Dietary intake and cardiovascular outcomes in patients with chronic vascular disease: insights from the COMPASS trial cohort

Abstract

Aims

Patients with coronary artery disease (CAD) and patients with peripheral artery disease (PAD) are at risk for major adverse cardiovascular events (MACE) and major adverse limb events (MALE). There are limited data regarding dietary patterns and the risk of recurrent MACE and MALE in CAD and PAD patients. We aimed to identify dietary patterns associated with MACE and MALE in patients with CAD and/or PAD.

Methods and results

We analysed data collected from patients enrolled into the Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) trial, in which diet was assessed by a short food frequency questionnaire (FFQ) at baseline. Two dietary pattern scores, the modified Alternate Healthy Eating Index (mAHEI) and Mediterranean Diet Score (mMDS), were calculated. We tested the association between mAHEI and mMDS and the incidence of MACE and/or MALE. The mean mAHEI score was 23.0 ± 7.7 (out of 70) overall and was similar comparing CAD and PAD patients. The incidence of MACE or MALE was 6.3% in the lowest diet quality quartile (as assessed by mAHEI) compared with 4.2% in the highest quartile over 30 months. In the fully adjusted model, the hazard ratio of a low diet quality (Quartile 1) compared with the highest (Quartile 4) for MACE or MALE was 1.27 (95% CI: 1.08–1.49; P = 0.004, Q1 vs. Q4). This excess hazard was primarily driven by higher MACE in both the CAD and PAD cohorts.

Conclusions

Poor diet quality as assessed by the mAHEI is independently associated with a higher risk of recurrent MACE and MALE in patients with chronic CAD and/or PAD.

Structured Graphical Abstract

Lay Summary

There are limited data regarding dietary patterns and the risk of recurrent major adverse cardiovascular and limb complications in patients with coronary artery disease (CAD) and peripheral artery disease (PAD). We show that

• A low-quality diet is associated with a higher risk of cardiovascular and limb-related complications

• This elevated risk is driven by higher rates of heart attack, stroke, and cardiovascular death in patients with a low-quality diet

Peripheral artery disease, Diet, Coronary artery disease, Nutrition

See the editorial comment for this article ‘Dietary patterns for cardiovascular secondary prevention: eat well to keep the doctor away’, by L. Diaz-Gonzalez and V. Bruña, https://doi.org/10.1093/eurjpc/zwad098.

Introduction

Patients with coronary artery disease (CAD) and peripheral artery disease (PAD) are recognized as a high-risk group and are at risk for recurrent major adverse cardiovascular events (MACE) and major adverse limb events (MALE).^1,2 Current guidelines advocate for the role of diet and lifestyle optimization as key pillars in the management^3,4 of patients with CAD and PAD, but evidence supporting specific food choices is limited. Whether diet quality is a risk factor among patients with established CAD and/or PAD is uncertain.

In patients with established CAD, a Mediterranean style diet⁵ and a high-quality diet as measured by the Alternate Healthy Eating Index (AHEI)⁶ are protective against recurrent cardiovascular events. However, the role of dietary quality in PAD patients has not been extensively studied.⁷

Nutritional risk scores have shown promise in predicting MACE and major vascular complications (e.g. amputations, hospitalization) of PAD in medically and surgically treated patients.^8–11 However, high-quality data supporting a causal association are limited. The Prevención con Dieta Mediterránea (PREDIMED) primary prevention trial conducted in a Spanish population with cardiovascular risk factors suggested that a Mediterranean diet (high in vegetables, olive oil, or nuts) reduces the risk of cardiovascular events including new onset PAD.¹²

The Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) randomized clinical trial included 27 395 patients with CAD and/or PAD.^1,13 In this analysis, we investigate if diet quality is associated with the incidence of MACE and MALE in patients with established CAD and/or PAD.

Methods

Trial design and population

The study design and results of the COMPASS trial have been previously reported.^1,14 The COMPASS trial was a multicentre, double-blind, randomized, placebo-controlled trial which compared the combination of low-dose rivaroxaban plus aspirin, or rivaroxaban alone, with aspirin alone for the prevention of MACE in patients with CAD and/or PAD.

The full details of inclusion and exclusion criteria were described in the original manuscript.¹ In brief, patients with CAD were defined as any history of myocardial infarction in the last 20 years or symptomatic multi-vessel disease with or without revascularization. Patients with PAD were defined as previous peripheral revascularization, limb or foot amputation for arterial disease, carotid stenosis or carotid revascularization, and intermittent claudication with an abnormal ankle–brachial index or imaging evidence of peripheral arterial stenosis. Patients were excluded if they had a high risk of bleeding, stroke within 1 month, an estimated glomerular filtration rate <15 mL/min, or the need for other anti-thrombotic therapeutic regimens based on comorbidities. Subject health and quality of life was assessed using validated questionnaires [Standard Assessment of Global Activities in the Elderly (SAGE), Montreal Cognitive Assessment (MoCA), DSS (Digital Symbol Substitution), European Quality of Life—5 Dimensions (EQ—5D), the Interheart Diet Questionnaire, and the International Physical Activity Questionnaire].

A total of 27 395 patients from 33 countries from North America, South America, Eastern and Western Europe, Australia, and Asia were enrolled (Figure 1).¹ Written informed consent was obtained from all of the participants. The COMPASS trial protocol was approved by relevant health authorities and institutional review boards. The trial was stopped early on the recommendation of the independent data and safety monitoring board due to a substantial benefit in patients treated with a combination of rivaroxaban plus aspirin compared with aspirin alone.

Figure 1

Consolidated standard of reporting trial (CONSORT) diagram of participants in the Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) trial.

Here, we report on the association between dietary quality measured by a modified Alternate Healthy Eating Index (mAHEI) and a modified Mediterranean diet score (mMDS) with major adverse cardiovascular events (MACE) and major adverse limb events (MALE).

Assessment of dietary quality

We recorded participants’ food intake at the time of trial randomization between May 2013 and May 2016 using a short food frequency questionnaire (FFQ) that contained 31 items with pre-defined portion sizes, once, at the baseline visit. Participants were asked ‘In the last 12 months, how often did you eat foods from each of the following categories?’ and a list of food items was given. To compute the daily food intake, the reported frequency of consumption for each food item was converted to daily intake and then was multiplied by the portion size. The FFQ has not been validated but was designed for use in international studies and therefore contained all main food groups including dairy, meat (unprocessed and processed red meat), poultry, fish, eggs, whole and refined grains, nuts, fruits, vegetables, and soft drinks. Standard portion sizes were used for all participants from different countries. We generated two dietary scores: (i) the mAHEI and (ii) the mMDS.^15–17

Modified Alternative Healthy Eating Index

The mAHEI was adapted from the Alternative Healthy Eating Index (AHEI) approach described by McCullough et al.^15,17 We measured six of the nine food items included in the AHEI. Of these, four variables were identical (vegetables, fruits, nuts, ratio of white meat red meat), two items were comparable (whole grains in place of cereal fibre, deep fried foods in place of trans fats), and two items were not collected (alcohol and multivitamin use). Of note, the whole grain group contained whole grain flours such as those made from wheat, rye, triticale, oats, barley, maize, finger and pearl millet, sorghum, buckwheat, or brown rice. Also, wheat products made from flours ideally contain all the components of the intact grain. Polyunsaturated fat intake was not included. The scoring system for each food item was similar to that of the AHEI scoring system (see Supplementary material online, Table S2). For fibre intake, assuming that each serving of whole grain contained 5 g of fibre,¹⁷ we assigned 10 points for 3 or more servings of whole grains and 0 points for no intake. Conversely, for deep-fried foods, the highest score was given for the lowest intake (10 points for ≤ 0.5 times/day and 0 point for ≥ 4 times/day). Since a short FFQ was administered, we were unable to compute daily nutrient intakes and total energy. Hence, we excluded the ratio of polyunsaturated fatty acids/saturated fatty acids. Finally, for each participant, the points for each item were summed and the total score was calculated. A healthy diet was indicated by better adherence to dietary recommendations and reflects a high intake of fruits, vegetables, whole grains, nuts, and a higher ratio of white meat to red meat and low intake of fried foods (Figure 2). With this scale, the lowest possible score is 0, and the highest possible score is 70 (see Supplementary material online, Table S2).

Figure 2

Major adverse cardiovascular and limb events stratified by modified Alternate Healthy Eating Index.

Mediterranean diet score

We included eight food groups into our Mediterranean diet scoring system. We classified fruits and nuts, vegetables, legumes, cereals, and fish as healthy foods. Red meat, poultry, confectionary, sugar, and sugared drinks were classified as unhealthy foods. We did not consider dairy as an unhealthy food as recent studies have indicated the inverse or neutral associations between dairy intake and health outcomes.¹⁸ We also excluded alcohol intake as our FFQ was unable to capture a valid quantifiable measure of alcohol. For each healthy food group, participants received a score of one if consumption was at or above the median; otherwise, a value of zero was assigned. For each unhealthy food group, participants received scores of one if consumption was below the median and zero if they were above. Scores were summed, and each participants’ Mediterranean diet score was calculated. A higher total score indicated a greater degree of adherence to the Mediterranean diet. With this scale, the minimum mMDS is zero, and the maximum mMDS is eight.

Outcomes

The primary outcome was a composite incidence of either MACE (cardiovascular death, myocardial infarction, or stroke) or MALE [severe limb ischaemia leading to an intervention (acute and chronic limb ischaemia) and major vascular amputation] using the definitions reported in the COMPASS trial.¹ The secondary outcomes were MACE and MALE evaluated separately.

Statistical analysis

Patient characteristics for the overall COMPASS cohort and each of the CAD and PAD populations (groups were not mutually exclusive) are summarized as means and standard deviations for normally distributed continuous variables and as counts and proportions for categorical variables, respectively. Patient characteristics are described for the overall study and by quartile of mAHEI score. Comparisons were made using tests appropriate for the level of measurement and distribution of variables.

Multivariable Cox proportional hazards models were used to assess the association between the mAHEI and mMDS on the composite outcome of MACE or MALE. Quartiles 1–3 were compared with Quartile 4, which served as the reference. Unadjusted models were first constructed, including the health status and mAHEI scores as well as the random effect for country and treatment arm assigned. Next, socio-demographic factors and cardiovascular risk factors were added based on prior models derived from the COMPASS cohort.¹⁹ Thus, the fully adjusted model covariates included country, treatment, age, sex, level of education, use of a lipid-lowering agent, BMI, smoking, diabetes, hypertension, heart failure, eGFR, history of a cardiovascular event or amputation, number of vascular beds involved, and antiplatelet use at baseline.

As a secondary analysis, these models were repeated separately for CAD and PAD cohorts. Because the CAD and PAD cohorts were not mutually exclusive (17.9% of patients had CAD and PAD), we stratified by eligibility diagnosis of CAD or PAD cohorts as has been done in previous analyses.^13,20 For the CAD cohort, Kaplan–Meier survival curves for the combined endpoint (MACE/MALE) and for MACE alone were constructed according to quartiles of mAHEI scores. The log rank test was used to test whether curves significantly differed by quartiles of these scores. Next, hierarchical Cox proportional hazard regression models were constructed to assess the association between the mAHEI scores and MACE/MALE in the CAD and PAD cohorts, respectively.

All tests were two-tailed, and significance levels were set at P < 0.05. All analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC).

Results

We included 26 539 patients, of which 24 119 had CAD and 7163 had PAD (Figure 1). The mean age of the population was 68.2 ± 7.9 years, 20 789 patients were men (78.3%), and the mean body mass index (BMI) was 28.5 ± 4.6 kg/m² (see Supplementary material online, Table S1). The mean mAHEI score was 23.0 ± 7.7, and compliance with healthcare was high (91.3% taking medication regularly and 92.2% seeing a physician regularly). Most patients were of White European origin (63.1%, n = 16 676).

Compared with patients with a low mAHEI, patients with a high mAHEI were more likely to be older, have a lower BMI, and were less likely to be current smokers, have hypertension, and have diabetes. Similar patterns were observed among patients with CAD and PAD (Table 1). There were a total of 1391 events recorded, of which 1262 were MACE and 140 were MALE.

Table 1

Baseline characteristics of patients according to quartiles of the Alternate Healthy Eating Index

	Q1 (n = 6536)	Q2 (n = 6687)	Q3 (n = 6655)	Q4 (n = 6661)	P for trend
Median score	14.6	19.3	24.6	32.4
Age, mean (SD), y	67.4 (8.4)	68.0 (8.1)	68.5 (7.7)	69.0 (7.4)	<0.0001
Male (%), no.	4963 (75.9)	5220 (78.1)	5308 (79.8)	5298 (79.5)	<0.0001
BMI, mean (SD)	29.0 (4.6)	28.8 (4.7)	28.5 (4.5)	27.8 (4.3)	<0.0001
Race (%), no.
White	3744 (57.3)	4270 (63.9)	4341 (65.2)	4373 (65.7)	<0.0001
Black	110 (1.7)	58 (0.9)	50 (0.8)	33 (0.5)
Asian	400 (6.1)	791 (11.8)	1198 (18.0)	1581 (23.7)
Other	2282 (34.9)	1568 (23.5)	1066 (16.0)	674 (10.1)
Risk factors (%), no.
Current smoking	1880 (28.8)	1477 (22.1)	1255 (18.9)	1045 (15.7)	<0.0001
Diabetes	2635 (40.3)	2664 (39.8)	2488 (37.4)	2248 (33.8)	<0.0001
Hypertension	5183 (79.3)	5176 (77.4)	4932 (74.1)	4748 (71.3)	<0.0001
MET min exercise per week, median	5362 ± 8565	5596 ± 8128	5890 ± 8529	6093 ± 9281	<0.0001

Incidence of major adverse cardiovascular events and major adverse limb events by Alternate Healthy Eating Index score

The incidence of the MACE and MALE composite outcome was highest in patients with a poor diet quality. Each five-point reduction in the mAHEI was associated with a 7% (HR 1.07; 95% CI 1.03–1.11; P = 0.001) increase in MACE and MALE (Table 2). Patients in the lowest quartile of the mAHEI score experienced a 27% increase in the risk for MACE and MALE (HR 1.27; 95% CI: 1.08–1.49; P = 0.004; Q1 vs. Q4) when compared with patients in the highest quartile (Table 3). The 30-month risk of an incident event in patients with the lowest diet quality was 6.3% compared with 4.2% in the highest dietary quartile group over 30 months. The 30-month Kaplan–Meier estimate for the incidence of MACE and MALE also increased progressively as dietary quality declined. The separation of curves was evident as early as 6 months following trial enrolment (Figure 2).

Table 2

Incremental risk of worsening dietary quality as a risk factor for major adverse cardiovascular and limb events

	mAHEI, per five-unit decrease	Minimally adjusted model		Fully adjusted model
		HR (95% CI)	P value	HR (95% CI)	P value
Overall population	MACE or MALE	1.11 (1.06–1.15)	<0.0001	1.07 (1.03–1.11)	0.001
	MACE	1.10 (1.05–1.14)	<0.0001	1.06 (1.02–1.11)	0.005
	MALE	1.16 (1.02–1.32)	0.02	1.11 (0.97–1.25)	0.12
CAD group	MACE or MALE	1.11 (1.06–1.15)	<0.0001	1.06 (1.02–1.11)	0.005
	MACE	1.10 (1.05–1.14)	<0.0001	1.06 (1.01–1.10)	0.01
	MALE	1.25 (1.06–1.46)	0.008	1.12 (0.95–1.31)	0.17
PAD group	MACE or MALE	1.13 (1.06–1.21)	0.0002	1.11 (1.04–1.19)	0.002
	MACE	1.14 (1.07–1.23)	0.0002	1.14 (1.04–1.20)	0.004
	MALE	1.05 (0.92–1.19)	0.49	1.02 (0.90–1.16)	0.78

CAD, coronary artery disease; HR, hazard ratio; mAHEI, modified Alternate Healthy Eating Index; MACE, major adverse cardiovascular events; MALE, major adverse limb events; PAD, peripheral artery disease.

Table 3

Diet quality as measured by the Alternate Healthy Eating Index as a risk factor for major adverse cardiovascular and limb events

	mAHEI score	Minimally adjusted model		Fully adjusted modela
		HR (95% CI)	P value	HR (95% CI)	P value
Overall population	Q1 (mAHEI <17)	1.45 (1.23–1.70)	<0.0001	1.27 (1.08–1.49)	0.0044
	Q2 (mAHEI 17–21)	1.25 (1.06–1.46)	0.01	1.12 (0.96–1.32)	0.15
	Q3 (mAHEI 22–27)	1.16 (0.99–1.36)	0.07	1.09 (0.93–1.28)	0.30
	Q4 (mAHEI ≥ 28)	1.00		1.00
CAD group	Q1 (mAHEI <17)	1.45 (1.22–1.73)	<0.0001	1.24 (1.04–1.48)	0.01
	Q2 (mAHEI 17–21)	1.31 (1.11–1.55)	0.002	1.16 (0.98–1.38)	0.08
	Q3 (mAHEI 22–27)	1.21 (1.02–1.43)	0.03	1.12 (0.95–1.33)	0.18
	Q4 (mAHEI ≥ 28)	1.00		1.00
PAD group	Q1 (mAHEI <17)	1.52 (1.18–1.94)	0.001	1.39 (1.08–1.79)	0.01
	Q2 (mAHEI 17–21)	1.37 (1.07–1.75)	0.01	1.29 (1.00–1.65)	0.05
	Q3 (mAHEI 22–27)	0.92 (0.70–1.20)	0.53	0.87 (0.66–1.13)	0.29
	Q4 (mAHEI ≥ 28)	1.00		1.00

CAD, coronary artery disease; HR, hazard ratio; mAHEI, modified Alternate Healthy Eating Index; PAD, peripheral artery disease.

Fully adjusted model includes country, treatment, age, sex, level of education, use of a lipid-lowering agent, BMI, smoking, diabetes, hypertension, heart failure, eGFR, history of a cardiovascular event or amputation, number of vascular beds involved, and antiplatelet use at baseline.

The increased event incidence was primarily driven by an excess of MACE in patients with low diet quality (Figure 3). After adjusting for potential confounding factors, the incidence of MACE was increased (HR 1.26; 95% CI 1.06–1.49; P = 0.008) when comparing the lowest vs. highest quartiles of dietary quality. The estimate in MALE was also higher but non-significant (HR 1.29; 95% CI 0.78–2.1) (see Supplementary material online, Table S3).

Figure 3

Kaplan–Meier curves of individual components [major adverse cardiovascular events (MACE) and major adverse limb events (MALE)] in the overall population.

In patients with CAD, a similar increase in the risk of MACE and MALE was observed in patients in the lowest quartile of dietary quality compared with the highest (HR 1.24; 95% CI 1.04–1.48; P = 0.01). This was once again driven by an excess of MACE experienced by patients in the lowest diet quality quartile (Figure 4).

Figure 4

Kaplan–Meier curve of major adverse cardiovascular and limb events in the coronary artery disease cohort.

Patients with PAD experienced a higher overall event rate compared with patients with CAD (7.4% vs. 5.2% over 30 months). The risk of MACE and MALE was 39% higher in the lowest mAHEI quartile compared with the highest mAHEI quartile (Figure 5). A significantly higher rate of MACE was observed (HR 1.47; 95% CI 1.11–1.95; P = 0.008), but there was no association between mAHEI and the incidence of MALE (HR 0.95; HR 0.58–1.56) (see Supplementary material online, Table S3).

Figure 5

Kaplan–Meier curve of major adverse cardiovascular and limb events in the peripheral artery disease cohort.

Mediterranean diet score

The median mMDS was 3.71 ± 1.47. Although numerically higher, the incidence of MACE and MALE comparing patients in the lowest Mediterranean diet score quartile to the highest quartile was not significantly higher in a fully adjusted model (HR 1.14; 0.98–1.33; P = 0.10). This was true for both the CAD and PAD subgroups (see Supplementary material online, Table S4).

Discussion

In the large COMPASS trial cohort, we found that low diet quality is a risk factor for recurrent MACE and MALE in patients with established CAD and/or PAD.

Our data reaffirm that low dietary quality is an independent risk factor for MACE/MALE in patients with established atherosclerotic vascular disease. Patients with a low-quality diet as assessed by the mAHEI experienced a 27% higher risk of recurrent MACE and MALE after adjustment for confounding factors, primarily driven by an excess of MACE. The estimate for MALE was similar, but not statistically significant. While it has been shown that improved dietary quality is associated with a lower risk of incident cardiovascular disease,^21,22 there is a paucity of evidence describing the association between dietary quality and secondary prevention of cardiovascular events, especially in PAD patients. A prior analysis of the Ongoing Telmisartan Alone and in Combination With Ramipril Global End Point Trial (ONTARGET) and Telmisartan Randomized Assessment Study in ACE-I Intolerant Subjects with Cardiovascular Disease (TRANSCEND) trial (total of 31 564 participants followed for 5 years) also reported a protective effect of a high-quality diet in preventing recurrent cardiovascular events in patients ≥ 55 years of age with established cardiovascular disease or diabetes mellitus and end-organ damage.⁶ Our findings in the COMPASS cohort with a large proportion of PAD patients are consistent with the findings of ONTARGET/TRANSCEND.

Patients with atherosclerotic PAD represent a high-risk subgroup of vascular disease. This is supported by higher baseline event rates in multiple large cardiovascular outcome trials.^2,13,23,24 However, despite a growing recognition that PAD is responsible for a significant burden of morbidity and mortality, much of the evidence in clinical management has been extrapolated from the larger body of evidence in CAD. While societal guidelines have highlighted dietary optimization as a key aspect of non-pharmacologic therapy in PAD, there is minimal evidence to describe the impact of dietary intake on vascular outcomes specific to the peripheral arterial beds.⁷ In the COMPASS PAD cohort, the rates of MACE and MALE were higher than those in the CAD only cohort, confirming previous findings that PAD patients are a high-risk subgroup.^2,13,24 The excess risk for MACE and MALE among PAD patients with a low diet quality is substantial among PAD patients and is driven primarily by the higher rate of MACE. There was no significant increase in MALE with a low mAHEI score, although the overall MALE incidence rate was low in COMPASS.

Using a modified Mediterranean diet score, our analysis demonstrated a trend towards reduction of MACE and MALE with a higher adherence to a Mediterranean diet which was not statistically significant in the fully adjusted model. The bulk of evidence supporting dietary guidelines in atherosclerotic cardiovascular disease comes from several trials assessing the utility of a Mediterranean diet. The Lyon Diet Heart Study suggested the protective effect of a Mediterranean style diet in reducing combined endpoints of cardiac death and non-fatal myocardial infarction in a secondary prevention population.⁵ Notably, our modified score was different from the original Mediterranean Diet Score described by Trichopoulou et al.,¹⁶ as some components were not captured in our FFQ.

A subgroup analysis of the PREDIMED study provided some insight into the role of dietary content into the development of PAD in a primary prevention population. This randomized controlled study suggested that the Mediterranean diet high in olive oil or nuts led to a significant reduction in the incidence of PAD.^12,25 Unfortunately, the study suffered from inconsistencies in methodology and randomization, leading to the eventual retraction and re-publication of the article, although the overall results were not significantly different.^12,26 This result has not been replicated, and our investigation of the role of diet quality in a secondary prevention PAD population is the largest of such analysis to date.

The mechanisms by which dietary quality affects vascular health are not fully understood. Several mechanisms involving pathogenesis of atherosclerosis have been proposed. It is generally accepted that dietary modification is an initial approach to augment the management of several risk factors for atherosclerosis including hypertension, diabetes, and dyslipidaemia. However, inflammation is now recognized as an important modulator of atherosclerosis. Omega-3 polyunsaturated fatty acids have been shown to demonstrate atheroprotective effects in molecular studies²⁷ and in the recent randomized clinical trial Reduction of Cardiovascular Events with Icosapent Ethyl—Intervention Trial (REDUCE-IT).²⁸ This trial reported that a highly purified form of eicosapentaenoic acid called icosapent ethyl given to patients with established cardiovascular disease or diabetes and additional risk factors resulted in a 25% reduction in the rate of cardiovascular death, myocardial infarction, or stroke.²⁸ While the mechanism responsible for this benefit has not been well established, several hypotheses exist including an antithrombotic effect, modulation of inflammatory pathways, and plaque stabilization. While it would be unlikely that a modified dietary intake or supplementation can replicate the doses of omega-3 polyunsaturated fatty acids used in this study, some elements of the protective effect of a high-quality diet may operate through similar mechanisms. In addition, there is a growing interest in the role of the intestinal microbiome, as some metabolites produced from dietary metabolism may lead to atherosclerotic generation and thrombosis.²⁹ Finally, further studies are necessary to identify causal pathways and potential therapies of dietary intake and cardiovascular disease, particularly among patients with PAD.

Clinicians are often asked by patients about dietary guidance once a disease has been established. Dietary recommendations have challenges as many foods are not applicable across ethnic groups, countries of origin, and availability of resources. However, our study indicates that the emphasis should be shifted to improving overall dietary quality rather than specific food types by suggesting greater consumption of fruits, vegetables, nuts, higher fibre foods, choosing white over red meat, and consumption of minimally processed foods. This may improve the applicability to a larger general population with a variety of cultural backgrounds and simplify advice to patients. Thus, clinicians should recognize and consider poor dietary quality as an important risk factor when stratifying patients with respect to vascular disease.

Limitations

Our analysis has several limitations. Our dietary assessment was performed at trial enrolment, and this analysis assumes that dietary patterns were relatively consistent throughout the trial period. Several potential risk factors were not captured in this study, and risk factors were only assessed once at trial enrolment. Hence, despite adjustments to our models, we may not have captured all relevant risk factors, and the reported estimates could be under-adjusted. The short FFQ has not been validated; however, the success in prediction and agreement of our findings with previous studies indicates that the short FFQ captured the intake of important food groups. In addition, the population was well treated with medical therapies and enrolled outpatients with PAD, resulting in low event rates, especially for MALE. Due to differences in the FFQ that was used, we had to modify the originally described AHEI, yet we believe our modifications are justified and reflect contemporary evidence supporting dietary quality, as has been shown in other studies.⁶ Notably, this was not a quantitative FFQ and energy intake could not be calculated nor accounted for. We did, however, account for body size using BMI which ranged from 27.8 to 29 and suggests that none of the participants had excessively low energy intake or were malnourished. While this is a large cohort of patients, these data remain observational and we have therefore described associations, rather than inferring causality. Finally, the Mediterranean diet score used in this analysis did not show that a classical Mediterranean diet is associated with a lower risk of MACE and MALE, as we needed to make adjustments in scoring based on the available data from the FFQ used.

Conclusions

Our analysis suggests that a low dietary quality is an independent risk factor for MACE and MALE in patients with established CAD and/or PAD. This appears to be driven by an excess of MACE, regardless of which vascular bed is affected. Healthcare providers should recognize dietary quality as an important risk factor with the potential to reduce the risk of MACE and MALE in patients with atherosclerotic vascular disease.

Author contributions

Darryl Wan (Visualization: lead; Writing—original draft: lead; Writing—review & editing: lead), Mahshid Dehghan (Methodology: equal; Writing—review & editing: equal), Russell J de Souza (Methodology: equal; Writing—review & editing: equal), Chinthanie Ramasundarahettige (Formal analysis: lead), John W Eikelboom (Writing—review & editing: supporting), Jackie Bosch (Writing—review & editing: supporting), Aldo P Maggioni (Writing—review & editing: supporting), Deepak L Bhatt (Writing—review & editing: supporting), Salim Yusuf (Writing—review & editing: supporting), and Sonia Anand (Supervision: lead; Writing—original draft: supporting; Writing—review & editing: lead).

Supplementary material

Supplementary material is available at European Journal of Preventive Cardiology online.

Acknowledgements

The authors acknowledge the work of Natalie Williams, who assisted with manuscript preparation and submission.

Data availability

The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.

Author notes

Conflict of interest: D.W. has received honoraria from Amgen Canada, HLS Therapeutics Inc., and Sanofi–Aventis Canada. M.D. has no disclosures. R.d.S. has received grants from the Canadian Foundation for Dietetic Research, the Canadian Institutes of Health Research, the Hamilton Health Sciences Corporation, and the Population Health Research Institute; receiving personal and/or travel fees from Health Canada, McMaster Children’s Hospital, the University of Toronto, and the World Health Organization; serving as a member of the Nutrition Science Advisory Committee to Health Canada and the Scientific Advisory Committee on Nutrition Subgroup on the Framework for the Evaluation of Evidence (Public Health England); serving as an external resource person to the World Health Organization Nutrition Guidelines Advisory Group on trans fats, saturated fats, and polyunsaturated fats; and serving as an independent director of the Helderleigh Foundation outside the submitted work. C.R. has no disclosures. J.E. reports grants and personal fees from Bayer, Boehringer Ingelheim, Bristol–Myers Squibb/Pfizer, and Daiichi Sankyo, during the conduct of the study, and grants and personal fees from Bayer, Boehringer Ingelheim, Bristol–Myers Squibb/Pfizer, Daiichi Sankyo, Janssen, Astra Zeneca, Eli Lilly, Glaxo Smith Kline, and Sanofi Aventis, outside the submitted work. J.B. has no disclosures. A.M. reports personal fees from Novartis, Bayer, Fresenius, and Cardiorentis. D.B. discloses the following relationships—Advisory Board: AngioWave, Bayer, Boehringer Ingelheim, Cardax, CellProthera, Cereno Scientific, Elsevier Practice Update Cardiology, Janssen, Level Ex, Medscape Cardiology, Merck, MyoKardia, NirvaMed, Novo Nordisk, PhaseBio, PLx Pharma, Regado Biosciences, and Stasys; Board of Directors: AngioWave (stock options), Boston VA Research Institute, Bristol Myers Squibb (stock options), DRS.LINQ (stock options), Society of Cardiovascular Patient Care, and TobeSoft; Chair: Inaugural Chair, American Heart Association Quality Oversight Committee; data monitoring committees: Acesion Pharma, Assistance Publique–Hôpitaux de Paris, Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute, for the PORTICO trial, funded by St. Jude Medical, now Abbott), Boston Scientific (Chair, PEITHO trial), Cleveland Clinic (including for the ExCEED trial, funded by Edwards), Contego Medical (Chair, PERFORMANCE 2), Duke Clinical Research Institute, Mayo Clinic, Mount Sinai School of Medicine (for the ENVISAGE trial, funded by Daiichi Sankyo; for the ABILITY-DM trial, funded by Concept Medical), Novartis, Population Health Research Institute, and Rutgers University (for the NIH-funded MINT Trial); honoraria: American College of Cardiology (Senior Associate Editor, Clinical Trials and News, ACC.org; Chair, ACC Accreditation Oversight Committee), Arnold and Porter law firm (work related to Sanofi/Bristol–Myers Squibb clopidogrel litigation), Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute; RE-DUAL PCI clinical trial steering committee funded by Boehringer Ingelheim; AEGIS—II executive committee funded by CSL Behring), Belvoir Publications (Editor in Chief, Harvard Heart Letter), Canadian Medical and Surgical Knowledge Translation Research Group (clinical trial steering committees), Cowen and Company, Duke Clinical Research Institute (clinical trial steering committees, including for the PRONOUNCE trial, funded by Ferring Pharmaceuticals), HMP Global (Editor in Chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (Guest Editor; Associate Editor), K2P (Co-Chair, interdisciplinary curriculum), Level Ex, Medtelligence/ReachMD (CME steering committees), MJH Life Sciences, Oakstone CME (Course Director, Comprehensive Review of Interventional Cardiology), Piper Sandler, Population Health Research Institute (for the COMPASS operations committee, publications committee, steering committee, and USA national co-leader, funded by Bayer), Slack Publications (Chief Medical Editor, Cardiology Today’s Intervention), Society of Cardiovascular Patient Care (Secretary/Treasurer), WebMD (CME steering committees), and Wiley (steering committee); other: Clinical Cardiology (Deputy Editor), NCDR–ACTION Registry Steering Committee (Chair), VA CART Research and Publications Committee (Chair); research funding: Abbott, Acesion Pharma, Afimmune, Aker BioMarine, Amarin, Amgen, AstraZeneca, Bayer, Beren, Boehringer Ingelheim, Boston Scientific, Bristol–Myers Squibb, Cardax, CellProthera, Cereno Scientific, Chiesi, CSL Behring, Eisai, Ethicon, Faraday Pharmaceuticals, Ferring Pharmaceuticals, Forest Laboratories, Fractyl, Garmin, HLS Therapeutics, Idorsia, Ironwood, Ischemix, Janssen, Javelin, Lexicon, Lilly, Medtronic, Merck, Moderna, MyoKardia, NirvaMed, Novartis, Novo Nordisk, Owkin, Pfizer, PhaseBio, PLx Pharma, Recardio, Regeneron, Reid Hoffman Foundation, Roche, Sanofi, Stasys, Synaptic, The Medicines Company, 89Bio; royalties: Elsevier (Editor, Braunwald’s Heart Disease); Site Co-Investigator: Abbott, Biotronik, Boston Scientific, CSI, Endotronix, St. Jude Medical (now Abbott), Philips, Svelte; Trustee: American College of Cardiology; unfunded research: FlowCo, Takeda. S.Y. has received research grants, honoraria, and travel expenses for lectures from Bayer, Boehringer Ingelheim, Bristol–Myers Squibb, and AstraZeneca. S.A. has received speaking honoraria and consulting fees from Bayer and speaking fees from Janssen.