SEARCH

SEARCH BY CITATION

Keywords:

  • hyperparathyroidism;
  • PTH;
  • HR-pQCT;
  • microarchitecture;
  • bone densitometry

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Patients with primary hyperparathyroidism (PHPT) have continuously elevated parathyroid hormone (PTH) and consequently increased bone turnover with negative effects on cortical (Ct) bone with preservation of trabecular (Tb) bone. High-resolution peripheral quantitative computed tomography (HR-pQCT) is a new technique for in vivo assessment of geometry, volumetric density, and microarchitecture at the radius and tibia. In this study we aimed to evaluate bone status in women with PHPT compared with controls using HR-pQCT. The distal radius and tibia of 54 women—27 patients with PHPT (median age 60, range 44–75 years) and 27 randomly recruited age-matched healthy controls (median age 60, range 44–76 years)—were imaged using HR-pQCT along with areal bone mineral density (aBMD) by dual-energy X-ray absorptiomentry (DXA) of the ultradistal forearm, femoral neck, and spine (L1–L4). Groups were comparable regarding age, height, and weight. In the radius, patients had reduced Ct area (Ct.Ar) (p = .008), Ct thickness (Ct.th) (p = .01) along with reduced total (p = .002), Ct (p = .02), and Tb (p = .02) volumetric density and reduced Tb number (Tb.N) (p = .04) and increased Tb spacing (Tb.sp) (p = .05). Ct porosity did not differ. In the tibia, no differences in HR-pQCT parameters were found. Moreover, patients had lower ultradistal forearm (p = .005), spine (p = .04), and femoral neck (p = 0.04) aBMD compared with controls. In conclusion, a negative bone effect of continuously elevated PTH with alteration of HR-pQCT assessed geometry, volumetric density, and both trabecular and cortical microarchitecture in radius but not tibia was found along with reduced aBMD by DXA at all sites in female patients with PHPT. © 2010 American Society for Bone and Mineral Research


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Primary hyperparathyroidism (PHPT) is a common endocrine disorder characterized by elevated serum calcium and inappropriately elevated serum parathyroid hormone (PTH). Today, skeletal involvement is typically a reduced bone mineral density (BMD) when measured by dual-energy X-ray absorptiometry (DXA) rather than classic osteoitis cystica.1

The continuously elevated PTH increases the frequency of activation of the bone-remodeling units leading to a high bone turnover state.2–4 Most histomorphometric studies on iliac crest bone biopsies have found preservation in cancellous bone microarchitecture but increased cortical porosity (Ct.Po) and/or reduced cortical width.2–8 Several epidemiologic studies have found an increased fracture risk in PHPT patients, suggesting a reduced bone biomechanical competence as a consequence of the high turnover state per se or owing to the subsequent bone loss.9–12

The access to parameters of bone quality synonymous with bone characteristics besides BMD13 has improved in recent years with the refinement of bone imaging techniques. With high-resolution peripheral quantitative computed tomography (HR-pQCT), 3D images of the radius and tibia with a resolution of 82 µm can be readily achieved.14 This provides in vivo access to parameters such as cortical thickness (Ct.Th), trabecular number (Tb.N), trabecular thickness (Tb.Th), and trabecular separation (Tb.Sp) normally restricted to histomorphometry or micro–computed tomographic (µCT) analysis of bone biopsy specimen. Simultaneously, parameters of bone geometry and volumetric BMD (vBMD) are achieved. HR-pQCT parameters display high accuracy compared with gold standard µCT,15 and HR-pQCT has been used clinically to gain insight into age- and gender-related bone characteristics,16, 17 as well as bone architectural changes in diseases of bone loss such as osteopenia and osteoporosis.14, 18, 19

However, the bone changes inflicted by the elevated PTH in PHPT using this technique are, to our knowledge, still to be explored. In this study we therefore aimed to gain insight into bone geometry, density, and microarchitecture using HR-pQCT in female patients with PHPT.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Patients

Twenty-seven women (median age 60 years, range 44 to 75 years) with biochemically proven PHPT referred for parathyroidectomy (PTx) at the Center for Endocrine Neck Surgery, Odense University Hospital, Denmark, from April 2009 to November 2009 agreed to participate in the study. Patients with a history of prior PTx, former or current use of oral glucocorticoids (equivalent to more than 3 months of 5 mg of prednisone daily), age below 18 years, kidney disease, or uncontrolled thyroid disease were excluded. No patients were treated with biphosphonates or cinacalcet.

Age-matched controls were invited by mail from a random sample of 900 women drawn from the Civil Registration System in the Municipality of Odense, Denmark, to participate in the study. A total of 69 women were invited, of whom 32 agreed to participate in the study (17 declined, and 20 did not reply). Four women were excluded because of known calcium metabolic disease (n = 1), other endocrine disorders potentially affecting bone (n = 2), or treatment with biphosphonates (n = 1). To exclude subjects with diseases potentially affecting bone, all controls were screened biochemically with measurements of serum potassium, serum creatinine, serum alkaline phosphatase (ALP), serum alanine transaminase, and serum ionized calcium and 25-hydroxyvitamin D [25(OH)D], as well as additionally with serum thyroid-stimulating hormone and serum PTH if osteoporosis was found on DXA examination. As a result of the biochemical screening, a single subject was excluded owing to hypercalcemia. Thus the control group consisted of 27 women (median age 60 years, range 44 to 76 years).

All participants gave verbal and written informed consent, and the study was performed according to guidelines from the Declaration of Helsinki and approved by the local ethics review board.

Dual-energy X-ray absorptiometry (DXA)

Areal BMD (aBMD) values for the lumbar spine (L1–L4), hip (trochanteric region, femoral neck, and total hip), and nondominant forearm (ultradistal, middle, proximal third, and total) were measured using DXA (Hologic 4500-A or Hologic Discovery, Waltham, MA, USA) and according to the World Health Organization (WHO) definition of osteoporosis, expressed as a T-score equivalent to the number of SDs below the young-adult mean. In our unit, the coefficient of variation (CV) for measurements of spine and total hip is 1.5% at both sites.

High-resolution peripheral quantitative computed tomography (HR-pQCT)

Assessment of geometry, vBMD, and microarchitecture of the nondominant distal radius and distal tibia (or in case of a previous fracture in the affected radius/tibia, the nonfractured limb) was obtained using a HR-pQCT system (Xtreme CT, Scanco Medical AG, Brüttisellen, Switzerland). The image acquisition, image analysis, and validation of the method have been described in detail previously.15, 20, 21 The manufacturer's default protocol for in vivo patient scanning was applied with provision of a 3D representation of 9.02 mm of the radius and tibia in the axial direction. Images analysis involves filtering with a Laplace-Hamming filter and binarization to separate the cortical and trabecular regions, as described by Laib and colleagues.22 Furthermore, images were analyzed with an automated segmentation method23 using two threshold values to extract the endosteal and periosteal edges of the cortex, which allows isolation of the cortical region and assessment of Ct.Po, as described by others.24, 25 The CVs for geometry [total bone area (Ar), Ct.Ar, Tb.Ar, and cortical perimeter], vBMD (total bone vBMD, Ct. vBMD, Tb. vBMD), and microarchitectural parameters (Ct.Th, Ct.Po, Tb.N, Tb.Th, and Tb.Sp) were in the range of 0.2% to 1.8%, 0.4% to 0.9%, and 0.6% to 7.2%, respectively.

Biochemical analysis

Blood samples were drawn in a nonfasting state and analyzed without delay. Serum ionized calcium was measured with a Triolab Nova 8 (Brondby, Denmark), ALP with a Roche Modular P system (Basel, Switzerland), and intact PTH with an IMMULITE 2000 (Deerfield, IL, USA) immunoassay, whereas [25(OH)D] was assayed with EZChrom Elite (Santa Clara, CA, USA) chromatography.

Statistical analysis

Data are presented as mean ± SD or median (range) as appropriate. Between-group differences are analyzed with Wilcoxon's nonparametric rank-sum test. p Values are two-sided, and the statistical significance level was set at .05. All statistics were performed using Stata Statistical Software Release 11.0 (StataCorp LP, College Station, TX, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Baseline and biochemical characteristics

Despite instructions to stay relaxed during HR-pQCT examination, one radius scan from a PHPT patient had poor scan quality because of motion artifacts and was excluded from the analysis. In another PHPT patient, the lower leg circumference was too large to fit the scanner opening, and no tibia scan could be performed. In all participants, aBMD by DXA was achieved without difficulty.

Known duration of PHPT disease as calculated from the first elevated serum PTH and until surgery was 12 months (3 to 99 months). Serum ionized calcium and PTH levels in the PHPT group were 1.46 ± 0.09 mmol/L and 11.9 ± 8.1 pmol/L with standard laboratory reference values of 1.19 to 1.29 mmol/L and 1.10 to 6.90 pmol/L, respectively. The diagnosis of PHPT was verified biologically in 25 of the 27 patients, all with the finding of parathyroid adenoma on pathologic examination. In the remaining 2 patients, normal parathyroid tissue was found, and both had elevated serum calcium postoperatively, whereas serum calcium was normalized in the remainder of the patients (data not shown).

The PHPT patients and controls were comparable according to age, weight, height, menopausal status, and years from menopause, as well as serum [25(OH)D], as shown in Table 1. Serum ALP did not differ between groups. Also in Table 1, aBMD variables by DXA for the forearm, spine, and hip are presented.

Table 1. Baseline Characteristics and aBMD by DXA
ParameterPHPTControlsp Valuea
  • Data are shown as mean ± SD or median (range).

  • p Values ≤ .05 are shown in boldface.

  • a

    Wilcoxon rank-sum test.

Subjects (n)2727 
Age (years)60 (44–75)60 (44–76).88
Weight (kg)75 ± 1972 ± 12.67
Height (cm)164 ± 6164 ± 6.76
Pre-/postmenopausal (n/n)4/233/24 
Years from menopause14 ± 515 ± 10.88
History of fractures (n)97.55
25-Hydroxyvitamin D (nmol/L)70 ± 2969 ± 30.98
Serum ionized calcium (mmol/L)1.46 ± 0.091.22 ± 0.03<.00001
Serum total alkaline phosphatase (U/L)82 ± 4668 ± 20.61
DXA   
Forearm   
 Forearm ultradistal BMC (g)1.85 ± 0.392.08 ± 0.38.02
 Forearm ultradistal BMD (g/cm2)0.307 ± 0.0620.345 ± 0.053.005
 Forearm middle BMC (g)5.53 ± 1.196.07 ± 1.15.13
 Forearm middle BMD (g/cm2)0.471 ± 0.0790.505 ± 0.064.09
 Forearm third BMC (g)2.87 ± 0.473.12 ± 0.34.06
 Forearm third BMD (g/cm2)0.585 ± 0.0850.633 ± 0.071.04
 Forearm total BMC (g)10.24 ± 1.9111.27 ± 1.77.07
 Forearm total BMD (g/cm2)0.452 ± 0.0730.490 ± 0.060.03
Spine   
 Spine (L1–L4) BMC (g)45.04 ± 15.5350.03 ± 12.74.21
 Spine (L1–L4) BMD (g/cm2)0.837 ± 0.1470.917 ± 0.130.04
Hip   
 Femoral neck BMC (g)3.34 ± 0.613.59 ± 0.60.09
 Femoral neck BMD (g/cm2)0.645 ± 0.1080.705 ± 0.117.04
 Trochanteric BMC (g)6.65 ± 1.357.53 ± 2.10.10
 Trochanteric BMD (g/cm2)0.600 ± 0.1070.652 ± 0.106.04
 Total-hip BMC (g)28.33 ± 4.9830.64 ± 4.58.08
 Total-hip BMD (g/cm2)0.795 ± 0.1250.852 ± 0.107.05
 Osteoporosis by WHO criteria (n)117.25

aBMD by DXA

As shown in Table 1, aBMD values of the ultradistal (p = .005), distal third (p = .04), and total forearm (p = .005), but not middle region (p = .09), of the radius were significantly lower in PHPT patients than in controls. Similarly, PHPT patients had lower L1–L4 aBMD values (p = .04). In the hip, aBMD values for the femoral neck, trochanteric region, and total hip were lower in PHPT patients than in controls (p = .04, p = .04, and p = .05, respectively). Applying WHO criteria for osteoporosis, 11 PHPT patients versus 7 controls were classified as osteoporotic.

Geometry, vBMD, and microarchitecture by HR-pQCT

Table 2 shows the geometry, vBMD values, and microarchitecture, as measured by HR-pQCT. In the radius, the total bone area was comparable between groups, whereas Ct.Ar was reduced in PHPT patients (p = .008). No difference could be found in radial trabecular area or in total bone area or cortical or trabecular area in the tibia.

Table 2. Geometry, Density, and Microarchitecture by HR-pQCT
  Radius 
ParameterPHPTControlsp Value
Number (n)2627 
Geometry   
 Total bone area (cm2)260.4 ± 48.7254.9 ± 38.1.83
 Cortical area (cm2)48.6 ± 12.558.2 ± 11.3.008
 Trabecular area (cm2)205.9 ± 50.4193.2 ± 39.5.44
 Cortical perimeter (mm)68.2 ± 6.666.9 ± 5.4.44
Volumetric BMD   
 Total bone density (mg/cm3)264.0 ± 79.0317.3 ± 73.7.002
 Cortical density (mg/cm3)844.6 ± 74.6892.1 ± 58.4.02
 Trabecular density (mg/cm3)101.0 ± 46.0125.2 ± 36.2.02
Microarchitecture   
 Trabecular bone volume/total volume0.08 ± 0.040.10 ± 0.03.02
 Ct.Th (mm)0.72 ± 0.220.87 ± 0.20.01
 Ct.Po (%)2.10 ± 1.081.71 ± 0.97.18
 Tb.N (mm−1)1.41 ± 0.561.71 ± 0.45.03
 Tb.Th (mm)0.06 ± 0.010.06 ± 0.01.43
 Tb.Sp (mm)0.74 ± 0.350.60 ± 0.32.05
  Tibia 
ParameterPHPTControlsp Value
  1. Data are shown as mean ± SD.

  2. p Values ≤ .05 are shown in boldface.

  3. aWilcoxon rank-sum test.

Number (n)2627 
Geometry   
 Total area (cm2)730.9  ±  123.9734.7 ± 126.9.96
 Cortical area (cm2)95.3 ± 25.6101.9 ± 23.4.32
 Trabecular area (cm2)626.7 ± 127.1625.4 ± 134.0.76
 Cortical perimeter (mm)106.2 ± 8.9106.6 ± 9.2.96
Volumetric BMD   
 Total bone density (mg/cm3)226.8 ± 59.5251.5 ± 49.6.07
 Cortical density (mg/cm3)803.0 ± 78.8826.7 ± 68.0.19
 Trabecular density (mg/cm3)125.5 ± 35.8144.2 ± 36.0.06
Microarchitecture   
 Trabecular bone volume/total volume0.10 ± 0.030.12 ± 0.03.06
 Ct.Th (mm)0.91 ± 0.270.97 ± 0.26.35
 Ct.Po (%)6.63 ± 2.916.70 ± 3.00.96
 Tb.N (mm−1)1.58 ± 0.491.78 ± 0.37.14
 Tb.Th (mm)0.07 ± 0.020.07 ± 0.01.76
 Tb.Sp (mm)0.64 ± 0.270.52 ± 0.13.12

For the vBMD measurements in the radius, we found reductions in both total bone and cortical and trabecular vBMD (p = .002, p = .02, and p = .02, respectively). In the tibia, no difference in vBMD could be found, yet we did observe a trend toward reduced total (p = .07) and trabecular density (p = .06).

For radial microarchitecture, patients with PHPT had reduced Ct.Th (p = .02), whereas Ct.Po was similar between groups. In the trabecular compartment, both the ratio of trabecular bone volume to tissue volume (BV/TV; p = .02) and Tb.N (p = .03) were lower, whereas Tb.Sp (p = .05) was higher in patients than in controls. In the tibia, no between-group differences in any of the microarchitectural parameters could be demonstrated, but a trend toward lower trabecular vBMD (p = .06) and BV/TV (p = .06) was observed in PHPT patients.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

In this HR-pQCT study, we found that women with PHPT had reduced Ct.Ar and Ct.Th, as well as total, cortical, and trabecular vBMD, and reduced Tb.N and increased Tb.Sp in the radius. In contrast, no difference in geometry, vBMD values, or microarchitecture was observed in the tibia. Furthermore, patients had reduced aBMD, as measured by DXA, in the forearm, spine, and hip.

From our results, it seems that the continuously elevated PTH decreases bone mass in the radius in both the cortical and trabecular compartments. For the trabecular compartment, we found that Tb.N was reduced along with increased Tb.Sp, whereas Tb.Th did not differ. This deterioration of trabecular architecture is opposed to findings from iliac crest biopsies, where several studies have found preserved trabecular integrity in PHPT patients.3, 4, 6, 8 The changes in the cortical compartment of the radius, with reduced Ct.Th seen in our study, are in line with some4, 26 but not other2 histomorphometric findings in biopsies from the iliac crest. In contrast to the histomophometric findings in biopsies from the iliac crest,2, 7 we did not observe an increase in Ct.Po in the radius. The inability to demonstrate a difference in Ct.Po could be due to a lack of power given the relatively small number of subjects evaluated because we did observe a 19% difference in Ct.Po in PHPT patients compared with controls (2.10% ± 1.08% in PHPT, 1.71% ± 0.97% in controls). Calculated from our data, a sample size of 109 patients and a similar number of controls would have been required for such a difference to be statistically significant (α = 0.05, β = 0.80). Alternatively, the lack of effect on Ct.Po could be the inability of HR-pQCT to capture Ct.Po given the inferior resolution compared with histomorphometry; however, Ct.Po. as assessed using HR-pQCT performed well in comparison with µCT (19-µm resolution) with an r2 value of 0.80 in regression analysis.25

The mechanism by which the geometry changes in the radius are inflicted cannot be addressed conclusively from our findings given the absence of dynamic indices. However, it is noted that radial cortical area was reduced significantly, along with both (insignificantly) increased total bone and trabecular area. From studies on the mechanism of intermittent PTH on bone in cynomolgus monkeys, endocortical resorption (trabecularization) combined with periosteal apposition has been demonstrated,27 a mode of action that could explain our findings.

For the tibia, no effect of excess PTH could be demonstrated for either cortical or trabecular bone. The preserved bone geometry in the distal tibia in PHPT patients is in agreement with a pQCT study by Charapoulos and colleagues, but they did report reductions in both cortical and trabecular density that were not found in our study.28 The inability to capture differences in the tibia again could be a matter of power, and a sample size estimation with the means and SDs found here and α = 0.05 and β = 0.80 revealed that evaluation of 36 and 56 women in each group would have been required to reveal significant between-group differences for tibial BV/TV and tibial total bone vBMD, respectively.

The reason for the negative effects of PHPT in the radius but not the tibia is unclear, yet a possible explanation is the different mechanical loading conditions that apply for the two sites, with the tibia being a weight-bearing bone. Mechanical loading is a prerequisite for normal bone formation, as well as for preservation, and studies in rodents have demonstrated a positive synergistic effect on both cortical and trabecular bone when combined mechanical loading and PTH treatment was instituted.29–33 However, the PTH exposure in those studies were intermittent and not continuous, as in PHPT, but our findings indicate that mechanical loading plays a role in the action of PTH and that mechanical stimuli are important for the effects mediated by PTH.

In our study we found reduced aBMD by DXA in the forearm, spine, and hip. Bone evaluation by DXA in patients with PHPT has been performed in several case-control and a few longitudinal studies, and most have reported reduced aBMD in the radius.34–36 For the spine and hip, the results have varied. In the spine, reduction was found in some28, 34, 37 but not all35, 36, 38 studies, and similarly for the hip, both aBMD reductions36, 37 and no difference34 have been reported. The aBMD findings in our study are in accordance with those previously described, but we did observed reductions in aBMD at all three skeletal sites.

From a clinical perspective, bone status in the individual patient with PHPT is important because of fracture risk12 and because evidence of bone loss is an indication for PTx, as stated in recent guidelines.39 The patients in this study had bone loss in the radius, hip, and spine but not in the tibia. In an epidemiologic study of fracture risk in PHPT patients, Khosla and colleagues demonstrated an increased overall risk of fracture primarily accounted for by an increase in vertebral, distal radius, rib, and pelvic fractures, whereas the risk of fractures in the tibia was not increased.12 Based on previous histomorphometric and densitometric findings, it has been hypothesized that trabecular bone is preserved compared with cortical bone in patients with PHPT. However, the majority of fractures in PHPT patients occur in skeletal sites rich in cancellous bone. It has been argued that the increased risk of fracture could be due to cortical thinning40 or the high turnover state per se.41, 42 Whereas the latter cannot be addressed from our findings, the cortical thinning in the radius found in our study could, at least in part, account for the observed increased risk of distal radius fractures. Furthermore, in accordance with the normal fracture risk in the tibia, we did not observe tibial bone loss. On the contrary, we found no evidence of trabecular preservation because reductions in aBMD also were seen in skeletal sites rich in trabecular bone such as the ultradistal forearm, spine, and trochanteric region of the hip. Therefore, based on our results, the increased risk of fracture may be accounted for by deteriorations in both cortical and trabecular bone.

Our study has some limitations. First, the findings are based on a fairly small number of subjects, and as outlined earlier, we cannot exclude that evaluation of a larger number of patients also could have resulted in between-group differences in the HR-pQCT indices in the tibia. Second, we have no measure of bone strength and so cannot directly address how the described bone changes relate to biomechanical competence.

Our study, however, also has some important strengths. First, we included a homogeneous group of PHPT patients in whom a pathologic diagnosis was confirmed in the vast majority. In addition, controls were recruited at random from the back-ground population to avoid recruitment bias.

In conclusion, we found evidence of a negative bone effect of continuously elevated PTH with alteration of geometry, volumetric density, and microarchitecture in both trabecular and cortical bone in the radius but not the tibia assessed by HR-pQCT, along with reduced aBMD by DXA in the forearm, spine, and hip in women with PHPT.

Disclosures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

All the authors state that they have no conflicts of interest.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Thanks to Anette Riis Madsen, Steffanie Anthony-Christensen, and Elizabeth Hanmann for excellent technical support. This work has received grants from the PhD School of Endocrinology, University of Southern Denmark, the Municipal Region of Southern Denmark, and The Lily Benthine Lunds Foundation (1/6/1978).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References