The Journal of Nutrition · Nov 2026

Watermelon supplementation lowers blood pressure: A systematic review and meta-analysis of randomized controlled trials

By Lei Y, Diao J, Xu M et al.

Background Watermelon contains L-citrulline, polyphenols, lycopene, and other bioactive compounds that may influence vascular function and blood pressure regulation. This systematic review and meta-analysis evaluated the effects of watermelon supplementation on blood pressure, arterial stiffness-related indices, autonomic function, lipid metabolism, and anthropometric parameters. Methods PubMed, Web of Science, Embase, and the Cochrane Library were searched from inception to May 13, 2026. Randomized controlled trials comparing watermelon, watermelon juice, or watermelon extract with non-watermelon control interventions were included. Methodological quality was assessed using the revised Cochrane Risk of Bias tool, and evidence certainty was evaluated using the GRADE framework. Continuous outcomes were synthesized using mean differences with 95% confidence intervals. Crossover trials were analyzed as paired data, with each participant counted once; when paired variance information was unavailable, a within-participant correlation coefficient of 0.50 was assumed. Primary analyses used post-intervention values, and analyses based on change-from-baseline values were conducted as supplementary analyses. Heterogeneity was explored using subgroup analysis, meta-regression, restricted cubic spline analysis, and leave-one-out sensitivity analysis. Trial sequential analysis was conducted using unique participant counts. Results Sixteen randomized controlled trials were included. Five trials were judged to have a low risk of bias, eight had some concerns, and three had a high risk of bias. In the post-intervention primary analysis, watermelon supplementation was associated with lower systolic blood pressure (SBP; 12 studies; WMD, −5.53 mmHg; 95% CI, −8.99 to −2.06; P=0.002; I²=89.7%) and diastolic blood pressure (DBP; 12 studies; WMD, −2.55 mmHg; 95% CI, −3.77 to −1.33; P<0.001; I²=57.1%). Significant pooled reductions were observed for first systolic peak pressure and pulse pressure, whereas second systolic peak pressure, mean arterial pressure, reflection time, heart rate, pulse-wave velocity, augmentation index, and anthropometric and lipid outcomes were not statistically significant. Among autonomic outcomes, RMSSD and high-frequency power increased, whereas SDNN, low-frequency power, and the LF/HF ratio did not show statistically significant changes; these outcomes were each informed by two studies. Exploratory subgroup and study-level meta-regression analyses showed nominal differences across study characteristics, but these analyses were based on sparse, potentially confounded evidence and were not adjusted for multiple comparisons. Restricted cubic spline analyses found no evidence of nonlinearity. In the TSA analysis, 236 unique participants represented 15.9% and 28.9% of the required information sizes for SBP and DBP, respectively, and neither cumulative Z-curve crossed the trial sequential monitoring boundary. Potential small-study effects could not be excluded. According to the current GRADE profile, certainty was very low for SBP and low for DBP. Conclusions Watermelon supplementation was associated with lower SBP and DBP, but substantial heterogeneity, limited accrued information, and low certainty of evidence restrict confidence in the estimated effects. Evidence for haemodynamic, autonomic, arterial-stiffness, lipid, and anthropometric outcomes was inconsistent and was usually based on few studies. The subgroup and meta-regression findings are exploratory study-level observations and do not establish patient-level effect modification or an optimal intervention regimen. Larger, well-designed randomized controlled trials with prespecified analyses are needed.

doi.org/10.1016/j.tjnut.2026.101944Interventions

AgingResearch.News