Correspondence
Effect of Estrogen Therapy on Bone Density in Elderly Women
N Engl J Med 1994; 330:715-716March 10, 1994
- Article
To the Editor:
The study by Felson and colleagues (Oct. 14 issue)1 is liable to be interpreted as showing a decline in the protective effect of estrogen, perhaps related to “catch up” bone loss. One of the chief problems with this cross-sectional study is that it included an inappropriate control group. By using as controls women over the age of 75 years who had not taken estrogen, the authors selected women likely to have high bone density. There is 20 percent excess mortality among women with low bone density,2,3 even in the absence of fractures. Thus, at least one fifth of an appropriate control group will have died. Excess mortality also biases other case-control studies that show an apparent waning of the protective effect of estrogen therapy on the risk of hip fracture4. It is therefore to be expected that even if the effect of estrogen persists indefinitely, the more elderly the population, the smaller the effect of menopausal estrogen therapy.
4 ReferencesJohn A. Kanis, M.D.
University of Sheffield Medical School, Sheffield S10 2RX, United KingdomJohn C. Stevenson, M.B., B.S.
Wynn Institute for Metabolic Research, London NW8 9SQ, United Kingdom1
Felson DT ,Zhang Y ,Hannan M ,Kiel DP ,Wilson PWF ,Anderson JJ . The effect of postmenopausal estrogen therapy on bone density in elderly women. N Engl J Med 1993;329:1141-1146
Full Text | Web of Science | Medline2
Browner WS ,Seeley DG ,Vogt TM ,Cummings SR . Non-trauma mortality in elderly women with low bone mineral density. Lancet 1991;338:355-358
CrossRef | Web of Science | Medline3
Cooper C ,Atkinson EJ ,Jacobsen SJ ,O'Fallon WM ,Melton LJ III . Population-based study of survival after osteoporotic fractures. Am J Epidemiol 1993;137:1001-1005
Web of Science | Medline4
Kanis JA ,Johnell O ,Gullberg B , et al. Evidence for efficacy of drugs affecting bone metabolism in preventing hip fracture. BMJ 1992;305:1124-1128
CrossRef | Web of Science | Medline
To the Editor:
Felson et al. reported that bone density in women less than 75 years old who had taken estrogen for seven or more years was higher than in untreated women of comparable age. Bone density in women 75 years or older who had taken estrogen for seven or more years was no higher than in untreated women of comparable age. The authors suggest that any early benefit in the older women was outweighed by bone loss after the estrogen had been discontinued. An alternative interpretation of these results is that the women who were 75 or older already had low bone density when estrogen therapy was started. These women may have had lower bone density because they started estrogen therapy later after the menopause and so had lost more bone when therapy was initiated. It is not possible to examine this alternative interpretation from the results provided.
Bone loss accelerates for several years after the menopause, particularly after surgically induced menopause.1 What was the duration of menopause when estrogen therapy was started in the two age groups? If the women 75 or older and those under 75 are further divided into those who began estrogen therapy within six years after the menopause and those who began therapy more than six years after the menopause, what was the bone density in the four groups? In particular, did the women 75 or older who started treatment within six years after the menopause have higher bone density than the women of the same age who never took estrogen?
Bone loss resumes after estrogen is discontinued. If bone loss proceeds at the same rate as it does in the absence of treatment, there should be a residual benefit, even years later. If the authors are correct in assuming that the lack of residual benefit was the result of bone loss after estrogen had been discontinued (rather than before it was started), bone loss must have been more rapid than it would have been had no treatment been given. If so, what is the mechanism of this more rapid bone loss?
1 ReferencesEgo Seeman, M.D.
Austin Hospital, Heidelberg 3084, AustraliaAuthor/Editor Response
The authors reply:
To the Editor: Drs. Kanis and Stevenson are correct in asserting that women with low bone density have a moderately higher mortality, leaving fewer survivors with low bone density. The trouble with this argument is that the higher mortality among such women is not limited to those who do not take estrogen (the women Kanis and Stevenson call controls). The effect of low bone density on survival is comparable in women who take estrogen and in those who do not 1 and is probably related either to frailty or to some correlate of it. Assuming that the increased mortality due to low bone density is independent of treatment with estrogen, we performed simulations using earlier cohorts in which larger percentages of women with osteopenia were alive. We found that if all the women had lived to the age of 75 or older, the residual effect of estrogen treatment on bone density would have been the same or marginally smaller than the effect we observed.
Dr. Seeman is concerned that the residual effect of estrogen may be marginal in women 75 or older, because these women started estrogen long enough after the menopause that they had already lost considerable bone. Since bone loss is accelerated during the first few years after the menopause, we compared the bone density in the 14 women 75 or over who had started estrogen therapy within two years after the menopause and had taken it for seven or more years with the bone density in women of a similar age who had never taken estrogen. The results were similar to those we reported for all older women who had received long-term estrogen therapy. As compared with the bone density in the women who had never taken estrogen, the adjusted bone density in those treated soon after their menopause was the same (a difference of less than 1 percent) in the femoral neck, trochanter, and Ward's triangle; 4 percent higher in the ultradistal radius; 5 percent higher in the spine; and 10 percent higher in the shaft of the radius. The results were similar in older women who had started taking estrogen within six years after their menopause.
1 ReferencesDavid T. Felson, M.D., M.P.H.
Yuqing Zhang, D.Sc.
Boston University School of Medicine, Boston, MA 02118-23941
Browner WS ,Seeley DG ,Vogt TM ,Cummings SR . Non-trauma mortality in elderly women with low bone mineral density. Lancet 1991;338:355-358
CrossRef | Web of Science | Medline
- Citing Articles (3)
Citing Articles
1
M Zitzmann, M Brune, V Vieth, E Nieschlag. (2002) Monitoring bone density in hypogonadal men by quantitative phalangeal ultrasound. Bone 31:3, 422-429
CrossRef2
Peter J. Nigro, Brian L. Warrick. (1996) Referee's Ear. Journal of Occupational & Environmental Medicine 38:4, 329
CrossRef3
Silvano Adami, John A. Kanis. (1995) Perspectives assessment of involutional bone loss: Methodological and conceptual problems. Journal of Bone and Mineral Research 10:4, 511-517
CrossRef
