What is the question?

• The question is: What is the relationship between types of dietary fat consumed and risk of cardiovascular disease?

What is the answer to the question?

Cardiovascular disease intermediate outcomes (blood lipids and blood pressure): Children

• Strong evidence demonstrates that diets lower in saturated fatty acids and cholesterol during childhood result in lower levels of total blood and low-density lipoprotein cholesterol throughout childhood, particularly in boys. 
• Moderate evidence indicates that diets higher in polyunsaturated fatty acids during childhood result in lower levels of total blood cholesterol throughout childhood, particularly in boys. 
• Insufficient evidence is available to determine the relationship between monounsaturated fatty acid intake during childhood and total blood and low-density lipoprotein cholesterol throughout childhood. 
• Insufficient evidence is available to determine the relationship between intake of types of dietary fat during childhood and blood pressure throughout childhood. 

Cardiovascular disease endpoint outcomes (cardiovascular disease event, diagnosis, or mortality): Children

• Insufficient evidence is available to determine the relationship between intake of types of dietary fat during childhood and cardiovascular disease health outcomes during adulthood. 

Cardiovascular disease intermediate outcomes (blood lipids): Adults

• Strong and consistent evidence from randomized controlled trials demonstrates that replacing saturated fatty acids with unsaturated fats, especially polyunsaturated fatty acids, in adults significantly reduces total and low-density lipoprotein cholesterol. Replacing saturated fatty acids with carbohydrates (sources not defined) also reduces total and low-density lipoprotein cholesterol, but significantly increases triglycerides and reduces high-density lipoprotein cholesterol. Since the 2015 Dietary Guidelines Advisory Committee review, evidence remains inadequate to differentiate among sources of carbohydrate and their impact on blood lipids. 
• Insufficient evidence is available to determine an independent relationship between dietary cholesterol intake in adults and blood lipids, given the co-occurrence of cholesterol with saturated fats in foods. 

Cardiovascular disease endpoint outcomes (cardiovascular disease event, diagnosis, or mortality): Adults

• Strong evidence demonstrates that replacing saturated fatty acids with polyunsaturated fatty acids in adults reduces the risk of coronary heart disease events and cardiovascular disease mortality. 
• Insufficient evidence is available to determine whether replacing saturated fatty acids with polyunsaturated fatty acids in adults affects the risk of stroke or heart failure. 
• Insufficient evidence is available to determine whether replacing saturated fatty acids with different types of carbohydrates (e.g., complex, simple) in adults affects the risk of cardiovascular disease. 
• Limited evidence is available regarding whether replacing saturated fatty acids with monounsaturated fatty acids in adults confers overall cardiovascular disease endpoint health benefits. Main sources of monounsaturated fatty acids in a typical American diet are animal fats, with co-occurrence of saturated fatty acids and monounsaturated fatty acids in these foods thereby obscuring the independent association of monounsaturated fatty acids with cardiovascular disease. Evidence reviewed from randomized controlled trials and prospective studies demonstrated benefits of plant sources of monounsaturated fats, including olive oil and nuts on cardiovascular disease risk. 
• Moderate evidence indicates that total intake of omega-3 polyunsaturated fatty acids, particularly eicosapentaenoic acid and docosahexaenoic acid from food sources, by adults is associated with lower risk of cardiovascular disease. 
• Limited evidence suggests that intake of linoleic acid, but not arachidonic acid, during adulthood may be associated with lower risk of cardiovascular disease, including cardiovascular disease mortality. 
• Insufficient evidence is available from randomized controlled trials to quantify an independent relationship between dietary cholesterol intake in adults and overall risk of cardiovascular disease. 

Why was this question asked?

• This important public health question was identified by the U.S. Departments of Agriculture (USDA) and Health and Human Services (HHS) to be examined by the 2020 Dietary Guidelines Advisory Committee.

How was this question answered?

• The 2020 Dietary Guidelines Advisory Committee, Dietary Fats and Seafood Subcommittee conducted a systematic review to answer this question with support from the Nutrition Evidence Systematic Review (NESR) team.

What is the population of interest? 

• For the intervention or exposure, generally healthy children, age 18 years and younger, and adults, age 19 years and older. For the outcome, those who had blood pressure measured during childhood, blood lipids measured during childhood or adulthood, and/or had CVD endpoint outcomes measured during adulthood.    

What evidence was found? 

• This review identified 228 articles that met inclusion criteria: 37 in children and 191 in adults.
• Children:
  • Most studies found that lower childhood intake of saturated fatty acids (SFA) and dietary cholesterol, as well as higher intake of polyunsaturated fatty acids (PUFA), were beneficial for total blood cholesterol and/or low-density lipoprotein (LDL) cholesterol. Results were more consistent in boys. The 2020 Committee determined this evidence was strong for SFA and dietary cholesterol and moderate for PUFA. 
  • Few studies examined monounsaturated fatty acid (MUFA) intake during childhood and its relationship to blood lipids and a conclusion could not be drawn.
  • Few studies examined types of dietary fat during childhood and either blood pressure or risk of CVD and a conclusion could not be drawn. 
• Adults: 
  • Most studies reported a beneficial effect of replacing SFA with MUFA or PUFA on total and LDL cholesterol. Most studies did not find an effect of replacing SFA with MUFA or PUFA on high-density lipoprotein (HDL) cholesterol or triglycerides. 
  • Many studies found that replacement of SFA with PUFA was related to lower risk of CVD mortality and CHD. No consistent relationship between replacement of SFA with MUFA or carbohydrates and risk of CVD was found.
  • The 2020 Committee could not draw a conclusion about replacement of SFA with PUFA and heart failure or stroke due to a lack of evidence. 
  • Several studies observed a beneficial relationship between higher omega-3 PUFA intake and lower risk of CVD, most consistently for eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). 
  • Few studies assessed the relationship between omega-6 PUFA intake and risk of CVD, tending to observe benefits with higher intake of linoleic acid but not with arachidonic acid.
  • The 2020 Committee could not draw a conclusion about the relationship between dietary cholesterol intake and blood lipids or risk of CVD  due to a small number of studies and inconsistent results. Additionally, dietary cholesterol is commonly found in foods that also contain SFA, making it difficult to assess dietary cholesterol independently.
  • This systematic review builds and expands on the work of the 2015 Committee. The 2020 Committee concurs with and updates the conclusions drawn by the 2015 Committee.

How up-to-date is this systematic review?

• This review searched for studies in children from January 1990 to October 2019 and studies in adults from January 2010 to October 2019.

Background

• This important public health question was identified by the U.S. Departments of Agriculture (USDA) and Health and Human Services (HHS) to be examined by the 2020 Dietary Guidelines Advisory Committee.
• The 2020 Dietary Guidelines Advisory Committee, Dietary Fats and Seafood Subcommittee conducted a systematic review to answer this question with support from the Nutrition Evidence Systematic Review (NESR) team.
• The goal of this systematic review was to examine the following question: What is the relationship between types of dietary fat consumed and risk of cardiovascular disease?

Conclusion statements and grades

Cardiovascular disease intermediate outcomes: Children

• Strong evidence demonstrates that diets lower in saturated fatty acids and cholesterol during childhood result in lower levels of total blood and low-density lipoprotein cholesterol throughout childhood, particularly in boys. (Grade: Strong)
• Moderate evidence indicates that diets higher in polyunsaturated fatty acids during childhood result in lower levels of total blood cholesterol throughout childhood, particularly in boys. (Grade: Moderate)
• Insufficient evidence is available to determine the relationship between monounsaturated fatty acid intake during childhood and total blood and low-density lipoprotein cholesterol throughout childhood. (Grade: Grade not assignable)
• Insufficient evidence is available to determine the relationship between intake of types of dietary fat during childhood and blood pressure throughout childhood. (Grade: Grade not assignable)

Cardiovascular disease endpoint outcomes: Children

• Insufficient evidence is available to determine the relationship between intake of types of dietary fat during childhood and cardiovascular disease health outcomes during adulthood. (Grade: Grade not assignable)

Cardiovascular disease intermediate outcomes: Adults

• Strong and consistent evidence from randomized controlled trials demonstrates that replacing saturated fatty acids with unsaturated fats, especially polyunsaturated fatty acids, in adults significantly reduces total and low-density lipoprotein cholesterol. Replacing saturated fatty acids with carbohydrates (sources not defined) also reduces total and low-density lipoprotein cholesterol, but significantly increases triglycerides and reduces high-density lipoprotein cholesterol. Since the 2015 Dietary Guidelines Advisory Committee review, evidence remains inadequate to differentiate among sources of carbohydrate and their impact on blood lipids. (Grade: Strong)
• Insufficient evidence is available to determine an independent relationship between dietary cholesterol intake in adults and blood lipids, given the co-occurrence of cholesterol with saturated fats in foods. (Grade: Grade not assignable)

Cardiovascular disease endpoint outcomes: Adults

• Strong evidence demonstrates that replacing saturated fatty acids with polyunsaturated fatty acids in adults reduces the risk of coronary heart disease events and cardiovascular disease mortality. (Grade: Strong)
• Insufficient evidence is available to determine whether replacing saturated fatty acids with polyunsaturated fatty acids in adults affects the risk of stroke or heart failure. (Grade: Grade not assignable)
• Insufficient evidence is available to determine whether replacing saturated fatty acids with different types of carbohydrates (e.g., complex, simple) in adults affects the risk of cardiovascular disease. (Grade: Grade not assignable)
• Limited evidence is available regarding whether replacing saturated fatty acids with monounsaturated fatty acids in adults confers overall cardiovascular disease endpoint health benefits. Main sources of monounsaturated fatty acids in a typical American diet are animal fats, with co-occurrence of saturated fatty acids and monounsaturated fatty acids in these foods thereby obscuring the independent association of monounsaturated fatty acids with cardiovascular disease. Evidence reviewed from randomized controlled trials and prospective studies demonstrated benefits of plant sources of monounsaturated fats, including olive oil and nuts on cardiovascular disease risk. (Grade: Limited)
• Moderate evidence indicates that total intake of omega-3 polyunsaturated fatty acids, particularly eicosapentaenoic acid and docosahexaenoic acid from food sources, by adults is associated with lower risk of cardiovascular disease. (Grade: Moderate)
• Limited evidence suggests that intake of linoleic acid, but not arachidonic acid, during adulthood may be associated with lower risk of cardiovascular disease, including cardiovascular disease mortality. (Grade: Limited)
• Insufficient evidence is available from randomized controlled trials to quantify an independent relationship between dietary cholesterol intake in adults and overall risk of cardiovascular disease. (Grade: Grade not assignable) 

Methods 

• Two literature searches were conducted using 4 databases (PubMed, Cochrane, Embase, CINAHL) to identify articles that evaluated the intervention or exposure of intake of types of dietary fat and the outcomes of cardiovascular disease (CVD) intermediate outcomes (blood lipids in children and adults; blood pressure in children) and CVD endpoint outcomes. Because this systematic review built on evidence reviewed by the 2015 Dietary Guidelines Advisory Committee, one search was designed to identify articles in children from January 1990 to December 2009 and the other search was designed to identify articles in children and adults from January 2010 to October 2019. A manual search was conducted to identify articles that may not have been included in the electronic databases searched. Articles were screened by two NESR analysts independently for inclusion based on pre-determined criteria. 
• Data extraction and risk of bias assessment were conducted for each included study, and both were checked for accuracy. The Committee qualitatively synthesized the body of evidence to inform development of conclusion statements, and graded the strength of evidence using pre-established criteria for risk of bias, consistency, directness, precision, and generalizability.

Summary of the evidence

Children

• This systematic review includes 37 articles, 22 articles from 7 randomized controlled trials (RCTs) and 16 articles from 14 prospective cohort studies (PCSs), published between January 1990 and October 2019 that examined the relationship between intake of types of dietary fat during childhood and cardiovascular disease (CVD) risk. (Note: One article from an RCT was also analyzed as a PCS.)
  • The RCTs modified child fat intake either through dietary counseling that focused primarily on reducing saturated fatty acids (SFA) and dietary cholesterol intake, with additional encouragement to increase polyunsaturated fatty acid (PUFA) intake, or through provision of food products (i.e., eggs, extra virgin olive oil, or oily fish) that differed in types of fat including SFA, monounsaturated fatty acids (MUFA), PUFA, and/or dietary cholesterol.
  • The PCSs primarily assessed SFA or PUFA intake, with fewer studies on MUFA or dietary cholesterol intake; only two studies modeled replacement between different types of fat or macronutrients.
• Most included studies assessed the relationship between intake of types of dietary fat during childhood and blood lipids.
  • Evidence from RCTs predominantly indicated that consuming less SFA and dietary cholesterol resulted in lower blood total cholesterol and low-density lipoprotein (LDL) cholesterol throughout childhood, particularly in boys; evidence from PCSs was consistent with the RCTs.
  • Although reduction of SFA intake was the primary focus of most RCTs, evidence from these RCTs also showed that higher PUFA intake resulted in decreased total blood cholesterol, particularly in boys; evidence from PCSs was broadly consistent with the RCTs. 
  • Few studies, RCTs or PCSs, focused on the relationship between MUFA intake and blood lipids and the results were predominantly null.
  • The majority of studies assessed blood lipids during childhood; few assessed intake of types of fat during childhood and blood lipids into early adulthood.
• Fewer studies assessed the relationship between intake of types of dietary fat during childhood and blood pressure.
  • It was difficult to discern the effect of consuming different types of fat in the RCT that contributed the most evidence due to additional advice to reduce sodium consumption.
  • Few PCSs were conducted on this topic and results were predominantly null.
• Only one study included in this review assessed the relationship between intake of types of dietary fat during childhood and CVD endpoint outcomes and methodological limitations related to the dietary assessment confounded interpretation of results. Therefore, no conclusion could be drawn.
• Limitations of this body of evidence:
  • Most articles did not report race and ethnicity, but those that did included predominantly White or Caucasian participants.
  • Some studies specifically recruited children with elevated or higher than average blood lipid levels, reducing generalizability.
  • RCTs had predominantly low risk of bias, but few pre-registered their analysis intentions and several RCTs did not provide information on allocation of randomization sequences.
  • Although many PCSs accounted for most or many key confounders, all PCSs did not account for at least one key confounder.
  • Approximately half of the diet assessment methods used in the PCSs were not validated; many PCSs had high attrition rates and did not provide information on those lost to follow-up.

Adults

Adults: CVD intermediate outcomes

• This systematic review includes 97 articles that examined the relationship between intake of types of dietary fat during adulthood and CVD intermediate outcomes, published between January 2010 and October 2019. Of these, 47 were from 47 parallel design RCTs, 46 were from 44 crossover design RCTs, and 5 were from non-randomized controlled trial designs. (Note: One parallel design RCT was also analyzed as a crossover design RCT.)  
  • The articles examined intake of SFA, MUFA, PUFA, and dietary cholesterol.
  • The majority of articles specifically examined types of fat from different food sources, including food sources that are predominantly fat (e.g., butter and olive oil). 
• The relationship between types of dietary fat and blood lipids varied by the type of fat examined and the comparator.
  • SFA intake: Predominantly null effects were reported for SFA intake when replacement was not considered or when SFA was partially replaced by carbohydrates. However, among the studies that detected significant effects, all reported significantly higher total, LDL, and high-density lipoprotein (HDL) cholesterol with higher intake of SFA, compared to either lower intake of SFA or substitution with carbohydrate. 
  • Replacement of SFA with MUFA: More than half of articles reported a beneficial effect of replacing a portion of SFA intake with MUFA intake on total and LDL cholesterol. Predominantly null effects were reported for HDL cholesterol and triglycerides.
  • Replacement of SFA with PUFA: More than half of articles reported a beneficial effect of replacing a portion of SFA intake with PUFA intake on total and LDL cholesterol. Predominantly null effects were reported for HDL cholesterol and triglycerides.
  • MUFA intake: Predominantly null effects were reported for MUFA intake when replacement was not considered or when MUFA was partially replaced with carbohydrates. 
  • Replacement of MUFA with PUFA: Predominantly null effects were reported in articles that examined partial replacement of MUFA intake with PUFA intake. 
    • Among the studies that found significant effects, the majority detected significantly lower levels of or significant decreases in total and LDL cholesterol when PUFA intake replaced a portion of MUFA intake. 
    • Studies replacing a portion of MUFA intake with PUFA intake predominantly detected significant decreases or greater decreases in HDL cholesterol and significantly higher levels of, smaller decreases in, or greater increases in triglycerides.
  • PUFA intake: The vast majority of articles that assessed the effect of PUFA intake (without considering replacement) on total, LDL, or HDL cholesterol and triglycerides were null. However, among the few articles that detected significant effects, total and LDL cholesterol were significantly lower with greater PUFA intake, compared with lower PUFA intake; HDL cholesterol significantly increased or had smaller decreases with greater PUFA intake; and triglycerides were significantly lower or had greater decreases with greater PUFA intake.
• Few articles published during the search years of the present review assessed the relationship between dietary cholesterol intake and blood lipids. 
  • Predominantly null effects were reported for dietary cholesterol. However, among the few articles that found significant results, higher intake of dietary cholesterol, compared to lower intake, significantly increased or resulted in higher levels of total, LDL, and HDL cholesterol. 
  • In several articles, it was not possible to isolate the independent effect of dietary cholesterol on blood lipids due to simultaneous changes in the total amount of fat or proportion of different types of fatty acids in the study diet. 
• Limitations of this body of evidence:
  • Several articles involved small sample sizes and lacked sufficient power.
  • Race or ethnicity was not consistently reported, but among those studies that provided this information, the majority included participants who were predominantly White or Caucasian. 
  • It was not possible to isolate the independent effect of SFA, MUFA, or PUFA on blood lipids in several articles due to simultaneous changes of those three types of fat.
  • The majority of articles did not control for other dietary components beyond the intervention or test meal.
• This systematic review builds and expands on the work of the 2015 Dietary Guidelines Advisory Committee, which answered the question “What is the relationship between intake of saturated fat and risk of cardiovascular disease?” and considered evidence from RCTs and PCSs from the 1960s to 2010.  This systematic review concurs with and updates the conclusions drawn by the 2015 Dietary Guidelines Advisory Committee.

Adults: CVD endpoint outcomes

• This systematic review includes 94 articles that examined the relationship between intake of types of dietary fat during adulthood and CVD endpoint outcomes, published between January 2010 and October 2019. Of these, 90 were from 47 PCSs and 4 were from 3 nested case-control studies.
  • The articles primarily examined SFA, total PUFA, omega-3 (n-3) PUFA (including alpha-linolenic acid [ALA], eicosapentaenoic acid [EPA], docosahexaenoic acid [DHA], or docosapentaenoic acid [DPA]), or MUFA intake. Fewer articles examined omega-6 (n-6) PUFA (including linoleic acid [LA] or arachidonic acid [AA]) or dietary cholesterol intake.
  • Several articles modeled replacement between different types of fats or macronutrients or, in some cases, between different sources of the same type of fat. Few articles specifically assessed food sources that are predominantly fat (i.e., butter and olive oil) or types of fat from different food sources. 
• The relationship between types of dietary fat and CVD endpoint outcomes varied by the type of fat examined and the specific outcome assessed, with the most consistent results observed when replacement was modeled.
  • Replacement of SFA with PUFA: In this review, replacement of SFA with PUFA (predominantly total PUFA) in many studies was associated with significantly lower risk of CVD mortality and/or coronary heart disease (CHD) or associations were null. Fewer articles in this review reported data regarding the relationship between replacement of SFA with PUFA and other specific types of CVD including heart failure or stroke, and results were predominantly null.
  • Replacement of SFA with carbohydrates: In this review, replacement of SFA with carbohydrates and CVD outcomes were predominantly null. Most articles did not specify or differentiate between the types of carbohydrate replacing SFA (e.g., complex or simple carbohydrates/sugar).
  • Replacement of SFA with MUFA: In this review, predominantly null associations were observed between replacement of SFA with MUFA and total CVD and CHD. However, among the few articles that differentiated plant and animal sources, MUFA from plants tended to be associated with lower risk.
• In addition to articles that reported on total PUFA intake, many specifically assessed n-3 PUFA and some assessed n-6 PUFA.
  • Total n-3 PUFA: Predominantly null or beneficial associations were observed between total n-3 PUFA intake and CVD outcomes.
  • Types of n-3 PUFA: When “long chain” n-3 PUFA (EPA, DHA, and sometimes DPA), primarily from marine sources, were assessed separately from ALA, more consistent associations with lower risk of CVD were observed.
  • Total n-6 PUFA: Associations between total n-6 PUFA intake and CVD were predominantly null.
  • Types of n-6 PUFA: In the few articles specifically assessing LA and AA separately, beneficial associations were more often observed for LA as compared to AA.
• Few articles, with inconsistent results, assessed the independent relationship between dietary cholesterol intake and CVD endpoint outcomes, thereby further confounding meaningful conclusions. Due to the co-occurrence of dietary cholesterol and SFA in animal source foods, disentangling independent associations between dietary cholesterol and CVD endpoint outcomes in these observational studies is challenging.
• Limitations of this body of evidence:
  • Many articles did not report race or ethnicity, but the majority of those that did involved participants who were predominantly White or Caucasian. Other important characteristics of participants in some included articles did not mirror those of the U.S. population, such as BMI and diet at baseline.
  • Although many articles accounted for the majority of key confounders, few accounted for all key confounders.
  • Some studies did not use validated dietary assessment methods or were limited by high attrition.
  • Although most studies included in this body of evidence were specifically designed to evaluate the relationship between diet and CVD, several articles were less direct as a result of being secondary analyses of RCTs or cohorts originally designed to assess outcomes other than CVD.
• This systematic review builds upon the work of the 2015 Dietary Guidelines Advisory Committee, which answered the question “What is the relationship between intake of saturated fat and risk of cardiovascular disease?” and considered evidence from RCTs and PCSs from the 1960s to 2010. 
  • Regarding the relationships between replacement of SFA with PUFA or carbohydrate, this systematic review concurs with and updates the conclusions drawn by the 2015 Dietary Guidelines Advisory Committee, providing additional context regarding specific CVD endpoint outcomes and the type of carbohydrate replacing SFA. 
  • Regarding the relationship between replacement of SFA with MUFA, this systematic review concurs with and updates the conclusions drawn by the 2015 Dietary Guidelines Advisory Committee.