Evaluation of the Blood Pressure Lowering-Effect of the Lactotripeptides Valine-Proline-Proline and Isoleucine-Proline-Proline in Non-Hypertensive Japanese Subjects through a Meta-Analysis of Randomized-Controlled Studies

Several studies have reported the ability of the lactotripeptides valine-prolineproline (VPP) and isoleucine-proline-proline (IPP) to lower systolic blood pressure (SBP), including in Japanese populations. But the magnitude of the reported changes differs across trials. Conclusions from a previous meta-analysis in Japanese subjects suggest that this may be due at least partly to differences in subjects’ blood pressure (BP) status. Therefore, we decided to resume this analysis, focusing only on non-hypertensive subjects and including newly-published eligible studies, in order to further evaluate the SBPlowering effect of VPP/IPP and study the influence of the ingested dose, type of ingredient (enzymatic or fermented) and food product (drink or supplement). The systematic search of four databases (including two in Japanese) allowed to identify 11 relevant randomized-controlled trials (581 subjects), which were included in the meta-analysis. Results reported a significant decrease in SBP following VPP/IPP intake in non-hypertensive Japanese individuals, with an estimated effect-size of −3.44 mm Hg (95% CI, −4.53 to −2.34, P < 0.0001) as compared to placebo. There was no indication of heterogeneity or publication bias. Furthermore, the type of food product and ingredient did not influence the SBP-lowering effect, which was significant and How to cite this paper: Chanson-Rolle, A., Aubin, F., Braesco, V., Takeda, R. and Saito, Y. (2018) Evaluation of the Blood Pressure Lowering-Effect of the Lactotripeptides Valine-Proline-Proline and Isoleucine-Proline-Proline in Non-Hypertensive Japanese Subjects through a Meta-Analysis of Randomized-Controlled Studies. Food and Nutrition Sciences, 9, 1221-1253. https://doi.org/10.4236/fns.2018.910088 Received: August 9, 2018 Accepted: October 26, 2018 Published: October 30, 2018 Copyright © 2018 by authors and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/


Introduction
High blood pressure (BP) represents an important concern for population health, and may exert a significant burden on healthcare costs in many countries, including in Japan and other Asian countries [1].Recommendations for the preliminary management of hypertension include modification of daily-life habits such as regular physical exercise and reduction of the consumption of salt or alcohol [2].Simultaneously, clinical data from the literature suggest that some dairy-based peptides, such as the lactotripeptides valine-proline-proline (VPP) and isoleucine-proline-proline (IPP), can lower systolic blood pressure (SBP) [3] [4] [5] [6] [7].These two peptides have initially been characterized as angiotensin-converting enzyme (ACE) inhibitors [8], and have been reported to display BP-lowering effects upon ingestion in animals, including in spontaneously hypertensive rats [9] [10] [11] [12].Accordingly, it has been proposed that the SBP-lowering effect of the two peptides may be related to their ACE-inhibiting ability [13], although they may also act through the release of vasodilative elements [14] [15] including bradykinin [16], through the stimulation of Mas-receptors [16], or through an impact on the sympathetic nervous system [17] [18].
The ability of VPP/IPP to lower SBP following consumption has been reported in several previously-published meta-analyses, with effect-sizes of −2.9 mmHg (95% confidence interval [CI], −4.2 to −1.7) [19], −3.7 mmHg (95% CI, −6.7 to −1.8) [3] or −4.8 mmHg (95% CI:, −6.0 to −3.7) [5], in comparison to placebo.These meta-analyses have also indicated that the magnitude of the change would vary across trials, and some of this heterogeneity may be explained by the individuals' BP status [20] [21].But few meta-analyses have specifically evaluated the impact of VPP/IPP on BP in subjects without overt hypertension.Furthermore, part of the heterogeneity may be due to ethnicity, with several meta-analyses reporting a larger efficacy in Asian individuals (from −5.5 1223 Food and Nutrition Sciences mmHg to −6.9 mmHg) as compared with Europeans (from −1.2 mmHg to −1.4 mmHg) [3] [19], and no effect reported in some trials performed in Dutch subjects (e.g.[22]).Although not clearly elucidated, this ethnic difference in the efficacy of VPP/IPP may be due to variations in the pharmacokinetic response of VPP/IPP in Asian and Europeans individuals, or to environmental disparities such as different dietary habits [3] [19].There have been numerous randomized-controlled studies in Asian individuals, especially in Japanese.However, several of them have only been available in Japanese journals which are not included in the major bibliographic databases, and most systematic reviews have not included searches for such journals [3] [4] [5].In a meta-analysis released in 2015, which included all studies published until October 2014, including those from Japanese journals, we confirmed the ability of VPP/IPP to decrease SBP in Japanese individuals, as well as the significant impact of the individuals' BP status at baseline, with a stronger efficacy observed in hypertensive (HT) individuals (−8.4 mm Hg) as compared with non-hypertensive (non-HT) individuals (−3.4 mm Hg) [21].Finally, VPP/IPP ingredients can be obtained by enzymatic hydrolysis or fermentation.Even though similar final products are obtained in either case, published findings indicate that the way of production may have an impact on the ingredient's ability to lower BP [19] [22].Moreover, VPP/IPP can be included in different kinds of food products (e.g.drinks or dietary supplements), which may also influence the size of their SBP-reducing effect.
The aim of the current meta-analysis was to perform an update of our previous meta-analysis published in 2015, and to more specifically evaluate the magnitude of the SBP variation occurring following consumption of VPP/IPP in healthy non-HT Japanese subjects.For that purpose, we performed an update of the literature search in order to include any newly published eligible studies, and we focused on studies in non-HT Japanese subjects.Additional objectives of the current meta-analysis were to examine the impact of the consumed amount of VPP/IPP, as well as the impact of the form of VPP/IPP (enzymatic or fermented ingredient) and type of food product (drink or dietary supplement), on the SPB-lowering effect.

Methods
The methodology of the systematic review has been described in a protocol available in the PROSPERO international database (CRD42014014322).This study has been performed following the PRISMA guidelines [23].The PROSPERO protocol, as well as the PRISMA worksheet, are provided as supplementary materials (see Protocol S1 and Checklist S1, respectively).http://jdream3.com).The set of key-words was as follows: [lactotripeptide OR lactotripeptides OR "dairy peptide" OR "dairy peptides" OR ("Ile-Pro-Pro" AND "Val-Pro-Pro") OR ("Isoleucyl-prolyl-proline" AND "valyl-prolyl-proline") OR ("Valine-proline-proline" AND "isoleucine-proline-proline") OR ("IPP" and "VPP") OR "fermented milk" OR "milk fermented" OR "sour milk"] AND (hypertension OR "blood pressure").The J-STAGE and J Dream III correspond to databases of Japanese journals.The words "Asian OR Japan OR Japanese"

Literature Searches
were included in the search queries for the two other databases.Articles written in all languages (without limitation) were searched for, and the lists of references of selected articles were also screened for any possibly relevant papers (hand search).

Screening Process
The screening of all retrieved papers was performed by three independent researchers (all with a doctoral degree), first on the basis of the reading of titles/abstracts, and secondly on the basis of full-texts for potentially relevant articles.Disagreements were addressed through discussion.We used professional English translations for articles published in Japanese.Inclusion criteria were as follows: randomized-controlled studies, with a single or double-blind design, in

Data Collection
Data were extracted by two independent scientists through the use of a pre-defined worksheet, and any disagreements were solved by discussion.Additional data were requested through direct contacts of authors for nine articles, and appropriate answers were received in all cases.The information collected for all studies was as follows:  [24].Finally, the Jadad-score was used to evaluate study quality [25].

Statistical Analysis
The primary outcome for the meta-analysis was the mean difference between groups consuming VPP/IPP and groups consuming placebo in the variation in office SBP between baseline and last endpoint.The secondary outcome was office DBP reported in the same way.The mean pooled effect-size of VPP/IPP and its 95% CI were calculated by using fixed and random effects, with the REstricted Maximum Likelihood (REML) estimator [26] [27].Since we assumed some heterogeneity from results of previous meta-analyses [3] [21], the random-effect model was defined as the main analysis.Trials were weighted according to the inverse of their variance.Heterogeneity across trials was estimated through tau 2 , I 2 , H 2 , and Cochran's Q test statistics [28].Publication bias was examined by means of a funnel plot (SE of effect plotted against estimate of effect-size for each trial) and through the Kendall's rank correlation test statistic (Kendall's tau) between the standardised effect-size and the SE values of the effect, according to Begg and Mazumdar [29].
We used adjusted meta-analyses, meta-regressions and sub-group analyses to examine the heterogeneity of the effect of VPP/IPP on SBP.More precisely, we studied the impact of the following parameters: type of individuals (NT or PHT, depending on their BP level at baseline), amount of VPP/IPP consumed daily, duration of VPP/IPP consumption, sort of VPP/IPP ingredient (enzymatic or fermented) and type of food product (drink or dietary supplement).Furthermore, the following sub-group analyses were performed: for NT and PHT individuals separately, within series which tested VPP/IPP amounts ≤ 5 mg/d (corresponding to amounts which can be ingested on a daily basis), and for each type of VPP/IPP ingredient and product separately.Finally, we also evaluated the impact of each study on the global findings by removing one study at a time, and 1 VPP equivalents are calculated as the IPP content in mg multiplied by 1.7 and added to the VPP content in mg.The 1.7 correction factor is to take into account the difference in potency of the two tripeptides to inhibit ACE activity in vitro.
(Table 1).Indeed, there were six series involving NT subjects and 11 series involving PHT subjects.Six studies (nine series) used amounts of VPP/IPP ≤ 5 mg/d 2 , which represent amounts that may possibly be ingested on a daily basis.
For the five other trials (eight series), consumed amounts were greater than the "usual" amounts (from 9 to 17 mg/d3 ).This is related to the fact that the main objective of these five trials was to assess the security of VPP/IPP when consumed at higher amounts than the usual ones.The type of ingredient tested was fermented VPP/IPP in four studies (four series) and enzymatic VPP/IPP in the seven remaining studies (13 series).VPP/IPP were consumed as a dietary supplement (tablets) in seven studies (12 series) and as a drink in the four remaining studies (five series).In total, 581 individuals were included within the 17 analyzed series, corresponding to 581 treatment-periods (304 for VPP/IPP and 277 for placebo).All selected trials were randomized-controlled studies that had a parallel and double-blind design, except one (that was single-blind [32]) (Table 1).All needed information was extracted from the original papers or retrieved from authors, and the quality of the studies was considered as adequate (Jadad-score ≥ 4 for all trials but two that displayed a score of 3 [32] [36], see Table 1).This suggests that the risk of bias should have been marginal within studies.For the primary outcome (variation in office SBP between baseline and last 2 For doses expressed both as the sum of the IPP and VPP contents or as "VPP equivalents" (as defined previously).S1 (in supplementary material), which also describes the outcome of the selection process for each article (including justification for exclusion).IPP: isoleucine-proline-proline.SBP: systolicblood pressure.VPP: valine-proline-proline.
endpoint), the mean difference between VPP/IPP and placebo differed across studies from −5.8 (± 3.1 SE) mm Hg in favour of VPP/IPP to +0.6 (±5.4 SE) in favour of placebo (Table 2 and Figure 2).Corresponding findings for DBP (secondary outcome) are described in supplementary material (Table S2 and Figure S1).

Effect of VPP and IPP on Blood Pressure
The findings of the main meta-analysis (SBP variations at last endpoint) reported a statistically significant larger effect of VPP/IPP on SBP as compared with placebo in Japanese non-HT individuals, of a size of −3.44 mm Hg (P < 0.0001, 95% CI, −4.53 to −2.34; random and fixed-effect models) (Figure 2).The estimated effect-size was lower for DBP though still significant [−1.50 mm Hg (95% CI, −2.55 to −0.44, P = 0.006 with the random-effect model)] (Figure S1).
There was no significant heterogeneity between series for SBP (I2 = 0.0%, tau2 = 0.0, Q = 3.7, P = 1.00) and for DBP (I2 = 9.0%, tau2 = 0.4, Q = 15.7,P = 0.47).Nevertheless, heterogeneity was further explored, as described below, since this should provide interesting input regarding the possible parameters that may impact the results.Abbreviations: ACE: angiotensin-converting enzyme.BP: blood pressure.D: double blinded.IPP: isoleucine-proline-proline.NT: normotensive (subjects with normal BP).n: number of subjects analyzed.PHT: pre-hypertensive (subjects with high-normal BP).S: single-blinded.VPP: valine-proline-proline.Y: years.b Type of food product in which VPP/IPP has been added to, namely drink or dietary supplement ("supplement", as tablets).c IPP content in mg + VPP content in mg.d Dose expressed in "VPP equivalents", calculated as the IPP content in mg multiplied by 1.7 added to the VPP content in mg.The 1.7 correction factor is to take into account the difference in potency between the two tripeptides to inhibit ACE activity in vitro [8] [24].

Heterogeneity and Sub-Group Meta-Analyses
Adjusted meta-analyses, meta-regressions and sub-group meta-analyses were used to further explore heterogeneity regarding the effect of VPP/IPP on SBP.The conclusions of these analyses are summarized below.First, the individuals' BP status at baseline was found to exert no significant impact (P = 0.93), and sub-group analyses showed that the size of the effect of VPP/IPP on SBP was similar in NT individuals [−3.54 mm Hg (95% CI, −6.07 to −1.01)] and PHT individuals [−3.41 mm Hg (95% CI, −4.63 to −2.20)], and remained significant in both groups (P = 0.006 and P < 0.0001, respectively) (Figure 3).Second, there was also no significant impact of the daily amount of VPP/IPP, independently of the way it was analyzed [i.e. as a continuous variable (P = 0.65) or as a categorical variable (≤ vs > 5 mg/d, i.e., "usual" vs higher than usual amounts, respectively; P = 0.74].Interestingly, in the sub-group analysis on the nine series in which usual daily amounts of VPP/IPP were consumed, the SBP-lowering effect was still significant (P < 0.0001), with a reduction of −3.34 mm Hg as compared to placebo (95%CI, −4.58 to −2.10) (Figure 4).Third, the form of VPP/IPP ingredient was shown to exert no significant impact (P = 0.65), and sub-group analyses reported that the effect on SBP was significant and of a similar magnitude with either type of ingredient.More precisely, the estimated effect-size was (Figure 5).The same overall conclusions were obtained when the analysis was restricted to series where subjects received usual amounts of VPP/IPP (i.e., ≤5 mg/d).There was still no significant impact of the ingredient type (P = 0.46), with a pooled-effect of −4.00 mm Hg (95% CI, −6.15 to −1.85) for fermented Figure 2. Forest plot of treatment effects of valine-proline-proline/isoleucineproline-proline in the meta-analysis of 17 series of findings of its effect on systolic blood pressure in non-hypertensive Japanese subjects.BP: blood pressure.CI: confidence interval.FE: fixed effect.NT: normotensive (subjects with normal BP).PHT: pre-hypertensive (subjects with high-normal BP).RE: random effect.SBP: systolic blood pressure.Series numbers are those indicated in Table 2.
Figure 3. Forest plot of treatment effects of valine-proline-proline/isoleucineproline-proline on systolic blood pressure in the sub-group analysis according to the baseline blood pressure status of the subject.BP: blood pressure.CI: confidence interval.NT: normotensive (subjects with normal BP).PHT: pre-hypertensive (subjects with high-normal BP).SBP: systolic blood pressure.Series numbers are those indicated in Table 2.   2.
VPP/IPP and of −3.01 mm Hg (95% CI, −4.53 to −1.49) for enzymatic VPP/IPP (Figure S2).The VPP/IPP-induced change in SBP also remained significant with either type of ingredient (P = 0.0003 and P < 0.0001 for fermented and enzymatic VPP/IPP, respectively).Fourth, and similarly, no significant impact of the food product type was observed (P = 0.94), the effect of VPP/IPP on SBP being of the same magnitude when consumed as a drink [−3.47 mm Hg (95% CI, −4.90 to −2.04)] or as a dietary supplement [−3.39 mm Hg (95% CI, −5.09 to −1.69) (Figure 6).Interestingly, the change in SBP induced by VPP/IPP was significant with either type of products (P < 0.0001 in both cases).The same conclusions were obtained if the analysis was restricted to series where subjects received usual amounts of VPP/IPP (i.e., ≤5 mg/d), with no significant influence of the type of food (P = 0.83), and a pooled-effect of −3.43 mm Hg with drink (95% CI, −4.92 to −1.94) and of −3.14 mm Hg with dietary supplement (95% CI, −5.37 to −0.90) (Figure S3).And the SBP reduction induced by VPP/IPP remained significant with either type of product (P < 0.0001 and P = 0.006 for drink and dietary supplement, respectively).Finally, the duration of the VPP/IPP ingestion was not shown to exert any significant impact on the effect of VPP/IPP on SBP (P = 0.96).

Analysis of the Influence of Individual Series and Publication Bias
The impact of omitting one series at a time and the computation of influence diagnostics found that no single series exerted a strong-enough impact on the results that may have generated bias in the observed conclusions.Indeed, VPP/IPP always exerted a significant effect (P < 0.0001) on SBP, independently on which series was omitted.Two series of PHT subjects were shown to exert the strongest influence on the results, namely series #701 and #801, due to their larger sample sizes (106 and 104 patients, respectively, see Table 2).Nevertheless, their results were consistent with those of other studies and with the overall estimate of the effect from the meta-analysis.Therefore, they did not introduce any heterogeneity, and the results remained consistent with the overall analysis when either of these two series was excluded from the analysis.
Finally, the funnel plot did not show any indication of asymmetry (Figure 7), and the Kendall's Tau statistic was not significant (Kendall's Tau = 0.21, P = 0.27).This suggests that there was no publication bias for SBP in the 11 trials included in our meta-analysis.The same was reported for DBP (Figure S4).

Discussion
Conclusions from the meta-analysis of the 11 randomized-controlled studies identified with data in non-HT Japanese subjects indicated that VPP/IPP ingestion produced a significant decrease in SBP in this population, when compared with placebo-control.The effect-size was estimated at −3.44 mm Hg (95% CI, −4.53 to −2.34, P < 0.0001) and there was no indication of heterogeneity or publication bias.A smaller but significant effect was also observed on DBP Figure 6.Forest plot of treatment effects of valine-proline-proline/isoleucineproline-proline (VPP/IPP) on systolic blood pressure in the sub-group analysis according to the type of food product.BP: blood pressure.CI: confidence interval.NT: normotensive (subjects with normal BP).PHT: pre-hypertensive (subjects with high-normal BP).SBP: systolic blood pressure.Series numbers are those indicated in Table 2.
Figure 7. Funnel plot used to explore the potential for publication bias in the meta-analysis of 17 series for the effect of valine-proline-proline/isoleucineproline-proline on systolic blood pressure in non-hypertensive Japanese subjects.Series numbers are those indicated in Table 2.
A −3.41 mm Hg).Results from this updated meta-analysis therefore confirm results from our previous meta-analysis published in 2015 [21], with a larger number of non-HT subjects included.
Furthermore, results from adjusted meta-analyses and sub-group analyses showed that the form of VPP/IPP ingredient (fermented or enzymatic) and the type of food product (drink or dietary supplement) did not influence significantly the SBP-lowering effect of VPP/IPP, which was significant and of the same order of magnitude with either type of ingredient and product.These findings, obtained in non-HT subjects, therefore do not support data from the literature suggesting that the mode of production of VPP/IPP (by fermentation or enzymatic hydrolysis) may influence the efficacy of the lactotripeptides [19] [22].Finally, the SBP-lowering effect of VPP/IPP observed in non-HT subjects remained significant in a sub-analysis that considered only studies in which usual daily amounts (≤5 mg/d) of VPP/IPP were ingested.Results from this updated meta-analysis therefore further support the ability of VPP/IPP to decrease SBP in non-HT Japanese subjects at amounts likely to be consumed on a daily basis, and independently of the types of ingredient and food product consumed.
In Japan, the lactotripeptides VPP/IPP have been permitted as functional components of products for Foods for Specified Health Uses (FOSHU) since 1997, under the health claim "suitable for a person with high-normal blood pressure" and for a labeled daily dose of 3.4 mg of VPP/IPP, expressed in "VPP equivalents"."VPP equivalents" was chosen to express the VPP/IPP dose because it was considered as easier to understand by consumers.Results from this meta-analysis support the VPP/IPP health claim for FOSHU under the above-mentioned wording and daily dose.
Significant relationships between SBP and risk of cardiovascular diseases (CVD) and between SBP and mortality have been reported in epidemiological studies performed in Asian individuals, including Japanese.For CVD, available data suggest that reducing SBP until 115 mm Hg can exert a possible beneficial impact [41] [42].Thus, the reduction in SBP induced by the consumption of VPP/IPP and reported here in non-HT individuals, although modest (−3.44 mm Hg), may be of interest at the scale of the population, especially regarding CVD prevention.In addition, some advocated and effective preventive strategies, such as the reduction of sodium dietary intakes, induce decreases in SBP of a same extent [43] [44].Conclusions of this updated meta-analysis, in which results from all existing trials carried out in Japanese subjects have been combined, therefore ascertain that VPP/IPP ingestion can be beneficial in order to help in maintaining a normal SBP or improving SBP regulation in Japanese subjects without overt hypertension.
Finally, it should be mentioned that few of the studies included in this meta-analysis have evaluated other outcomes beside office BP.For instance, no trial was investigated in one study only [40].Findings from this study showed a significant increase in FMD following VPP/IPP intake when compared with control, therefore suggesting that VPP/IPP may exert a beneficial impact on vascular endothelial function in healthy non-HT subjects [40].
Our meta-analysis displays several strengths.First, two Japanese databases were used for the literature search, which allowed to retrieve eligible trials written in Japanese and which are not included in the MEDLINE and Cochrane databases.This strategy permitted to identify more studies than the previous meta-analyses with results in Japanese subjects (e.g., [3]).This also allowed to evaluate more accurately the magnitude of the impact of VPP/IPP on SBP in Japanese individuals.Second, we focused on results obtained in non-HT subjects, which allowed us to characterize the extent of the variation in SBP induced by VPP/IPP in healthy subjects without overt hypertension.Besides, robustness of our meta-analysis is supported by the fact that we did not find any indication of publication bias and that no individual study was shown to exert a high impact on the global findings.Moreover, all included trials were randomized-controlled studies published in peer-reviewed journals, all but one with a double-blind design.This lowered the risk of bias, which was further limited through the fact that every data had been extracted from the original papers or retrieved from authors.
Nevertheless, this meta-analysis displays a few weaknesses that should be mentioned since they may have influenced the analysis of the data.Despite a reasonable number of included trials and individuals in the overall analysis, the small dataset of some of the sub-group analyses should lead to consider their conclusions with some caution.For instance, in the sub-analyses on the impact of the type of food product and ingredient within series in which usual amounts of VPP/IPP were consumed, only two series were available for fermented VPP/IPP at doses ≤ 5 mg/d.Besides, the number of subjects in individual study groups was low for some included studies.This is related to the fact that we purposely chose to not apply any exclusion criteria on the basis of the number of subjects included in the eligible studies for our meta-analysis.This was done in the objective to exploit the totality of the data available in the literature, which allowed us to evaluate the effect of VPP/IPP on BP more accurately.Finally, although there was no indication of any publication bias, some eligible trials may nevertheless be unpublished.
Therefore, conclusions from this updated meta-analysis confirm that the lactotripeptides VPP and IPP can decrease office SBP in a significant way in healthy Japanese individuals without overt hypertension, with a magnitude which was statistically and clinically significant, and which may lower the risk of CVD at the population level [45].Moreover, the SBP decrease was reported for amounts of VPP/IPP that can potentially be consumed on a regular daily basis, and independently of the types of ingredient (enzymatic or fermented) and food product (drink or dietary supplement) consumed.These findings suggest that the com-

Date of publication of this version 02 August 2018
Details of any existing review of the same topic by the same authors This systematic review is an update of the systematic review which was published on the PROSPERO register under the following title "Influence of the lactotripeptides Isoleucine-Proline-Proline and Valine-Proline-Proline on blood pressure in Asian subjects: a systematic review and meta-analysis of randomized controlled trials" and under registration number CRD42014014322.

Results (paragraph on characteristics of included studies)
and Table 1 Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: a) simple summary data for each intervention group b) effect estimates and confidence intervals, ideally with a forest plot.
Results (paragraph on characteristics of included studies), Table 2, Figure 2 IPP content in mg + VPP content in mg.b Dose expressed in "VPP equivalents", calculated as the IPP content in mg multiplied by 1.7 added to the VPP content in mg.The 1.7 correction factor is to take into account the difference in potency between the two tripeptides to inhibit ACE activity in vitro [8] [24].
Figure S4.Funnel plot used to explore the potential for publication bias in the meta-analysis of 17 series for the effect of valine-proline-proline/isoleucine-proline-proline on diastolic blood pressure in non-hypertensive Japanese subjects.Series numbers are those indicated in Table S2.

Figure 1 .
Figure 1.Flow diagram of study selection.The list of the 18 articles selected for full-text evaluation is available in TableS1(in supplementary material), which also describes the outcome of the selection process for each article (including justification for exclusion).IPP: isoleucine-proline-proline.SBP: systolicblood pressure.VPP: valine-proline-proline.

Figure 4 .
Figure 4. Forest plot of treatment effects of valine-proline-proline/isoleucineproline-proline (VPP/IPP) on systolic blood pressure in the sub-group analysis within the nine series that tested usual daily doses of VPP/IPP (i.e., ≤5 mg/d).BP: blood pressure.CI: confidence interval.NT: normotensive (subjects with normal BP).PHT: pre-hypertensive (subjects with high-normal BP).RE: random effect.SBP: systolic blood pressure.Series numbers are those indicated in Table2.

Table 1 .
Characteristics of the 11 studies included in the meta-analysis of randomized-controlled trials of the effect of valine-proline-proline and isoleucine-proline-proline on systolic blood pressure in non-hypertensive Japanese subjects a .

Table 2 .
Effect of valine-proline-proline and isoleucine-proline-proline on systolic blood pressure at final endpoint in non-hypertensive Japanese subjects a .
[24]breviations: ACE: angiotensin-converting enzyme.BP: blood pressure.IPP: isoleucine-proline-proline.n:number of subjects.NA: not available.NT: normotensive (subjects with normal BP).PHT: pre-hypertensive (subjects with high-normal BP).SBP: systolic blood pressure.SD: standard deviation.SE: standard error.VPP: valine-proline-proline. b IPP content in mg + VPP content in mg.c Dose expressed in "VPP equivalents", calculated as the IPP content in mg multiplied by 1.7 added to the VPP content in mg.The 1.7 correction factor is to take into account the difference in potency betweenthe two tripeptides to inhibit ACE activity in vitro[8][24].

Table S2 .
Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).Effect of valine-proline-proline and isoleucine-proline-proline on diastolic blood pressure at final endpoint in non-hypertensive Japanese subjects.Abbreviations: ACE: angiotensin-converting enzyme.BP: blood pressure.DBP: diastolic blood pressure.IPP: isoleucine-proline-proline.n: number of subjects.NA: not available.NT: normotensive (subjects with normal BP).PHT: pre-hypertensive (subjects with high-normal BP).SD: standard deviation.SE: standard error.VPP: valine-proline-proline.