Sugar-sweetened beverages increases the risk of hypertension among children and adolescence: a systematic review and dose–response meta-analysis

In the current systematic review and meta-analysis, we summarized the studies that evaluated the effects of sugar-sweetened beverages (SSBs) intake on blood pressure among children and adolescents. In a systematic search from PubMed, Scopus, Embase and Cochrane electronic databases up to 20 April 2020, the observational studies that evaluated the association between sugar-sweetened beverages intake and hypertension, systolic or diastolic blood pressure (SBP, DBP) were retrieved. A total of 14 studies with 93873 participants were included in the current meta-analysis. High SSB consumption was associated with 1.67 mmHg increase in SBP in children and adolescents (WMD: 1.67; CI 1.021–2.321; P < 0.001). The difference in DBP was not significant (WMD: 0.313; CI −0.131– 0.757; P = 0.108). High SSB consumers were 1.36 times more likely to develop hypertension compared with low SSB consumers (OR: 1.365; CI 1.145–1.626; P = 0.001). In dose–response meta-analysis, no departure from linearity was observed between SSB intake and change in SBP (P-nonlinearity = 0.707) or DBP (P-nonlinearity = 0.180). According to our finding, high SSB consumption increases SBP and hypertension in children and adolescents.


Background
The increasing prevalence of obesity and weight gain in pediatric population, as a major health problem, is associated with insulin resistance, hypertension, atherogenic dyslipidemia, and pro-inflammatory state [1]. Hypertension, as one of the major component of metabolic syndrome is also associated with obesity state and has an increasing prevalence in youth [2]. Hypertension in children and adolescents is defined as average systolic blood pressure (SBP) and/or diastolic blood pressure (DBP) greater than 95th percentile for gender, age, and height on ≥ 3 occasions, while prehypertension is defined as average SBP or DBP levels that are greater than 90 th percentile but less than 95th percentile [3]. Hypertension and elevated blood pressure among children is associated with cardiovascular risk factors and obesity as well. Although major final outcome of CVD such as death and cardiovascular disability do not occur in hypertensive children, they are encountered with increased risk of intermediate markers of target organ damage, such as left ventricular hypertrophy, retinal vascular changes, thickening of the carotid vessel wall, and even subtle cognitive changes [4]. It is widely recognized that blood pressure

Open Access
Journal of Translational Medicine *Correspondence: abbasalizad_m@yahoo.com 1 Drug Applied Research Center, Tabriz University of Medical Sciences, Tabriz, Iran Full list of author information is available at the end of the article Page 2 of 18 Farhangi et al. J Transl Med (2020) 18:344 levels are influenced by genetic as well as by environmental factors [5,6]. In this regard, more than 90 different genetic polymorphisms have been identified to be associated with high blood pressure [7]. For example, a recent study reported that polymorphism of aldosterone synthase gene is linked with the development of hypertension through increasing the aldosterone level and aldosterone/renin ratio [5]. On the other hand, among environmental parameters, obesity, smoking, alcohol consumption, diet, and physical inactivity likely play a major role in development of hypertension [6]. The role of sugars in developing cardio-metabolic disorders and hypertension in children has been actively investigated. However, recently the role of sugar-sweetened beverages (SSBs) in developing hypertension particularly in children and adolescents is highlighted [8][9][10][11][12]. SSBs, as a liquid form of carbonated or noncarbonated energy beverages, are the principle source of added sugar in diets [13]. For instance, a cross-sectional study from China showed that SSBs provide 10-15% of total calorie intake of school students [9]. Another study in Taiwan indicated that adolescents are also one of the major groups who consume a high amount of SSBs [13]. The US Nutrition Examination Survey showed that approximately 64% of the pediatric and adolescents aged 2-19 years have daily SSB consumption contributing to 8.4% of the daily energy intake [14]. In Iran, the average SSBs intake among children and adolescents was 38.5 ± 75.0 g per day with the mean daily SSB intake of 98 ml in boys and 70 ml in girls [15].
In Australia, the average amount of 217 mL of SSB per day is consumed by youth contributing to 5.5% of their total energy intake [16]. In Mexico, SSB intake as one of the main sources of added sugar intake contributes to 8.3% of the total energy intake among children and adolescents [17]. Therefore, SSBs contain excessive amounts of energy, in the form of simple sugar. All of these figures have exceeded the recommended intake of free sugars that has been proposed by the World Health Organization to be less than 5% of total energy intakes [18]. Increased sympathetic nervous system activity [19], significant increase in blood pressure due to potential antinatriuresis effect of fructose affecting salt metabolism [9] and increased serum uric acid due to fructose metabolism [20][21][22] are several suggested mechanisms of the association between SSBs intake and hypertension among children and adolescents. Although numerous studies confirmed the role of high SSBs consumption in developing hypertension in youth [9,13,[23][24][25], there are several inconsistencies reporting no significant association between SSB intake and blood pressure [13,26,27]. Moreover, childhood and adolescence are critical periods for the acquisition of healthy behaviors; therefore, the study of several indices and their co-occurrence in this ages should be a priority. In the current systematic review and meta-analysis, we aimed to summarize the studies that evaluated the association between SSBs intake and blood pressure among children and adolescents in twoclass and dose-response meta-analysis.

Materials and methods
The current study was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [28]. The completed checklist has been provided in the Additional file 1 (Additional file 1: Table S1); moreover, the abstract was written according to the 12-item PRISMA extension checklist [29].

Data sources
A systematic search using PubMed, Scopus, Embase and Cochrane electronic databases was performed to find the studies evaluated the association between sugar-sweetened beverages intake and hypertension up to 1 April 2020. No language and time restrictions were applied. Moreover, hand-searching from reference lists of all relevant papers, previous reviews and meta-analyses was performed to cover all relevant publications. Strategy search was created using a combination of the MeSH (Medical Subject Headings) terms from the PubMed database and free text words.

Search strategy
For the search purpose, we used MeSH (Medical Subject Heading) and non-MeSH keywords including the following: ( Table S2). The reviewed literatures were inserted into the EndNote software (version X8, for Windows, Thomson Reuters, Philadelphia, PA, USA). For each electronic database, search strategy was adopted.

Study selection
In the current systematic review and meta-analysis, observational studies with the design of cross-sectional, case control or cohort evaluating the association between sugar-sweetened beverages (SSB) and hypertension (HTN), systolic blood pressure (SBP) and diastolic blood Since there is no official definition for SSBs, they were defined as any type of above-mentioned drinks. Initially, retrieved citations were merged, duplications were eliminated and the review process was facilitated. Accordingly, the titles and abstracts of all articles were evaluated independently by 2 reviewers (MAF, LN). Full-texts of relevant articles were retrieved if meeting the eligibility criteria, and then were re-evaluated. Any disagreements were discussed and resolved by consensus.

Risk of bias and quality assessment
The quality of cross-sectional studies was assessed by Agency for Healthcare Research and Quality (AHRQ) checklist [30]. There was no quality criteria for inclusion of the studies in the current meta-analysis. The items were scored "1" if the answer was "YES, " and "0" if the answer was "NO" or "UNCLEAR. " The final quality assessments scores were as follows: low quality = 0-3; moderate quality = 4-7; high quality ≥ 8. The details of the studies' quality assessment are presented in Additional file 1: Table S3.

Data synthesis and analysis Two class meta-analysis of the comparison of SBP and DBP between SSB categories
The comparison of SBP and DBP between highest versus lowest category of SSB was performed by measuring the unstandardized mean differences as the effect size calculated by pooled estimate of weighted mean difference (WMD) with 95% confidence interval (CI), and the fixed effects and random effects models according to level of heterogeneity. When the mean values were missed and median and range were provided, we used the method provided by Hozo et al. [31] considering the median values as best estimate of mean for sample size more than 25 and calculating SD as follows: . When SD of the mean difference was not available from the studies, we calculated it using the following formula: SD change = square root [(SD baseline 2 [32], SD = IQR/1.35 (symmetrical data distribution) and SD = SEM × sqrt (n), where n is number of participants, IQR is interquartile range and SEM is standard error of the mean. When the number of individuals in each category of SSB was not provided in the manuscript, we assumed that equal number of participants is enrolled in each group. When the odds of hypertension in SSB consumers versus non-consumers were provided, ORs and 95% CIs were used to estimate the combined effects. Subgroup analysis was also performed to identify possible sources of heterogeneity according to the study setting, SSB dose, and baseline values of SBP or DBP, design, health status, sample size, region, quality score of study, gender and study design. The dose of SSB intake was converted to gram of intake per day according to food agriculture organization (FAO) guidelines for converting units, denominators and expressions [33].
Cochran's Q test and I squared test was used to identify between-study heterogeneity; I 2 ˂ 25%, no heterogeneity; I 2 = 25-50%, moderate heterogeneity; I 2 > 50% large heterogeneity [34]. The heterogeneity was considered significant if either the Q statistic had P value < 0.1 or I 2 > 50%. Sensitivity analysis by exclusion of one study at a time was applied to test the influence of each individual study on overall pooled estimates and heterogeneity [35]. Begg's funnel plots was assessed to evaluate the publication bias followed by the Egger's regression asymmetry test and Begg's adjusted rank correlation for formal statistical assessment of funnel plot asymmetry. The data were analyzed using STATA version 13 (STATA Corp, College Station, TX, USA), and P-values less than 0.05 were considered as statistically significant.

Dose-response meta-analysis of the association between SSB dose and change in SBP or DBP
For dose response meta-analysis, the eligible studies had been reported the mean (SD) of continuous variable (e.g. SBP, DBP) in at least three categories. The median point in each SSB category was also identified. If medians had not been reported in the manuscript, then approximate medians were estimated, using the midpoint of the lower and upper limits. If the highest study category was open-ended, its SSB dose was calculated by assuming that the interval was the same as the closest category. The lowest categories of SSB intake was considered as the reference dose for each study. Any potential non-linear associations of SSB intake were performed by fractional polynominal modelling (polynomials) to explore the nonlinear potential effects of SSB dosage (g/d) and the studyspecific parameter [36].

Flow of studies
Our search strategy identified 1661 potentially relevant articles. Thereafter 857 manuscripts were remained for full text screening after removing duplicates and exclusion according to the title and abstract reading. Totally, 671 manuscripts were excluded because of their irrelevant subject, inappropriate design, being reviews including meta-analysis or systematic reviews, conferences and seminars, not relevant age groups, not evaluating the association of studied parameters. A final number of 14 manuscripts were included in the current meta-analysis ( Fig. 1).

Finding from the dose-response meta-analysis of the association between SSB dose and blood pressure
The details of dose-response meta-analysis are shown in Table 2 and the results for the SBP and DBP are presented in Figs. 5 and 6, respectively. According to the results of dose-response meta-analysis, no evidence of departure from linearity was observed for the association between dose of SSB with mean change in SBP (P-nonlinearity = 0.707) or DBP (P-nonlinearity = 0.180).

Publication bias
The funnel plots are presented in Additional file 1: Figure

Discussion
According to our finding, high SSBs intake among children and adolescents was associated with higher SBP and odds of hypertension. Moreover, no evidence of departure from linearity was observed in the dose-response meta-analysis of change in SBP or DBP according to SSB dosage. A total of 14 studies with 93,873 participants were included in the current meta-analysis. SSBs such as sugar sodas and juices are one of the main sources of excess sugar consumption containing 22 to 39 g of sugar per serving [40,41]. The American Academy of Pediatrics (AAP) has recommended that young children refrain from intake of SSB because of its potential adverse effects on obesity and related disorders [42]. According to the last update of the clinical practice guideline which is issued by AAP, the prevalence of pediatric prehypertension and hypertension has increased to 14.8% and 16.3%, respectively [2]. In our work, high SSB intake was associated with increased systolic blood pressure and odds of hypertension; numerous trials also evaluated the effects of reduced SSB intake on blood pressure; in the study by Chen L et al. reduction in SSB intake of 1 serving/day over 18 months was associated with a 1.8 and 1.1 mmHg reduction in SBP and DBP, respectively [43]. Chiu S et al. also reported reduced systolic blood pressure after replacing sugar sweetened sodas with milk in young male adolescents [44]. Accumulating evidence has linked SSB consumption during childhood to unhealthy weight gain which itself associated with risk of health outcomes such as type 2 diabetes, metabolic syndrome, cardiovascular diseases and other obesity-related disorders in later life [45]. Therefore, intake of SSB should be limited in children and adolescents to reduce obesityrelated chronic disease risk.
By using subgroup analyses, we could successfully identify possible sources of heterogeneity; such that the setting, region, gender and study quality were associated with a significant source of heterogeneity for SBP and SSB dosage, baseline DBP values, study quality, gender and design were possible source of heterogeneity across studies for DBP. Although the effect of high SSB intake on DBP was not significant, while subgrouping, the results were significant for the studies performed in apparently healthy and Asian populations, school setting, with high baseline DBP values and in large sample size studies. So, potential sources of bias were detected with the help of subgroup analyses. It seems that school is one of the best environments for children's psychological, physical and social development [46]. Since children spend so much of their day predominantly in the school setting, the school food environment can contribute in reversing the trend towards childhood obesity [47]. Research has shown that children consume nearly 35-47% of their daily dietary intake and they are exposed to less healthful food and beverages such as SSBs and energy dense food (pizza, french fries, chips and candies) while at school [48]. It seems that improvement to the school food environment through decreasing availability of SSBs and less healthful nutritional practices can be considered as a strategy to reduce obesity and its-related complications in children and adolescents [47]. Numerous school base studies have effective strategies combating against children health problems [49][50][51]. WHO recommends that reduction of SSB intake among children should be implemented initially in schools by developing rules about consuming soft drinks in schools, removing vending machines selling soft drinks from school premises, provision of safe drinking water fountains in schools and other locations where children gather and promoting healthy dietary behavior in classrooms [52]. Moreover, children with higher baseline DBP values showed higher association of SSB intake with DBP; this finding showed that possibly the adverse effect of high SSB intake increased by increase in baseline blood pressure. In our research, the association between mean difference in SBP or DBP with SSB dosage did not exert a non-linear association. Therefore, increase in SBP or DBP is not a dose-dependent event after SSB consumption; this finding was also similar to the previous  [54]. It seems that the role of FFQ as a self-reported data collection tool for estimating the serving sizes might be a source of bias, this is mostly because of the difference in the FFQs structure and items and also difference in the serving definition in numerous studies. Also, in different studies outcome of study was adjusted for wide heterogenic confounders that may have affected the accuracy of dose-response estimates [53]. In the present meta-analysis, we found that SSB consumption is associated with the elevated SBP and DBP among apparently healthy subjects. However, we should take into account that the most of studies had included healthy participants in their researches and only one study performed among diabetic subjects. Therefore, the observed results may not reflect the true relationship regarding the subjects' health status. Since the previous studies have shown that SSB intake is positively associated with diabetes and other health outcome [55], these data support the benefits of lower intake of SSBs.
Region was also another important factor affecting the SSB and DBP association. Our meta-analysis found that in the studies that performed in Asia there was a potent effect of high SSB intake on DBP, while this association was not significant for the studies that performed in USA/Oceania. Interestingly, this finding was also similar for SBP subgrouping. This finding is possibly due to this fact that most of the studies were form Asia and this high number of studies give greater power to Asian studies; also, in the previous report of global, regional, and national consumption of sugar-sweetened beverages in 187 countries, the SSB intake among Asian countries was lower than European and American countries and these findings were strongly dependent to age, country and sex of participants [56]; therefore, the role of these confounders in explaining the association between SSB intake and burden of disease should be considered. On the other hand, cultural differences among the lifestyle and sociodemographic factors play an important role in dietary intakes especially sugar; and it has been proposed as an explanation for the disparities in disease risk among ethnically diverse population [57,58]. It seems that cultural factors by influencing on food preferences and choices may contribute to diet quality and subsequently health inequalities [57]. On the other hand, according to the latest data, childhood obesity prevalence, which coincides with the highest prevalence of hypertension and other metabolic disorders, in Latin American is among the highest in the world [59]. However, only one study from Latin American countries was included in our meta-analysis and as a result, we missed information on the relationship between SSB intake and hypertension among children and adolescents in this geographical region. Several potential mechanisms may describe how SSB consumption could results in increasing the risk of hypertension. Hyperuricemia which is induced by a higher fructose load from sugar-sweetened beverages may leads to acute endothelial dysfunction and chronic Na retention and consequently predisposes individuals to hypertension [60,61]. In this regard, findings from a human study showed a significant increase in blood pressure after acute administration of fructose while this effect was not seen with glucose [62]. Therefore, it has been hypothesized that the fructose in SSBs is responsible for their association with elevated blood pressure. Heredity appears to play a major role in the development of metabolic abnormalities such as hypertension especially in childhood and reports have shown heritability of childhood hypertension is estimated at 50 percent [63]. However, from included studies in our meta-analysis, only one citation [9] had included those who didn't have a history of hypertension. On the other hand, none of included studies have adjusted for family history, thus our finding in the present meta-analysis should be interpreted with caution. Additionally, SSB consumption has been shown to be a part of an overall unhealthy dietary pattern and is correlated with unfavorable socioeconomic status [64]. There is limited research has directly compared the effect of SSB intake to other foods with regard to the risk of cardio-metabolic risk factors such as elevated blood pressure [65]. For example, Amini et al. reported western dietary pattern which contains high amount of SSB is associated with greater odds of having increased blood pressure [65]. Besides, the Dietary Approach to Stop Hypertension (DASH) which emphasizes on higher consumption of vegetables, fruits, nuts, legumes, fish, chicken, whole grains, low-fat dairy products, and lower consumption of SSBs and red meat, has been shown to be negatively associated with hypertension in adults and children [66].
Recently accumulating evidence has linked the maternal diet during pregnancy and breastfeeding to food and tastes preferences of children [67]. The fetus experiences maternal diet tastes and smells through amniotic fluids during pregnancy and afterward by breast milk [68]. Thus, maternal intake in pregnancy could program taste preference of the child towards SSB and health care providers should pay particular attention to educating women in this area.
The association between high SSBs intake and higher odds of hypertension among children and adolescents was another main finding in the present research. A large number of studies have shown that blood pressure in childhood predicts the future hypertension in adulthood [69,70]. Hence, early interventions are warranted.

Strength and limitations
The current systematic review and meta-analysis for the first time evaluated the dose-response association between sugar-sweetened beverage intake and hypertension in children and adolescents. Due to growing prevalence of hypertension in this population, this study has clinical and social implications regarding developing preventive strategies against high SSB consumption in children and adolescents. However, several limitations of the current meta-analysis should also be mentioned; first, using different kinds of FFQ for extraction of SSB intake is a matter of bias because this information is self-reported and has different structures and definitions between studies. Second, there were different kinds of SSBs in these studies and subgrouping according to SSB types were not possible. Moreover, different studies have reported the SSB intake with different units and these conversions might be a cause of error in estimating the accurate dosage of SSB consumption. Additionally, there were different adjustments for confounders in different studies that might affect the results.

Conclusion
The current meta-analysis, for the first time revealed that high SSBs consumption is associated with increased SBP and odds of hypertension among children and adolescents. Although further large prospective studies and well-designed intervention studies are recommended to confirm the observed relationships, the results of the present study support recommendations to decrease the consumption of SSB to prevent and control hypertension and its complications. Developing strategic programs to reduce SSBs consumption particularly in school settings is suggested to reduce the disease burden in this population.