Oral calcium and blood pressure: a controlled intervention,2

Transcription

1 Oral calcium and blood : a controlled intervention,2 ECH van Beresteyn, c, G Schaafsma, PhD, and H de Waard, PhD Introduction ABSTRACT In a double-blind, placebo-controlled trial with 58 normotensive female students, the effect of oral-calcium supplementation (1500 mg Ca4/day for 6 wk) on diastolic and systolic blood was studied while students were consuming a low-calcium diet (500 mg Ca/day) by restricting the intake ofdairy products. Results show that, in both the calcium- and placebo-supplemented groups, blood values decreased slightly and no effect of oral-calcium supplementation on blood could be demonstrated. In addition, at baseline neither systolic nordiastolic blood correlated with habitual calcium intake. Diastolic but not systolic blood correlated significantly with body mass index (r=0.3l,p= 0.01). It is concluded that oral-calcium supplementation for 6 wk does not influence blood in young, healthy normotensive females consuming low-calcium diet. Am J C/in Nutr 1986;44: KEY WORDS Calcium, blood, normotensive women In recent years, interest has been shown in the effect ofdietary calcium on blood. Epidemiological studies have suggested an inverse relationship between dietary calcium and blood (1-5). A significant higher blood level in calcium-deprived rats has been described (6, 7) and spontaneously hypertensive rats showed a reduced increase in blood values when the calcium content of their diet was raised (8, 9). Low serum concentrations of ionized calcium were found in patients with hypertension (10); on the other hand, a positive correlation between serum-calcium concentration and blood was found in an epidemiological survey (1 1). The few intervention studies described in the literature (12, 13) indicate a blood lowering effect of calcium supplementation (1000 mg Ca/day, 3-9 wk), but a short-term study (1500 mg Ca/day, 3 wk) carried out in our Institute did not show any effect in young, healthy normotensive males (private observation). Because of some contradictions in the results of observational studies and unconvincing evidence from clinical trials, the reported hypothesis that dietary calcium influences blood needs further investigation. This article describes a controlled, doubleblind, placebo-controlled intervention trial made to investigate the effect on diastolic and systolic blood of oral-calcium supplementation (1500 mg Ca/day) in young, healthy normotensive females on a lowcalcium diet (500 mg Ca/day). Materials and methods Subjects and experimenza/ protoco/ The 58 normotensive subjects examined were students of the Training College for Dietitians. They were yr ofage, kg in body weight, and not receiving any medical treatment at the time of recruitment. The experiment was carried out over 7 wk. Baseline data on blood, body weight, and urinary excretion were collected during the first week ofthe experiment while subjects were on their usual diet. The habitual intake ofdairy products The 1 rom the Netherlands Institute for Dairy Research, Netherlands. 2 Address reprint requests to: ECH van Beresteyn, Netherlands Institute for Dairy Research, P0 Box 20, 6710 BA EDE, The Netherlands. Received December 19, Accepted for publication March 18, Am J C/in Nuir l986;44: Printed in USA American Society for Clinical Nutrition 883

2 884 VAN BERESTEYN ET AL was estimated by means of a specially developed questionnaire (CIVO/TNO Department for Toxicology and Nutrition, Zeist, The Netherlands). Calcium intake from dairy products was calculated using a computerized foodcomposition table (14) and amounted to about 75% of total calcium intake. Subjects were assigned to two groups according to a randomized block design that accounted for habitual calcium intake and body mass index, each ofthese variables at two levels (15). Median values ofbody mass index and habitual calcium intake served as classification criteria. Subjects were given detailed instructions on how to reduce their calcium intake to 500 mg/day during the course of the experimental period (6 wk). The calcium-supplemented group received a powder containing 500 mg Ca three times a day. Each portion ofthe powder comprised the following: calcium carbonate ( g), citric acid ( g), sodium-hydrogen carbonate (0.5 g), and dextrose (2.88 g). The other group was given a placebo. Each portion ofthe placebo comprised the following: citric acid (0.85 g), sodium-hydrogen carbonate (0.5 g), dextrose (4.5 g), and comflour(0. 1 g). The powders were consumed with lemonade or apple juice. Blood s were taken five times at each session, after a standardized rest period of5 mmn in dorsal position, at the right arm using an automatic device (Copal digital sphygmomanometer, model UA 231, Adquipment Medical, Rotterdam, The Netherlands). After each session, the mean ofthe five repeated measurements was taken. Blood s were measured twice during the baseline period and in the 6th wk ofthe experiment, once a week during the intermediate period (except in week 5 when students were on holiday). or each subject, the blood measurements obtained during the baseline period and during the 6th wk were averaged and each mean was considered as the initial and final value, respectively. Body height was measured. Body weight was recorded once a week and the energy intake was adjusted as required to prevent changes in body weight. Body mass index (weight/height2, kgjm2) was calculated. To control compliance to the calcium-restricted diet and calcium supplementation and to obtain information about the intake of sodium, potassium, and magnesium, each subject collected 24-h urine samples, one during the baseline period and two during the experimental period. Urine samples were analyzed for total volume and concentrations of creatinine, calcium, phosphorus, sodium, potassium, and magnesium by standard laboratory meth- TABLE 1 Mean (±SD) baseline and final values for placebo and calcium groups ods. Urinary excretion of each variable was expressed as its ratio to creatinine. Statistics In a pilot study, the standard deviation of paired differences in blood (measured as indicated above) was estimated in six healthy young adults. This standard deviation was 7 mmhg for both diastolic and systolic. rom this value it was computed that, with 29 participants in both experimental and placebo groups, a treatment effect of3 mmhg could be detected with a type- I error (a) of 5% and a statistical power of 50%. The statistical significance ofdifferences between groups was assessed by Student s t test(two-sided) for unpaired data. A paired t test was used to assess changes within groups. The effect oforal calcium versus placebo on systolic and diastolic blood pressure was evaluated by two-way unbalanced analysis ofvariance using a randomized block design (15), as indicated above. Two variables were subjected to this analysis: 1) the difference for each individual between baseline blood and final blood and 2) the individual change in blood during the experiment as mdicated by the regression coefficient (slope) obtained from linear regression analysis of blood versus time during the experimental period. Correlation coefficients were calculated between baseline blood and habitual calcium intake with partial correlation coefficients computed to control for differences in body mass index and intake ofsodium, potassium, and magnesium. Results Information on basal values Basal information by treatment group is given in Table 1 and igure 1 No differences between groups were found in the variables collected during the baseline period except for the urinary excretion of calcium, which was significantly higher for the calcium group (p < 0.05). Placebo gro up (n = 29) Calcium group (n = 29) Basal values inal values Basal values inal valucs Age (yr) Body weight (kg) Body mass (kg/m2) Systolic blood (mmhg) Diastolic blood (mmhg) Calcium intake (mg/day) S rom dairy products only (range, mg/day). I SD I SD I SD I SD ± ±

3 ORAL CALCIUM AND BLOOD PRESSURE 885 Ca/creat. (mglg) P/creat. (mglg) * weeks IG la. Urinary excretion ofcalcium and phosphorus expressed as their ratios to creatinine (mean ± SEM) versus time. A, calcium group; t, placebo group. Significance level between groups:, p < 0.05;, p < Body weight Table 1 shows that final body weight did not differ significantly between groups. During the study, no significant weight changes were observed. Blood effects Tables 2 and 3 show the analysis of variance for both variables tested: the individual differences between baseline blood and final blood and the individual regression coefficients. igure 2 shows blood changes of the placebo and calciumsupplemented groups. No effect of calcium treatment on either systolic or diastolic blood could be found. or both groups, blood values decreased slightly during the experimental period as compared to baseline values. Nalcreat. (mglg) Mglcreat (mglg) Kicreat (mglg) The relationship between habitual calcium intake from dairy products and blood can be expressed as a correlation coefficient. The magnitude of this correlation was for systolic blood and 0.03 for diastolic blood. Neither of these correlations was significant. Blood values were not significantly different between a subgroup with a low dairy intake (< 600 mg calcium/day, n = 1 1) and a subgroup with a high dairy intake (> 1000 mg calcium/day, n = 24); they were 1 12/64 and 1 14/65 mmhg, respectively. No other correlations were found except that diastolic blood appeared to correlate with body mass index (r = 0.31, p = 0.01). This relation persisted after taking calcium intake and excretion of sodium, potassium, and magnesium into account in multiple linear-regression analyses weeks IG lb. Urinary excretion of sodium, magnesium, and potassium expressed as their ratios to creatinine (mean ± SEM) versus time. A, calcium group; t, placebo group.

4 886 VAN BERESTEYN ET AL TABLE 2 Analysis of variance for individual differences between baseline and final blood TABLE 3 Analysis of variance for individual regression coefficients of blood versus time Systolic blood Diastolic blood Syst olic blood Diast olic blood Source of variation df Total 57-4 Supplementation (t) (t) Replications (t) (t) Samplingerror S Mean squares. t Not significant, p > Urine variables igure 1 shows the urinary excretion of calcium, phosphorus, sodium, magnesium, and potassium, expressed as their ratio to creatinine during the experimental period. Compared to baseline values, calcium cxcretion decreased in the placebo group (p < 0.05) and increased slightly in the calcium group (not significant). The differences between the two groups were significant (p < 0.01). Phosphorus excretion decreased in the calcium group (p < 0.01) but decreased less in the placebo group (not significant). The two groups differed significantly (p < 0.01). There were no significant differences in cxcretion of sodium, potassium, or magnesium between the two groups. Compared to baseline values, potassium excretion decreased, but not significantly, in both groups. S change from basal values systolic I Sourceofvsriation df diastolic Total 57 Supplementation (t) (t) Replications (t) (t) Sampling error C Mean squares. t Not significant, p > Discussion The mean calcium intake from dairy products only (75% of total calcium intake) amounted to 950 mg/day, well above the recommended daily allowance of800 mg in both The Netherlands and the USA (16, 17). Results of this study showed that supplementing the diet with oral calcium for 6 wk did not affect either the systolic or the diastolic blood ofnormotensive, healthy young females. In addition, restriction ofdietary calcium in the placebo group did not result in a rise of blood. The difference in calcium intake between the two groups (1 500 mg Ca/day) was greater than that used in other intervention studies [1000 mg Ca/day, (12, 13)]. Moreover, the calcium intake in the calcium-supplemented group was 1000 mg/ day higher than the habitual calcium intake I weeks IG 2. Mean percentage changes of systolic and diastolic blood from basal values versus time. A, calcium group;, placebo group.

5 ORAL CALCIUM AND BLOOD PRESSURE 887 Similar results were obtained in a similar cxperiment in young healthy men (mean age, 33 yr; mean blood, 1 12/68 mmhg; duration, 20 days) performed earlier in our Institute (private observation). or both groups, the first blood measurement during the baseline period was slightly higher than the second measurement, probably due to psychological factors. This might explain the slight fall in blood in both groups after the first week. We have no explanation for the decrease in diastolic blood in both the placebo and cxperimental groups after 3 wk. Probably this should be considered as a time-of-measurement effect. Also the difference in basal values of urinary-calcium excretion between the two groups cannot be explained. Compliance with the dietetic instructions and calcium supplementation appeared to be good, according to the urinary excretion of calcium and phosphorus. Dairy products are rich in calcium and phosphorus. Restriction oftheir intake reduces calcium and phosphorus excretion in the urine. Calcium supplementation prevents the decrease in urinary-calcium excretion, but not the decrease in urinary-phosphorus excretion. At a given change of dietary calcium, urinary excretion of calcium will change by approximately 6% ofthe dietary change (1 8, 19). The changes in urinary excretion of calcium found in this study agree with this finding, although the increase for the calcium group was somewhat less than anticipated because fractional calcium absorption decreases as calcium intake increases (18). It is known from studies in animals (20) and in man (2 1) that decreasing dietary-calcium intake leads to an increase in urinaryphosphorus excretion. This explains why phosphorus excretion decreased less in the placebo group than in the calcium group. Urinary excretion ofsodium and potassium reflects their intake very well because intestinal absorption ofthese minerals is high. Inasmuch as the results do not show any differences in body weight and urinary excretion of sodium, potassium, or magnesium, these factors, known to influence blood, cannot be considered as confounding factors in this study. It is concluded that our results do not show any effect of oral calcium on blood over a 6-wk period. The question may arise whether an experimental period of6 wk is long enough to show the possible effect of calcium on blood. It has been demonstrated (22) that, within 4 days, a reduction in calcium intake of about 1000 mg a day caused a fall in plasma- and urinary-calcium values and a rise in plasma- PTH, plasma-1,25 dihydroxy-cholecalciferol, and urinary-camp values in young healthy volunteers. Thus, it might be expected that a 6-wk period of low calcium intake should be long enough to produce a rise in blood if adaptation to a low calcium intake is associated with this effect. In the study of Belizan et al (12), 9 wk of 1000 mg oral-calcium supplementation were required before a significant effect on blood was obtained in a group of young, healthy, normotensive females, whereas only 3 wk were required to produce such an effect in their male counterparts. This difference may indicate that females are less sensitive than males to a blood -lowering effect of oral calcium. To shorten the period needed to observe an effect of oral calcium on blood, our study design was sensitized by subjecting the placebo and experimental groups to a low dietary-calcium intake (< 500 mg/day) and by supplementing the experimental group with not less than mg Ca/day. In spite of the big difference in calcium intake between the two groups, no effect on blood could be observed. This result does not exclude completely the possibility that more than 6 wk are needed to observe an effect of calcium on blood in young, healthy, normotensive females, even with the present study design. We do not feel this is likely, even though results of some observational studies ( 1-5) may suggest that a long-term, lowcalcium status is required before blood increases, in which case a long-term study would be required to establish any effect of calcium deprivation on blood. If that were the case, we might have expected a negative correlation between habitual calcium intake and basal blood in our study population. This correlation was not found.

6 888 VAN BERESTEYN ET AL Nor were differences in blood found between groups who were at the extremes of calcium-intake range (intake < 600 mg Caj day or > 1000 mg Ca/thy). The results ofthis study do not exclude any effect in persons with hypertension. This aspect needs further investigation. I] We thank T Lossonszy, M Dijkmeier, T Dijkstra, and the second-year students ofthe Training College for Dietitians, The Hague. We are indebted to Rob Dekker, who assisted in blood measurements; Joke van der Heiden, who estimated habitual intake ofdairy products; Romana Visser, who made urine measurements; and Rinus van Schaik, who did statistical analysis. References 1. Langford HG, Watson RL, Douglas BH. actors affeeling blood in population groups. Trans Assoc Am Physicians l968;81:l Ackley 5, Barret-Conner E, Suarez L. Dairy products, calcium, and blood. Am J Clin Nutr l983;38: McCarron DA, Morris CD, Cole C. Dietary calcium in human hypertension. Science l982;217: Garcia-Palmieri MR, Costas R, Cruz-Vidal M, Sorlie PD, Tillotson J, Havlik J. Milk consumption, calcium intake, and decreased hypertension in Puerto Rico: Puerto Rico heart health program study. Hypertension 1984;6: Kromhout D, Bosschieter EB, Coulander CdL. Potassium, calcium, alcohol intake and blood : the Zutphen study. Am J Clin Nutr 1985;41 : McCarron DA. Blood and calcium balance in the Wistar-Kyoto rat. Life Sci 1982;30: Belizan JM, Pineda 0, Sainz E, Menendez LA, Villar J. Rise of blood in calcium-deprived rats. Am J Obstet Gynecol 198 l;141: Ayachi S. Increased dietary calcium lowers blood in the spontaneously hypertensive rat. Metabolism l979;28: Anderson 5, Grady JR, Ellison DE, McCarron DA. Calcium balance and parathyroid hormone-mediated vasodilation in the spontaneously hypertensive rat. Hypertension 1983;5(Suppl I): McCarron DA. Low serum concentrations of ionized calcium in patients with hypertension. N Engl J Med l982;307: Kesteloot H, Gebroers J. Calcium and blood. Lancet 1982;i: Belizan JM, Villar J, Pineda 0, et al. Reduction of blood with calcium supplementation in young adults JAMA l983;249(9):l Resnick LM, Laragh JH. The hypotensive effect of short-term oral calcium loading in essential hypertension. Clin Res l983;3l:334(abstract). 14. Commissie UCV s Gravenhage: Voorlichtingsbureau vcor de Voeding UCV-tabel (Dutch computerized food composition table) Snedecor OW, Cochran WG. Statistical methods Iowa State University Press, Ames, IA, USA Nederlands Voedingsmiddelen Tabel, 34* gewijzigde druk. Vcorlichtingsbureau voor de Voeding. s-graven-hage, 1983: Recommended Dietary Allowances, 9th ed. National Academy of Sciences, Washington, DC, Heany RP, Saville PD, Reeker RR. Calcium absorption as a function ofcalcium intake. J Lab Gin Med 1975;88: Lemann J, Adams ND, Gray RW. Urinaiy calcium excretion in human beings. N EngI J Med 1979; 301(l0): Gark I, Rivera-Cordero. Effects of endogenous parathyroid hormone on calcium, magnesium, and phosphate metabolism in rats. Endocrinology 1973;92: Spencer H, Kramer L, Osis D. Effect of calcium on phosphorus metabolism in man. Am J Clin Nutr l984;40: Adams ND, Gray RW, Lemann J. The effects of oral CaCO3 loading and dietary calcium deprivation on plasma l,25-dihydroxy-vitamin D concentrations in healthy adults. J Clin Endocrinol Metab 1979;48(6):