Comparing dietary strategies to manage cardiovascular risk in primary care: a narrative review of systematic reviews

Background Nutrition care in general practice is crucial for cardiovascular disease (CVD) prevention and management, although comparison between dietary strategies is lacking. Aim To compare the best available (most recent, relevant, and high-quality) evidence for six dietary strategies that are effective for primary prevention/absolute risk reduction of CVD. Design and setting A pragmatic narrative review of systematic reviews of randomised trials focused on primary prevention of cardiovascular events. Method Studies about: 1) adults without a history of cardiovascular events; 2) target dietary strategies postulated to reduce CVD risk; and 3) direct cardiovascular or all-cause mortality outcomes were included. Six dietary strategies were examined: energy deficit, Mediterranean-like diet, sodium reduction (salt reduction and substitution), the Dietary Approaches to Stop Hypertension (DASH) diet, alcohol reduction, and fish/fish oil consumption. Reviews were selected based on quality, recency, and relevance. Quality and certainty of evidence was assessed using GRADE. Results Twenty-five reviews met inclusion criteria; eight were selected as the highest quality, recent, and relevant. Three dietary strategies showed modest, significant reductions in cardiovascular events: energy deficit (relative risk reduction [RRR] 30%, 95% confidence interval [CI] = 13 to 43), Mediterranean-like diet (RRR 40%, 95% CI = 20 to 55), and salt substitution (RRR 30%, 95% CI = 7 to 48). Still, some caveats remain on the effectiveness of these dietary strategies. Salt reduction, DASH diet, and alcohol reduction showed small, significant reductions in blood pressure, but no reduction in cardiovascular events. Fish/fish oil consumption showed little or no effect; supplementation of fish oil alone showed small reductions in CVD events. Conclusion For primary prevention, energy deficit, Mediterranean-like diets, and sodium substitution have modest evidence for risk reduction of CVD events. Strategies incorporated into clinical nutrition care should ensure guidance is person centred and tailored to clinical circumstances.


Background
Nutrition care in general practice is crucial for cardiovascular disease (CVD) prevention and management, although comparison between dietary strategies is lacking.

Aim
To compare the best available (most recent, relevant, and high-quality) evidence for six dietary strategies that are effective for primary prevention/ absolute risk reduction of CVD.

Design and setting
A pragmatic narrative review of systematic reviews of randomised trials focused on primary prevention of cardiovascular events.

Method
Studies about: 1) adults without a history of cardiovascular events; 2) target dietary strategies postulated to reduce CVD risk; and 3) direct cardiovascular or all-cause mortality outcomes were included.Six dietary strategies were examined: energy deficit, Mediterranean-like diet, sodium reduction (salt reduction and substitution), the Dietary Approaches to Stop Hypertension (DASH) diet, alcohol reduction, and fish/fish oil consumption.Reviews were selected based on quality, recency, and relevance.Quality and certainty of evidence was assessed using GRADE.

Results
Twenty-five reviews met inclusion criteria; eight were selected as the highest quality, recent, and relevant.Three dietary strategies showed modest, significant reductions in cardiovascular events: energy deficit (relative risk reduction [RRR] 30%, 95% confidence interval [CI] = 13 to 43), Mediterranean-like diet (RRR 40%, 95% CI = 20 to 55), and salt substitution (RRR 30%, 95% CI = 7 to 48).Still, some caveats remain on the effectiveness of these dietary strategies.Salt reduction, DASH diet, and alcohol reduction showed small, significant reductions in blood pressure, but no reduction in cardiovascular events.Fish/fish oil consumption showed little or no effect; supplementation of fish oil alone showed small reductions in CVD events.

Conclusion
For primary prevention, energy deficit, Mediterranean-like diets, and sodium substitution have modest evidence for risk reduction of CVD events.Strategies incorporated into clinical nutrition care should ensure guidance is person centred and tailored to clinical circumstances.

Introduction
The World Health Organization estimates that 75% of cardiovascular events may be preventable.However, cardiovascular disease (CVD) is a major source of morbidity and mortality globally. 1Appropriate primary care management of absolute CVD risk reduces associated death and disability, 2 and contemporary CVD risk management focuses on reducing absolute risk (that is, 5-year risk of CVD including multiple risk factors), rather than individual risk factors. 3,4et is a key factor in managing absolute CVD risk, 4 and is recognised for the primary prevention of CVD, 1,5,6 but the relative effectiveness of different dietary strategies remains unclear.Direct comparison of different dietary strategies will support clinicians to make evidence-informed diet recommendations for patients looking to manage CVD risk.Although other dietary approaches (for example, vegetarian or vegan diets) 7 may influence CVD risk, this narrative review of systematic reviews focuses on six strategies identified as potentially influencing absolute cardiovascular risk by the National Heart Foundation of Australia and an expert advisory panel. 8,9These are outlined in Box 1 along with their proposed cardiovascular effects.Given the heterogeneous nature of these interventions, a narrative review was conducted to allow top-level comparison between dietary strategies to inform clinical decisions.
In this pragmatic narrative review of systematic reviews, the aim was to identify and descriptively compare the most relevant, best available, highest-quality systematic reviews for six dietary strategies postulated to be RESEARCH | e200 effective for primary prevention/absolute risk reduction of CVD: • energy deficit; • Mediterranean-like diet; • sodium reduction (salt reduction and substitution); • Dietary Approaches to Stop Hypertension (DASH) diet; • alcohol reduction; and • fish/fish oil consumption. 8

Method
This narrative review has been reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 18 broad methodological approach was prospectively developed as part of a larger programme of commissioned work to update Australian absolute CVD risk guidelines.As a result of this, the protocol was not registered.Pragmatic decisions were made to enable identification of the most recent, relevant, and high-quality available evidence.In brief, studies published after 2014 were included to limit screening and identify the most recent evidence; the review was limited to six dietary strategies in line with the funder's priorities; 8,9 it utilised a targeted search approach (that is, a two-pronged approach) to identify included reviews; and used AMSTAR 1 rather than AMSTAR 2 to assess quality as it allows for an overall score.These decisions

Alcohol reduction
Reduction in usual alcohol intake or elimination of alcohol from diet Decreased alcohol is thought to lower blood pressure.Alcohol consumption has a complex relationship with cardiovascular health and excess consumption is associated with many CV diseases, so reduction may reduce CV disease risk 15,16 Fish/fish oil consumption Diets high in fish, or supplemental fish oils Omega chain fatty acids, commonly found in fish, are thought to be cardioprotective 17 CV = cardiovascular.DASH = Dietary Approaches to Stop Hypertension.
and full justifications are summarised in Supplementary Information S1.

Identification of target dietary interventions
Target dietary interventions (outlined in Box 1) relevant to clinical practice were identified jointly by general practice, cardiovascular and nutrition and dietetics experts from the authorship team, their networks, and the National Heart Foundation cardiovascular expert committee. 8arch strategy A two-pronged citation analysis approach was used to identify reviews for each dietary strategy. 19First, 'similar articles' searches in PubMed were conducted from known seed articles. 20Second, for each review identified, a forward and backward citation analysis was conducted using the SpiderCite automation tool, 21 and results were filtered to include only systematic reviews.

Screening and study selection
Systematic reviews of randomised controlled trials (RCTs) published from January 2014 to May 2022 were screened for eligibility against PICO (population, intervention, comparator, outcome) criteria in Box 2. If a review for a target diet reporting direct cardiovascular or mortality outcomes was not found, reviews with indirect outcomes (for example, blood pressure and blood lipids) were considered.
At prong 1, one reviewer (the first author) screened systematic reviews for those relevant to the PICO of each target dietary intervention.At prong 2, two reviewers (the first author and the second or third author) independently screened title and abstract to select eligible reviews, including those identified in prong 1.The full text was obtained and was independently assessed for eligibility and risk of bias by two reviewers (the first author and the second or third author).Discrepancies in screening were resolved by discussion or referral to another author (the senior author).
Selection of key systematic review/s for each dietary intervention was made jointly by all authors, accounting for recency, quality, and relevance (that is, how closely the review question matched the PICO of the six selected dietary approaches).If the outcomes in a review did not include both indirect (for example, blood pressure) and direct (for example, CVD events or mortality) measures of cardiovascular risk, the authors considered and included more than one review to summarise findings where appropriate.If studies were related and included additional information (for example, subsequent trials or reanalysis), these results are also reported in the current article.

Review methodological quality (risk of bias)
AMSTAR 1 was used to assess methodological quality of systematic reviews across 11 domains, and it is considered a pragmatic, time-efficient method for comparing quality between studies. 22Two authors (the first author and the second or third author) independently assessed risk of bias for all full-text results.

Data extraction
Where available, the following outcome data were extracted: dietary strategy, number of included trials and participants, participant hypertension status (hypertensive, mixed), intervention, length of intervention, adherence to intervention, comparator, all relevant outcomes as reported (that is, total [all-cause] mortality), CVD mortality, CVD events, blood pressure (systolic, diastolic), body weight, blood lipid concentration, change in alcohol consumption, length of study follow-up, and effect estimate (hazard ratio [HR]  or risk ratio [RR]) with 95% confidence intervals (CIs) and/or P-values.

Evaluation of evidence certainty
The certainty or quality of the body of evidence for each reported outcome was assessed using GRADE. 23GRADE provides a rating reflecting how certain the authors are that estimated effect aligns with the true effect: very low, low, moderate, or high.Where original authors had conducted GRADE, results were reviewed and retained if sufficiently detailed and applicable.For reviews without existing GRADE assessment, two authors (the first and second authors) completed GRADE independently, with disputes resolved by discussion or referral to another author (the third or senior author).

Data synthesis
Existing outcomes for the dietary strategies were tabulated to allow direct descriptive comparison between strategies.A narrative overview of findings compares the different strategies for primary CVD prevention.No new statistical analyses were planned because of the heterogeneity of interventions, timeframes, review questions, and populations.

Results
The selection process is presented in Figure 1.Eight reviews were included for the six dietary strategies.Supplementary Table S1   Alcohol reduction Direct: Acin et al (2020) 30 Indirect: Roerecke et al (2017) 31 Fish and fish oil Abdelhamid (2020) 32 British Journal of General Practice, March 2024 0.09; high-density lipoprotein [HDL]: MD 0.02 mmol/L, 95% CI = -0.04 to 0.08).Compared with other dietary interventions, Mediterranean-like diet had little to no clear effect on blood pressure or lipids.
For people with normotensive blood pressure, there is moderate-to-high certainty evidence that sodium reduction does not meaningfully reduce CVD events (RR 0.84, 95% CI = 0.64 to 1.10) or all-cause mortality (RR 0.90, 95% CI = 0.58 to 1.40).These findings are driven by one trial assessing large-scale salt reduction by institutional low-sodium alternative substitution (salt substitution).

DASH diet
Filippou et al (2020) 29 reported moderate-certainty evidence that, over an average of 15.3 weeks, the DASH diet (compared with a control diet) had a small effect on blood pressure reduction (SBP MD -3.2 mmHg, 95% CI = -4.2 to -2.3; DBP MD -2.5 mmHg, 95% CI = -3.5 to -1.5); this was consistent regardless of baseline blood pressure.The blood pressure lowering effect of the DASH diet was more pronounced when participants' baseline sodium intake was >2400 mg/ day (P = 0.003), and when participants were aged <50 years (P<0.001).There was no moderating effect of weight, and no evidence was reported for the impact of DASH diet for primary prevention of cardiovascular events or mortality.

Alcohol reduction
Acin et al (2020) 30 reported little to no effect of alcohol reduction on all-cause mortality (RR 0.70, 95% CI = 0.20 to 3.20) or cardiovascular events (RR 0.80, 95% CI = 0.36 to 1.79), but evidence was of low certainty and drawn from a subgroup of one trial with mostly male participants.No included studies reported CVD mortality or blood pressure.Roerecke et al (2017) 31 reported low-certainty evidence that reduced alcohol consumption meaningfully decreased blood pressure (SBP MD -3.1 mmHg, 95% CI = -3.9 to -2.3; DBP MD -2.0 mmHg, 95% CI = -2.7 to -1.4).Substantial heterogeneity was largely explained when stratified by alcohol consumption at baseline.When consuming ≤2 drinks/day at baseline, reduction had no significant effect on blood pressure, whereas for those consuming ≥3 drinks/day at baseline alcohol reduction resulted in meaningful reductions in blood pressure.This effect
Bernasconi et al's (2020) 33 reanalysis of supplementation alone (average dose 1221 mg/day) found reduced risk of CHD events, fatal and non-fatal myocardial infarction, and CHD mortality, but not CVD events.Some outcomes were dose dependent, whereby higher doses of fish oil increased protection from CVD events and myocardial infarction.Although there was no clear evidence for an impact on CVD of including fish in the diet, when supplementation alone was considered, there was a small, dose-dependent, protective cardiovascular effect of fish oil supplementation, although number needed to treat for additional benefit was high. 32,33

Summary
This pragmatic narrative review of systematic reviews descriptively compares the effectiveness of six dietary strategies on cardiovascular outcomes in primary prevention populations.
Although no singular dietary strategy produced significant risk reductions across all CVD risk or mortality outcomes, all dietary strategies showed some potential benefit to meaningfully reduce CVD events and improve indirect CVD outcomes (for example, blood pressure), overall reducing CVD risk; however, there are some caveats to note within and across dietary strategies.
Three dietary strategies showed modest, significant reductions in cardiovascular events: energy deficit (relative risk reduction [RRR] 30%, 95% confidence interval [CI] = 13 to 43), Mediterranean-like diet (RRR 40%, 95% CI = 20 to 55), and salt substitution (RRR 30%, 95% CI = 7 to 48).The evidence certainty for energy deficit was very low, and the composite outcome used (cardiovascular complications + recommencing antihypertensives) may inflate the estimate. 24Mediterranean-like diets compared with low-fat diets showed hazard reduction, but this was for stroke only, and from one large study (reanalysis of PREDIMED).The salt substitution effect estimate is from a large trial 28 and is not a pooled estimate.Although the point estimates for sodium reduction (via salt reduction or salt substitution), alcohol reduction, or fish/fish oil also suggest reduced CVD event risk, the CIs were too imprecise to confirm or exclude important differences.
Most strategies that reported blood pressure showed small but meaningful reductions in SBP. 24,25,27,31,34The most promising strategy reviewed is salt substitution, a type of sodium-reduction strategy. 27The reported SBP reduction (6 mmHg to 9 mmHg) is sufficient to reduce CVD mortality, based on estimates that systolic reductions of 1 mmHg and 4 mmHg translate to reductions in CVD mortality of 2% and 8%, respectively. 35rtality outcomes were not available for all dietary strategies.For CVD mortality, salt substitution showed a significant reduction in CVD mortality, although this estimate is from a large trial, not a meta-analysis. 28Other available point estimates show small reductions in CVD mortality risk, but CIs are too imprecise to determine if these are meaningful.For all-cause mortality, although available point estimates show a small risk reduction, no strategy was associated with significantly reduced all-cause mortality.Some caveats across all dietary approaches should be highlighted.First, the length of the studies in each of the included reviews are arguably not long enough to capture lifelong prevention of CVD, and the distinction between intervention and follow-up can be ambiguous because of the assumption of permanent dietary change. 36Second, very few studies within the included reviews reported adherence to the dietary intervention or assessed dietary intake.Dietary intake is an important indication of intervention fidelity and is required to differentiate a dietary intervention's CVD risk-reduction effects compared with other physiological effects.Finally, very few RCTs reported outcome measures that reflect potential mechanisms of action of the dietary intervention, for example, whether energy deficit lowers CVD event risk via indirect lowering of blood pressure or via direct mechanisms.

Strengths and limitations
First, in the current study the authors elected to take a pragmatic approach to identify the most recent, highest-quality reviews for a selection of six relevant dietary strategies, with the goal to approximate effect sizes of each dietary strategy using best available evidence, not all available evidence.However, it is recognised that this approach does not provide a definitive measure of effect for each dietary strategy considering all available evidence, which would be a substantially larger, more resource-intensive project.
Second, the six dietary strategies were pragmatically selected by an expert committee for being relevant to clinical practice and postulated to effectively reduce absolute cardiovascular risk.However, it is highlighted that this is not an exhaustive list of dietary strategies that may reduce cardiovascular risk.Other examples of dietary strategies touted for beneficial cardiovascular effects that were outside the scope of this work include low carbohydrate and intermittent fasting, although recent Cochrane reviews suggest no major cardiovascular benefit. 37,38ird, only reviews of RCTs were examined because of quality limitations inherent to observational studies.However, RCT designs for dietary studies also have limitations, such as difficulties with blinding, control of intervention fidelity, long-term follow-up, uniformity, independence of effects, and control over comparator groups. 39,40 for recommending dietary strategies to manage cardiovascular risk in primary prevention populations.
Finally, there may be some overlap between dietary strategies, for example, the Mediterranean diet may be used as an energy deficit strategy.However, the authors of the current study suggest there is a distinction between dietary strategies with the primary aim of reducing weight via energy deficit (such as the reduced-calorie diets in the included energy deficit study) 24 and approaches such as the Mediterranean diet or DASH that may incidentally result in weight loss but this is not the primary aim.Although this justifies the authors' inclusion of both dietary strategies, it does not erase possible overlapping effects that should be considered in interpretation of findings.
Comparison with existing literature Given the social and economic burden of CVD, and the central role diet plays in the prevention and management of CVD, 1,5,6 consideration of how to support primary care patients to optimise their diet is essential.To provide positive, cost-effective, and sustainable dietary change, current evidence supports individual, community, and system-level strategies. 41,42The complexity of dietary interventions examined in this narrative review varied from simple behaviour change (for example, salt substitution that requires only direct replacement of regular salt with salt substitute) to more complex behaviour change (for example, Mediterranean or DASH diets that require adopting a comprehensive lifestyle approach).Dietary interventions that use behaviour change science are likely to better facilitate improvements in health outcomes. 43e selected reviews rarely considered the complexity of behaviour change required to successfully implement different dietary strategies, and clinicians should be guided by behaviour change theories when supporting patients to make dietary changes.There is inherent complexity in the variability of dietary behaviours, which are highly individual and influenced by personal (for example, taste and food preference), social (for example, familial/cultural preferences), and environmental factors (for example, ability to obtain, store, prepare, and cook foods appropriately). 44For GPs exploring diet change for cardiovascular health with patients, selecting dietary strategies to match their patients' needs, preferences, access, and social determinants of behaviour may maximise person-centred improvements in diet and health outcomes. 45trition care is not provided as often as clinically beneficial, and patients expect primary care clinicians to be competent in supporting them to optimise their diet. 46This narrative review shows that several dietary approaches are protective for cardiovascular health, so improving primary care clinicians' skills to support patient dietary change is worth pursuing.Clinicians can also advocate for positive nutrition policies at the community and system levels. 47This may include identifying misinformation, promoting the relevance and importance of healthy eating, and linking patients to evidence-based reputable sources of further support, including referral to a registered or accredited dietitian, which is shown to be clinically effective for reducing blood lipid levels. 48,49plications for practice All dietary interventions reviewed showed promising but modest effects on direct or indirect CVD outcomes and may be recommended by GPs for cardiovascular risk reduction, although, as discussed, there are some caveats to consider.The choice of dietary strategy for patients will depend on preferences and circumstances that may enable sustained behaviour change.Assessments of current weight (for energy deficit strategy), sodium intake (for salt reduction or substitution in people with hypertension), and alcohol intake (for high alcohol intake) should be conducted, and a Mediterranean diet may be useful whatever the background factors.Crucially, dietary strategies are not mutually exclusive: multiple compatible strategies (for example, sodium substitution and alcohol reduction) may be applied simultaneously to potentially maximise the benefit to cardiovascular health and absolute risk reduction.

Figure 1 .
Figure 1.Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) study flow diagram to select key reviews.DASH = Dietary Approaches to Stop Hypertension.Med = Mediterranean.SR = systematic review.

Figure 2 .
Figure 2. Summary of key cardiovascular outcomes for target dietary strategies, including forest plot of HR/RaR/ RR and associated 95% CIs to visually display effect of dietary strategies on CVD event outcomes.a Combined endpoint including CVD complications.b Stroke only.c NR in systematic review; results of subsequent randomised controlled trial, Neal et al (2021). 28BP = blood pressure.CVD = cardiovascular disease.DASH = Dietary Approaches to Stop Hypertension.DBP = diastolic blood pressure.HR = hazard ratio.MD = mean difference.Med = Mediterranean.NR = not reported.RaR = rate ratio.RR = risk ratio.SBP = systolic blood pressure.