Mediterranean Diet Leads the CRP Evidence, With Caveats

The Mediterranean diet has the best evidence among named diets for lowering C-reactive protein (CRP), a blood marker of systemic inflammation. That claim has limits: two recent umbrella reviews agree on the ranking, but neither proves that the diet prevents inflammatory disease, and certainty ranges from weak to low for many results.

The more surprising finding is who did not qualify. Intermittent fasting, ketogenic, Nordic, and Paleolithic diets showed no statistically significant inverse association with CRP in the CRP-focused umbrella review, and the broader review judged their evidence across inflammatory markers inconclusive or limited. Popularity did not translate into a settled anti-inflammatory result.

Two reviews, one hierarchy

Tran et al.'s 2024 umbrella review examined 27 systematic reviews containing 30 meta-analyses across seven dietary patterns: Mediterranean, vegetarian/vegan, energy-restricted, intermittent fasting, ketogenic, Nordic, and Paleolithic diets. In its randomized-trial-only analysis, the Mediterranean diet produced the largest CRP reduction, followed by vegetarian/vegan and energy-restricted diets. All three received a "weak" grade on the review's five-level certainty framework.

Weak does not mean the Mediterranean result failed statistical testing. It means the underlying evidence did not meet the review's higher thresholds for certainty. The abstract does not provide the pooled CRP change in mg/L, confidence intervals, or heterogeneity estimates for each diet, so readers should resist turning this ranking into a precise promise about how much their CRP will fall.

Reyneke et al.'s 2025 umbrella review reached the same broad conclusion using a different method. It covered 30 reviews, 225 primary studies, and 60 unique meta-analyses across 15 dietary patterns, grading certainty with GRADE. The Mediterranean diet had evidence of effects on CRP, interleukin-6 (IL-6), and adiponectin, with certainty ranging from high to low depending on the biomarker. Vegetarian diets had low-to-very-low-certainty evidence for CRP only. The other patterns had inconclusive or limited evidence across the biomarker panel.

Agreement between the reviews matters because they did not use interchangeable certainty systems. Tran et al.'s "weak" label and Reyneke et al.'s GRADE ratings are not a numerical replication. Still, both reviews put the Mediterranean pattern first, vegetarian eating behind it, and the fashionable low-carb and fasting patterns outside the evidence-backed tier.

CRP is useful, but it is not a clinical finish line

CRP rises with many conditions, including infection, obesity, and chronic inflammatory disease. Lower circulating CRP can signal less systemic inflammation, but it is a biomarker outcome, not proof that a diet prevents heart attacks, dementia, autoimmune disease, or any other clinical endpoint. The reviews evaluated circulating inflammatory markers; they did not establish treatment effects on disease outcomes.

A focused 2025 meta-analysis of randomized Mediterranean-diet trials offers direct corroboration of the leading pattern's inflammation signal in adults. It is especially useful because randomized trials can better separate the intervention from pre-existing differences between people who choose different diets. Yet randomization does not erase practical problems: dietary adherence varies, comparator diets differ, and trial duration can be short.

A broader 2021 review by Aleksandrova et al., which included observational and intervention studies, found that patterns rich in plant foods, healthy fats, and whole grains consistently associated with lower markers of inflammation and oxidative stress. Observational associations cannot prove cause and effect, but they fit the direction seen in Mediterranean-diet trials.

Why fasting and keto remain unsettled

The umbrella-level result does not show that intermittent fasting, keto, Nordic, or Paleo diets biologically cannot affect inflammation. It shows that the available evidence did not support a statistically significant CRP-lowering effect for those patterns in the general adult population in Tran et al.'s review, while Reyneke et al. found their multi-marker evidence inconclusive or limited.

This distinction is more than academic. A dedicated 2025 review and network meta-analysis of intermittent-fasting protocols—including alternate-day fasting, time-restricted feeding, and 5:2 fasting—described prior fasting meta-analyses as producing mixed results. Mixed findings can arise when protocols, participant health, weight change, and control diets vary enough that one pooled label, "intermittent fasting," obscures meaningful differences.

Population also changes the question. A 2024 randomized-trial meta-analysis of fasting and calorie restriction in people with overweight or obesity addresses a group in which weight loss itself may lower CRP. Similarly, a 2024 meta-analysis of ketogenic-diet trials in adults with overweight or obesity evaluates a population where changes in adiposity can confound any direct dietary effect on inflammation. Results in these higher-BMI groups should not be casually applied to general adults, nor should a general-population umbrella null be used to dismiss dietary treatment in obesity care.

The likely food-level explanation is plausible, not proven

The Mediterranean pattern usually emphasizes olive oil, fish, legumes, vegetables, fruit, nuts, and whole grains. Vegetarian patterns also tend to deliver more plant foods. That overlap has led researchers to ask whether polyphenols, fiber, fat quality, omega-3 intake, or their combination drives the inflammation signal.

There is a reasonable basis for studying polyphenols, but not for declaring them the entire answer. A 2025 meta-analysis of polyphenol-rich interventions in adults with overweight or obesity examined inflammatory and oxidative-stress markers alongside gut-microbiota outcomes. A separate 2025 meta-analysis in people with PCOS evaluated dietary polyphenol administration and inflammatory and oxidative-stress measures in that clinical population. These studies support a plausible role for polyphenol-rich foods, but neither proves that polyphenols explain the Mediterranean diet's effects in the general adult population.

Diet labels can hide this problem. A ketogenic diet built around processed meats, butter, and refined low-carb products differs materially from one built around fish, olive oil, nuts, and non-starchy vegetables. The label alone cannot identify the active ingredient—or guarantee an inflammatory outcome.

What the evidence can and cannot support

The best-supported practical choice is a Mediterranean-style eating pattern built around diverse plant foods, legumes, whole grains, nuts, olive oil, and fatty fish. It has the strongest available umbrella-level evidence for lower CRP and is the only named pattern in the broader review with evidence spanning CRP, IL-6, and adiponectin.

That recommendation should stay proportionate. In Tran et al.'s review, only 15 of 30 included meta-analyses met high-quality AMSTAR 2 criteria. Comparator diets were not detailed in the available abstracts, so the apparent benefit may differ when Mediterranean eating replaces a typical Western diet versus a low-fat or habitual diet. The abstracts also do not establish the minimum intervention duration, per-pattern trial counts behind the null findings, or the size and precision of each CRP change.

For someone trying to lower systemic inflammation through food, Mediterranean-style eating is the most defensible dietary bet. Treat fasting, keto, Nordic, and Paleo claims as unproven for CRP in general adults—not disproven forever, and not a substitute for evidence.