| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Division of Mineral Metabolism (C.R.P., D.J.H.) and Departments of Clinical Chemistry (P.J.B.) and Radiology (K.J.F.), City Hospital, Nottingham NG5 1PB, United Kingdom
Address all correspondence and requests for reprints to: Cornelle R. Parker, M.D., Airedale General Hospital, Skipton Road, Steeton, Keighley, West Yorkshire BD20 6TD, United Kingdom. E-mail: cparker{at}doctors.org.uk.
Abstract
We investigated the effect of alendronate on calcium, PTH, and bone mineral density in 27 female and 5 male patients with primary hyperparathyroidism. The treatment group [n = 14; T score 
-2.5 SD at the femoral neck (FN) or T -1.0 SD plus previous nonvertebral fracture] was given alendronate 10 mg/d for 24 months. The second group (n = 18; T score > -2.5 SD at the FN) was untreated. Biochemistry was repeated at 1.5, 3, 6, 12, 18, and 24 months, and dual-energy x-ray absorptiometry at 12 and 24 months.
There were no significant between-group baseline differences in calcium, creatinine, or PTH. Alendronate-treated patients gained bone at all sites [lumbar spine (LS), 1 yr gain, +7.3 ± 1.7%; P < 0.001; 2 yr, +7.3 ± 3.1%; P = 0.04). Untreated patients gained bone at the LS over 2 yr (+4.0 ± 1.8%; P = 0.03) but lost bone elsewhere. Calcium fell nonsignificantly in the alendronate group between baseline (2.84 ± 0.12 mmol/liter) and 6 wk (2.76 ± 0.09 mmol/liter), with a nonsignificant rise in PTH (baseline, 103.5 ± 14.6 ng/liter; 6 wk, 116.7 ± 15.6 ng/liter). By 3 months, values had reverted to baseline.
In primary hyperparathyroidism, alendronate is well tolerated and significantly improves bone mineral density at the LS (with lesser gains at FN and radius), especially within the first year of treatment. Short-term changes in calcium and PTH resolve by 3 months.
THERE HAVE BEEN several different approaches to the medical management of primary hyperparathyroidism. Estrogen therapy has been shown to normalize serum calcium without significant alterations in PTH (1, 2, 3), and short-term studies with pamidronate (4) and risedronate (5) have assessed the efficacy of bisphosphonates in lowering serum calcium, elevating PTH, and reducing bone resorption, but often at the expense of a further rise in PTH. Only one recent study has examined the long-term effects of a bisphosphonate in primary hyperparathyroidism with bone mineral density (BMD) as an end-point. Twenty-six elderly patients with primary hyperparathyroidism were randomized to alternate-day oral alendronate (10 mg) or no treatment for 2 yr (6). There were significant reductions in serum calcium, phosphate, and urinary calcium excretion at 3 months, with a significant rise in PTH during the first year of treatment. Reported gains in BMD at the lumbar spine (LS) and total hip at 2 yr were 8.6% and 4.8%, respectively.
The aims of our study were 3-fold: 1) to investigate the effects of alendronate on BMD, 2) to assess the biochemical responses to long-term oral alendronate therapy in primary hyperparathyroidism, and 3) to evaluate the safety of alendronate in this population.
Subjects and Methods
Subjects
We invited 63 patients with primary hyperparathyroidism under regular review in our bone metabolism outpatient clinics to attend for osteoporosis risk assessment. This comprised our entire outpatient population of primary hyperparathyroid patients with the exception of one patient awaiting parathyroidectomy. At the time of recruitment, it was departmental policy to review all patients with primary hyperparathyroidism in our bone metabolism clinics on a long-term basis rather than actively discharging patients to primary care. All patients had elevated serum calcium and PTH, confirming the diagnosis of primary hyperparathyroidism.
Forty-five patients attended for the initial screening visit. We asked patients a series of questions assessing osteoporotic risk through an in-house questionnaire ascertaining prior low-trauma fracture, family history of fracture, exposure to drugs associated with increased risk of osteoporosis (especially steroids), past or current therapy with bone modifying agents, and, for female patients, menstrual history.
We excluded two patients who were on bisphosphonate therapy (both female), one female patient unwilling to discontinue HRT for the duration of the study, and one female patient with bilateral hip replacements. Two patients with serum creatinine concentrations greater than 160 µmol/liter were excluded, because alendronate was unlicensed for serum creatinine concentrations in excess of approximately 180 µmol/liter. Patients were not excluded if they had a history of either dyspepsia or renal tract calculi, or were treated with thiazide diuretics or inhaled glucocorticoids. Six patients declined to participate, and one patient allocated to alendronate treatment withdrew at 6 wk due to dyspepsia and was excluded from analysis. Altogether, the present study is based on 32 patients (27 female, 5 male) with primary hyperparathyroidism. Informed consent was obtained, and local ethical committee approval was granted.
We divided patients into two groups depending on baseline BMD and the presence of a prevalent fracture. The treatment group (n = 14) had a femoral neck (FN) T score of -2.5 SD or less (n = 12), or a FN T score of -1 SD or less plus a low-trauma (nonvertebral) fracture (n = 2). This group was given open-label oral alendronate 10 mg/d for 24 months. The untreated group (n = 18) had a T score above -2.5 SD at the FN (Table 1
).
|
Fracture assessment
Prevalent and incident nonvertebral fractures were documented through self-report and confirmed by x-ray. Vertebral fractures were identified by plain radiography. Only low-trauma fractures, that is fractures occurring as a result of a fall from a standing height or less, were included.
Radiology
All patients had radiographs of the thoracolumbar spine (anteroposterior and lateral views) to assess the prevalence of vertebral fractures and to look for spondylotic change or aortic calcification. Vertebral fractures were classified as anterior, central, or posterior and graded according to the percentage reduction in vertebral height: grade 1, 0–25%; grade 2, 25–50%; grade 3, 50–75%; grade 4, 75–100% (7). Patients were classified as having minimal, mild, or moderate spondylosis. Because this was predominantly an elderly population, individuals were not excluded if they had evidence of osteoarthritic change. Plain pelvic radiographs (to include both hip joints) were also performed to exclude artifactual influence on BMD measurements.
BMD
Patients underwent DEXA using the Hologic QDR 2000 densitometer (Hologic, Inc., Waltham, MA) at the LS (L1-L4 in AP direction), hip, total hip, FN, intertrochanteric, ultradistal radius (UDR), and mid-shaft radius. BMD was expressed in grams per square centimeter as SD units as T scores (comparison with young adult mean), or as Z scores (comparison with age-matched mean). Precision errors, established with a local normal population (females: age range, 45–59 yr; spine and hip sites, n = 410; radial sites, n = 277), were as follows: L1–4, 1.48%; total hip, 0.99%; FN, 1.77%; mid-radius, 1.07%; UDR, 1.82%; and distal third radius, 1.17%. Precision errors for male subjects were based on the manufacturers’ data. Spinal scans were standardized daily against a phantom with a precision error of ±1.5%.
Biochemistry
Serum samples were taken in the nonfasting state. Patients brought two sequential 24-h urine collections completed on the morning of clinic attendance and two second-void fasting urine samples to each clinic visit. A mean of the paired samples was taken.
Serum bone-specific alkaline phosphatase (BALP) was determined by wheat germ lectin precipitation (Sigma-Aldrich Co. Ltd, Poole, Dorset, UK) Laboratory-derived intra-assay coefficient of variation (CV; n = 36) is 5.6% at 91.8 U/liter and 8.5% at 142 U/liter. Interassay CV (n = 36) is 9.9% at 139 U/liter and 10.2% at 223 U/liter.
Serum osteocalcin was measured by two-site immunoradiometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). Manufacturer-derived intra-assay CV (n = 20) is 5.2% at 1.6 µg/liter and 3.9% at 14.2 µg/liter. Interassay CV (n = 20) is 6.7% at 1.5 µg/liter and 5.5% at 14.2 µg/liter.
Total hydroxyproline (Hypro) was assayed in urine by HPLC. Laboratory-derived intra-assay CV (n = 20) is 4.1% at 30 µmol/liter, 3.5% at 100 µmol/liter, 3.0% at 500 µmol/liter, and 3.2% at 2000 µmol/liter. Interassay CV is 10.7% at 110.9 µmol/liter (n = 38) and 17.9% at 587.5 µmol/liter (n = 32).
Free deoxypyridinoline (DPyd) was measured on the Chiron Diagnostics ACS:180 automated immunoassay analyzer (Chiron Diagnostics Ltd., Halstead, Essex, UK). Manufacturer-derived intra-assay CV is 3.7% at 28.2 nmol/liter, 1.8% at 75.5 nmol/liter, and 1.1% at 272.9 nmol/liter. Interassay CV is 10.0% at 28.2 nmol/liter, 7.3% at 75.5 nmol/liter, and 4.0% at 272.9 nmol/liter.
Intact PTH (reference range, 12–72 ng/liter) was measured on a DPC-Immulite automated immunoassay analyzer (Diagnostics Products, Los Angeles, CA). Laboratory-derived intra-assay CV is 3.4% at 56.9 ng/liter and 4.3% at 142.8 ng/liter (n = 20). Interassay precision (derived between January and July 1999) is 6.6% at 44.1 ng/liter (n = 15) and 5.8% at 391.1 ng/liter (n = 15).
Routine biochemical measurements including serum creatinine, calcium, and phosphate, and 24-h urinary calcium were measured using an Olympus AU 800 automated multichannel analyzer (Olympus Optical Company, UK Ltd., Hampshire, UK). Calcium was corrected for serum albumin concentration. Samples were taken in the nonfasting state.
Statistical analysis
Changes in BMD were expressed as a percentage of the baseline value (measured in grams per square centimeter). Paired and unpaired t tests were used to detect differences in BMD within and between groups, respectively. Serial biochemical data were analyzed by ANOVA for repeated measures. 
2 analysis was used to identify between-group differences in the prevalence of dyspepsia, exercise, smoking, and prior low-trauma fracture. Fisher’s exact test was used to test for between-group differences for dichotomous variables. Pearson correlation coefficients tested associations between markers and between baseline BMD (grams per square centimeter) and percentage change in absolute BMD at 12 and 24 months.
A P value of less than 0.05 was considered to be significant. All results are two-tailed, and results are expressed as means (±SEM).
Results
Table 2 compares available data for the 31 patients with primary hyperparathyroidism not included in the study (either through exclusion or declining to participate) with the study population. Comparison BMD data are only available for those patients who attended the initial screening visit (n = 13; n = 12 for FN results because one patient had bilateral hip replacements). There were no significant differences between the two populations in age, BMD, PTH, serum calcium, or creatinine.
|
Biochemical data are complete for all subjects for the first 12 months. During the second 12 months, patients withdrew as indicated above; in addition, one patient from each group was not sampled at 18 months. At 24 months, there were 10 complete sets of serum biochemical results (9 urine) in treated patients and 13 (12 urine) in untreated patients.
Five patients allocated to the treated group and four to the untreated group had evidence of moderate spondylotic change in the thoracic and/or LS on plain x-ray. There were three subjects in the treated group and four in the untreated group with mild spondylotic changes.
Baseline characteristics for the two groups are shown in Table 3. The treated group was significantly further past the menopause and had greater self-reported height loss. Two female patients (one from each group) had been on estrogen replacement, but this was discontinued on study commencement.
|
. There was a significant difference in total numbers of fractures at baseline between the two groups (P = 0.001), but the difference in baseline vertebral fracture prevalence did not reach statistical significance (P = 0.07). There were no significant differences between groups at baseline in PTH, calcium, phosphate, creatinine, and 24-h urinary calcium excretion.
BMD (Fig. 1)
In both groups, there was a greater reduction in baseline BMD at cortical sites, with the lowest values at the mid-radius, intermediate reductions at the FN, and the smallest reduction at the LS (treatment group: mid-radius, mean T score, -4.20 ± 0.35 SD; FN, mean T score, -3.77 ± 0.19 SD; LS, mean T score, -2.66 ± 0.32 SD).
|
Because of the discrepancy in baseline BMD between groups, we investigated the relationship between baseline BMD (grams per square centimeter) and percentage change in absolute BMD. We found statistically significant correlations at the FN for the treatment group at 12 months (P = 0.003) and 24 months (P = 0.04) and at the UDR at 24 months in the nontreatment group (P = 0.02).
We examined the relationship between change in BMD (expressed both as percentage change and as absolute difference in BMD (grams per square centimeter) over the first 12 months of the study) and percentage change in the biochemical markers. We found no significant associations in alendronate-treated patients (except for percentage change in DPyd at the mid-radius). In untreated patients, there were significant negative associations between BALP and Hypro and percentage change in hip BMD. This relationship remained significant at the intertrochanteric and total hip sites only for BALP when BMD changes were expressed in absolute terms (grams per square centimeter).
There were two male patients in the treated group and three in the untreated group. There were no significant differences in baseline T scores between male and female subjects, with the exception of LS T score being significantly higher at baseline in male patients in the untreated group. However, subsequent change in BMD at the LS did not differ significantly between groups.
In treated patients, there were no significant differences in change in BMD over the 2 yr between male and female patients. In untreated subjects, there was a significantly greater increase in FN BMD in male patients in the first (but not the second) year and also a significantly greater reduction in BMD in male patients at the mid-radius in both the first and second years of the study.
Serum calcium, phosphate, PTH, creatinine, and 24-h urinary calcium excretion (Fig. 2, A–D)
In the alendronate-treated group, mean serum calcium fell between baseline (2.84 ± 0.03 mmol/liter) and 6 wk (2.77 ± 0.02 mmol/liter), with a corresponding rise in mean PTH over the same time period (baseline, 103.5 ± 14.6 ng/liter; 6 wk, 116.7 ± 15.6 ng/liter), however these differences failed to reach statistical significance. By 3 months, mean serum calcium and PTH had reverted to baseline values. Serum calcium remained stable for the remainder of the study, but there was a gradual nonsignificant rise in PTH between 3 and 18 months with an increment of approximately 40 ng/liter. Patient withdrawal by 24 months (n = 10 remaining) made the apparent increase in calcium and decrease in PTH at this time point less reliable. Phosphate showed a nonsignificant decline at 3 and 12 months, with a subsequent return to baseline over the second year of the study. In the untreated group, serum calcium, PTH, and phosphate showed no statistical change over 2 yr from baseline.
|
Biochemical markers of bone turnover (Fig. 3, A–D)
There were no significant correlations across all patients at baseline between the two markers of bone formation or the two markers of bone resorption (data not shown). When examining cross-correlations between markers of bone formation and resorption, the only relationship that reached statistical significance was between osteocalcin and Hypro (r = 0.44; P = 0.01).
|
In the untreated group, there was less fluctuation by comparison with baseline values. The only significant change was for osteocalcin with a reduction at 12 months compared with baseline (P = 0.05). Although there was a decline in BALP between 3 and 12 months, this difference was not statistically significant.
Between-group comparisons demonstrated significant differences between the two groups at baseline only for serum osteocalcin (P = 0.04). Bone formation was significantly lower in treated patients at 3 months (P = 0.003 for osteocalcin, 0.009 for BALP). Hydroxyproline followed a similar trend, indicating a decline in resorption in alendronate-treated patients (P = 0.01 at 3 months). These differences remained significant for all three markers over the first 12 months of the study. Deoxypyridinoline showed lower mean values in treated patients, but the differences between groups did not reach statistical significance at any time point.
Safety and tolerability
During the course of the study, four patients on alendronate developed new or worsening dyspepsia, but all were able to continue with treatment. Symptoms resolved spontaneously in two patients, lasting 2 months and 13 months, respectively. In one patient, dyspepsia continued for the duration of the study but was tolerated and did not necessitate discontinuation. One patient, who withdrew from the study after 18 months, developed severe dyspepsia and an iron-deficient anemia after 6 months of treatment. Upper gastrointestinal endoscopy was normal, except for the presence of Helicobacter pylori. The patient continued with the study drug for a further 12 months but then withdrew from the trial because of persistent dyspepsia. One patient developed a generalized macular rash within days of starting alendronate, which was well tolerated but persisted for the duration of the study. Alendronate was discontinued in one patient (severe dyspepsia) who was withdrawn from the study within the first 6 wk and excluded from analysis.
In the nontreatment group, one patient developed worsening dyspepsia over the course of the study.
Discussion
Bisphosphonates have been avoided in the past in the management of primary hyperparathyroid-related osteoporosis, largely because of concern about hypocalcemia with a subsequent further elevation of PTH and amplification of bone loss.
The 32 patients who participated in the study did not differ in age, calcium, PTH, creatinine, or in a more limited comparison, spinal or FN BMD, from the remainder of our local population of patients with primary hyperparathyroidism.
Our study subjects were divided into two groups according to BMD and fracture history to assess the effects of bisphosphonate treatment on those individuals with the lowest BMD (± low trauma nonvertebral fractures). It is acknowledged that bisphosphonates improve BMD most effectively in osteoporotic patients with very low BMD (8, 9). Our ability to make valid intergroup comparisons for BMD is limited in that it would be expected that the patients allocated to treatment might gain BMD at above-average rates in response to bisphosphonate therapy. However, the only instance of baseline BMD influencing response to alendronate therapy was identified at the FN at 12 and 24 months. The correlation between baseline BMD and change in BMD described in the untreated group only reached significance at 24 months, with no trend toward significance at 12 months, and is therefore of doubtful relevance.
Difficulties also arise in female patients when trying to determine the relative contribution to reduced bone mass made by postmenopausal age-related changes and by primary hyperparathyroidism. Similar proportions of female patients in each group were postmenopausal, but those in the treatment group were significantly further past the menopause (P = 0.03), and this may in part account for their lower BMD. Although both pathogenetic mechanisms will contribute to reduced BMD, providing there is an improvement in bone density in response to a bisphosphonate, it is of limited practical consequence which mechanism predominates.
Because our study included a small number of male patients, BMD results were also analyzed with these patients excluded. This made no significant difference in treated patients but did influence the untreated group. The significantly greater gain at the FN in male patients during the first year suggests that losses in female patients are underestimated in our results. Similarly, greater reductions at the mid-radius throughout the 2 yr of the study would suggest our results overestimate losses at this site.
In accordance with observations made by other investigators (3, 8, 9), in both groups of patients BMD was predominantly reduced at cortical sites (mid-radius), with relative preservation at the LS. Patients on alendronate gained bone compared with baseline at all measured sites over the study period, although these only reached statistical significance at the LS (+7.3 ± 1.7%; P < 0.001) at 1 yr. In common with bisphosphonate treatment of postmenopausal osteoporosis (10), the largest gains in bone mass were evident at the trabecular-rich LS by comparison with the FN, with the mid-radius showing the poorest response. Thus, sites with the lowest baseline BMD and potentially most to gain in terms of fracture reduction show the least benefit from treatment.
During the second year of alendronate treatment, lumbar spinal BMD stabilized at a level similar to that at 12 months, but there was a trend toward a progressive gain at hip and mid-radius suggesting that more prolonged therapy may benefit cortical sites. The gain of 7.26% during the first year at the LS with alendronate treatment exceeds that of 5.42% reported by Liberman et al. (10) in alendronate-treated postmenopausal women. At 2 yr, our patients also compare favorably (7.31% vs. 7.06%). Our results are similar to those of Rossini et al. (6) who document gains of 8.6% at the LS over 2 yr in alendronate-treated elderly patients with primary hyperparathyroidism. The same authors report lesser gains at the total hip (4.8%, compared with 2.1% in our patients). These gains are to be expected, because the tendency in postmenopausal osteoporotic patients (10, 11, 12) is for bisphosphonates, through in-filling of the resorption space, to induce the greatest increments in BMD during the first year of therapy. The size of the resorption space is directly proportional to the rate of bone remodeling, so higher rates of bone turnover in primary hyperparathyroidism result in greater gains in BMD. The resorption space diminishes as bone turnover is suppressed by bisphosphonate therapy, and contraction of this space would explain the reduced response in the second year of treatment. Another contributory factor may be earlier escape from bisphosphonate-induced suppression of bone turnover in primary hyperparathyroid patients due to the increased numbers of bone remodeling units activated.
The natural history of untreated primary hyperparathyroidism may be an initial decline in LS BMD and subsequent maintenance of BMD because of an anabolic effect of PTH on trabecular bone (9). This would appear to be supported by our results in untreated subjects who gained 2.8% in spinal BMD over 2 yr. The effects of PTH on bone are complex and appear to differ in trabecular and cortical bone. With the advent of PTH as a therapeutic agent, these effects are being further elucidated. Intermittent dosing with PTH has an anabolic effect on trabecular bone leading to gains in trabecular bone mass (13, 14). In primary hyperparathyroidism, trabecular bone is also increased with increased resorption of cortical bone (15), offering an explanation for the differences between BMD at vertebral and cortical-rich sites.
In the alendronate-treated group, the decline in serum calcium between baseline and 6 wk occurs as a result of inhibition of osteoclast-mediated bone resorption, with a corresponding rise in PTH, a pattern described by a number of authors (4, 5, 6, 16). Although serum calcium returned to the baseline level by 3 months and remained stable, there was an unexpected, nonstatistically significant, further rise in PTH. Interestingly, Rossini et al. (6) also observed a gradual rise in PTH over the first year of treatment. This increase in PTH may account for the observed decline in serum phosphate over the initial 12 months of the current study in the presence of stable calcium homeostasis (serum and urinary calcium). In untreated patients, serum calcium remained unchanged throughout the study, but PTH although initially stable, again showed a gradual (nonsignificant) increase between 3 and 18 months. Renal function, as assessed by serum creatinine, was unchanged in both groups, but 24-h urinary calcium excretion although unchanged in alendronate-treated patients fell at 12 and 18 months in untreated patients. Reduced dietary calcium intake may account for the variation in urinary calcium excretion and PTH in untreated patients.
Although we found no baseline correlation between osteocalcin and BALP, or between Hypro and DPyd, the markers (with the exception of DPyd) performed consistently in response to treatment. Although these changes rarely achieved statistical significance, certain patterns emerged. We noted reductions in bone formation markers (reaching statistical significance for BALP) and the resorption marker urinary Hypro in alendronate-treated patients at 6 wk, with a further progressive decline until 12 months. Values then gradually rose again and reached baseline levels for osteocalcin and Hypro by 18 months, in keeping with the serum calcium and PTH responses to treatment noted above. They reflect inhibition of osteoclastic bone resorption and, through a resultant reduction in available growth factors, subsequent inhibition of osteoblast-mediated formation. The increase between 12 and 24 months may reflect escape from the effect of bisphosphonate therapy. Urinary DPyd behaved differently from the other markers, with little variation in treated patients. One possible explanation is that short-term bisphosphonate treatment results in markedly decreased excretion of the peptide-bound cross-link fraction, with no effect on excretion of free pyridinolines (17, 18).
An additional primary aim was to assess safety and tolerability of alendronate in patients already predisposed, as a result of their primary hyperparathyroidism, to the principal alendronate-related adverse effect of dyspepsia. Considering the high pretreatment prevalence of dyspepsia (9 of 14 treatment patients, and 12 of 18 nontreatment patients) and the open-label design of the study, alendronate was well tolerated. As anticipated, the major side effect from therapy was dyspepsia, but despite being reported by four patients, this only necessitated discontinuation of alendronate in one case. In addition, one patient not included in our final analysis withdrew because of severe indigestion before 6 wk had elapsed.
In summary, alendronate treatment of primary hyperparathyroid-related osteoporosis results in short-term fluctuations in serum calcium and PTH that return to baseline values by 3 months. Therapy with alendronate substantially improves BMD at the LS, but virtually all of this gain appears to occur within the first year of treatment. There are lesser gains at hip and radial sites. It is disappointing that benefit is greatest at the LS rather than the cortical-rich sites, which need it most. However, alendronate may have a role in those patients with reduced BMD who are not suitable candidates for parathyroidectomy. A larger scale randomized placebo-controlled study would help determine whether there is a sustained, significant effect at sites other than the LS. Alendronate is also generally well tolerated in patients with primary hyperparathyroidism, many of whom have preexisting dyspepsia, and does not adversely affect renal function.
Acknowledgments
We thank the patients with primary hyperparathyroidism who participated in this study. We also thank the Department of Bone Densitometry, City Hospital (Nottingham, United Kingdom).
Footnotes
Abbreviations: BALP, Bone-specific alkaline phosphatase; BMD, bone mineral density; CV, coefficient of variation; DEXA, dual-energy x-ray absorptiometry; DPyd, deoxypyridinoline; FN, femoral neck; Hypro, hydroxyproline; LS, lumbar spine; ns, not significant; UDR, ultradistal radius.
Received March 7, 2001.
Accepted June 27, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  G. J. Strewler A 64-Year-Old Woman With Primary Hyperparathyroidism JAMA, April 13, 2005; 293(14): 1772 - 1779. [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Khan, J. P. Bilezikian, A. W. C. Kung, M. M. Ahmed, S. J. Dubois, A. Y. Y. Ho, D. Schussheim, M. R. Rubin, A. M. Shaikh, S. J. Silverberg, et al. Alendronate in Primary Hyperparathyroidism: A Double-Blind, Randomized, Placebo-Controlled Trial J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3319 - 3325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Bilezikian and S. J. Silverberg Asymptomatic Primary Hyperparathyroidism N. Engl. J. Med., April 22, 2004; 350(17): 1746 - 1751. [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 |
