| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
National Dairy Council, Rosemont, Illinois (G.D.M., D.D.D.)
Division of Nephrology, Hypertension, and Clinical Pharmacology, Oregon Health Sciences University, Portland, Oregon (M.E.R., D.A.M.)
Address reprint requests to: David A. McCarron, M.D. Division of Nephrology—PP262 Oregon Health Sciences University 3314 SW US Veterans Hospital Road Portland OR 97201-2940
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUMMARY OF SCIENTIFIC DATASUMMARYREFERENCES |
|---|
Key words: calcium-dietary, blood pressure, hypertension, dairy foods, review
Key teaching points
• Epidemiologic evidence first indicated a relationship between calcium intake and blood pressure control that was later supported by results of both laboratory and clinical studies.
• The DASH study assessed the effects of dietary patterns on blood pressure and observed the greatest reductions with diets containing high levels of dairy foods, fruits and vegetables.
• The DASH findings are consistent with the earlier epidemiologic, laboratory and clinical intervention studies.
• Meta-analyses of observational surveys and randomized controlled trials predict blood pressure decreases at the levels observed in the DASH study.
• Three to four daily servings of dairy products are recommended as adequate intake for optimal blood pressure control and bone health as well as prevention of other diseases including osteoporosis and some forms of cancer.
|
SUMMARY OF SCIENTIFIC DATA |
|---|
|
TOPABSTRACTSUMMARY OF SCIENTIFIC DATASUMMARYREFERENCES |
|---|
Epidemiologic surveys initially identified the dairy product/blood pressure relationship through the observation that the intakes of minerals, particularly calcium and potassium, were inversely linked to blood pressure and/or the incidence of hypertension [see reviews 1,2]. Because dietary calcium, of which 70% to 75% of its biological source is dairy products, acts as a surrogate for dairy product consumption, many of the epidemiologic databases specifically identified low dairy intake as the nutritional pattern predicting a low intake of calcium and, in some cases, potassium, magnesium and phosphorus in individuals and/or populations with higher arterial pressure and/or hypertension prevalence.
The initial series of epidemiologic reports prompted a rapid expansion in laboratory investigations of blood pressure control. These studies, summarized below, provided supportive evidence that inadequate exposure to the mineral profile associated with dairy product consumption had a consistent and statistically significant effect on arterial pressure in normotensive experimental models as well as in classic hypertensive models of arterial pressure control. Furthermore, these studies demonstrate that it is inadequate intake of minerals including calcium, potassium, magnesium and phosphorus, rather than higher levels of electrolytes, that is associated with increased blood pressure [3–5].
In parallel with the growing data from observational and laboratory studies, numerous research programs undertook clinical studies that sought to define two issues in humans. The first was whether there was evidence from metabolic studies that inadequate mineral balance, principally insufficient calcium intake, was linked to arterial pressure. The second was whether improving mineral intake would result in reductions in blood pressure in persons with high or high-normal blood pressure levels. These clinical investigations established that in individuals with increased arterial pressure or in populations at increased risk of hypertension, there is consistent and compelling evidence that impaired mineral balance is associated with higher blood pressure [1,6–8].
This body of evidence, generated by investigators throughout the world, served as the stimulus for the National Heart, Lung, and Blood Institute (NHLBI) to initiate the multicenter, randomized clinical trial study "Dietary Approaches to Stop Hypertension" (DASH) which was designed to assess overall dietary patterns and blood pressure control [9]. On the basis of its conclusion that a diet low in essential minerals and fiber and high in fat was associated with increased blood pressure, the DASH Steering Committee identified the "ideal diet" for reducing blood pressure as one that is high in low-fat dairy products and fruits and vegetables. DASH was a 10-week dietary intervention study comparing the typical American diet—low in fruits, vegetables and dairy products, and thus low in essential minerals and fiber and high in fat—to a diet that was high in fruit and vegetable content and to a diet high in fruits, vegetables and low-fat dairy products. This latter diet, the "DASH diet", was distinguished by its higher contents of potassium, magnesium, calcium and fiber, lower fat and minimally increased protein.
The findings from DASH, reported in the New England Journal of Medicine in April 1997 [10], form the cornerstone of this summary as this definitive study in persons with high-normal blood pressure demonstrated an unequivocal beneficial effect of a diet rich in low-fat dairy products and fruits and vegetables on arterial pressure control in these individuals and, more so, in individuals with established hypertension. In the words of the DASH investigators, the observed effect on blood pressure of the ideal diet was equal to or exceeded that of single-drug therapy for the management of hypertension.
The summary estimates of increasing dairy product intake in the diet from a range matching the lower 25 percentile of the U.S. population to an intake meeting current recommendations resulted in both systolic and diastolic blood pressure reductions that are consistent with the observational data and the previously reported randomized clinical trials that tested the blood pressure response to either increased dietary sources of minerals or calcium supplements. Thus, the DASH findings are consistent with the predicted ranges of response from the observational studies noted above and below and consistent with the prior dietary intervention studies that increased calcium intake using dairy products rather than calcium supplements. This internal consistency is critical as it provides the most compelling evidence of the biological plausibility: that the predictions from observational data are confirmed by state-of-the-art intervention studies such as DASH and that DASH is consistent with other intervention studies that have modified dietary calcium intake through increased dairy product consumption.
Finally, the evidence provided here demonstrates that the level of dairy product consumption needed to achieve the range of calcium in the diet that effects these blood pressure changes is consistent with the current recommendations of the Food and Nutrition Board of the National Academy of Sciences [11, 12]. In addition, the level of dairy product consumption shown to achieve this higher calcium intake is within the range recommended by the National Institutes of Health Consensus Conference on Optimal Calcium Intake [13]. In conclusion, the evidence from the biomedical literature cited in this review provides consistent and compelling evidence that a dietary pattern that includes three to four servings of dairy products per day is important for optimal blood pressure regulation in humans.
Dietary Approaches to Stop Hypertension (DASH) Study
The NHLBI multicenter study Dietary Approaches to Stop Hypertension (DASH) was designed to assess the blood pressure-lowering effect of dietary patterns as opposed to individual nutrients or nutrient supplements [9,10]. Prior dietary intervention studies have suggested causal relationships between diet and blood pressure, but the results have been inconsistent. The DASH Study Steering Committee postulated that these discrepancies may be due to a variety of factors including
First, the blood pressure-lowering effect of single nutrients may be too small to detect in small-scale clinical trials. Second, when several nutrients...are consumed together as in observational studies..., their additive effect may be sufficiently large to be detectable. Third, interactions could exist among nutrients to amplify the effect of combinations. Fourth, untested or unknown nutrients in plant food may lower blood pressure. Fifth, nutrient supplements may not affect blood pressure to the same extent as do the same nutrients occurring naturally in foods [9].
Investigative groups at Harvard, Johns Hopkins, Duke and Louisiana State universities were selected to execute the DASH trial, and the Kaiser Center for Health Research in Portland, Oregon, served as the study coordinating center. Participants in the trial were individuals above the age of 22 years with average systolic blood pressure less than 160 mmHg and a diastolic pressure of 80 to 95 mmHg. Persons with drug-treated hypertension were enrolled if they met the inclusion criteria for blood pressure after supervised withdrawal of the medication. The major exclusion criteria were poorly controlled diabetes, hyperlipidemia and cardiovascular events within the previous six months, chronic diseases that might interfere with participation, pregnancy or lactation, body mass index of more than 35, use of medications that affect blood pressure, unwillingness to stop taking vitamin or mineral supplements or antacids containing magnesium or calcium, renal insufficiency and alcoholic intake of more than 14 drinks per week. Because of the disproportionate burden of hypertension in the minority population, particularly among Blacks, the goal of the trial was to recruit a cohort in which two-thirds of the subjects were members of a racial ethnic minority.
The four clinical sites screened a total of 8,813 persons of whom 502 started the run-in phase. Of these, 459 were randomized and completed the trial. Table 1 portrays the baseline characteristics of the subjects in each diet group. For each group, the trial was conducted in three phases (screening, run-in, and intervention); at the three screening visits, standing blood pressure was measured by certified staff members using random zero sphygmomanometers and standard techniques including paired blood pressure measurements at each visit. During the three-week run-in all subjects received the control diet, comprising the nutritional composition of typical American diets. Potassium, magnesium and calcium levels approximated the 25th percentile for the U.S. consumption based on National Center for Health Statistics data from the 1980s. The macronutrient profile and fiber content corresponded to average consumption. At the end of the run-in, participants were randomized to one of three diets and learned of their assignments on the first day of the eight-week intervention.
|
Table 2 describes the nutrient targets, menu analysis and average daily servings of foods according to diet. The feeding protocol in the DASH trial at each of the four centers included a seven-day menu cycle with 21 meals at four caloric levels for each diet. Commonly available foods were used in all three diets, and all of the centers used the same brand of given food items. The food was prepared in a research kitchen each weekday; the subjects ate lunch and dinner on-site and were given meals to be consumed off-site on Fridays and weekends. Subjects were instructed to have no more than three caffeinated beverages and no more than two alcoholic drinks per day. They were given two packets of salt each containing 0.2 grams of sodium for daily discretionary use. Weight was measured each weekday and kept stable by changing caloric levels and by adding snacks with nutrient contents that corresponded to subjects’ diets. For each day of the controlled feeding phase, the subjects recorded their intake of discretionary items (beverage and salt) and any non-study foods, and whether they did not eat all the study food.
|
The DASH study achieved remarkable adherence; the percentage of subjects to complete the intervention phase in the control, fruits-and-vegetables and combination diets was 95.5%, 97.4% and 98.7% respectively. In the three groups, attendance at on-site meals was 95.8%, 95.4% and 96.1%, and, according to subjects’ reports, they adhered strictly to the study diets on 94.6%, 93.9% and 93.2% of the person days. Urinary analysis revealed that potassium excretion increased significantly from the run-in to the intervention phase on the fruits-and-vegetables and combination diets and that magnesium excretion increased in the combination diet. Urinary urea nitrogen excretion, indicative of total protein intake, increased substantially with the combination diet. Urinary calcium excretion decreased in the control diet and fruits-and-vegetables diet. Urinary phosphorus excretion increased significantly in the combination diet, and, in each group, urinary sodium excretion changed little between baseline and intervention; mean changes in weight from the end of run-in to the end of intervention were minimal and not significant between the groups.
Blood pressure results are shown in Table 3. As hypothesized in the study design, the combination diet compared to the control group resulted in highly significant reductions in blood pressure; with the combination diet systolic pressure was reduced by 5 mmHg more and diastolic pressure by 3.0 mmHg more than with the control diet. Blood pressure reductions with the fruits-and-vegetables diet compared to the control were also highly significant, but were only about half (2.8 mmHg systolic and 1.1 mmHg diastolic) of those achieved with the combination diet. The reductions with both the intervention diets were observed within the first two weeks and were sustained for the remaining six weeks of the intervention (Fig. 1).
|
|
Of particular clinical significance, the blood pressure reduction observed in the hypertensive subjects on the combination diet compared to the control diet was 11.4 mmHg systolic and 5.5 mmHg diastolic blood pressure. The comparable changes in hypertensive subjects on the control diet was 0.72 mmHg systolic and 0.28 mmHg diastolic blood pressure. The comparison of the combination diet to the fruits-and-vegetables diet in hypertensive persons was also significantly different, with a reduction of 4.1 mmHg systolic blood pressure and 2.6 mmHg diastolic blood pressure greater in the combination diet than in the fruits-and-vegetables diet. There were no clinically significant complications reported. Interestingly, gastrointestinal symptoms which might have been thought to have been greater in the fruits-and-vegetables or combination diets, were actually reported less in these two diets than in the control group.
Thus, the DASH investigators concluded that the DASH trial "demonstrated that certain dietary patterns can favorably affect blood pressure in adults with average systolic blood pressures of less than 160 mmHg and diastolic blood pressures of 80 to 95 mmHg. Specifically, a diet rich in fruits, vegetables and low-fat dairy products and with reduced saturated and total fat lowered systolic blood pressure more than a control diet. A diet rich in fruits and vegetables also reduced blood pressure but to a lesser extent" [than the combination diet]. The DASH investigators further noted that "known diet-related determinants of blood pressure (sodium chloride, body weight and alcohol) could not have accounted for the reductions in blood pressure, because changes in these potential confounders were small and similar for all the diets."
Among other important findings were that blood pressure reductions were observed within two weeks of starting the intervention and maintained throughout the study; the pattern blood pressure of reduction among diets was similar in men and women and members of non-minority groups using an average of 24-hour ambulatory blood pressures as the outcome; and attendance rates and reported diet adherence (substantiated by urinary analyses) were excellent. The DASH investigators noted that while urinary calcium excretion decreased with the control and fruits-and-vegetables diet and was comparable to baseline values with the combination diet, it was likely calcium absorption was impeded by the higher fiber content of the first two diets. They went on to emphasize, however, that "the substantial increase in urinary phosphorus excretion with the combination diet suggested the subjects in this group did consume more dairy products than the other participants."
The DASH investigators concluded that the DASH results "...should be broadly applicable to the U.S. population. The study population was demographically heterogeneous, with a range of blood pressure that includes approximately 40% of the U.S. adults [the majority of the remainder are below the entry criteria, not above], 49% of the subjects were women, 60% were Black and 30% had household incomes of less than $30,000 a year." Furthermore, because the combination diet comprises commonly available foods including meat, it could likely be adopted by the general population. The number of servings of fruits and vegetables in both the fruits-and-vegetables and combination diets (8–10 servings) was nearly twice the current U.S. average for adults (4.3 servings), and the combination diet included similarly more servings (2.7) per day of dairy products than the current average adult consumption of 1.5 servings per day. The results of the DASH study demonstrate, in a demographically representative population of adults, that consuming a diet that meets current dietary guidelines, although significantly higher than typical in fruit, vegetable and low-fat dairy product content, is broadly applicable and achievable across the U.S. population. Furthermore, this dietary pattern has been previously recommended for reduction of other disease states such as coronary artery disease, colon cancer and osteoporosis.
While the DASH investigators appreciated the 10-week intervention may raise questions as to its applicability for long-term adherence, they still concluded that their results have both clinical and public health implications. One of these is that the range of blood pressure reductions observed in the DASH subjects without hypertension suggests that "...following the DASH combination diet might be an effective nutritional approach to preventing hypertension." Because these blood pressure reductions occurred in the setting of stabilized weight and sodium chloride intake and limited alcoholic consumption, they state that "...adoption of the DASH combination diet should complement, rather than supplant, what is currently recommended" for hypertension prevention, including weight control and reduced sodium chloride intake and alcoholic intake.
They further note that blood pressure reductions in hypertensive subjects with the combination diet were similar in magnitude to those reported in trials of drug monotherapy for mild hypertension. "Hence, following the DASH combination diet might be an effective alternative to drug therapy in people with stage I hypertension and might prevent or delay the initiation of drug therapy in people with blood pressure levels that straddle the thresholds for drug treatment."
From the perspective of a public health strategy, the DASH investigators stated that "...adoption of the combination diet could potentially shift the population distribution of blood pressure downward, reducing the occurrence of blood pressure-related cardiovascular disease." They estimated "that a population-wide reduction in systolic or diastolic blood pressure with the magnitude observed in the combination diet would reduce incident coronary heart disease by approximately 15% and stroke by approximately 27%." These investigators concluded that "...a diet rich in fruits and vegetables and low-fat dairy foods and with reduced saturated and total fat can substantially lower blood pressure. Such a diet offers an additional nutritional approach to the prevention and treatment of hypertension."
The DASH study provides compelling evidence that dairy products reduce blood pressure in normotensive and stage I hypertensive persons for the following reasons. The combination diet achieved significantly greater reductions in systolic and diastolic blood pressure than did the fruits-and-vegetables diet. The principle difference between these two diets was the inclusion of almost three servings a day of dairy products (predominantly low-fat) and reductions in total and saturated fat. The effect of the combination diet was even greater in hypertensive subjects than in those who were normotensive. In subgroups such as non-minority and normotensives, the fruits-and-vegetables diet failed to achieve a consistent and significant reduction of blood pressure, whereas the combination diet achieved highly significant reductions in systolic and diastolic blood pressure in each of these groups.
The remaining question is whether the lower saturated and total fat content of the combination diet compared to the other two diets in and of itself could have accounted for the blood pressure reductions. This conclusion is not supported by the data in the biomedical literature. The observational studies that have identified patterns of low mineral intake associated with increased arterial pressure have almost universally included assessment of the relationship of dietary fat consumption to blood pressure in the populations studied, and none have reported a clear effect of dietary fat. Clinical trials have also failed to identify a specific effect of fat intake on blood pressure regulation. The Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of Hypertension (JNC VI) concluded that on the basis of available information from randomized controlled studies, "...diets varying in total fat and proportions of saturated to unsaturated fats have had little, if any, effect on blood pressure" [14].
A review of dietary fat intervention studies by the National Kidney Foundation (NKF) Nonpharmacologic Management of Hypertension Organizing Committee [15–17] revealed that of 11 randomized controlled studies, the two largest reported no effect on blood pressure of varying fat content or the polyunsaturated-to-saturated fat intake; "...of the other nine studies, only two found any effect on blood pressure, and these have been criticized on methodologic grounds". While these conclusions do not preclude the possibility of a minimal effect of dietary fat, such an effect would not account for the 2.7 mmHg difference in systolic and 0.9 mmHg difference in diastolic blood pressure observed between the DASH combination and fruits-and-vegetables diets. Furthermore, it would not account for the much greater reductions in hypertensive blood pressure, both systolic and diastolic, for the combination diet compared to the fruits-and-vegetables diet.
Additional evidence that the reduction of fat would not account for the majority of the blood pressure reductions observed in the DASH study comes from laboratory investigations of mechanisms of blood pressure regulation. There is no strong or compelling evidence in the database from animal models that moderation of total and saturated fat intake can achieve blood pressure reductions of the magnitude observed in the DASH study. Finally, the range of blood pressure reductions achieved with the DASH combination diet are remarkably consistent with the summary estimates of the predicted blood pressure reduction that would be achieved with an increased dietary calcium from food reported in numerous observational studies and summarized in the meta-analysis by Birkett discussed below [18].
Furthermore, the range of blood pressure reduction is also consistent with the eight previous clinical intervention trials that assessed blood pressure effects of increasing calcium from food sources, principally dairy products. This comparability of the DASH experience to other randomized controlled trials is demonstrated in the recent meta-analysis by Griffith et al. [7] discussed below. Thus, the absence of evidence associating the reduction of fat with reduction in blood pressure from observational, intervention or laboratory studies, combined with the compatibility and consistency of the DASH blood pressure reductions in the combination diet with estimates based on observational and intervention trials of dietary calcium, i.e., dairy products, make it highly likely that blood pressure reductions achieved with the DASH combination diet can be ascribed largely to the inclusion of dairy products.
The findings of the DASH study were received by the biomedical community with a great deal of enthusiasm, and essentially no criticism has been published to date. There are several likely reasons for the general acceptance of the DASH findings, including
The possible criticism of the short term of the study is largely negated, first, by the fact that the blood pressure reductions observed were entirely compatible with the data from observational studies where the relationship between reported food intake and arterial pressure reflects long-term effects. While the observational database does not provide the tight control that an intervention study such as DASH does, it does offer strong evidence for dietary patterns over a lifetime which are associated with various disease states, in this case specifically hypertensive cardiovascular disease. As described in this paper, both the observational and intervention studies provide estimates of blood pressure reductions which are remarkably similar to those observed in the DASH cohort. Second, the absence of any adverse effects and the clear compatibility of the DASH diet with other diets recommended for the reduction of cardiovascular disease, cancer and osteoporosis in this country make it difficult to be critical of the short duration of the study. Adherence to the DASH diet could be achieved with effort similar to that of virtually all the other diets recommended for reduction of chronic diseases in the U.S., as outlined in such documents as the Dietary Guidelines for Americans [19] and the most recent report of the Food and Nutrition Board of the Institute of Medicine [11].
While an eight-week intervention period is relatively short, it is within the range of prior NIH trials for dietary strategies to reduce blood pressure and equal to that used for a substantial number of pharmaceutical interventions. The fact that blood pressure was reduced dramatically within the first two weeks and sustained for the following six weeks is compatible with the long-term effect and is consistent with the longstanding principles of nutrition and chronic disease; that is, repletion of dietary inadequacies result in a reversion of the biological parameters to some baseline value, as would appear to be the case in the DASH study where both normotensive and hypertensive subjects achieved highly significant reductions in blood pressure. Finally, there is no evidence in the current database from human studies, observational or clinical, or in the laboratory findings that there is a limiting effect of the blood pressure reduction that occurs with increasing the intake of dairy products. Specifically from the animal studies, the effects of increasing mineral intakes on blood pressure have been sustainable over 20 to 30 weeks, which are comparable to the lifetime of an adult [3]. There is no evidence of regression to baseline blood pressures in these studies and, thus, no reason to anticipate that incorporation of this dietary pattern, three to four servings of dairy products, would achieve anything but sustained reductions in blood pressure.
The acceptance of the findings of DASH is well demonstrated by the multiple authoritative statements that have now acknowledged their importance. Among these are
Thus, within a relatively short period of time, since April 1997, federal and voluntary health advisory groups have confirmed the relevance of the DASH findings to public health efforts to reduce hypertensive cardiovascular disease and its complications through appropriate dietary measures.
Consistency Between DASH and Prior Evidence in the Biomedical Literature
The findings of DASH outlined above are markedly consistent with an extensive body of literature that includes observational studies, randomized controlled trials, metabolic studies in humans and extensive laboratory investigations, each of which is summarized below. The observational studies demonstrate that a dietary pattern of low mineral intake is associated with an increased risk of hypertension worldwide and that inadequate intake of dairy products is the dietary pattern most often linked to the low mineral intake of high risk populations. The recent meta-analysis of these studies [18] provides estimates of the blood pressure reduction that would occur with increased intake of calcium from food sources such as dairy products. As described below, the range of the predicted benefits is remarkably consistent with the findings of DASH (Fig. 2 [23]).
|
Finally, laboratory investigations of mineral intake conducted over the past 20 years in a wide variety of animal models provide evidence of the biological possibility that increasing mineral intake through dairy products would be effective for the human population and that deficiencies in the overall nutrient profile are more important to blood pressure control than individual excesses [3–5]. These data further document that cardiovascular conditions associated with hypertension in the laboratory are reversed by increasing mineral intake. This evidence derived from animal models, including those of genetic hypertension, salt sensitivity and renal insufficiency, is compatible with the human data indicating that improving mineral balance can improve blood pressure levels.
Meta-analyses of Observational Studies
As noted in the introductory summary, the observational data linking inadequate dairy product consumption with increased risk of arterial pressure is remarkably consistent among age groups, genders and ethnic groups and has been identified across the range of blood pressures in adults, infants, the elderly and in pregnant women [1, 8]. That observational database was summarized in 1995 by Cappuccio et al. [24]. Their meta-analysis has recently been revised by Birkett [18] assessing the same studies, but correcting several methodologic errors identified in the first analysis; these corrections have been acknowledged as appropriate by Cappuccio et al. in an accompanying editorial [25].
While the Cappuccio meta-analysis did reveal a very weak association between increasing dietary calcium from food sources such as dairy products with a reduction in arterial pressure, when the appropriate corrections were made to that meta-analysis, this estimated effect was shown to be almost 30-fold greater than the original analysis. Of even greater significance, the Birkett meta-analysis provides an estimate effect (mmHg of blood pressure reduction per 100 mg of dietary calcium increase) which is remarkably consistent with the results of the DASH study, i.e. the observed reduction in both systolic and diastolic blood pressure with the approximate 800 mg increase in dietary calcium in the DASH combination diet.
Birkett abstracted the data from each paper used in the Cappuccio meta-analysis [26–44] to permit derivation of the unadjusted regression slope and standard error relating dietary calcium intake to blood pressure. The protocol employed to convert the standardized regression coefficients, correlations, and the like to regular regression slopes and standard deviations used the published methods of Cappuccio et al. [24], comparing abstracted data to that published in the original meta-analysis. In cases of disagreement, the original papers were reviewed by colleagues to determine which abstract values reflected the correct interpretation of data. The results from the individual studies were then combined to produce an overall estimate of the regression slope relating calcium intake to blood pressure. As noted above, Birkett identified several critical deviations from the data presented in the Cappuccio meta-analysis, the first based on the Zutphen Study by Kromhout et al. [26]. Birkett and colleagues noted that this received 99.99% of the total weight of the overall analysis. When they made the appropriate correction for this overweighting, pooled estimates for changes in systolic blood pressure in men increased from -0.01 to -0.32 mmHg per 100 mg increase in dietary calcium, and -0.009 to -0.212 mmHg per 100 mg of dietary calcium for diastolic blood pressure.
In addition, Birkett found that five of the included studies [27–31] had been given inappropriate weight in the analysis, were included more than once or included data inappropriate for meta-analysis. Correcting for these errors and using the most stringent inclusion criteria based on a modification of Cappuccio’s original meta-analysis, Birkett arrived at an estimate of a reduction in systolic blood pressure of -0.39 mmHg per 100 mg of calcium with a range of -0.47 to -0.31, and an estimate of -0.35 mmHg reduction in diastolic pressure per 100 mg of dietary calcium with a range of 0.67 to -0.02. Using these estimates based on "per 100 mg of dietary calcium intake," and 800 mg calcium per day increase in the DASH diet, Birkett’s findings would predict a 3.1 mm Hg greater reduction in systolic blood pressure when low-fat dairy products were added to the fruits-and-vegetables diet; the DASH study reported a 2.7 mmHg reduction in systolic blood pressure.
For diastolic blood pressure, Birkett’s analysis would predict a further 2.8 mmHg reduction with the addition of low-fat dairy products to the fruits-and-vegetables diet, and in fact, the DASH study observed a 1.9 mmHg further reduction. Thus, the most recently published, corrected estimates of the effect of higher dietary calcium, and therefore dairy product consumption, on blood pressure are remarkably consistent with and supportive of the DASH findings. This provides important evidence that the internal consistency of the data supports the biological plausibility of the effect and suggests that the addition of further studies of increasing dairy product consumption to increase mineral intake will continue to affirm what has now been established by published epidemiologic, clinical and laboratory evidence.
There now exists strong concordance and consistency between the data from observational studies and randomized, controlled clinical trials and, particularly, the DASH study. There appears to be no discrepancy within the observational database, which has identified a consistent benefit of increasing mineral intake, particularly calcium, as a means of reducing blood pressure. As noted above, virtually all of these trials have identified inadequate dairy product intake as the primary source of dietary calcium linked to hypertension risk. Specifically nine of the reports explicitly identified dairy products as the source of the dietary calcium linked to improved blood pressure control. The strong agreement that now exists between the observational studies and the clinical intervention trials obviates explaining any internal inconsistencies that might have otherwise existed and were previously noted by other reviews such as the 1995 report from the Federation of American Societies for Experimental Biology Life Sciences Research Office [2] and the original Cappuccio meta-analysis [24], which has now been corrected by the Birkett meta-analysis [18].
This interpretation of the concordance of the DASH findings with the extensive prior experience from observational studies is best documented by the actual experience of increasing dietary calcium from food sources. This concordance is depicted in Fig. 3. The studies originally summarized by Cappuccio, Cutler, and colleagues [24] and corrected by Birkett [18] are listed in Table 4 and in the references of this review.
|
|
The DASH study addressed each of these theoretical explanations for the heterogeneity. Their relevance to accounting for the heterogeneity is borne out by the DASH results. DASH included a greater proportion of minority subjects, particularly African-Americans, and individuals with high normal blood pressure, used solely commonly available foods rich in calcium, specifically low-fat dairy products, to increase the mineral content of the diet of the subjects and used a control diet with levels of mineral content that reflect the 25th percentile of the current U.S. population and thus reflected individuals under-consuming the nutrients of interest.
Having raised the issues noted above, Bucher and colleagues at McMaster University recently updated their 1996 meta-analysis of randomized control trials of increasing calcium intake for the purpose of lowering arterial pressure [7]. In the re-analysis they used the justification and guidelines published in an editorial in the New England Journal of Medicine [45] which called for "cumulative meta-analyses". The editorial pointed out limitations of pooled analysis in clinical trials, but acknowledged the importance of using this technique to identify trends which could be further defined in better designed trials in the future. They also called for the incorporation of new data into existing meta-analyses in order to better define relationships between intervention and dose response. Using that objective, the McMaster research group conducted a new meta-analysis incorporating the findings of DASH, as well two other clinical trials that have been published since the publication of their previous analysis.
The McMasters research group is part of the Cochrane Collaboration, which set the international standards for summarizing data from randomized controlled trials [46]. Their analyses use state-of-the-art statistical techniques which provide the most conservative estimates of the effect of an intervention on the proposed outcome. In the re-analysis, they specifically excluded studies in which a) blood pressure was measured in less than 80% of the patients randomized, b) the intervention was shorter than two weeks, c) intervention groups ingested a total of less than 1,000 mg of calcium/day, d) there were duplicate reports on the same patient and/or e) study results were not available in a form to allow combining with the other studies. For this meta-analysis, they again used methodological quality assessment that included concealment of random allocation, blinding of participants, care givers and blood pressure measurers, number of blood pressure measurements, use of the random zero sphygmomanometer, formal training in blood pressure measuring and specifications of measurement technique. Prior to completing the analysis, these investigators identified several potential sources of heterogeneity including play of chance, age and gender of subjects, normotensive versus hypertensive subjects, baseline calcium, dietary versus non-dietary calcium supplementation and methodological qualifications of a study.
The new meta-analysis identified 66 randomized trials in nonpregnant study populations [7]. Eleven of these had not been included in their prior meta-analysis. Of the 66, 42 proved to be eligible for the overview [10, 47–87]. As listed in Table 5, nine of these used dairy foods and other dietary manipulations to increase mineral intake in the diet [10, 80–87], and these were subsequently compared to the 33 randomized controlled trials that used supplemental calcium [47–79]. Analysis of all the eligible studies revealed significant reductions in systolic blood pressure, -1.44 mm Hg (95% confidence interval [CI] -2.20 to 0.68; p < 0.001), and in diastolic blood pressure, -0.84 mm Hg (CI -1.44 to -0.24; p < 0.001).
|
In the nine randomized controlled trials that have used food sources to increase mineral intake, compared with the 33 supplementation trials, there was evidence of a differing effect on blood pressure. In the dietary studies, the change in systolic blood pressure was -2.10 mm Hg (CI -2.93 to -1.26) compared to -1.09 mm Hg (CI -2.12 to -0.06) observed in the supplement studies (p = 0.14). The meta-analysis detected a slightly greater effect on diastolic blood pressure in the dietary trials, -1.09 mm Hg (CI -1.67 to -0.52), compared to -0.87 mm Hg (CI -1.71 to -0.03) in those that used supplements (p=0.67).
Of greatest significance was the observation that in the trials using dietary calcium, both systolic and diastolic blood pressure responses were remarkably homogeneous; that is, the range of blood pressure reduction in the nine trials was consistent, with each trial exhibiting approximately the same range of reduction. However, in trials utilizing supplemental calcium, the marked heterogeneity in blood pressure response persisted, indicating that factors tracking with calcium from dietary sources were essential to elicit a optimal and consistent beneficial effect of increasing calcium intake. Thus, the second meta-analysis demonstrated that the blood pressure-lowering effect of dietary calcium is more consistent than that of calcium supplementation. These findings are consistent with the results of the DASH study and confirm the DASH hypothesis that dietary mineral sources and their nutrient interactions would have a stronger effect on blood pressure than supplementation of a single nutrient. Furthermore, the finding of greater homogeneity in the dietary trials was crucial to confirming the consistency of the DASH findings with prior trials in which dairy foods were used to increase mineral intake.
From this updated McMasters meta-analysis, we can conclude that the significantly greater blood pressure reductions that occurred when low-fat dairy products were added to the fruits-and-vegetables diet in DASH are consistent with the other eight published randomized clinical trials that tested the blood pressure effects of increasing calcium from the diet. Thus, the results of DASH cannot be discounted as simply one trial. DASH represents the predictable conclusion of more than 15 years of clinical investigation that has explored the influence of adequate intake of calcium and other minerals uniquely available in dairy products on blood pressure regulation. Furthermore, the initial and now extensive identification of a possible cardiovascular benefit of dietary calcium, as a surrogate for dairy product intake, that has been reported from observational studies of populations is also validated by the DASH findings. As could be anticipated under the best circumstances, the biological association between blood pressure and calcium intake in the observational studies, as reported in the Birkett meta-analysis [18], has been clearly confirmed by DASH and the other randomized controlled trials.
This conclusion is further strengthened by the quality of the DASH trial and the rigorous statistical methods employed by the authors of both the McMaster clinical trials analysis [7] and the Birkett observational studies analysis [18]. Both research groups employed the stringent criteria of the Cochrane Collaboration, identical to that used by the U.S. and Canadian Task Force on preventive health strategies as well as other outcome groups throughout the world to set public health policy and practice guidelines. While some of the smaller studies incorporated in the dietary trials in the McMaster analysis could be criticized, the finding of internal consistency best demonstrated by the lack of heterogeneity in these studies makes this argument moot. Furthermore, without exception, all of the reported dietary intervention studies have demonstrated a significant effect of increasing minerals in the diet.
The most important finding from the comparison of the McMaster’s meta-analysis of all prior dietary calcium trials with the DASH study is that their summary estimates of -2.2 mm Hg systolic and -0.95 mm Hg diastolic blood pressure reductions are remarkably consistent with the net differences between the DASH fruits-and-vegetables diet and the DASH combination diet which were -2.7 mm Hg systolic and -1.9 mm Hg diastolic blood pressure. This concordance of estimates versus actual outcomes in the DASH study is compelling evidence of the consistency of the beneficial effect of increasing dietary calcium through dairy products on blood pressure in normotensive and hypertensive humans.
The compatibility of these findings speaks specifically to the health goals set by expert groups in the federal government, national health agencies and scientific organizations. DASH demonstrated that a diet rich in low-fat dairy products and fruits and vegetables reduced blood pressure in both high-normal and hypertensive persons. On the combination diet, the DASH participants increased their intake of dairy products from a level representative of the lowest quartile of consumption in the United States to a level consistent with current national recommendations, including those of the NIH [13], the Institute of Medicine [11] and the National Research Council [12]. In addition, while DASH was limited to an eight-week intervention, the consistency of DASH results with the summary estimates from the observational studies [18] suggests that the long-term benefits of increasing dietary calcium from dairy products would be sustainable over a lifetime.
The importance of the DASH findings to public health strategies designed to reduce cardiovascular disease is echoed in the words of the DASH investigators who stated that the impact of the combination diet was independent of other current dietary recommendations for blood pressure control including limiting sodium, calorie and alcohol intake, all of which were controlled for in the DASH study design. These investigators concluded that the magnitude of the blood pressure reduction was comparable to typical mono-drug therapy for hypertension, suggesting that a diet high in low-fat dairy products and fruits and vegetables for the management of high blood pressure could significantly reduce if not eliminate medication requirements for persons at risk. Importantly, unlike many antihypertensive drugs which are limited in their utility to specific subgroups of patients, the effect of the DASH diet was generalizable to the entire population. This conclusion is what would be expected for a dietary intervention that is within the range of current recommendations and that corrects for what is generally regarded as an important nutritional deficiency, i.e., inadequate mineral intake.
Beyond the demonstration by observational and clinical studies that dairy products improve blood pressure control within the current range of human consumption, the biological plausibility of this effect is supported by metabolic studies of humans which have identified disturbances of mineral metabolism in association with increased arterial pressure. In parallel, laboratory studies in experimental models of hypertension have confirmed in large part these observations in humans. While not the subject of the present review, these data from other areas of investigation offer additional evidence that assuring consumption of dairy products at levels adequate to achieve sufficient mineral intake is an important, established and biologically plausible factor in optimal blood pressure regulation in humans.
Scientific Consensus
The above conclusion is supported by the recommendations and statements of several representative scientific bodies published within the past five years. Six current authoritative documents have indicated that an association exists between a diet adequate in dairy product content and a reduced risk of hypertension. Primary among these is the Sixth Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of Hypertension, the authoritative periodic report of the NHLBI High Blood Pressure Education Program published in 1997 [14]. For the first time, this committee has recommended dietary rather than single-nutrient means for managing hypertension. JNC VI specifically recommends the DASH diet, a diet rich in low-fat dairy products and fruits and vegetables, for the prevention and treatment of high blood pressure. The contents of the DASH diet are included in the JNC VI report.
The second authoritative document is the summary review from the Life Sciences Research Office [2]. In this document, the expert panel reviewing the literature available at that time concluded that the scientific data indicated an association between diets poor in calcium and dairy products and both higher arterial pressure and increased prevalence of hypertension. They noted that this relationship was particularly prevalent in segments of the population at higher risk of hypertension, including African-Americans, pregnant women and elderly subjects, groups that typically consume less than the recommended levels of calcium.
The published rationale for the DASH Study prepared by the DASH Steering Committee is the third report of an expert panel documenting the association between dairy product intake and blood pressure control [9]. In their published rationale for the study design of DASH, the Steering Committee concluded that, based on available scientific information, the "ideal diet" for prevention and management of hypertension is one rich in low-fat dairy products and fruits and vegetables. The DASH Steering Committee comprises members selected by peer review and includes NHLBI staff; none of the university-based members of this committee had previously been directly involved in the development of the research base that identified low-fat dairy products as the component of the ideal diet or what was later designated the combination diet.
The fourth authoritative document is the National Kidney Foundation Syllabus on Nonpharmacologic Management of Hypertension [15]. This summary review of the available scientific literature relating life-style and nutritional factors to the prevention and control of hypertension was originally published in 1988 by the NKF Council on Hypertension. In its most recent revision in 1996, this syllabus concluded that "the daily consumption of the current RDA of calcium (1000–1200 mg) is associated with a reduction in the risk of developing hypertension, and lowers blood pressure levels in a subset of individuals with existing essential hypertension." The NKF’s assessment of the biomedical literature as it pertains to diet and hypertension is consistent with the conclusion that a diet rich in dairy products is associated with a reduction in arterial pressure and incidence of hypertension.
The fifth source is the 1997 Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Chloride from the Food and Nutrition Board of the Institute of Medicine of the National Academy of Sciences [11]. This document, citing the DASH study, concurred that "a diet with increased low-fat dairy products lowered blood pressure when fed to normotensive and hypertensive adults."
The sixth authoritative statement is the recent update of dietary guidelines for health American adults from the Nutrition Committee of the American Heart Association [20]. This committee report of current AHA recommendations for appropriate dietary practices for the prevention and management of cardiovascular disease states that "...the DASH diet convincingly reaffirms the importance of overall diet to blood pressure control" and that the blood pressure-lowering effects of diet can be amplified "...by assuring adequate intakes of foods such as fruits, vegetables, and low fat dairy products."
Recommended Levels of Dairy Product Intake
The quantity of dairy products necessary to reduce blood pressure or hypertension risk is consistent with the recent National Academy of Sciences recommendations for adults: three to four servings per day of dairy products. Three or four servings of dairy products have been specifically recommended by at least two national health agencies as the range for adequate or optimal intake. The first of these, the National Institutes of Health Consensus Conference on Optimal Calcium Intake [13], states that the "preferred approach to attaining optimal calcium intake is through dietary sources" and specifically identifies dairy products as the ideal source of calcium in the diet.
The second national recommendation of at least three servings of low-fat dairy products per day is in the recently published Dietary Reference Intakes for Calcium, Phosphorus, Magnesium, Vitamin D, and Chloride from the Institute of Medicine [11]. That document sets an adequate level of calcium at 1,000 mg per day for adults under the age of 50, which is consistent with a minimum of three servings of dairy products per day. For adults over the age of 50, the recommendation is 1,200 mg, consistent with an intake of four servings of dairy products per day.
|
SUMMARY |
|---|
|
TOPABSTRACTSUMMARY OF SCIENTIFIC DATASUMMARYREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTSUMMARY OF SCIENTIFIC DATASUMMARYREFERENCES |
|---|
