• Open Access

Vitamin D Supplementation After Parathyroidectomy: Effect on Bone Mineral Density—A Randomized Double-Blind Study

Authors

  • Sophie Norenstedt,

    Corresponding author
    1. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
    2. Department of Breast and Endocrine Surgery, Karolinska University Hospital, Stockholm, Sweden
    • Address correspondence to: Sophie Norenstedt, MD, PhD, Department of Breast and Endocrine Surgery, Karolinska University Hospital, P9:03, SE-171 76 Stockholm, Sweden. E-mail: Sophie.Norenstedt@capiostgoran.se

    Search for more papers by this author
  • Ylva Pernow,

    1. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
    2. Department of Endocrinology, Metabolism, and Diabetes, Karolinska University Hospital, Stockholm, Sweden
    Search for more papers by this author
  • Jan Zedenius,

    1. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
    2. Department of Breast and Endocrine Surgery, Karolinska University Hospital, Stockholm, Sweden
    Search for more papers by this author
  • Jörgen Nordenström,

    1. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
    2. Department of Breast and Endocrine Surgery, Karolinska University Hospital, Stockholm, Sweden
    Search for more papers by this author
  • Maria Sääf,

    1. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
    2. Department of Endocrinology, Metabolism, and Diabetes, Karolinska University Hospital, Stockholm, Sweden
    Search for more papers by this author
  • Fredrik Granath,

    1. Department of Medicine, Karolinska Institutet, Stockholm, Sweden
    Search for more papers by this author
  • Inga-Lena Nilsson

    1. Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
    2. Department of Breast and Endocrine Surgery, Karolinska University Hospital, Stockholm, Sweden
    Search for more papers by this author

ABSTRACT

Patients with primary hyperparathyroidism (PHPT) have higher bone turnover, lower bone mineral density (BMD), and an increased risk of fractures. They also have a high incidence of low vitamin D levels (25-OH-vitamin D <50 nmol/L) that could worsen the negative effect on the bone. In this double-blinded clinical trial, 150 patients with PHPT were randomized, after successful parathyroidectomy (PTX), to 1-year daily treatment with either cholecalciferol 1600 IU and calcium carbonate 1000 mg (D +) or calcium carbonate alone (D–). BMD was measured in the lumbar spine, femoral neck, total hip, distal and 33% radius using dual-energy X-ray absorptiometry (DXA) before surgery and after 1 year of study medication. Median age was 60 (range 30–80) years and there were 119 (79%) women and 31 (21%) men; 76% had 25-OH-D <50 nmol/L before PTX and 50% had persistent elevated parathyroid hormone (PTH) 6 weeks after PTX. A similar increase in BMD in the lumbar spine, femoral neck, and total hip was observed in both groups (D + : 3.6%, 3.2%, and 2.7%, p < 0.001, respectively; and D–: 3.0%, 2.3%, and 2.1%, respectively, p < 0.001). Patients with vitamin D supplementation also increased their BMD in distal radius (median 2.0%; interquartile range, −1.7% to 5.4%; p = 0.013). The changes in BMD, especially in the hips, were correlated to the baseline concentrations of PTH, ionized calcium, and bone markers (p < 0.001). A benefit from vitamin D substitution was observed among patients with a persistent postoperative PTH elevation, who also improved their BMD at 33% radius and radius ultradistal (p < 0.05). In conclusion, except for a minor improvement of radius BMD, our data show no beneficial effect on BMD or bone turnover markers of vitamin D supplementation after PTX. Preoperative PTH seems to have the strongest association with improvement in BMD. © 2014 American Society for Bone and Mineral Research.

Introduction

In primary hyperparathyroidism (PHPT), the level of parathyroid hormone (PTH) is inappropriately increased in relation to the level of calcium. Parathyroid adenomectomy (PTX) leads to normalization of PTH in most cases. PHPT is associated with increased bone turnover, resulting in bone loss and an increased risk of fractures.[1-5] The reduction in bone mineral density (BMD) seems to vary between skeletal sites, with the greatest effect at sites rich in cortical bone. A number of cohort studies have reported an increased risk of fractures at several sites in patients with PHPT, even 10 years before diagnosis.[6-8] The fracture risk is not only increased at sites rich in cortical bone, as suggested by the dual-energy X-ray absorptiometry (DXA) findings, but also in sites rich in cancellous bone, such as the lumbar spine and hip.

There are no randomized studies on the effect of surgery on fracture risk in PHPT, but three cohort studies have shown a decreased risk of fractures of the hip, femur, forearm, and upper arm.[3, 9, 10] After PTX, BMD increases and bone turnover markers normalize. The earliest improvements occur at sites rich in cancellous bone, such as the lumbar spine.[2, 11, 12]

Low vitamin D levels are more common among patients with PHPT than in the general population.[13-16] Low 25-hydroxyvitamin D (25-OH-D) levels result in a compensatory elevation of PTH[17, 18] and has been associated with a more advanced disease, such as larger parathyroid adenomas, higher PTH and calcium levels, and an increase in bone turnover and fracture risk.[2, 7, 19-21] Furthermore, low levels of 25-OH-D have been associated with persistent PTH elevation after PTX, indicating a coexistent secondary hyperparathyroidism.[13, 22] In a Swedish study, 28% of the patients had a persistent high level of PTH 8 weeks after PTX.[23] The combination of PTH above or in the upper reference interval and low vitamin D levels has been associated with an increased risk of fractures in normocalcemic, postmenopausal women.[24] It has been suggested that vitamin D supplementation may have a beneficial effect on PTH and BMD in PHPT patients.[25, 26]

Our aim was to study whether vitamin D supplementation after successful PTX can provide a positive effect on bone recovery.

Patients and Methods

Patients with PHPT subjected to PTX at the Karolinska University Hospital in Stockholm, Sweden, were enrolled in a double-blinded, randomized clinical trial. During the period from April 2008 to November 2010, 159 consecutive patients with PHPT subjected to PTX were included. A total of 460 PHPT patients were subjected to PTX during the study period at our clinic (Fig. 1). Exclusion criteria were age under 18 years, manifest osteoporosis at PHPT diagnosis, persistent hypercalcemia after surgery, postoperative hypocalcaemia requiring vitamin D treatment, glomerular filtration rate (GFR) <40 mL/min, pregnancy, breast-feeding, or the treating physician's assessment that it was unsuitable for the patient to participate for other reasons.

Figure 1.

Flowchart of the study.

Nine patients were excluded after surgery, before randomization. Two patients had been inappropriately included despite lithium treatment and osteoporosis with a previous fracture. Seven patients were excluded because of persistent hypercalcemia after PTX. A total of 150 patients were randomized at 6 ± 2 weeks after PTX, 75 into each group, to 1 year of oral treatment with either calcium carbonate 500 mg × 2 alone (D– group), or calcium carbonate 500 mg + cholecalciferol 800 IU × 2 (D+ group) (Fig. 1). Patients had to withdraw any current supplementation with vitamin D during the study period. Patients on vitamin D treatment, prescribed for medical reasons were not included in the study. Recip AB, Solna, Sweden, delivered the study medication. All tablets were identical in appearance, the tins were numbered, and randomization followed a list compiled by an independent clinical research support organization. A total of 135 patients had a complete follow-up. The 15 patients who were lost to complete follow-up were followed for a median of 6 (1–9, minimum–maximum) months. Reasons for termination were the patient's own will (11), emigration (1), deceased (2), and symptomatic vitamin D deficiency (1). Baseline data did not differ between dropouts and those who completed the study (data not shown).

Blood samples were drawn after overnight fasting at 6 ± 2 weeks before surgery, at randomization, and after 6 and 12 months of treatment. Two trained nurses measured body height and weight. Body mass index (BMI) was calculated at baseline as weight (in kg) divided by the square of height (in meters). BMD was measured 6 ± 2 weeks before surgery and after 12 months of treatment with the study medication. The primary endpoint was the change in PTH after PTX and treatment with the study medication. Secondary end points were vitamin D levels, biochemical markers of bone turnover, and BMD.

The study was blinded to all the researchers, physicians, nurses, and patients. External study monitoring assured compliance with Good Clinical Practice (GCP). Each patient gave written consent to participation. The study complied with the Ethical Principles of the World Medical Association Declaration of Helsinki, and was approved by the Medical Products Agency in Sweden and by the Local Ethics Committee, Regionala etikprövningsnämnden, EPN, Stockholm, Sweden.

BMD measurements

Areal BMD (aBMD, g/cm2) of the total body, total hip, femoral neck, lumbar spine, and nondominant forearm (ultradistal [UD] and 1/3 proximal radius) was estimated using DXA. The same instrument (Lunar Prodigy Advance, #PA + 41562; GE Healthcare, Diegem, Belgium) was used for all the patients. Osteoporosis was defined as a T-score at any site that exceeded −2.5 SD below the value for white women/men aged 20 to 29 years.

The precision error (SD) was 0.009 g/cm2 in the lumbar spine, 0.010 g/cm2 in the total hip, 0.007 g/cm2 in the femoral neck, 0.035 g/cm2 in 33% radius, and 0.028 g/cm2 in radius UD.

Laboratory methods

Blood and urine samples were collected after overnight fasting 6 weeks before and after PTX and after 6 and 12 months of treatment. Plasma concentrations of total calcium, alkaline phosphatase (ALP), phosphate, and creatinine were measured using the Synchron LX 20 system (Beckman Coulter Inc., Brea, CA, USA). Renal function was estimated by calculating the glomerular filtration rate (GFR; mL/min/1.73 m2) according to Cockroft-Gault's formula: GFR = (140–age in years) × (weight in kg/plasma creatinine) × (1.23 in men, 1.04 in women). Serum ionized calcium (Ca2+) was analyzed on an ABL 800 (Radiometer, Copenhagen, Denmark). Plasma concentrations of intact PTH and serum concentrations of procollagen type 1 amino-terminal propeptide (P1NP) and beta C-terminal telopeptide of type 1 collagen (βCTx) were determined with electrochemiluminescence immunoassay on the Modular E170 system (Roche Diagnostics GmbH, Mannheim, Germany). Serum concentrations of 25-OH-D were measured by chemiluminescence on Liason XL (DiaSorin, Inc., Stillwater, MN, USA); interassay coefficient of variation percent (%CV) is 4.6% at 15.5 nmol/L and 2.7% at 68.3 nmol/L, intraassay %CV is 4.4% at 15.5 nmol/L and 2.6% at 68.3 nmol/L. Values below 50 nmol/L were considered to represent low vitamin D levels.[27] In order to minimize interassay variation, the preoperative and postoperative samples of 25-OH-D, P1NP, and βCTx were analyzed in the same series on serum previously frozen at −70°C.

Statistics

Statistical analysis was performed with the IBM SPSS Statistics, version 20. Because data did not follow a normal distribution, they are expressed as median and interquartile range. Intraindividual analyses were performed with the Wilcoxon signed rank sum test. For comparisons between groups, the Mann-Whitney U test for unpaired data was used. The Kruskal-Wallis one-way analysis of variance was used for comparisons with respect to independent categorical variables with more than two levels, and chi-square tests for comparisons of the distributions of categorical variables. Bivariate associations between continuous variables were assessed with Spearman's ρ-correlation test. Partial correlations were used to assess the relationship between ΔBMD and PTH and 25-OH-D (controlled for age, gender, weight, smoking, and creatinine).

All tests were done two-tailed, and p < 0.05 was considered to be statistically significant.

Sample size calculation

Based on data from one European study showing that 90% of a population of patients with PHPT had low vitamin D levels[14] and a Swedish study in which 28% had an increased level of PTH 8 weeks after PTX,[23] we expected PTH to be within the normal range after PTX in 72% of the patients not receiving vitamin D and in 97% of those with vitamin D treatment. Because data on the effect of vitamin D on postoperative PTH levels are scarce, we assumed a normal PTH level in two-thirds of the patients with vitamin D supplementation. Thus, with a significance level of 0.05 and a power of 80%, we calculated a sample size of 71 patients in each group. To compensate for dropouts during the study, we chose to enroll 75 patients per group.

Results

Clinical characteristics and biochemical data at baseline are presented in Tables 1 and 2. All but 5 patients (3 women) were above the age of 40 years. The baseline concentration of PTH correlated positively to the ionized calcium level and to the weight of the adenoma (r = 0.35 and r = 0.44; p < 0.001) and there was a weak negative correlation to 25-OH-D (r = −0.21, p = 0.009). The calcium level was normalized after PTX in all patients, but 50% had a persistent elevation of PTH (>65 ng/L) 6 weeks after PTX. In this group, 85% had a preoperative 25-OH-D <50 nmol/L compared to 51 of 75 (68%) in the group with normal PTH (p = 0.02).

Table 1. Clinical Characteristics at Baseline (Before Surgery) (n = 150)
  1. BMI = body mass index.
Age, years, median (minimum–maximum)60 (30–80)
Women/men, n119/31
Women ≤50 years/ > 50 years, n19/100
BMI, kg/m2, median (minimum–maximum)26 (17–44)
Weight of adenoma, mg, median (minimum–maximum)450 (75–27800)
Multiglandular disease, n4
Vitamin D <50 nmol/L, n (%)106 (71)
Diabetes mellitus, n (%)8 (5)
Renal stone, n (%)14 (9)
Antihypertensive treatment, n (%)67 (45)
Osteoporosis, n (%)69 (46)
Smokers, n (%)23 (15)
Medication, n 
Bisphosphonates3
Corticosteroids3
Estrogen: systemic/local6/8
Table 2. Biochemistry at Baseline and the Median Change After PTX (n = 150 Patients)
 Baseline (6 weeks before PTX)Change after PTXpa
MedianIQRMedianIQR
  1. There was no difference between D+ and D–.
  2. PTX = parathyroidectomy; IQR = interquartile range; P-calcium = plasma concentration of calcium; S-Ca2+ = serum concentration of ionized calcium; P-PTH = plasma concentration of parathyroid hormone; P-ALP = plasma concentration of alkaline phosphatase; P-albumin = plasma concentration of albumin; P-creatinine = plasma concentration of creatinine; GFR = glomerular filtration rate; Pt(U)-calcium = 24-hours urine calcium; S-25-OH-D = serum concentration of 25-hydroxy-vitamin D; S-P1NP = serum concentration of procollagen type 1 amino terminal propeptide, βCTx = beta C-terminal telopeptide of type 1 collagen.
  3. aWilcoxon signed rank sum test for paired data.
  4. b1 µkat/L = 58.8235 units/L.
  5. cNormal values of P-creatinine: females <90 µmol/L; males <100 µmol/L.
P-calcium (2.15–2.50 mmol/L)2.582.51 to 2.56−0.31−0.41 to −0.22<0.001
S-Ca2+ (1.15–1.33 mmol/L)1.431.39 to 1.43−0.18−0.23 to −0.15<0.001
P-PTH (10–65 ng/L)11689 to 145−45−65 to −31<0.001
P-phosphate (0.75–1.4 mmol/L)0.830.74 to 0.920.180.08 to 0.28<0.001
P-ALP (<1.9 µkat/L)b1.21.0 to 1.4−0.1−0.2 to 0.0<0.001
P-albumin (36–45 g/L)3938 to 41−1−2 to 1<0.001
P-creatinine (µmol/L)c6556 to 761−5 to 60.400
GFR creatinine (mL/min/1.73 m2)9779 to 1170−8 to 80.900
Pt(U)-calcium (0.7–7.0 mmol/d)7.04.5 to 9.9no data  
S-25-OH-D (75–250 nmol/L)4031 to 491−4 to 80.004
S-P1NP (µg/L)6247 to 89−4−14 to 2<0.001
βCTx (ng/L)545388 to 707−193−312 to −98<0.001

At randomization, patients with 25-OH-D <50 nmol/L and patients with postoperative elevation of PTH were evenly distributed between the D+ and D– groups. Biochemical data before and after 1 year of study medication are shown in Table 3. The only significant differences between groups were the expected increase of 25-OH-D and the decrease of PTH in the D+ group (Table 3).

Table 3. Biochemistry 6 Weeks After Parathyroid Adenomectomy and Median Change After 1 Year of Study Medication
 Vitamin D+Vitamin D–Vitamin D+ versus D–
Randomization (n = 75)Change after 1 year (n = 66)Randomization (n = 75)Change after 1 year (n = 69)Randomization1 year
MedianIQRMedianIQRpaMedianIQRMedianIQRpapbpb
  1. IQR = interquartile range; S-Ca2+ = serum concentration of ionized calcium; P-PTH = plasma concentration of parathyroid hormone; P-phosphate = plasma concentration of phosphate; P-ALP = plasma concentration of alkaline phosphatase; P-creatinine = plasma concentration of creatinine; S-25-OH-D = serum concentration of 25-hydroxyvitamin D; S-P1NP = serum concentration of procollagen type 1 amino-terminal propeptide; S-βCTx = serum concentration of beta C-terminal telopeptide of type 1 collagen.
  2. aWilcoxon signed rank sum test, paired data.
  3. bMann-Whitney U test, independent data.
  4. c1 µkat/L = 58.8235 units/L.
  5. dNormal values of P-creatinine: females <90 µmol/L; males <100 µmol/L.
S-Ca2+ (1.15–1.33 mmol/L)1.241.21 to 1.270.00−0.01 to 0.040.0661.251.22 to 1.27−0.01−0.02 to 0.040.0280.5910.928
P-PTH (10–65 ng/L)6650 to 90−23−36 to −11<0.0016556 to 81−13−29 to −1<0.0010.824<0.05
P-phosphate (0.75–1.4 mmol/L)1.00.9 to 1.10.0−0.1 to 0.10.4691.00.9 to 1.10.1−0.1 to 0.20.0170.3710.606
P-ALP (<1.9 µkat/L)c1.21 to 1.4−0.2−0.3 to −0.1<0.0011.10.9 to 1.1−0.2−0.4 to −0.1<0.0010.4090.654
P-creatinine (µmol/L)d6658 to 743−4 to 80.0656858 to 773−3 to 80.0110.3560.536
S-25-OH-D (75–250 nmol/L)4033 to 523325 to 47<0.0014535 to 5460 to 11<0.0010.285<0.001
S-P1NP (µg/L)5643 to 79−30−50 to −17<0.0015739 to 73−27−41 to −14<0.0010.4070.806
S-βCTx (ng/L)314213 to 460−107−210 to −31<0.001318221 to 466−96−192 to −25<0.0010.9240.539

Two patients in the D+ group had 25-OH-D below 50 nmol/L compared to 36 patients in the D– group. Nineteen percent (n = 26; 20 women) of all the patients had a PTH level above the normal range (n = 17 [24.6%] in the D– group, n = 9 [13.6%] in the D+ group, p = 0.129). Twelve of the patients with a PTH level above the normal range had a 25-OH-D level below 50 nmol/L, all in the D– group. The combination of PTH elevation and a 25-OH-D level >75 nmol/L was seen in 6 patients (5 in the D+ group), all with normal ionized calcium (minimum 1.17; maximum 1.30 mmol/L) and GFR >60 mL/min/1.73 m2.

Biochemical bone turnover markers

In 79 patients, P1NP and βCTx were within the normal range at baseline, whereas 70 patients had an increased level of P1NP (n = 46), βCTx (n = 1), or both (n = 23). βCTx and P1NP decreased significantly in both groups (Tables 2 and 3). βCTx changed most after PTX, whereas the decrease in P1NP was more pronounced after 1 year of study medication. Patients with postoperative PTH >65 ng/L had higher levels of βCTx and ALP at baseline (βCTx: 634 [475–795] versus 490 [343–633], p = 0.026; and ALP: 1.3 [1.1–1.5] versus 1.2 [0.9–1.4], p = 0.041) and a tendency to higher P1NP (68 [51–94] versus 58 [43–80], p = 0.063).

BMD

BMD at baseline was similar in the D+ and D– groups. Men had higher BMD than women at all measured sites (p < 0.05), but a comparison of Z-scores showed no difference in the lumbar spine, femoral neck, or 33% radius. Women had lower Z-scores in the UD radius (−1.4 [−2.1 to −0.4] versus −0.8 [−1.4 to 0.3], p = 0.038) and men had lower Z-scores in the total hip (women: −0.2 [−1.0 to 0.4] versus men: −0.7 [−1.1 to −0.2], p = 0.029). After PTX and 12 months of study medication, median BMD had increased significantly in the lumbar spine, the total hip, and the femoral neck in both the D+ and the D– group (Table 4). In the D+ group the BMD of the UD radius also increased, but there was no other significant difference in response between the D+ and D– groups. The improvement in bone density (ΔBMD) did not correlate to the baseline BMD except for in the radius(r = −0.24; p = 0.006). Patients with a PTH >65 ng/L 6 weeks after PTX had a greater improvement in BMD in total hip, femoral neck, and distal forearm than patients with normalized PTH levels. Their BMD increased at all measured sites in the D+ group, whereas in the D– group BMD did not improve in the forearm (UD and 33% radius), without regard to vitamin D status at baseline.

Table 4. BMD 6 Weeks Before Surgery (Baseline) and the Median Change After 1 Year of Study Medication
 Vitamin D+Vitamin D–Vitamin D+ versus D–
BaselineChange at 1 yearBaselineChange at 1 yearBaseline1 year
MedianIQR%IQR (%)paMedianIQR%IQR (%)papbpb
  1. BMD = bone mineral density; IQR = interquartile range; UD = ultradistal.
  2. aWilcoxon signed rank sum test, paired data.
  3. bMann-Whitney U test, unpaired data.
BMD lumbar spine (g/cm2)1.0670.951 to 1.2423.60.5 to 6.0<0.0011.0420.938 to 1.1823.00.5 to 6.2<0.0010.1800.839
Z-score0.1−0.8 to 0.8   −0.4−1.2 to 0.3     
T-score−0.9−1.9 to 0.3   −1.1−2.2 to 0     
BMD hip, total (g/cm2)0.9150.823 to 1.0182.81.5 to 4.7<0.0010.8890.797 to 0.9712.11.2 to 4.3<0.0010.1940.376
Z-score−0.2−0.9 to 0.5   −0.4−1.0 to 0.3     
T-score−0.9−1.6 to −0.1   −1.1−1.9 to −0.4     
BMD femoral neck (g/cm2)0.8520.773 to 0.9483.21.0 to 4.9<0.0010.8450.749 to 0.9402.30.3 to 4.0<0.0010.5490.092
Z-score−0.3−0.9 to 0.2   −0.5−1.0 to 0.2     
T-score−1.4−1.9 to −0.6   −1.3−2.0 to −0.7     
BMD radius UD (g/cm2)0.4050.330 to −0.4582.0−1.7 to 5.40.0130.3710.331 to 0.4401.1−2.2 to 5.10.0910.1920.449
Z-score−1.0−1.9 to 0.3   −1.3−2.1 to −0.7     
T-score−1.7−3.0 to −0.5   −2.2−3.1 to −1.3     
BMD radius 33% (g/cm2)0.8010.694 to 0.8980.2−2.0 to 3.20.5290.7660.656 to 0.9010.3−1.7 to 2.70.3810.3910.911
Z-score−0.2−1.2 to 0.5   −0.5−1.5 to 0.0     
T-score−1.2−2.2 to −0.2   −1.5−2.6 to −0.5     

When comparing patients with or without 25-OH-D <50 nmol/L, an improvement in the distal radius was seen solely in the group with low vitamin D levels; otherwise the changes in BMD did not differ (Supplementary Figure 1). No significant additive effect from vitamin D substitution was observed.

Not all patients improved their BMD; in the lumbar area, 77 patients (60 women) increased their BMD above the least significant change (LSC = 0.026 g/cm2), 40 (33 women) remained stable, and 14 (13 women) had a significant decrease in lumbar spine BMD at follow-up. In the hip, the BMD increased above the LSC (≥0.028 g/cm2) in 52 cases (42 women), remained stable in 74 (60 women), and 4 (3 women) had lower values. In 33% radius and radius UD, the BMD was unchanged according to the LSC in the majority of patients. Patients with 25-OH-D <50 nmol/L were evenly distributed among these groups. There was no significant difference between the D+ and D– group (data not shown).

The changes in BMD, especially in the hips, were correlated with the baseline concentrations of PTH, ionized calcium, and bone turnover markers, but not with vitamin D (data not shown). When controlling for age, gender, smoking, weight, and creatinine, the correlation between PTH and ΔBMD remained (r = 0.38, p < 0.001).

After PTX and study medication the number of patients with osteoporosis decreased from 30 to 24 in the D+ group and from 39 to 32 in the D– group. The increase in BMD did not differ either between patients with or without osteoporosis, or between men and women (data not shown).

Discussion

The major finding of this double-blinded randomized trial was that vitamin D supplementation had no obviously beneficial effect on bone recovery after PTX except for a minor improvement of radius BMD. The circulating concentrations of PTH and the bone turnover marker βCTX predicted the improvements in BMD, preferentially seen in the lumbar and hip areas. To our knowledge, this is the first reported randomized double-blind study evaluating the effect of vitamin D supplementation in PHPT patients after successful parathyroid surgery. The strengths of this interventional study are the prospective randomized design, the close and standardized follow-up with good compliance, and the achievement of adequate vitamin D levels in the D+ group. Furthermore, the diagnosis was verified by PTX in all cases.

The increase in BMD did not differ between patients with or without low vitamin D levels. Neither did the increase in BMD differ either between patients with or without osteoporosis or between genders. Instead, the change in BMD correlated to the baseline circulating concentrations of PTH, ionized calcium, and bone markers. The high incidence of low vitamin D levels among our PHPT patients is in accordance with a previous study among Swedish patients with mild PHPT.[13] The prevalence of low vitamin D levels may differ from PHPT patients from other geographic areas, although similar figures have been reported from other parts of the world.[21, 22, 28, 29]

The clinical importance of persistent PTH elevation after curative PTH remains to be explored.[30] Several factors are probably causally involved in normocalcemic PTH elevation. One such factor is the interval after PTX. Another is secondary hyperparathyroidism as a result of low vitamin D values, which was the case in some, but not all patients. High postoperative levels of PTH have been associated with larger adenomas and high preoperative PTH, as in our patients, and might be due to an increased need for calcium in the remineralization of the bone or to an increased peripheral resistance to PTH.[31, 32] The high percentage of PTH elevations 6 weeks after successful PTX in association with high concentrations of bone turnover markers and improvement in BMD at all measured sites supports the theory of an adaptive role of PTH in bone mineralization. Except for 1 patient with biochemically persistent or recurrent disease, the explanation for the cause and clinical importance of the persistent PTH elevation in nearly 20% of our patients more than 1 year after PTX is more complicated. Our results are in line with a long-term follow-up of 99 patients after PTX, in which 19% had high PTH at 1 year and 16% at 5 years after PTX.[23] A relative vitamin D insufficiency may be present even with vitamin D levels above 50 nmol/L. The relationship between the circulating concentrations of PTH and vitamin D was recently revised.[17] In contrast to earlier findings, the PTH levels continue to show a decline even when the 25-OH-D level is set at 75 nmol/L.

Based on recently published data, vitamin D supplementation with ≥800 IU per day was found to be favorable in the prevention of hip and nonvertebral fractures in persons 65 years of age or older; a 25-OH-D concentration above 60 nmol/L was recommended.[33] It is not clear to what extent the negative effects of low vitamin D values on bone are mediated by elevation of PTH. We cannot exclude the possibility that the lower PTH concentration in the D+ group will have a positive effect on bone recovery in the long run. PTH has both anabolic and catabolic effects on bone. Histomorphometric studies have shown increased remodeling and imbalance between the amount of bone removed by osteoclasts and the amount replaced by osteoblasts.[34, 35] The high levels of PTH in PHPT are thought to increase bone turnover by 50% to 60%.[36] Studies of BMD, using DXA, microcomputed tomography, and analyses of iliac crest bone biopsies in patients with PHPT, show that cortical bone undergoes reductions of cortical width and porosity which recover after PTX, whereas the cancellous bone is relatively preserved.[37, 38] Recently, Stein and colleagues[39] used high-resolution peripheral quantitative computed tomography (HR-pQCT) at the radius and tibia, and found both trabecular and cortical abnormalities, resulting in decreased whole bone and trabecular stiffness. HR-pQCT accordingly demonstrated improvements in the microarchitecture after PTX in both cortical and trabecular bone.[1] In accordance with our results, the potential for improvements in the microarchitecture and bone strengths were related to the baseline levels of PTH and bone turnover markers. In patients with PHPT and low vitamin D levels, higher PTH and greater catabolic effects were observed in cortical bone and greater anabolic effects were observed in cancellous bone.[19] Whether these findings were due to a pure PTH effect or to a combination of high PTH and low vitamin D is not known. These findings are consistent with the correlation between ΔBMD and the baseline circulatory concentration of PTH and bone markers, observed by us and others.[4, 40, 41] The behavior of the markers of bone turnover—with an earlier decrease of βCTx, a marker of bone resorption, and a later decrease of P1NP, an anabolic marker—is expected considering the remodeling space and is consistent with previous reports.[11, 12] There is a possibility that vitamin D supplementation has a beneficial effect in certain subgroups, for example those with a high PTH level after PTX; they showed a greater improvement in BMD and a beneficial effect in the forearm from vitamin D supplementation. The entire group with vitamin D supplementation also had a positive effect on BMD of the UD radius. In a Danish study on patients with PHPT, high levels of 1,25(OH)2D were inversely correlated to BMD in the distal radius.[42]

This raises the question of whether vitamin D and/or PTH have differential effects on different skeletal compartments; eg, weight-bearing and non-weight-bearing skeletal sites or cortical versus cancellous bone.

A limitation of our study is the use of calcium carbonate instead of placebo. It is conceivable that the calcium supplementation interfered with the results. However, it is also important to ensure a sufficient calcium intake to enable optimal mineralization. Our results are comparable to previously reported effects of parathyroid surgery only, confirming the positive effects on BMD in sites rich in cancellous bone, such as the lumbar spine and total hip.[5, 43-46] Another limitation may be the time interval between operation and randomization. We chose this time of randomization to make sure that the patients were cured before starting the study medication. The very rapid decline in serum βCTx could indicate that a shorter interval might have been favorable for the early mineralization of the bone in some patients. By using an immunoassay method for the measurement of the 25-OH-D concentration, the prevalence of low vitamin D levels in our cohort may be overestimated. However, although spectrometry analysis may have the advantage of potential greater accuracy, immunoassay is still the most commonly used method.[47, 48] Another limitation may be the risk of type 2 error, considering the ΔBMD variations within the groups and the precision error for bone density measurements. We cannot exclude effects of vitamin D supplementation in subgroups of PHPT patients.[46, 49] Neither can we rule out the possibility that a higher dosage of vitamin D or a prolonged period of study medication would have affected the outcome.

Conclusion

BMD increased after PTX and the improvements were related to the baseline concentrations of PTH and bone markers. Despite a high incidence of low vitamin D levels in the PHPT cohort, we observed no obviously beneficial effects of vitamin D supplementation on BMD recovery. However, vitamin D may have a positive effect on BMD in certain subgroups, such as patients with a persistent high PTH concentration after surgery. Our results support a correlation between bone recovery and the preoperative level of PTH.

Disclosures

All authors state that they have no conflicts of interest.

Acknowledgments

This work was funded by Novo Nordisk Foundation, Wiberg Foundation, Capio, Magnus Bergvall Foundation, Fredrik and Ingrid Thuring Foundation, Karolinska Institutet. Recip AB provided the study medication but was not involved in the design, conduction, implementation, data analysis, or manuscript writing. We express our sincere gratitude to the research nurses Wiveca Åberg, Agneta Eriksson, and Lisa Ånfalk, for taking care of the patients in a professional manner. We also thank all other staff members who contributed to the study.

Authors' roles: Study design: SN, YP, JZ, JN, MS, FG, and I-LN. Study conduct: SN, YP, JZ, JN, MS, and I-LN. Data collection: SN, YP, JZ, JN, MS, and I-LN. Data analysis: SN, I-LN, and FG. Data interpretation: SN, YP, JZ, FG, and I-LN. Drafting manuscript: SN, I-LN, and YP. Revising manuscript content: SN, YP, JZ, JN, MS, FG, and I-LN. Approving final version of manuscript: SN, YP, JZ, JN, MS, FG, and I-LN. SN, I-LN, and YP take responsibility for the integrity of the data analysis.

Ancillary