JOURNAL OF BONE AND MINERAL RESEARCH Volume 6, Supplement 2, 1991 Mary Ann Liebert, Inc., Publishers

Asymptomatic Primary Hyperparathyroidism BRUCE H. MITLAK, MARGARET DALY, JOHN T. POTTS, JR., DAVID SCHOENFELD, and ROBERT M. NEER

ABSTRACT Of 118 consecutive white patients referred for asymptomatic primary hyperparathyroidism, the diagnosis was clinically confirmed in 100, of whom 85 adults had a serum calcium less than 3.0 mM (12 mg/dl) and no skeletal, rheumatic, or significant neuropsychiatric symptoms, azotemia, or other significant illnesses. Among these 85, 68 had both asymptomatic and medically uncomplicated hyperparathyroidism, whereas 17 had historical, radiographic, or ultrasonic evidence of renal stone disease. The 20% with past or present renal calculi concentrated their urine significantly better than the 68 others (p = 0.05), but these two groups were otherwise not distinguished by the tests we performed, so all 85 patients were analyzed together. Systolic and diastolic blood pressures were normal, but premature osteopenia and/or impaired renal function were present in 29-36'70 of the patients. Micrometer measurements of metacarpal radiographs and 12' photon absorptiometry at the shaft of the radius revealed cortical osteopenia. Osteopenia was equally significant in the distal radius (cortical plus trabecular bone). These quantitative measurements were superior to routine bone radiography, and ROC analysis showed that '*'I absorptiometry at either site was superior (p < 0.01) to metacarpal cortex measurements for detecting premature osteopenia, which was present in more than a third of these patients. Creatinine clearances (24 h) and maximum urine concentrating capacity (overnight dehydration plus the synthetic vasopressin analog DDAVP) were each significantly reduced, despite all patients' normal BUN and serum creatinine levels. Sequential performance of a 24 h creatinine clearance and a urine concentration test revealed abnormalities in the renal function of 27 of 74 patients (36Vo), with a specificity of 95% and a higher sensitivity than either test alone (27-29%). Testing the kidneys' ability to acidify the urine to pH 5.3 or less identified only 1 additional abnormal patient. Silent complications of primary hyperparathyroidism are thus common in "asymptomatic" patients with mild hypercalcemia. Their presence was only weakly correlated with urinary calcium excretion (measured on a low-calcium diet) and serum alkaline phosphatase. From these data, it is not possible to predict whether these complications progress in the absence of treatment or diminish with successful treatment of the primary hyperparathyroidism.

INTRODUCTION HE ENHANCED ABILITY to detect hyperparathyroidism in an asymptomatic individual was brought about, several decades ago, by the widespread introduction of automated calcium measurements. Subsequently, improvements in assays for parathyroid hormone (PTH) have enabled the rapid, reliable confirmation of hyperparathyroidism in these individuals without surgery. With improvement in diagnostic skill came the yet to be resolved dilemma sur-

T

rounding the management of these patients: since they are asymptomatic, when if ever is operation appropriate? Most physicians consider parathyroid surgery appropriate if the hyperparathyroidism produces medically significant secondary complications, such as impaired renal function or premature osteopenia, even if these d o not yet cause symptoms. It is therefore important to detect asymptomatic skeletal and renal complications of the disease efficiently. We have analyzed data obtained during the initial evaluation of 68 patients with mild asymptomatic primary

Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts.

S103

MITLAK ET AL.

s104

118

hyperparathyroidism and 17 patients with mild primary hyperparathyroidism with nephrolithiasis and compared the utility of techniques utilized to detect the silent complications of hyperparathyroidism.

I

Inclusion criteria

I

/

Id0

No stones

I

78

I I

\

Stones

I

22

I

Exclusion criteria 68

I

17

MATERIALS AND METHODS A total of 118 consecutive white patients were referred

for evaluation of asymptomatic primary hyperparathyroidism beginning in January 1972 and concluding in January 1983. The diagnostic criteria included serum Ca > 2.6 mmol/liter on at least two occasions plus an elevated or inappropriately normal serum immunoreactive PTH concentration plus exclusion of other causes for hypercalcemia. After an initial screening consisting of history and physical examination, automated chemistry panel, and renal ultrasound or radiograph of the abdomen, 18 patients did not meet the inclusion criteria. Of these, 9 were suspected or presumed to have familial hypocalciuric hypercalcemia and 9 others had no evidence of elevated serum calcium or PTH concentrations. Of the 100 patients who had confirmed hyperparathyroidism, 22 had radiologic or ultrasonographic evidence of nephrolithiasis when evaluated by us or previously or had a prior episode of flank pain and hematuria consistent with the passage of a renal stone. These patients were analyzed separately from those without prior evidence of nephrolithiasis. We excluded from further analysis 10 of the 78 nonstone patients: 3, less than 60 years of age, had fractures unassociated with major trauma, 2 had bone pain due to radiologically obvious osteitis fibrosa cystica, and 1 had an elevated serum creatinine concentration (> 133 pmol/liter) without other explanation. These abnormalities were attributed to their primary hyperparathyroidism. Another 2 nonstone patients were excluded because their hyperparathyroidism was arbitrarily judged to be severe (Ca > 3.0 mmol/liter), and 2 adolescents (ages 14 and 17) were excluded because of their atypical age. Of the 22 stone patients, 5 were also excluded from further analysis: 2 had bone pain with osteitis fibrosa cystica, 1 had an elevated serum creatinine concentration, 1 had intractable, severe diarthea, and 1 had hypopituitarism. The study group therefore comprised 68 asymptomatic patients with mild primary hyperparathyroidism who were without obvious significant complications on routine“’ medical evaluation, plus 17 who met the same criteria but had past or present evidence of renal stone disease. None of the patients had neuropsychiatric symptoms as judged by the patients’ subjective assessment of their personality, work, performance, life-style, or ability to concentrate. None had bone pain or arthritis.

PTH assays Initially, serum PTH was measured with a radioimmunoassay employing antiserum GP-1, which is directed toward antigenic determinants in the amino-terminal region of human PTH and additional determinants in the carboxyl-terminal region. It thus recognizes intact PTH and circulating PTH fragments.”) Subsequently, an amino-terminal radioimmunoassay employing antiserum CK-67 was utilized. This recognizes antigenic determinants in the regions of human PTH-(18-25) [hPTH-(l8-25)] and is thus insensitive to circulating carboxyl-terminal PTH fragment^.'^' Calcium, phosphorus, albumin, alkaline phosphatase, magnesium, electrolytes, and creatinine were analyzed in the hospital’s general chemistry laboratory. Urine for calcium was measured by atomic absorption spectrophotometry.(4-6)To accommodate changes in the PTH assay and in reporting of serum alkaline phosphatase over the period of the study, we express these values as a multiple of the appropriate maximal normal values. Renal function was evaluated by determining 24 h creatinine clearance, maximal urinary concentrating capacity, and urinary acidifying capacity during admission to the hospital’s clinical research center.

Maximal urinary concentrating capacity Patients were required to swallow nothing from 8 p.m. until the test was completed the following morning. Desmopressin acetate (DDAVP) (20 pg) was instilled intranasally at 10 p.m. and at 6 a.m., or pitressin tannate in oil ( 5 units) was injected intramuscularly at 10 p.m. The osmolarity of the second and third morning urine was measured in the hospital’s chemistry laboratory by freezing-point depression. Simultaneously, serum osmolarity and sodium were determined to assure that the patient was not water intoxicated as a result of surreptitious drinking.

Urinary acidifying capacity Urine pH was checked with each void until a reading of pH 5.3 or less was recorded. If this did not occur, oral ammonium chloride (100 mg/kg) was given and urine pH was measured by pH meter over the next 3 h or until a pH of 5.3 or less was recorded.(” Simultaneously, serum bicarbonate was measured to document an adequate decline.

s 10s

ASYMPTOMATIC PRIMARY HYPERPARATHYROIDISM

Skeletal status Bone mineral densitometry of the forearm was performed at the junction of the proximal two-thirds and distal one-third of the nondominant radius and at a distal site 2.0 cm proximal to the tip of the ulnar styloid process utilizing an l Z 5 l single-photon absorptiometer (Norland Instruments, Fort Atkinson, WI). Radiographs of the skull, hands, and spine were performed and analyzed routinely in the radiology department of the hospital. Subcortical resorption or cortical tunneling prompted the decision to report an abnormality. Osteopenia or vertebral compression fracture in a patient over age 65 was not taken as evidence for hyperparathyroid-related bone loss. The outer ( D ) and inner ( d ) cortex were measured by two independent observers at the midpoint of the second metacarpal on a magnification hand radiograph using a needle-point micrometer (Helios Instruments, Hamburg, Germany). The fractional cortical area was expressed as (D’ - d 2 ) / D z . We considered a priori that hyperpatathyroid bone loss may be related to the severity of the hyperparathyroidism (elevation in serum calcium or PTH concentration) or the severity of phosphate depletion (fasting serum inorganic phosphate concentration) and might be predicted by the level of urinary calcium excretion or bone remodeling rate (urinary hydroxyproline and serum alkaline phosphatase concentration) on a restricted calcium intake. Because these variables are sensitive to diet and environment, they were measured after hospitalization on the clinical research center and 3 days equilibration to a constant low-gelatin diet containing 200 mg Ca, 600 mg P, and 100 mEq Na per day to reduce the variability arising from differences in food absorption and life-style.

Statistical analysis The results are reported as the mean f SD (standard deviation). To distinguish disease-specific changes from

those attributable to age alone, Z-score transformations were performed using age- and sex-specific mean values and standard deviations for forearm bone mineral density at the radial shaft and metaphysis [ n = 356, 811(891],fractional cortical area of the second metacarpal [ n = 2500(101],renal concentrating capacity [n = 97(1t1], creatinine clearance [ n = 648(l21 3 ) ] , and blood pressure [ n > 10,000(141]. The normal age- and sex-specific mean wa\ subtracted from each patient’s measurement, and the resulting value was divided by the approptiate standard deviation to yield the Z score. An individual with a test result equal to the appropriate mean value would therefore have a Z score of 0. The normality of the Z score distribution\ was determined by the Kolomogorov statistic D. Groups were cornpared by t-test; one-sided tests were used to test patients against the normal population, and two-sided tests were used to compare stone and nonstone patients. Variables were correlated using Pearson product-moment correlations (SAS Institute, Carey, NG).

RESULTS Clinical and biochemical data The 68 patients without renal or apparent skeletal complications on routine medical evaluation and the 17 patients with stone disease are described in Table l . These groups were indistinguishable by all criteria measured, although the nonstone patients could concentrate their urine slightly better (p = 0.051 compared by age- and sexmatched 2 scores). Power calculations indicated that we could have detected differences as small as 1.0 SD unit between the two group means with a probabillty of 58-98% (Table 1). We therefore subsequently analyzed the data from the two groups together.

TABLE 1. CHARACTERISTICS (MEANf SD) OF T W O GROUPSOF PATIENTS

Nonstone (N = 68)

WITH M t t D PRtMARY #YPERPARATHYROtDlSM*

Storle (N = 17)

P

Power (%)

~~~~

Calcium, mg/dl Phosphorus, mg/dl Alkaline phosphatase, 070 of NL,, 24 h urine calcium PTH, % of NL,, SPA distal radius Z score S P A radial shaft 2 score MCA Z score Creatinine clearance 2 score Maximum urinary concentrating capacity 2 score Age Sex, F/M

10.74 2.81 0.83 206 1.36 -1.27 -1.35 -0.92

-1.19 - 1.47

2t

0.53 0.43

f 0.39 i 113

0.76 k 1.02 k

&

1.23

i 1.5 1.3

+ + 0.87

54 k 13 5/1

10.74 f 0.66 2.76 f 0.49 0.88 i 0.27 191 + 87 1.22 -+ 1.2 -1.28 i 1.47 -1.40 + 1.16 -0.87 & 1,O

*

-1.0 1.9 -1.95 h 0.99 53 f 14 4/ I

0.97 0.69 0.56

0.66 0.65 0.98 0.87

0.90 0075 0.051

58 82 58 80

98

aThe two groups were 68 patients without evidence of nephrolithiasls and 17 patients with evidence of nephrolithlasls or prior passage of renal calculi. The results of a two-sided I-test comparing the groups are shown for each parameter. The power calculations demonstrate the probability to detect a difference between the two groups as small as 1 .O SD unit with ad ci error of 0.05.

s 106

MITLAK ET AL.

Blood pressure Systolic and diastolic blood pressure Z scores were normally distributed and had mean values of -0.23 & 1 . 1 and -0.23 & 1.0, respectively (Fig. I ) . The diastolic blood pressure was slightly lower than that in the normal reference population (N = 85, p = 0.046). Power calculations (a = 0.05, one-sided test) indicate that we could have detected an increase of l l mmHg in mean systolic pressure with a likelihood of 99% and increases of 7 mmHg in diastolic pressure with a likelihood of 99% compared to ageand sex-matched values.‘L4’

Skeletal status The second metacarpal cortical area Z scores were normally distributed, and the mean of -0.91 + 1.4 was significantly lower than that for the reference population (N = 75, p < O.OOO1; Fig. 2). The mean Z scores for bone mineral density at the metaphysis and shaft of the radius were - 1 . 3 & 1 . 1 and -1.4 & 1.2, respectively; each was significantly lower than values in the normal reference population ( N = 71,p < O.OOO1 and N = 82, p < O.OOO1, respectively). Bone mineral density at the radial shaft was negatively and significantly (p < 0.01) correlated with urinary calcium and alkaline phosphatase measured after equilibration to a calcium-restricted diet, but the correlations were weak (each r = -0.4). There was no correlation with the serum PTH concentration. To analyze further the information provided by the tests of skeletal status, decision points of three diagnostic tests for hyperparathyroid-associated bone loss were arranged in order of increasing sophistication and expense, which might represent a physician’s approach for evaluating a newly diagnosed asymptomatic patient with primary hyperparathyroidism (Fig. 3, left). A Z score < -2.0 (2.5th percentile for the normal reference population) was taken as the threshold for abnormality at each of the decision points when Z scores were available. Of the 56 patients who had all three tests performed, 22 (39%) were abnormal by at least one of the three tests evaluating skeletal status. Radiographs of the skull and hands detected asymptomatic subperiosteal resorption and/or “salt and pepper” demineralization of the skull in 6 of 56 ( 1 1 Yo) of the patients, but when measurement of the second metacarpal cortical area was also performed the number of patients with abnormalities increased to 12 of 56 (22%). Bone densitometry at the radial shaft detected an additional 10 patients who had significant osteopenia despite normal routine radiographs and a normal second metacarpal cortical area. Bone mineral density at the radial shaft as a single test detected significant osteopenia in 16 of 56 (29%) of patients. However, if this test of bone mineral density had been the only test performed, abnormalities would have been missed in 7 patients: 4 whose metacarpal cortical area Z score was < -2.0 and 3 who had subperiosteal resorption and/or salt and pepper demineralization of the skull. Because physicians use different thresholds to decide when a measurement is abnormal or warrants intervention,

Z-SCORE

4

-3

-2

.l

0

1

2

3

2-SCORE

FIG. 1. Age- and sex-matched score distributions of systolic (top) and diastolic (bottom) blood pressure. The shaded area represents a Gaussian distribution of Z scores in a normal population of equal size.

we also compared the metacarpal cortical measurements and the two bone density tests using ROC curves. Such curves display the sensitivity and specificity of the tests using all possible thresholds to discriminate an individual from a reference population, ranging from thresholds that exclude all normal individuals (100% specificity) to thresholds that include all hyperparathyroid individuals (100% sensitivity). By expressing the data as Z scores, we constructed receiver operating characteristic (ROC) curves that also take into account the age- and sex-specific nature of the normal values (Fig. 4).The accuracy of the tests was compared by determining the area under each ROC curve. The relative areas were 0.77, 0.77, and 0.67 for single-photon absorptiometry (SPA) of the radial shaft, SPA of the radial metaphysis, and measurement of the metacarpal cortex, respectively. By this analysis there was no difference in the accuracy of densitometry at the radial shaft and the distal radius to detect osteopenia; however, the measurement of metacarpal cortical area was significantly less accurate than the other two tests (p < 0.01).

Renal function Creatinine clearance and maximal urinary concentrating capacity Z scores were -1.2 1.4 and -1.6 & 0.9, re-

*

ASYMPTOMATIC PRIMARY HYPERPARATHYROIDISM

A

"1 12

10

G

5 ' 1) $

E

6

4

2

0

o

t

2

3

4

2-SCORE

S107

renal function by one of these three tests. Measurement of creatinine clearance detected abnormalities in 18 of 74 (24%) of the patients. Of the 56 who were found to have a normal creatinine clearance, 9 (12%) had a defect in maximal urinary concentrating capacity ( Z scores < -2.0). The determination of urinary acidification capacity added little to the two prior tests, identifying only 1 additional abnormal patient. Measurement of both creatinine clearance and concentrating capacity provided complementary information. The combined sensitivity of the two tests performed in sequence was 3670, which is higher than the sensitivity of either test alone (27 and 29%), and the specificity of the two tests combined was 95% or hetter, nearly as good as for each test separately.* There was a weak negative correlation between the concentrating capacity Z scorc and the urinary calcium excretion on a low-calcium diet ( r = -0.5, p = 0.0007). There was no correlation between the P T H concentration and either test of renal function.

DISCUSSION

4

-3

4

-1

0

1

2

J

4

1

2

J

1

2-SCORE

4

-5

-2

-1

0

2-SCORE

FIG. 2. Age- and sex-matched Z score distributions of (A) bone mineral densitometry of the radial metaphysis, (B) of the radial shaft at the one third distal site, and (C) the metacarpal cortical area determined from radiographs. The shaded areas represent a Gaussian distribution of Z scores in a normal population of equal size.

spectively; each was significantly lower than values in the reference population (N = 80, p < 0.0001 and N = 82, p < O.OOO1, respectively; Fig. 5). Measurement of creatinine clearance and maximal urinary concentrating capacity detected a Z score < -2.0 in 22 of 82 patients (27%) and 23 of 80 (29%) patients, respectively. Four patients were unable to acidify their urine to pH 5.3 or less. Figure 3 (right) compares the results of three renal function tests in 74 patients who had each test performed. Of the 74 patients who had a normal serum creatinine, 46 (38%) had abnormal

Previous studies examining patients with mild primary hyperparathyroidism have differed in inclusion criteria, exclusion criteria, and patient referral base. Patients have been identified by screening populations or patients at primary medical care centers'l) or studying patients referred for medical or surgical speciality care of suspected parathyroid disease.(1s-19)Some studies include both primary and speciality care referral patients.(*0)The serum calcium concentration at which patients were included also differed among studies. Scholz and Purnell") included patients within the range of 2.52-2.75 mmol/liter; we and others included patients with serum calcium concentrations of up to 3.0 mmol/liter.'15) In some studies there was no clearly stated level of hypercalcemia at which patients were included or exc1uded.('6-'8) Renal stone disease was not considered an exclusion criterion in some s t u d i e ~ " ~and - ~ ~was ~ an exclusion criterion in others only if symptomatic during the or within the year before the study.'ll We initially analyzed patients with nephrolithiasis separately, because it has been hypothesized that they represent a distinct subgroup of patients.lz1) This has not been confirmed by subsequent i n ~ e s t i g a t o r s , ( ~however, ' ~ ~ ~ ) and we were unable to distinguish these patients from the nonstone patients by any criterion. Hypertension was not an exclusion criterion for our study, and our patients' blood pressure Z scores followed a Gaussian distribution, suggesting that subgroups of patients with varying degrees of hyperparathyroid-associated hypertension were not present and that involuntary referral bias had not skewed or truncated the distribution of blood pressures. The mean systolic and diastolic blood

*Performing multiple tests likely increases the number o f false positive results. Because we chose a specificity of 97.5% for each test separately (-2.0 SD as the threshold for coding an abnormality), the maximum false positive rate of the two tests combined is 2.5 + 2.5 = 5 % (specificity 95% or better).

MITLAK ET AL.

S108 Renal I-unition

Bone S t a t u s

I

(50 niarirnal urinary

concenlrallng capacity 2~scorc

X

c o r t i c a l area 2nd m e l a c a r p a l

-9

II7Xi

&b"l

"l"

__

2-score

P

U r i n e pH mlnlrnum

I

bone rnlneral density o f !he radial shalt2-score

I

I IIX)

lo'181' ab"l

"I

I

54

46

FIG. 3. (Left) Decision tree for tests of skeletal status in the 56 patients who had all three tests performed. Routine radiographs were coded abnormal if subperiosteal resorption or cortical osteopenia was detected on hand films or if salt and pepper demineralization of the skull was seen. Z scores of metacarpal cortical area and bone mineral density were coded abnormal if less than -2.0. (Right) Decision tree for tests of renal function in the 74 patients who had all three tests performed. Z scores of creatinine clearance and maximal urinary concentrating capacity were coded abnormal if less than -2.0, as was the inability to acidify the urine to p H 5.3 or less. The denominator of the percentages is 74.

IOiI

I1

,

4 11ll1

, / , , , . , . , . , ,

xi1

hlI

,

,

T

m

_

,

,

.

.

411

.

,

.

.

,

,

,,,,,, ,,,,, !I1

I1

srmwrry FIG. 4. ROC curves comparing sensitivity and specificity of bone mineral density at the radial shaft (solid line), metaphysis (dotted line), and metacarpal cortical area determined from radiographs (bottom line) at various detection thresholds, based on data from the 56 patients who had all three tests.

pressure Z scores were normal. This finding is in contrast with several reports showing a high prevalence of hypertension in patients referred for tertiary care of primary hyperparathyr~idism''*-'~.~~) but is in agreement with the population-based in which hypercalcemia was milder. Calculating blood pressure Z scores avoids using arbitrary thresholds for diagnosing hypertension, and power calculations indicated we should have been able to detect hypertension if present. Renal tubular function, as measured by urinary concentrating capacity, and glomerular function, as measured by creatinine clearance, were each abnormal in a significant number of our patients. The tests provided complementary information; a significant number of abnormalities was detected by one test and not by the other. This suggests that tubular and glomerular dysfunction may appear independently in these patients. The extent to which these abnormalities are reversible upon correction of the hypercalcemia cannot be established from the present observations. It is of note in this regard that abnormalities in maximal urinary concentrating capacity were somewhat worse the

ASYMPTOMATIC PRIMARY HYPERPARATHYROIDISM

.:::1

A ”

18

g

l:I 10

P a e

.h -4

-a

.2

.1

0

1

3

T I

3

1

2-SCORE

s109

these silent abnormalities predispose patients to later symptomatic illness. It is now clear that postmenopausal osteopenia in the forearm increases women’s risk of future fractures. ( 3 2 - 3 5 1 The relationship between osteopenia and future fracture has now also been demonstrated in This is presumably also true in patients with osteopenia caused by mild primary hyperparathyroidism, but the presumption must be verified by data. Because bone undergoing rapid remodeling as a result of mild primary hyperparathyroidism may have more effective repair of microfractures, the mechanical strength may be different from that in normal bone of similarly reduced density. Asymptomatic abnormalities in skeletal mass and/or renal function are considered, by some physicians, indications for surgical or other treatment of the causative hyperparathyroidism. This is reasonable only if treatment reverses the abnormalities or stops their progression. At present it is unclear how often these silent abnormalities progress and how often they are reversed by cure of the hyperparathyroidism. These uncertainties and the documented difficulty of maintaining careful long-term followup of patients with mild asymptomatic hyperparathyroidism underscore the need for clinically useful prognostic information.

ACKNOWLEDGMENT 4

4

-2

-1

0

1

2

2-SCORE

FIG. 5. Age- and sex-matched Z scores of (A) maximal urinary concentrating capacity and (B) creatinine clearance. The shaded areas represent a Gaussian distribution of Z scores in a normal population of equal size.

This study was supported by National Institutes of Health Grant No. RR 1066.

REFERENCES 1. Scholz DA, Purnell DC 1981 Asymptomatic primary hyper-

higher the urinary calcium excretion on a low-calcium diet. This correlation, although statistically significant, was too weak to be valuable in medical decision-making. Because the measurement of urinary acidification capacity added little to the ability to detect an abnormality in renal function, we no longer perform this test to evaluate our patients. Bone mineral densitometry at the radial shaft (cortical bone) and metaphysis (cortical and trabecular bone) were equally sensitive in detecting osteopenia. This contrasts with observations by some investigators, who reported measurements at a distal site, including trabecular bone, were more ~ e n s i t i v e . ( ~ ~Other ~ ’ ~ - ~investigators ~’ have reported that cortical bone is lost preferentially in primary hyperparathyroidism, with relative sparing of trabecular bone. ( 1 9 - 3 1 1 The reasons for these discrepancies are unclear. Our data emphasize that mild primary hyperparathyroidism, although asymptomatic and superficially uncomplicated, often causes premature osteopenia and/or abnormal renal function. In our population of 100 patients nearly one-third were found to have silent complications attributable to their mild primary hyperparathyroidism. Prospective studies are necessary to determine whether

2.

3.

4.

5.

6.

7. 8.

9.

parathyroidism 10-year prospective study. Mayo Clin Proc 56t473-478. Segre GV, Habener JF, Powell D, Tregear GW, Potts J T Jr 1972 Parathyroid hormone in human plasma: Immunochemical characterization and biological implications. J Clin Invest 51:3163-3172. Potts JT Jr, Segre GV, Enders DB 1983 Current concepts: Assessment of parathyroid function with an N-terminal specific radioimmunoassay for intact parathyroid hormone. Nichols Institute Reference Laboratories. Case Records of the Massachusetts General Hospital 1974 Normal reference laboratory values. N Engl J Med 290: 39-49. Case Records of the Massachusetts General Hospital 1978 Normal reference laboratory values. N Engl J Med 298: 34-45. Case Records of the Massachusetts General Hospital 1980 Normal reference laboratory values. N Engl J Med 302: 37-48. Wrong 0, Davies H E F 1959 The excretion of acid in renal disease. Q J Med 28:259-313. Mazess RB, Cameron JR 1973 Bone mineral content in normal US whites. In: Mazess RB (ed.) International Conference on Bone Mineral Measurement. DHEW Publication No. (NIH) 75-683. Johnston JO, Goldsmith NF 1973 Osteoporosis prevalence study. In: Mazess RB (ed.) International Conlerence o n Bone

MITLAK ET AL.

SllO

10.

II.

12.

13.

14.

15.

16.

17.

18.

19. 20.

21.

22.

23.

24.

Mineral Measurement. DHEW Publication No. (NIH) 75-683. Garn SM, Poznanski AK, Larson K 1973 Metacarpal lengths cortical diameters and area from the 10 state nutrition survey. In: Jaworski ZFG (ed.) Proceedings of the First Workshop on Bone Morphometry. University of Ottawa Press. Rowe JW, Shock NW, DeFronzo RA 1976 The influence of age on the renal response to water deprivation in man. Nephron 17270-278. Rowe JW, Andres R, Tobin JD, Norris AH, Shock NW 1976 The effect of age on creatinine clearance in men: A cross sectional and longitudinal study. J Gerontol 31:155-163. Rowe JW, Andres R, Tobin JD, Norris A H , Shock NW 1976 Age-adjusted standards for creatinine clearance (letter). Ann Intern Med 84567-569. Roberts J , Maurer K 1977 Blood pressure levels of persons 6-74 years, United States, 1971-4. Vital and health statistics. Series I I , Data from the National Health Survey. DHEW Publication No. (HRA) 78-1648. Rao DS, Wilson RJ, Kleerekoper M, Parfitt AM 1988 Lack of biochemical progression or continuation of accelerated bone loss in mild asymptomatic primary hyperparathyroidism: Evidence for biphasic disease course. J Clin Endocrinol Metab 67: 1294- 1298. Posen S, Clifton-Bligh P, Reeve TS, Wagstaffe C , Wilkinson M 1985 Is parathyroidectomy of benefit in primary hyperparathyroidism? Q J Med 54241 -251. Rohl PG, Wilkinson M, Clifton-Bligh P , Posen S 1981 Hyperparathyroidism: Experiences with treated and untreated patients. Med J Aust 1:519-521. Van’t Hoff W, Ballardie FW, Bicknell EJ 1983 Primary hyperparathyroidism: The case for medical management. Br J Med 287:1605-1608. Adams P H 1982 Conservative management of primary hyperparathyroidism. J R Coll Physicians Lond 16:184-190. Palmer M, Ljunghall S, Akerstrom G , Adami HO, Bergstrom R, Grimelius L, Rudberg C , Johansson H 1987 Patients with hyperparathyroidism operated on over a 24-year period: Temporal trends of clinical and laboratory findings. J Chron Dis 40:121-130. Broadus AE, Horst RL, Lang R, Littledike ET, Rassmusen H 1980 The importance of circulating 1,25-dihydroxyvitamin D in the pathogenesis of hypercalciuria and renal-stone formation in primary hyperparathyroidism. N Engl J Med 302:421-426. Thakker RV, Fraher LJ, Adami S, Karmali R, O’Riordan J L 1986 Circulating concentrations of 1.25-dihydroxyvitamin D, in patients with primary hyperparathyroidism. Bone Miner 1: 137- 144. Pak CY, Nicar MJ, Peterson R, Zerwehk JE, Snyder W 1981 A lack of unique pathophysiologic background for nephrolithiasis of primary hyperparathyroidism. J Clin Endocrinol Metab 53536-542. Diamond TW, Botha JR, Wing J , Meyers AM, Kalk WJ 1986 Parathyroid hypertension, a reversible disorder. Arch Intern Med 146:1709- I7 12.

25. Dalen N, Hjern B 1974 Bone mineral content in patients with

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

primary hyperparathyroidism without radiological evidence of skeletal changes. Acta Endocrinol (Copenh) 75997-304. Hahn T J , Boisseau VC, Avioli LV 1973 Effect of chronic corticosteroid administration on diaphyseal and metaphyseal bone mass. J Clin Endocrinol Metab 39274-282. Seeman E, Wahner HW, Offord KP, Kumar R, Johnson WJ, Riggs BL 1982 Differential effects of endocrine dysfunction on the axial and the appendicular skeleton. J Clin Invest 69: 1302- 1309. Parfitt AM, Rao DS, Kleerekoper M, Walczak N, Levin N, Oliver I , Frame B 1980 Midshaft and distal radius bone mineral in primary and secondary hyperparathyroidism: Diagnostic value and response to treatment. In: Mazess RB (ed.) Proceedings Fourth International Conference on Bone Measurement. Publication No. (NIH) 809-1938. Silverberg SJ, Shane E, de la Cruz L, Dempster W , Feldman F, Seldin D, Jacobs T P , Siris ES, Cafferty M, Parisien MV, Lindsay R, Clemens TL, Bilezikian J P 1989 Skeletal disease in primary hyperparathyroidism. J Bone Miner Res 4283290. Bressot C, Courpron P, Edouard C, Meunier P 1976 Histomorphometrie des osteopathies endocriniennes. Association Corporative des Etudiants en Medicine de Lyon, Lyon. pp. 75-82. Richardson ML, Genant HK. Cann CE, Kolb FO, Ettinger B, Gordon GS 1984 Noninvasive assessment of skeletal mass. In: Christiansen C, Arnaud CD, Nordin BEC, Parfitt AM, Peck WA, Riggs BL (eds.) Proceedings of the Copenhagen International Symposium on Osteoporosis. Glostrup Hospital, Denmark. pp. 371-375. Wasnich RD, Ross PD, Heilbrun LK, Vogel J M 1985 Prediction of postmenopausal fracture risk with use of bone mineral measurements. Am J Obstet Gynecol 153:745-751. Gardsell P, Johnell 0, Nilsson BE 1989 Predicting fractures in women by using bone densitometry. Calcif Tissue Int 4: 235-242. Hui SL, Slemenda P H , Johnston C C J r 1989 Baseline measurement of bone mass predicts fracture in white women. Ann Intern Med 111:355-361. Cummings SR, Black DM, Nevitt MC, Browner WS, Cauley JA, Genant HK, Mascioli SR, Scott JC, Seeley DG, Steiger P , Vogt TM, Study of Osteoporotic Fractures Research Group 1990 Appendicular bone density and age predict hip fracture in women. JAMA 263:665-668. Gardsell P, Johnell 0, Nilsson BE 1990 The predictive value of forearm bone mineral content measurements in men. Bone 11 ~229-232.

Address reprint requests to: Robert M . Neer, M . D . Endocrine Unit Buifinch 3 Massachusetts General Hospital Fruit Street Boston, M A 02114