of sodium fluoride, estrogen, calcitonin, or bis-phosphonates. Follow-up was 96%.


1022 women were assigned to alendronate (5 mg/d for 24 mo and then 10 mg/d), and 1005 women were assigned to placebo.

Main outcome measures

Change in bone mass density and incidence of new vertebral (clinical and morphometric), nonvertebral, hip, and wrist fractures.

Main results

Compared with women receiving placebo, women receiving alendronate had increased mean bone mass measured at the femoral neck (4.1% difference), total hip (4.7% difference), posterior-anterior lumbar spine (6.2% difference), whole body (1.8% difference), and proximal forearm (1.6% difference) (P < 0.001 for all comparisons). 78 women (8%) in the alendronate group compared with 145 women (15%) in the placebo group had ≥ 1 morphometric vertebral fracture (P < 0.001). {This absolute risk reduction (ARR) of 7% means that 14 patients would need to be treated (NNT) with alendronate (compared with placebo) for 3 years to prevent 1 additional new morphometric fracture (95% CI 10 to 23); the relative risk reduction (RRR) was 47%, CI 31% to 59%.}* Similar results were obtained for new clinical vertebral fractures (2% vs 5%, {P = 0.001; ARR 3%; NNT 35, CI 22 to 85; RRR 55%, CI 27% to 72%}*), any clinical fracture (14% vs 18% {P < 0.05; ARR 4%; NNT 22, CI 13 to 70; RRR 25%, CI 9% to 39%}*), hip fractures (1% vs 2%, {P < 0.05; ARR 1%; NNT 90, CI 43 to 8724; RRR 51%, CI 1% to 76%}*), and wrist fracture (2% vs 4%, {P < 0.05; ARR 2%; NNT 52, CI 28 to 234; RRR 47%, CI 13% to 68%}*). The groups did not differ for non-fracture-related adverse events.


Alendronate safely reduced the frequency of vertebral and nonvertebral fractures in postmenopausal women with low bone mass density and ≥ 1 vertebral fracture.

Source of funding: Merck Research Laboratories.

For article reprint: Dr. D.M. Black, Department of Epidemiology and Biostatistics, Box 0886, University of California-San Francisco, San Francisco, CA 94143, USA. FAX 415-597-9213.

*Numbers calculated from data in article.


The well-conducted RCT by Black and colleagues provides convincing evidence that the bisphosphonate alendronate reduces the risk for clinically important fractures in postmenopausal women with established osteoporosis. Similar high-quality evidence of effectiveness does not exist for any other osteoporosis therapy. Previous trials of osteoporosis treatments have all used changes in bone mass density or asymptomatic, radiographically detected vertebral fractures as the outcome variables. The study by Black and colleagues is the first RCT that has had sufficient power to use the patient-centered outcome of symptomatic fractures. It is hoped that this study heralds a new era in clinical research into the treatment of osteoporosis.

Estrogen is currently the mainstay of osteoporosis treatment, even though no large RCTs have shown that estrogen reduces the risk for symptomatic fractures. Should alendronate replace estrogen therapy in women with established osteoporosis? Head-to-head RCTs of these 2 drugs are needed to definitively answer this question. It should also be remembered that estrogen has cardiovascular, as well as skeletal, benefits (1). On the basis of current evidence, alendronate should certainly be carefully considered for women with established osteoporosis who are reluctant to take estrogen.

Clinical experience suggests that alendronate may cause upper gastrointestinal symptoms and, rarely, esophageal ulceration. It is reassuring that no association was found between these conditions and use of alendronate.

Many unanswered questions remain. The authors have initiated an RCT of alendronate in a large group of postmenopausal women with mild or moderate osteoporosis, and we will soon know the effectiveness of this treatment for these patients. Because bisphosphonates remain in bone for a long time, concern exists that use of these drugs for many years may induce a mineralization defect. However, not enough persons have been taking alendronate for an adequate period for us to know whether this is likely to be a real clinical problem.

Robert G. Cumming, MB, BS, MPH, PhD
University of SydneySydney, New South Wales, Australia

Robert G. Cumming, MB, BS, MPH, PhD
University of Sydney
Sydney, New South Wales, Australia


1. Grady D, Rubin SM, Petitti DB, et al. Ann Intern Med. 1992;117:1016-37.