Effect of Pigmented Rice Consumption on Cardiometabolic Risk Factors: A Systematic Review of Randomized Controlled Trials

Purpose of Review Dietary patterns that include polyphenols may help manage cardiometabolic risk factors. Pigmented rice contains phenolic acids and flavonoids that contribute to its antioxidant properties. This review examined the effect of polyphenol-containing pigmented rice on antioxidant status, lipid profile, glucose/insulin, blood pressure, and weight among adults. Four electronic databases including PubMed, ProQuest, EBSCOhost, and Google Scholar were systematically searched for relevant articles published in English since 2000, using PRISMA guidelines (PROSPERO registration: CRD42022358132). Two-staged screening resulted in the inclusion of seventeen (seven acute, ten chronic) randomized controlled trials. A random effects model was conducted on cardiometabolic outcomes reported in at least three studies. Recent Findings Acute intake increased plasma antioxidant activity and lowered postprandial glucose and insulin levels. Chronic consumption was associated with reductions in fasting glucose (WMD: -1.60 mg/dL; 95% CI:-3.05,-0.14, p = 0.03, k = 5, n = 349), weight (WMD: -0.23 kg, 95% CI: -0.44, -0.02, p = 0.03, k = 3, n = 182), and diastolic blood pressure (WMD: -1.39 mmHg, 95% CI: -2.21, -0.56, p = 0.001, k = 3, n = 185). No effect on total cholesterol, low-density lipoprotein, high-density lipoprotein, triglycerides, body mass index, and systolic blood pressure was found. Summary The consumption of pigmented rice may improve cardiometabolic risk factors. However, the small number of studies and differences in study design, including participants’ health status, form of rice utilized, and duration of intervention, support the need for more high-quality trials to further investigate these findings. Supplementary Information The online version contains supplementary material available at 10.1007/s13668-023-00496-7.

Animal models have identified several mechanisms through which pigmented rice may reduce cardiometabolic risk including the regulation of enzymes involved in metabolic pathways that influence lipid and glucose metabolism.Lower total cholesterol (TC), LDL, and triglycerides (TG) were attributed to a reduction in the expression of fatty acid synthase (FAS) [25,26], an enzyme that catalyzes de novo synthesis of fatty acids, and an increase in carnitoyl transferase (CPT), which is essential for fatty acid oxidation [27].Also demonstrated was increased levels of adiponectin [26], which is associated with improved insulin sensitivity, and reduced TG via increasing lipoprotein lipase activity [28].Lastly, reduced adipose tissue and improved lipid levels were associated with down-regulation of peroxisome proliferator activated receptor-γ (PPAR-γ) [29].Several human trials have recently been conducted to investigate the effect of pigmented rice consumption on lipids, glucose and body weight.However, the overall effect of pigmented rice on cardiometabolic risk factors has yet to be summarized.Thus, this review aimed to systematically review the body of evidence on the effect of pigmented rice on antioxidant status, cholesterol, glucose/insulin, BP, and weight in adults.

Materials and Methods
This review was conducted according to PRISMA guidelines, and the protocol is registered in the PROSPERO database (CRD42022358132).

Search Strategy
A core search strategy was first developed in PubMed by identifying MeSH terms for each concept (adult, pigmented rice, control, cholesterol, glucose, BP, phenolics, weight, and waist circumference (WC)), and using Boolean operators (See supplementary information).This was then used to systematically search for relevant studies in the following databases: PubMed, ProQuest, EBSCOhost, and Google Scholar (search completed October 25, 2022).In addition, online clinical trial registries (ClinicalTrials.gov,RIAT, Phil.Health Research Registry, and Australian New Zealand Clinical Trials Registry (ANZCTR)), reference lists of identified studies (pearling), and other sources of grey literature were searched to ensure all relevant studies were included.No limit on language was applied, but date of publication was limited from 2000 to the present as research outputs on the relationship between rice consumption and health increased from 2000 thereafter [20].

Eligibility Criteria/Study Selection
Table 1 presents the summarized eligibility criteria used in the present study.Randomized-controlled trials with (P) adult participants (> 18 years old) using (I) pigmented rice (red, purple, and/or black) in any form (including cooked, extract, powdered) as an intervention against a (C) control (placebo, white rice, brown rice, maltodextrin, or usual diet), and reporting at least one of the following outcomes (O): antioxidant status (total phenol index (TPI), total antioxidant/radical activity/capacity (TAC)), cholesterol (TC, LDL, high-density lipoprotein (HDL), TG), glucose/insulin (fasting blood glucose, glycated hemoglobin, insulin, insulin sensitivity), BP and anthropometry (weight, body mass index (BMI), WC/waisthip ratio (W:H)) were included.Studies were excluded if they combined pigmented rice with other foods or intervention or did not compare to a control, such that the specific effects of pigmented rice could not be determined.
Studies were only eligible if published in full text in a peer-reviewed journal, in English (or with English translation).Animal and in vitro studies were excluded.
All studies were independently screened by title and abstract, then by full text by two reviewers (DMS; AMH, EVM, and RAR), and any disagreement was settled by a third reviewer (AMH, EVM, or RAR) using COVIDENCE software.

Data Extraction
Two reviewers independently extracted data (DMS; EVM and FPP) using a pilot-tested template, checked for discrepancies, and resolved by consensus.Extracted data included the following: author's name, year of publication, country where the study was conducted, study design, study duration, blinding, study population characteristics (sex, age, co-morbidities,

Risk of Bias Assessment
The quality of included studies was assessed independently by two reviewers (DMS; EVM and RAR using the Cochrane Collaboration Risk of Bias Version 2 (RoB 2) tool, and a consensus resolved disagreements.Studies were assessed as low risk ( +), high risk (-) or some concerns (!), based on the following domains: (D1) bias arising from randomization process, (D2) bias due to deviations from intended interventions, (D3) bias due to missing outcome data, (D4) bias in measurement of outcome, and (D5) bias in selection of the reported result [30].Crossover studies included assessment on bias arising from period carryover effects (DS).

Data Analysis
Chronic and acute study data were assessed separately for each outcome.Antioxidant activity in acute studies were analyzed by identifying the time points for initial increase of antioxidant activity, peak response, and duration of response.Pre-and post-intervention mean and standard deviation (SD) for TC, LDL, HDL, TG, glucose, BP, weight, and BMI were used to compute weighted mean differences (WMD) with their 95% confidence interval, and overall effect size for chronic studies.Standard deviations were calculated for studies that reported only standard error of the mean (SEM) using the formula: SD = SEM x square root (n).Studies without reported change and/or SD were calculated using the following formulas assuming a modest correlation coefficient (R) of 0.5 [31].
A random effects model was conducted on cardiometabolic outcomes reported in at least three studies using Review Manager version 5.4 [32], with a P-value of < 0.05 considered statistically significant.Heterogeneity was assessed using chi-squared test and I 2 statistics.An I 2 result of > 50% was considered substantial heterogeneity [33].Forest plots were produced to present the summarized information on each study, heterogeneity, and overall effect size for each outcome.Due to the low number of studies included in the analyses, funnel plots were not generated.
Quality of evidence for each outcome was rated high, moderate, low, or very low based on the following criteria: (1) risk of bias, (2) inconsistency of results, (3) indirectness of evidence, (4) imprecision, and (5) publication bias [34], using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) criteria [35].

Study Selection
Figure 1 presents the flow of study selection based on PRISMA guidelines [36].A total of 6,184 studies from the four databases and 40 from additional searches were identified.Duplicates were removed using a reference manager (Mendeley).A total of 4,537 studies were imported to COV-IDENCE for screening.A total of 42 studies were identified for full-text review, of which 26 were excluded based on the following reasons: wrong intervention (n = 20), non-RCT (n = 3), wrong outcomes (n = 1), no English translation (n = 1), and with co-intervention (n = 1).Seventeen eligible studies were included in the systematic review.

Assessment of Risk of Bias
Figure 2 presents the results of the risk of bias assessment for parallel studies.Four studies were assessed as having some concerns attributed to the randomization process [48, [50][51][52].Despite only two studies providing a complete discussion on how randomization was carried out [45, 47•] and three providing information on blinding [45, 47•, 53], all studies provided details on baseline characteristics of participants which assisted with assessing bias arising from the randomization process (D1).All studies, except for Syarief et al. [51] provided details on the sham intervention provided for the control which aids in assessing bias arising from deviations from intended intervention (D2).All study outcomes are considered observer reported outcomes not involving judgement which are not likely to be affected by knowledge of intervention received by the participants [31].
Figure 3 presents the results of the risk of bias assessment for crossover studies.Two of eight studies were assessed as having some concerns as they did not provide detailed descriptions of how randomization was done.Most studies (n = 6) reported at least a one-week wash out period in between treatment groups which was considered sufficient for minimizing carry-over effects.

Effect on Antioxidant Status
Nine studies (seven acute and two chronic) investigated the effect of pigmented rice consumption on antioxidant status   [38-41, 42••, 43, 44, 48, 53].Of the acute studies, three used Riceberry, a rice cultivar with deep-purple pigment [38,39,41], three provided black rice [44,48,53], and one study investigated both purple and red rice [40].All acute studies reported an increase in antioxidant activity after 30 min of ingestion compared to control (brown rice, wheat bread, sucrose, or yogurt).However, peak antioxidant activity varied among studies, ranging from 30 to 360 min after ingestion (see Table 4).Duration of antioxidant activity lasted from 180 min [38,41,44] after ingestion, to as long as 240 [40] or 360 [39] minutes.Four acute studies demonstrated significant increases in several measures of antioxidant activity compared to control at various time points [38,39,41,44], and one study reported a significant increase in FRAP compared to baseline [40].
Wang et al. [53] measured antioxidant activity after six months of intervention with black rice fraction among CHD patients.They demonstrated significantly higher FRAP activity compared to control (1.29 ± 2.96 103 u/L vs. -0.61± 1.69 103 u/L; p < 0.01) but no differences in superoxide dismutase (SOD).Similarly, Kim et al. [48] also showed no effect on SOD, but observed significantly higher glutathione peroxidase among women with obesity who consumed black rice meal replacements for six weeks compared to white rice meal replacements (15.36 ± 5.63 U/g Hb vs. 3.52 ± 5.41 U/g Hb, p < 0.05).

Effect on Glucose
Eleven studies reported glucose as an outcome (five acute and six chronic intake).All data for postprandial glucose and insulin were reported in line graphs, which limited data extraction (only one of the contacted authors provided data).Only the chronic studies provided sufficient data for meta-analysis.

GRADE
The certainty assessment for each outcome is presented in Table 5.All outcomes were assessed to have no serious risk of bias.However, TC, LDL, TG, Glucose, and SBP have serious concerns for inconsistency and indirectness attributed to differences in the study population, and delivery of the intervention.Lastly, imprecision was assessed as serious for TC, LDL, BMI, weight, and DBP, and very serious for HDL, TG, and SBP.Anuyahong et al. 2020 [38] Anuyahong et al. 2022 [39] Calcott et al. 2019 [40] Chusak et al 2020 [41] Muangchan et al. 2021 [42] Se  • Significantly lower compared to control at 30, 60, 90 and 120 min

Discussion
This study systematically reviewed the effect of pigmented rice consumption on antioxidant status and cardiometabolic risk factors in adults.Meta-analysis of chronic intake studies demonstrated significant reductions in glucose, weight, and diastolic BP, but no significant effects on TC, LDL, TG, HDL, BMI, or systolic BP.All acute studies included in the review demonstrated that pigmented rice consumption increases antioxidant activity 30 min after ingestion and is sustained for at least 180 min.Moreover, chronic consumption of pigmented rice (12-24 weeks) was shown to increase total antioxidant activity compared to control [48,53].This finding may explain the role of pigmented rice in improving blood pressure by reducing oxidative stress which promotes endothelial dysfunction [55].Reactive oxygen species decrease nitric oxide production and promote cellular damage, resulting to inflammation, vasoconstriction, vascular lesion and ultimately atherosclerosis and CVDs [56].
The antioxidant activity of pigmented rice is attributed to its polyphenol content such as phenolic acids that can stabilize electrons, and flavonoids (including anthocyanins) that have the ability to donate electrons and stop chain reactions [20].Various studies have shown pigmented rice promotes radical scavenging activity (DPPH) [44], has reducing power (FRAP) [38,39,41,53], ability to inhibit radical induced oxidation (ORAC [38] and ABTS [44]), ability to neutralize radical cation (TEAC) [38][39][40], and reduces oxidative stress (MDA) [38][39][40][41].Despite having low bioavailability, once absorbed in the gut anthocyanins are metabolized in the liver, secreted and reabsorbed in the enterohepatic circulation resulting in molecular intermediates that contribute to their biologic actions [57].The meta-analysis showed significant beneficial effects on glucose, weight, and diastolic BP, but not cholesterol (TC, LDL, HDL, TG), BMI, or systolic BP.Black rice anthocyanins may help reduce glucose levels by delaying carbohydrate absorption through inhibition of α-amylase and α-glucosidase [58].This can be seen in the acute intake studies where the majority demonstrated significantly lower postprandial glucose and insulin levels at various timepoints compared to the control.This may imply that pigmented rice consumption may help maintain lower blood glucose levels post meal.However, it should be noted that despite significant reductions in fasting blood glucose levels in chronic studies, the magnitude of effect was small and unlikely to be considered clinically meaningful.Participants in this analysis had normal blood glucose levels at baseline; it is possible that greater benefit may be seen in persons with elevated glucose levels.Additionally, polyphenols may assist with managing obesity through increased energy expenditure, appetite suppression, and regulation of lipid metabolism [59].Several mechanisms of action have been postulated for how pigmented rice consumption may reduce lipid levels.These include regulation of fatty acid synthesis [25,26], transcription factors [29], and lipid metabolism [26].However, while most studies included in this review reported reductions in lipids, these were not statistically different to the control, and collectively this meta-analysis did not show any benefit on lipids.Manach et al. [60] argues that doses in animal studies are often higher than what human tissue may be exposed to.Additionally, participants in many of the studies included in this review had normal TC, LDL, TG, and glucose levels, as opposed to most animal studies which induce hyperlipidemia or hyperglycemia.
Moreover, considerable differences in study characteristics were noted.While all studies in the meta-analysis used black rice, they varied in form (cooked, extract, fermented, glutinous) and dosage (15 mg to 320 mg anthocyanin).Phenolic acids, mostly ferulic acid, and anthocyanins which are mostly cyanidin 3-O-glucodside and peonidin 3-O-glucoside, are highly concentrated in the bran [20].Further, different rice forms (extract, with giant embryo, glutinous) and effect of processing (cooked, fermented, incorporated to a snack bar) may contribute to differences in concentration and bioavailability of polyphenols.Black sticky rice was reported to have higher antioxidant activity compared with red, and black rice [19 •].Germination of rice also increase antioxidant activity, total phenolics and flavonoids [61], while fermentation of grains is hypothesized to increase bioavailability of phenolic compounds [62].The study of Fauziyah et al. [50] which utilized fermented glutinous black rice reported the greatest reduction in LDL among all studies included in this review, which was significantly lower compared to the control.Most of the chronic studies matched their control with the intervention, either using white/polished rice in the same form as the intervention (i.e., whole, powdered, meal replacement mix) or using placebo/ maltodextrin capsules identical to the extracts administered to the intervention group.However, some studies did not report the specific control used [50,51], as noted in the assessment of risk of bias, or did not attempt to match the rice form given to the intervention group [52].Lastly, included studies differed in duration of intervention, and presence or absence of co-intervention.Apart from the rice, one study also provided diet counselling to both groups [50]; one study induced energy restriction in both groups [48].Seesen et al. [52], provided health education to the control in lieu of the intervention.Due to the limited number of studies, subgroup analyses could not be conducted to determine whether these differences impacted on the cardiometabolic outcomes.Altogether, these differences influence the certainty of evidence of the outcomes.Therefore, further randomized controlled human clinical trials are warranted to support the clinical value of pigmented rice consumption for reducing cardiometabolic risk factors.
The present meta-analysis has some limitations that should be noted.First, some of the included studies have some concerns based on Cochrane Collaboration's Risk of Bias, specifically in relation to how randomization was carried out.Secondly, no meta-analysis was conducted on antioxidant status, or postprandial glucose and insulin due to availability of data from included studies.Outcomes with meta-analyses have heterogeneity (TC, LDL, TG, BMI, and systolic BP) which is likely attributed to differences in study design.

Conclusion
This systematic review showed that acute intake of pigmented rice increases antioxidant activity and lowers postprandial glucose and insulin levels.Meta-analysis demonstrated significant reductions in glucose, weight, and diastolic BP following chronic pigmented rice consumption, but no significant effects on TC, LDL, TG, HDL, BMI, or systolic BP.More high quality randomized controlled trials are warranted to further investigate the effect of pigmented rice consumption on cardiometabolic risk factors in adults; additional benefit may be observed in those with established clinical conditions such as dyslipidemia, pre-diabetes/diabetes, and hypertension.

Fig. 1
Fig. 1 Flow diagram of study selection

Fig. 2
Fig. 2 Result of the assessment for Risk of Bias of Parallel Studies.Green represents low risk of bias; yellow represents some concerns for risk of bias

•Fig. 4
Fig. 4 Effect of pigmented rice on lipid profile

Table 1
PICOS criteria used to define the eligibility criteria

Table 5
GRADE Assessment for Certainty of Evidence