Monounsaturated fatty acids, olive oil and health status: a systematic review and meta-analysis of cohort studies

Background The aim of the present meta-analysis of cohort studies was to focus on monounsaturated fat (MUFA) and cardiovascular disease, cardiovascular mortality as well as all-cause mortality, and to distinguish between the different dietary sources of MUFA. Methods Literature search was performed using the electronic databases PUBMED, and EMBASE until June 2nd, 2014. Study specific risk ratios and hazard ratios were pooled using a inverse variance random effect model. Results Thirty-two cohort studies (42 reports) including 841,211 subjects met the objectives and were included. The comparison of the top versus bottom third of the distribution of a combination of MUFA (of both plant and animal origin), olive oil, oleic acid, and MUFA:SFA ratio in each study resulted in a significant risk reduction for: all-cause mortality (RR: 0.89, 95% CI 0.83, 0.96, p = 0.001; I2 = 64%), cardiovascular mortality (RR: 0.88, 95% CI 0.80, 0.96, p = 0.004; I2 = 50%), cardiovascular events (RR: 0.91, 95% CI 0.86, 0.96, p = 0.001; I2 = 58%), and stroke (RR: 0.83, 95% CI 0.71, 0.97, p = 0.02; I2 = 70%). Following subgroup analyses, significant associations could only be found between higher intakes of olive oil and reduced risk of all-cause mortality, cardiovascular events, and stroke, respectively. The MUFA subgroup analyses did not reveal any significant risk reduction. Conclusion The results indicate an overall risk reduction of all-cause mortality (11%), cardiovascular mortality (12%), cardiovascular events (9%), and stroke (17%) when comparing the top versus bottom third of MUFA, olive oil, oleic acid, and MUFA:SFA ratio. MUFA of mixed animal and vegetable sources per se did not yield any significant effects on these outcome parameters. However, only olive oil seems to be associated with reduced risk. Further research is necessary to evaluate specific sources of MUFA (i.e. plant vs. animal) and cardiovascular risk. Electronic supplementary material The online version of this article (doi:10.1186/1476-511X-13-154) contains supplementary material, which is available to authorized users.


Background
The most common monounsaturated fatty acids (MUFA) in daily nutrition is oleic acid, followed by palmitoleic acid, and vaccenic acid. Moreover, oleic acid represents the topmost MUFA provided in the diet (~90% of all MUFA). No dietary recommendations for MUFA are given by the National Institute of Medicine, the United States Department of Agriculture, the European Food and Safety Authority and the American Diabetes Association. In contrast, the Academy of Nutrition and Dietetics as well as the Canadian Dietetic Association both promote <20% MUFA of daily total energy consumption, while the American Heart Association sets a limit of 20% MUFA in their respective guidelines [1][2][3]. One reason for specific MUFA recommendations might be their potential benefit in the primary and secondary prevention of cardiovascular diseases. However, previous meta-analyses of cohort studies reported inconsistent results of MUFA on coronary heart disease (CHD). Jakobsen et al. [4] observed that replacement of SFA by MUFA marginally increased the risk of coronary events, whereas no significant effects on coronary death could be observed. These results are in strong discrepancy with another meta-analysis of cohort studies, were Mente et al. [5] reported a significant correlation between MUFA intake and a decrease in the relative risk for CHD. Skeaff and Miller [6] did not observe any effects of MUFA-rich diets on relative risks of CHD events and CHD death. Likewise, the most recent meta-analysis by Chowdhury et al. including nine cohort studies found no significant associations between MUFA intake, circulating MUFA and risk of CHD [7].
One explanation for these inconclusive data might be that different sources of MUFA were not taken into account. Adopting a western diet means that MUFA is predominantly supplied by foods of animal origin, while in south European countries, extra virgin olive oil is the most dominant source of this type of fatty acid [8].
Results of the recently published PREDIMED trial demonstrated major cardiovascular benefits of olive oil and nuts when compared to a low-fat diet [9]. As a major outcome parameter, the risk of stroke was reduced, an event which has not been included in the meta-analyses mentioned above. In addition, a recent cohort study observed a significant association between dietary olive oil, higher plasma oleic acid and reduced risk of stroke [10]. Extra virgin olive oil is regarded to be the genuine driver of the Mediterranean diet and was found to be associated with a 26% reduced risk of all-cause mortality in the Spanish branch of the EPIC study [11]. The aim of the present meta-analysis of cohort studies was to focus on MUFA and CVD (combining CHD and stroke), cardiovascular mortality, and all-cause mortality, and to distinguish between the different dietary sources of MUFA (e.g. olive oil).

Literature search
Queries of literature were performed using the electronic databases PUBMED, and EMBASE (until 2nd June 2014, respectively) with no restrictions to language, and calendar date using the following search terms: ("dietary fat" OR "fatty acids" OR "monounsaturated fat" OR "mufa" OR "olive oil" OR "oleic acid" OR "mediterranean diet") AND ("cardiovascular disease" OR "myocardial infarction" OR "coronary heart disease" OR "stroke" OR "mortality") AND ("incidence" OR "cohort" OR "follow-up" OR "prospective" OR "risk ratio" OR "hazard ratio" OR "rate ratio"). Moreover, the reference lists from retrieved articles, systematic reviews, and meta-analyses were checked to search for further relevant studies. This systematic review was planned, conducted, and reported in adherence to standards of quality for reporting meta-analyses [12]. Literature search was conducted independently by both authors, with disagreements resolved by consensus.

Eligibility criteria
Studies were included in the meta-analysis if they met all of the following criteria: (i) cohort study design; (ii) data related to dietary consumption of MUFA, MUFA: SFA ratio, olive oil, and oleic acid; (iii) the primary outcomes were: all-cause mortality, CVD mortality, combined CVD events (cardiovascular mortality, cardiovascular morbidity (non-fatal myocardial infarction, angina, stroke, heart failure, peripheral vascular events)); the secondary outcomes were: coronary heart disease, and stroke; (iv) adjusted relative risks (RRs), and hazard ratios (HRs) with corresponding 95% confidence intervals (95% CIs) or the data necessary to calculate these; (v) when a study appeared to have been published in duplicate, the version containing the most comprehensive information was selected.
Data extraction and quality assessment The following data were extracted from each study: the first author's last name, year of publication, study origin, outcome parameter, sample size, study length, age at entry, sex, specification of MUFA, adjustment factors, quality score, and risk estimates (HR, RR; highest vs. lowest category) with their corresponding 95% CIs. If separate risk estimates for males and females or separate risk estimates for ages were available in one study, the data were pooled and treated as one study. When a study provided several risk estimates, the maximally adjusted model was chosen. To assess the study quality, a 9-point scoring system according to the Newcastle-Ottawa Scale (NOS) was used. Hence, the full score was 9, and a high-quality study in the present analysis was defined by a threshold of ≥ 7 points [13]. Data extraction and quality assessment were performed by one author (L.S).

Statistical analysis
The meta-analysis was performed by combining the multivariable adjusted RR or HR of the highest compared with the lowest MUFA, MUFA:SFA ratio, oleic acid, or olive oil category based on random effects model using DerSimonian-Laird method, which incorporated both within and between study variability [14]. To ensure a transparent approach to meta-analysis and interpretation of findings in this review, RR/HR estimates for association of fatty acids and primary/secondary outcomes that were often differently reported by each study (such as per-unit or per-1-SD change or comparing quintiles, quartiles, thirds, and other groupings) were transformed, using methods previously described [7]. These transformed estimates consistently corresponded to the comparison of the top versus bottom third of MUFA, MUFA:SFA ratio, olive oil, and oleic acid distribution in each study. To evaluate the weighting of each study, the standard error for the logarithm HR/RR of each study was calculated and regarded as the estimated variance of the logarithm HR/RR using an inverse variance method [14]. Studies were grouped according to the different clinical outcomes (all-cause mortality, cardiovascular mortality, combined cardiovascular events, coronary heart disease, and stroke). Subgroup analysis was performed for total MUFA, MUFA:SFA ratio, oleic acid, and olive oil. Heterogeneity was estimated by the Cochrane Q test together with the I 2 statistic. An I 2 value >50% indicates substantial heterogeneity across studies [15]. The heterogi command in STATA was used to calculate the confidence intervals for the heterogeneity estimates. Funnel plots were used to assess potential publication bias. To determine the presence of publication bias, we assessed the symmetry of the funnel plots in which mean differences were plotted against their corresponding standard errors. In addition, Egger test was performed to test for potential publication bias [16]. Sensitivity analyses were performed assuming statistical heterogeneity with the metaan command in STATA [17]. All analyses were conducted using the Review Manager by the Cochrane Collaboration (version 5.2) and STATA 13.0 (Stata-Corp, College Station, TX).

Missing data
Dr. Goldbourt (personal communication) provided the 23 year follow-up all-cause mortality and cardiovascular mortality data of the Israeli civil cohort for the highest vs. lowest quintile MUFA: SFA ratio [18].

Literature search and study characteristics
A total of 32 cohort studies (42 reports) met the inclusion criteria and were included in the meta-analysis [10,11,. Full search strategy for PUBMED is given in the Additional file 1. General study characteristics are given in Table 1. Sample size varied between 161 and 161,808 with a follow-up time ranging from 3.7 to 30 years. The total number of subjects in the included studies was 841,211.

Main outcomes
According to the different clinical outcomes, overall risk of all-cause mortality was evaluated in seventeen cohorts, cardiovascular mortality in fourteen cohorts, combined cardiovascular events in twenty-eight cohort studies, coronary heart disease in fifteen cohorts, and stroke in eleven cohorts, respectively.

Publication bias
The Egger's linear regression tests provided evidence for a potential publication bias for combined cardiovascular events (p = 0.018), all-cause mortality (p = 0.041), and cardiovascular mortality (p = 0.12) following comparison of the top versus bottom third combined MUFA, olive oil, oleic acid, and MUFA:SFA ratio. No evidence of publication bias could be detected for risk of CHD (p = 0.28) and stroke (p = 0.28). All funnel plots indicate little to moderate asymmetry, suggesting that publication bias cannot be completely excluded as a factor of influence on the present meta-analysis (Additional file 1: Figures S1,  S2, S3, S4 and S5). It remains possible that small studies with inconclusive results have not been published or failed to do so.

Discussion
In the present meta-analysis, comparison of the top versus the bottom third of combined MUFA subgroups (MUFA, olive oil, oleic acid, and MUFA:SFA) was associated with reduced risk of all-cause mortality (11%), cardiovascular mortality (12%), combined cardiovascular events (9%), and stroke (17%). In the ensuing subgroup analyses, this significant correlation could only be observed between higher intakes of olive oil and reduced risk of all-cause mortality, cardiovascular events, and stroke, respectively. In contrast, monounsaturated fatty acids of mixed animal and plant origin did not result in any significant effects with respect to these outcome parameters. Thus, it seems possible that olive oil represents the crucial factor of influence for the protective health effects observed in the primary analysis. However, one has to keep in mind the limitations of the present systematic review and meta-analysis summarized at the end of this section, especially the fact that the specific sources of MUFA have not been indicated in every study.
In order to properly evaluate the potential beneficial or detrimental effects of MUFA with respect to cardiovascular diseases, it seems of importance to consider the source of food providing these fatty acids. In the Nurses' Health Study, MUFA intake was highly correlated with SFA intake (correlation coefficient of 0.81) but only moderately correlated with intakes of PUFA (correlation coefficient of 0.30), suggesting that fat was primarily of animal origin [58]. In the different EPIC cohorts, MUFA intakes ranged between approximately 10% of daily total energy consumption (TEC) in The Netherlands and~20% of TEC in Greece. In general, intake of MUFA was higher in southern European countries as compared central or northern cohorts. However, another distinguishing feature seems to be the predominant source of MUFA in the respective cohorts. In Greece, Spain, and Italy, fat of plant origin (mainly olive oil) provided up to 64% of MUFA intake, whereas in most other EPIC centers, the main contributors to total MUFA intake were meat and meat products, added fats, and dairy products [8]. This might also provide an explanation for the somewhat mixed results provided by systematic reviews and meta-analyses in the past. Thus, a diet rich in MUFA was found to have beneficial effects on a broad range of CVD risk factors, not only in the primary prevention of CVD [1,[59][60][61]. On the other hand, no association between total and individual MUFA and CHD was reported in a meta-analysis of studies assessing both dietary intake and circulating fatty acid composition, [7] while a meta-analysis of observational studies suggested that replacing SFAs with PUFAs might have a greater benefit than replacement of SFAs by MUFA [4]. There is some evidence drawn from prospective studies of an adverse association between MUFA and coronary events, but this correlation might be influenced by high amounts of MUFA of animal origin [4]. A number of in-vivo and in-vitro studies examined the health effects of extra virgin olive oil, the potential "Unique Selling Proposition" of a genuine Mediterranean diet. Thus, the Di@bet.es study demonstrated that individuals who consumed olive oil had a significantly lower risk of developing obesity, impaired glucose metabolism, hypertriglyceridemia, and lower HDL cholesterol levels as compared to a group consuming sunflower oil [62]. In addition, results from experimental studies provide evidence that olive oil consumption improves several CHD risk factors [63,64]. The PREDIMED dietary intervention trial aimed a intake of 50 g/d or more of extra virgin olive oil observed a significant risk reduction of both combined cardiovascular events as well as primary stroke, but not of CHD, indicating a consistency with the results of the present meta-analyses of cohort studies [9]. In a long-term intervention trial by Esposito et al., a higher regression in as well as a lower rate of progression of the intima-media thickness of the carotid artery was found in the group adopting a Mediterranean diet as compared to a low-fat diet reference arm [65].
Apart from oleic acid, olive oil contains a number of bioactive compounds such as polyphenols which are especially prominent in virgin and extra-virgin olive oil, but not in refined olive oil [64,66]. A key olive oil polyphenol is oleuropein (a compound that generates tyrosol and hydroxytyrosol), which accounts for approximately 80% of olive oil phenolic content and is a potent scavenger of superoxide radicals and inhibits LDL oxidation [67,68]. There is a causal link between oxidative stress, inflammation, endothelial dysfunction, and CVD/CHD [69]. A meta-analysis of intervention trials provide evidence that an MD decreases inflammation and improves endothelial function [70]. When focusing on virgin olive oil consumption, the inverse correlation between olive oil and CHD risk found in the present meta-analysis is consistent with the fact that olive oil is not just a supplier of MUFA but of other biologically active components as well.
Several limitations should be taken into account when interpreting the results of the present meta-analysis. MUFA coexist with SFA in several food sources. In addition, cis-and trans-isomers of MUFAs were sometimes categorized together in cohort studies. Furthermore, moderate to substantial heterogeneity could be observed in the present meta-analysis. Potential sources of heterogeneity include combining MUFA/olive oil/oleic acid/ MUFA:SFA ratio in the same analysis, heterogeneous risk estimates, heterogeneous populations/ages/gender, sample sizes as well as follow-up periods of the included studies. No unpublished data were considered for the present meta-analysis, and it cannot be excluded that these results may influence the effect size estimates. Examination of funnel plots showed little to moderate asymmetry suggesting that publication bias cannot be completely excluded as a confounder of the present meta-analysis (e.g. it remains possible that small studies yielding inconclusive data have not been published). In addition, the specific food sources of MUFA could not always identified, limiting the validity of any general recommendation towards MUFAs of plant origin (it is most likely olive oil, but it might be other types of food as well, e.g. nuts, canola oil or a specific variety of sunflower oil). Conversely, it might be that results from studies using mixed sources of MUFA might be biased by non-identified olive oil, making MUFA appear to be beneficial in general when some sources are not. Furthermore, observational studies including cohort studies assessing outcome events affected by nutrition should be interpreted with caution, since reliance on nutritional assessment methods with validity and reliability is lower when compared to randomized controlled trials. However, the present study has some complementary strengths as well. Compared to cohort studies, dietary intervention trials are often limited by lack of double blinding, non-compliance, cross-over, and high drop-out rates. Therefore, well-designed analyses in prospective cohort studies could also provide important evidence with respect to long-term clinical outcomes. Another strength of this work is the inclusion of an overall population >800,000 subjects. To the best of our knowledge, this represents the most comprehensive summary of the evidence on MUFA, olive oil, MUFA:SFA on hard clinical outcome parameters.

Conclusion
The results of the present meta-analysis indicate an overall risk reduction of all-cause mortality (11%), cardiovascular mortality (12%), cardiovascular events (9%), and stroke (17%) when comparing the top vs. bottom thirds of a combination of MUFA, olive oil, oleic acid, and MUFA: SFA ratio. Monounsaturated fat of mixed animal and vegetable sources per se did not yield any significant effects on these outcome parameters. Subgroup analysis indicated that only olive oil (the primary monounsaturated fat source in south European countries) is was associated with a significant risk reduction for several outcomes. These data provide evidence that the source and origin of MUFA within a specific diet should be taken into account in order to evaluate the potential benefits of this type of fatty acids. Further studies are required evaluating specific food sources of MUFA and risk of all-cause mortality and CVD events.