The Influence of Growth Hormone Deficiency, Growth Hormone ...

39 downloads 144 Views 167KB Size Report
Hormone Replacement Therapy, and Other Aspects of ... Key words: hypopituitarism, growth hormone deficiency, fracture rate, bone mass, growth hormone ther-.
JOURNAL OF BONE AND MINERAL RESEARCH Volume 16, Number 2, 2001 © 2001 American Society for Bone and Mineral Research

The Influence of Growth Hormone Deficiency, Growth Hormone Replacement Therapy, and Other Aspects of Hypopituitarism on Fracture Rate and Bone Mineral Density ¨ STER,1 ROGER ABS,2 BENGT-ÅKE BENGTSSON,3 HELGE BENNMARKER,4 CHRISTIAN WU ULLA FELDT-RASMUSSEN,5 ELIZABETH HERNBERG-STÅHL,4 JOHN P. MONSON,6 BJØRN WESTBERG,4 and PATRICK WILTON,4 ON BEHALF OF THE KIMS STUDY GROUP AND THE KIMS INTERNATIONAL BOARD

ABSTRACT To assess the influence of factors affecting fracture risk and bone density in adult hypopituitary patients with growth hormone deficiency (GHD), data from a large-scale pharmacoepidemiological survey (the Pharmacia & Upjohn International Metabolic Database [KIMS]) were analyzed and compared with data from a control population (the European Vertebral Osteoporosis Study [EVOS]). The KIMS group consisted of 2084 patients (1112 men and 972 women) with various types of pituitary disease and EVOS consisted of 1176 individuals (581 men and 595 women). Fracture and bone mineral density (BMD) data were available from 2024 patients from the KIMS group and 392 patients from EVOS. The prevalence of fractures in patients with hypopituitarism was 2.66 times that in the non–GH-deficient EVOS population. Adult-onset hypopituitarism with GHD was associated with a higher fracture risk than childhood-onset disease, and patients with isolated GHD had a similar prevalence of fractures to those with multiple pituitary hormone deficiencies. Hormonal replacement therapy with L-thyroxine, glucocorticoids, and sex steroids did not affect the risk of fracture in KIMS patients. In addition, fracture rates in KIMS were independent of body mass index (BMI) and the country of origin. However, smoking was associated with a higher fracture rate in this group. In summary, this is the first large-scale analysis to support the hypothesis of an increased fracture risk in adult patients with hypopituitarism and GHD. This increased risk appears to be attributable to GHD alone, rather than to other pituitary hormone deficiencies or to their replacement therapy. (J Bone Miner Res 2001;16:398 – 405) Key words:

hypopituitarism, growth hormone deficiency, fracture rate, bone mass, growth hormone therapy, osteoporosis

INTRODUCTION YPOPITUITARISM OF childhood or adult onset combined with untreated growth hormone deficiency (GHD) is associated with reduced bone mass.(1–7) These findings are

H

clinically important because both bone mineral content and bone mineral density (BMD) have been shown to correlate with the risk of fractures.(8,9) Indeed, there have been reports that fracture rates are increased in patients with hypopituitarism and untreated GHD compared with control popula-

1

Department of Internal Medicine 1, Endocrinology and Metabolism, University Medical Clinic, Heidelberg, Germany. Department of Endocrinology, University Hospital, Antwerp, Belgium. 3 Division of Endocrinology, Sahlgrenska Hospital, Go¨teborg, Sweden. 4 KIMS Outcomes Research Department, Metabolic Diseases, Pharmacia & Upjohn, Stockholm, Sweden. 5 Clinic of Endocrinology, National University Hospital, Copenhagen, Denmark. 6 Department of Endocrinology, St Bartholomew’s Hospital, London, UK. 2

398

GROWTH HORMONE DEFICIENCY AND FRACTURE RISK

tions,(10,11) although the numbers of patients involved in these studies were small for assessing fracture risk (122 and 107 patients, respectively). The cause of reduced bone mass in patients with hypopituitarism and untreated GHD is not fully understood and many contributory factors may exist. For example, the use of too high a dose of glucocorticoids for replacement therapy,(12) untreated hypogonadism,(13,14) or the use of other pharmaceutical agents such as cytotoxic drugs(15) are all known to reduce bone mass. However, several studies(2,5,16) have shown no difference in BMD between patients with isolated GHD and those with multiple pituitary hormone deficiencies, suggesting that GHD itself is a key factor in reduced bone mass. A direct link between GHD and reduced bone mass in hypopituitarism is supported by reports that GH replacement therapy can improve BMD in these patients.(17–25) Typically, BMD remains constant or decreases during the first year of GH replacement therapy but increases in the longer term. In the study of the longest duration published to date, patients were assessed after 4 years of GH treatment, at which point age- and sex-related BMD at the lumbar spine had increased by more than 10% from baseline.(22) Although GH replacement therapy is known to have a beneficial effect on bone mass, there have been no reports to date regarding the impact of GH therapy on fracture rates in patients with hypopituitarism. Ideally, the efficacy of any therapeutic agent should be assessed in a randomized, double-blind, placebo-controlled trial involving an adequately large patient group. However, this protocol is not always practical. To establish whether a treatment reduces the risk of fractures requires at least 1000 patients and a prolonged study period. With regard to GH replacement therapy, there would be difficulties in recruiting a large number of suitable patients, because of the low prevalence of hypopituitarism. Furthermore, the requirement for an extended period of placebo treatment may be considered unethical because of the known beneficial effects of GH on quality of life, serum lipid profiles, fat mass, and muscle mass in adults with GHD.(26) In circumstances where prospective studies are impractical, insights can be gained by analyzing data from large-scale pharmacoepidemiological surveys. One such survey is (the Pharmacia & Upjohn International Metabolic Database [KIMS]), which aims to monitor the long-term safety and efficacy of GH replacement therapy in adults. At present, KIMS contains data from over 4500 GHD patients from 26 countries, providing a valuable resource for epidemiological analysis. Descriptive demographic and treatment responses of the KIMS patient cohort have been published previously.(27–29) In the present analysis, we used data from patients registered in KIMS to assess BMD and the prevalence of fractures and the factors affecting these variables, in patients with hypopituitarism and GHD. For purposes of comparison, data from a non–GH-deficient population (European Vertebral Osteoporosis Study [EVOS]) were used. In addition, we looked for signs of a beneficial treatment effect in two separate patient populations: a group that had received

399

GH therapy before entry into KIMS and a group that had not received GH therapy for at least 6 months before entry.

MATERIALS AND METHODS KIMS patients Enrollment of patients into KIMS began in 1994 and, although the number of countries represented in KIMS continues to grow, most participants are currently from Sweden, Germany, United Kingdom, The Netherlands, Belgium, and Austria. After enrollment, patients visit their local clinic at a frequency determined by the treating physician; however, a minimum of one visit per year is mandatory. At each clinic visit, data are collected on specially designed case report forms and are entered into a central database. This data collection process is monitored by representatives of the sponsoring company. In addition, the accuracy of data entry into the KIMS database has been confirmed by internal audit as well as externally by the physician members of the KIMS Strategic Scientific Committee. For entry into KIMS, patients must have confirmed GHD. Patients with ongoing malignant neoplasm are excluded. Having given written informed consent, registered patients receive individualized doses of GH (Genotropin; Pharmacia & Upjohn, Stockholm, Sweden) and are expected to be followed for a minimum of 5 years, regardless of whether treatment is continued throughout this period. Patients receiving GH therapy at the time of enrollment are referred to as “nonnaive”; this group often includes adults with childhood-onset GHD who have received GH for several years. Patients classified as “naive” have never received GH therapy or this treatment has been discontinued for at least 6 months before enrollment. Patient characteristics are shown in Table 1. The present analysis involved data from 2084 patients (1112 men and 972 women) enrolled in KIMS with a mean age of 50 ⫾ 11.7 years (range, 18 – 82 years). Of these, 813 were classed as naive and 1271 were classed as nonnaive. There were no significant differences in clinical characteristics between patients from different countries or patients in the naive and nonnaive groups before enrollment in KIMS. The mean duration of GH therapy in the nonnaive group was 1.8 years (range, 0.1–10.1 years). Fracture data were available from 2024 KIMS patients (1079 men and 945 women): 1518 had adult-onset GHD and 506 had childhood-onset GHD and 790 were classed as naive and 1234 as nonnaive. BMD data were available from 392 patients. There were no differences between the groups that provided fracture and BMD data with regards to sex distribution, the proportions of naive and nonnaive patients, and the proportions of adult-onset GHD and childhood-onset GHD patients.

Assessments Details of all previous fractures experienced by KIMS patients were recorded using a questionnaire. On request, radiological documentation was sought for each patient who had experienced a fracture. Stress fractures were not taken into account for this analysis. Fracture prevalence was an-

¨ STER ET AL. WU

400

TABLE 1. CLINICAL CHARACTERISTICS

OF THE

KIMS

AND

EVOS POPULATIONS

KIMS

Total numbers of individuals (n) Age (years) BMI (kg/m2) Numbers of age-matched subgroups (n) Age of age-matched subgroups (years) Numbers with adult-onset GHD (n) Numbers with childhood-onset GHD (n) Numbers of naı¨ve patients (n) Numbers of nonnaı¨ve patients (n)

EVOS

Females

Males

Females

Males

972 49 ⫾ 11 28 ⫾ 6 108 65.1 ⫾ 3.4 729 243 379 593

1112 51 ⫾ 12 28 ⫾ 4 156 65.8 ⫾ 3.9 834 278 434 678

595 63.7 ⫾ 7.7 28 ⫾ 5 595 63.7 ⫾ 7.7 — — — —

581 65.0 ⫾ 8.6 28 ⫾ 3 581 65.0 ⫾ 8.6 — — — —

alyzed subsequently by age (decade) and age at time of fracture (⬍20 years, 20 –50 years, and ⬎50 years). Data from KIMS patients were compared with sex and agematched data from EVOS to assess the effect of hypopituitarism on the prevalence of fractures. EVOS is a population-based study of the prevalence of vertebral osteoporosis in 19 countries (countries that also are represented in KIMS).(30) An identical questionnaire to determine fracture history is used in KIMS and EVOS. Fracture data from 1176 individuals (581 men, mean age, 65.0 ⫾ 8.6 years; 595 women, mean age, 63.7 ⫾ 7.7 years) in EVOS were available. Participants in EVOS were, on average, 20 years older than the KIMS patients; thus, fracture data from EVOS were compared with fracture data from an older subgroup of 264 KIMS patients (156 men, mean age, 65.8 ⫾ 3.9 years; 108 women, mean age, 65.1 ⫾ 3.4 years) who were all ⬎60 years of age. Dual-energy X-ray absorptiometry (DXA) was used to measure BMD at the lumbar spine in patients enrolled in KIMS (several devices were used), and BMD Z scores (i.e., the difference in SDs from the mean BMD for age- and sex-matched healthy controls) were calculated according to the manufacturer’s reference curves. In addition, the prevalence of Z scores of ⬍ ⫺2 SD were analyzed. The following factors, which might affect fracture prevalence and BMD, also were assessed: body mass index (BMI; ⱕ27 kg/m2 vs. ⬎27 kg/m2), number of pituitary hormone deficiencies, glucocorticoid therapy (patients receiving vs. those not requiring treatment), L-thyroxine therapy (patients receiving vs. those not requiring treatment), estrogen replacement therapy (women receiving vs. those not requiring treatment), testosterone replacement therapy in all hypogonadal men, and smoking habit. In addition, comparisons were made between patients with and without diabetes insipidus. Comparing naive and nonnaive patient groups assessed the effect of GH replacement therapy on BMD and fracture risk.

Statistical analysis All statistical analyses were performed using SAS version 6.12 (SAS Institute, Cary, NC, USA). Treatment effects were analyzed by paired t-tests. Intergroup comparisons

were performed by ƒ tests or nonparametric statistics, when appropriate. For all tests, values of p ⬍ 0.05 were considered significant.

RESULTS Prevalence of fractures in KIMS patients A total of 578 fractures were counted within KIMS, 317 fractures in males (28.5%) and 261 fractures in females (26.8%). The sites of fractures were located at almost any bone, the majority being fractures of the radius (n ⫽ 192 or 9.5%), 94 in males (8.4%) and 98 in females (10.1%). There were 30 hip fractures (1.48%). The prevalence of fractures in the subgroup of KIMS patients over 60 years old was similar to the prevalence of fractures in all KIMS patients (Fig. 1). The prevalence of fractures in KIMS patients over 60 years of age was 2.66 times that in the age-matched EVOS cohort. The difference was significant in men (Fig. 1). Similar differences were found when comparisons of KIMS and EVOS fracture data from individual countries were made. No significant differences were seen in the prevalence of fractures in KIMS patients between different countries, except when patient numbers were less than 100.

Influence of additional pituitary hormone deficiencies The prevalence of fractures and the percentage of patients with BMD Z scores below ⫺2.0 SD in KIMS were similar in patients with isolated GHD and in those with multiple pituitary hormone deficiencies (Fig. 2). Furthermore, there was no evidence of any trend between these parameters and the severity of hypopituitarism, as measured by the number of additional deficiencies. The effects of replacement therapy for pituitary hormone deficiencies other than GHD on the prevalence of fractures in KIMS patients are shown in Fig. 3. There were no significant differences in the prevalence of fractures in patients receiving glucocorticoid, L-thyroxine, estrogen, or testosterone (data not shown) replacement therapy compared with KIMS patients without these hormonal deficiencies.

GROWTH HORMONE DEFICIENCY AND FRACTURE RISK

FIG. 1. Comparisons of the prevalence of (A) all fractures in EVOS participants and in KIMS patients over the age of 60 years and (B) all fractures in all naive and nonnaive KIMS patients and of fractures of the radius in naive and nonnaive patients and in patients with adultonset (AO) and childhood-onset (CO) disease.

Diabetes insipidus (antidiuretic hormone deficiency) was associated with a significantly higher fracture risk in men, but not in women (Fig. 4). However, overall, the proportion of patients with low BMD was similar in the diabetes insipidus and nondiabetes insipidus groups (14% vs. 13%).

Influence of age of onset of pituitary hormone deficiency Patients with hypopituitarism of adult onset had a higher prevalence of fractures of the radius than patients with childhood-onset disease (Fig. 1B), but spinal and hip BMD Z scores were not significantly different between these

401

FIG. 2. The effect of the number of pituitary hormone deficiencies in addition to GHD on (A) the prevalence of all fractures and (B) the percentage of patients with low BMD (⬍ ⫺2 SD).

groups (spinal, ⫺0.42 ⫾ 1.67 SD for adult-onset disease vs. ⫺1.20 ⫾ 1.62 SD for childhood-onset disease; hip ⫺0.35 ⫾ 1.1 SD for adult-onset disease vs. ⫺1.16 ⫾ 0.7 SD for childhood-onset disease).

Other factors affecting BMD and the prevalence of fractures in KIMS Obese patients (BMI ⬎ 27 kg/m2) had a similar prevalence of fractures than nonobese patients (30% vs. 27%) although the obese group had a higher BMD at the lumbar spine than the nonobese group (BMD Z score: men, ⫺0.10 ⫾ 1.10 SD vs. ⫺1.15 ⫾ 1.35 SD, p ⬍ 0.05; women, ⫺0.15 ⫾ 1.10 SD vs. ⫺0.75 ⫾ 1.33 SD, not significant). Smoking had a significant impact on fracture risk, with nonsmokers having a lower prevalence of fractures than smokers (24.6% vs. 33.6%, p ⬍ 0.0001).

¨ STER ET AL. WU

402

compared with the nonnaive group; this difference was significant in men (Fig. 1B). Lumbar spine BMD was lower in the naive group than in the nonnaive group (BMD Z score, ⫺0.50 ⫾ 1.67 SD vs. ⫺0.13 ⫾ 1.68 SD, respectively; p ⫽ 0.06). However, the proportion of patients with Z scores below ⫺2 SD was similar in the two groups (13% vs. 15%, respectively, not significant).

DISCUSSION

FIG. 3. The effect of glucocorticoid, L-thyroxine, and estrogen replacement therapy on the prevalence of fractures in KIMS patients.

FIG. 4. Prevalence of fractures in KIMS patients with and without diabetes insipidus.

Difference between naive and nonnaive patients Patients in the naive group had a similar prevalence of fractures than those in the nonnaive group (Fig. 1B). However, this difference became statistically significant when patients ⬎50 years of age were considered (34% in naive vs. 27% in nonnaive; p ⬍ 0.05). Men ⬍30 years of age, who were classed as naive, also had a significantly higher prevalence of fractures than their nonnaive counterparts (32% in naive vs. 17% in nonnaive; p ⬍ 0.05). The prevalence of fractures of the radius also was higher in the naive group

Our report is the first large-scale analysis to show that patients with hypopituitarism and GHD have a higher prevalence of fractures than the general population. This finding is based on a comparison of age-matched data from KIMS and a non–GH-deficient population, EVOS (Fig. 1). The difference in the level of significance between men and women might be caused by a smaller-size problem based on the smaller number of KIMS women and the greater prevalence of fractures in the EVOS women compared with EVOS men. This is shown by the change in absolute fracture rates between men and women in the total versus age-matched populations (Fig. 1 vs. Fig. 4). Females have higher rates than males in the older subgroup (Fig. 1) compared with the total population (Fig. 4) in which males have higher rates then females. Overall, one limitation of this present study is the analyses of the effect of age on fracture rates caused by GHD. As the GH production and secretion rates normally decline with age, it is feasible that the effects of GHD are more apparent in younger individuals than older individuals. Because our control group (EVOS) is older, this might tend to minimize the true impact of GHD on fracture risks. In agreement with other studies,(2,5,16,31) our results suggest that other pituitary hormone deficiencies do not have as great an impact on fracture rate as untreated GHD. Posterior pituitary damage, identified by the presence of diabetes insipidus, was shown to be an additional risk factor for fracture in men only. Of particular interest is the finding that patients classed as naive (i.e., those who had not received GH therapy for at least 6 months before enrolling into KIMS) appeared to be at greater risk of fracture than patients classed as nonnaive. To our knowledge, this is the first evidence to suggest that GH replacement therapy can reduce fracture risk. Further studies would be required to confirm this beneficial effect of GH therapy. However, it should be noted that differences between naive and nonnaive patients with regards to homogeneity might partially explain the differences in fracture rates. However, naive and nonnaive patients may be different in many respects, especially because the nonnaive patients had been those previously included into clinical trials. Those patients had to fulfill special inclusion criteria and had to agree to take part in a clinical trial, which by itself makes them different from other patients. We therefore thought that this might be a nonmeasurable bias potentially influencing the effects on fracture rates. Ideally, a randomized, placebo-controlled study should be used to evaluate the effect of GH replacement therapy on clinical outcome in patients with hypopituitarism. In the

GROWTH HORMONE DEFICIENCY AND FRACTURE RISK

case of fracture risk, such a study would require a large number of patients and a long duration. For example, Black and coworkers(32) studied over 2000 patients for 3 years to evaluate the effect of bisphosphonate treatment on fracture risk in women with previous vertebral fractures. Recruiting such a large number of patients with hypopituitarism is unlikely to be feasible. Given these difficulties, the value of pharmacoepidemiological surveys such as KIMS is clear. By using data from KIMS, we were able to establish a minimum estimate of the prevalence of fractures in a large group of patients with hypopituitarism and GHD and to assess many factors that could influence fracture rates. One limitation of our study was the retrospective nature of fracture assessment. The prevalence of vertebral fractures, the most common fractures in osteoporosis, is difficult to assess retrospectively, because these fractures are difficult to diagnose and often go unnoticed. Therefore, the reported prevalence of fractures in KIMS is a minimum estimate, and so the difference in fracture rates we describe between KIMS and EVOS also is a minimum. In addition, interpretation of the bone density data must be made with care because of some limitations of the analysis. Bone densities were available from a relatively small number of patients and it is possible that in many of these cases BMD was measured because osteoporosis was suspected. Furthermore, a number of different devices were used to measure BMD in KIMS patients and these generally employ different reference values. National differences in reference values also may complicate the assessment of BMD. Two previous studies involving much smaller patient numbers than the present study have shown that fracture frequency is increased in patients with hypopituitarism.(10,11) Because these previous studies only involved approximately 100 patients, the results are open to criticism. Our analysis, based on 264 patients with hypopituitarism and GHD over the age of 60 years, confirms an increased fracture risk, compared with healthy individuals of a similar age; this increase may be almost 3-fold. Furthermore, because the fracture prevalence in KIMS patients older than 60 years did not differ greatly from the fracture prevalence in all 2024 patients enrolled in KIMS, it is reasonable to assume that hypopituitarism is associated with an even higher relative risk of fracture in younger patients with GHD compared with the normal population. It is possible that the increased prevalence of fractures in KIMS patients is not caused by endocrine factors directly. Patients with hypopituitarism have a higher than normal prevalence of impaired vision and use psychotropic agents to a greater extent than the general population; this might have resulted in a higher frequency of falling, a factor that was not assessed by the questionnaire.(33–35) The number of pituitary hormone deficiencies other than GHD did not influence fracture risk or bone density, which suggests that, assuming an endocrine cause, GHD itself has a significant impact on bone and may be considered as a risk factor for osteoporosis. Similar conclusions have been made previously.(2,5,31) Interestingly, patients receiving glucocorticoids did not have a significantly higher prevalence of fractures, suggesting that the doses used were not excessive.

403

The use of other pituitary hormone replacement therapies also had little effect on the prevalence of fractures. In our study, patients with adult-onset hypopituitarism had more fractures of the radius than those with childhoodonset disease, although there was an apparent lack of significant differences in BMD, which might be simply a sample size problem. One reason for these apparently conflicting results could be the use of DXA to measure BMD. Bone density, as measured by DXA, is based on a twodimensional image, but this areal BMD is dependent on bone volume; as bone grows in childhood, areal BMD increases.(36,37) Patients with hypopituitarism of childhood onset are likely to be short for their age because of suboptimal GH replacement therapy early in life. Thus, areal BMD measurements in these patients are expected to be low compared with the age- and sex-matched reference values used to calculate Z scores. An alternative explanation may be that BMD is not as good a marker of fracture risk as previously thought in young patients.(38) Many authors suggest that bone quality (i.e., the strength of bone) is dependent on other factors, such as architecture and microdamage,(39,40) in addition to mass. Conflicting bone density and fracture results also were found in obese and nonobese patients. Again, DXA methodology may explain this anomaly because fat tissue can affect the estimation of BMD using this technique.(41) Alternatively, the nonobese patients in KIMS may have had improved bone quality when compared with the obese patients, despite having lower bone density. We expected that patients with diabetes insipidus would have a higher prevalence of fractures than other patients with hypopituitarism because this condition is associated with more extensive pituitary damage and may be a marker for visual impairment. Such patients might have an increased risk of falling. In addition, it may have been expected that diabetes insipidus would be associated with a greater prevalence of epilepsy and, thus, a greater risk of falling. However, there was no difference in the use of antiepileptic drugs between the patients with and those without diabetes insipidus. Our analysis showed that diabetes insipidus was associated with a significantly increased fracture rate in men only. The reason for this apparent difference between the sexes is not clear and needs further investigation. Our analysis also highlighted the fact that smoking can have a detrimental effect on bone. In the present study, patients in the naive group had a lower mean Z score than patients in the nonnaive group, but similar proportions of these patients had Z scores below ⫺2 SD. Previous assessments of GH replacement therapy have shown that at least 2 years’ treatment is required to show significant improvements in bone mass.(17–20,22,23,25) Because the mean duration of GH treatment in the naive group was only 1.8 years, we did not expect to see a significant difference in BMD between the naive and nonnaive groups. The difference in the prevalence of fractures between the naive and nonnaive groups is an important result. The data suggest a favorable treatment effect, particularly in men, but analyses involving longer treatment periods are required to confirm this. Another explanation could be that the increase

¨ STER ET AL. WU

404

in muscle mass associated with GH therapy may help to protect bones from fracture.(25) In conclusion, our analysis shows that patients with hypopituitarism and GHD are at high risk of experiencing fractures. Furthermore, it would appear that this increased risk is mainly attributable to GHD, assuming a direct endocrine cause. This study is, to our knowledge, the first to report that GH replacement therapy may reduce the risk of fracture in patients with hypopituitarism and GHD.

REFERENCES 1. Johansson AG, Burman P, Westermark K, Ljunghall S 1992 The bone mineral density in acquired growth hormone deficiency correlates with circulating levels of insulin-like growth factor I. J Intern Med 232:447– 452. 2. Kaufman J-M, Taelman P, Vermeulen A, Vandeweghe M 1992 Bone mineral status in growth hormone-deficient males with isolated and multiple pituitary deficiencies of childhood onset. J Clin Endocrinol Metab 74:118 –123. 3. Bing-You RG, Denis M-C, Rosen CJ 1993 Low bone mineral density in adults with previous hypothalamic–pituitary tumors: Correlations with serum growth hormone responses to GHreleasing hormone, insulin-like growth factor I and IGF binding protein 3. Calcif Tissue Int 32:326 –330. 4. Rose´n T, Hansson T, Granhed H, Szucs J, Bengtsson B-Å 1993 Reduced bone mineral content in adult patients with growth hormone deficiency. Acta Endocrinol (Copenh) 129: 201–206. 5. de Boer H, Blok GJ, van Lingen A, Teule GJ, Lips P, van der Veen EA 1994 Consequences of childhood-onset growth hormone deficiency for adult bone mass. J Bone Miner Res 9:1319 –1326. 6. Holmes SJ, Economou G, Whitehouse RW, Adams JE, Shalet SM 1994 Reduced bone mineral density in patients with adult onset growth hormone deficiency. J Clin Endocrinol Metab 78:669 – 674. 7. Wu¨ster Chr 1995 Growth hormone, insulin-like growth factors and bone metabolism. Endocrinol Metab 2:3–12. 8. Melton LJ III, Atkinson EJ, O’Fallon WM, Wahner HW, Riggs BL 1993 Long-term fracture prediction by bone mineral assessed at different skeletal sites. J Bone Miner Res 8:1227– 1233. 9. Wasnich R 1993 Bone mass measurement: Prediction of risk. Am J Med 95:6S–10S. 10. Wu¨ster C, Slenczka E, Ziegler R 1991 Erho¨hte pra¨valenz von osteoporose und arteriosklerose bei konventionell substituierter hypophysenvorderlappeninsuffizienz: Bedarf einer zusa¨tzlichen wachtumshormonsubstitution? Klin Wochenschr 69: 769 –773. 11. Rose´n T, Wilhelmsen L, Landin-Wilhelmsen K, Lappas G, Bengtsson B-Å 1997 Increased fracture frequency in adult patients with hypopituitarism and GH deficiency. Eur J Endocrinol 137:240 –245. 12. Eastell R, Reid DM, Compston J, Cooper C, Fogelman I, Francis RM, Hosking DJ, Purdie DW, Ralston SH, Reeve J, Russell RG, Stevenson JC, Torgerson DJ 1998 A UK Consensus Group on management of glucocorticoid-induced osteoporosis: An update. J Intern Med 244:271–292. 13. Richelson LS, Wahner HW, Melton LJ III, Riggs BL 1984 Relative contributions of aging and estrogen deficiency to post-menopausal bone loss. N Engl J Med 311:1273–1275. 14. Jackson JA Kleerekoper M 1990 Osteoporosis in men: Diagnosis, pathophysiology, and prevention. Medicine 69:137– 152. 15. Wolinsky-Friedland M 1995 Drug-induced metabolic bone disease. Endocrinol Metab Clin North Am 24:395– 420.

16. Wu¨ster C 1993 Growth hormone and bone. Acta Endocrinol 128(Suppl 2):14 –18. 17. Vandeweghe M, Taelman P, Kaufman JM 1993 Short and long-term effects of growth hormone treatment on bone turnover and bone mineral content in adult growth hormonedeficient males. Clin Endocrinol (Oxf) 39:409 – 415. 18. Baum H, Biller B, Finkelstein J, Baker Carnistraro K, Oppenheim D, Schonfeld D, Hoskins Michel T, Wittink H, Klibanski A 1996 Effects of physiologic growth hormone therapy on bone density and body composition in patients with adultonset GH deficiency. Ann Intern Med 125:883– 890. 19. Johannsson G, Rose´n T, Bosaeus I, Sjo¨stro¨m L, Bengtsson B-Å 1996 Two years of growth hormone (GH) treatment increases bone mineral content and density in hypopituitary patients with adult-onset GH deficiency. J Clin Endocrinol Metab 81:2865–2873. 20. Bravenboer N, Holzmann P, de Boer H, Roos JC, van der Veen EA, Lips P 1997 The effect of growth hormone (GH) on histomorphometric indices of bone structure and bone turnover in GH-deficient men. J Clin Endocrinol Metab 82:1818 –1822. 21. Ter Maaten JC, de Boer H, Roos JC, Lips P, van der Veen EA 1997 Long-term effects of growth hormone treatment on bone density. Endocrinol Metab 4(Suppl A):8. 22. Kann P, Piepkorn B, Schehler B, Andreas J, Lotz J, Prellwitz W, Beyer J 1998 Effect of long-term treatment with GH on bone metabolism, bone mineral density and bone elasticity in GH-deficient adults. Clin Endocrinol (Oxf) 48:561–568. 23. Rahim A, Holmes SJ, Adams JE, Shalet SM 1998 Long-term change in the bone mineral density of adults with adult onset growth hormone (GH) deficiency in response to short or longterm GH replacement therapy. Clin Endocrinol (Oxf) 48:463– 469. 24. Rosen C, Wu¨ster C 1996 Growth hormone, insulin-like growth factors—potential applications and limitations in the management of osteoporosis. In: Marcus R, Feldman D (eds.) Osteoporosis. Academic Press, San Diego, CA, USA, pp. 1313– 1333. 25. Wu¨ster C, Ha¨rle U, Rehn U, Mu¨ller C, Knauf K, Schwabe C, Ziegler R 1998 Benefits of growth hormone treatment on bone metabolism, bone density and bone strength in growth hormone deficiency and osteoporosis. Growth Horm IGF Res 8(Suppl A):87–94. 26. Carroll PV, Christ ER, Bengtsson B-Å, Carlsson L, Christiansen JS, Clemmons D, Hintz R, Ho K, Laron Z, Sizonenko P, So¨nksen PH, Tanaka T, Thorner M 1998 Growth hormone deficiency in adulthood and the effects of growth hormone replacement: A review. J Clin Endocrinol Metab 83:382–395. 27. Abs R, Bengtsson B-Å, Hernberg-Ståhl E, Monson J, Tauber JP, Wilton P, Wu¨ster Chr 1999 GH replacement in 1034 growth hormone deficient adults: Demographic and clinical characteristics, dosing and safety. Clin Endocrinol (Oxf) 50: 703–713. 28. Bengtsson BÅ, Abs R, Bennmarker H, Monson J, FeldtRasmussen U, Hernberg-Ståhl E, Westberg B, Wilton P, Wu¨ster Chr 1999 The effects of treatment and the individual responsiveness to growth hormone (GH) replacement therapy in 665 GH-deficient adults. J Clin Endocrinol Metab 84:3929 – 3935. 29. Bengtsson B-Å, Koppeschaar HPF, Abs R, Bennmarker H, Hernberg-Ståhl E, Westberg B, Wilton P, Monson JP, FeldtRasmussen U, Wu¨ster Chr 1999 Growth hormone replacement therapy is not associated with any increase in mortality. J Clin Endocrinol Metab 84:4291– 4292. 30. Scheidt-Nave C, Ziegler R, Raspe H 1998 Epidemiologie der Osteoporose. Med Klin 93(Suppl 2):7–11. 31. Bachrach LK 1998 Bone mineral, histomorphometry, and body composition in adults with growth hormone receptor deficiency. J Bone Miner Res 13:415– 421.

GROWTH HORMONE DEFICIENCY AND FRACTURE RISK 32. Black DM, Cummings SR, Karpf DB, Cauley JA, Thompson DE, Nevitt MC, Bauer DC, Genant HK, Haskell WL, Marcus R, Ott SM, Torner JC, Quandt SA, Reiss TF, Ensrud KE, for the Fracture Intervention Trial Research Group 1996 Randomised trial of effect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 348:1535–1541. 33. Felson DT, Anderson JJ, Hannan MT, Milton RC, Wilson PW, Kiel DP 1989 Impaired vision and hip fracture. J Am Geriatr Soc 37:495–500. 34. Dargent-Molina P, Favier F, Grandjean H, Baudoin C, Schott AM, Hausherr E, Meunier PJ, Bre`art G 1996 Fall-related factors and risk of hip fractures: The EPIDOS prospective study. Lancet 348:145–149. 35. Cumming RG 1998 Epidemiology of medication-related falls and fractures in the elderly. Drugs Aging 12:43–53. 36. Glastre C, Braillon P, David L, Cochat P, Meunier PJ, Delmas PD 1990 Measurement of bone mineral content of the lumbar spine by dual energy x-ray absorptiometry in normal children: correlations with growth parameters. J Clin Endocrinol Metab 70:1330 –1333. 37. Miller JZ, Slemenda CW, Meaney FJ, Reister TK, Hui S, Johnston CC 1991 The relationship of bone mineral density and anthropometric variables in healthy male and female children. Bone Miner 14:137–152. 38. Wang XD, Masilamani NS, Mabrey JD, Alder ME, Agrawal CM 1998 Changes in the fracture toughness of bone may not

405

be reflected in its mineral density, porosity, and tensile properties. Bone 23:67–72. 39. Cooper C 1993 The epidemiology of fragility fractures: Is there a role for bone quality? Calcif Tissue Int 53(Suppl 1):S23–S26. 40. Schnitzler CM 1993 Bone quality: A determinant for certain risk factors for bone fragility. Calcif Tissue Int 53(Suppl 1):S27–S31. 41. Manzoni P, Brambilla P, Pietrobelli A, Beccaria L, Bianchessi A, Mora S, Chiumello G 1996 Influence of body composition on bone mineral content in children and adolescents. Am J Clin Nutr 64:603– 607.

Address reprint requests to: Professor Christian Wu¨ster, M.D. Department of Internal Medicine 1 Endocrinology and Metabolism University Medical Clinic Heidelberg Karlstrasse 41 D-65185 Wiesbaden, Germany Received in original form March 15, 2000; in revised form August 25, 2000; accepted September 21, 2000.