The Effect of a Low-fat, Plant-based Lifestyle Intervention (CHIP) on Serum HDL Levels and the Implications for Metabolic Syndrome Status

A Cohort Study

Lillian Kent; Darren Morton; Paul Rankin; Ewan Ward; Ross Grant; John Gobble; Hans Diehl


Nutr Metab. 2013;10(58) 

In This Article


The most striking observation of the present study is that when people move towards a low-fat, plant-based diet, HDL levels tend to decrease while all other measures of cardiovascular risk improve. This large study is the first to highlight the implications of this effect for MetS classification. Furthermore, the findings raise questions about the usefulness of measuring HDL levels for predicting cardiovascular risk, especially among specific groups such as vegetarian populations.

Further evidence that the cardiovascular risk of the participants in this study improved despite the trend for HDL to decrease, is seen in the improvements in the overall TC: HDL and LDL: HDL ratios. A meta-analysis of 61 prospective studies with vascular deaths as an endpoint suggested that the TC: HDL ratio was more predictive than either HDL or non-HDL cholesterol sub-fractions, and two times more predictive than TC.[14] However, in the present study, this ratio, as well as the ratio of LDL: HDL, significantly increased among participants who were newly characterised with MetS and low HDL levels. In addition, there were significant reductions in the number of participants who met the MetS criteria for BMI, BP, TG and FPG levels (11–29%), as compared to the 30% increase in the number of participants who meet the HDL criterion. These results show that in the context of interventions that emphasise a low-fat, plant-based eating pattern, changes in HDL levels may be misleading, whether measured as an absolute value or calculated as a ratio of TC or LDL.

The consistently strong inverse association between low HDL levels and the risk of cardiovascular events observed in epidemiological studies,[15,16] has traditionally been explained by its role in reverse cholesterol transport (RCT), also known as cholesterol efflux.[16] Notwithstanding this role, HDL more recently has been shown to have many anti-atherogenic properties, which include anti-inflammatory, anti-apoptotic, nitric oxide promoting, prostacyclin-stabilizing, and platelet-inhibiting functions.[17,18] Indeed, it has been postulated that the anti-inflammatory properties of HDL and its ability to protect LDL from oxidation may be just as important as its role in RCT.[19]

Despite the documented anti-atherogenic properties of HDL, there is evidence that questions the relationship between HDL levels and risk of cardiovascular events, as many individuals who suffer coronary atherosclerotic events have normal or even elevated HDL levels.[18,20] Specifically, the Framingham study demonstrated that more than 40% of cardiac events occurred in men and women with normal HDL levels.[20] In addition, other studies have shown that when HDL levels are raised pharmacologically, the elevated levels do not always correlate with reduced risk of coronary heart disease.[21,22]

The result of this study, which suggests that HDL levels may not be helpful for predicting cardiovascular risk in individuals consuming a low-fat, plant-based diet, is supported by other epidemiological and clinical studies. Over 30 years ago, Connor observed that the Tarahumara Indians of Mexico, who consumed a largely plant-based diet comprising approximately 12% fat, 13% protein (predominantly from corn and beans) and 85% carbohydrate, had very low rates of vascular disease and blood lipids, including HDL.[23] However, blood lipids, including HDL, were observed to significantly increase after only five weeks when their traditional diet was changed to a Western diet.[24] It was argued that the increase in HDL was the "normal response to a high-fat diet" and that low-HDL in concert with low-LDL in a low-fat diet is associated with a low risk of coronary disease. Other epidemiological findings also show that individuals who consume a plant-based diet are at lower risk of CVD and type 2 diabetes mellitus, despite having lowered HDL levels.[19,25] In addition, the Lifestyle Heart Trial,[7] which incorporated a plant-based diet with less than 10% fat, showed a 7.9% improvement in measured coronary artery percent diameter stenosis after five years despite a 13% reduction in HDL. Similarly, individuals with diagnosed CVD and a recommendation for bypass surgery who participated in the Pritikin residential program, which recommends a plant-based diet (10% fat), experienced a 16% reduction in HDL but decreases in symptomatic angina.[26] These patients averted surgery for more than five years after program entry despite sustained lowered HDL levels.

The value of increasing HDL levels has been further questioned as its range of functions has become better understood.[11,17,18] With its heterogeneous structure and subtypes, varying functions are now being demonstrated.[18] To date, a number of apolipoproteins and antioxidant enzymes have been identified in the HDL structure that may explain its anti-inflammatory (e.g. ApoA-1 and paraoxonase 1) and inflammatory properties (e.g. ApoA-II and ApoC-III).[16,18,27] Although the mechanisms governing HDL inflammatory and anti-inflammatory ratios are yet to be fully elucidated, these may at least in part be influenced by the presence of oxidized lipids and oxidants, which inhibit or directly damage the anti-inflammatory molecules on HDL.[27] Noteworthy, antioxidants and phytonutrients abundant in plant foods may increase the activity of HDL enzymes or counter the adverse effect of oxidants on apoA-1 and/or the pro-inflammatory effect on LDL lipids.[28] Given that people following a low-fat, plant-based diet typically have lowered levels of serum TC and LDL, the need for elevated HDL levels may be diminished from an RCT perspective. Indeed, lifestyle factors have been shown to influence the subpopulations of HDL.

Additional evidence against the role of HDL in CVD risk has come from genetic studies, designed to examine the associations of LDL, HDL and TG with CVD risk.[29,30] Early studies showed that rare mutations that encode LCAT (lecithin cholesterol acyl transferase) and ABCA1 (ATP-binding cassette transporter, also known as cholesterol efflux regulatory protein (CERP)), which profoundly reduce HDL levels, were inconsistently associated with CVD risk.[29] Subsequent Mendelian randomization (MR) analyses examined several DNA variants that affect HDL levels and found no associations with major adverse CVD end points and risk of myocardial infarction.[29] In contrast, genetics variants that raise LDL levels significantly increased CVD risk. More recent MR analyses, using carotid intima-media thickness (CIMT) as a surrogate for atherosclerosis, confirmed a strong relationship between LDL and CIMT, but not with HDL and TG.[30]

There is also growing evidence that lifestyle interventions may be able to modulate the inflammatory or anti-inflammatory properties of HDL. In patients at risk of CVD, the anti-inflammatory properties of HDL improved following lifestyle modification, despite reductions in HDL.[28] In another study, the HDL shifted from pro-inflammatory to anti-inflammatory in obese men, with MetS, who underwent a three-week intervention involving a low-fat, high- fibre diet and exercise.[19] More specifically, consumption of saturated fat reduces the anti-inflammatory potential of HDL, but consumption of polyunsaturated fat has been shown to increase it.[31] The regulation and function of HDL appears more complex than originally thought-- although high HDL levels are associated with reduced CVD at a population level, at an individual level HDL function may be more important than the actual HDL levels.[32]