Risk factors for postmenopausal osteoporosis

Share Embed


Descrição do Produto

Risk Factors for Postmenopausal Osteoporosis

JOHN F. ALOIA, M.D. Mineola,

New

York

STANTON H. COHN, Ph.D. Upton,

New

York

ASHOK VASWANI, M.D. Mineola

and Upton,

New

York

JAMES K. YEH, Ph.D. Mheola,

New

York

KAPO YUEN, B.S., M.S. KENNETH ELLIS, Ph.D. Upton,

New

York

From Nassau Hospital, Mineola, New York, and Brookhaven National Laboratory, Upton, New York. This work was supported in part by Contracts EY-76-C-02-0016 and DE-ACOP76CH00016 from the Unlted States Department of Energy. Calcium studies were supported by National Institutes of Health Grant l-ROl-AG 00919-04. Requests for reprints should be addressed to Dr. John F. Aloia, Nassau Hospital, 259 First Street, Mineola, New York 11501. Manuscript accepted July 19, 1964.

‘Fifty-eight women with postmenopausal osteoporosis (crush fracture of the spine) were compared with 58 age-matched normal women. The osteoporotic women had lower total-body calcium levels and bone mineral content of the radius, had undergone an earlier menopause, smoked cigarettes more, and had breast-fed less often. They also had lower levels of estrone, estradiol, and testosterone and reduced levels of 25hydroxyvitamin D, 24,25-dihydroxyvitamin D, and 1,2!Gdihydroxyvitamin D. These findings suggest the presence of changeable risk factors for the development of osteoporosis. Smoking should be discouraged. An adequate intake of calcium and vitamin D should be ensured. It is the opinion of the authors that those women who have had an early menopause or who have a low bone mass at the time of menopause should be given the choice of medically supervised replacement therapy with estrogen and progesterone. Postmenopausal osteoporosis is a significant cause of morbidity and mortality in the United States, leading to fractures of the hip, spine, and wrist. Osteoporosis may be defined as osteopenia of such magnitude as to have resulted in structural failure of bone (fracture). A variety of epidemiologic studies in osteoporotic populations have attempted to delineate the factors that place some persons at high risk of fracture. More recently, it has become possible to perform similar studies using newly developed, quantitative methods for the measurement of bone mass. These techniques permit evaluation of the magnitude of osteopenia, rather than limiting the analysis to the extreme of fracture rate. In the current study, both total-body neutron activation analysis and the technique of single-photon absorptiometry of the radius were used to measure bone mass. Various endocrine and historical parameters were also determined in women with postmenopausal osteoporosis and in age-matched control subjects. It was hoped in this manner to identify those factors that result in the development of vertebral crush fractures in some women in the seventh decade of life and not in other women. SUBJECTS AND METHODS recruited with announcements in the press, conSubJects. The subjects, sisted of 58 white women with vertebral crush fractures and 58 healthy, white women with no history of osteoporosis. Persons were excluded from either group if they had any chronic illness or were receiving any medication associated with accelerated bone loss. All osteoporotic women had experienced back pain and at least one vertebral crush fracture. Subjects with a vertebral crush fracture on radiography were excluded from the control group.

January

1965

The American

Journal

of Medlcine

Volume

76

95

RISK FACTORS

FOR OSTEOPOROSIS-ALOIA

ET AL

Diet and Physical Activity. Seven-day diet histories were available for analysis by a nutritionist for 38 of the osteoporotic women and 38 of the control subjects. Data from the “Handbook of Agriculture,” Number 8, were used. Physical activity was estimated by a scoring system during the initial history. Total-Body Neutron Activation Analysis and Whole-Body Countlng (Total Bone Mass). The details of neutron activation analysis and the characteristics of the Brookhaven 54detector whole-body counter have been previously described [ 1,2]. The patients underwent uniform exposure to a beam of partially moderated fast neutrons. The resulting induced activities were then measured with the whole-body counter, permitting determination of total-body calcium in an anthropomorphic phantom with a precision of f 1.1 percent SD. The radiation dose for each activation is 0.28 rem. Photon Absorptlometry (Regional Bone Mass). The bone mineral content and the width of the radius were measured by the Cameron-Sorenson technique with a Norland Instruments absorptiometer. Measurements were performed at the 8 cm radius site. The accuracy of this technique is f5 percent and the precision (reproducibility) is f2.5 percent SD [31. Biochemical Measurements. Routine laboratory chemical studies included complete blood count, urinalysis, serum fasting calcium, phosphorus, urea nitrogen, creatinine, alkaline phosphatase, aminotransferase, thyroxine, cholesterol, and triglycerides. A 24-hour urine sample was collected for hydroxyproline peptides, calcium, phosphorus, and creatinine on an outpatient basis on the third day of a low hydroxyproline diet. A maturation index and Papanicolaou smear were performed in each patient. Fasting serum samples were analyzed for growth hormone, prolactin, estrone, estradiol, estriol, progesterone, testosterone, follicle-stimulating hormone, luteinizing hormone, parathyroid hormone, 25hydroxyvitamin D, 24,25dihydroxyvitamin D, and 1,25dihydroxyvitamin D. The growth hormone was assayed using a Kallestad Laboratories radioimmunoassay kit (Chaska, Minnesota). Testosterone and prolactin were assayed using a Serono Laboratories radioimmunoassay kit (Braintree, Massachusetts). Parathyroid hormone, lutenizing hormone, and follicle-stimulating hormone were assayed using Cambridge Nuclear Radiopharmaceutical radioimmunoassay kits. The antiparathyroid hormone antibody in this assay is directed against the mid-region of the parathyroid hormone molecule. Serum 25-hydroxyvitamin D was measured by the method of Haddad and Chyu [4]. Serum 1,25dihydroxyvitamin D and 24,25dihydroxyvitamin D were measured by the method of Eisman et al [5]. Urinary total and nondialyzable hydroxyproline were determined by the method of Prockop and Udenfriend [6]. Serum estrone, estradiol, and estriol were measured by radioimmunoassay. The solvent extraction and column chromatographic separation were performed by the method of Tsai and Yen [7]. The antiestrone and antiestradiol antiserum was obtained through the courtesy of Dr. G. D. Niswender (Colorado State University, Fort Collins, Colorado). The antiestriol serum was obtained from C.I.D. Radiopharmaceuticals, Inc. (Bedford, Massachusetts). Statistical Analysis. For the comparison of body composition, chemical values, and hormonal and dietary variables 96

January

1985

The American

Journal

of Medlcine

Volume

between the age-matched control subjects and osteoporotic women, two-sample t tests were performed. All p values were twc&iled. The relationships between the historical data, biochemical variables, and bone mass were assessed by calculating Pearson correlations in the control and osteoporotic groups separately and also in the combined population. Stepwise logistic regression was used to examine the risk factors for the development of osteoporosis, as well as to develop a probabilistic model to identify asymptomatic women who are at risk of development of osteoporosis. The estimated partial odds ratio of a particular risk factor was computed by taking the exponent of the product of its coefficient in the logistic equation with the difference within the variable. RESULTS

Clinical Characteristics. The mean ages of the control group and the osteoporotic patients were nearly identical (Table I). The mean values of height and weight of the osteoporotic group were slightly lower. Cessation of the menses occurred at a younger age in the osteoporotic women than in the normal control subjects. Cigarette smoking was more prevalent among osteoporotic women than among the control subjects. There was a higher number of pregnancies in the age-matched control group, but the difference was not statistically significant. The age-matched control subjects had breast-fed their children more often than did the osteoporotic women. Diet and Physical Activity. The mean activity scores did not differ in the two groups of patients. The osteoporotic women had a higher caloric intake with a significantly higher intake of protein, calcium, phosphorus, and vitamin D. Many of the osteoporotic patients had knowingly increased their intake of milk and dairy products after they received the diagnosis of osteoporosis. The daily calcium intake of the osteoporotic and control groups was 911 f 54 mg and 616 f 30 mg, respectively. Their daily vitamin D intake was 218 f 18 IU and 135 f 10 IU, respectively. Body Composition and Blood Chemical Values (Table II). Total bone mass (total-body calcium) and compact bone (bone mineral content) of the radius were both significantly reduced in the osteoporotic women. The bone mineral content and total-body calcium level were 16 and 20 percent lower, respectively, in the women with vertebral

crush

fractures.

Mean serum

chemical

values were almost identical in the two groups except for alkaline phosphatase, which was significantly higher in the osteoporotic women. Urine chemical values did not differ in the groups; there was no difference in total or nondialyzable hydroxyproline excretion. Quantitative histologic analysis of anterior iliac crest bone biopsy specimens, including double tetracycline labeling, was available in four of eight patients with hyperphosphatasia. These bone biopsy specimens had in common higher levels of bone resorption surfaces 78

RISK

without evidence of osteomalacia. Additionally, the mean parathyroid hormone level in these eight patients was almost 100 pg/ml higher than the osteoporotic group mean value. Hormonal Measurements. The estrone and estradiol levels were lower in the osteoporotic group than in the normal group (Table III). The serum thyroxine level was higher in the osteoporotic women. Although the serum parathyroid hormone level was lower in the normal postmenopausal women, this was not statistically significant. The mean values for each vitamin D metabolite were lower in the osteoporotic women than in the normal women. Only one of the subjects has 25-hydroxyvitamin D levels below 15 rig/ml. Analysis of the vitamin D data by separating values according to the season in which blood samples were drawn did not result in a different interpretation. The 1,25dihydroxyvitamin D and 24,25-dihydroxyvitamin D levels were reduced in the osteoporotic women; the reduction was proportional to the reduction in the 25-hydroxyvitamin D level. Relationship among Variables. The following correlations were noted in the postmenopausal osteoporotic women. Alkaline phosphatase level was inversely related to bone mineral content (r = -0.33) and total-body calcium level (r = -0.34). The total-body calcium level in the group with postmenopausal osteoporosis was related to the serum estradiol level (r = 0.24). The combined population of patients demonstrated similar correlations of the bone mass measurements with age and height to those previously observed (total-body calcium level versus height, r = 0.69, total-body calcium level versus age, r = -0.22). The relationship between bone mineral content and total-body calcium level was highly correlated (r = 0.75) (Figure 1). Calcium intake was not related to parathyroid hormone or 1,25dihydroxyvitamin D levels. Discrimination of Osteoporotic Women from Nonosteoporotic Women. A multiple logistic regression model was developed for estimating the likelihood of the presence or absence of osteoporosis. The variables selected were total-body calcium levels, lactation, cigarette smoking, and 25-hydroxyvitamin D level. Age was not included as a factor, since our subjects were age-matched. Similarly, the bone mineral content measurement did not improve the separation of the two groups of women because of the relationship of bone mineral content to total-body calcium level. Other variables could be included or substituted in the model. An example of a model developed follows: Logit prob (normal) = 0.0237 X total-body calcium + 0.746 X lactation - 0.601 X smoking i- 0.218 X 25-hydroxyvitamin D - 21 .Q (likelihood goodness of fit chi-square = 67.45 with 102 degrees of freedom, and p value = 0.997). Smoking was indicated as + 1 for smokers and - 1 for nonsmokers and lactation as i- 1 for those who January

TABLE I

FACTORS

FOR OSTEOPOROSIS-ALOIA

ET AL

Clinical Characteristics of Age-Matched Control Subjects and Osteoporotic Women (mean f SE) Age-Matched Control

Variable Age (years) Height (m) Weight (kg) Age at menopause (years) Percent who had pregnancies Percent smokers Percent who breast-fed

63.9 1.61 65.6 49.0

Postmenopausal Osteomrosls

f 0.8 f 0.01 f 1.6+ f 0.5” 84 30% 35+

64.5 1.54 60.6 45.8

l

f f f f 75 59 16

1.2 0.01 1.7 1.1

l p
Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.