Beneficial Effects of Dairy Fat Intake?
Dairy Fat Characteristics
Dairy fat contains approximately two-thirds of SFA, with the majority in the range of 4:0 to 18:0, and the unsaturated fat mostly represented by 18:1n-9. Small amounts of CLA are present but levels are probably too low to affect metabolic risk.[108] Ruminant milk also contains minor amounts (<5%) of TFA, mostly trans-vaccenic acid (trans-18:1n-7), and the health effects from these are under debate at present. According to adverse effects on blood lipid profile, TFA from dairy fat seems equal to industrial TFA with regard to CVD risk,[109] although dairy TFA has a greater effect on raising LDL-C and industrial TFA has a greater effect on lowering HDL-C.[68] Plausibly, there is also heterogeneity among different TFA, as indicated in a recent cohort study where trans- 16:1n-7, solely produced in ruminants and strongly associated with whole-fat dairy intake, was also associated with lower diabetes risk.[110] Due to the complex composition of nutrients in bovine milk and limited evidence for harmful metabolic effects from finite amounts of dairy intake,[111] dairy has been suggested as part of a healthy diet, albeit with its high SFA content.[42] Low-fat dairy is also included in the Dietary Approaches to Stop Hypertension (DASH) diet (also high in fruit and vegetables, whole grains, nuts and legumes, and lower in processed and red meats, sweets and salt) and current recommendations for dietary therapy in hypertension,[112] as a result of data indicating that low-fat dairy foods could decrease blood pressure.[113] Low-fat dairy is also a part of several food patterns associated with reduced mortality and CVD risk. Mechanisms for possible beneficial effects are not established but dairy content of calcium and vitamin D has been suggested to regulate adipocyte fatty acid metabolism and reduce adiposity.[114] Absorption and/or excretion of intestinal fatty acids and bile acids may be affected by dietary calcium,[115] and whey (specifically leucin) content in milk may have an angiotensin-converting enzyme-inhibiting effect and possibly inhibit 11β-HSD1, which has been considered a link between glucocorticoids and visceral adiposity.[114] Some fermented milk may even lower LDL-C levels and blood pressure,[116] whereas cream intake has been demonstrated to induce postprandial oxidative stress, endotoxin and TLR 4 expression,[117] besides increasing fasting blood lipoproteins. The rather high proportion of 16:0 in bovine milk is a potential caveat. It is primarily located in the sn-2 position of TGs, which has been demonstrated to have nutritional advantages, such as increasing the intestinal absorption of calcium,[42] but may concomitantly have a stronger cholesterol-raising effect compared with vegetable fat, in which there is no cholesterol and 16:0 is mainly in the sn-1 and sn-3 positions.[68] This distinction also applies for 14:0, located to approximately 50% in sn-2 position in dairy but primarily in sn-1 or sn-3 positions in coconut and palm kernel oil.
Observational Evidence
In a Swedish, cross-sectional study of 70-year-old men, 15:0 serum cholesterol ester levels were associated with dairy fat intake, and inversely associated with BMI, waist circumference, lipoproteins and fasting plasma glucose.[118] In another Swedish prospective, case–control study in men and women, 15:0 and 17:0 were associated with lower risk of first MI, especially in women.[119] This was partly confirmed with inverse associations between MI and cheese and fermented milk intake.[119] This is in line with a recent meta-analysis of milk and dairy intake, suggesting a slight but reduced risk for overall mortality, CHD and Type 2 diabetes.[120] A recent Australian prospective cohort study on dairy intake confirms this inverse relationship with CVD mortality, but did not include butter.[121] Notably, it is unknown whether such inverse relationships are explained by any of the different SFA found in dairy foods or if it merely reflects SFA-independent relationships (i.e., other unknown 'milk factors' or nutrients) or associated lifestyle habits (residual confounding factors). One cautionary note is the observed associations with increased risk of CVD and peripheral vascular disease from butter and cream intake in retrospective case–control studies on dietary patterns,[122,123] although not confirmed in prospective cohort studies.[120] Different dairy products may also exert different effects. For instance, cheese has a lower cholesterol-raising effect than butter[124] and has been suggested to have an overall neutral effect on CHD risk,[116] although evidence is limited. Since the fatty acid compositions among different dairy products are similar, other components may explain these differences concerning metabolic effects (e.g., fermentation) and total fat content. In a recent cohort study, butter and dairy intake did not predict all-cause and ischemic heart disease mortality in men, and slightly increased risk in women, whereas fermented full-fat milk was inversely associated with mortality in both men and women.[125] Further research is needed to elucidate these relationships.
Divergent Evidence from Randomized Trials
Despite inconsistent evidence from observational studies regarding harmful effects of dairy consumption on cardiometabolic risk, there is robust evidence from feeding studies to suggest that replacing dairy fat with unsaturated vegetable fat (e.g., rapeseed [canola] oil) improves blood lipid profile.[50,51] This suggests that relationships between dairy intake and CVD from observational studies are not in accordance with short-term RCTs using cardiovascular risk markers (where, however, a risk for publication bias may exist), nor with long-term RCTs, which often included significant amounts of dairy fat and suggest that replacing SFA with PUFA reduces CVD risk.[85] The seemingly different metabolic effects from butter and fermented or low-fat milk, for example, may contribute to this inconsistency and also possibly dilute the associations between total SFA and CVD in epidemiological studies. Another possibly overlooked confounding factor is the growing use of milk in combination with coffee. In Sweden, this phenomenon is increasingly prevalent and, to date, more than a quarter of all milk is consumed with coffee. Coffee has several established metabolic effects and is inversely related to diabetes incidence in epidemiological studies. Thus, in populations where such habits are prevalent, we suggest that coffee intake should be taken into account and probably be adjusted for in future epidemiological studies on dairy intake and cardiometabolic outcomes. In summary, high-fat dairy (e.g., butter) is to be considered as a less than optimal food choice for reducing CVD risk, which is especially relevant for persons with increased CVD risk (e.g., in secondary prevention). However, observational biomarker studies indicate a decreased risk from dairy intake. Dietary recommendations should consider all available knowledge and be regularly scrutinized and updated.
Clin Lipidology. 2011;6(2):209-223. © 2011 Future Medicine Ltd.
