Homocysteine as a Predictive Factor for Hip Fracture in Older Persons
http://www.100md.com
《新英格兰医药杂志》
To the Editor: In their study of homocysteine as a predictive factor for hip fracture in older persons, McLean et al. (May 13 issue)1 do not thoroughly discuss the role of physical activity, which may influence the association between homocysteine levels and the risk of fracture. Several previous studies and systematic reviews suggested that exercise increases bone mineral density and reduces or prevents bone loss in patients with osteoporosis.2,3 Physical activity is also effective in preventing falls.4 Plasma total homocysteine levels are inversely related to exercise, and this relation is not substantially changed by adjustment for intake of vitamin supplements, fruits, and vegetables.5 Thus, in addition to nutritional status, including folic acid, cobalamin, and pyridoxine, homocysteine levels are dependent on a person's level of physical activity. This observation implies that homocysteine may be an innocent bystander reflecting confounding variables that actually modify the risk of fracture.
Johann Auer, M.D.
Gudrun Lamm, M.D.
Bernd Eber, M.D.
General Hospital Wels
A-4600 Wels, Austria
johann.auer@klinikum-wels.at
References
McLean RR, Jacques PF, Selhub J, et al. Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med 2004;350:2042-2049.
Prince RL, Smith M, Dick IM, et al. Prevention of postmenopausal osteoporosis: a comparative study of exercise, calcium supplementation, and hormone-replacement therapy. N Engl J Med 1991;325:1189-1195.
Bonaiuti D, Shea B, Iovine R, et al. Exercise for preventing and treating osteoporosis in postmenopausal women. Cochrane Database Syst Rev 2002;3:CD000333-CD000333.
Gillespie LD, Gillespie WJ, Robertson MC, Lamb SE, Cumming RG, Rowe BH. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev 2001;3:CD000340-CD000340.
Nygard O, Vollset SE, Refsum H, et al. Total plasma homocysteine and cardiovascular risk profile: the Hordaland Homocysteine Study. JAMA 1995;274:1526-1533.
To the Editor: The articles by McLean et al. and van Meurs et al. (May 13 issue)1 and the accompanying editorial2 on homocysteine levels and the risk of osteoporotic fracture do not recognize the major effect of the age-related decline in the glomerular filtration rate on homocysteine levels.3 Although, in the study from the Netherlands, van Meurs et al. entered serum creatinine levels into their models, serum creatinine overestimates the glomerular filtration rate in elderly persons such as those they studied.4 Indeed, in the study from the Framingham Study group, which specified age in the homocysteine quartile, there was a progressive increase in age with each increasing quartile of homocysteine level, as shown in Table 2 of their article. Thus, the effect of age on the risk of fracture is what was actually measured. The lack of association between homocysteine levels and bone density (i.e., osteoporosis) may support such an interpretation.
Michael Bursztyn, M.D.
Hadassah University Hospital, Mount Scopus
Jerusalem 91240, Israel
bursz@cc.huji.ac.il
References
van Meurs JBJ, Dhonukshe-Rutten RAM, Pluijm SMF, et al. Homocysteine levels and the risk of osteoporotic fracture. N Engl J Med 2004;350:2033-2041.
Raisz LG. Homocysteine and osteoporotic fractures -- culprit or bystander? N Engl J Med 2004;350:2089-2090.
Friedman AN, Bostom AG, Selhub J, Levey AS, Rosenberg IH. The kidney and homocysteine metabolism. J Am Soc Nephrol 2001;12:2181-2189.
Bostom AG, Kronenberg F, Ritz E. Predictive performance of renal function equations for patients with chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol 2002;13:2140-2144.
To the Editor: McLean et al. suggest that the homocysteine level is an important risk factor for hip fracture in the elderly. A substantial percentage of older people have deficiencies of folic acid and vitamin B that are associated with the high prevalence of atrophic gastritis in this population.1 This constellation of findings is associated with hyperhomocysteinemia.
In a case–control study in which patients with hip fracture were compared with patients without hip fracture (mean
age, 83±5 vs. 82±10; homocysteine level, 12.5±5 vs. 15±5 μmol per liter, respectively), an inverse relation between homocysteine level and bone mass was observed in patients with hip fracture (r=0.475, P=0.02). Nevertheless, the significance disappeared with adjustment for age.2 Furthermore, no relation between bone mass and homocysteine level was observed in a younger population with diabetes mellitus type 2 whose intake of folic acid and vitamin B followed recommended daily allowances.3 Thus, we would submit that it is likely that homocysteine is not a risk factor independent of age.
José L. Pérez-Castrillón, M.D., Ph.D.
María L. Arranz-Pe?a, M.D., Ph.D.
Daniel D. Luis, M.D., Ph.D.
University Hospital Rio Hortega
47010 Valladolid, Spain
castrv@terra.es
References
Clarke R, Grimley Evans J, Schneede J, et al. Vitamin B12 and folate deficiency in later life. Age Ageing 2004;33:34-41.
Pérez-Castrillón JL, Velayos J, Arranz ML, et al. Masa ósea y homocisteina en pacientes con fractura de cadera. REEMO 2001;10:Suppl:S22.
De Luis Roman DA, Aller R, Pérez Castrillón JL, et al. The effects of dietary intake and life style on bone density in patients with diabetes mellitus type 2. Ann Nutr Metab 2004;48:141-145.
To the Editor: Raisz's thoughtful editorial on the effect of homocysteine level on osteoporotic fractures included the statement that randomized, placebo-controlled trials of nutritional supplements to reduce homocysteine levels in patients with osteoporosis would be unethical, because effective treatments for osteoporosis are available. But such trials in which all patients receive known effective treatment (e.g., a bisphosphonate plus calcium and vitamin D) and are randomly assigned to receive in addition either the supplement or a placebo should raise no ethical issues. Because effective treatments for osteoporosis reduce but do not eliminate the risk of fracture, such a trial could demonstrate whether the supplement would further reduce the risk of fracture.
It is a common misconception that placebo recipients in a randomized, placebo-controlled trial receive no treatment whatsoever. However, such trials are often used to evaluate "add-on" therapies — those intended to improve outcomes when added to known effective treatments. For example, new treatments for human immunodeficiency virus infection are evaluated in patients who are receiving standard antiviral regimens. Similar study designs are used to evaluate new treatments for cancer, acute myocardial infarction, heart failure, and other diseases.
Susan S. Ellenberg, Ph.D.
Food and Drug Administration
Rockville, MD 20852
David G. Orloff, M.D.
Robert J. Temple, M.D.
Food and Drug Administration
Rockville, MD 20857
Drs. McLean and Kiel reply: Although we agree with Pérez-Castrillón et al. and Bursztyn that age is an important potential confounder of the relation between plasma homocysteine and the risk of hip fracture, we contend that the results in our study cannot be explained by age alone. Although Bursztyn correctly points out that, in our study, mean age increased with increasing homocysteine quartiles, we did adjust for age in our regression models. Furthermore, in the same issue of the Journal, van Meurs et al. observed an almost identical effect in an analysis with the use of age-specific quartiles.
Bursztyn also points out that with increasing age there is a corresponding decline in the glomerular filtration rate that results in increased homocysteine concentrations. There is no evidence, however, that the age-related decline in kidney function is independently associated with the risk of hip fracture. Only among patients with end-stage renal disease who are receiving dialysis has an increased risk of hip fracture been observed.1 None of the participants in our study had kidney-function impairment of this magnitude.
Although we recognize the importance of physical activity as a determinant of the risk of hip fracture, its relation to the homocysteine concentration is not as clear. The results of population-based studies have shown both inverse2 and direct3 relations between level of physical activity and homocysteine concentration. Despite this conflicting evidence, physical activity may still have been a potential confounder in our study. Auer et al. wisely point out that the level of physical activity influences the risk of falling. As we stated in our article, homocysteine is associated with cardiovascular disease and cognitive dysfunction, conditions that contribute to increased frailty and, in turn, to the propensity to fall. Because it appears that homocysteine affects the risk of hip fracture independently of bone mineral density, falls due to increased frailty represent a possible alternative causal mechanism. In our article, we described an additional analysis in which we attempted to account for the effects of frailty by controlling for physical function. This additional adjustment did not, however, alter the effect of homocysteine on the risk of hip fracture.
Robert R. McLean, M.P.H.
Douglas P. Kiel, M.D.
Framingham Study
Framingham, MA 01702
rmclean@mail.hrca.harvard.edu
References
Alem AM, Sherrard DJ, Gillen DL, et al. Increased risk of hip fracture among patients with end-stage renal disease. Kidney Int 2000;58:396-399.
Nygard O, Vollset SE, Refsum H, et al. Total plasma homocysteine and cardiovascular risk profile: the Hordaland Homocysteine Study. JAMA 1995;274:1526-1533.
de Bree A, Verschuren WMM, Blom HJ, Kromhout D. Lifestyle factors and plasma homocysteine concentrations in a general population sample. Am J Epidemiol 2001;154:150-154.
Dr. van Meurs and colleagues reply: We acknowledge the fact that age-related decreases in the glomerular filtration rate can have an effect on homocysteine levels. We addressed this issue by including serum creatinine in our models. In addition, we performed a stringent adjustment for age by defining age-specific quartiles (as was not done in the study by McLean et al.) that resulted in homocysteine quartiles, each of which had a similar mean age and the same age distribution. Therefore, we can rule out the possibility that the association between homocysteine levels and fracture risk is simply an age-related effect. The fact that our analysis found that this association (with the clinical end point of osteoporosis) is independent of bone mineral density emphasizes the potential clinical importance of homocysteine as an independent, potentially modifiable risk factor for osteoporotic fracture.
Joyce B.J. van Meurs, Ph.D.
Huibert A.P. Pols, M.D., Ph.D.
André G. Uitterlinden, Ph.D.
Erasmus Medical Center
3000 DR Rotterdam, the Netherlands
j.vanmeurs@erasmusmc.nl
The editorialist replies: The suggestion from Dr. Ellenberg and colleagues, of the Food and Drug Administration, that the efficacy of lowering homocysteine levels to reduce the risk of osteoporotic fractures could be tested in an "add-on" study is certainly valid. Current therapies for osteoporosis generally result in a reduction of 30 to 60 percent in the incidence of fracture, so there is room for improvement. A test of the hypothesis that a high homocysteine level is a culprit and causes skeletal fragility might be carried out by adding folic acid to an antiresorptive regimen. Designing such a study would require an estimate of the effect size. If the effect were relatively small, then the number of patients required to demonstrate any additional reduction in fractures due to reduced homocysteine levels could be prohibitively large. It would also be necessary to restrict the study to subjects who not only have osteoporosis but also have high initial levels of homocysteine. This restriction would further complicate the design. Because of these concerns, I suggested that we try to obtain more data on the incidence of fracture from prospective studies of patients in whom homocysteine levels are altered. Were these studies to show an effect on the incidence of fracture, a prospective add-on trial would certainly be desirable.
Lawrence G. Raisz, M.D.
University of Connecticut Health Center
Farmington, CT 06032
raisz@nso.uchc.edu
Johann Auer, M.D.
Gudrun Lamm, M.D.
Bernd Eber, M.D.
General Hospital Wels
A-4600 Wels, Austria
johann.auer@klinikum-wels.at
References
McLean RR, Jacques PF, Selhub J, et al. Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med 2004;350:2042-2049.
Prince RL, Smith M, Dick IM, et al. Prevention of postmenopausal osteoporosis: a comparative study of exercise, calcium supplementation, and hormone-replacement therapy. N Engl J Med 1991;325:1189-1195.
Bonaiuti D, Shea B, Iovine R, et al. Exercise for preventing and treating osteoporosis in postmenopausal women. Cochrane Database Syst Rev 2002;3:CD000333-CD000333.
Gillespie LD, Gillespie WJ, Robertson MC, Lamb SE, Cumming RG, Rowe BH. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev 2001;3:CD000340-CD000340.
Nygard O, Vollset SE, Refsum H, et al. Total plasma homocysteine and cardiovascular risk profile: the Hordaland Homocysteine Study. JAMA 1995;274:1526-1533.
To the Editor: The articles by McLean et al. and van Meurs et al. (May 13 issue)1 and the accompanying editorial2 on homocysteine levels and the risk of osteoporotic fracture do not recognize the major effect of the age-related decline in the glomerular filtration rate on homocysteine levels.3 Although, in the study from the Netherlands, van Meurs et al. entered serum creatinine levels into their models, serum creatinine overestimates the glomerular filtration rate in elderly persons such as those they studied.4 Indeed, in the study from the Framingham Study group, which specified age in the homocysteine quartile, there was a progressive increase in age with each increasing quartile of homocysteine level, as shown in Table 2 of their article. Thus, the effect of age on the risk of fracture is what was actually measured. The lack of association between homocysteine levels and bone density (i.e., osteoporosis) may support such an interpretation.
Michael Bursztyn, M.D.
Hadassah University Hospital, Mount Scopus
Jerusalem 91240, Israel
bursz@cc.huji.ac.il
References
van Meurs JBJ, Dhonukshe-Rutten RAM, Pluijm SMF, et al. Homocysteine levels and the risk of osteoporotic fracture. N Engl J Med 2004;350:2033-2041.
Raisz LG. Homocysteine and osteoporotic fractures -- culprit or bystander? N Engl J Med 2004;350:2089-2090.
Friedman AN, Bostom AG, Selhub J, Levey AS, Rosenberg IH. The kidney and homocysteine metabolism. J Am Soc Nephrol 2001;12:2181-2189.
Bostom AG, Kronenberg F, Ritz E. Predictive performance of renal function equations for patients with chronic kidney disease and normal serum creatinine levels. J Am Soc Nephrol 2002;13:2140-2144.
To the Editor: McLean et al. suggest that the homocysteine level is an important risk factor for hip fracture in the elderly. A substantial percentage of older people have deficiencies of folic acid and vitamin B that are associated with the high prevalence of atrophic gastritis in this population.1 This constellation of findings is associated with hyperhomocysteinemia.
In a case–control study in which patients with hip fracture were compared with patients without hip fracture (mean
age, 83±5 vs. 82±10; homocysteine level, 12.5±5 vs. 15±5 μmol per liter, respectively), an inverse relation between homocysteine level and bone mass was observed in patients with hip fracture (r=0.475, P=0.02). Nevertheless, the significance disappeared with adjustment for age.2 Furthermore, no relation between bone mass and homocysteine level was observed in a younger population with diabetes mellitus type 2 whose intake of folic acid and vitamin B followed recommended daily allowances.3 Thus, we would submit that it is likely that homocysteine is not a risk factor independent of age.
José L. Pérez-Castrillón, M.D., Ph.D.
María L. Arranz-Pe?a, M.D., Ph.D.
Daniel D. Luis, M.D., Ph.D.
University Hospital Rio Hortega
47010 Valladolid, Spain
castrv@terra.es
References
Clarke R, Grimley Evans J, Schneede J, et al. Vitamin B12 and folate deficiency in later life. Age Ageing 2004;33:34-41.
Pérez-Castrillón JL, Velayos J, Arranz ML, et al. Masa ósea y homocisteina en pacientes con fractura de cadera. REEMO 2001;10:Suppl:S22.
De Luis Roman DA, Aller R, Pérez Castrillón JL, et al. The effects of dietary intake and life style on bone density in patients with diabetes mellitus type 2. Ann Nutr Metab 2004;48:141-145.
To the Editor: Raisz's thoughtful editorial on the effect of homocysteine level on osteoporotic fractures included the statement that randomized, placebo-controlled trials of nutritional supplements to reduce homocysteine levels in patients with osteoporosis would be unethical, because effective treatments for osteoporosis are available. But such trials in which all patients receive known effective treatment (e.g., a bisphosphonate plus calcium and vitamin D) and are randomly assigned to receive in addition either the supplement or a placebo should raise no ethical issues. Because effective treatments for osteoporosis reduce but do not eliminate the risk of fracture, such a trial could demonstrate whether the supplement would further reduce the risk of fracture.
It is a common misconception that placebo recipients in a randomized, placebo-controlled trial receive no treatment whatsoever. However, such trials are often used to evaluate "add-on" therapies — those intended to improve outcomes when added to known effective treatments. For example, new treatments for human immunodeficiency virus infection are evaluated in patients who are receiving standard antiviral regimens. Similar study designs are used to evaluate new treatments for cancer, acute myocardial infarction, heart failure, and other diseases.
Susan S. Ellenberg, Ph.D.
Food and Drug Administration
Rockville, MD 20852
David G. Orloff, M.D.
Robert J. Temple, M.D.
Food and Drug Administration
Rockville, MD 20857
Drs. McLean and Kiel reply: Although we agree with Pérez-Castrillón et al. and Bursztyn that age is an important potential confounder of the relation between plasma homocysteine and the risk of hip fracture, we contend that the results in our study cannot be explained by age alone. Although Bursztyn correctly points out that, in our study, mean age increased with increasing homocysteine quartiles, we did adjust for age in our regression models. Furthermore, in the same issue of the Journal, van Meurs et al. observed an almost identical effect in an analysis with the use of age-specific quartiles.
Bursztyn also points out that with increasing age there is a corresponding decline in the glomerular filtration rate that results in increased homocysteine concentrations. There is no evidence, however, that the age-related decline in kidney function is independently associated with the risk of hip fracture. Only among patients with end-stage renal disease who are receiving dialysis has an increased risk of hip fracture been observed.1 None of the participants in our study had kidney-function impairment of this magnitude.
Although we recognize the importance of physical activity as a determinant of the risk of hip fracture, its relation to the homocysteine concentration is not as clear. The results of population-based studies have shown both inverse2 and direct3 relations between level of physical activity and homocysteine concentration. Despite this conflicting evidence, physical activity may still have been a potential confounder in our study. Auer et al. wisely point out that the level of physical activity influences the risk of falling. As we stated in our article, homocysteine is associated with cardiovascular disease and cognitive dysfunction, conditions that contribute to increased frailty and, in turn, to the propensity to fall. Because it appears that homocysteine affects the risk of hip fracture independently of bone mineral density, falls due to increased frailty represent a possible alternative causal mechanism. In our article, we described an additional analysis in which we attempted to account for the effects of frailty by controlling for physical function. This additional adjustment did not, however, alter the effect of homocysteine on the risk of hip fracture.
Robert R. McLean, M.P.H.
Douglas P. Kiel, M.D.
Framingham Study
Framingham, MA 01702
rmclean@mail.hrca.harvard.edu
References
Alem AM, Sherrard DJ, Gillen DL, et al. Increased risk of hip fracture among patients with end-stage renal disease. Kidney Int 2000;58:396-399.
Nygard O, Vollset SE, Refsum H, et al. Total plasma homocysteine and cardiovascular risk profile: the Hordaland Homocysteine Study. JAMA 1995;274:1526-1533.
de Bree A, Verschuren WMM, Blom HJ, Kromhout D. Lifestyle factors and plasma homocysteine concentrations in a general population sample. Am J Epidemiol 2001;154:150-154.
Dr. van Meurs and colleagues reply: We acknowledge the fact that age-related decreases in the glomerular filtration rate can have an effect on homocysteine levels. We addressed this issue by including serum creatinine in our models. In addition, we performed a stringent adjustment for age by defining age-specific quartiles (as was not done in the study by McLean et al.) that resulted in homocysteine quartiles, each of which had a similar mean age and the same age distribution. Therefore, we can rule out the possibility that the association between homocysteine levels and fracture risk is simply an age-related effect. The fact that our analysis found that this association (with the clinical end point of osteoporosis) is independent of bone mineral density emphasizes the potential clinical importance of homocysteine as an independent, potentially modifiable risk factor for osteoporotic fracture.
Joyce B.J. van Meurs, Ph.D.
Huibert A.P. Pols, M.D., Ph.D.
André G. Uitterlinden, Ph.D.
Erasmus Medical Center
3000 DR Rotterdam, the Netherlands
j.vanmeurs@erasmusmc.nl
The editorialist replies: The suggestion from Dr. Ellenberg and colleagues, of the Food and Drug Administration, that the efficacy of lowering homocysteine levels to reduce the risk of osteoporotic fractures could be tested in an "add-on" study is certainly valid. Current therapies for osteoporosis generally result in a reduction of 30 to 60 percent in the incidence of fracture, so there is room for improvement. A test of the hypothesis that a high homocysteine level is a culprit and causes skeletal fragility might be carried out by adding folic acid to an antiresorptive regimen. Designing such a study would require an estimate of the effect size. If the effect were relatively small, then the number of patients required to demonstrate any additional reduction in fractures due to reduced homocysteine levels could be prohibitively large. It would also be necessary to restrict the study to subjects who not only have osteoporosis but also have high initial levels of homocysteine. This restriction would further complicate the design. Because of these concerns, I suggested that we try to obtain more data on the incidence of fracture from prospective studies of patients in whom homocysteine levels are altered. Were these studies to show an effect on the incidence of fracture, a prospective add-on trial would certainly be desirable.
Lawrence G. Raisz, M.D.
University of Connecticut Health Center
Farmington, CT 06032
raisz@nso.uchc.edu