| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Laboratory for Research of Musculoskeletal System (I.C., S.T., G.T., P.R., K.K., G.P.L.), University of Athens, KAT Hospital, 14561 Athens, Greece; Endocrinology Department (P.K.), Metaxa Hospital, 18537 Piraeus, Greece; and Metabolic Bone Disease Unit (G.S.), First Department of Orthopedics, University of Athens, Attikon Hospital, 12410 Athens, Greece
Address all correspondence and requests for reprints to: Symeon Tournis, M.D., Laboratory for Research of Musculoskeletal System "Th. Garofalidis," University of Athens, KAT Hospital, 10 Athinas Street, Kifissia, 14561, Athens, Greece. E-mail: stournis{at}med.uoa.gr.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to examine the effect of PH on quantitative and qualitative bone characteristics using peripheral quantitative computed tomography at the tibia in postmenopausal women with PH and healthy controls.
Design and Setting: We conducted a cross-sectional study at a tertiary referral center.
Patients: Fifty-two postmenopausal women with PH and 56 healthy controls, comparable for age and anthropometric measures, participated.
Intervention: There was no intervention.
Main Outcome Measure: We assessed volumetric bone mineral density (vBMD), bone mineral content (BMC), cortical thickness, cortical and trabecular area, peri- and endosteal circumference, and polar stress strength index assessed by peripheral quantitative computed tomography of the left tibia at 4% (cancellous), 14% (transition zone), and 38% (cortical) from the distal end.
Results: At 4%, there was a significant decrease of trabecular BMC and vBMD (P < 0.001), effect particularly evident in hypercalcemic patients, whereas trabecular area was comparable. At 38%, cortical BMC (P < 0.01), vBMD (P < 0.01), area (P < 0.05), and thickness (P < 0.001) were reduced in the PH group, particularly in hypercalcemic patients. Endosteal circumference increased (P < 0.001), whereas periosteal circumference was comparable, indicating cancellization of cortical bone. At 14%, polar stress strength index was significantly decreased (P < 0.01) in hypercalcemic patients, indicating impairment of bone mechanical properties.
Conclusions: Normocalcemic PH is characterized by catabolic actions at both cortical and cancellous sites (38 and 4%, respectively), an effect accentuated in hypercalcemic patients. Cortical geometric properties are adversely affected even in normocalcemic patients, whereas trabecular properties are generally preserved.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Apart from the aforementioned changes, continuous PTH excess alters several determinants of bone strength (6, 7), such as bone turnover, trabecular number, thickness, connectivity, and bone geometry. Indeed, Parfitt (6), using hand x-ray and single-photon absorptiometry, demonstrated increased net bone formation beneath the periosteum in the peripheral skeleton. These findings indicate that aBMD significantly underestimates the skeletal effects of PH.
It is well established that after PH comes to clinical attention because of hypercalcemia, biochemical and densitometric indices remain stable for at least 10 yr (3, 8). Thus, it is proposed that PH has a biphasic course (9), the earlier phase characterized by normocalcemia with elevated PTH, followed by the emergence of hypercalcemia. Another possible explanation is that the two subgroups differ in unexplored aspects of calcium handling, such as vitamin D receptor or calcium-sensing receptor polymorphisms. Indeed, no study prospectively established that patients with PH progress from a stage of normocalcemic to hypercalcemic hyperparathyroidism. Nevertheless, it is possible that even in the normocalcemic subgroup, continuous PTH excess might affect several aspects of bone strength (7), especially at cortical sites. However, there are no studies reporting changes in bone geometry in the putative different stages or subgroups of PH.
Peripheral quantitative computed tomography (pQCT) has the advantage of simultaneous separate assessments of trabecular and cortical components as well as geometric characteristics of the peripheral skeleton (10, 11). Although several groups reported pQCT data in various metabolic bone diseases, there is a limited number of reports concerning patients with PH (12, 13). Furthermore, previous studies assessed bone characteristics in both pre- and postmenopausal women at the radius (10, 12), a site subject to different mechanical forces compared with the tibia. Given that PTH modulates the response of osteoblasts to mechanical loading causing up-regulation of bone formation (14), it is possible that the tibia is a better site for assessing bone geometry and the effect of mechanical loading in patients with PH.
The present study was designed to investigate the effect of PH on quantitative and qualitative bone characteristics using pQCT at the tibia in postmenopausal women with PH and also to assess differences in skeletal disease in normocalcemic and hypercalcemic patients with PH.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Controls were healthy Caucasian postmenopausal women evaluated in our department for osteoporosis, matched for age, years after menopause, weight, and height with patients with PH. All subjects had normal plasma PTH and calcium concentrations and no identifiable cause of secondary osteoporosis. None of the subjects was receiving any specific therapy for osteoporosis. The protocol was designed according to the Declaration of Helsinki and approved by the Ethics Committee of Athens University. All subjects gave written informed consent.
Anthropometric variables and biochemical analysis
All patients and controls had a full medical history and physical examination at baseline by a trained physician. Height and weight were measured using standard procedures. Each subject reported to the laboratory after an overnight fast on two different occasions for blood sampling and 24-h urinary calcium and creatinine determinations. Serum total calcium was measured by colorimetry using a Roche Hitachi 902 analyzer (Roche, Indianapolis IN), with intra- and interassay coefficients of variation of 0.9 and 1.5%, respectively. Serum inorganic phosphate was quantified by colorimetry using a Roche Hitachi 902 analyzer, with intra- and interassay coefficients of variation of 0.9 and 1.4%, respectively. Plasma iPTH was measured by an electrochemiluminescence immunoassay (Roche). The sensitivity was 1.2 pg/ml, and the intra- and interassay coefficients of variation were 4 and 4.3%, respectively. Serum 25OHD3 levels were determined by enzyme immunoassay (Immunodiagnostic System’s OCTEIA, 25 OHD3), with a sensitivity of 5 nmol/liter and intra- and interassay coefficients of variation of 5.3 and 4.6%, respectively.
aBMD measurements by DXA
aBMD was measured by DXA using Lunar Prodigy Pro (GE Lunar Corp., Madison) at the lumbar spine (L2–L4). Data are reported for both absolute BMD values and Z- and T-scores (SD values from the mean for a sex- and age-matched and young reference population, respectively). The in vivo precision of the DXA at the spine derived from 25 postmenopausal women subjected to duplicate measurements within 1 month was 0.92 ± 0.43.
Volumetric BMD (vBMD) and geometric properties estimation by pQCT
pQCT bone mineral measurements and analyses were performed at the left tibia, using XCT-3000 device (Stratec Gmbh, Pforzheim, Germany) (16, 17). A single-energy x-ray source was used. All computed tomography scans had a slice thickness of 2.4 mm and a voxel size of 0.5 mm3. The distal end of the tibia was used as an anatomical marker; the bone cross-sectional area was imaged at 4, 14, 38, and 66% of the tibia length, proximal to this point. Analyzing each slice, vBMD (mg/cm3), corresponding bone mineral content (BMC) (mg), and cross-sectional area (CSA) (mm2) of tibia bone section were estimated as well as cortical thickness (mm), endosteal (mm) and periosteal circumference (mm), and polar stress strength index in torsion (SSIp) (mm3). Image analysis was performed using integrated software (STRATEC XCT-3000, version 5.4). The total (from the periosteum, including area of the bone and bone marrow), trabecular, and cortical bone density in milligrams per cubic centimeter and the CSA of the corresponding bone portions in square millimeters were calculated by the following procedure: 1) voxels outside the bone (soft tissue) with lower attenuation coefficients than the selected threshold (181 mg/cm3) are removed within the region of interest, and 2) the cortical and trabecular structures are separated by the areal distribution of both bone structures. By default, 55% of the outer bone area is concentrically separated and defined as cortical-subcortical region. The remaining 45% of the inner core is defined as trabecular bone. Therefore, bone marrow is included in the estimation of total and trabecular area but eliminated in the estimation of the respective BMC. To calculate pure cortical density and area without including subcortical area, all voxels within the ROI that have an attenuation coefficient below the threshold 710 mg/cm3 density are removed. Cortical thickness was defined as the mean distance between inner and outer edge of the cortical shell. SSIp lies within the theory of stability of mechanical structures against bending or torsion. From computed tomography cross-sectional images, the determination of bone strength is based on the calculation of the cross-sectional moment of inertia. Division of cross-sectional moment of inertia by the maximum distance of any voxel from the center of gravity (rmax) yields the section modulus that is directly proportional to maximum stress in bone. To take also the material properties into consideration, the section modulus is multiplied by the quotient of calculated cortical density and normal physiological cortical density of 1200 mg/cm3, yielding the calculation of SSIp. To calculate the relative decrease of trabecular vs. cortical bone, the ratio of BMC at slice 4% (M1) to slice 38% (M3) was determined (M1/M3 ratio), ideally being equal to 1. At the 66% slice, muscle CSA was calculated. The long-term in vitro (phantom) precision of the pQCT in 12-month daily measurements was 0.12% for total vBMD and 0.3% for trabecular vBMD. The in vivo precision derived from 25 postmenopausal women subjected to duplicate measurements within 1 month was as follows: total vBMD, 0.2 ± 0.13; trabecular vBMD, 0.46 ± 0.26; cortical area, 0.34 ± 0.19; SSIp, 1.02 ± 0.63; and cortical thickness, 0.83 ± 0.48.
Statistical analysis
Data are presented as mean ± SD. Between-group differences were analyzed by ANOVA with Bonferroni correction for multiple comparisons concerning continuous variables and 
2 test for categorical variables as appropriate. Generalized linear modeling univariate regression analysis was used to adjust all the BMC comparisons for changes in the respective areas, whereas endocortical circumference was adjusted for periosteal circumference changes and vice versa. Bivariate correlations were estimated by Pearson correlation test. All tests were two tailed, and P < 0.05 was considered significant. All data analysis was performed using the Statistical Package for Social Sciences (version 10.0) software (SPSS Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Of the 65 patients initially evaluated for PH, eight were receiving bisphosphonates, two were receiving steroid therapy for rheumatic diseases, and three were premenopausal, thus leaving 52 postmenopausal women with PH fulfilling the inclusion criteria. The two groups were comparable according to age, weight, height, and menopausal years. Women with PH had significantly higher serum and urinary calcium concentrations, higher iPTH (P < 0.001), and lower phosphate levels (P < 0.001) compared with controls (Table 1
). Furthermore, there was a trend for lower 25OHD3 levels in patients with PH (P = 0.08).
|
Comparison of aBMD between groups by DXA
Patients with PH had significantly reduced aBMD at lumbar spine (P < 0.001) compared with controls (Table 1
). Furthermore, both T- and Z-scores were significantly lower in patients with PH (lumbar spine, P = 0.001 and P = 0.001, for T- and Z-score, respectively). Finally, using the World Health Organization criteria for the diagnosis of osteoporosis (18), 44.2% of patients with PH vs. 21.4% of controls had osteoporosis (P < 0.05).
Comparison of vBMD and bone geometry between groups by pQCT
Trabecular site (4% of the tibia length).
The tibia is rich in trabecular bone at the 4% site and cortical bone at the 38% site and has intermediate amounts of both types of bone tissue at the 14% site. Thus, at 4%, both total vBMD and total BMC as well as trabecular BMC and vBMD were significantly reduced (–15.1 and –15.8%, respectively) in patients with PH compared with controls (P < 0.001) (Table 2). No significant difference was observed between the two PH groups (Table 2).
|
Transition zone (14% of the tibia length).
Concerning the 14% site, both trabecular and cortical density were significantly reduced (P < 0.001) in patients with PH compared with controls. Finally, the M1/M3 ratio was significantly reduced in patients with PH, indicating increased cancellous bone loss relative to cortical bone loss compared with controls. However, when we analyzed the M1/M3 ratio according to baseline calcium levels (Table 2), this effect was mainly observed in hypercalcemic patients compared with controls (P < 0.05).
Bone geometry
Trabecular site (4% of the tibia length).
Total CSA and trabecular CSA were comparable between groups (P > 0.05). Although no statistically significant difference was detected between normo- and hypercalcemic patients (Table 2), when we compared trabecular area in patients with calcium levels greater than 2.75 mmol/liter, there was a significant reduction of trabecular area compared with normocalcemic patients with PH (454.55 ± 56.22 vs. 498.2 ± 47.9 mm2; P = 0.01) and marginal reduction compared with controls (P = 0.07).
Cortical site (38% of the tibia length). At the 38% site, cortical CSA was significantly reduced (–7.4%; P < 0.05). Concerning differences according to baseline calcium levels, there was a significant decrease of CSA in hypercalcemic patients compared with controls (P < 0.01). Furthermore, endosteal circumference increased (8.8%; P < 0.001), whereas cortical thickness decreased (–9.7%; P < 0.001). These findings were evident in both PH groups compared with controls, particularly, however, in hypercalcemic patients. Yet periosteal circumference was comparable between the two groups (1.16%). Finally SSIp, an index of bone strength, was comparable between groups.
Transition zone (14% of the tibia length). At the 14% site, total CSA and trabecular CSA were comparable between groups (P > 0.05), whereas cortical area was reduced (P = 0.001), a finding evident in both groups with PH compared with controls (P < 0.05). SSIp at the 14% site was reduced only in hypercalcemic patients compared with controls (P < 0.01).
Correlation analyses
The correlation between PTH levels with biochemical as well as quantitative and qualitative bone characteristics gives an insight of the altered PTH regulation in PH and the skeletal effects of continuous PTH excess. Concerning the association between biochemical variables, in patients with PH, iPTH correlated significantly with calcium (r = 0.38; P < 0.01) and phosphate levels (r = –0.28; P < 0.05). On the contrary, in controls, iPTH showed negative association with calcium (r = –0.27; P < 0.05), phosphate (r = –0.38; P < 0.01), and 25OHD3 levels (r = –0.31; P < 0.05). In patients with PH, iPTH levels correlated negatively with trabecular vBMD at all sites tested (4%: r = –0.28; P < 0.05). However, no association was detected between trabecular CSA and iPTH. In contrast, in controls, trabecular vBMD at the 4% site had no association with iPTH, whereas the respective trabecular CSA had significant positive association with iPTH (r = 0.32; P < 0.05). Regarding indices of cortical bone assessed at the 38% site, iPTH showed significant positive association with endosteal circumference (r = 0.37; P < 0.01).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Continuous endogenous PTH excess is characterized by site-specific skeletal effects concerning both bone density and structure, depending on the severity and duration of PH and possibly on the background bone disease and other hormonal factors that might modulate PTH effects on the skeleton. In general, contemporary PH is characterized by particular bone loss at cortical sites, especially at the distal radius, whereas aBMD at sites enriched in cancellous bone is less affected (19). Given the dual actions of PTH on the skeleton, it is clear that aBMD assessed by DXA underestimates important aspects of bone strength (20, 21). Thus, pQCT, an alternative to invasive measures of bone quality, seems more appropriate for the evaluation of skeletal disease in PH.
In the present study, the catabolic actions of continuous PTH excess on cortical bone were evident. At the respective 38% site of the tibia, both cortical content and area were reduced by 9.7 and 7.4%, whereas cortical density decreased by 2.3%. These findings are in accordance with previous studies using DXA (2, 3), histomorphometry (21, 22), or pQCT (12). Concerning cortical geometric properties, the most striking abnormality was a significant increase of endosteal circumference, whereas periosteal circumference was comparable. That led to reduction of cortical thickness. Furthermore, PTH levels correlated positively with endosteal circumference, indicating that at peripheral weight-bearing sites, continuous PTH excess is associated with increased endosteal resorption but might not result in significant periosteal apposition. This finding contrasts with those reported at the radius using either pQCT (12) or single-photon absorptiometry (6) and at the second metacarpal by x-rays (6). Indeed, Chen et al. (12), using pQCT at the radius, reported significant reduction of cortical vBMD and cortical thickness, increased endo- and periosteal circumference in 36 women with PH, whereas bone strength, assessed by SSIp, was preserved. Furthermore, Adami et al. (23), in a combined cross-sectional and longitudinal study in postmenopausal women with PH using digitalized x-rays of the second metacarpal, reported a significant increase in the Z-score of the outer diameter and increased subperiosteal apposition rate in the follow-up. However, absolute outer diameter values were comparable between patients and controls. Thus, it is possible that differences in the number of patients studied (type II error) and in methods used account for the observed variations. An alternative explanation could be that periosteal acquisition might be a secondary adaptation to increased cortical porosity, thus being a late phenomenon in the natural history of PH. However, no significant difference was detected between hypercalcemic and normocalcemic patients. Given the cross-sectional design of our study, we cannot exclude the possibility of increased subperiosteal apposition rate in weight-bearing sites in patients with PH. Nevertheless, at the 38% site, bone strength assessed by SSIp was not reduced, indicating favorable adaptation to the catabolic actions of PTH and thus relative preservation of cortical mechanical properties.
Our data concerning the effects of PH on trabecular bone mass argue against an anabolic action. Indeed, at all sites tested, and especially at the 4% site, trabecular content and vBMD were significantly reduced compared with controls. Yet trabecular area was comparable, indicating apparent demineralization of the bone. Although a number of investigators, including Chen et al. (12), reported similar results (24, 25, 26), these findings contrast with the general concept of relative preservation of cancellous bone in PH. However, when we examined the same indices according to calcium levels, normocalcemic patients had a moderate reduction of trabecular BMC, although area was preserved, whereas hypercalcemic patients showed significant reduction of trabecular BMC compared with controls and marginally compared with normocalcemic subjects (P = 0.09), while trabecular area decreased significantly only in a subgroup of hypercalcemic patients with overt hypercalcemia (>2.75 mmol/liter). Although these results are obtained from subgroup comparisons, they raise the possibility that during the early stages of PH, trabecular bone mass at weight-bearing sites is mildly affected, whereas geometric properties, such as trabecular area, are preserved. However, during the late stages, BMC and possibly area tend to decrease, indicating profound demineralization of the bone and loss of the favorable adaptation of bone geometric properties.
There are some limitations to this study. First, although the number of patients is comparable with other studies reporting bone geometric properties in patients with PH, the possibility of type II error cannot be excluded. A second limitation refers to the cross-sectional design of the study. Thus, although baseline calcium levels are considered the hallmark of the different stages/groups of PH and therefore probably account for both severity and duration of continuous PTH excess, we cannot ascertain individual longitudinal changes of bone geometric properties. Moreover, some limitations inherent to the pQCT technique must be noted. Although the threshold method for the separation of cortical from trabecular bone has been validated (27), there is still difficulty obtaining an accurate estimation of the endosteal portion of the cortex. However, when evaluating the relation of cortical area with bone strength, it does not matter whether endosteal area is included (13). In addition, bone marrow area is included in the estimation of total and trabecular area but not on the respective BMC. However, to our knowledge, there is no evidence for a primary bone marrow disorder in patients with PH. The issue of multiple t testing without adjustments concerning pQCT-derived variables merits comment. Given that these variables are interrelated and may be differentially affected by continuous PTH excess, there is a possibility that significant P values represent significant differences for other interrelated variables. However, when we adjusted BMC results for changes in the respective areas and periosteal results for changes in endocortical circumference, no significant alteration was detected (Table 2). Finally, the data of the present study concern only postmenopausal women with PH, and results should not be extrapolated in either premenopausal women or males with PH.
In conclusion, normocalcemic PH is characterized by catabolic actions at cortical and cancellous sites, an effect accentuated in hypercalcemic patients. Cortical geometric properties are adversely affected even in normocalcemic patients, whereas trabecular properties are generally preserved.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online February 21, 2006
Abbreviations: aBMD, Areal bone mineral density; BMC, bone mineral content; CSA, cross-sectional area; DXA, dual-energy x-ray absorptiometry; iPTH, intact PTH; 25OHD3, 25-hydroxyvitamin D3; PH, primary hyperparathyroidism; pQCT, peripheral quantitative computed tomography; SSIp, polar stress strength index; vBMD, volumetric bone mineral density.
Received September 20, 2005.
Accepted February 15, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. C. Hsiao, B. M. Boudignon, W. C. Chang, M. Bencsik, J. Peng, T. D. Nguyen, C. Manalac, B. P. Halloran, B. R. Conklin, and R. A. Nissenson From the Cover: Osteoblast expression of an engineered Gs-coupled receptor dramatically increases bone mass PNAS, January 29, 2008; 105(4): 1209 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Bollerslev, S. Jansson, C. L. Mollerup, J. Nordenstrom, E. Lundgren, O. Torring, J.-E. Varhaug, M. Baranowski, S. Aanderud, C. Franco, et al. Medical Observation, Compared with Parathyroidectomy, for Asymptomatic Primary Hyperparathyroidism: A Prospective, Randomized Trial J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1687 - 1692. [Abstract] [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 |
