Hyperparathyroidism in the hemodialysis population: A survey of 612 patients

Hyperparathyroidism in the Hemodialysis Population: A Survey of 612 Patients Mahmoud

M. Salem, MD

0 There are no epidemiologic studies documenting the prevalence of hyperparathyroidism in the US hemodialysis population. We looked at a random sample of 612 hemodialysis patients enrolled in 10 outpatient dialysis units in Mississippi. Fifty percent of the patients surveyed had an intact serum parathyroid hormone (PTH) level more than three times normal (mean, 622 pg/mL). Another 25% had a less than normal PTH level (mean, 33 pg/mL), suggesting adynamic bone disease. Abnormal serum calcium was also common. Seventeen percent of patients were hypocalcemic and 14% were hypercalcemic. These high point prevalences occurred despite widespread use of calcium supplements and/or vitamin D (used in 90% of the patients surveyed). Black patients tended to have a lower serum calcium and higher PTH level than white patients. We also found that diabetic patients are less likely to have an elevated PTH level than nondiabetic patients. Elevated serum phosphorus was the most important factor correlating with the development of secondary hyperparathyroidism. Causes of inadequate control of hyperparathyroidism in this population require further study. 0 1997 by the National Kidney Foundation, Inc. INDEX WORDS: hormone; serum

Hyperparathyroidism; calcium; serum

end-stage renal phosphorus; epogen;

R

ENAL bone disease has been recognized for a long time. The spectrum of bone disease in the dialysis population has been changing.’ In addition to secondary hyperparathyroidism and aluminum bone disease, adynamic bone disease has emerged as a separate entity. Potent vitamin D preparations administered both intravenously and orally are now widely used. Changing dialysate calcium concentration and use of large doses of oral calcium preparations instead of aluminum salts also has made an impact. Research protocols involving all the above measures have found a significant improvement in renal bone disease. Outcome research involving large numbers of dialysis patients in an outpatient setting to document epidemiology of renal bone disease and how effective these measures are in real practice has been lacking. This study aims to describe the current status of bone disease in a large number of hemodialysis patients from Mississippi. MATERIALS We reviewed the complete modialysis patients enrolled

AND

METHODS

computerized files in 10 hemodialysis

of 612 hecenters in

From the Department of Medicine, University of Mississippi Medical Center, Jackson, MS. Received July 16, 1996; accepted in revisedform February 7, 1997. Address reprint requests to Mahmoud M. Salem, MD, Department of Medicine, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216. 0 1997 by the National Kidney Foundation, Inc. 0272-6386/97/2906-0007$3.00/O 862

American

Journal

disease; morbidity.

hemodialysis;

chronic

renal

failure;

parathyroid

Mississippi. All 10 units are run by Kidney Care, Inc (Jackson, MS). The average of the last two intact serum parathyroid hormone (PTH) levels, and concomitant serum calcium, albumin, and other pertinent laboratory parameters were extracted from the files and used in the analysis. The two PTH assays were 3 months apart. Hyperparathyroidism was defined as a PTH level more than three times normal (> 195 pg/mL). Hypocalcemia was defined as total serum calcium less than 8.5 mg/dL, normocalcemia as between 8.5 and 10.5 mg/dL, and hypercalcemia as more than 10.5 mg/dL. Patients were observed and doses of vitamin D and calcium were prescribed by several practicing nephrologists according to the standards of care. The standards of care regarding bone disease in our hemodialysis patients include the following goals: serum calcium within normal limit, serum phosphorus below 5 mg/dL, and lack of symptoms alluding to bone disease. Our general overall management plan is as follows: calcium supplements are used initially in the majority of patients to increase serum calcium and bind phosphorus. Patients are instructed to take it with meals, unless their serum calcium is extremely low. Vitamin D is usually added at the discretion of the nephrologist for patients with persistent elevation of PTH levels and/or hypocalcemia despite compliance with calcium. We attempt to lower serum phosphorus before giving high doses of vitamin D. All the vitamin D doses were administered intravenously at the end of the dialysis session. Unpaired Student’s t-test was used for comparisons involving two groups. Simple and multiple regressions were used to asses factors correlating with PTH levels. Unless otherwise stated, all data are reported as the mean i SD, and P < 0.05 was considered significant. Stat-View computer software (Abacus Concepts, Inc, Berkeley, CA) was used.

RESULTS

Description

of the Study Population

Demographics, dialysis duration, and reported causes of end-stage renal disease in our study of Kidney

Diseases,

Vol 29, No 6 (June),

1997:

pp 862-865

HYPERPARATHYROIDISM

863

IN HEMODIALYSIS

population are shown in Table 1. All patients were dialyzed with cuprophan-based dialyzers. The average dialysis session was 236 minutes (approximately 4 hours) three times a week. Our standard dialysate bath is 3.5 mEq/L. The average urea reduction ratio was 0.59. Ninety percent of the patients were receiving either calcium supplements, vitamin D, or both (Table 2). The average serum calcium and other bone-related variables are shown in Table 3. In general, the average serum calcium was within normal limits, while serum phosphorus was high. Bicarbonate level was low. The mean PTH level was above three times normal. The albumin level was normal. Prevalence of Hyperparathyroidism One hundred fifty-three patients (25% of the total) had a serum intact PTH level within the normal range (> 65 pg/mL). Another 25% had a serum PTH level more than one but less than three times normal (65 to 195 pg/mL). Three hundred five patients (50%) had PTH levels more than three times normal (> 195 pg/mL). The mean PTH levels for the above three groups were 33, 119, and 622 pg/mL, respectively. Seventeen percent of the patients had hypocalcemia, 69% had normocalcemia, and 14% had hypercalcemia. Black patients tended to have higher serum PTH levels and lower serum calcium than white patients (Table 4). Factors Correlating

With PTH Level

In simple regression, serum phosphorus correlated positively with PTH level, while calcium, Table

1. Characteristics

of the

Study

Population No. of Patients, N=612(%)

Males Females Blacks Whites Cause of ESRD Hypertension Diabetes mellitus Glomerulonephritis Polycystic kidney Other

disease

NOTE. The mean patient mean time on dialysis was

292 320 550 62

(48) (52) (90) (10)

317 177 48 12 58

(51) (29) (8)

age was 56 I 15 years. 49 ? 51 months.

Table

2. Use

of Vitamin D and Calcium in the Study Population No. of Patients

On On On On

vitamin D alone calcium alone vitamin D and calcium neither calcium nor vitamin D Total

The

53 (9) 250 (41) 245 (40)

3.1 1.7 2.312.6

64 (10) 612 (100)

o/o 1.711.3

*The dose of calcium is given in grams of elemental calcium per day. The dose of vitamin D is given in micrograms per week.

bicarbonate, urea reduction ratio, and age correlated negatively with PTH level (Table 5). In multiple regression, phosphorus continued to correlate positively with PTH level and age had a negative correlation. All other variables had no correlation to PTH level in multiple regression. It is of note that the serum PTH level was significantly higher in patients receiving vitamin D (383 ?c 448) than in those not receiving vitamin D (315 + 402) (P = 0.04). Diabetic Versus Nondiabetic

Patients

Diabetic patients as a group had a lower serum PTH level than nondiabetic patients despite similar levels of serum calcium, phosphorus, and bicarbonate (Table 6). Analysis of the difference in PTH level depending on diabetic status and taking all the significant factors into consideration (as covariates in ANCOVA) again showed the same difference (F value, 8.3; P = 0.003). This may be due to a stronger negative feedback of serum calcium on PTH level, as shown by a significant correlation coefficient of -0.20 (P =

3. Bone-Related

Factor

(2)

Mean Dose of Vitamin D/Calcium*

VW

Table

(10)

Preparations

Calcium (mg%) Phosphorus (mg%) Bicarbonate (mEq/L) Aluminum (mg%) Albumin (9%) PTH (pg/m L)

Variables Mean

9.3 6.2 19.6 18.3 4.0 348

2 SD

t -c + + -c i

1.0 1.8 3.3 16.5 0.4 426

MAHMOUD

864 Table

4. Bone-Related and White

Factor Calcium

Parameters Patients Whites

Blacks

(mg%)

9.3

Phosphorus (mg%) Bicarbonate (mEq/L) Aluminum (mg%) Albumin (9%) PTH (pg/mL)

6.2 19.6 18 4.0 363

9.6 6.1 19.8 21 3.9 214

i- 1.0

2 I 2 2 +-

1.8 3.2 16 0.4 440

Table

in Black

t i ? i i IT

Factor

P Value

0.99 1.9 3.2 19 0.3 226

0.03 0.59 0.70 0.14 0.19 0.008

PTH (pg/mL) Serum calcium O-w% Serum phosphorus b-W@ Serum bicarbonate (m EN-)

0.005) in the diabetic patients versus r of -0.08 (P = 0.06) in the nondiabetic patients. DISCUSSION

We chose a plasma intact PTH level above 195 pg/mL as our cutoff point since at this level the positive predictive value is at 88% for the diagnosis of mild and severe hyperparathyroidism. A value below 65 pg/mL has a positive predictive value of 78% for the diagnosis of adynamic bone disease.2 Our results indicate a high prevalence of hyperparathyroidism in the hemodialysis population despite widespread use of calcium supplements and vitamin D. Fifty percent of the patients had a serum PTH level more than three times normal. Our group of patients with more than three times the upper limit of PTH level had an average PTH level of 622 pg/mL. This is comparable to the group of biopsy-proven parathyroid bone disease reported by Pei et a1.3 We did not find any comparable epidemiologic study addressing the prevalence of hyperparathyroidism in the US hemodialysis population. A study from Canada involving 268 patients found evidence of parathyroid bone disease in 72 (61%) of 117 hemodialysis patients. The prevalence of

Table

5. Factors Parathyroid

Correlating Hormone

With

Multiple Regression

Factor

Simple Regression

Calcium Phosphorus Bicarbonate Urea reduction

-0.12* 0.18* -0.13* -0.08

-0.07 -0.15* -0.06 -0.07

-0.1.5*

-0.12*

ratio

Age *Statistically

significant

value:

r

P < 0.05.

6. Bone-Related Compared With

r

Variables Nondiabetic

Diabetics

265

i 338

M. SALEM

in Diabetic Patients

Nondiabetics

382

?I 453

P Value

0.002

9.4 lb 0.9

9.3 2 1.0

0.39

6.2 IT 1.7

6.2 k 1.9

0.93

2 3.4

0.54

19.5

-t 3.1

19.7

secondary hyperparathyroidism was less in peritoneal dialysis patients (48 of 142 patients [34%]). The average serum calcium in our patients was within the normal range at 9.3 mg/dL. It may be that higher than normal serum calcium is required to control hyperparathyroidism in these patients.4 The lack of correlation between serum calcium and PTH level in multiple regression suggests an abnormal calcium-parathyroid hormone (CaPTH) feedback. This has been shown to be a factor in the genesis of secondary hyperparathyroidism.5 More evident was the positive correlation between the serum phosphorus and PTH level in both simple and multiple regression. This suggests elevated serum phosphorus as the main pathogenetic mechanism for secondary hyperparathyroidism. In addition, the mean serum phosphorus in our population was elevated at 6.2 mg/dL. In 1971, Slatopolsky et al6 stressed the role of phosphorus retention in the pathogenesis of secondary hyperparathyroidism. In the same year, Fournier et al7 found a correlation between serum phosphorus and PTH level in the chronic hemodialysis patient. This was thought to lower serum calcium and consequently activate PTH. Recently, Fine et al* showed the important role serum phosphorus plays in the pathogenesis of secondary hyperparathyroidism in chronic renal failure, irrespective of serum calcium level. All the above results are in accordance with our findings. Acidosis (reflected by a low serum bicarbonate) was associated with a higher PTH level in simple regression. The importance of this new finding and the magnitude of contribution of the acidosis to hyperparathyroidism is difficult to ascertain from our limited study. It is of note that in multiple regression, the bicarbonate level no

HYPERPAFiATHYROlDlSM

865

IN HEMODIALYSIS

longer correlated with the PTH level, suggesting that the effect of acidosis is not an independent contributor to PTH (it may act through influencing other factors). Younger patients had a higher PTH level. The effect of age was evident even after adjusting for serum calcium, serum phosphorus, and other variables known to influence the PTH level in multiple regression. This suggests that the difference is true and reflects a more biologically active parathyroid gland in the young. Adequacy of dialysis as reflected by urea reduction ratio also appeared to ameliorate the increase in the PTH level. This may act through multiple mechanisms, including higher serum calcium and more phosphate clearance or even PTH clearance. We did not study the effect of the dialysis membrane on PTH level since the vast majority of our patients are dialyzed with the same cellulose-based membrane. Patients with normal PTH levels may be suffering from adynamic bone disease. Twenty-five percent of our patients had a PTH level within the normal range. Black patients tended to have a higher PTH level than white patients. This may be due to a lower average serum calcium in black patients. We did not find any difference in any other bone-related parameters between black and white patients. However, only 10% of our patients are white, and we may be missing an otherwise significant difference due to the disparity in size of the two populations. Our results show a higher PTH level in patients receiving vitamin D. We believe this may represent a bias introduced by treatment. Patients with a higher PTH level tended to receive vitamin D preferentially. In conclusion, the findings of our study agree with prior evidence’ regarding the importance of elevated serum phosphorus in the genesis of

secondary hyperparathyroidism. They also reveal a high prevalence of the disease despite widespread use of calcium and vitamin D preparations. The reasons for this phenomenon are not clear and require further study. ACKNOWLEDGMENT The author thanks Kidney Care, Inc (Jackson, MS) and Dr John Bower for support of this research. The author acknowledges the help and expertise of Edna Curry, RN, and Michelle Curry, without whom this work could not have been completed.

REFERENCES 1. Sherrard DJ, Hertz G, Pei Y, Maloney N, Greenwood C, Manuel A, Saiphoo C, Fenton S, Segre G: The spectrum of bone disease in end-stage renal failure-An evolving disorder. Kidney Int 43:436-442, 1993 2. Hutchison AJ, Wbitehouse RW, Boulton HF: Correlation of bone histology with parathyroid hormone, vitamin D and radiology in end-stage renal disease. Kidney Int 44:10711077, 1993 3. Pei Y, Hertz G, Greenwood C, Segre G, Manuel A, Saiphoo C, Fenton S, Sherrard D: Risk factors for renal osteodystrophy. J Bone Miner Metab 10: 149-156, 1995 4. Brown EM, Wilson RE, Eastman R: Abnormal regulation of parathyroid hormone release by calcium in secondary hyperparathyroidism due to chronic renal failure. J Clin Endocrinol Metab 54:172-177, 1982 5. Falsenfeld AJ, Llach F: Parathyroid gland function in chronic renal failure. Kidney Int 43:771-789, 1993 6. Slatopolsky E, Caglar S, Pennel JP: On the pathogenesis of hyperparathyroidism in chronic experimental renal insufficiency in the dog. J Clin Invest 50:492-499, 1971 7. Foumier AE, Amaud CD, Johnson WJ: Etiology of hyperparathyroidism and bone disease during chronic hemodialysis. II. Factors affecting serum immunoreactive parathyroid hormone. J Clin Invest 50:599-605, 1971 8. Fine A, Cox D, Fontaine B: Elevation of serum phosphate affects parathyroid hormone levels in only 50% of hemodialysis patients, which is unrelated to changes in serum calcium. J Am Sot Nephrol 3:1947-1953, 1993 9. Yi H, Fukagawa M, Kurokawa K: Mild dietary phosphorus restriction directly prevents enhanced parathyroid hormone secretion and synthesis and proliferation of parathyroid cells in chronic renal failure in rats. J Am Sot Nephrol3:703707, 1992