点击显示 收起
1 From the Departments of Nutrition (WKA-D, ER, WCW, and FBH) and Epidemiology (ER, WCW, MJS, and FBH), Harvard School of Public Health, Channing Laboratory (ER, WCW, and MJS), Department of Medicine, Harvard Medical School and Brigham and Women’s Hospital, Boston.
2 Supported by research grants HL24074, HL34594, DK36798, and CA87969 from the National Institutes of Health. 3 Reprints not available. Address correspondence to WK Al-Delaimy, Department of Nutrition, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115. E-mail: wael{at}hsph.harvard.edu.
ABSTRACT
Background: Calcium intake is thought to have a protective effect on the risk of developing ischemic heart disease (IHD), but the existing data are inconsistent.
Objective: The objective was to assess the relation between calcium intake and risk of IHD among men.
Design: Men in the Health Professionals Follow-up Study who returned a dietary questionnaire in 1986 (n = 39 800) were followed up for 12 y. Intakes of calcium and other nutrients were assessed in 1986, 1990, and 1994. The endpoints of total IHD (nonfatal myocardial infarction and fatal IHD) incidence were ascertained by medical record review. Other IHD risk factors were recorded biennially.
Results: During 12 y of follow-up (415 965 person-years), we documented 1458 cases of IHD: 1030 of nonfatal myocardial infarction and 428 of fatal IHD. After control for standard IHD risk factors, the relative risk of developing IHD among men in the highest (median intake = 1377 mg/d) compared with the lowest (median intake = 523 mg/d) calcium intake quintile was 0.97 (95% CI: 0.81, 1.16; P for trend = 0.64), for vitamin D intake was 1.00 (95% CI: 0.80, 1.24; P for trend = 0.66), and for total dairy product intake was 1.01 (95% CI: 0.83, 1.23; P for trend = 0.57). For supplemental calcium intake, the relative risk of developing IHD in a comparison of the highest quintile with nonusers of supplements was 0.87 (95% CI: 0.64, 1.19; P for trend = 0.31).
Conclusion: The results suggest that neither dietary nor supplemental intakes of calcium are appreciably associated with the risk of IHD among men.
Key Words: Calcium • ischemic heart disease • diet • supplements • Health Professionals Follow-up Study • men
INTRODUCTION
Increased calcium intakes have been shown to reduce blood pressure, although weakly, in prospective studies (1) and randomized trials (2, 3), and hypertension is a known risk factor for ischemic heart disease (IHD) (4). Therefore, one might suspect that a high calcium intake would also protect against IHD. However, the literature is not consistent regarding the relation between calcium intake and IHD risk (5–9). A follow-up study of 2605 persons aged > 28 y found no relation between calcium intake and IHD mortality (5), whereas another 8-y follow-up study of 43 486 women found a slight apparent protective effect of calcium intake against IHD mortality (6). Vitamin D is closely related to calcium metabolism, and dairy products are major sources of calcium and vitamin D. However, dairy products may also be related to an increased risk of IHD as a result of their high saturated fatty acid contents (10), and very high vitamin D intakes in addition to excess calcium intakes may increase the risk of atherosclerosis according to animal studies (9).
The aim of this analysis was to investigate the association between intakes of calcium, vitamin D, and dairy products and the risk of IHD [fatal IHD and nonfatal myocardial infarction (MI)] among the men participating in the Health Professionals Follow-up Study.
SUBJECTS AND METHODS
Subjects
The Health Professionals Follow-up Study is a prospective cohort initiated in 1986, when 51 529 predominantly white men aged 40–75 y answered a detailed questionnaire by mail on diet and medical history. This cohort consists of dentists (57.6%), veterinarians (19.6%), pharmacists (8.1%), optometrists (7.3%), osteopathic physicians (4.3%), and podiatrists (3.1%). Every 2 y, follow-up questionnaires were mailed to all surviving cohort members, up to 6 times per follow-up cycle for nonrespondents, to update data on medical conditions and exposures.
To form the cohort for the analysis, we excluded men with implausibly high or low scores for total food intake (outside the range of 800–4200 kcal/d) or with 70 items left blank on the 1986 baseline dietary questionnaire (11). In addition, men with cancers (excluding nonmelanoma skin cancer) diagnosed at baseline or before the development of IHD (during follow-up) were excluded because these men may have changed their diets as a result of their cancer. In addition, those with myocardial infarction (MI) or other cardiovascular diseases at baseline were excluded. A total of 11 729 men were excluded at baseline, and the remaining 39 800 men were eligible for follow-up. The follow-up rate for this cohort averaged 94% per follow-up cycle during the 5 biennial cycles from 1986 to 1996. The National Death Index was used to determine vital status for nonrespondents, and the remaining nonrespondents were assumed to be alive and at risk of IHD. The study was approved by the Brigham and Women’s Hospital Human Subjects Committee.
Dietary intake
To assess dietary intake, we used a 131-item semiquantitative food-frequency questionnaire (11) that is an expanded version of a previously validated questionnaire (12). The baseline dietary questionnaire was administered in 1986, and information on dietary intake was updated in 1990 and 1994. The questionnaire assesses the average frequency of intake over the previous year. For each man, we calculated energy and nutrient intakes by multiplying the frequency that each food item was reported by the energy or nutrient content for the specified portion size. We asked about the use of specific supplements of calcium and vitamin D in addition to multivitamin supplements. Total calcium and vitamin D intakes were calculated as dietary and all supplemental intakes. To determine dairy calcium intake, we summed the calcium intake from the following items on the food-frequency questionnaire: whole milk, skim or low-fat milk, yogurt, ice cream, cottage cheese, and other cheese.
The food-composition database used to calculate the nutrient values is based primarily on US Department of Agriculture publications (13) supplemented with other published data in the literature and manufacturers’ data.
The validity of the food-frequency questionnaire was evaluated in a random sample of 127 men from the Health Professionals Follow-up Study living in the Boston area. In that study, nutrient intakes computed from the questionnaire were compared with nutrient intakes from two 1-wk diet records spaced 6 mo apart (11). A correlation coefficient of 0.64 between questionnaires and diet records was observed for total calcium intake.
Ascertainment of endpoints
On each questionnaire, the participants indicated whether they had been diagnosed with any major cancer (eg, prostate or colon cancer), heart disease, or other medical conditions.
As described elsewhere in detail (14), the endpoints were fatal IHD (including sudden death) and nonfatal MI. In the present study, we included events that occurred between the return of the 1986 questionnaire and 31 January 1998. From those participants who reported an incident MI on a follow-up questionnaire, permission was requested to review their medical records. Nonfatal MI was confirmed with the use of World Health Organization criteria (15): symptoms plus either elevated cardiac enzyme concentrations or typical electrocardiographic changes.
Deaths were reported by next of kin, coworkers, or postal authorities or in the National Death Index (16). Fatal IHD was confirmed by medical records, autopsy reports, or the death certificate if IHD was the underlying cause and a diagnosis of IHD was confirmed by other sources. Deaths due to sudden death, within 1 h of the onset of symptoms in men with no other apparent cause of death (other than IHD), were also included.
Statistical analysis
We computed person-time of follow-up for each participant from the return date of the 1986 questionnaire to the date of the IHD diagnosis, to the day of death from any cause, or 31 January 1998, whichever came first. In the main analysis, exposure categories were updated every 2 y in all analyses. The incidence rate for each category of calcium, vitamin D, and dairy product intakes was calculated as the number of cases with IHD divided by the person-time of follow-up. These nutrients were all energy-adjusted (17). Cutoffs for the different groupings of calcium, vitamin D, and dairy product intakes were obtained by dividing each into quintiles. To adjust for age (5-y categories) and other covariates, we used pooled logistic regression (18). This approach is asymptotically equivalent to the Cox regression model with time-dependent covariates given short time intervals and low probability of the outcome within the interval. Total energy intake was also included in multivariate models to minimize extraneous variation introduced by underreporting or overreporting on the food-frequency questionnaire. In multivariate analyses, in addition to age, we included time period (2-y intervals), smoking history [never smoker, past smoker, or current smoker (1–14, 15–24, or 25 cigarettes/d)], alcohol consumption (0, 1–4.9, 5–29, and 30 g/d), history of diabetes, history of hypercholesterolemia (physician diagnosed or history of medication), parental history of MI before the age of 65 y, body mass index (calculated as weight in kilograms divided by the square of height in meters and included as an updated variable in the analyses in the following categories: < 21, 21–22.9, 23–24.9, 25–26.9, 27–28.9, 29–31, and > 31), aspirin intake (yes or no), vitamin E intake quintiles, and total energy intake quintiles. Physical activity was measured by the time per week engaged in 10 specified physical activities and in 4 sedentary activities during the previous year (19). Using these activities, we calculated a weekly metabolic equivalent task score for total physical activities. The validity of the questionnaire in assessing physical activity was described elsewhere (19).
We conducted further analyses to adjust for quintiles of the ratio of dietary polyunsaturated to saturated fatty acids (P:S), trans fatty acids, protein, n-3 fatty acid, -linolenic acid, folate, and cereal fiber.
We examined the intakes of calcium, vitamin D, and dairy products in relation to the incidence of IHD by updating the baseline dietary data with information from subsequent questionnaires (in 1990 and 1994). In these analyses, dietary data from the 1986 questionnaire were used to predict outcomes during the period from 1986 to 1990. The average of dietary intakes in 1986 and 1990 was used to predict outcomes during the period from 1990 to 1994, and the average of intakes in 1986, 1990, and 1994 was used for subsequent cases (ie, 1994–1998). Cumulative averaging reduces within-person variation and thus can better represent long-term intake (20).
Mantel extension tests for trend (21) were obtained by assigning the median value for each category and modeling this variable as a continuous variable with the use of pooled logistic regression for multivariate analyses at 2-y intervals. All P values are two-sided. The association between calcium intake and total IHD was stratified by vitamin D intake, because it has been shown that vitamin D intake can modify this association (9). The -2 log-likelihood test for interaction was used to assess effect modification by vitamin D of the association between calcium and IHD.
RESULTS
During 415 965 person-years of follow-up of 39 800 men over 12 y (1986–1998), we documented 1458 cases of total IHD (1030 nonfatal MI cases and 428 fatal IHD cases). The characteristics of the study population according to calcium and vitamin D intakes in 1986 are shown in Table 1. Men in the highest quintile of calcium intake had higher vitamin E intakes than did those in the lower quintiles, but they drank less alcohol. Similar trends were seen for vitamin D intake. Those with higher intakes of calcium and vitamin D consumed less trans fatty acids and more animal protein, were less likely to smoke, and were physically more active. All trends of variables across calcium and vitamin D quintiles were significant because of the large sample size.
View this table:
TABLE 1 . Characteristics of men according to energy-adjusted calcium and vitamin D quintiles 1, 3, and 5 in 19861
The association between calcium intake and total IHD is shown in Table 2. In age-adjusted analyses, no significant trend in the risk of IHD was seen with increasing quintiles of total calcium intake. The relative risk (RR) of developing IHD for the highest (median = 1377 mg/d) compared with the lowest (median = 523 mg/d) quintile of calcium intake was 0.90 (95% CI: 0.76, 1.06; P for trend = 0.17). A similar lack of association was seen in the multivariate analyses when other IHD risk factors (diabetes history, high cholesterol history, smoking history, family history of MI, aspirin intake, body mass index, alcohol intake, physical activity, vitamin E intake, and total energy intake) were included (RR: 0.94; 95% CI: 0.79, 1.11; P for trend = 0.43). History of hypertension was not included in the final multivariate model because it may be in the causal pathway between calcium and IHD, but including this variable did not change the results. Further adjustment for nutrient variables (the P:S and intakes of trans fatty acids, protein, n-3 fatty acids, -linolenic acid, folate, and cereal fiber) did not alter the results: the RR for the highest compared with the lowest quintile of calcium intake was 0.97 (95% CI 0.81, 1.16; P for trend = 0.64). We repeated the above analyses for baseline and simple updated calcium intakes and found similar results. In both age-adjusted and multivariate-adjusted analyses, dietary calcium and calcium from dairy products (after exclusion of any calcium supplement users) were not significantly associated with the risk of IHD (Table 2). The age-adjusted analysis for supplemental calcium intake was significant, but became nonsignificant when adjusted further for other risk factors.
View this table:
TABLE 2 . Age- and multivariate-adjusted relative risk (RR) of developing ischemic heart disease according to quintiles of cumulative total calcium, dietary calcium, dairy calcium, and calcium supplement intakes among men in the Health Professional’s Follow-up Study1
When fatal IHD and nonfatal MI were examined separately, the RRs did not appear to differ significantly (Table 3). However, there was an apparent inverse association between supplemental calcium intake and fatal IHD; the RR for the highest quintile of calcium supplement intake compared with the nonusers was 0.61 (95% CI: 0.34, 1.10), but among the calcium supplement users there was no dose-response relation (P = 0.35).
View this table:
TABLE 3 . Multivariate-adjusted relative risk (RR) of developing nonfatal myocardial infarction (MI) and fatal ischemic heart disease (IHD) according to quintiles of cumulative total calcium, dietary calcium, dairy calcium, and calcium supplement intakes among men in the Health Professional’s Follow-up Study1
The associations with total vitamin D intake, dietary vitamin D intake, intake from vitamin D supplements, and total intake of dairy products largely followed patterns similar to that for calcium (data not shown). The multivariate-adjusted (including nutrients) RR of IHD for the men in the highest quintile of total vitamin D intake was 1.00 (95% CI: 0.80, 1.24; P for trend = 0.66), for dietary vitamin D intake was 1.04 (95% CI: 0.83, 1.30; P for trend = 0.74), and for total intake of dairy products was 1.01 (95% CI: 0.83, 1.23; P for trend = 0.57). For intakes from vitamin D supplements, the RR for the highest quintile compared with the nonusers of supplements was 0.71 (95% CI: 0.53, 0.96; P for trend = 0.03). However, there was no dose-response relation when the analysis was limited to supplement users; the RR for the highest compared with the lowest quintile was 0.88 (95% CI: 0.58, 1.33; P for trend = 0.21). The association between total calcium intake and total IHD did not appear to be modified by vitamin D intake (P for interaction = 0.9).
DISCUSSION
In this prospective cohort study, we found no significant association between intake of calcium and IHD risk before or after adjusting for IHD risk factors (including nutrient variables other than calcium intake). Total and dietary vitamin D intakes and total dairy product intake were also not associated with the risk of IHD. Analyses of calcium supplement intake and total IHD showed similar negative results, although an apparent inverse association with fatal IHD was observed. This inverse relation was not evident among the calcium supplement users after exclusion of the nonusers of supplements.
Our results are consistent with those of the Dutch follow-up study, which found no association between calcium intake and IHD mortality risk (RR for the lowest compared with the highest tertile of calcium intake: 0.90; 95% CI: 0.6, 1.6). However, they had a much smaller sample size (n = 2605) than did our study and measured calcium intake only with a 1-wk food-frequency-recall questionnaire (5). The other major US cohort in which the association between calcium intake and IHD risk was evaluated with a range of total calcium intakes similar to those used in the current study is that of the Iowa Women’s Health Study (6). Bostick et al, in a comparison of extreme quintiles, found that total calcium intake was inversely associated with IHD mortality (RR: 0.67; 95% CI: 0.47, 0.94; P for trend = 0.09). Even though the relation between dietary and calcium supplement intakes and IHD mortality showed a similar trend in their study (6), neither was significant. Their study did not evaluate nonfatal MI.
The null results are unlikely to be explained by chance because this finding was consistent when different calcium dietary intake variables were used (which were recorded and updated 3 times during the follow-up). Recall bias would not have influenced our results because all of the dietary data were collected prospectively. Food-frequency questionnaires are subject to inaccuracies, which might account for a null finding. However, our questionnaire was validated and found to be adequately accurate (11). Using the same dietary assessment, we found that dietary calcium is consistently associated with a reduced incidence of renal stones; a high intake of calcium reduces the urinary excretion of oxalate, which is thought to lower the risk of renal stones (22, 23). Thus, the lack of an observed association cannot be attributed to our inability to measure calcium intake. We adjusted for possible confounders in the association between calcium intake and IHD, but there was still the possibility of residual or unmeasured confounding. An unmeasured source of calcium intake in this cohort was from water; however, this was unlikely to have influenced the results because the intake of calcium from water is negligible compared with that from the diet.
In conclusion, our results suggest that calcium intake does not have an important relation to the risk of IHD in men. In addition, neither vitamin D intake nor total dairy product intake was significantly associated with the risk of IHD.
ACKNOWLEDGMENTS
The contributions of the authors to the manuscript were as follows: study concept and design (WKA-D and FBH), data collection (ER and WCW), data analyses (WKA-D and FBH), draft of manuscript (WKA-D), and critical revision of manuscript (ER, WCW, MJS, and FBH). None of the authors had any conflict of interest from a financial, personal, or professional aspect in relation to the findings of this study.
REFERENCES