Down-Regulation of the Serum Stimulatory Components of the Insulin-like Growth Factor (IGF) System (IGF-I, IGF-II, IGF Binding Protein [BP]-3, and IGFBP-5) in Age-Related (Type II) Femoral Neck Osteoporosis†
Leuven University Center for Metabolic Bone Diseases, Katholieke Universiteit Leuven, Leuven, Belgium
Arthritis and Metabolic Bone Disease Research Unit, Katholieke Universiteit Leuven, Leuven, Belgium
Laboratory for Experimental Medicine and Endocrinology, Katholieke Universiteit Leuven, Leuven, Belgium
The data have been partially presented at the 1996 Annual Meeting of the American Society for Bone and Mineral Research, Seattle, Washington, U.S.A.
Both a decrease in bone formation and an increase in bone resorption have been implicated in the pathogenesis of age-related (type II) femoral neck osteoporosis. While the increase in the bone resorption rate has been shown to be partially related to secondary hyperparathyroidism, the mechanisms underlying the decline in bone formation have not yet been identified. The aim of the present study was to test the hypothesis that the bone formation deficit associated with type II osteoporosis might be due to secondary hyperparathyroidism and/or to a deficiency of the insulin-like growth factor (IGF) system. Circulating concentrations of IGF-I, IGF-II, IGF binding protein (IGFBP)-3, IGFBP-4, IGFBP-5, 25-hydroxycholecalciferol (25(OH)D3), and intact parathyroid hormone (PTH) were measured in 50 elderly women after sustaining a hip fracture and in 50 healthy age-matched controls. In addition, serum levels of osteocalcin (OC), skeletal alkaline phosphatase, and N-terminal procollagen peptide and urinary pyridinium cross-links were determined as markers of bone remodeling, and bone mineral density (BMD) was assessed at the proximal femur. In the patient group, serum was drawn within 18 h of the fracture and prior to surgery. Circulating protein concentrations did not change over this time frame. No difference was found between mean IGFBP-4 serum levels in the two groups studied, while mean levels of IGF-I, IGF-II, IGFBP-3, IGFBP-5, 25(OH)D3, and markers of bone formation were significantly lower (p < 0.006) in patients as compared with healthy subjects. Serum PTH and urinary pyridinium cross-links, however, were markedly increased (p < 0.001) in the osteoporotic group. In pooled data from the normal and osteoporotic populations, age-adjusted multiple regression models based on IGF-I, IGF-II, IGFBP-3, and IGFBP-5 were found to be highly predictive of serum OC (R2 = 19%, p < 0.001) and BMD of femoral neck (R2 = 49%, p < 0.0001), consistent with an effect of the anabolic IGF components on overall bone formation rate. Similar models based on 25(OH)D3 and PTH, however, were statistically unrelated to OC. To address further the potential impact of trauma on circulating IGF system components, we measured IGF system component levels in 10 male patients within 18 h following tibial fracture and in 10 age-matched normal male subjects. There was no significant difference in serum level of any of the IGF system components between the two groups. Although limited by its cross-sectional design, the present study suggests that, in addition to bone resorption resulting from secondary hyperparathyroidism, impaired bone formation associated with deficiency of the IGF system might predispose elderly women to fragility fracture of the proximal femur.
Fractures of the proximal femur are a significant cause of frailty, morbidity, and even mortality, particularly in the rapidly growing segment of women over the age of 65 years.(1) Among other factors, lack of 25-hydroxycholecalciferol (25(OH)D3) has been implied in the induction of secondary hyperparathyroidism of the elderly and in turn to bone resorption and fracture risk.(2,3) Previous studies have indeed provided evidence for an age-related increase in circulating parathyroid hormone (PTH)(4,5) and urinary pyridinium cross-links(5) among elderly osteoporotic women. Moreover, substitution with calcium and cholecalciferol has been documented to reduce PTH-mediated bone resorption(6) and to lower the frequency of hip fractures in selected elderly women.(7) In addition to this increase in the overall resorption rate, a decrease in bone formation has also been implicated in the pathogenesis of the progressive loss of bone that occurs with age.(8,9) This conclusion is based on strong histologic evidence for an age-associated decline in the wall thickness of the remodeling site,(10–12) reflecting the inability of osteoblasts to reconstitute adequately the previously resorbed bone in the individual remodeling units. However, the mechanisms responsible for this defect in osteoblastic function and the resulting remodeling imbalance have not yet been identified.
One of these mechanisms could be secondary hyperparathyroidism itself, inasmuch as animal studies have provided evidence for impaired bone formation in calcium depletion–induced hyperparathyroidism.(13,14) However, uncoupling of formation to resorption might also be related to the age-associated decline in serum insulin-like growth factor (IGF)-I and IGF-II,(15,16) endocrine factors considered to be anabolic for bone tissue.(17,18) In vitro, both IGF-I and IGF-II have indeed been shown to stimulate human osteoblast proliferation and differentiation.(19) Similarly, in vivo administration of IGF-I to elderly women increases biochemical markers of osteoblast function.(20,21) In serum and body fluids, IGFs are known to be bound to binding proteins (IGFBPs).(22) Six structurally distinct but homologous IGFBPs have been identified and are referred to as IGFBP-1 through IGFBP-6. Various lines of evidence suggest that these IGFBPs may be involved in targeting IGF to specific cell types(23,24) and in modulating the bioavailability of locally secreted IGFs to IGF target cells.(25,26) In this regard, both IGFBP-3(27) and IGFBP-5(28) have been shown to potentiate the mitogenic action of IGFs on bone cells. IGFBP-4, however, has an inhibitory effect on IGF-I activity in bone cultures.(29) The fact that IGFBPs modulate IGF actions either positively or negatively supports the concept that the balance between the stimulatory and inhibitory classes of IGFBPs will determine the degree and extent of IGF-induced cellular responses in target tissues such as bone.
The aim of the present study was to test the hypothesis that the bone formation deficit associated with type II osteoporosis might be due to secondary hyperparathyroidism and/or to a decrease in IGF action. Such a decrease could be due to a deficit of the serum stimulatory IGF system components IGF-I, IGF-II, IGFBP-3, or IGFBP-5 and/or to increased circulating levels of the inhibitory component IGFBP-4. To this end, we assessed the 25(OH)D3–PTH axis and we measured systemic components of the IGF system in 50 women after sustaining a hip fracture and in an equal number of elderly nonfractured controls. In addition, formation and resorption of bone were assessed by different markers of bone remodeling.(30) To minimize the potentially confounding effects of the trauma and the subsequent treatment, samples from osteoporotic patients were obtained within 18 h after fracture, prior to surgery.
MATERIALS AND METHODS
The study was conducted in 50 patients with hip fracture (mean age 74 years, range 60–95 years) and 50 elderly controls (mean age 75.8 years, range 70–90 years). Time since menopause (26.4 ± 6.8 vs. 25.4 ± 7.4 years) was not different between the control and fracture groups. Both fracture cases and control subjects were sampled throughout the same period (September–December). Informed consent was obtained from all patients and controls, and all procedures were approved by the institutional ethical committee.
Women admitted to the Department of Traumatology following a subcapital fracture of the proximal femur were recruited consecutively in the province of Vlaams-Brabant, an area with 500,000 inhabitants in Belgium. To be eligible for participation, women had to be over 60 years of age, to be previously ambulatory, and to have suffered a fall resulting in a radiologically confirmed first hip fracture. All osteoporotic patients were studied before surgery and within 18 h after fracture. Patients were excluded if they met any of the following criteria: having been admitted with a pathologic fracture or fracture resulting from trauma other than a fall; having sustained a previous hip fracture; hemodynamic instability and/or need for intravenous fluids or blood products prior to surgery; nonosteoporotic metabolic bone disease; diabetes or thyroid disease, whether controlled or uncontrolled; calcium, fluoride or vitamin D supplements; or history of use of thiazides, glucocorticoids (≥5 mg of prednisone or equivalent per day), estrogens (or estrogen-related drugs), anabolic steroids or calcitonins for more than 3 months. The study population of 50 patients represented 72% of the women admitted to the hospital following a subcapital fracture of the proximal femur over the 4-month recruitment period.
An age-matched sample of 50 control subjects was selected from a study population of 245 healthy elderly women recruited from 50 general practices in the province of Brabant, Belgium. To be eligible for participation, women had to be over 70 years of age, noninstitutionalized, functionally independent, and free from diseases and medications known to affect the musculoskeletal system or the somatotrophic axis. Details of the recruitment of the subjects have been described previously.(31) The 50 control subjects selected for this study were representative of the overall healthy study population of 245 individuals in terms of body mass index (BMI), IGF-I and IGF-II, 25(OH)D3, PTH, and bone density.
To address further the potential impact of trauma on circulating IGF system components, additional serum analyses were performed in 10 male patients within 18 h following a fracture of the tibia and in an equal number of age- and gender-matched healthy controls.
Measurements were made of height and body weight. BMI was calculated as body weight divided by height squared (kg/m2).
Fasting blood samples were collected in the morning from all subjects. In the patients with a fracture of the proximal femur or tibia, samples were obtained before surgical treatment and within 18 h after fracture. Total serum calcium, inorganic phosphate, albumin, and creatinine were determined by standard analytical methods. Creatinine clearance was estimated according to Cockroft and Gault, relying on serum creatinine, weight, and age.(32) 25(OH)D3 (reference range 10–60 ng/ml) was measured by a previously reported competitive binding assay.(33) Serum intact PTH (reference range 3–40 pg/ml) was measured by a two-step immunochemical method, involving an amino-terminal capture and a midregional detecting antibody, as described previously.(34) Human osteocalcin (OC) was determined by a previously reported radioimmunoassay (RIA).(35) Serum levels of skeletal alkaline phosphatase were determined by the heat inactivation method.(36) The sensitivity of the assay was <0.2 U/l and averaged intra-assay coefficient of variation (CV) was 6.2 U/l and interassay variation (CV) was 10.4 U/l. Measurement of amino-terminal propeptide of human type-I procollagen was performed by a RIA that recognizes only an intact form of the propeptide.(37) The sensitivity of the RIA is 2 μg/l, and the averaged intra-assay and interassay CV reported by the manufacturer was 8.75 and 5.05, respectively. Pyridinium cross-links (pyridinoline and deoxypyridinoline), corrected for creatinine, were measured on hydrolized urine extract by fluorescent detection after high-pressure liquid chromatography as previously described.(38) IGF-I and IGF-II were measured by RIAs after separation of IGFBPs, as previously described.(39) Because IGFBPs produce artifacts in IGF radioligand assays, it is essential to separate completely the IGFBPs from the IGFs in order for the IGF determinations to be valid. The separation of the IGFBPs from IGFs was achieved by a recently developed and validated rapid acid gel-filtration protocol.(39) To measure serum levels of IGFBP-3, IGFBP-4, and IGFBP-5, we recently developed valid RIAs, using recombinant human IGFBP-3, IGFBP-4, and IGFBP-5 as antigens, tracer, and standard.(40,41) None of the other purified IGFBPs exhibited significant cross-reactivity in any of these assays.
Both women with fractured neck of femur (within 1 week after surgery) and control subjects were scanned using dual-energy X-ray absorptiometry of the femoral neck region and the trochanteric area. Areal BMD was measured using the Lunar DPX-L scanner (Lunar Radiation, Inc., Madison, WI, U.S.A.). Standard positioning was used with anterior–posterior scanning of the right proximal femur, except in the event of hip replacement when the left femur was scanned. The precision of femoral BMD measurements in elderly women using our dual-energy X-ray absorptiometry equipment is 3.1% at the neck and 2.6% at the trochanter.(31)
Differences between patients and controls were evaluated with Student's t-test. The relationship between the biochemical parameters were evaluated by calculating Pearson (r) correlation coefficients, based on logarithmic transformations of IGF-I, IGFBP-4, 25(OH)D3, PTH, and OC. In view of the fact that no normalizing transformation was found for the parameters time (the interval elapsed after fracture) and age, Spearman rank correlation (rs) was used to assess the effect of these parameters on the biochemical variables. Age-adjusted multiple regression models were constructed with serum OC or femoral BMD as response, and the stimulatory components of the IGF system (IGF-I, IGF-II, IGFBP-3, and IGFBP-5) as regressors. All statistical analyses were conducted with the use of SAS (Statistical Analysis Systems, Inc., Cary, NC, U.S.A.). Reported p values are two-sided. The nominal significance level was set at 0.05.
Characteristics of the population
Clinical and biochemical data from the elderly controls and the hip fracture patients are indicated in Table 1. According to the 95% reference interval of a healthy young reference group,(34) 40% of the fracture patients had elevated levels of PTH (serum PTH >40 pg/ml) compared with none of the elderly controls. Vitamin D deficiency, as evidenced by serum 25(OH)D3 levels below 12 ng/ml,(42) was observed in 64% of the osteoporotic patients compared with 8% of the control subjects. Within the 4-month sampling period, no relationship was observed between the month of specimen collection and vitamin D status (data not shown).
Table Table 1. Subject Characteristics of the Elderly Women Participating in the Study and Results of the Biochemical and Bone Density Measurements
Differences in biochemical and bone density parameters between patients and controls
A comparison of both groups is presented in Table 1. No significant differences were observed for mean age between patients and controls. Osteoporotic patients, however, had a lower BMI. Serum albumin was significantly lower in fractured women as well. Total calcium and phosphate levels, serum creatinine, and calculated creatinine clearance were not statistically different. Serum concentrations of 25(OH)D3 were lower in patients than in the control subjects. In contrast, intact PTH concentrations were significantly increased in the fracture group. Osteoporotic patients had marked decreases in serum levels of IGF-I, IGF-II, IGFBP-3, and IGFBP-5. These decreases remained highly significant for all anabolic IGF components after adjusting for BMI, 25(OH)D3, and PTH (p = 0.001–0.002). Similarly, significant differences in IGF-I and IGF-II persisted after adjustment for IGFBP-3 by calculating the ratio of IGF to IGFBP-3 (p = 0.04 and p < 0.001, respectively). In contrast to the anabolic components of the IGF system, mean circulating levels of IGFBP-4 were not statistically different. Finally, there was a decrease of the different markers of bone formation and femoral BMD in hip fracture patients compared with elderly controls, while the resorption markers pyridinoline and deoxypyridinoline were significantly increased.
Age-associated changes in serum IGF components, 25(OH)D3, and PTH
In the control group, serum levels of IGF-I (rs = −0.28, p = 0.04), IGF-II (rs = −0.55, p = 0.0001), IGFBP-3 (rs = −0.43, p = 0.002), and IGFBP-5 (rs = −0.44, p = 0.001) decreased significantly as a function of age. No significant relationship, however, was observed between age and IGFBP-4 (rs = −0.24, p = 0.09). In contrast, fracture patients showed an age-associated increase in IGFBP-4 (rs = 0.33, p = 0.02), whereas IGF-I (rs = −0.23, p = 0.10), IGF-II (rs = −0.17, p = 0.24), IGFBP-3 (rs = −0.17, p = 0.23), or IGFBP-5 (rs = −0.17, p = 0.24) did not change with age. No significant age-dependent changes were observed in circulating 25(OH)D3 (rs = −0.25, p = 0.07 and rs = −0.25, p = 0.08 in controls and patients, respectively) or serum PTH (rs = 0.28, p = 0.05 and rs = −0.01, p = 0.90, respectively).
Relationship between the IGF system and the 25(OH)D3–PTH axis
The correlation matrices of the IGF components and the 25(OH)D3–PTH(1–84) axis in control subjects and fracture patients are shown in Table 2 and Table 3, respectively. In both groups, serum concentrations of IGF-I, IGF-II, IGFBP-3, and IGFBP-5 were highly positively interrelated. Circulating IGFBP-4, however, was unrelated to the other components of the IGF system. Serum PTH correlated inversely with 25(OH)D3 in patients but not in controls, presumably reflecting the adequate vitamin D status in the control group.(35) In both groups, serum PTH was found to be positively related to circulating IGFBP-4 (Fig. 1), whereas no relationship was observed between PTH and the anabolic components of the IGF system. Serum 25(OH)D3, however, was negatively related to IGFBP-4 in patients, but this association did not persist after adjusting for PTH(1–84) (data not shown). No significant correlations were observed between 25(OH)D3 and any of the anabolic IGF components in either group, except for a positive relation with IGFBP-5 in control subjects.
Table Table 2. Correlation Matrix Between IGF System Components, 25(OH)D3, and PTH(1–84) in Control Subjects
Table Table 3. Correlation Matrix Between IGF System Components, 25(OH)D3, and PTH(1–84) in Fracture Patients
Relationship between the 25(OH)D3–PTH axis and (deoxy)pyridinoline
In the hip fracture group, 25(OH)D3 was inversely related to pyridinoline (r = −0.40, p = 0.008) and deoxypyridinoline (r = −0.42, p < 0.001), while PTH correlated positively with both resorption markers (r = 0.32, p = 0.04 and r = 0.35, p = 0.03). In contrast, no statistically significant relationships were observed between the 25(OH)D3–PTH axis and (deoxy)pyridinoline in the control group (data not shown).
Relationship between the IGF system components and BMI
Both in the elderly controls and in the fracture patients, IGF-I (r = 0.33, p = 0.02 and r = 0.31, p = 0.03, respectively) and IGFBP-3 (r = 0.34, p = 0.02 and r = 0.33, p = 0.02, respectively) were found to be positively associated with BMI. No other statistically significant correlations were noted (data not shown).
Effect of time elapsed since fracture
There was no significant correlation between the time elapsed after fracture and serum albumin (rs = −0.15, p = 0.29), IGF-I (rs = 0.08, p = 0.56), IGF-II (rs = 0.12, p = 0.39) or any of the IGFBPs (rs = 0.02–0.10, p = 0.85–0.46). Similarly, serum OC (rs = 0.13, p = 0.37) did not correlate with the time elapsed between sustaining the fracture and the sampling.
Prediction of markers of bone formation and femoral BMD by the anabolic components of the IGF system
Considering the study population as a whole (n = 100), multiple regression models based on IGF-I, IGF-II, IGFBP-3, and IGFBP-5 were found to be highly predictive of serum OC (R2 = 19%, p < 0.001) and BMD at the femoral neck (R2 = 49%, p < 0.001) (Table 4) or the trochanteric region (data not shown). Similar predictive results were obtained when skeletal alkaline phosphatase or N-terminal procollagen peptide were included into these models instead of OC (data not shown). Adjustment for BMI did not alter these findings nor did the exclusion of individuals with impaired renal function (data not shown). Similar relationships were statistically not significant when separate (potentially underpowered) analyses were performed on fracture patients and elderly controls (data not shown), emphasizing the exploratory character of the regression analyses on pooled data.
Table Table 4. Age-Adjusted Multiple Regression Models of the Anabolic IGF Components on Serum Osteocalcin and Femoral Neck BMD, Respectively, in Pooled Data
Prediction of serum OC and femoral BMD by the 25(OH)D3–PTH axis
No significant relationships were observed between the 25(OH)D3–PTH axis and serum OC or femoral BMD in fracture patients and elderly controls when analyzed separately (data not shown). In multiple regression analyses on pooled data, models based on 25(OH)D3 and PTH were statistically unrelated to serum OC (R2 = 3%, p = 0.11) but highly predictive of femoral neck BMD (R2 = 32%, p < 0.001).
Differences in IGF system component levels between tibia fracture patients and age- and gender-matched healthy controls
To assess that the down-regulation of serum stimulatory components of the IGF system is related to osteoporosis and not to the stress induced by fracture, we evaluated serum IGF system component levels in 10 healthy subjects within 18 h of tibial fracture and 10 age-matched normal subjects. No significant differences were observed for any or the IGF system components between patients and controls (Table 5).
Table Table 5. Circulating IGF System Components in Healthy Men Following a Traumatic Fracture of the Tibia and Age- and Gender-Matched Control Subjects
The salient findings of this study are that elderly osteoporotic women with femoral neck fracture, compared with healthy age-matched controls with normal bone density, had, in addition to the increase in bone resorption, a significant decrease in all bone formation parameters, along with a highly significant deficit of the stimulatory components of the IGF system. The depression of bone formation is surprising because, generally, an increase in resorption is followed by a corresponding increase in bone formation.(17,18) The findings of this study that all three bone formation parameters studied were decreased by 20–32%, while the bone resorption parameter was increased by 30% in fracture patients compared with controls demonstrate the large magnitude of depression of bone formation in these patients. These findings together suggest that the decrement in bone density in senile osteoporosis patients with femoral neck fracture may be a consequence of both increased bone resorption(4–5,7,43) and decreased bone formation, as demonstrated in this study.
With regard to the potential etiology for the increase in bone resorption in the femoral neck fracture patients, secondary hyperparathyroidism appears to be a primary contributing factor.(4–5,7,43) The findings that the mean serum PTH level was 3-fold higher in fracture patients compared with controls, and that the serum PTH levels showed significant positive correlation with urinary deoxypyridinoline levels, support the view that secondary hyperparathyroidism contributes to bone loss and osteoporosis in the elderly population. Many of our patients had a serum deficiency of 25(OH)D3. However, this is an unlikely explanation for the cause of the observed secondary hyperparathyroidism in its entirety, because patients without vitamin D deficiency in our study still had evidence of hyperparathyroidism and other studies have shown the presence of secondary hyperparathyroidism in elderly osteoporotic patients with normal serum 25(OH)D3 levels.(4–5,42,43)
With regard to the potential etiology for the decreased bone formation in femoral neck fracture patients, we have found significantly lower serum levels of IGF-I, IGF-II, IGFBP-3, and IGFBP-5 in fracture patients compared with age-matched controls. We speculate that the lower levels of serum stimulatory components of the IGF system observed in the present study reflect pre-existing deficiencies unrelated to short-term metabolic alterations induced by the trauma for the following reasons: recent studies have shown that serum proteins, including albumin and OC, did not change over the 18-h period between fracture and sampling(44); none of the serum IGF system components showed significant correlation with the time elapsed between sustaining fracture and the sampling; in contrast to the stimulatory IGF system components, serum level of inhibitory IGFBP-4 did not change significantly in the fracture patients, although serum levels of IGFBP-4 showed significant positive correlation with serum levels of PTH, as observed in our previous studies(16,45); and we observed no stress-induced down-regulation of the stimulatory components in healthy young subjects within 18 h after fracture of tibia. Nevertheless, we acknowledge that it is impossible to control completely the effect of trauma in a cross-sectional analysis and that confirmation would be required using a longitudinal study.
The findings that the serum level of stimulatory IGF system components show significant positive correlations with serum OC levels (R2 = 19%, p < 0.0001) and BMD (R2 = 49%, p < 0.0001) implicate a role for the deficit in the IGF system in impaired bone formation and a corresponding remodeling imbalance. The mechanisms by which the deficit in the IGF system could lead to an impairment in bone formation can only be speculated. In this regard, the decrease in serum levels of IGF system components could lead to a decrease in the endocrine actions of IGFs. The finding that systemic administration of IGF-I, which increases serum level of IGF-I, causes a marked increase in bone formation in both animals and humans(20–23) provide evidence for such endocrine actions of IGFs. In addition, we have indirect evidence that the age-related decrease in serum levels of stimulatory IGF system components may, in part, reflect decreased bone cell production (i.e., local) of stimulatory IGF system components.(16) Thus, the decrease in both local and endocrine actions of IGFs could lead to a decrease in osteoblast cell proliferation and a subsequent reduction in bone formation. Further studies are needed to establish a cause and effect relationship between IGF system deficit and reduction in bone formation in femoral neck fracture patients.
The potential causes for the significant decrease in the serum level of various stimulatory IGF system components are not known at this time. In this regard, the lack of association between the serum levels of stimulatory IGF system components and serum 25(OH)D3 or PTH suggest that the down-regulation of the IGF system in the fracture patients is independent of PTH–vitamin D axis. The findings that fracture patients had a lower BMI and lower serum albumin concentrations implicate nutritional deficiency as a cause for down-regulation of the IGF system. Poor nutrition and protein depletion are indeed common findings in hip fracture patients on admission to surgical services(46,47) and nutrition-dependent changes have been observed in both serum IGF-I and IGFBP-3.(48,49) The above findings and the findings that serum levels of IGF-I and IGFBP-3 showed significant positive correlations with BMI in both fracture and control patients are consistent with the idea that nutritional deficiency could in part contribute to the deficit in the IGF system. Nutritionally dependent changes have indeed been observed in both IGF-I and IGFBP-3.(48) Numerous studies have even provided evidence for a direct role of the IGF system in mediating changes in the anabolic response to nutrient intake,(50) suggesting that components of this system may be particularly sensitive to nutritional deprivation. However, further studies are needed to establish this relationship, since a formal assessment of nutritional status was not included in our study protocol.
We would like to emphasize that serum bone markers reflect the overall turnover rate of the skeleton and not specific parameters of bone formation. Moreover, the relationship between serum levels of bone markers and the degree of bone formation is not stoichiometric because the relationship is affected by several variables, including clearance rate, presence of multiple fragments, and reincorporation into bone matrix. These issues must be borne in mind in interpreting our bone formation marker data.
In summary, the low bone density in elderly women with femoral neck (type II) osteoporosis is associated with increased bone resorption and decreased bone formation. The mechanisms responsible for the decrease in bone formation in elderly fracture patients is unknown and this study has revealed a clue, namely down-regulation of the IGF system. The next important step will be to determine the cause of this change in the IGF system.
The expert data management by S. Breemans is gratefully acknowledged. We are grateful to T. Pham, J. Rung-Aroon, and E. Van Herck for their skillful laboratory assistance. We also would like to acknowledge J. Kirsten Ciechanowski for her help in preparing this manuscript. Dr. S. Boonen is holder of the Chair for Metabolic Bone Diseases at the Katholieke Universiteit Leuven, founded by Merck Sharp and Dohme. This work was supported by funds from the National Institutes of Health (grant AR31062) and the Jerry L. Pettis VA Medical Center.