Survival after the Diagnosis of Hyperparathyroidism: A Population-based Study1

CLINICAL STUDIES

Survival after the Diagnosis of Hyperparathyroidism: A Population-based Study* Robert A. Wermers, MD, Sundeep Khosla, MD, Elizabeth J. Atkinson, MS, Clive S. Grant, MD, Stephen F. Hodgson, MD, W. Michael O’Fallon, PhD, L. Joseph Melton, III, MD BACKGROUND: Reports of increased mortality from cardiovascular disease and malignancy in primary hyperparathyroidism have been based primarily on patients who have undergone parathyroidectomy. In order to assess the true impact of primary hyperthyroidism on mortality in the general population, we assessed survival in a large inception cohort of Rochester, Minnesota residents with primary hyperparathyroidism initially diagnosed over a 28-year span, the majority of whom were followed with uncomplicated disease. METHODS: All Rochester residents with primary hyperparathyroidism first recognized in 1965 to 1992 were identified through the Rochester Epidemiology Project medical records linkage system. Included as cases were patients with pathologic confirmation of hyperthyroidism, hypercalcemia with inappropriately elevated parathyroid hormone levels, or hypercalcemia for more than a year with no other cause. Survival was estimated using the Kaplan Meier product-limit method. The Cox proportional hazards model was used to determine associations, as relative hazards (RR) with 95% confidence intervals (CI), of various risk factors with time to death.

RESULTS: During the study period, 435 cases of primary hyperparathyroidism were identified. Altogether, parathyroid surgery was performed on 126 patients (29%), with a mean delay between the initial elevated serum calcium level and surgery of 3.3 years. Patients who underwent surgery had higher maximum serum calcium levels than the patients who were observed (mean 6 SD, 11.3 6 0.7 versus 10.7 6 0.4 mg/dL, P ,0.001), but their mean ages were similar (54 6 16 versus 56 6 17 years). Overall survival in the patients with primary hyperthyroidism was better than expected (P 5 0.02), but by ageadjusted multivariate analysis, higher maximal serum calcium level was an independent predictor of mortality (RR 5 1.3 per mg/dL; 95% CI: 1.1–1.6; P ,0.02). CONCLUSION: Overall survival is not adversely affected among unselected patients with mild primary HPT in the community, although patients with more severe disease, as manifested by higher serum calcium levels, may have an increased risk of death. Am J Med. 1998;104:115–122. q1998 by Excerpta Medica, Inc.

T

mild to moderate hypercalcemia who did not undergo surgery (5) and among more seriously affected individuals who underwent parathyroidectomy (6–10). In 1 study, the risk of death appeared to normalize approximately 5 years after surgery, suggesting that primary hyperparathyroidism may cause damage that is not clinically obvious and that surgery, even in mild cases may improve survival (8). The aim of the present investigation was to assess survival in a large inception cohort of Rochester, Minnesota residents with primary hyperparathyroidism initially recognized over a 28-year span, 1965 to 1992. During this period, the introduction of automated serum calcium measurements (in July 1974) produced a sharp increase in the incidence of hyperparathyroidism (11,12), which was accompanied by a change in its clinical spectrum: the proportion of patients with complications at diagnosis fell from 23% in the prescreening era to 2% of patients in 1983 to 1992, while the proportion managed surgically fell from 29% to 13% (12). The large number of patients in our cohort with mild hypercalcemia who were treated with observation allowed us to assess the risk of death in the typical patient seen in today’s clinical practice.

he impact of primary hyperparathyroidism on survival, especially among patients with uncomplicated disease, is controversial. While an increasing number of uncomplicated patients with mildly elevated serum calcium levels are being followed without parathyroidectomy (1), a number of reports have described increased mortality due to cardiovascular disease and malignancy in patients with primary hyperparathyroidism (2–7). An increased risk of death has been seen both among patients with

*Access the ‘‘Journal Club’’ discussion of this paper at: http://www. elsevier.com/locate/ajmselect/ From the Division of Endocrinology/Metabolic Diseases and Internal Medicine(RAW, SK, SFH); Department of Health Sciences Research(EJA, WMO, LJM); and Department of Surgery (CSG), Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Supported in part by Research Grants AG 04875 and AR 30582 from the National Institutes of Health, United States Public Health Service. Requests for reprints should be addressed to Sundeep Khosla, MD, Mayo Clinic, 200 First Street SW, 5-164 West Joseph, Rochester, Minnesota 55905. Manuscript submitted November 26, 1996 and accepted in revised form July 23, 1997. q1998 by Excerpta Medica, Inc. All rights reserved.

0002-9343/98/$19.00 115 PII S0002-9343(97)00270-2

Survival After Hyperparathyroidism/Wermers et al

METHODS Population-based research is feasible in Rochester because medical care is virtually self-contained within the community and there are relatively few providers. Most endocrinologic care, for example, is provided by the Mayo Clinic, which has maintained a common medical record with its 2 large affiliated hospitals (St. Mary’s and Rochester Methodist) for 90 years. Recorded diagnoses and surgical procedures are indexed, including the diagnoses made for outpatients seen in office or clinic consultations, emergency room visits or nursing home care, as well as the diagnoses recorded for hospital inpatients, at autopsy examination or on death certificates (13). Medical records of the other providers who serve the local population, most notably the Olmsted Medical Center (Olmsted Medical Group and its affiliated Olmsted Community Hospital), are also indexed and retrievable. Thus, details of the medical care provided to the residents of Rochester are available for study through this medical records linkage system (the Rochester Epidemiology Project) as described elsewhere (14). Study criteria described in detail elsewhere (12) resulted in an inception cohort of 435 cases of primary hyperparathyroidism recognized among the residents of Rochester during the 28-year period, 1965 to 1992. For each potential case, the complete (inpatient and outpatient) medical record in the community was reviewed by one of us (RAW). Patients were accepted as cases of primary hyperparathyroidism if they met one or more of the following criteria: (1) histopathological confirmation of parathyroid adenoma or hyperplasia; (2) hypercalcemia (normal, 8.9 to 10.1 mg/dL) with an inappropriately elevated serum immunoreactive parathyroid hormone (iPTH) level [.2.1 pmol/L by two-site immunochemiluminometric assay (15) or .20 mL eq/mL by C-terminal radioimmunoassay (16,17) or (3) hypercalcemia for more than 1 year without another evident cause (eg, thiazide diuretics, malignancy, creatinine . 2.0 mg/dL, lithium therapy, etc.). Of the 435 cases identified using these criteria, 104 (24%) were in this last category. In 73 patients, serum PTH levels were not measured by the clinician taking care of the patient. In the remaining 31 patients, serum PTH levels were measured, but these patients had intermittently elevated serum calcium levels, and serum calcium was in the normal range at the time that the PTH level was obtained. Thus, they were categorized in the third group rather than the second. Patients with familial benign hypercalcemia who had previously been identified in an extensive study at our institution were excluded (18,19). In addition, subjects had to have established residency in Rochester for at least 1 year prior to the initial serum calcium elevation, thus minimizing any potential impact of ill patients with unrecognized dis116

February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

ease moving into the community for care at the Mayo Clinic. Since screening for serum calcium was introduced in June 1974, 2 incidence cohorts were created, as previously described (1965 to June 1974 and July 1974 to 1992) (12). Group characteristics such as patient age and gender were compared using the chi-square test. Continuous variables were described using the mean and standard deviation. The Cox proportional hazards model was used to determine the associations of time to death with age, gender, maximum serum calcium, complications of hyperparathyroidism, decision to operate at the time of detection, surgery (yes/no), and time period (1965 to June 1974 vs July 1974 to 1992); results are expressed as relative hazards (RR) with 95% confidence intervals (CI). Surgery (yes/no) was analyzed as a time-dependent variable. Stepwise model selection was used to build a multivariate model. Interactions, higher order terms, and proportional hazards assumptions were examined where appropriate. Lifetable methods were used to estimate survival and cumulative incidence of surgery in this cohort (20). The estimated mortality rates were compared using the one-sample log-rank test. Expected rates were based on the experience of Minnesota white residents, taking into account the age and gender distribution of the hyperparathyroidism subjects. Observed and expected survival was also examined for presurgery time (follow-up was censored at the time of surgery) and postsurgery time (follow-up began at the time of surgery).

RESULTS Clinical Spectrum During the 28-year study period, 1965 to 1992, 435 cases of primary HPT were identified among the residents of Rochester, Minnesota. Almost all of the patients were white, reflecting the racial composition of the community (96% white in 1990). The majority of patients were women (329, 76%) and most were 45 years of age or older (339, 78%). The mean age was 56.1 years (median, 57.3 years; range, 15.8 to 89.4 years). The proportion of women (75% versus 76%) and the mean (6SD) age (53.6 6 16.0 versus 56.5 6 16.3 years) were similar in the prescreening era (1965 to June 1974) before the introduction of automated serum calcium measurements in July 1974 and afterward (July 1974 to 1992). The average maximum serum calcium levels recorded in the 2 time periods were identical (10.9 6 0.5 versus 10.9 6 0.6 mg/dL). Serum calcium levels did not differ in young (less than 40 years) compared with middle aged (40 to 60 years) or elderly (.60 years) women or men (data not shown). The majority of patients had either an inappropriately elevated PTH in the face of hypercalcemia (209 patients,

Survival After Hyperparathyroidism/Wermers et al

Figure 1. Cumulative incidence of parathyroid surgery among Rochester, Minnesota residents with definite primary hyperparathyroidism recognized in the prescreening era (1965 to June 1974) versus the postscreening era (July 1974 to 1992) (P 5 0.11 for difference between the two curves).

48%) or a histologic diagnosis (122 patients, 28%). The proportion of patients who had histologic proof of their diagnosis, however, fell from 38% in 1965 to June 1974 to 26% in July 1974 to 1992 (P 5 0.07). This was related to the decline in the frequency of parathyroidectomy (see below). Among those with pathologic confirmation, the diagnosis was parathyroid adenoma in 91% of cases, with parathyroid hyperplasia making up the remainder. There were also fewer patients who presented with complications of hyperparathyroidism in the second time period (23.0% in 1965 to June 1974 versus 6.7% in July 1974 to 1992, P ,0.0001). Complications included nephrolithiasis (25), osteoporosis or fractures (5), hypercalcemic crisis (5), peptic ulcer disease (2), pseudogout (1), and band keratopathy (1).

through 1992 (P 5 0.02 for the difference in the recommendation for surgery between the 2 time periods). Altogether, there was little difference in the mean age of those patients who were observed (56 6 17 years) compared with patients undergoing parathyroid surgery (54 6 16 years). However, the patients who had surgery had higher maximum serum calcium levels compared with the patients who were observed (11.3 6 0.7 mg/dL versus 10.7 6 0.4 mg/dL, P ,0.001). In addition, a greater proportion of the patients who had surgery had a maximum serum calcium of 11.2 mg/dL or greater (47%) compared with the patients who were observed (13%, P ,0.0001).

Survival Following the initial elevated serum calcium level, the patients were followed for 5680 person-years (1171 person-years among the 1965 to June 1974 cases and 4509 person-years for the July 1974 to 1992 cases). When compared with white Minnesota residents, overall age and gender-adjusted survival was better than expected, with a relative risk of death of 0.69 (95% confidence intervals [CI] 5 0.57– 0.83). For the patients recognized in the

Parathyroid Surgery Parathyroid surgery was ultimately performed on 126 patients (29%). The cumulative incidence of surgery by time after the initial elevated serum calcium level is illustrated in Figure 1. Overall, the cumulative incidence of surgery was 28% by 10 years. However, the likelihood of surgery was somewhat greater in the prescreening era than afterward (at 10 years, 37% versus 27%, P 5 0.11). The difference appeared greatest within the first few years following recognition of the condition (Figure 1). The mean (6SD) time between the initial elevated serum calcium level and the date of surgery, among those who underwent surgery, was 3.3 6 4.0 years. Surgery was recommended as the initial form of management for 43% of the patients in the prescreening era and was scheduled within 6 months after the clinician’s diagnosis of primary hyperparathyroidism in 30%; the other patients refused surgery or were too ill to operate. Surgery was recommended for only 28% of the patients diagnosed in July 1974

Figure 2. Top panel: survival among Rochester, Minnesota residents with primary hyperparathyroidism in the prescreening era (1965 to June 1974) compared with that expected for Minnesota white residents of similar age and gender (P 5 0.71 for difference between observed versus expected survival). Bottom panel: survival among Rochester, Minnesota residents with primary hyperparathyroidism in the postscreening era (July 1974 to 1992) compared with that expected for Minnesota white residents of similar age and gender (P ,0.02 for difference between observed versus expected survival). February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 117

Survival After Hyperparathyroidism/Wermers et al

Table 1. Age and Gender-Adjusted Relative Risk of Death, by Diagnosis, among Rochester, Minnesota Residents with Primary Hyperparathyroidism, Compared with Expected Risk in White Minnesota Residents Cause of Death

Relative Risk (95% Confidence Interval)

Cancer Cardiovascular disease Respiratory disease Gastrointestinal disease All other causes All causes

0.58 (0.39–0.89) 0.60 (0.45–0.79) 0.97 (0.53–1.63) 0.88 (0.29–2.06) 1.00 (0.63–1.51) 0.69 (0.57–0.83)

prescreening era, survival was slightly, but not significantly, better than expected (Figure 2, top panel). Survival was clearly better than expected among the cases first recognized between July 1974 and 1992 (P ,0.02; as shown in Figure 2, bottom panel). When patients from the 2 time periods were compared, however, there was no significant difference in survival between the groups (P 5 0.23). Mortality in patients with primary hyperparathyroidism was lower than expected for cancer and for cardiovascular diseases (Table 1). We also compared survival in those patients who had histological confirmation of hyperparathyroidism or hypercalcemia with an inappropriately elevated serum PTH level versus those with presumptive primary HPT on the basis of hypercalcemia for more than 1 year without another cause being found. Survival rates were similar in these two groups. When the group with a presumptive diagnosis was excluded from the analysis, the relative risk of death among those with HPT diagnosed by more definitive criteria was identical to that for the whole group (RR 5 0.69, 95% CI 5 0.56 – 0.85). In univariate analyses, older age and higher maximal

serum calcium levels were associated with worse survival, whereas men had better survival (Table 2). In the multivariate analysis, however, only age and maximal serum calcium level were associated with mortality (Table 2). Neither the presence of complications nor the time period of diagnosis (before or after 1974) were significant predictors of mortality. Since serum calcium level was an independent predictor of mortality, this was explored further by plotting survival based on quartiles of serum calcium (Figure 3). As evident in the figure, survival was quite similar among the patients in the 3 lower quartiles, but was significantly worse among the patients in the highest quartile (11.2 to 16.0 mg/dL, P ,0.001). Nevertheless, when compared with expected survival (adjusting for age and gender), survival even among the patients in the highest quartile was not significantly impaired either before (Figure 4A) or after (Figure 4B) surgery. However, there were nonsignificant trends towards decreased survival in the patients who did not undergo surgery, especially after 10 years following the initial serum calcium elevation (Figure 4A). Within this highest quartile, serum calcium levels were significantly skewed, with 75% of the patients having serum calcium levels between 11.2 and 11.8, and only 25% having serum calcium levels between 11.8 and 16.0. We also analyzed the survival data in this group to test whether there was a cutoff beyond which there was a clear increase in mortality, but could not identify such as value.

DISCUSSION The introduction of automated serum calcium measurements in the 1970s led to the identification of a large number of individuals with mild, asymptomatic primary hyperparathyroidism (11). This has resulted

Table 2. Predictors of All-Cause Mortality among Rochester, Minnesota Residents with Primary Hyperparathyroidism Variable Univariate Model Age (per 10 year increase) Male gender Highest calcium level (per mg/dL increase) Complications* Diagnosis year after June 1974 Multivariate Model Age Highest calcium level

Relative Hazard (95% Confidence Interval)

P Value

2.7 (2.2, 3.2) 0.5 (0.3, 0.9) 1.8 (1.5, 2.2) 1.0 (0.5, 1.7) 1.4 (0.8, 2.2)

,0.0001 0.02 ,0.0001 0.88 0.23

2.6 (2.2, 3.1) 1.3 (1.1, 1.6)

,0.0001 0.02

* Included nephrolithiasis, osteoporosis, hypercalcemic crisis, peptic ulcer disease, pseudogout, or band keratophy. 118

February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

Survival After Hyperparathyroidism/Wermers et al

Figure 3. Survival as a function of quartile of serum calcium levels (I 5 10.2 to 10.5 mg/dL; II 5 10.6 to 10.7 mg/dL; III 5 10.8 to 11.1 mg/dL; and IV 5 11.2 to 16.0 mg/dL) among Rochester, Minnesota residents with primary hyperparathyroidism. Group IV had significantly reduced survival (P ,0.001) compared with the other 3 groups.

in an ongoing controversy regarding the most appropriate management of these patients because, in general, the majority of patients with mild disease do not develop clinical complications. Scholz and Purnell’s prospective study on 147 asymptomatic patients with primary hyperparathyroidism showed that only 23% required parathyroidectomy for disease progression (21). Furthermore, during the 10-year period of follow-up, only 12% of patients had increased serum calcium levels. Many studies have confirmed this initial impression (22–29). The development of renal calculi has been noted in just 2% to 6% of patients with mild primary hyperparathyroidism, and decreased renal function has been reported in from 0 to 11% of patients. The changes in renal function are generally mild (22–27). Hypercalcemic crisis is very uncommon, occuring in less than 2% of patients (21–23). The effects of mild hyperparathyroidism on bone density and fractures are more controversial. While fractures can occur with severe disease (30), the overall risk of fracture was not increased following the diagnosis of hyperparathyroidism in Rochester (31). Other studies have shown both an increase (30,32) and a decrease in the risk of vertebral fractures (33). Likewise, Larsson and colleagues demonstrated an increase in forearm fractures (34) but found that primary hyperparathyroidism was not a risk factor for hip fractures in women (35). More recent data using bone densitometry indicate that patients with primary hyperparathyroidism may have lower cortical bone mass (eg, the distal radius), with preservation of cancellous bone mass [eg, the lumbar spine (36)]. This has also been confirmed by histomorphometric studies showing preservation of cancellous elements (37). However, a

subgroup of patients with vertebral osteopenia that improves following parathyroidectomy has also been recently described (38). Finally, there are conflicting data on whether bone density remains stable (29) or decreases (39) over time in these patients. Thus, while most patients with primary hyperparathyroidism have stable serum calcium levels, the issue of skeletal involvement and the potential for ongoing bone loss in these patients remains an area of controversy. In addition to potential effects on bone and renal function, several other concerns have been raised regarding the long term consequences of untreated hyperparathyroidism (40,41), including the potential impact on overall mortality (42,43). Palmer and colleagues (5) reported that survival was reduced by more than 10% in patients with mild to moderate hypercalcemia, primarily as a result of an increase in cardiovascular disease. In a separate study, they (9) noted a 5% reduction in survival in 441 patients who underwent parathyroidectomy between 1956 to 1979, again primarily due to cardiovascular disease. Several other reports based on patients who had

Figure 4. Survival among Rochester, Minnesota residents with definite primary hyperparathyroidism in the highest serum calcium level quartile (11.2 to 16.0 mg/dL) compared with that expected for Minnesota white residents of similar age and gender before parathyroid surgery (A) and following parathyroid surgery (B) (P 5 0.91 and 0.41, respectively for difference between observed versus expected survivals). February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 119

Survival After Hyperparathyroidism/Wermers et al

parathyroid surgery have also indicated an increased mortality primarily due to cardiovascular disease (6,7,10,44). Moreover, recent studies using echocardiography have indicated that patients with hyperparathyroidism may have an increased prevalance of left ventricular hypertrophy, calcific deposits in the myocardium, and aortic and mitral valve calcifications (45). Many of these changes may regress following parathyroidectomy (46). Several studies have also suggested that primary HPT may be associated with an increased risk of malignancy (2– 6,47). For example, Palmer et al found a relative risk of 1.6 for the development of a malignancy in surgical patients with primary HPT (47). More recently, Hedback and colleagues described an increased risk of mortality in patients after parathyroidectomy, due to malignancy and cardiovascular disease (6). Several different malignancies have been reported in these series, including adenocarcinomas of the gastrointestinal tract, renal carcinomas, breast cancers, endocrine tumors, and multiple myeloma (6,38,48). In contrast to these previous data, however, our study demonstrates that neither overall mortality nor mortality specifically due to cancer or cardiovascular disease is increased among unselected patients with primary HPT in a community that is generally representative of the US white population. There are several potential explanations for this discrepancy, the most likely one being in the type of patients included in the different studies. Whereas the majority of previous studies reviewed only surgical cases, our patient population consisted of all community residents with primary hyperparathyroidism, whether or not they underwent surgery. Moreover, our patients were detected through a rather intensive surveillance because, on average, about 22% of community residents had a serum calcium determination each year during the period 1984 to 1993. Thus, while most previous investigators evaluated a group of patients with more severe disease, who might be expected to have worse survival, most of our patients had uncomplicated, asymptomatic disease. We did find, however, that even though overall mortality was not increased in patients with primary HPT, increasing serum calcium levels were associated with worse survival in a multivariate analysis. Patients in the highest quartile of serum calcium levels ($11.2 mg/dL) had significantly worse survival than the patients in the three lower quartiles, although this group showed only a trend toward reduced survival when compared to age- and gender-matched controls. Based on these data, while parathyroid surgery may be indicated to prevent complications of HPT such as nephrolithiasis or osteopenia, a potential increase in mortality should no longer be an important concern in the conservative management of patients with mild, uncomplicated disease. 120

February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

As noted earlier, patients were accepted as having primary HPT in this study if they had histopathological confirmation of parathyroid adenoma or hyperplasia, hypercalcemia with an elevated serum PTH level, or hypercalcemia for more than 1 year without another cause being found. These were the original criteria used by Heath and co-workers (11) from our institution in defining the incidence of the disease following the introduction of automated serum calcium measurements, and we used the same criteria in our follow-up study on subsequent changes in the incidence of hyperparathyroidism in this community (12). However, when we excluded patients who did not have histopathological confirmation or elevated PTH levels from the analysis, identical results were obtained, effectively eliminating the possibility that potential misclassification of patients might account for the differences between our findings and those of some of the previous studies. In addition to assessing its impact on mortality, our study reaffirms many of the previously described features of primary hyperparathyroidism. It remains a disease that most commonly affects postmenopausal women and, in recent years, the majority of patients are without complications and have only modest elevations of serum calcium. The shift in clinical spectrum, which was precipitated by the introduction of automated serum calcium determinations here in July 1974 (11), has been reflected in patient management at our institution: whereas parathyroid surgery was commonly performed in the prescreening era, most patients currently are being observed. Thus, the routine measurement of serum calcium levels in the population has changed both the patient profile and management of primary hyperparathyroidism. Just recently, however, the Health Care Financing Administration has dictated that multichannel autoanalyzer tests no longer be performed for Medicare patients but that specific chemistry determinations be ordered individually. As a consequence, the proportion of tests that include a serum calcium determination is likely to fall. If this policy is extended to other payers, we may see a return to the former situation, where the only patients who are clinically recognized are those who present with symptoms or complications of primary hyperparathyroidism. In summary, in contrast to previous studies, our data indicate that primary hyperparathyroidism in unselected patients from the community does not shorten life, probably because most of our patients had mild disease that was found incidentally on routine serum calcium screening. As the frequency of testing for serum calcium declines, there may be a shift in the clinical spectrum of primary hyperparathyroidism back to that seen in the prescreening era (12). Although only randomized trials of observation versus surgery can define the risks to bone

Survival After Hyperparathyroidism/Wermers et al

and possibly renal function, our results suggest that the current practice of observing patients with uncomplicated mild primary hyperparathyroidism does not compromise survival.

ACKNOWLEDGMENT The authors wish to thank Mrs. Cindy Crowson for data analysis and Mrs. Mary Roberts for assistance in preparing the manuscript.

REFERENCES 1. Anonymous. NIH Conference. Diagnosis and management of asymptomatic primary hyperparathyroidism: Consensus development conference statement. Ann Intern Med. 1991;114:593– 597. 2. Farr HW, Fahey TJ Jr, Nash AG, Farr CM. Primary hyperparathyroidism and cancer. Am J Surg. 1973;126:539 –543. 3. Kaplan L, Katz AD, Ben-Isaac C, Massry SG. Malignant neoplasms and parathyroid adenoma. Cancer. 1971;28:401– 407. 4. Wajngot A, Werner S, Granberg P-O, Lindvall N. Occurrence of pituitary adenomas and other neoplastic diseases in primary hyperparathyroidism. Surg Gynecol Obstet. 1980;151:401– 403. 5. Palmer M, Adami HO, Bergstrom R, et al. Survival and renal function in untreated hypercalcaemia. Population-based cohort study with 14 years of follow-up. Lancet. 1987;1:59 – 62. 6. Hedback G, Tisell LE, Bengtsson BA, et al. Premature death in patients operated on for primary hyperparathyroidism. World J Surg. 1990;14:829 – 836. 7. Ronni-Sivula H. Causes of death in patients previously operated on for primary hyperparathyroidism. Ann Chir Gynaecol. 1985;74: 13–18. 8. Hedback G, Oden A, Tisell LE. The influence of surgery on the risk of death in patients with primary hyperparathyroidism. World J Surg. 1991;15:399 – 407. 9. Palmer M, Adami HO, Bergstrom R, et al. Mortality after surgery for primary hyperparathyroidism: A follow-up of 441 patients operated on from 1956 to 1979. Surgery. 1987;102:1–7. 10. Uden P, Tibblin S. Mortality in patients surgically treated for primary hyperparathyroidism due to solitary adenoma. Annales Chirurgiae et Gynaecologiae. 1990;79:123–128. 11. Heath H III, Hodgson SF, Kennedy MA. Primary hyperparathyroidism: Incidence, morbidity, and potential economic impact in a community. NEJM. 1980;302:189 –193. 12. Wermers RA, Khosla S, Atkinson EJ, et al. The rise and fall of primary hyperparathyroidism in Rochester, Minnesota. A population-based study, 1965–1992. Ann Intern Med. 1997;126:433– 440. 13. Kurland LT, Molgaard CA. The patient record in epidemiology. Sci Am. 1981;245:54 – 63. 14. Melton LJ III. History of the Rochester Epidemiology Project. Mayo Clin Proc. 1996;71:266 –274. 15. Kao PC, van Heerden JA, Grant CS, et al. Clinical performance of parathyroid hormone immunometric assays. Mayo Clin Proc. 1992; 67:637– 645. 16. Arnaud CD, Tsao HS, Littledike T. Radioimmunoassay of human parathyroid hormone in serum. J Clin Invest. 1971;50:21–34. 17. Kao PC, Jiang N-S, Klee GG, Purnell DC. Development and validation of a new radioimmunoassay for parathyrin (PTH). Clin Chem. 1982;28:69 –74. 18. Law WM Jr, Heath H III. Familial benign hypercalcemia (hypocalciuric hypercalcemia). Clinical and pathogenetic studies in 21 families. Ann Intern Med. 1985;102:511–9.

19. Heath H III. Familial benign (hypocalciuric) hypercalcemia. A troublesome mimic of mild primary hyperparathyroidism. Endocrinol Metab Clin North Am. 1989;18:723–740. 20. Kalbfleisch JD, Prentice RL. The Statistical Analysis for Failure Time Data. New York: J Wiley & Sons; 1980:1–321. 21. Scholz DA, Purnell DC. Asymptomatic primary hyperparathyroidism. 10-year prospective study. Mayo Clinic Proceedings. 1981;56: 473– 478. 22. Corlew DS, Bryda SL, Bradley EL III, DiGirolamo M. Observations on the course of untreated primary hyperparathyroidism. Surgery. 1985;98:1064 –1071. 23. Palmer M, Jakobsson S, Akerstrom G, Ljunghall S. Prevalence of hypercalcaemia in a health survey: a 14-year follow-up study of serum calcium values. Eur J Clin Invest. 1988;18:39 – 46. 24. Rohl PG, Wilkinson M, Clifton-Bligh P, Poser S. Hyperparathyroidism: experiences with treated and untreated patients. Med J Aust. 1981;1:519 –521. 25. Adams PH. Conservative management of primary hyperparathyroidism. J Royal Coll Phys Lond. 1982;16:184 –190. 26. Posen S, Clifton-Bligh P, Reeve TS, et al. Is parathyroidectomy of benefit in primary hyperparathyroidism? Q J Med. 1985;54:241– 251. 27. Van’t Hoff W, Ballardie FW, Bicknell EJ. Primary hyperparathyroidism: the case for medical management. BMJ. 1983;287: 1605–1608. 28. Rao DS, Wilson RJ, Kleerekoper M, Parfitt AM. Lack of biochemical progression or continuation of accelerated bone loss in mild asymptomatic primary hyperparathyroidism: evidence for biphasic disease course. J Clin Endocrinol Metab. 1988;67:1294 – 1298. 29. Silverberg SJ, Gartenberg F, Jacobs TP, et al. Longitudinal measurements of bone density and biochemical indices in untreated primary hyperparathyroidism. J Clin Endocrinol Metab. 1995;80:723– 728. 30. Dauphine RT, Riggs BL, Scholz DA. Back pain and vertebral crush fractures: An unemphasized mode of presentation for primary hyperparathyroidism. Ann Intern Med. 1975;83:365–367. 31. Melton LJ III, Atkinson EJ, O’Fallon WM, Heath H III. Risk of age-related fractures in patients with primary hyperparathyroidism. Arch Intern Med. 1992;152:2269 –2273. 32. Kenny AM, MacGillivray DC, Pilbeam CC, et al. Fracture incidence in postmenopausal women with primary hyperparathyroidism. Surgery. 1995;118:109 –114. 33. Wilson RJ, Rao S, Ellis B, et al. Mild asymptomatic primary hyperparathyroidism is not a risk factor for vertebral fractures. Ann Intern Med. 1988;109:959 –962. 34. Larsson K, Lindh E, Lind L, et al. Increased fracture risk in hypercalcemia: bone mineral content measured in hyperparathyroidism. Acta Orthop Scand. 1989;60:268 –270. 35. Larsson K, Ljunghall S, Krusemo UB, et al. The risk of hip fractures in patients with primary hyperparathyroidism: a population-based cohort study with a follow-up of 19 years. J Intern Med. 1993;234: 585–593. 36. Silverberg SJ, Shane E, de la Cruz L, et al. Skeletal disease in primary hyperparathyroidism. J Bone Miner Res. 1989;4:283–291. 37. Parisien M, Silverberg SJ, Shane E, et al. The histomorphometry of bone in primary hyperparathyroidism: preservation of cancellous bone. J Clin Endocrinol Metab. 1990;70:930 –938. 38. Silverberg SJ, Locker FG, Bilezikian JP. Vertebral osteopenia: A new indication for surgery in primary hyperparathyroidism. J Clin Endocrinol Metab. 1996;81:4007– 4012. 39. Guo CY, Thomas WE, al-Dehaimi AW, et al. Longitudinal changes in bone mineral density and bone turnover in postmenopausal women with primary hyperparathyroidism. J Clin Endocrinol Metab. 1996;81:3487–3491. February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 121

Survival After Hyperparathyroidism/Wermers et al 40. Solomon BL, Schaaf M, Smallridge RC. Psychologic symptoms before and after parathyroid surgery. Am J Med. 1994;96:101– 106. 41. Kleerekoper M, Bilezikian JP. A cure in search of a disease: parathyroidectomy for nontraditional features of primary hyperparathyroidism. Am J Med. 1994;96:99 –100. (Editorial.) 42. Rastad J, Akerstrom G, Ljunghall S. Mortality of untreated primary hyperparathyroidism—a nontraditional indication for parathyroid surgery? Am J Med. 1995;99:577–578. (Letter.) 43. Kleerekoper M, Bilezikian JP. Mortality of untreated primary hyperparathyroidism–a nontraditional indication for parathyroid surgery? Am J Med. 1995;99:578. (Reply). 44. Romanus R, Heimann P, Nilsson O, Hansson G. Surgical treatment of hyperparathyroidism. Progr Surg. 1973;12:22–76.

122

February 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

45. Stefenelli T, Mayr H, Berger-Klein J, et al. Primary hyperparathyroidism: incidence of cardiac abnormalities and partial reversibility after successful parathyroidectomy. Am J Med. 1993;95: 197–202. 46. Stefenelli T, Abela C, Frank H, et al. Cardiac abnormalities in patients with primary hyperparathyroidism: implications for followup. J Clin Endocrinol Metab. 1997;82:106 –112. 47. Palmer M, Adami HO, Krusemo UB, Ljunghall S. Increased risk of malignant diseases after surgery for primary hyperparathyroidism: A nationwide cohort study. Am J Epidemiol. 1988;127:1031–1040. 48. Hoelzer DR, Silverberg AB. Primary hyperparathyroidism complicated by multiple myeloma. Arch Intern Med. 1984;144:2069 – 2071.