(Hypertension. 1999;34:484-490.)
© 1999 American Heart Association, Inc.
Scientific Contributions |
Serum Calcium and Cardiovascular Risk Factors and Diseases
The Tromsø Study
From the Departments of Medicine (R.J.) and Clinical Chemistry (J.S.), University Hospital of Tromsø, and the Institute of Community Medicine (K.H.B.), University of Tromsø, Tromsø, Norway; and the Center for Clinical Epidemiology and Biostatistics (R.J., P.F.), Faculty of Medicine and Health Sciences, University of Newcastle, Newcastle, Australia.
Correspondence to Rolf Jorde, MD, Department of Internal Medicine, University Hospital of Tromsø, 9038 Tromsø, Norway. E-mail medrj{at}rito.no
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract—Total serum calcium levels were measured in 12 865 men and 14 293 women, between the ages of 25 and 97 years, in the Tromsø Study during 1994 and 1995. With the use of a sex-specific multiple linear regression model with age, calcium, body mass index, cholesterol, HDL cholesterol, triglycerides, systolic and diastolic blood pressure, and pulse as possible covariates, serum calcium was significantly (P<0.001) and positively associated with systolic and diastolic blood pressure, serum cholesterol, and HDL cholesterol in both sexes. A similar but weaker association was observed between serum calcium and triglycerides in men (P<0.01). In all age groups, serum calcium levels were higher in men with a history of myocardial infarction than in those without, and the difference was significant (P<0.0001) in a linear regression analysis adjusted for age. When all the other variables were also included in a logistic regression model, serum calcium was a highly significant (P<0.0001) predictor of myocardial infarction in men, with an odds ratio of 1.2 per 0.1 mmol/L increase in serum calcium. In women, a nonsignificant trend was again seen. Because the free or ionized form of calcium is the physiologically important form and serum calcium was not corrected for serum albumin in our study, the results must be interpreted with caution. However, it appears likely that serum calcium is a predictor of cardiovascular disease in men.
Key Words: calcium • blood pressure • cholesterol • triglycerides • cardiovascular diseases
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In epidemiological studies, serum calcium has been found to increase with increasing levels of systolic and diastolic blood pressure (BP).1 2 Furthermore, in patients with hyperparathyroidism and therefore chronic hypercalcemia, hypertension is prevalent3 and an increased mortality rate exists, particularly from cardiovascular diseases.4 Serum calcium has also been found to correlate with serum cholesterol and blood glucose2 and therefore is apparently associated with the metabolic syndrome. Recently, serum calcium has been reported to be an independent, prospective risk factor for myocardial infarction in middle-aged men.5
In the municipality of Tromsø, which is located in northern Norway, 4 large health surveys have been performed, the first in 1974 and the last in 1994–1995. The Tromsø studies have primarily focused on cardiovascular risk factors and diseases, and in the last study, serum calcium was also analyzed. Because >27 000 persons were examined, a large database was therefore available to study relationships between serum calcium and risk factors for coronary heart disease.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
In 1994–1995, all men and women who were >24 years of age and who lived in the municipality of Tromsø were invited to participate in a health survey performed by the National Health Screening Service in cooperation with the University of Tromsø and local health authorities. Generally, this health survey was conducted in a manner similar to previous studies.6 A questionnaire that included a medical history and past and present medication was completed at home and returned by mail. All subjects gave their written informed consent to participate, and the study was approved by the regional ethics committee.
Measurements
The participants were not requested to fast. BP and heart
rate were measured as previously described in detail.6
Blood samples were drawn, and cholesterol, HDL
cholesterol, and triglycerides were measured as
previously described.6 The reference range for
cholesterol in our laboratory is 3.2 to 7.4 mmol/L for
those aged 25 to 29 years, 3.7 to 8.3 mmol/L for those aged 30 to
39 years, and 4.1 to 8.7 mmol/L for those aged >40 years. The
reference range for HDL cholesterol is 0.80 to 2.00
mmol/L for men and 0.95 to 2.20 mmol/L for women; the range for
triglycerides is 0.20 to 1.80 mmol/L. Serum calcium
was measured on Hitachi 917 equipment with reagents from
Boehringer Mannheim (reference range, 2.20 to 2.60
mmol/L).
Statistical Analyses
Differences between age and serum calcium groups were evaluated
with ANOVA. Linear trend was evaluated with linear regression. Linear
association was evaluated by Pearson correlation coefficients. A
sex-specific multiple linear regression model was used to control for
confounding factors and included age, calcium, body mass index (BMI),
cholesterol, HDL cholesterol,
triglycerides, systolic BP, diastolic
BP, and pulse as potential predictors. The regression coefficient is
given as the standardized ß coefficient (the coefficient of the
independent variables when all variables are expressed in
standardized [Z score] form). Comparison between
unadjusted calcium levels in disease states was performed with the use
of linear regression, with calcium as the dependent variable, age
as the covariable, and the disease in question as the fixed factor.
A sex-specific logistic regression model was used to study the
independent association between calcium and a self-reported history of
myocardial infarction, angina pectoris, and stroke, and it controlled
for the same variables as in the linear regression model. Unless
otherwise stated, the data presented are mean±SD. All tests
were done 2-sided, and P<0.05 was considered statistically
significant. Statistical analyses were performed with SPSS
version 8.0 (SPSS Inc).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
The total number of subjects examined was 12 865 men and 14 293 women, which were 74.2% and 79.0%, respectively, of the eligible population. The characteristics of the study population are given in Tables 1 and 2. The exclusion of patients with heart disease, stroke, or diabetes, pregnant women, and patients taking antihypertensive drugs had no effect on any of the results presented. Therefore, all subjects examined were included in the analyses.
|
|
There was a significant decrease in men and a significant increase in women in serum calcium with age (P<0.001) (Tables 1 and 2).
Serum Calcium and BP
There was a significant linear relation between serum calcium and
BP in both men and women (correlation coefficients of 0.059 and 0.206
for systolic BP in the 2 genders and of 0.062 and 0.198 for
diastolic BP, respectively; P<0.001) (Table 3). Because BP is strongly dependent on
age, an age-stratified analysis was performed. The relation was
found to be present within each age group and was similar in men
and women (Figures 1 and 2). The association was still highly
significant (P<0.001) after adjustment for all the
other variables included in the multiple linear regression model
(Tables 4 and 5).
|
|
|
|
|
Serum Calcium and Cholesterol
Cholesterol levels increased with increasing calcium
levels in both men and women (Pearson correlation coefficients of 0.149
and 0.265, respectively; P<0.001) (Table 3), which
was similar to that observed for BP. This significance remained after
adjustment for all the other variables included in the model
(Tables 4 and 5).
Serum Calcium and HDL Cholesterol
In both men and women, a significant (P<0.001)
association existed between serum calcium and HDL
cholesterol (Pearson correlation coefficients of 0.060 and
0.041, respectively) (Table 3), which remained after adjustment
for other variables (Tables 4 and 5). However, the
association appeared weaker than that between cholesterol
and calcium.
Serum Calcium and Triglycerides
In both men and women, there was a significant linear correlation
between serum calcium and triglycerides (correlation
coefficients of 0.075 and 0.115, respectively; P<0.001)
(Table 3). In men, this association remained significant after
adjustment for the other variables (Table 4), whereas in
women, this correlation was no longer significant after adjustment for
cholesterol (Table 5).
Serum Calcium and Cardiovascular Diseases
Within each age group, both men and women with a history of
myocardial infarction had higher serum calcium levels than those
without infarction (Table 6). When linear
regression with age as a covariant was used, this difference
was significant (P<0.0001) in men but not in women.
Similarly, with a logistic regression model that included the same
variables as in the multiple linear regression analysis,
serum calcium was statistically significant as an independent predictor
of myocardial infarction in men (estimated odds ratio=0.1503,
P<0.0001) but not in women (odds ratio=0.4812,
P=NS). For myocardial infarction in men, this resulted in an
odds ratio of 6.65 per unit (mmol/L) increase in serum calcium or an
odds ratio of 1.208 per 0.1 mmol/L increase in serum calcium.
|
Similar to that observed for myocardial infarction, the serum calcium levels in men with angina pectoris were higher within each age group (Table 6). This difference was statistically significant (P=0.024) in the age-adjusted linear regression model described above but not in the logistic regression model. In women, a similar but nonsignificant trend was observed (Table 6). No significant difference was found in serum calcium levels in regard to patients with a history of stroke.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In the present study, serum calcium was affected by age in both genders, with a gradual decrease with age in men and an increase in calcium at the time of menopause in women. This age-related pattern is similar to that described in other epidemiological studies.1 7
We found a highly significant association between serum calcium and both systolic and diastolic BP in both genders, an association that persisted after adjustment for all the other variables in the linear regression model. When the importance of calcium in all aspects of physiology is considered, together with the huge amount of research done on hypertension, there are remarkably few studies on the relation between total serum calcium levels and BP. However, in the few available studies, 1 2 8 the results are similar to those presented in this study.
Calcium exists in 3 major forms in plasma. Approximately 50% is in the free or ionized form, which is the physiologically important fraction, 40% is bound to albumin, and the remaining 10% is in soluble complexes with anions such as bicarbonate, phosphate, and lactate.9 Thus, total serum calcium will not only reflect calcium physiology but also be a function of the serum albumin level. Numerous algorithms have therefore been derived to correct for this effect,10 but unfortunately, serum albumin data are not included in our study. This could be of considerable importance because in a report by Hu et al,11 serum albumin also showed a consistent and strong association with systolic and diastolic BP.
In other epidemiological studies on calcium and BP, some investigators have corrected for serum albumin concentration. Thus, in the study by Kesteloot and Geboers1 of 9321 men in the Belgian army, serum calcium values were not corrected, whereas in the study from Sweden by Lind et al,2 which included >18 000 men and women, values were corrected and the association was still present and highly significant. On the other hand, in the British Regional Heart Study of 7735 healthy middle-aged men, Phillips and Shaper8 found that the small but significant association between serum calcium and BP diminished after they corrected for serum albumin and became nonsignificant after they included serum globulins and hematocrit in the model.
There are several reports that ionized serum calcium actually is lower in hypertensive patients than in normal controls.12 13 To reconcile this with our findings, because there is a strong correlation between total and ionized serum calcium, one would have to assume that the binding characteristics for calcium and its carrier proteins are abnormal in hypertension. This has been found in a study of 28 hypertensive subjects compared with normotensive controls,14 but until this finding is confirmed in other studies, it must be considered highly speculative.
To make the matter even more complicated, there seems to be a negative association between the intake of dietary calcium and BP,15 and the serum calcium level has been reported to increase with calcium intake.16 To settle these questions, large studies, which consider all of these variables, are needed.
In regards to serum calcium and lipids, there was a particularly strong association with cholesterol in both genders even after correction for age, BMI, and other possible risk factors. Thus, the standardized regression coefficient ß for calcium, with cholesterol as dependent variable, was second only to age and the other lipids in the model. Serum calcium was also found to be an independent, significant predictor of the HDL cholesterol level, but the standardized regression coefficient was only half that seen for total cholesterol. In contrast, the association between serum calcium and triglycerides was weaker and disappeared in women, but not in men, in the linear regression analysis.
To the best of our knowledge, the relation between serum calcium and triglycerides has not been previously reported, whereas similar results regarding cholesterol were found by Lind et al.2 As with hypertension, our results are confounded by the lack of correction for serum albumin, which has been reported to be positively correlated with serum cholesterol.17 However, the similarity of the results in our study with the study by Lind et al,2 in which they corrected for serum albumin, makes it unlikely that serum albumin had a major influence.
Our findings on BP and lipids are in agreement with findings in chronic hypercalcemia. Thus, hypertension18 and hyperlipidemia19 are common in patients with hyperparathyroidism, and these patients also have an increased mortality rate from cardiovascular diseases.4 Moreover, serum calcium levels are higher in non–insulin-dependent diabetes mellitus20 and are also positively associated with blood glucose in the general population.2 An association between serum calcium and the metabolic syndrome has therefore been suggested.2
If such an association exists, one would expect serum calcium to be related to cardiovascular mortality and morbidity. Indeed, this has been demonstrated prospectively for myocardial infarction in a study in which serum calcium appeared as an independent risk factor in middle-aged men followed for 18 years.5 Furthermore, in a study by Leifsson and Ahrén21 that followed 22 000 men for a mean period of 10 years, the risk of premature death in men <50 years of age increased when serum calcium levels rose, even if the increase was within the normal range. As in hyperparathyroidism, this increase was mainly due to cardiovascular diseases.
With our data, we can thus far only perform a retrospective analysis on serum calcium in relation to cardiovascular diseases. In the questionnaire completed by the patients, questions on myocardial infarction, angina pectoris, and stroke were included. Not surprisingly, serum calcium levels were higher in those with infarction and angina, but significantly so only for men with myocardial infarction when analyzed with logistic regression. In this respect, serum albumin is probably not a confounding factor, because serum albumin is found to be inversely related to cardiovascular diseases.22 However, a positive association is no proof of causative relation. In the case of calcium and hypertension, one should be especially careful because there is an apparent discrepancy between results obtained with ionized and total serum calcium levels.
In men, BP increased with age whereas calcium decreased with age. Accordingly, BP must be influenced by other factors that in this respect are more important than calcium. It must also be recognized that serum calcium could merely be a marker for other factors of greater importance for hypertension and cardiovascular diseases. Thus, in the study by Resnick et al,13 subjects with hypertension and high plasma renin activity also had elevated serum ionized calcium levels, and Alderman et al23 reported that a 2 unit increase in plasma renin activity increased the overall incidence of myocardial infarction by 25%. Furthermore, the calcium-regulating hormones calcitonin, calcitriol, and parathyroid hormone have been related to hypertension, partly in association with plasma renin activity.24 Finally, there may be interactions between serum calcium, phosphate, and magnesium,13 25 26 as well as between calcium and sodium.
Our results are average values of large groups of individuals, and effects in subgroups may not be seen, as shown in the study by Hunt et al.27 Thus, in their study on 875 subjects, no correlation between plasma ionized calcium and BP was found. However, after the older subjects were divided into tertiles on the basis of renin activity, there was a significant inverse correlation between ionized calcium and BP in the low renin group and a significant positive relation with systolic BP in the high renin group.
In addition to relations between serum calcium and cardiovascular risk factors, certain other observations should be noted. Despite a highly significant correlation between systolic and diastolic BP (correlation coefficient in men of 0.681), in men there was a negative association between cholesterol and systolic BP and a positive correlation between cholesterol and diastolic BP. For systolic BP, this is contrary to that reported by others.28 29 However, as shown in Table 1, there was an increase in systolic BP up to the age of 90 years, whereas diastolic BP and serum cholesterol peaked in those in their 60s and thereafter showed a decline. Because the age of our study population was high, this age-related pattern may explain the discrepancy with previous reports.
Furthermore, there was a significant correlation between pulse rate and BP in both genders. Because a highly significant relation between heart rate and the calcium/phosphate ratio has been described,26 this could have an effect on the relation between calcium and BP. However, when pulse rate was excluded from the multiple linear regression analysis, this increased the corresponding standardized ß coefficient in both genders by only 7%.
In conclusion, total serum calcium is strongly related to BP and serum lipids. This relation might be explained in part by covariation with serum albumin and may also be due to the association between serum calcium and other substances related to hypertension and heart disease. Regardless of this, however, total serum calcium appears to be a marker of cardiovascular disease, at least in men.
|
Acknowledgments |
|---|
This study was supported by a grant from the Norwegian Research Council.
Received February 2, 1999; first decision March 1, 1999; accepted May 10, 1999.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Kesteloot H, Geboers J. Calcium and blood pressure. Lancet. 1982;1:813–815.[Medline] [Order article via Infotrieve]
2. Lind L, Jakobsson S, Lithell H, Wengle B, Ljunghall S. Relation of serum calcium concentration to metabolic risk factors for cardiovascular disease. BMJ. 1988;297:960–963.
3. Rosenthal FD, Roy S. Hypertension and hyperparathyroidism. BMJ. 1972;4:396–397.
4. Palmér M, Adami H-O, Bergström R, Jakobsson S, Åkerström G, Ljunghall S. Survival and renal function in untreated hypercalcaemia: population-based cohort study with 14 years of follow-up. Lancet. 1987;1:59–62.[Medline] [Order article via Infotrieve]
5. Lind L, Skarfors E, Berglund L, Lithell H, Ljunghall S. Serum calcium: a new, independent, prospective risk factor for myocardial infarction in middle-aged men followed for 18 years. J Clin Epidemiol. 1997;50:967–973.[Medline] [Order article via Infotrieve]
6.
Bønaa KH, Arnesen E. Association between heart rate
and atherogenic blood lipid fractions in a population: the Tromsø
Study. Circulation. 1992;86:394–405.
7. Christensson T, Hellström K, Wengle B, Alveryd A, Wikland B. Prevalence of hypercalcemia in a health screening in Stockholm. Acta Med Scand. 1976;200:131–137.[Medline] [Order article via Infotrieve]
8. Phillips AN, Shaper AG. Serum calcium and blood pressure. J Hum Hypertens. 1991;5:479–484.[Medline] [Order article via Infotrieve]
9. Moore EW. Ionized calcium in normal serum, ultrafiltrates and whole blood determined by ion-exchange calcium electrodes. J Clin Invest. 1970;49:318–334.
10.
Ladenson JH, Lewis JW, Boyd JC. Failure of total
calcium corrected for protein, albumin, and pH to correctly
assess free calcium status. J Clin Endocrinol Metab. 1978;46:986–993.
11.
Hu H, Sparrow D, Weiss S. Association of serum
albumin with blood pressure in the normative aging study.
Am J Epidemiol. 1992;136:1465–1473.
12. McCarron DA. Low serum concentrations of ionized calcium in patients with hypertension. N Engl J Med. 1982;307:226–228.[Medline] [Order article via Infotrieve]
13. Resnick LM, Laragh JH, Sealey JE, Alderman MH. Divalent cations in essential hypertension: relations between serum ionized calcium, magnesium, and plasma renin activity. N Engl J Med. 1983;309:888–891.[Abstract]
14.
Folsom AR, Smith CL, Prineas RJ, Grimm RH. Serum
calcium fractions in essential hypertensive and matched normotensive
subjects. Hypertension. 1986;8:11–15.
15. Osborne CG, McTyre RB, Dudek J, Roche KE, Scheuplein R, Silverstein B, Weinberg MS, Salkeld AA. Evidence for the relationship of calcium to blood pressure. Nutr Rev. 1996;54:365–381.[Medline] [Order article via Infotrieve]
16.
Kesteloot H, Joossens JV. Relationship of dietary
sodium, potassium, calcium, and magnesium with blood pressure: Belgian
interuniversity research on nutrition and health.
Hypertension. 1988;12:594–599.
17. Gillum RF. The association between serum albumin and HDL and total cholesterol. J Nat Med Assoc. 1993;85:290–292.[Medline] [Order article via Infotrieve]
18. Lind L, Wengle B, Wide L, Sörensen OH, Ljunghall S. Hypertension in primary hyperparathyroidism—reduction of blood pressure by long-term treatment with vitamin D (alphacalcidol): a double blind, placebo-controlled study. Am J Hypertens. 1988;1:397–402.[Medline] [Order article via Infotrieve]
19. Ljunghall S, Lithell H, Vessby B, Wide L. Glucose and lipoprotein metabolism in primary hyperparathyroidism: effects of parathyroidectomy. Acta Endocrinol. 1978;89:580–589.
20. Levy J, Stern Z, Gutman A, Naparstek Y, Gavin JR III, Avioli LV. Plasma calcium and phosphate levels in an adult noninsulin-dependent population. Calcif Tissue Int. 1986;39:316–318.[Medline] [Order article via Infotrieve]
21. Leifsson BG, Ahrén B. Serum calcium and survival in a large health screening program. J Clin Endocrinol Metab. 1996;81:2149–2153.[Abstract]
22.
Kuller LH, Eichner JE, Orchard TJ, Grandits GA,
McCallum L, Tracy RP. The relation between serum albumin levels
and risk of coronary heart disease in the multiple risk factor
intervention trial. Am J Epidemiol. 1991;134:1266–1277.
23. Alderman MH, Ooi WL, Cohen H, Madhavan S, Sealey JE, Laragh JH. Plasma renin activity: a risk factor for myocardial infarction in hypertensive patients. Am J Hypertens. 1997;10:1–8.[Medline] [Order article via Infotrieve]
24. Resnick LM, Muller FB, Laragh JH. Calcium-regulating hormones in essential hypertension: relation to plasma renin activity and sodium metabolism. Ann Intern Med. 1986;105:649–654.
25. Hvarfner A, Ljunghall S, Mörlin C, Wide L, Bergström R. Indices of mineral metabolism in relation to blood pressure in a sample of a healthy population. Acta Med Scand. 1986;219:461–468.[Medline] [Order article via Infotrieve]
26.
Kesteloot H, Joossens JV. Relationship of serum sodium,
potassium, calcium, and phosphorus with blood pressure: Belgian
interuniversity research on nutrition and health.
Hypertension. 1988;12:589–593.
27. Hunt SC, Williams RR, Kuida H. Different plasma ionized calcium correlations with blood pressure in high and low renin normotensive adults in Utah. Am J Hypertens. 1991;4:1–8.[Medline] [Order article via Infotrieve]
28.
Kesteloot H, van Houte O. An epidemiological survey of
arterial blood pressure in a large male population group.
Am J Epidemiol. 1974;99:14–29.
29.
Bønaa KH, Thelle DS. Association between blood
pressure and serum lipids in a population: the Tromsø Study.
Circulation. 1991;83:1305–1314.
This article has been cited by other articles:
 |
|
 |
|
  E. Hagstrom, E. Lundgren, J. Rastad, and P. Hellman Metabolic abnormalities in patients with normocalcemic hyperparathyroidism detected at a population-based screening. Eur. J. Endocrinol., July 1, 2006; 155(1): 33 - 39. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||