Abstract

OBJECTIVES:

• To review all prospective studies which published information about total dairy, milk, and other dairy products in relationship with cardiovascular disease risk and mortality/survival.
• To define the amount of consumption found to be related with possible effects on cardiovascular disease.
• To define the modifying effect of dairy fat on the relation between dairy (product) consumption and CVD.
• To define possible effect modification by confounders.

DATA SOURCE: The Pubmed database was searched (No start date - May 25th, 2010) for relevant articles using the keywords "butter, cheese, cream, dairy, dietary calcium, dietary fat, lactose, margarine, milk, saturated fat, or yoghurt" combined with "prospective, cohort, follow-up, or longitudinal". The exact search term is described Here.
Prospective studies published in the English language were included. Reference lists were searched for additional articles.
RESULTS: 50 articles were found which provided information about 29 different cohorts. Of these, 0 articles were excluded.
Because of little data, results are described for the combined end point of disease & mortality, unless stated otherwise.

• Total dairy. No evidence was found for an association between total dairy and total CVD. Suggestive evidence was found for an increased CHD risk among men at an intermediate level of consumption (237-405 g/day). And suggestive evidence was found for increased risk of CHD mortality from high consumption. In contrast, high consumption possibly protects against stroke (≥ 128 mg dairy calcium/day). And suggestive evidence was found for a protective effect against ischemic stroke (116-≥ 296 mg dairy calcium/day). No consistent differences in effect were found between full-, and low-fat dairy for any variable.
• Milk. No evidence was found for an association between milk and total CVD, CHD risk, CHD mortality, and stroke risk. Milk consumption possibly protects against stroke mortality at consumption ≥ 4-5 times/week. No consistent differences in effect were found between full-, and low-fat milk for any variable.
• Cheese. No evidence was found for an association between cheese and total CVD, CHD risk, CHD mortality, and stroke mortality. High consumption of cheese possibly protects against ischemic stroke risk. No consistent differences in effect were found between full-, and low-fat cheese for any variable.
• Cream. Inconclusive findings were done, but suggestive evidence was found for a protective effect among men when CHD and stroke were considered one end point.
• Butter vs margarine. No independent effects on CHD were found. Within-cohort comparisons showed margarine to be more protective than butter in 1 out of 6 studies. In contrast, margarine possibly protects against stroke mortality among women at consumption 4-7 times/week, and butter possibly increases risk of inctracerebral hemorrhage.
• Other dairy items. No evidence was found for an association between yoghurt, ice cream, or other dairy items and any end point.

CONCLUSION: Suggestive evidence was found for an increased CHD risk, and for increased CHD mortality (+ 13%) of dairy consumption. The association with disease risk was found among men and at an intermediate level of dairy consumption (237-405 g/day). In contrast, dairy possibly protects against total stroke (- 31%), and suggestive evidence was found for a protective effect against ischemic stroke. Both effects were found at relatively low consumption (≥ 128-296 mg dairy calcium/day).
Milk consumption possibly protects against stroke mortality at consumption ≥ 4-5 times/week (- 20%). High consumption of cheese possibly protects against ischemic stroke risk (- 15%). And suggestive evidence was found for a protective effect of high cream consumption among men when CHD- and stroke risk were considered one end point (- 18%).
No evidence was found that margarine is beneficial over butter for CHD, but this was not the case for stroke. Margarine possibly protects against stroke mortality (- 27%) among women at consumption 4-7 times/week, while butter possibly increases risk of intracerebral hemorrhage.
No evidence was found for a modifying effect of dairy-, milk-, and cheese fat on the association between the related dairy items and any end point. But effects may differ by race: The evidence for increased risks was totally restricted to US/European populations. Protective effects of dairy/milk were found among Asian populations, and protective effects of cheese/cream were found among US/European populations.
PERSPECTIVE: Most evidence was restricted to findings from either Asian or European/US cohorts. In addition, some evidence was found at intermediate levels of consumption and not among both sexes. Very little data was available about the relation with total CVD, and all evidence was limited to specific CVD end points. Current evidence should be looked at in this perspective: Associations with specific end points may be balanced by opposing effects on other CVD end points.
Current literature does not allow evidence based advice on the relation between dairy or dairy fat and CVD, especially among US/European cohorts. Though some increased risks were found with total dairy consumption in these populations, no increased risks were found with specific dairy items. Surprisingly, some protective effects were found among the dairy items which contain a relatively large amount of saturated fats (cream, cheese).
New analysis are needed, preferably pooled analysis, to determine which dairy items may be related to CVD, and to specify at which levels of consumption these effects can be found. Pooled analysis also allows for looking at relatively low levels of consumption. And could answer the following question: Is the protective effect of dairy/milk against stroke among Asian populations due to the relatively low consumption, or due to other variables?

Introduction.

The American Heart Association advices limiting the intake of saturated fat to < 7% of energy. Based on this, 1 out of 3 practical advices reads:
"selecting fat-free (skim), 1%-fat, and low-fat dairy products" (American Heart Association Nutrition Committee).
This advice is based solely on the assumption that saturated fats increase LDL cholesterol, and by doing so would automatically increase CVD risk. While giving this advice, the AHA completely ignores the large body of evidence on the direct relation between dairy consumption and CVD from prospective studies.

The current evidence on the relation between saturated fats and cholesterol & on the relation between dairy and CHD, will shortly be discussed. Results and conclusions from this systematic review can be found by checking the related items in the menu.

|Reference: American Heart Association Nutrition Committee. Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation. 2006 Jul 4;114(1):82-96. Full text|

Cholesterol and heart disease.

Milk and total dairy contain saturated fats which have been linked to increased cholesterol levels in the past, and increased cholesterol levels have been linked to increased chance of heart disease.
In 2003, a meta-analysis of 60 controlled trials was published. The relationship between dietary fats and cholesterol levels was examined. The analysis showed that saturated fat did indeed increase cholesterol levels compared to carbohydrates, but both the "good" (HDL), and "bad" (LDL) cholesterol increased. No effect was found on the ratio total:HDL cholesterol. The authors of the analysis stress the fact the ratio of total:HDL cholesterol is considered more important than the total or lipoprotein cholesterol concentrations in estimating the risk of CAD (Mensink RP 2003).

The authors of the former meta-analysis warn against using effects on cholesterol alone as a risk marker of CAD. Quote:

Also, they advice the use of studies on the direct relationship between diet and CVD. Quote:

Conclusion: Saturated fatty acids do not affect the ratio of total:HDL cholesterol, and therefore do probably not influence CHD risk compared to carbohydrate consumption, based on this mechanism. Studies on the direct relation between diet and CVD are a more reliable source of information than studies of the effects on biomarkers, because the chance of missing effects from known or unknown markers for disease risk is excluded in direct relations.

|Reference: Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. Am J Clin Nutr. 2003 May;77(5):1146-55. Full text|

Studies about the direct relationship between dairy and CVD.

No randomized trial exist of the relation between diary (product) consumption and CVD. Though some intervention trials exist in which individuals were randomized to low-fat instead of high-fat dairy (e.g., "No authors listed" 1968; Frantz ID Jr 1989), but dairy fat was not the only variable that changed. The interventions also included changes in other variables (e.g., meat consumption, addition of vegetable oils).

When no data from randomized trials is available, the next best line of evidence consists of data from prospective (= cohort) studies (Sunny Downstate Medical Center). Preferably in the form of a systematic review of the literature. In the last few years, a few systematic reviews were published:

• One systematic review found no association between milk consumption and CHD, based on the Bradford Hill Criteria for assessing a potential causal relationship (Mente A 2009).
• Another systematic review also found no consistent evidence that dairy food consumption was associated with CHD (Gibson RA 2009).
• But a very recent systematic review found a significant protective effect of high milk/dairy intake against heart disease: 0.92 (0.80-0.99). (Elwood PC 2010).

The most recent systematic review, on the relation between milk/dairy intake and CHD, found a significant protective effect of high dairy/milk intake. Hardly any association was found in any individual cohort included in this analysis, but meta-analysis did show an association (Elwood PC 2010).
In my own systematic review, I included data from 4 different cohorts in which increased risks were found (See Menu). My review includes the following cohorts:

• Fraser GE (1994) RH = 1.33 (P = < 0.07).
• Kelemen LE (2005) RR = 1.41 (1.07-1.87; P = 0.02).
• Knekt P (1994) A significantly increased risk (P = < 0.001).
• Buckland G (2009) HR = 1.51 (1.21-1.89; P = < 0.001).

The first 2 cohorts were found by Elwood. But they were omitted because of an absence of detailed data (Fraser GE), and because it was a test of dietary substitution (Kelemen LE).
The third study was not included, probably because of a lack of specific data (Knekt P). And The fourth study was not included, probably because it was published after the literature review was completed (Buckland G).

Conclusion: 2 out of 3 recent systematic reviews on the relation between dairy/milk and CHD show no association with this disease. The most recent review missed data from 4 cohorts in which increased risks were found. Though plausible reasons were given/suggested for not using these cohorts, including them - and other cohorts missed after the literature search by Elwood - in the analysis would shift the RR closer to 1, showing little effect (RR's for total dairy would become 1.10, and 1.13 for CHD risk, and CHD mortality, respectively. And RR's for milk would become 0.74, and 0.96 for CHD risk, and CHD mortality, respectively).

|References:
-Buckland G, González CA, Agudo A, Vilardell M, Berenguer A, Amiano P. Adherence to the Mediterranean diet and risk of coronary heart disease in the Spanish EPIC Cohort Study. Am J Epidemiol. 2009 Dec 15;170(12):1518-29. Abstract
-Elwood PC, Pickering JE, Givens DI, Gallacher JE. The Consumption of Milk and Dairy Foods and the Incidence of Vascular Disease and Diabetes: An Overview of the Evidence. Lipids. 2010 Apr 16. [Epub ahead of print] Abstract
-Frantz ID Jr, Dawson EA, Ashman PL, Gatewood LC, Bartsch GE, Kuba K. Test of effect of lipid lowering by diet on cardiovascular risk. The Minnesota Coronary Survey. Arteriosclerosis. 1989 Jan-Feb;9(1):129-35. Full text
-Fraser GE. Diet and coronary heart disease: beyond dietary fats and low-density-lipoprotein cholesterol. Am J Clin Nutr. 1994 May;59(5 Suppl):1117S-1123S. Full text
-Gibson RA, Makrides M, Smithers LG, Voevodin LG, Voevodin M, Sinclair AJ. The effect of dairy foods on CHD: a systematic review of prospective cohort studies. Br J Nutr. 2009 Nov;102(9):1267-75. Link.
-Kelemen LE, Kushi LH, Jacobs DR Jr, Cerhan JR. Associations of dietary protein with disease and mortality in a prospective study of postmenopausal women. Am J Epidemiol. 2005 Feb 1;161(3):239-49. Full text
-Knekt P, Reunanen A, Järvinen R, Seppänen R, Heliövaara M, Aromaa A. Antioxidant vitamin intake and coronary mortality in a longitudinal population study. Am J Epidemiol. 1994 Jun 15;139(12):1180-9. Abstract
-Mente A, de Koning L, Shannon HS, Anand SS. A systematic review of the evidence supporting a causal link between dietary factors and coronary heart disease. Arch Intern Med. 2009 Apr 13;169(7):659-69. Full text
-No author listed. Controlled trial of soya-bean oil in myocardial infarction. Lancet. 1968 Sep 28;2(7570):693-9. Abstract
-Sunny Downstate Medical Center. Guide to Research Methods. The Evidence Pyramid. Link|

Results from the systematic review. Overview of the effects stratified by continent.

The frequency at which an effect was found, was evaluated.
The tables below show the frequency at which a (non)significant effect was found at high vs low consumption of the different variables.
For more specific data, You are referred to the menu.