The effect of pecan consumption on cardiometabolic risk factors: a systematic review and meta-analysis of randomized controlled trials Authors: Ziwei Zhang, Mohammad Hassan Sohouli Published: 2026 ABSTRACT BackgroundPecans are rich in monounsaturated fats, phytosterols, fiber, and polyphenols, suggesting potential cardiometabolic benefits. However, evidence from randomized controlled trials (RCTs) remains inconsistent, and no prior synthesis has comprehensively quantified their effects on major metabolic risk factors. Therefore, we aimed to systematically evaluate the effects of pecan consumption on cardiometabolic outcomes in adults using data from RCTs.MethodsA systematic search of PubMed, Scopus, Web of Science, Embase, and Cochrane CENTRAL was conducted through November 2025. Eligible studies were RCTs assessing pecan consumption versus control diets and reporting changes in glycemic indices, anthropometric measures, lipid profiles, or blood pressure. Random-effects models were used to calculate weighted mean differences (WMDs).ResultsA total of 10 randomized controlled trials were included. Pecan consumption significantly reduced total cholesterol (7 studies; WMD: − 8.49 mg/dL; 95% CI: − 12.61 to − 4.37; I²=0.0%), LDL-cholesterol (7 studies; WMD: − 7.42 mg/dL; 95% CI: − 10.83 to − 4.01; I²=0.0%), and triglycerides (7 studies; WMD: − 14.06 mg/dL; 95% CI: − 22.37 to − 5.75; I²=0.0%). Little evidence of an effect was observed for HDL-cholesterol (7 studies; I²=0.0%), fasting glucose (4 studies; I²=71.2%), insulin (4 studies; I²=44.6%), HOMA-IR (3 studies; I²=12.8%), body weight (4 studies; I²=0.0%), BMI (3 studies; I²=0.0%), waist circumference (4 studies; I²=0.0%), systolic blood pressure (4 studies; I²=88.8%), or diastolic blood pressure (4 studies; I²=89.6%). Subgroup analyses showed consistent LDL-C reductions across all doses and durations, whereas triglyceride reductions were evident only in interventions > 4 weeks. Sensitivity analyses confirmed the robustness of findings, and no publication bias was detected.ConclusionsPecan consumption was associated with improvements in total cholesterol, LDL-cholesterol, and triglycerides but does not significantly influence glycemic control, adiposity, or blood pressure. Pecans represent a beneficial lipid-modifying component of cardioprotective dietary patterns. Larger, long-term RCTs in high-risk populations are warranted.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13098-026-02212-1. Introduction Cardiometabolic diseases (including atherosclerotic cardiovascular disease and type 2 diabetes mellitus) remain leading causes of global morbidity and mortality, and dietary patterns are central determinants of both incidence and progression of these conditions [1, 2]. Nuts are nutrient-dense whole foods, rich in unsaturated fatty acids, dietary fiber, plant sterols, tocopherols, and polyphenolic compounds, that have been consistently associated with favorable cardiometabolic outcomes in prospective cohorts and randomized trials. Large umbrella reviews and recent meta-analyses report that regular nut consumption (≈ 15–40 g/day) is associated with lower risks of cardiovascular disease, coronary heart disease and all-cause mortality, and that nut interventions reliably improve intermediate risk markers such as total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-C) and triglycerides (TG) [3–6]. Despite this robust literature for nuts collectively, evidence for individual nut species remains uneven. Pecans (Carya illinoinensis) contain high amounts of monounsaturated fats, tocopherols, plant sterols, and polyphenols, nutrients that may contribute to favorable cardiometabolic effects [7–9]. However, randomized controlled trials (RCTs) and mechanistic studies specifically focused on pecans have been fewer and smaller than for other nuts, producing some heterogeneity in reported outcomes (especially for glycemic endpoints and blood pressure) and limiting confident, pecan-specific dietary guidance [8, 10–12]. Randomized feeding and free-living intervention trials of pecan supplementation have reported consistent cholesterol-lowering effects (reductions in total cholesterol, LDL-C and triglycerides) and some favorable changes in lipoprotein subfractions and postprandial lipid responses; yet effects on fasting glycemia, anthropometry and vascular function are mixed across trials of varying duration, dose and participant cardiometabolic risk [8–10, 13, 14]. Given the established cardiovascular benefits of nuts generally, the biologic plausibility offered by pecans’ unique nutrient and phytochemical composition, and the presence of several modest-size RCTs with divergent endpoints and follow-up times, there is a clear and timely need for a focused systematic review and meta-analysis of randomized trials evaluating pecan on cardiometabolic risk factors. The current manuscript addresses this evidence gap by systematically identifying RCTs of pecan intake and applying meta-analytic methods to produce precise, clinically interpretable estimates that can guide future trials and dietary guidance. Methods Search strategy This systematic review and meta-analysis was performed in accordance with PRISMA 2020 guidelines [15]; a comprehensive electronic search was conducted in PubMed/MEDLINE, Embase, Scopus, Web of Science and Cochrane CENTRAL from database inception through 30 November 2025 (In addition, an updated supplementary search was performed prior to manuscript submission to identify any newly published eligible studies, and no additional randomized controlled trials meeting the inclusion criteria were identified) without language or publication-date restrictions, supplemented by searches of gray literature sources (Google Scholar, OpenGrey, ProQuest Dissertations) and manual screening of reference lists of included studies and relevant reviews to capture additional trials; search strategies combined MeSH/Emtree headings and free-text terms for pecans and cardiometabolic outcomes (e.g., “pecan” OR “Carya illinoinensis” OR “tree nut*” OR “nut-enriched diet”) AND (“cholesterol” OR “LDL-C” OR “HDL-C” OR “triglyceride*” OR “blood glucose” OR “insulin” OR “HOMA-IR” OR “blood pressure” OR “body weight” OR “BMI” OR “waist circumference”) AND trial design filters (“randomized controlled trial” OR “randomized” OR “clinical trial” OR “cross-over”), and all retrieved records were imported into EndNote for deduplication prior to screening. Eligibility criteria Two reviewers independently screened titles, abstracts and full texts according to predefined PICOS criteria and included randomized controlled trials (parallel or crossover) in adults (≥ 18 years) that administered pecans as the sole or primary dietary intervention and compared them with a nut-free, habitual, or isocaloric non-pecan control diet, reporting at least one prespecified cardiometabolic outcome [glycemic indices: fasting blood glucose (FBG), insulin, Homeostatic Model Assessment for Insulin Resistance (HOMA-IR); lipid parameters: TC, LDL-C, high density lipoprotein-cholesterol (HDL-C), TG; anthropometrics: body weight, body mass index (BMI), waist circumference (WC); blood pressure: systolic/diastolic blood pressure (SBP/DBP)]; studies were excluded if they were non-randomized, combined pecans with other active interventions whose effects could not be separated, evaluated mixed-nut interventions without isolating pecan effects, lacked extractable quantitative data, were animal/in vitro reports, or were duplicate/overlapping publications (in which case the most complete or longest follow-up report was retained). Data extraction Two investigators independently extracted data using a standardized form and resolved disagreements by consensus with a third reviewer; extracted items included first author, year, country, study design (parallel vs. crossover), number randomized and analyzed per arm, participant characteristics (mean age, sex distribution, baseline BMI, health status), intervention details (pecan dose, form, frequency, comparator diet, duration), outcome measures and units, and numerical data (means and SDs at baseline and follow-up or mean changes and corresponding SDs); when only SEMs or 95% CIs were reported, SDs were calculated as SD = SEM × √n or derived from CIs, and when only baseline and final means (without change SD) were available, change-SDs were computed using standard formulas assuming an appropriate correlation coefficient informed by prior nut intervention trials. Quality assessment Risk of bias for individual RCTs was independently assessed by two reviewers using the Cochrane Risk-of-Bias 2.0 (RoB 2) tool [16] across the domains of randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of reported results, with each domain judged as low risk, some concerns, or high risk and disagreements resolved by discussion with a third reviewer. Data synthesis and statistical analysis Statistical pooling was performed in STATA v12.0 using weighted mean differences (WMDs) with 95% confidence intervals for continuous outcomes and a random-effects model (DerSimonian–Laird) chosen a priori to accommodate clinical and methodological heterogeneity. If standard deviations (SDs) were missing, they were estimated using standard formulas from reported standard errors (SE), confidence intervals, or p-values. For example, SDs were estimated by multiplying SE by the square root of the sample size (n), or calculated from baseline and final SDs using the following formula: SD_change = √[(SD_baseline^2 + SD_final^2) - (2 * R * SD_baseline * SD_final)], assuming a correlation coefficient (R) of 0.5 when not reported [17, 18]. To evaluate the robustness of imputations based on assumed within-group correlations, additional sensitivity analyses were conducted using alternative plausible correlation coefficients (R = 0.3 and R = 0.7), in accordance with Cochrane Handbook recommendations. When SDs were not directly reported, but p-values were available, we calculated SDs from the corresponding test statistics (e.g., t-value) or from SEs derived using the reported p-value and sample size, in accordance with the Cochrane Handbook guidance. Between-study heterogeneity was evaluated using Cochran’s Q (with p < 0.10 considered suggestive of heterogeneity) and the I² statistic, interpreted as negligible (0–25%), low (26–50%), moderate (51–75%) or high (> 75%) [19]. Subgroup analyses explored potential sources of heterogeneity by intervention duration (≤ 4 weeks vs. >4 weeks) and pecan dose (< 60 g/day vs. ≥60 g/day) when reported. The duration threshold was selected because four weeks represented the most common minimum intervention period among the included trials and allowed comparison of shorter- versus longer-term interventions. The dose threshold was chosen based on the distribution of pecan intake across studies, with approximately 60 g/day representing the median and most frequently investigated intake level, thereby providing balanced subgroup comparisons. Additional sensitivity analyses were performed by excluding studies with fewer than 20 participants per study arm and separately excluding studies judged as having a high overall risk of bias according to the RoB 2 tool, in order to evaluate the robustness of pooled estimates against small-study effects and methodological limitations. Publication bias was assessed visually with funnel plots and statistically with Egger’s regression test [20], with the trim-and-fill method applied where asymmetry suggested potential missing studies. When necessary, outcome units were harmonized and conversions applied to ensure comparability across studies, and all analyses were two-sided with α = 0.05. For crossover trials, paired comparisons were handled in accordance with Cochrane Handbook recommendations. Change-from-baseline values were preferentially extracted, and within-subject correlation was accounted for when estimating the variance of treatment effects. When the correlation coefficient was not reported, a conservative correlation coefficient (R = 0.5) was assumed to calculate SDs of change scores and avoid unit-of-analysis errors associated with treating crossover studies as independent parallel-group comparisons . The certainty of evidence for each outcome was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework, which classifies evidence as high, moderate, low, or very low based on risk of bias, inconsistency, indirectness, imprecision, and publication bias [21]. Results Figure 1 presents the flow diagram of the study selection process and the application of exclusion criteria. A total of 619 records were initially retrieved through systematic searches of electronic databases. Following the removal of duplicate records, 447 unique studies were retained for screening. Title and abstract screening led to the exclusion of 428 records that did not meet the predefined eligibility criteria. Subsequently, 29 articles underwent full-text assessment for eligibility. Of these, 19 studies were excluded for reasons including the absence of an appropriate control group or insufficient outcome data. Ultimately, 10 randomized controlled trials [7, 9–11, 13, 22–26] fulfilled all inclusion criteria and were included in the final quantitative meta-analysis. Fig. 1Flow chart of study selection process Flow chart of study selection process Descriptive Summary of eligible studies (based on Table 1) A total of 10 randomized controlled trials (RCTs) [7, 9–11, 13, 22–26] were included in the final analysis, conducted predominantly in the United States and Brazil. The studies were published between 2000 [26] and 2025 [10] and collectively evaluated the effects of pecan consumption on cardiometabolic outcomes across diverse adult populations. Sample sizes varied substantially across trials, ranging from 8 [24] to 69 [10] participants per study arm, reflecting considerable heterogeneity in study scale and statistical power. The populations enrolled encompassed a broad spectrum of cardiometabolic risk profiles, including patients with stable coronary artery disease [11, 23], adults with overweight and obese [9], hyperlipidemic individuals [24], sedentary adults [22], healthy men and women [7, 13, 25, 26], and adults at increased cardiometabolic risk [10]. Mean participant age ranged from approximately 37 [26] years in younger healthy cohorts to over 62 [9] years in studies involving older or higher-risk populations. Sex distribution varied markedly, with male representation ranging from 21% [26] to 81% [9], and several studies enrolling mixed-sex populations, while others focused primarily on men or postmenopausal women. Most trials employed a parallel-group [7, 10, 11, 13, 22–24, 26] randomized design, with only two studies adopting a cross-over design [9, 25]. Intervention durations ranged from 4 [7, 9, 13, 25] to 12 weeks [10, 11, 23], with the majority of trials clustering around 4–8 weeks, and a smaller number extending to 12 weeks. These durations reflect short- to medium-term dietary interventions rather than long-term exposure. Across studies, pecan intake was administered in relatively consistent doses, typically ranging from 30 g/day [11, 23] to 68 g/day [7, 9, 13, 22, 24–26], either as a fixed daily quantity or as a percentage of total energy intake (e.g., 15% of total daily energy). Some trials instructed participants to replace habitual snacks with pecans, while others incorporated pecans into a pecan-enriched or heart-healthy dietary pattern. Control groups generally followed usual diets, nut-free diets, or isocaloric control diets, with several studies explicitly instructing control participants to avoid all nuts. Most interventions were conducted under free-living conditions, although dietary guidance and monitoring varied across studies. Baseline body mass index (BMI) ranged from values consistent with normal weight in healthy adult cohorts to overweight and obese classifications in higher-risk populations. Table 1Characteristics of eligible studies.First author et alYearsCountry PopulationMean ageyearSex (male %)Sample size studyFollow up of intervention (Weeks)Type of RCTsIntervention groupControl groupInterventionControl1. Campos2019BrazilPatients with stable coronary artery disease (CAD)59.677686712ParallelHealthy diet plus 30 g day of pecan nutsHealthy diet2. Cogan2023USAAdult men and postmenopausal women5929.521234Parallel68 g of pecans/dAvoided all nuts (control3. McKay2018USAOverweight and obese adults62.6580.715114Cross-over desigPecan-rich diet (15% of the total energy (~42.5 g/2000 kcal)).Isocaloric control diet4. Eastman2005USAhyperlipidemic individuals4623.5898Parallel68g of pecans/daySelf-selected diets ( not consuming nut)5. Morgan2000USAPeopls with normal lipid levels37211098Parallel68g of pecans/dayUsual diets ( not consuming nut)6. Rajaram2001USAHealthy men and women3860.823234Cross-over desig68g of pecans/dayUsual diets ( not consuming nut)7. Hart2025USAAdults at increased risk of cardiometabolic diseases4641696912Parallel57 g/d of pecans and instructed to replace the snacks usuallyUsual diet8. Santos2021BrazilPatients with stable coronary artery disease (CAD)59.677686712ParallelHealthy diet plus 30 g day of pecan nutsHealthy diet9. Cogan2023USAHealthy aging adults5929.521204Parallel68 g of pecans/dAvoided all nuts (control10. Guarneiri2022USASedentary men and women4730.630328ParallelADD (n 30) consumed pecans (68 g/d) as part of a free-living dietNut-free control group (n 32) Table 1Characteristics of eligible studies.First author et alYearsCountry PopulationMean ageyearSex (male %)Sample size studyFollow up of intervention (Weeks)Type of RCTsIntervention groupControl groupInterventionControl1. Campos2019BrazilPatients with stable coronary artery disease (CAD)59.677686712ParallelHealthy diet plus 30 g day of pecan nutsHealthy diet2. Cogan2023USAAdult men and postmenopausal women5929.521234Parallel68 g of pecans/dAvoided all nuts (control3. McKay2018USAOverweight and obese adults62.6580.715114Cross-over desigPecan-rich diet (15% of the total energy (~42.5 g/2000 kcal)).Isocaloric control diet4. Eastman2005USAhyperlipidemic individuals4623.5898Parallel68g of pecans/daySelf-selected diets ( not consuming nut)5. Morgan2000USAPeopls with normal lipid levels37211098Parallel68g of pecans/dayUsual diets ( not consuming nut)6. Rajaram2001USAHealthy men and women3860.823234Cross-over desig68g of pecans/dayUsual diets ( not consuming nut)7. Hart2025USAAdults at increased risk of cardiometabolic diseases4641696912Parallel57 g/d of pecans and instructed to replace the snacks usuallyUsual diet8. Santos2021BrazilPatients with stable coronary artery disease (CAD)59.677686712ParallelHealthy diet plus 30 g day of pecan nutsHealthy diet9. Cogan2023USAHealthy aging adults5929.521204Parallel68 g of pecans/dAvoided all nuts (control10. Guarneiri2022USASedentary men and women4730.630328ParallelADD (n 30) consumed pecans (68 g/d) as part of a free-living dietNut-free control group (n 32) Characteristics of eligible studies. Risk of bias assessment (based on Table 2) The overall risk of bias across the included RCTs was judged to be low to moderate, according to the Cochrane Collaboration’s risk of bias assessment tool. Most studies were rated as low risk for random sequence generation, suggesting adequate randomization procedures. Selective reporting bias was also generally assessed as low, indicating that prespecified outcomes were appropriately reported. However, blinding of participants and personnel was frequently rated as high or unclear risk, which is common in dietary intervention trials involving whole foods such as nuts. The distinct sensory characteristics of pecans and the nature of dietary instructions likely limited the feasibility of effective blinding, thereby increasing the potential for performance bias. Allocation concealment was often reported as unclear, reflecting insufficient methodological detail in several publications. Similarly, incomplete outcome data and blinding of outcome assessment were commonly categorized as unclear, particularly in studies with small sample sizes or short follow-up durations. The high risk of bias assigned to the domain “deviations from intended interventions” largely reflects the lack of participant blinding and the free-living nature of most dietary trials, rather than documented non-adherence or protocol deviations. Only a limited number of trials [23, 24] were judged to have a high overall risk of bias, whereas the majority were classified as low or unclear risk across most domains. Overall, the methodological quality of the included studies can be considered acceptable, though not without limitations. These findings underscore the importance of cautious interpretation of pooled results and support the application of sensitivity analyses to assess the robustness of the meta-analytic conclusions. We acknowledge that dietary intervention trials inherently involve challenges related to food substitution and participant blinding; therefore, judgments for the domain “deviations from intended interventions” were intentionally applied conservatively in accordance with RoB 2 guidance. Table 2Risk of bias assessment according to the Cochrane risk of bias assessment tool Table 2Risk of bias assessment according to the Cochrane risk of bias assessment tool Risk of bias assessment according to the Cochrane risk of bias assessment tool Meta-analysis Glucose metabolism outcomes [7, 9, 10, 23] The meta-analysis findings indicated that pecan consumption did not provide clear evidence of an effects on glycemic indices. No meaningful changes were observed in FBG (WMD: − 2.28 mg/dL; 95% CI: − 6.19 to 1.64; P = 0.255; I²=71.2%), insulin levels (WMD: − 0.78 µU/mL; 95% CI: − 1.87 to 0.31; P = 0.161; I²=44.6%), or HOMA-IR (WMD: − 0.30; 95% CI: − 0.76 to 0.15; P = 0.193; I²=12.8%) compared with the control group (Fig. 2). Notably, heterogeneity for FBG was substantial (I² = 71.2%), likely reflecting differences in baseline metabolic status, comparator diets, and intervention characteristics across the limited number of available trials. Fig. 2Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) glucose, (b) Insulin, (c) HOMA-IR. WMD: weighted mean Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) glucose, (b) Insulin, (c) HOMA-IR. WMD: weighted mean Anthropometric measurements [7, 10, 22, 23] Regarding anthropometric indicators, the analysis showed no significant differences between the pecan and control groups. Body weight (WMD: − 1.59 kg; 95% CI: − 4.66 to 1.48; P = 0.311; I²=0.0%), BMI (WMD: − 0.09 kg/m²; 95% CI: − 1.26 to 1.09; P = 0.885; I²=0.0%), WC (WMD: − 0.44 cm; 95% CI: − 2.66 to 1.79; P = 0.698; I²=0.0%) remained unchanged between groups (Fig. 3). Fig. 3Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) weight, (b) BMI, and (c) WC. WMD: weighted mean Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) weight, (b) BMI, and (c) WC. WMD: weighted mean Lipid profile parameters [9–11, 13, 24–26] The effects of pecan intake on HDL-C were modest and not statistically significant (WMD: 0.97 mg/dL; 95% CI: − 0.25 to 2.20; P = 0.120; I²=0.0%). Interestingly, significant reductions were observed in TC (WMD: − 8.49 mg/dL; 95% CI: − 12.61 to − 4.37; P < 0.001; I²=0.0%), TG (WMD: − 14.06 mg/dL; 95% CI: − 22.37 to − 5.75; P = 0.001; I²=0.0%), and LDL-C levels (WMD: − 7.42 mg/dL; 95% CI: − 10.83 to − 4.01; P < 0.001; I²=0.0%) following pecan consumption (Fig. 4). Fig. 4Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) total cholesterol, (b) LDL-C, (c) HDL-C, (d) TG. WMD: weighted mean Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) total cholesterol, (b) LDL-C, (c) HDL-C, (d) TG. WMD: weighted mean Blood pressure indicators [7, 9, 10, 22] The effect of pecan consumption on blood pressure was assessed using a random-effects model. However, no significant changes were detected in SBP (WMD: − 1.40 mmHg; 95% CI: − 4.22 to 1.41; P = 0.329; I²=88.8%) or DBP (WMD: 1.08 mmHg; 95% CI: − 0.82 to 2.97; P = 0.265; I²=89.6%) (Fig. 5). Notably, heterogeneity for both SBP and DBP was very high (I² >85%), indicating substantial between-study variability. Therefore, the pooled null findings for blood pressure outcomes should be interpreted cautiously, as only four studies were available for analysis and substantial clinical and methodological heterogeneity existed across trials, including differences in participant cardiometabolic status, intervention characteristics, and blood pressure assessment procedures. Fig. 5Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) SBP, and (b)DBP. WMD: weighted mean Forest plots from the meta‐analysis of clinical trials investigating the effects Pecan on (a) SBP, and (b)DBP. WMD: weighted mean Subgroups analysis Subgroup analyses (Based on the subgroup analysis results from Supplementary Table 2) demonstrated that pecan consumption significantly reduced TC across all subgroups, including interventions lasting ≤ 4 weeks (WMD: − 10.49 mg/dL; 95% CI: − 19.26 to − 1.72) and > 4 weeks (WMD: − 7.92 mg/dL; 95% CI: − 12.59 to − 3.25), as well as at intake levels < 60 g/day (WMD: − 7.54 mg/dL; 95% CI: − 12.08 to − 3.21) and ≥ 60 g/day (WMD: − 13.81 mg/dL; 95% CI: − 24.95 to − 2.67). For triglycerides, statistically significant reductions were observed in studies with intervention durations > 4 weeks (WMD: − 16.49 mg/dL; 95% CI: − 26.42 to − 6.55). However, these subgroup findings should be interpreted cautiously because study duration and dose categories were not independent across trials, and the limited number of available studies precluded reliable multivariable subgroup or meta-regression analyses to disentangle these effects. LDL-C levels decreased significantly across all subgroups, including ≤ 4 weeks (WMD: − 10.72 mg/dL; 95% CI: − 18.69 to − 2.76), > 4 weeks (WMD: − 6.68 mg/dL; 95% CI: − 10.45 to − 2.90), < 60 g/day (WMD: − 6.42 mg/dL; 95% CI: − 10.03 to − 2.81), and ≥ 60 g/day (WMD: − 15.91 mg/dL; 95% CI: − 26.41 to − 5.41). In contrast, HDL-C did not show any significant changes in any subgroup, as all confidence intervals included zero, whether for durations ≤ 4 weeks (WMD: − 0.06 mg/dL; 95% CI: − 2.75 to 2.63) and > 4 weeks (WMD: 1.25 mg/dL; 95% CI: − 0.14 to 2.63), or intake levels < 60 g/day (WMD: 0.98 mg/dL; 95% CI: − 0.51 to 2.47) and ≥ 60 g/day (WMD: 0.77 mg/dL; 95% CI: − 3.27 to 4.81). Sensitivity analysis A leave-one-out sensitivity analysis was conducted for all pooled outcomes, including glycemic parameters (FBG, insulin, and HOMA-IR), lipid profile measures, anthropometric outcomes, and blood pressure indices, to determine the influence of individual studies on the pooled effect estimates. The overall findings remained stable after sequential exclusion of each study, with no single trial materially altering the direction or significance of the pooled results. Additional sensitivity analyses were conducted after excluding studies with fewer than 20 participants per arm and studies classified as having a high overall risk of bias. The pooled estimates for total cholesterol, LDL-cholesterol, and triglycerides remained statistically significant and directionally consistent with the primary analyses, although effect sizes were modestly attenuated and confidence intervals widened slightly due to reduced sample size. No materially different findings were observed for glycemic, anthropometric, or blood pressure outcomes. These analyses support the robustness of the primary findings and suggest that the overall conclusions were not disproportionately influenced by small studies or higher-risk trials. Additional sensitivity analyses using alternative assumed correlation coefficients (R = 0.3 and R = 0.7) for imputed change-score SDs produced materially similar pooled estimates, indicating that the main findings were robust to the selected correlation assumptions (Supplementary Figs. 1–4 and Supplementary Table 3). Publication bias Visual inspection of the funnel plots revealed no apparent asymmetry, suggesting a low risk of publication bias in relation to the effects of pecan interventions. This observation was further supported by Egger’s regression test, which showed no statistically significant evidence of publication bias for any of the analyzed parameters. The P-values were as follows: weight (P = 0.624), BMI (P = 0.174), WC (P = 1.00), LDL-C (P = 0.881), HDL-C (P = 0.453), TC (P = 0.453), TG (P = 0.176), insulin (P = 0.174), HOMA-IR (P = 0.117), SBP (P = 0.142), and DBP (P = 0.624). However, the result for FBG was borderline/significant (P = 0.050), suggesting the possibility of small-study effects or publication bias for this parameter. Nevertheless, trim-and-fill analysis did not identify any missing studies and did not materially change the pooled effect estimate for FBG. The corresponding funnel plots are presented in Supplementary Figs. 5–8. Moreover, the trim-and-fill analysis detected no missing studies, further confirming the robustness of the publication bias assessment. The certainty of evidence for each outcome was assessed using the GRADE approach (Supplementary Table 1). The certainty of evidence for lipid outcomes, including total cholesterol, LDL-cholesterol, triglycerides, and HDL-cholesterol, was judged to be low. Although the effects were generally consistent across studies and statistical heterogeneity was low for several lipid parameters, the certainty of evidence was downgraded due to imprecision related to modest sample sizes and short intervention durations, as well as indirectness arising from the inclusion of predominantly healthy or mildly at-risk populations and the use of surrogate cardiometabolic outcomes. The certainty of evidence for glycemic outcomes (fasting blood glucose, insulin, and HOMA-IR) was rated as low, primarily due to imprecision and indirectness, as most included trials enrolled metabolically healthy or mildly at-risk participants. Similarly, the certainty of evidence for anthropometric outcomes (body weight, BMI, and waist circumference) and blood pressure outcomes (SBP and DBP) was judged to be low, owing to imprecision, population indirectness, and, for blood pressure, substantial inconsistency across studies. No evidence of publication bias was detected for any outcome. Discussion The present systematic review and meta-analysis synthesized evidence from RCTs examining the effects of pecan consumption on cardiometabolic health outcomes, including glycemic control, anthropometric measures, lipid profiles, and blood pressure, in adult populations. Overall, consumption of pecans led to statistically significant reductions in total cholesterol, LDL-cholesterol, and triglycerides, whereas no significant effects were observed for HDL-cholesterol, fasting glucose, insulin, HOMA-IR, body weight or adiposity markers, or systolic and diastolic blood pressure. In this meta-analysis, pecan consumption did not produce statistically significant changes in FBG, fasting insulin, or HOMA-IR compared with control diets. These findings align with broader evidence from nut intervention studies, in which glycemic benefits tend to be small and often confined to populations with established insulin resistance or longer intervention durations [3–5, 12]. Mechanistically, there are plausible pathways by which pecans could influence glucose metabolism. Pecans are rich in monounsaturated fatty acids (MUFA), especially oleic acid, along with dietary fiber and antioxidant polyphenols. These nutrients have been hypothesized to improve insulin sensitivity by attenuating postprandial glycemia, modulating incretin hormone responses, and reducing oxidative stress and low-grade inflammation that contribute to insulin resistance [27–30]. Several aspects of the included RCTs may explain the absence of significant glycemic effects. First, many studies were short in duration (often ≤ 4–8 weeks). Improvements in fasting markers of glucose homeostasis generally require sustained metabolic adaptation, which may not occur in shorter interventions [31]. Second, in many trials pecan intake was added without strict controls on overall caloric intake or dietary composition, potentially diluting any glycemic benefit due to compensatory dietary adjustments. Third, most participants were metabolically healthy or at low cardiometabolic risk at baseline, reducing the potential to observe clinically meaningful improvements in glycemic control. Importantly, while this meta-analysis addresses fasting indices, pecan intake may still influence postprandial glucose responses, a hypothesis supported by studies on other nuts where postprandial glycemia improves even in the absence of changes in fasting glucose [28, 32]. The meta-analysis revealed no significant effects of pecan consumption on body weight, BMI, or WC. These findings are consistent with numerous nut intervention studies, where the high energy density of nuts does not translate into weight gain when incorporated into habitual diets [5, 33, 34]. The neutral effects of pecan consumption on body weight and adiposity are consistent with established mechanisms observed for nuts in general. Despite their high energy density, pecans promote satiety and partial dietary energy compensation, leading to reduced subsequent energy intake. In addition, the complex food matrix of nuts limits complete fat absorption due to cell wall encapsulation, resulting in lower metabolizable energy than predicted. Finally, MUFA-rich foods such as pecans may modestly increase postprandial energy expenditure and fat oxidation. Collectively, these mechanisms help explain why pecan supplementation does not lead to weight gain and may support weight maintenance, even in free-living dietary interventions [8, 35, 36]. This reinforces the concept that weight maintenance, rather than weight gain, may be a typical outcome of nut supplementation, even when study durations extend for several weeks. Additionally, the design of many RCTs, often isocaloric or with instructions to maintain habitual activity and energy intake, reduces the likelihood of observing weight change regardless of potential metabolic benefits [33, 34]. In contrast, interventions that combine pecan intake with caloric restriction or enhanced physical activity may yield weight loss benefits, but such combined interventions were not the focus of the current analysis [37]. In clinical practice, incorporating pecans into meals or snacks is unlikely to contribute to weight gain and may support weight stability. Their inclusion can be recommended without the fear of adverse effects on body weight, particularly when part of calorie-controlled diets. The most robust and consistent findings from this meta-analysis relate to improvements in the lipid profile with pecan consumption. Specifically, TC, LDL-cholesterol, and TG were significantly reduced, while HDL-cholesterol did not change meaningfully. These findings are in line with several pecan-specific RCTs. Morgan et al. (2000) observed LDL-C lowering in participants consuming pecans, and McKay et al. (2018) reported favorable changes in lipoprotein profiles with pecan-enriched diets. Additionally, Campos and colleagues (2020) showed lipid improvements when pecans were included as part of a heart-healthy dietary pattern in individuals with coronary artery disease [5, 7, 9, 11, 26]. Mechanistically, these lipid-lowering effects can be attributed to multiple nutritional properties of pecans: MUFA content, phytosterols, soluble fiber, and antioxidants [30, 38–40]. Pecan consumption was also associated with a significant reduction in triglycerides. The effect on TG appears more time-dependent than that on LDL-C, with larger reductions observed in trials of longer duration (> 4 weeks). This temporal pattern is biologically plausible [41]. Possible mechanisms for TG reduction include improved insulin sensitivity at the hepatic level (even if not sufficient to change fasting insulin or HOMA-IR), shifts in fatty acid oxidation toward greater utilization of MUFA, and reductions in de novo lipogenesis with substitution of refined carbohydrates or saturated fats by pecans [42]. Unlike HDL-C responses to other dietary interventions (e.g., exercise or omega-3 fatty acids), pecan consumption did not significantly raise HDL-cholesterol. This is consistent with a large body of nut research showing that HDL-C is relatively insensitive to specific food interventions in the absence of weight loss or pharmacologic treatment [5, 43]. HDL-C appears to be more strongly influenced by global lifestyle patterns rather than by single dietary components. For instance, Guasch-Ferré et al. (2023) reported similar findings for total nut interventions across RCTs, reinforcing that the lipid benefits of pecans are part of a broader pattern among tree nuts [12]. This meta-analysis did not detect significant effects of pecan consumption on SBP and DBP. Unlike certain dietary modifications (e.g., increased potassium intake, DASH dietary pattern), pecans do not contain exceptionally high levels of potassium or nitrate, nutrients implicated in blood pressure reduction [44, 45]. Therefore, from a mechanistic standpoint, significant blood pressure lowering attributable solely to pecan intake is not strongly supported. Additionally, the relatively short duration of many trials and the inclusion of participants without hypertension at baseline further limit the potential to detect meaningful effects [5, 46]. Observational data and nut-focused RCTs show mixed or small effects on blood pressure, suggesting that nuts alone are unlikely to be potent antihypertensive agents in the absence of broader dietary changes or sodium reduction [3, 47]. Additional sources of heterogeneity may include variation in baseline blood pressure status across studies, as some trials enrolled normotensive healthy adults whereas others included participants with overweight, obesity, coronary artery disease, or elevated cardiometabolic risk. Differences in intervention duration, pecan dose, comparator diets, dietary counseling intensity, and blood pressure measurement protocols may also have contributed to the observed variability. Furthermore, the small number of available studies limited the ability to conduct meaningful subgroup analyses or meta-regression to formally identify effect modifiers. Consequently, the pooled estimates for SBP and DBP should be interpreted cautiously. Subgroup and moderator insights Subgroup analyses revealed that reductions in LDL-C and total cholesterol occurred across a range of doses (< 60 g/day and ≥ 60 g/day) and study durations (≤ 4 weeks and > 4 weeks), supporting the robustness of the lipid effect. Triglyceride reductions were more prominent in longer studies, reinforcing the idea that TG metabolism may require sustained dietary engagement to shift appreciably. Although formal meta-regression analyses were not feasible due to limited numbers of trials and reporting variability, certain patterns are biologically plausible. For example, baseline LDL-C levels may modulate the degree of LDL-C lowering, with those having higher initial LDL-C showing greater absolute reductions. Similarly, metabolic phenotype (e.g., insulin resistance status) and adiposity measures could influence triglyceride and glycemic responses but require targeted examination in future studies. The consistency of lipid improvements across diverse populations and settings suggests that pecan intake exerts a generalizable effect, although individual responses may vary based on genetic, metabolic, and environmental factors. Clinical implications From a clinical perspective, the observed reductions in LDL-cholesterol may be clinically relevant when sustained over time, particularly in the context of established associations between LDL-C lowering and cardiovascular risk reduction. Large-scale meta-analyses of statin and non-statin interventions demonstrate that each 1 mmol/L (≈ 38.7 mg/dL) decrease in LDL-C is associated with an approximately 20–25% reduction in major cardiovascular events. Accordingly, the pooled LDL-C reduction of approximately 7–15 mg/dL observed with pecan consumption may correspond to an estimated 4–9% reduction in cardiovascular risk when sustained over time [39]. Although modest in absolute terms, such reductions are clinically relevant at the population level, particularly when achieved through dietary modification rather than pharmacotherapy. Similarly, the observed triglyceride reductions (14–16 mg/dL), especially in longer-duration interventions, may contribute to additional cardiometabolic benefits, given the established role of triglyceride-rich lipoproteins in atherosclerotic cardiovascular disease. However, based on the currently available randomized trial evidence, pecans should not be considered a primary intervention for improving fasting glycemic control or promoting weight loss. Their cardiometabolic benefits appear to be largely lipid-specific and should be viewed as complementary to, rather than a substitute for, other evidence-based dietary and lifestyle interventions. Strengths and limitations A major strength of this meta-analysis is the exclusive focus on randomized controlled trials investigating pecan intake, rather than combining mixed nut interventions. This specificity enhances the interpretability of the findings regarding pecans specifically. Additionally, consistent methodology across trials for lipid outcomes and generally low heterogeneity increase confidence in the robustness of cholesterol and triglyceride effects. Several limitations of this meta-analysis should be considered when interpreting the findings, particularly with respect to generalizability. Most included randomized controlled trials enrolled metabolically healthy adults or individuals with only mild dyslipidemia, limiting the ability to extrapolate results to populations with established type 2 diabetes, metabolic syndrome, or uncontrolled hypertension. In addition, intervention durations were relatively short, typically ranging from 4 to 12 weeks, which constrains inference regarding long-term cardiometabolic effects and clinical endpoints such as incident cardiovascular disease. Therefore, while the observed lipid improvements are robust within the studied populations, caution is warranted when applying these findings to higher-risk clinical populations or to long-term disease prevention. In addition, the borderline Egger’s test result for fasting blood glucose suggests potential small-study effects; therefore, findings related to glycemic outcomes should be interpreted cautiously. Another important limitation relates to the free-living nature of most included interventions. Although participants were instructed to consume pecans according to study protocols, adherence may have varied across individuals and studies, potentially influencing the magnitude of observed effects. In addition, the cardiometabolic effects attributed to pecan consumption may partly reflect dietary substitution patterns, as pecans may have replaced foods with differing nutritional profiles (e.g., refined carbohydrate snacks or foods rich in saturated fat). Because detailed information regarding food substitutions and adherence was not consistently reported across studies, residual confounding cannot be completely excluded. Conclusions This meta-analysis demonstrates that pecan consumption is associated with significant improvements in lipid profiles, notably reductions in total cholesterol, LDL-cholesterol, and triglycerides, without significant effects on HDL-cholesterol, glycemic indices, adiposity markers, or blood pressure. The lipid benefits are biologically plausible based on MUFA content, phytosterols, and fiber, and are consistent with broader evidence on tree nuts and cardiometabolic health. While pecans should be considered a heart-healthy food choice, their role in glycemic control remains uncertain, while the currently available evidence regarding blood pressure outcomes is highly heterogeneous and insufficient to determine whether pecan consumption meaningfully affects blood pressure. Future research should prioritize longer-term randomized controlled trials in populations at higher cardiometabolic risk, including individuals with type 2 diabetes, metabolic syndrome, or hypertension, to determine whether the lipid benefits observed with pecan consumption extend to improvements in glycemic control, blood pressure, and long-term clinical outcomes. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 : Figures S1-S8. Supplementary Material 1 : Figures S1-S8. Supplementary Material 2 : Tables S1-S3. Supplementary Material 2 : Tables S1-S3.