This research was conducted as an SR with meta-analysis, focusing on cross-sectional studies carried out in Latin America. The methodology strictly adhered to the guidelines established by the PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement,12 with necessary adaptations for a systematic review centered on prevalence studies. This approach ensures a rigorous and transparent synthesis of available evidence on the prevalence of prehypertension and high-normal blood pressure in the Latin American region.

A comprehensive search was conducted across four academic databases: Scopus, Web of Science, Embase, and PubMed/Medline until September 24, 2024. Key search terms included: “prevalence,” “prehypertension,” “pre-hypertension,” “high normal blood pressure,” “elevated blood pressure,” and “Latin America” or specific Latin American country names. The detailed search strategy for each database is provided in Supplementary Material 1.

Each strategy was meticulously adapted for each database, employing controlled vocabulary (such as MeSH terms in PubMed) when available, in combination with free-text searching. Publications in English, Spanish, and Portuguese were included, with no restrictions on publication year to ensure complete coverage of relevant literature.

To complement the electronic search, manual searches were conducted in the reference lists of included articles and relevant previous systematic reviews. This approach allowed for the identification of additional studies that might have been overlooked in the initial electronic search.

Studies were considered eligible for inclusion if they met the following specific criteria: (1) cross-sectional observational studies published as full articles, focusing on evaluating the prevalence of prehypertension and/or high-normal blood pressure; (2) studies had to include adult participants (≥18 years) of both sexes, residing in Latin American countries; (3) studies had to assess prevalence using systolic blood pressure (SBP) and diastolic blood pressure (DBP) measurements. The definition of prehypertension had to follow JNC 7 criteria (SBP of 120–139mmHg or DBP of 80–89mmHg),2 while HNBP had to conform to European guidelines (SBP of 130–139mmHg and/or DBP of 85–89mmHg); (4) probabilistic sampling methods were required, including simple random, stratified, cluster, or multistage sampling; (5) studies had to provide sufficient data to calculate the prevalence of prehypertension and/or high-normal blood pressure, or directly report these measures with their respective confidence intervals; and (6) studies published in English, Spanish, or Portuguese were considered.

Conversely, the following were excluded: (1) narrative reviews, editorials, letters to the editor, commentaries, and case studies; (2) studies using non-probabilistic sampling methods; (3) research focused exclusively on populations with specific medical conditions, such as hospitalized patients or individuals with pre-existing cardiovascular diseases; (4) studies not providing original data, such as predictive models based on secondary data; and (5) preclinical or animal studies.

Rayyan software (https://rayyan.qcri.org) was used to manage and store articles identified in the databases. This tool facilitated the organization and tracking of studies throughout the selection process.

Study selection was conducted in two phases. First, two independent reviewers examined the titles and abstracts of identified manuscripts. This phase was conducted blindly to minimize bias. Studies that both reviewers considered to meet inclusion criteria were selected for detailed review. In case of disagreement, a third reviewer intervened as arbitrator to resolve the discrepancy. Subsequently, full-text review was conducted, where a comprehensive evaluation of the complete text of all preselected articles was performed. Two reviewers carried out this process independently, following the same protocol as the initial phase. An Excel spreadsheet was used to record inclusion or exclusion decisions, along with specific reasons for exclusion when applicable. Discrepancies at this stage were resolved through discussion between reviewers, and in cases of lack of consensus, a fourth reviewer was consulted for the final decision.

To ensure comprehensive selection, an additional manual search was conducted in the reference lists of included articles and previous systematic reviews on related topics. Studies identified through this method underwent the same selection process described above.

In cases of multiple publications based on the same study population, the most recent article or the one providing more complete information on the prevalence of prehypertension and/or high-normal blood pressure was included.

The entire selection process was meticulously documented, including the number of studies identified, screened, assessed for eligibility, and included in the review, along with reasons for exclusion at each stage.

Data extraction was performed using a Microsoft Excel 2023 template specifically designed for this systematic review. Two independent reviewers conducted this process to ensure accuracy and comprehensiveness of extracted data. The template was piloted with a sample of included studies to ensure its suitability before complete extraction.

Extracted data encompassed detailed information about study characteristics and their results, including: author(s) and publication year, Latin American country(ies) involved, study design and data collection period, sample size and demographic characteristics (age, sex), probabilistic sampling method used, diagnostic criteria employed for prehypertension and/or high-normal blood pressure, blood pressure measurement method (device, number of measurements), reported prevalence of prehypertension and/or HNBP.

Discrepancies in data extraction or qualitative interpretation between the two reviewers were resolved through discussion and consensus. In cases of persistent disagreement, a third reviewer was consulted for the final decision.

Risk of bias assessment was conducted independently by two investigators for all studies included in the systematic review. For this purpose, Munn et al.’s risk of bias assessment tool,13 specifically designed for prevalence studies, was used. This tool was selected due to its suitability for evaluating the methodological quality of prevalence studies and its wide acceptance in the field of systematic reviews.

The assessment was based on ten specific criteria covering key aspects of prevalence study methodology. These criteria include sample representativeness of the target population, appropriateness of the sampling frame, random selection process of participants, minimization of non-response bias, data collection directly from subjects, case definition acceptability, study instrument reliability and validity, consistency in data collection mode, appropriateness of shortest prevalence period length, and appropriateness of the denominator.

For each criterion, reviewers assigned a response of “Low risk,” “High risk,” or “Unclear.” A total score was calculated for each study, awarding one point for each “Low risk” response. The level of risk of bias was categorized as follows: studies with 0–3 points were considered high risk, 4–6 points medium risk, and 7–10 points low risk of bias.

Additionally, publication bias was assessed through visual inspection of funnel plots and application of Egger's test, when appropriate. These assessments were conducted to determine if there was evidence of bias in the literature included in the systematic review.

All quantitative analyses were performed using R statistical software version 4.2.2. For the meta-analysis, all studies providing data on the prevalence of prehypertension and/or HNBP, assessed through systolic and diastolic blood pressure measurements, were included. Both total sample size (n) and number of cases (r) were extracted for each study.

The meta-analysis was conducted using the ‘metaprop’ function from the ‘meta’ package. The Freeman–Tukey double arcsine transformation method (sm=“PFT”) was employed for proportion transformation. Confidence intervals were calculated using the Clopper–Pearson method (method.ci=“CP”), which provides exact confidence intervals for proportions.

Given the expectation of significant variability due to differences in study populations, blood pressure measurement methods, and other contextual factors across Latin American countries, a random-effects model with the DerSimonian and Laird method (method.tau=“DL”) was used to estimate between-study heterogeneity. The Hartung–Knapp correction (hakn=TRUE) was applied to adjust standard errors. Additionally, heterogeneity between studies was assessed using the I2 statistic and Cochran's Q test, which are automatically calculated in the ‘metaprop’ function. Thus, meta-analysis results were presented with their respective 95% confidence intervals and visualized through forest plots using the ‘forest’ function from the ‘meta’ package.

Additionally, meta-regressions were conducted to examine the influence of continuous variables on prevalence estimates, specifically publication year and sample size. The weighted least squares method was used, with weights inversely proportional to each study's variance. For each meta-regression, a mixed-effects model was fitted using the “rma” function from the “metafor” package in R. To visualize the results, bubble plots were generated where the size of each bubble is proportional to the study's weight in the analysis.

This systematic review and meta-analysis was based exclusively on the analysis of aggregated data from previously published studies on the prevalence of prehypertension and HNBP in the region. This methodological approach did not involve the collection of individual participant data or the conduct of new interventions or experiments. Consequently, specific ethical approval was not required for the execution of this synthesis study.

A total of 1108 publications were found. After removing duplicates (760), 348 manuscripts were analyzed based on title and abstract. After excluding 322 studies, 26 full articles were obtained. Finally, after applying the selection criteria, 17 articles were selected.8,10,14–28 The process can be visualized in Fig. 1.

Figure 1.

Forest plot of the prevalence of arterial prehypertension in Latin America.

Table 1 presents a summary of the characteristics of studies included in this SR. Of these, 15 studies focused on evaluating prehypertension, one focused exclusively on HNBP, and one study addressed both conditions. This distribution underscores the predominant interest of the scientific community in prehypertension within the Latin American context.

The temporal spectrum of analyzed studies spans more than a decade and a half of research, providing a valuable longitudinal view of the field. The oldest study, conducted by Cynthia Pérez14 in 2008 in Puerto Rico, marks the beginning of this research period. At the most recent end, we find several studies conducted in 2023, including those by Hidalgo-Benites,26 and Karl Peltzer,27 along with a study by Vera-Ponce dated 2024.28

Regarding sample sizes, considerable variability is observed, reflecting different research approaches and resources. The most extensive study, conducted by Vera-Ponce in 2024 in Peru, included a sample of 275,759 participants.28 At the other end of the spectrum, Lizarazu-Dieazgranados's 2013 study in Colombia worked with a sample of 124 participants.17

Gender representation in the studies shows variations. While most studies included both men and women, Carmo Rocha's 2013 study18 in Brazil stands out for focusing exclusively on women, with a 100% female sample. In contrast, Moraes-Bezerra's 2017 study23 in Brazil presented the lowest proportion of women, at 46.4%.

In terms of geographical distribution, Peru is the most represented country in this review, with six studies,8,22,25,26,28 followed closely by Brazil with five.16,18,19,23,24 In contrast, countries such as Cuba, Puerto Rico, Argentina, Venezuela, and Ecuador are represented by only one study each.

The selection criteria of the studies show some notable similarities. Most focused on adult populations, generally defined as individuals over 18 years, although Salazar MR's study included participants from age 15.10 At the other extreme, Hidalgo-Benites's 2023 study in Peru focused on an older population, with an average age of 60.4 years.26 Additionally, a common exclusion criterion among selected studies was the presence of previously diagnosed hypertension or use of antihypertensive medications, allowing a more precise focus on prehypertension and HNBP. Several studies also excluded individuals with pre-existing cardiovascular diseases or conditions that could affect blood pressure measurements.

Finally, regarding blood pressure measurement methods, general consistency is observed in the use of automatic digital devices, with a notable preference for OMRON brand equipment. Most studies performed multiple measurements, typically two or three, and used their average for analyses, increasing measurement reliability. However, some variations in specific protocols were noted. Some studies, such as those by Cynthia Perez14 and Salazar MR,10 opted for mercury or aneroid sphygmomanometers, possibly reflecting more traditional clinical practices or resource limitations. There were differences in rest times before measurements (varying from 5 to 15min) and in intervals between measurements (from 1 to 10min).

Regarding bias analysis, evaluated in Table 2, of the 17 studies evaluated, all obtained scores of 8, except for Lizarazu-Dieazgranados's study (2013),17 due to the small sample size presented. Thus, all were classified in the low risk of bias category.

According to the random-effects model, the combined prevalence of prehypertension in Latin America is estimated at 27.98% (95% CI: 21.17–35.34%). However, heterogeneity was very high (I2=100%, p<0.01).

When examining results by country, different patterns are observed. Brazil, for example, shows a wide range of estimates, oscillating between 20.59% and 54.99%, with a combined estimate of 30.72% (95% CI: 15.16–48.96%). Peru presents a similar panorama, with prevalences ranging from 18.13% to 33.44%, and a combined estimate of 28.39% (95% CI: 18.64–39.29%).

Some countries stand out for their extreme values. Puerto Rico, represented by Cynthia Perez's study (2008),14 shows the highest individual prevalence at 46.10% (95% CI: 42.73–49.50%). At the other extreme, Cuba, with Miguel-Soca's study (2016),21 presents the lowest prevalence at 3.55% (95% CI: 2.81–4.44%). Meanwhile, the other countries included in the analysis, Venezuela, Colombia, and Ecuador show prevalences of 29.69%, 45.97%, and 22.81% respectively.

According to the random-effects model, the combined prevalence of HNBP in these countries is estimated at 19.23% (95% CI: 11.43–28.49%). Individually, Salazar MR's study (2016)10 in Argentina shows the highest HNBP prevalence at 23.47% (95% CI: 21.02–26.06%). Meanwhile, in Peru, two studies are included. Hidalgo-Benites's study (2023)26 shows a prevalence of 16.70% (95% CI: 14.41–19.19%), while Vera-Ponce's study (2024)28 indicates a slightly higher prevalence of 17.96% (95% CI: 17.81–18.10%). However, heterogeneity between studies is considerable (I2=91%, p<0.01).

The distribution of points in the plot exhibits noticeable asymmetry, deviating from the ideal inverted funnel shape that would be expected in the absence of bias. This asymmetry, along with the considerable dispersion of points, especially at the lower part of the plot, suggests the presence of significant heterogeneity among the studies included. The existence of some points clearly separated from the main cluster, particularly on the right side, indicates the presence of studies with unusual or extreme results that warrant special attention. Furthermore, the clustering of several studies at the top of the plot suggests that some high-precision studies, likely with large sample sizes, yield similar effect estimates (Figs. 2 and 3).

Figure 2.

Forest plot of the prevalence of HNBP in Latin America.

Figure 3.

Funnel plot of prehypertension studies in Latin America.

Fig. 4 illustrates the relationship between the year of publication and the prevalence of prehypertension across various studies conducted in Latin American countries. This bubble plot shows a slightly decreasing trend in prehypertension prevalence over time, as indicated by the red dashed regression line. However, the wide gray band surrounding this line suggests considerable uncertainty in this trend. The studies are represented by circles whose size reflects the sample size, and the colors indicate different countries. A large variability in the reported prevalence is observed, ranging from less than 10% to over 50%.

Figure 4.

Meta-regression of the prevalence of prehypertension by year of publication in Latin American countries.

Fig. 5 illustrates the relationship between sample size (on a logarithmic scale) and the prevalence of prehypertension across various studies conducted in Latin American countries. The red dashed line represents the overall trend, showing a slight decrease in prevalence as sample size increases. However, the wide shaded gray area around this line indicates considerable uncertainty in this trend. The studies are represented by points of different colors based on the country, allowing for easy comparison between nations.

Figure 5.

Meta-regression of the prevalence of prehypertension by sample size in Latin American countries.

In this meta-analysis of Latin American studies, we found a combined prevalence of prehypertension of 27.98% (95% CI: 21.17–35.34%), indicating that more than a quarter of the adult population in the region might be at elevated risk of developing hypertension. The prevalence of HNBP was estimated at 19.23% (95% CI: 11.43–28.49%), although this estimate is based on a more limited number of studies. We observed considerable heterogeneity among countries, with prehypertension prevalences ranging from 3.55% in Cuba to 46.10% in Puerto Rico. Meta-regression analyses suggested a slight decreasing trend in prehypertension prevalence over time and with increasing sample size, although these trends were accompanied by substantial uncertainty. These findings underscore the significant magnitude of prehypertension and HNBP in Latin America while revealing important variations between countries that warrant further investigation.

In interpreting our results, it is crucial to contextualize our findings with existing literature on prehypertension and HNBP at global and regional levels. Our combined prevalence estimates of 28.76% for prehypertension in the region is comparable with findings from other international studies. For instance, a global meta-analysis conducted by Guo et al. found a worldwide prehypertension prevalence of 36% (95% CI: 31.0–41.2%).29 Our results suggest that prevalence in Latin America might be slightly lower than these global averages, although the heterogeneity observed between countries indicates that some nations in the region may be close to or even exceed these figures.

The significant variability we found among Latin American countries is consistent with findings from previous studies that have examined regional differences in cardiovascular risk factors. For example, Chow et al.’s PURE study highlighted important differences in hypertension prevalence among low- and middle-income countries, with ranges from 32.5% to 40.8% across different regions.30 Our findings of prehypertension prevalences ranging from 3.55% in Cuba to 46.10% in Puerto Rico reflect even greater variability, underscoring the importance of considering country-specific factors when interpreting these data.

Regarding HNBP, our estimate of 19.23% aligns with findings from European studies, although it is important to note that literature on HNBP is less abundant than that on prehypertension. Additionally, it is noteworthy that more studies with probabilistic sampling have been conducted on prehypertension than on HNBP, likely because the definition of the latter is recent.3 Mancia et al. reported an HNBP prevalence of 13.5% in an Italian cohort,31 while the PAMELA study found a prevalence of 11.6% in another Italian population.32 Our results suggest that HNBP prevalence in Latin America might be slightly higher than these European estimates, although the limited number of studies included in our HNBP analysis indicates the need for more research in this specific area in the Latin American context.

The slightly decreasing trend in prehypertension prevalence over time observed in our meta-regression analysis is an interesting finding that deserves further exploration. This trend might reflect improvements in public health strategies and increased awareness about the importance of blood pressure control in the region. However, it contrasts with findings from some longitudinal studies such as Vasan et al., who observed an increase in hypertension incidence over time in individuals with prehypertension.33 This apparent discrepancy might be due to differences in study designs, examined populations, or time periods considered.

The variations observed in prehypertension and HNBP prevalence among Latin American countries can be attributed to a series of complex and interrelated factors. First, differences in lifestyles and dietary habits play a crucial role. A study by Carrillo-Larco and Bernabe-Ortiz34 reported that salt consumption in Latin America was more than double that recommended by the World Health Organization. Similarly, Bernabé-Ortiz et al.8 reported in Peru that excessive alcohol consumption and physical inactivity, especially in urban areas, were important risk factors for prehypertension or HNBP.

Socioeconomic disparities may also explain part of the observed variability. Chambergo-Michilot et al. demonstrated in Mexico that hypertension prevalence was lower in higher socioeconomic groups.35 This pattern might be extrapolated to other countries in the region, where economic inequalities are pronounced. Furthermore, access to health services and their quality vary significantly between and within Latin American countries, as pointed out by Atun et al. in their analysis of the region's health systems.36

Genetic and ethnic factors may also contribute to the observed differences. A study by Quaresma et al. in Brazil found variations in prehypertension prevalence among different ethnic groups, with higher rates in Afro-descendant populations.37 These genetic differences may interact with environmental factors, resulting in variations in susceptibility to prehypertension and HNBP.

Urbanization and associated changes in physical activity patterns may also explain part of the variability. Miranda et al. observed in Peru that migration from rural to urban areas was associated with an increase in cardiovascular risk factors, which may include prehypertension.38 This phenomenon might be occurring to different degrees across countries in the region.

Finally, methodological differences between studies, such as sample selection criteria, blood pressure measurement methods, and operational definitions of prehypertension and HNBP, may contribute to the observed variability. For instance, Hernández-Vásquez et al. noted that the choice of measurement device and protocol followed can significantly influence prevalence estimates.39

First, the combined high prevalence of prehypertension (28.76%) suggests that a substantial proportion of the Latin American population is at elevated risk of developing hypertension and its associated complications. This aligns with projections by Olsen et al., who anticipated a significant increase in the cardiovascular disease burden in the region over the coming decades.40 Therefore, it is imperative that public health systems in Latin America implement primary prevention strategies focused on individuals with prehypertension and HNBP.

Early identification of these pre-hypertensive states is crucial. As noted by Egan and Stevens-Fabry, prehypertension not only increases the risk of progression to hypertension but is also associated with a higher risk of cardiovascular events even before the development of frank hypertension.41 This underscores the need for broader and more effective screening programs in primary care, such as those proposed by Rubinstein et al. in their study on cost-effective interventions for non-communicable diseases in Latin America.42

The observed variability among countries in the prevalence of prehypertension and HNBP indicates that public health strategies must be adapted to specific national and local contexts. Thus, interventions should consider the cultural, socioeconomic, and health system particularities of each country. This could range from culturally adapted public education campaigns to nation-specific food regulation policies.

Moreover, it is essential to address the modifiable risk factors associated with prehypertension and HNBP. Lifestyle interventions, such as promoting healthy diet and increasing physical activity, have proven effective in preventing progression to hypertension. The ENCORE study by Blumenthal et al. demonstrated that these interventions can be particularly effective in individuals with prehypertension.43 Therefore, public policies that facilitate healthy lifestyle choices should be considered, such as improving public spaces for physical activity or regulating the advertising of unhealthy foods.

Finally, it is crucial to enhance public awareness about the significance of prehypertension and HNBP. Studies, such as that by Chow et al., found that hypertension awareness in Latin America was suboptimal,30 and awareness of pre-hypertensive states is likely even lower. Thus, public health campaigns should educate the population about the importance of regular blood pressure monitoring and adopting healthy lifestyles from an early age.