| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Bone and Mineral Research Unit, Department of Medicine (S.A.W., L.M.M., E.S.O) and Department of Public Health and Preventive Medicine (L.M.M.), Oregon Health and Science University, Portland, Oregon 97239; Department of Medicine and Division of Epidemiology, University of Minnesota (K.E.E.), Minneapolis, Minnesota 55454; Section of General Internal Medicine, Minneapolis Veterans Affairs Medical Center (K.E.E.), Minneapolis, Minnesota 55417; Departments of Epidemiology (J.A.C.) and Orthopedic Surgery (M.T.V.), University of Pittsburgh, Pittsburgh, Pennsylvania 15260; Department of Epidemiology and Biostatistics; University of California at San Francisco (D.M.B.), San Francisco, California 94105; Kaiser Permanente Center for Health Research Northwest/Hawaii Division (T.A.H.), Portland, Oregon 97227; and Departments of Medicine and Epidemiology and Preventive Medicine, University of Maryland School of Medicine, and Medical Service, Maryland VA Health Care System (M.C.H.), Baltimore, Maryland 21201
Address all correspondence and requests for reprints to: Stacey Wainwright, M.D., Oregon Health and Science University, 3181 S.W. Sam Jackson Park Road, Mail Code CR113, Portland, Oregon 97239. E-mail: s_wainwright{at}comcast.net.
 |
Abstract |
|---|
 Top Abstract IntroductionParticipants and MethodsResultsAppendixReferences |
|---|
Among all participants, 17% had osteoporosis (total hip BMD T-score 
–2.5). Of the 243 women with incident hip fracture, 54% were not osteoporotic at start of follow-up. Nonosteoporotic women who fractured were less likely than osteoporotic women with fracture to have baseline characteristics associated with frailty. Nevertheless, among nonosteoporotic participants, several characteristics increased fracture risk, including advancing age, lack of exercise in the last year, reduced visual contrast sensitivity, falls in the last year, prevalent vertebral fracture, and lower total hip BMD.
These findings call attention to the many older women who suffer hip fracture but do not have particularly low antecedent BMD measures and help begin to identify risk factors associated with higher bone density levels.
|
Introduction |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsAppendixReferences |
|---|
–2.5 (4)] in fracture prevention efforts. However, a growing number of reports suggest that many women who fracture have BMD higher than that usually associated with osteoporosis (5, 6, 7, 8, 9, 10, 11). Better understanding the characteristics of these women with fracture could lead to improved interventions to reduce fracture risk. Recently Miller et al. (10) described an approach to identifying the risk of fracture in postmenopausal women with peripheral BMD T-scores –2.5 to –1.0, and Robbins et al. (12) described hip fracture risk factors for women with high hip BMD among women older than 74 yr. In this investigation, we focused on the hypotheses that women without very low hip BMD who subsequently suffer hip fracture might be especially frail, physically active, or may have a genetic predisposition to fracture, putting them at increased risk for fracture.
In the Study of Osteoporotic Fractures (SOF) (13), a large cohort of postmenopausal women, we identified incident hip fracture cases during a period of up to 5 yr of follow-up and determined which proportion of these fractures occurred in women with total hip BMD at the start of observation above that usually associated with hip osteoporosis. Then, to learn more about the characteristics of these nonosteoporotic women with fracture, we compared baseline participant characteristics and potential risk factors for hip fracture among hip fracture cases with and without hip osteoporosis, as well as identified hip fracture risk factors for women without hip osteoporosis among potential risk factors previously described in this population.
|
Participants and Methods |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsAppendixReferences |
|---|
Participants in SOF were women aged 65 yr or older recruited from population-based listings and health maintenance membership lists at four sites in the United States: Baltimore, MD; Minneapolis, MN; the Monongahela Valley near Pittsburgh, PA; and Portland, OR. Black women were excluded due to their lower incidence of hip fracture, as were women who were unable to walk without the assistance of another person, had bilateral hip replacements, or were institutionalized. Written informed consent was obtained from all participants after the appropriate institutional review boards approved the study protocol.
Between October 1986 and October 1988, 9704 women attended the first study examination. Examinations were conducted approximately every 2 yr. The second study examination was attended by 9339 women (98% of survivors) from January 1989 through December 1990 and was the first study examination at which hip BMD was measured. Follow-up for hip fracture began at this second examination among the 8065 participants who had adequate total hip BMD measurements. Circumstances resulting in inadequate BMD data were varied; primarily these women completed the study examination questionnaire only.
Ascertainment of fracture cases
Participants were contacted every 4 months by telephone or mail to determine whether any fractures had occurred in the preceding 4-month period. In addition, participants were asked to notify the clinical center as soon as possible after any fracture. All hip fractures were radiographically confirmed. Fractures due to severe trauma (mainly as a result of motor vehicle accident) were excluded. After 5 yr, hip fracture follow-up remained over 98% complete for surviving participants. Details of fracture ascertainment methods have been published (14). Cases for this study were women who experienced a new hip fracture during the 5 yr of follow-up after total hip BMD measurement at the second SOF study examination.
Because a disproportionate loss to follow-up, particularly loss to death, might bias the proportion of hip fracture cases with and without osteoporosis as a competing risk for fracture, we assessed the number of confirmed deaths during follow-up among participants at risk for fracture. Deaths were documented by review of official death certificates and hospital records, if available. In 5 yr of follow-up, there were 191 confirmed deaths (14%) among women with osteoporosis and 511 (8%) among women without osteoporosis at the start of follow-up. Among participants who died during the observation period and had not experienced a hip fracture, 173 had osteoporosis, whereas 490 did not have osteoporosis. The proportion that terminated their participation in the study was approximately 1% and did not differ among those with and without osteoporosis.
Definition of osteoporosis
Total hip BMD was quantified at the second SOF examination using dual-energy x-ray absorptiometry (QDR-1000, Hologic, Inc., Bedford, MA). Interscanner precision was good for measurement of block phantom (coefficient of variation 0.25–0.77%) and anthropomorphic femoral neck phantom (coefficient of variation 0.93%) (15). Details of bone density measurement methods have been published (14, 15, 16).
To categorize total hip BMD in familiar terms, we used the World Health Organization BMD-based osteoporosis classification (4) to define osteoporosis. Total hip BMD T-scores were calculated using the third National Health and Nutrition Examination Survey Caucasian female mean BMD aged 20–29 yr as reference peak BMD (17) [T-score = (BMD–peak BMD)/peak BMD SD]. We then divided participants by BMD T-score into two groups, one with osteoporosis (total hip BMD T-score –2.5) and one without osteoporosis (total hip BMD T-score > –2.5). In additional analyses, we identified participants with total hip BMD T-score greater than –2.0 and greater than –1.0 because these are cut points referred to in some diagnostic and treatment guidelines.
We focus on total hip BMD measurements in these investigations because BMD measurement at this site has been described as the best predictor of all types of hip fracture, is associated with low precision error, and represents an assessment of both cortical and trabecular bone (18, 19). However, to determine which proportion of hip fracture cases had osteoporosis at the femoral neck or lumbar spine, we also measured BMD at these sites using dual-energy x-ray absorptiometry at the second study examination. Femoral neck BMD T-scores were calculated as described for the total hip BMD T-scores. Lumbar BMD T-scores were calculated using the manufacturer’s Caucasian female reference mean for age 25 yr as peak BMD. For both the femoral neck and lumbar spine measurements, osteoporosis was defined as BMD T-score –2.5 or less.
Other measurements
We assessed baseline demographics, potential risk factors for hip fracture previously described in the SOF cohort (7, 20, 21) and several measures of strength and fall propensity. Participants completed a questionnaire and were interviewed for self-assessment of health, exercise, physical activity, falls, medical history, habits, and medication use. Current and past activity levels were assessed using a modified Paffenbarger survey (22, 23). Weekly caloric expenditure was determined by converting values for type, frequency, and duration of weight-bearing activities for the preceding 12 months. Intensity-weighted lifetime activity was derived from designating activities as low intensity (walking or gardening), medium intensity (dancing or tennis), or high intensity (jogging or skiing) and multiplying the reported frequency of the activity by 2.5, 5.0, and 7.5, respectively, for several time periods (past week, past 12 months, at about 50 yr of age, at about 30 yr of age, and as a teenager). Caffeine intake was tabulated assuming caffeine content of 95 mg per cup of coffee, 55 mg per cup of tea, and 45 mg per cola drink. Weight was measured using a balance beam scale. Height was assessed using a standard held-expiration technique with a wall-mounted Harpenden stadiometer. Resting heart rate was measured in the supine position. Neuromuscular function was tested by determining whether participants could rise from a chair five times without using their arms for support. Grip strength was measured using an adjustable handgrip dynamometer. Knee extension strength was tested using a hand-held isometric dynamometer. Distance depth perception was measured using a Howard-Dolman apparatus and reported as the SD of four trials (20, 24). Contrast sensitivity was measured using a VCTS 6500 wall chart and light meter (Vistech Consultants, Inc., Dayton, OH) and the average score calculated separately for high and low spatial frequencies (25). Mental status was assessed using a modified version of the Mini-Mental State Examination with a maximum score of 26 (26, 27). Prevalent vertebral fracture was defined using radiographs, as lateral, cross-sectional height of a measured thoracic or lumbar vertebra exceeding 3 SD below the mean at that vertebra for normal women (21).
Because many of the potential characteristics of interest were measured at only the first study examination, baseline information for all case characteristics and risk factors was collected from this examination, with the exception of age, which was updated at the time of hip BMD measurement during the second study examination.
Main risk factors for hip fracture were derived from 16 previously described BMD-independent risk factors for hip fracture in the SOF cohort (20, 21): a low self-rated health score; no walking for exercise; being on one’s feet no more than 4 h each day; previous hyperthyroidism; previous fracture after age 50 yr; maternal history of hip fracture; caffeine intake of greater than 190 mg each day; current use of long-acting benzodiazepines; current weight less than that at age 25 yr; height at age 25 yr at least 168 cm; inability to rise from chair without the use of one’s arms; being in the lowest quartile of the cohort for distance depth perception or for low-frequency contrast sensitivity; resting pulse rate greater than 80 beats/min; being at least 80 yr of age; and prevalent vertebral fracture.
Statistical analysis
We first conducted a case-case analysis restricted to hip fracture cases grouped by osteoporosis status at the start of observation, comparing baseline characteristics and hip fracture risk factors. To examine these relationships in a different way, we made similar comparisons after dividing cases by total hip BMD tertile and quartile at the start of observation. The resulting inferences were similar to those for comparisons between cases grouped by hip osteoporosis classification. Therefore, for the remainder of this report, we refer to the two fracture case groups defined as having osteoporosis or not at the start of observation.
To control for a number of potential confounders in this case-case analysis, we additionally used multivariable logistic regression to quantify the association between baseline characteristics and risk factors and prevalence of total hip BMD T-score greater than –2.5 (no osteoporosis), compared with prevalence of total hip BMD T-score –2.5 or less (osteoporosis) at the start of observation. Here the odds ratios (ORs) from logistic regression estimate the magnitude of the difference in the frequency of fracture risk factors and distribution of the characteristics among cases without osteoporosis, compared with cases with osteoporosis. For risk factors previously identified in the SOF cohort (20, 21), we modeled variables as they had been categorized in those analyses. For additional characteristics considered, we modeled variables as continuous if there was evidence for a linear trend in the log OR for that variable. When there was not a trend in OR, we examined the OR in categories of the continuous variable (by creating quartiles or equal increments). To express the OR and 95% confidence intervals (CIs) for continuous variables, units of change were chosen to be approximately 1 SD in the distribution of that variable for all participants. The multivariable model was generated by first examining groups of related variables in the main areas of interest in our investigation of participant characteristics (physical activity, family history of fracture, and fall propensity) for associations with total hip BMD T-score greater than –2.5 (no osteoporosis). When more than one variable within a group was associated with total hip BMD T-score greater than –2.5, we examined these variables for multicollinearity and determined which parsimoniously explained the association of variables in these groups with total hip BMD T-score greater than –2.5. These selected variables were then included in the full multivariable analysis. Other variables from Table 1
were examined and included in the model if they were associated with total hip BMD T-score greater than –2.5 independent of the selected variables in the main areas of interest, or behaved as confounders to the relationship between the variables in the main areas of interest and total hip BMD T-score greater than –2.5.
|
After completing the case-case comparison, we used univariate and multivariable Cox proportional hazards regression analyses to quantify the relationship between baseline characteristics and hip fracture risk among women with and without osteoporosis. We approached the classification of variables and the modeling strategy as we had for the logistic regression analyses described above. To improve the proportional hazards regression models, a variable for total hip BMD was added to address potential confounding by BMD.
All statistical analyses used SAS (version 8.1, SAS Institute Inc., Cary, NC).
|
Results |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsAppendixReferences |
|---|
). The mean total hip BMD and BMD T-score were 0.76 g/cm2 (SD 0.13) and –1.51 (SD 1.07), respectively; 6667 (83%) of the participants were without hip osteoporosis. BMD in hip fracture cases
During 5 yr of observation, 243 participants experienced a new hip fracture. Among the women with incident hip fracture, median age at the start of follow-up was 77 yr (range 67–95) and mean baseline weight was 63.0 kg (SD 12.5). Mean time of observation from BMD measurement to fracture was 2.8 yr (SD 1.4). Crude incidence rates of hip fracture were 17.7 per 1000 person-years among women with hip osteoporosis and 4.1 per 1000 person-years among those without osteoporosis.
Although the average BMD in the hip fracture cases was lower than that among all participants, of the 243 incident hip fracture cases, 54% (131) did not have hip osteoporosis (Fig. 1), 32% had total hip BMD T-scores greater than –2.0, and 6% had total hip BMD T-scores greater than –1. To examine whether the length of follow-up influenced these results, we restricted the analysis to the hip fracture cases identified during the first 2 yr of follow-up. These results were similar; in 2 yr of follow-up, 49% of 76 hip fracture cases did not have hip osteoporosis and 28% had total hip BMD T-scores greater than –2.0 at start of observation. With the exception of the oldest women, after 5 yr of follow-up, the majority of hip fracture cases was without hip osteoporosis regardless of age; hip osteoporosis at start of observation was absent in 58% of the 26 fracture cases aged 65–69 yr at the time of BMD measurement, 55% of the 65 women aged 70–74 yr, 66% of the 74 women aged 75–79 yr, 44% of the 22 women aged 80–84 yr, and 32% of the 28 women aged 85 and older (Fig. 2).
|
|
); of those with incident hip fracture in 5 yr of follow-up, 54% were without osteoporosis at the lumbar spine, and 42% did not have osteoporosis at the lumbar spine or for the total hip. Osteoporosis at the femoral neck BMD measurement site was not present for 37% of hip fracture cases. Osteoporotic vs. nonosteoporotic hip fracture cases
Compared with cases with osteoporosis at the total hip at start of observation, those without osteoporosis were younger, heavier, and taller; had better distance depth perception, grip strength, and knee extension force; and less frequently had weight loss since age 25 yr, previous hyperthyroidism, or prevalent vertebral fracture (Table 1) at baseline. Hip fracture cases with and without osteoporosis did not differ with regard to self-rated health, time on feet less than 4 h each day, weighted lifetime activity, weekly activity in the last year, mental status score, ability to stand without using one’s arms, any falls in the last year, or maternal history of hip fracture.
The results of the final multivariable logistic regression model (Table 2) indicated that independent of other variables in this model, hip fracture cases without hip osteoporosis were significantly less likely than fracture cases with hip osteoporosis to have had a history of hyperthyroidism or prevalent vertebral fracture and had greater body mass index (BMI) and grip strength. In addition, these data suggest that hip fracture cases without osteoporosis may have been less likely to be age 80 yr or older than cases with osteoporosis.
|
). We note, however, that all fracture cases with osteoporosis and most fracture cases without osteoporosis (96%) had at least one non-BMD risk factor and that a greater proportion of fracture cases without osteoporosis had at least five fracture risk factors than for all participants without osteoporosis.
|
A number of previously identified risk factors for hip fracture (20, 21) were associated with increased fracture risk when multivariable analysis was restricted to women without hip osteoporosis (Table 3) or with hip osteoporosis (Table 4). Independent of other variables in the model, risk factors for hip fracture among women without osteoporosis included advancing age, reduced visual contrast sensitivity, falls in the last year, prevalent vertebral fracture, and lower total hip BMD. Furthermore, there was a tendency toward decreased risk associated with walking for exercise vs. no exercise activity in the past year and significantly decreased risk for other exercise activity alone or in addition to walking in the previous year.
|
|
In this observational study of a large cohort of older women, many (54%) hip fracture cases that occurred in 5 yr of follow-up did not have osteoporosis at the total hip at the start of observation. Furthermore, many did not have osteoporosis at the lumbar spine or a combination of axial BMD measurement sites. These findings are important because they call attention to the number of women with hip fracture but without particularly low antecedent BMD. Moreover, current therapies for reducing fracture risk have been evaluated primarily in women with low BMD levels, but in postmenopausal women without osteoporosis it has been difficult to demonstrate the antifracture efficacy of antiresorptive treatment (1, 2). Even if effective, widespread therapy of large populations with a low absolute risk of fracture would be prohibitively expensive. Thus, an important segment of women who will experience fractures does not have hip osteoporosis, and therapies for the prevention of fracture are not proven for these women with higher BMD.
The finding that fractures occur in older women without osteoporosis at the start of follow-up is not inconsistent with the well-documented relationship between BMD and fracture risk. It is clear that those with low total hip BMD are at higher risk. Indeed, in our study the hip fracture incidence rate was more than 3 times higher among women with osteoporosis. Similarly, hypertension is strongly associated with the risk of stroke, but strokes also occur in normotensive individuals (28, 29).
A negative association between hip BMD and age and positive associations between hip BMD and weight and strength have been described previously in the SOF cohort (30). Therefore, our similar findings in analyses restricted to fracture cases might be expected. However, we hypothesized that on average women without osteoporosis who suffer hip fracture might have a greater propensity for falling, might be more active (and hence more likely to fall), or might have a genetic predisposition to fracture and as a result, an increased fracture risk despite higher levels of BMD. Instead, baseline fall frequency, family history of fracture, and activity were similar in fracture cases with and without osteoporosis at the start of observation, and we were not able to demonstrate increased risk associated with activity among women without osteoporosis. Furthermore, we considered that cases without osteoporosis might have more risk factors for fracture than those who fracture with lower BMD at the start of follow-up. On the contrary, although most cases had at least one non-BMD risk factor for fracture, cases without osteoporosis at start of follow-up had fewer baseline risk factors, and those with osteoporosis who fractured appeared generally frailer at baseline.
It is important to note that whereas women without hip osteoporosis who experience hip fracture are not as frail as might be anticipated, several factors are associated with an increased risk of fracture in this group, including advancing age, lack of exercise activity in the last year, reduced visual contrast sensitivity, falls in the last year, prevalent vertebral fracture, and lower total hip BMD. These results are consistent with previous reports from this population that emphasized that these factors were associated with fracture risk independent of BMD and point to the potential clinical usefulness of assessing information concerning these factors even in the absence of osteoporosis.
Although in these initial analyses we focused on previously established hip fracture risk factors for all women (with and without osteoporosis), through further study it will be important to identify possible new hip fracture risk factors specific to nonosteoporotic women. Candidates for such investigations include geometric, structural, or material properties of bone.
This study has important strengths. It is based on a large, community-based, well-characterized population of postmenopausal women followed prospectively, with follow-up after 5 yr of observation remaining more than 98% complete for surviving participants. A large number of validated fractures are available for analysis. The results reported here should be applicable to a large segment of the population of women at risk for fracture. On the other hand, the study also has several limitations. First, a disproportionate loss to follow-up, specifically loss as a result of death, may have affected the proportion of hip fracture cases with and without osteoporosis. This disproportionate loss to death is not unexpected because an inverse relationship between mortality and BMD has been described (31, 32, 33). However, even if we assume that all women with osteoporosis at start of observation who died before completing fracture follow-up in the 5 yr after BMD measurement would have experienced a hip fracture and that none of the women without osteoporosis who died in the same follow-up would have fractured, we would still observe that 32% of those with hip fracture did not have osteoporosis at the start of observation. Therefore, the large proportion of hip fractures without osteoporosis is not explained completely by loss to death.
Because many of the characteristics of interest were assessed at only the first study examination, we analyzed baseline characteristics and risk factors from the first examination but BMD from the second examination. Some risk factors, such as activity, health status, vision, and fracture history, may have changed during that 2-yr interval. However, it seems unlikely that major changes in health status occurred in a large number of women in this short period.
Finally, women enrolled in this study were primarily Caucasian, community-dwelling volunteers in the United States, and our findings may not be generalizable to populations of older women.
In summary, a large proportion of older women who experience hip fracture has antecedent total hip BMD measurements, and other axial BMD measurements, which are not dramatically low, suggesting that the health care burden represented by hip fractures in women without osteoporosis may be large. And, interestingly, whereas it might be assumed that the group with fracture but without hip osteoporosis would be older and frailer than those with fracture and osteoporosis, our results suggest that on average just the opposite is true. Fracture cases without hip osteoporosis at start of observation were younger and seemed to be less frail at baseline than women with hip osteoporosis who suffered a fracture. Still, several factors, including advancing age, lack of exercise in the last year, reduced visual contrast sensitivity, falls in the last year, prevalent vertebral fracture, and lower total hip BMD, were found to be associated with increased fracture risk in women without hip osteoporosis. Together, these findings highlight the complex etiology of hip fracture and help begin to identify risk factors associated with higher bone density levels.
|
Appendix |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsAppendixReferences |
|---|
University of California, San Francisco (Coordinating Center): S. R. Cummings (principal investigator), M. C. Nevitt (coinvestigator), D. C. Bauer (coinvestigator), K. L. Stone (coinvestigator), D. M. Black (study statistician), H. K. Genant (director, central radiology laboratory), T. Blackwell, B. Blunt, M. Dockrell, S. Ewing, C. Fox, M. Jaime-Chavez, S. Litwack, L.Y. Lui, P. Mannen, L. Nusgarten, L. Palermo, M. Rahorst, C. Schambach, J. Schneider, R. Scott.
University of Maryland: M. Hochberg (principal investigator), L. Makell (project director), C. Boehm, L. Finazzo, R. Nichols, T. Page, S. Trusty, B. Whitkop.
University of Minnesota: K. Ensrud (principal investigator), M. Homan (coinvestigator), P. Bowman (project coordinator), S. Love (clinical research director), E. Mitson (clinic coordinator), C. Bird, D. Blanks, C. Burchkhardt, M. Cardenas, J. Holmes, F. Imker-Witte, K. Jacobson, K. Moen, N. Nelson, H. Peterson, M. Slindee.
University of Pittsburgh: J. A. Cauley (principal investigator), L. H. Kuller (co-principal investigator), M. Vogt (coinvestigator), L. Harper (project director), L. Buck (clinic coordinator), C. Bashada, N. Chiarvalle, A. Githens, M. Gorecki, D. Lee. D. Medve, C. Newman, D. Stewart, N. Watson.
The Kaiser Permanente Center for Health Research, Portland, Oregon: T. Hillier (principal investigator), E. Harris (co-principal investigator), E. Orwoll (coinvestigator), H. Nelson (coinvestigator), M. Aicken (coinvestigator), J. Van Marter (project administrator), M. Rix (clinic coordinator), K. Canova, T. Constantin-Suvalcu, R. Garza, P. Legarda, K. Pedula, K. Redden, J. Rehinhardt, J. Rizzo, K. Snider, J. Wallace.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online February 22, 2005
Abbreviations: BMD, Bone mineral density; BMI, body mass index; CI, 95% confidence interval; OR, odds ratio; SOF, Study of Osteoporotic Fractures.
Received August 5, 2004.
Accepted February 15, 2005.
|
References |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsAppendixReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. K. Ivaska, J. Lenora, P. Gerdhem, K. Akesson, H. K. Vaananen, and K. J. Obrant Serial Assessment of Serum Bone Metabolism Markers Identifies Women with the Highest Rate of Bone Loss and Osteoporosis Risk J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2622 - 2632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. G. Bone, M. A. Bolognese, C. K. Yuen, D. L. Kendler, H. Wang, Y. Liu, and J. San Martin Effects of Denosumab on Bone Mineral Density and Bone Turnover in Postmenopausal Women J. Clin. Endocrinol. Metab., June 1, 2008; 93(6): 2149 - 2157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. D. Rozental, E. C. Makhni, C. S. Day, and M. L. Bouxsein Improving Evaluation and Treatment for Osteoporosis Following Distal Radial Fractures. A Prospective Randomized Intervention J. Bone Joint Surg. Am., May 1, 2008; 90(5): 953 - 961. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Cauley, M. C. Hochberg, L.-Y. Lui, L. Palermo, K. E. Ensrud, T. A. Hillier, M. C. Nevitt, and S. R. Cummings Long-term Risk of Incident Vertebral Fractures JAMA, December 19, 2007; 298(23): 2761 - 2767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Robbins, A. K. Aragaki, C. Kooperberg, N. Watts, J. Wactawski-Wende, R. D. Jackson, M. S. LeBoff, C. E. Lewis, Z. Chen, M. L. Stefanick, et al. Factors Associated With 5-Year Risk of Hip Fracture in Postmenopausal Women JAMA, November 28, 2007; 298(20): 2389 - 2398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. SILVERMAN Selecting Patients for Osteoporosis Therapy Ann. N.Y. Acad. Sci., November 1, 2007; 1117(1): 264 - 272. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Assessing Fracture Risk in Individual Patients Journal Watch Women's Health, August 30, 2007; 2007(830): 1 - 1. [Full Text]
|
|
|
|
|
|
N. D. Nguyen, J. A. Eisman, J. R. Center, and T. V. Nguyen Risk Factors for Fracture in Nonosteoporotic Men and Women J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 955 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Lewiecki and A. J. Laster Clinical Applications of Vertebral Fracture Assessment by Dual-Energy X-Ray Absorptiometry J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4215 - 4222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Siris Osteopenia: A Risk Factor that Deserves Some Respect IBMS BoneKEy, December 1, 2005; 2(12): 42 - 44. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
