New England Journal of Medicine

Vol. 322 No. 12

March 22, 1990

Editorial: Fluoride and Bone - Quantity Versus Quality

POSTMENOPAUSAL osteoporosis is a major public health problem, whose impact is expected to reach epidemic proportions during the early part of the next century as the population ages. Consequently, whereas the prevention of bone loss will probably remain the most effective approach to therapy, efficacious treatment of patients who already had fractures is sorely needed. The ideal therapy would eliminate the risk of future fractures by restoring bone mass and repairing the alterations in architecture that had already occurred. Current treatments such as those using estrogen and calcitonin are antiresorptive - that is, they act by preventing further loss of bone - but treatment with neither of these agents results in much net gain of bone mass or repair of bone structure. Great hope, therefore, has been placed on sodium fluoride, an agent known to increase bone density, especially in the spine. (1) Unfortunately, the use of fluoride is often associated with side effects, including abnormalities of bone structure. Fluoride is incorporated into hydroxyapatite, altering the size and structure of the crystals and perhaps thereby decreasing the mechanical competence of the bone. When calcium intake is inadequate, the administration of fluoride also results in an impairment of mineralization. The bone formed in response to fluoride use may be somewhat disorganized, at least before remodeling, resembling immature woven bone rather than adult lamellar bone. Therefore, although the risk of fractures normally rises as the bone mass declines, increases in bone mass with fluoride treatment may not reduce fracture rates.

Until recently, there had been almost no well-controlled studies of the effects of fluoride in women with postmenopausal osteoporosis that used fracture frequency as an end point. In the early 1980s, with this in mind, the National Institutes of Health requested proposals for clinical studies of fluoride in the treatment of osteoporosis. The results of one such study appear in this issue of the Journal. (2) The randomized, placebo-controlled, double-blind study of Riggs et al. was designed to determine whether sodium fluoride given at doses of 75 mg per day (60 mg and 90 mg on alternate days) would alter bone mass and reduce the rate of fractures over a four-year period in women with postmenopausal osteoporosis. Although bone density in the lumbar spine increased by approximately 35 percent in the fluoride-treated women, the overall rate of vertebral fractures did not decrease significantly, implying that the new bone was structurally unsound. These women had a smaller change in bone density in the femoral neck and a worrisome but not statistically significant increase in the rate of hip fractures. There has been a disputed suggestion that fluoride treatment increases the risk of hip fracture, and a surprisingly high frequency of such fractures among treated patients was found in one study. (3) The increased number of peripheral fractures in the fluoride-treated group is equally disturbing. Although most were incomplete fractures, the relative risk for complete fractures was twice that in the placebo group. Besides the study's failure to establish the efficacy of fluoride against fractures, there was a high incidence of side effects, primarily gastrointestinal distress and a pain syndrome of the lower extremities. Eleven of the 19 patients with the latter problem had incomplete fractures at the site of pain, but the underlying cause of this syndrome in the others remains obscure. Unfortunately, the study did not include quantitative histologic data that might have aided the interpretation of the results, nor were the results compared in the women who had increased bone mass and those who did not.

The inescapable conclusion from this study is that sodium fluoride in the dosage used is not an effective or safe treatment for postmenopausal osteoporosis. The results were similar in a parallel but smaller study design carried out at the Henry Ford Hospital in Detroit. (4) Attempts to combine the data by using techniques of meta-analysis might be helpful in clarifying the findings that were not statistically significant in the two studies. Other studies have suggested that fluoride therapy is beneficial in postmenopausal women with osteoporosis, but a detailed review of the literature reveals mixed conclusions. Among recently published reports, a multicenter study conducted in France found no significant decrease in the occurrence of vertebral fractures in women treated with fluoride (50 mg per day as enteric-coated capsules) as compared with women treated with several standard regimens, but the probability of having a new fracture was significantly lower in the fluoride group according to a survival analysis. (5)

It is easy to find contentious issues in any clinical study of therapy for osteoporosis, and the several recent fluoride studies are no exception. The arguments will focus on the dosage, preparation, and duration of treatment. Other fluoride preparations, such as monofluorophosphate or slow-release capsules, or lower doses remain options for future investigation. However, it is important to note that the restoration of bone mass with fluoride, regardless of dose or preparation, cannot be taken as a surrogate measure for reduction of the fracture rate. Indeed, these studies raise the broader question of whether any therapeutic approach that increases bone mass in a skeleton already damaged by discontinuities in trabecular architecture can ever restore mechanical competence. (6) Further investigation of the basic effects of fluoride is clearly warranted, however, since fluoride remains the only agent that can induce the formation of new bone, apparently by acting as a specific mitogen for bone cells. (7) Finally, the issue of whether the addition of fluoride to more standard antiresorptive regimes is beneficial remains untested in prospective studies. Nevertheless, until studies designed as rigorously as that of Riggs et al. are performed with other preparations, formulations, or dosages, it is difficult to recommend the continued use of fluoride in clinical practice.

Thus, we do not yet have an ideal therapy for postmenopausal osteoporosis. The mainstays of such therapy - estrogen (with a progestin for women who have not had a hysterectomy) and calcitonin - are most effective if given early to prevent bone loss. (7,8) At that stage, estrogen has antifracture efficacy, the effect being most marked in women receiving long-term therapy. Recent evidence that validates the use of bone-mass measurements in predictions of the risk of fractures allows the physician to target treatment to postmenopausal women with low bone mass. (9) The wider use of estrogen will depend on other issues, such as its role in preventing ischemic heart disease or as a risk factor for breast cancer. (10) Estrogen treatment also appears to slow the rate of bone loss in more elderly women and those who have already had osteoporotic fractures. (7) Calcitonin must be given by subcutaneous injection, thereby limiting its usefulness, though intranasal administration may surmount this limitation if its efficacy is confirmed. (8) However, there have been no prospective, controlled studies of the recurrence of fractures in patients with osteoporosis who are given either estrogen or calcitonin. The impression that patients treated with estrogen sometimes feel better is related in part to the effect on other systems rather than to the inhibition of bone loss. The analgesic effect of calcitonin is also beneficial. In addition to the use of antiresorptive agents, attention must be given to the adequacy of calcium intake and rehabilitation for patients with fractures, including physical therapy and strategies to reduce falls.

Unfortunately, there are few alternatives to these two antiresorptive agents. The intermittent or continuous use of diphosphonates, which also have anti-resorptive properties, may be no better than the use of estrogen or calcitonin. One diphosphonate drug, etidronate, clearly represents an appealing treatment, because it is effective when given orally and has few if any extraskeletal effects. The manipulation of the remodeling cycle by the cyclic administration of agents that activate and then inhibit remodeling (coherence therapy) remains a hypothetical, unproved concept. Similarly, the value of parathyroid hormone or its analogues and 1,25-dihydroxyvitamin D, agents that may stimulate bone formation, has not yet been proved. Finally, the use of growth factors or other cytokines, even if feasible, remains years away.

The current studies emphasize that there are many areas in which our knowledge about postmenopausal osteoporosis is lacking. As a matter of priority, there is a clear need for targeted research in the areas of bone biology, age-related bone loss, and the epidemiologic and pathophysiologic features and treatment of postmenopausal and other forms of osteoporosis. The alternative is a not-too distant future in which the incidence of osteoporotic fractures will reach epidemic proportions, and costs may escalate beyond our capacity to pay

Robert Lindsay, M.B., CH.B., PH.D.
Helen Hayes Hospital
West Haverstraw, NY 10993


1. Riggs BL. Treatment of osteoporosis with sodium fluoride: an appraisal. In: Peck WA, ed. Bone and Mineral Research, Annual 2. New York: Elsevier, 1984:366-93.
2. Riggs BL, Hodgson SF, O'Fallon WM, et al. Effect of fluoride treatment on fracture rate in postmenopausal women with osteoporosis. N Eng J Med 1990; 322:802-9.
3. Hedlund LR, Gallagher JC. Increased incidence of hip fracture in osteoporotic women treated with sodium fluoride. J Bone Miner Res 1989; 4:223-5.
4. Kleerekoper M, Peterson E, Phillips E, Nelson D, Tilley B, Parfitt AM. Continuous sodium therapy does not reduce vertebral fracture rate in postmenopausal osteoporosis. J Bone Miner Res 1989; 4:Suppl 1:S376. abstract.
5. Mamelle N, Meunier PJ, Dusan R, et al. Risk-benefit ratio of sodium fluoride treatment in primary vertebral osteoporosis. Lancet 1988; 2:361-5.
6. Dempster DW, Shane E, Horbert W, Lindsay R. A simple method for correlative light and scanning electron microscopy of human iliac crest bone biopsies: qualitative observations in normal and osteoporotic subjects. J Bone Miner Res 1986; 1:15-21.
7. Lindsay R. Sex steroid in the pathogenesis and prevention of osteoporosis. In: Riggs BL, Melton LJ, eds. Osteoporosis: etiology, diagnosis, and management. New York: Raven Press, 1988:333-58.
8. Azria M. The calcitonins: physiology and pharmacology. Basel: Karger, 1989.
9. Johnston CC, Melton LJ, Lindsay R, Eddy DM. Clinical applications for bone mass measurements. J Bone Miner Res 1989; 4:Suppl 2:1-28.
10. Utian WH. Analysis of hormone replacement therapy. In: Studd JWW, Whitehead MI, eds. The menopause. Boston: Blackwell Scientific, 1988: 262-70.

Return to Fluoride Bibliography