Relationship between serum parathyroid hormone, serum calcium and ...

Download PDF
43 downloads 126 Views 88KB Size Report
Comment
(PTH) levels and arterial blood pressure (BP) in patients with primary ... PTH levels Пr ¼ 0.39, F ¼ 88.36ч, (ii) age and BP, both systolic Пr ¼ 0.61, F ¼ 118.16ч ...
European Journal of Endocrinology (2002) 146 643–647

ISSN 0804-4643

CLINICAL STUDY

Relationship between serum parathyroid hormone, serum calcium and arterial blood pressure in patients with primary hyperparathyroidism: results of a multivariate analysis F Lumachi, M Ermani1, G Luisetto2, A Nardi3, S M M Basso, V Camozzi2 and G Favia Endocrine Surgery Unit, Department of Surgical and Gastroenterological Sciences, 1Biostatistics Section, Department of Neurosciences, 2 Endocrinology Section, Department of Medical and Surgical Sciences, University of Padua School of Medicine, 35128 Padua, Italy and 3 Bone Metabolism Unit, City Hospital, Rovigo, Italy (Correspondence should be addressed to F Lumachi; Email: [email protected])

Abstract Objective: To evaluate the possible relationship between serum calcium, serum parathyroid hormone (PTH) levels and arterial blood pressure (BP) in patients with primary hyperparathyroidism (HPT). Design: A retrospective population-based study. Methods: Charts of 194 patients with proven primary HPT were reviewed, and the main clinical and biochemical parameters were recorded. There were 48 men (24.7%) and 146 women (75.3%), with a median age of 59 years (range 23– 82 years). Patients who used antihypertensive drugs or hormone replacement therapy had been previously excluded. All patients underwent successful parathyroidectomy, and were cured of their disease. Results: There were no differences ðP ¼ NSÞ between men and women in systolic ð143:3^19:1 vs 145:4^17:1 mmHgÞ and diastolic ð87:1^12:3 vs 88:4^9:9 mmHgÞ BP, and in the main biochemical parameters. A significant ðP , 0:01Þ correlation was found between (i) serum calcium and serum PTH levels ðr ¼ 0:39; F ¼ 88:36Þ; (ii) age and BP, both systolic ðr ¼ 0:61; F ¼ 118:16Þ and diastolic ðr ¼ 0:48; F ¼ 64:5Þ; and (iii) body mass index (BMI) and BP (r ¼ 0:45 and 0.36 respectively). There was no significant association of serum calcium levels with systolic ðr ¼ 0:0974; t ¼ 1:3422; P ¼ 0:18Þ or diastolic ðr ¼ 0:1117; t ¼ 1:5409; P ¼ 0:12Þ BP, and of serum PTH levels with systolic ðr ¼ 20:0349; t ¼ 20:4783; P ¼ 0:63Þ or diastolic ðr ¼ 20:0793; t ¼ 21:0913; P ¼ 0:28Þ BP. Multivariate analysis confirmed that none of the independent biochemical parameters significantly correlated with BP, both systolic and diastolic. Conclusions: In patients with primary HPT there is no relationship between PTH, calcium and BP. Thus, in hyperparathyroid patients, BP should be considered as an independent variable, mainly related to age and BMI. European Journal of Endocrinology 146 643–647

Introduction

Subjects and methods

A relationship between primary hyperparathyroidism (HPT) and arterial hypertension has long been suspected, but no convincing studies are available to confirm this hypothesis. Since the prevalence of hypertension in patients with primary HPT is higher than in the general population, both serum calcium (s-calcium) and serum parathyroid hormone (s-PTH) levels could be related to arterial blood pressure (BP). The aim of this study was to analyze whether a correlation exists between BP values, both systolic and diastolic, and the main clinical and biochemical parameters in a population of patients with confirmed primary HPT who underwent successful parathyroidectomy (PTx).

Charts of 194 consecutive patients (48 men (24.7%) and 146 women (75.3%)) with confirmed primary HPT were reviewed. Patients with a history of myocardial infarction, angina, stroke, diabetes mellitus, and those who used hormone replacement therapy, had been previously excluded. Also, we excluded from this study 18 patients (median age 58 years, range 52 – 78) who were treated for hypertension because BP values, modified by treatment, can potentially affect the correlation analysis, and a period of washout from treatment was impossible in patients undergoing surgery the next day. Their main clinical and biochemical data did not differ significantly, according

q 2002 Society of the European Journal of Endocrinology

Online version via http://www.eje.org

644

F Lumachi and others

EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 146

Table 1 Main clinical and laboratory data in the overall population. Parameter

Reference intervals

Mean 6 S.D.

Median

Range

— — — — 53–97 1.9 –2.6 10–55

57.7^12.9 24.7^12.6 144.8^17.5 88.0^10.5 81.9^25.3 2.95^0.33 188.4^154.5

59 25.5 150 90 78 2.83 140

23–82 22.4–29.2 105 –180 60–110 28–109 2.62–4.60 60–990

Age (years) BMI (kg/m2) Systolic BP (mmHg) Diastolic BP (mmHg) Creatinine (mmol/l) Calcemia (mmol/l) PTH (ng/l)

to age, with respect to patients who did not use antihypertensive drugs. Sixty-seven of 194 patients (34.5%) were asymptomatic, and two (1.0%) had multiple endocrine neoplasia type 1. According to the 1990 NIH Consensus Conference suggestions the main criteria for surgery in asymptomatic patients were a value of s-calcium $ 3.0 mmol/l, or decrease of bone mass to 2 S.D. less than the mean for age- and sex-matched control persons (1). After patients had rested in the supine position for at least 10 min (between 0800 and 0900 h) venous blood was sampled, and measurements of serum creatinine (s-creatinine), s-calcium and intact s-PTH (PTH(184)) were obtained. S-PTH was measured using a two-site chemiluminescent immunometric assay, while s-calcium and s-creatinine were measured by standard laboratory methods. Arterial BP was recorded using an automatic device. Three recordings were made at 2– 3 min intervals the day before the operation, and both mean systolic and diastolic values (mmHg) were recorded in a database of information. BP was also recorded using the same modalities on the last postoperative day, 2– 5 days after PTx. Age, body mass index (BMI) and the main clinical and biochemical data are reported in Table 1. To better evaluate the influence of age on the relationship between s-PTH, s-calcium and BP the population was divided into five age-groups, as shown on Table 2. All patients underwent successful PTx, and were cured of their disease. The reported data are expressed as means^S.D. A two-tailed Student’s t-test for unpaired data was used to compare means of grouped data. ANOVA was used to evaluate the linear relationship between pairs of variables. The regression results

were expressed using the standardized b coefficient and the partial correlation coefficient (r ) calculation. Furthermore, a measure of collinearity was provided by R 2i, which is the proportion of the variability in xi that is explained by the other variables when xi is the dependent variable in a regression on the others x values. The significant level was set at P , 0:01:

Results There were no differences ðP ¼ NSÞ between men and women in systolic ð143:3^19:1 vs 145:4^17:1 mmHg) and diastolic ð87:1^12:3 vs 88:4^9:9 mmHgÞ BP, and in the main biochemical parameters such as s-creatinine ð79:8^34:8 vs 76:1^18:0 mmol=lÞ; s-calcium ð3:06^0:46 vs 2:91^0:27 mmol=lÞ; and s-PTH ð210:3^176:6 vs 181:3^146:6 ng=lÞ: Women were significantly ðP , 0:01Þ older (59:6^11:6 vs 51:9^15:0 years) than men. In the overall population, ANOVA showed a significant linear correlation between age and BP, both systolic ðr ¼ 0:61; F ¼ 118:16; P , 0:01Þ and diastolic ðr ¼ 0:48; F ¼ 64:5; P , 0:01Þ: Also, a positive correlation of s-calcium with s-PTH levels ðr ¼ 0:38632; F ¼ 33:68; P , 0:01Þ was found. There was no correlation between s-calcium and BP, and between s-PTH and BP, both systolic and diastolic. As shown on Table 2, s-calcium levels were lower ðP , 0:01Þ in older patients (Group 5 vs Groups 1, 2 and 3), while s-creatinine and s-PTH values did not differ ðP ¼ NSÞ among the age-groups of patients. No relationship ðP ¼ NSÞ between BP and s-calcium, and between BP and s-PTH, was found in either age group, although the r value was relatively higher

Table 2 Main characteristics of different age-groups of patients means^S.D.

Group

Age (years)

No. patients (%)

S-creatinine (mmol/l)

S-calcium (mmol/l)

S-PTH (ng/l)

Systolic BP (mmHg)

Diastolic BP (mmHg)

1 2 3 4 5

23–39 40–49 50–59 60–69 .69

19 (9.8%) 26 (13.4%) 56 (28.9%) 59 (30.4%) 34 (17.5%)

83.4^14.8 73.8^24.4 83.6^34.4 82.3^20.7 84.2^20.2

3.04^0.34 2.98^0.33 3.04^0.44 2.92^0.26 2.79^0.20

199.5^155.6 199.7^140.5 178.8^159.4 175.1^127.6 212.7^198.3

120.0^9.4 137.1^13.1 142.6^15.0 146.3^14.8 153.3^14.6

72.6^9.2 79.5^10.9 84.4^9.0 87.4^8.4 91.8^8.7

www.eje.org

Blood pressure in primary hyperparathyroidism

EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 146

645

Table 3 Main characteristics of hypertensive and normotensive patients (means^S.D. ). Patients Blood pressure (mmHg) Systolic . 160 # 160 P Diastolic . 90 # 90 P

n

%

Age (years)

S-creatinine (mmol/l)

S-calcium (mmol/l)

S-PTH (ng/l)

25 169 —

12.9 87.1 —

68.9^7.6 56.1^12.8 0.00

81.1^25.4 88.5^33.7 0.29

2.96^0.34 2.96^0.42 1.00

187.5^155.4 225.3^213.4 0.39

69 125 —

35.6 64.4 —

64.2^10.6 54.3^12.9 0.00

86.1^29.6 79.3^23.7 0.08

2.98^0.43 2.94^31.3 0.46

207.9^192.5 183.9^146.7 0.33

(systolic BP vs calcemia) in Groups 4 ðr ¼ 0:17392Þ and 5 ðr ¼ 0:21955Þ: A higher but not significant ðP ¼ NSÞ r value (systolic BP vs s-calcium) was found in Groups 4 ðr ¼ 0:11739Þ and 5 ðr ¼ 0:13656Þ: Postoperative BP values, both systolic ð144:8^17:5 vs 141:8^18:4 mmHg; t ¼ 1:646; P ¼ 0:10Þ and diastolic ð88:0^10:5 vs 86:9^11:3 mmHg; t ¼ 0:993; P ¼ 0:32Þ did not differ significantly ðP ¼ NSÞ from preoperative BP values. Patients with systolic BP . 160 mmHg and diastolic BP . 90 mmHg were significantly ðP , 0:01Þ older, while s-creatinine, s-calcium and s-PTH levels did not differ ðP ¼ NSÞ between the subgroups of hypertensive and normotensive patients (Table 3). Multivariate analysis showed that no independent biochemical parameters correlated with BP, and confirmed the linear correlation between BP, both systolic and diastolic, age and BMI (Table 4). Lower BP values were recorded in younger patients (, 50 years) with higher calcium and PTH levels; this subgroup represented less than a quarter of the total. Since age is a major determinant of BP and the mean age of the whole group is 60 years, the relationship between s-PTH, s-calcium and BP was assessed in subgroups of patients aged , 50 and $ 50 years (Table 5). The results did not differ from those obtained in the overall population.

Discussion Primary HPT represents the most common cause of hypercalcemia, and its estimated age-adjusted incidence is 42 per 100 000 (2). Arterial hypertension is common in patients with primary HPT with reported rates ranging between 10 and 40%, and an average being about 30% (3). In 1982 Nainby-Luxemore et al. (4) compared the frequency of hypertension in a group of 124 patients with primary HPT with that found in cases matched for age and conditions considered of comparable surgical magnitude. The rate of hypertension was significantly ðP , 0:01Þ higher (73 vs 43%), but the BP elevation was modest. Possible mechanism which have been postulated for the supposed hypertension include the direct effect of calcium and PTH and the consequences of renal damage produced by the hypercalcemia. Ude´n et al. (5), reporting clinical and biochemical data concerning 250 patients with primary HPT, found that the rate of hypertension was significantly higher (47 vs 28%), in patients older than 60 with respect to younger patients. However, since 60 –70% of the patients with primary HPT were over 60 years of age, different causes of hypertension should be investigated, as BP is strongly related to age (1, 5, 6). In the present series, 35% of the patients

Table 4 Results of multivariate analysis for systolic and diastolic BP in the overall population.

Systolic BP Age (years) BMI (kg/m2) S-creatinine (mmol/l) S-calcium (mmol/l) S-PTH (ng/l) Diastolic BP Age (years) BMI (kg/m2) S-creatinine (mmol/l) S-calcium (mmol/l) S-PTH (ng/l)

b

r

R2

t

P

0.637175 0.473122 0.141095 0.085547 20.029155

0.614035 0.454172 0.154018 0.097427 20.034863

0.139471 0.082453 0.295462 0.244135 0.173642

10.6669 6.49521 2.13729 1.34223 20.47830

0 0.0001 0.03 0.18 0.63

0.500002 0.382214 0.159169 0.109398 20.074098

0.481275 0.361471 0.155933 0.111681 20.079340

0.139471 0.082453 1.295462 0.244135 1.173642

7.52810 5.97311 2.16452 1.54093 21.09129

0 0.002 0.03 0.12 0.28

www.eje.org

646

F Lumachi and others

EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 146

Table 5 Results of multivariate analysis for systolic and diastolic BP in patients ,50 and $50 years.

Systolic BP , 50 years ðn ¼ 45Þ S-creatinine (mmol/l) S-calcium (mmol/l) S-PTH (ng/l) $ 50 years ðn ¼ 149Þ S-creatinine (mmol/l) S-calcium (mmol/l) S-PTH (ng/l) Diastolic BP , 50 years ðn ¼ 45Þ S-creatinine (mmol/l) S-calcium (mmol/l) S-PTH (ng/l) $ 50 years ðn ¼ 149Þ S-creatinine (mmol/l) S-calcium (mmol/l) S-PTH (ng/l)

b

r

R2

t

P

0.163560 0.046475 20.054952

0.161540 0.046275 20.054552

0.374824 0.379852 0.383034

1.02224 0.28930 20.341184

0.31 0.77 0.73

0.134894 0.126059 20.024166

0.124958 0.122632 20.024973

0.329394 0.260853 0.177989

1.50608 1.47761 20.298724

0.13 0.14 0.76

0.221015 0.127989 20.143433

0.223371 0.131011 20.146162

0.374824 0.379852 0.383034

1.43111 0.825277 20.922480

0.16 0.41 0.36

0.134894 0.126059 20.024166

0.124958 0.122632 20.024973

0.329394 0.260853 0.177989

1.50608 1.47761 20.298724

0.13 0.14 0.76

had diastolic BP . 90 mmHg. This finding can more likely be due to the inclusion of hypertensive patients not treated, but could also be ascribed to the preoperative stress, although the mean short-term postoperative BP did not differ ðP ¼ NSÞ from preoperative BP. BMI is a known determinant of BP. It has been shown that a BMI $ 25 was associated with substantially higher BP, and a gain of 1.25 –1.7 kg/m2 in BMI may correspond to an elevation of 1 mmHg in BP, in some studies (7, 8). No data are available regarding possible connections between primary HPT and BMI. PTH is an important regulator of calcium homeostasis, involving bone and kidney. A direct relationship between PTH and hypertension has also been hypothesized, and a positive association of calcemia with PTH in patients with primary HPT is well known (9 – 11). In healthy human subjects chronic infusion of PTH has been shown to result in increasing BP levels, and moreover a co-action of PTH and PTH hypertensive factor has also been suggested in experimental studies (12– 14). Several recent clinical studies report that in hypertensive patients calcium metabolism is altered, and PTH levels are frequently increased (14 – 16). In fact, calciotropic hormones, such as PTH and calcitriol, have been shown to be increased in some forms of hypertension. St John et al. (16), in a group of hypertensive elderly subjects, showed that PTH was an independent significant predictor of mean BP. Moreover, PTH levels, which are elevated both in patients with HPT and in those with essential hypertension, can probably influence intracellular calcium levels and subsequently smooth muscle contraction (17). Significant relationships were also found between cytosolic free calcium and magnesium and BP, independent of age (18). Although intracellular calcium may play a role in the genesis of essential hypertension facilitating vascular smooth muscle contraction, no correlation www.eje.org

between BP and intracellular calcium in patients with primary HPT was found (19). Hunt et al. (20) in a group of healthy normotensive individuals found that there was no linear correlation of BP with plasma ionized calcium. However, a significant inverse correlation between ionized calcium and BP in patients with low renin levels was observed. More recently, Jorde et al. (21), in a population of 946 unselected patients who were living in Norway, confirmed a linear correlation between age and BP, but no association between BP and s-calcium was found. In women with elevated PTH levels due to causes other than HPT, both systolic and diastolic BP were higher ðP , 0:01Þ than in those with normal PTH levels, and patients with hypertension had s-PTH levels significantly higher than the normotensive (21, 22). We did not find such difference between male and female patients with primary HPT, in spite of the women being older than the men. In a group of 583 elderly patients untreated for hypertension and who did not suffer from primary HPT, calcium and PTH levels significantly correlated with mean BP, and PTH correlated positively with age (16). Similar results were obtained in patients with essential hypertension, confirming the presence of alteration of calcium metabolism in those patients (20). Finally, the relationships between age, PTH and BP have also been investigated in normotensive patients by Brickman et al. (23), who found a significant correlation between PTH and both systolic and diastolic BP. Multivariate analysis showed that PTH maintained significant correlation with mean arterial BP. Few studies consider the possible relationship between BP, calcium and PTH in patients with primary HPT, who typically are hypercalcemic with elevated s-PTH levels. It may probably be justified by the known chemical heterogeneity of hypertension in

Blood pressure in primary hyperparathyroidism

EUROPEAN JOURNAL OF ENDOCRINOLOGY (2002) 146

patients with primary HPT, since both low and high renin activity values were observed in different primary hyperparathyroid hypertensive subjects (20). In a group of 441 patients with primary HPT studied by Lind et al. (24) the preoperative systolic BP was significantly ðP , 0:05Þ correlated with s-calcium only when the influence of age was not considered. Therefore, they concluded that there is no simple cause-andeffect relationship accounting for hypertension in primary HPT. In our series both systolic and diastolic BP were related exclusively to age and BMI, and there was no correlation between BP and the main biochemical parameters. Sancho et al. (25) found that 12 of 56 patients (21.8%) with HPT were hypertensive, but the clinical and biochemical parameters that were considered did not differ significantly ðP ¼ NSÞ between the subgroups of normotensive and hypertensive patients. Similar results were described by Ude´n et al. (5) for patients . 60 years and , 60 years. In conclusion, in patients with primary HPT the hypothesized correlation between s-calcium and BP, and between s-PTH and BP does not seem to exist, and thus in such patients arterial BP should be considered as an independent variable, mainly related to age.

Acknowledgements This work was presented at the 82nd Annual Meeting of the Endocrine Society, Toronto, Canada, 21-24 June 2000.

8

9 10 11

12 13 14 15 16

17 18 19

References 1 NIH Consensus Conference. Diagnosis and management of asymptomatic primary hyperparathyroidism: consensus development conference statement. Annals of Internal Medicine 1991 114 593– 597. 2 Arnaud CD. The calciotropic hormones and metabolic bone disease. In Basic and Clinical Endocrinology, edn 4, ch 5, pp 227–306. Eds FS Greenspan & JD Baxter. East Norwalk, CT, USA: Appleton & Lange, 1994. 3 Dluhy RG. Uncommon forms of secondary hypertension in older patients. American Journal of Hypertension 1998 11 52S –56S. 4 Nainby-Luxemore JC, Langford HG, Nelson NC, Watson RL & Barnes TY. A case-control study of hypertension and hyperparathyroidism. Journal of Clinical Endocrinology and Metabolism 1982 55 303 –306. 5 Ude´n P, Chan A, Duh Q-Y, Siperstein A & Clark OH. Primary hyperparathyroidism in younger and older patients: symptoms and outcome of surgery. World Journal of Surgery 1992 16 791–798. 6 Lumachi F, Ermani M, Basso S, Zucchetta P, Borsato N & Favia G. Localization of parathyroid tumours in the minimally invasive era: which technique should be chosen? Population-based analysis of 253 patients undergoing parathyroidectomy and factors affecting parathyroid gland detection. Endocrine-Related Cancer 2001 8 63 –69. 7 Kumanyika SK, Landis JR, Mattews YL, Weaver SL, Harlan LC & Harlan WR. Secular trends in blood pressure among blacks and whites aged 18 –34 years in two body mass index strata, United

20

21

22 23 24

25

647

States, 1960–1980. American Journal of Epidemiology 1994 139 141–154. Doll S, Paccaud F, Bovet P, Burnier M & Wietlisbach V. Body mass index, abdominal adiposity and blood pressure: consistency of their association across developing and developed countries. International Journal of Obesity and Related Metabolic Disorders 2002 26 48 –57. Mok LL, Nickols GA, Thompson JC & Cooper CW. Parathyroid hormone as a smooth muscle relaxant. Endocrine Reviews 1989 10 420 –436. Resnick LM. Calcitrophic hormones in human and experimental hypertension. American Journal of Hypertension 1990 3 171S–178S. Lumachi F, Zucchetta P, Marzola MC, Boccagni P, Angelini F, Bui F et al. Advantages of combined technetium-99 m-sestamibi scintigraphy and high-resolution ultrasonography in parathyroid localization: comparative study in 91 patients with primary hyperparathyroidism. European Journal of Endocrinology 2000 143 755 –760. Hulter HN, Melby JC, Peterson JC & Cooke CR. Chronic continuous PTH infusion results in hypertension in normal subjects. Journal of Clinical Hypertension 1986 2 360–370. Pang PK, Lewanczuk RZ & Benishin CG. Parathyroid hypertensive factor. Journal of Hypertension 1990 8 S155–S159. Morfis L, Smerdely P & Howes LG. Relationship between serum parathyroid hormone levels in the elderly and 24-h ambulatory blood pressure. Journal of Hypertension 1997 15 1271–1276. Young EW, McCarron DA & Morris CD. Calcium regulating hormones in essential hypertension. Importance of gender. American Journal of Hypertension 1990 3 161S–166S. St John A, Dick I, Hoad K, Retallack R, Welborn T & Prince R. Relationship between calcitrophic hormones and blood pressure in elderly subjects. European Journal of Endocrinology 1994 130 446–450. Brickman AS, Nyby MD, von Hungen K, Eggena P & Tuck ML. Calcitropic hormones, platelet calcium, and blood pressure in essential hypertension. Hypertension 1990 16 515–522. Barbagallo M, Gupta RK, Dominguez LJ & Resnick LM. Cellular ionic alterations with age: relation to hypertension and diabetes. Journal of the American Geriatric Society 2000 48 1111–1116. Fardella C & Rodriguez-Portales JA. Intracellular calcium and blood pressure: comparison between primary hyperparathyroidism and essential hypertension. Journal of Endocrinological Investigation 1995 18 827 –832. Hunt SC, Williams RR & Kuida H. Different plasma ionized calcium correlations with blood pressure in high and low renin normotensive adults in Utah. American Journal of Hypertension 1991 4 1–8. Jorde R, Bonaa KH & Sundsfjord J. Population based study on serum ionised calcium, serum parathyroid hormone, and blood pressure. The Tromsø study. European Journal of Endocrinology 1999 141 350–357. Jorde R, Sundsfjord J, Haug E & Bonaa KH. Relation between low calcium intake, parathyroid hormone, and blood pressure. Hypertension 2000 35 1154–1159. Brickman AS, Nyby MD, von Hungen K, Eggena P & Tuck ML. Parathyroid hormone, platelet calcium and blood pressure in normotensive subjects. Hypertension 1991 18 176 –182. Lind L, Hvarfner A, Palmer M, Grimelius L, Akerstrom G & Ljunghall S. Hypertension in primary hyperparathyroidism in relation to histopathology. European Journal of Surgery 1991 157 457 –459. Sancho JJ, Rouco J, Riera-Vidal R & Sitges-Serra A. Long-term effects of parathyroidectomy for primary hyperparathyroidism on arterial hypertension. World Journal of Surgery 1992 16 732–736.

Received 6 December 2001 Accepted 13 February 2002

www.eje.org