A reappraisal of the prevalence of pediatric hypertension through a nationwide database in Taiwan

Hypertension in childhood and adolescence is associated with adult cardiovascular morbidity and mortality. However, the reported prevalence of pediatric hypertension varies considerably. We conducted a pioneer nationwide population-based study to investigate the prevalence of hypertension among children and adolescents. Pediatric patients who had been diagnosed with hypertension between 2000 and 2013 were selected from the National Health Insurance Research Database in Taiwan. Other metabolic syndrome-related diseases that would increase cardiovascular risk, including diabetes mellitus (DM), hyperlipidemia, and obesity, were also retrieved for further evaluation. In total, 10,364 children and adolescents diagnosed with hypertension were identified. The prevalence of pediatric hypertension in Taiwan ranged from 0.19 to 0.38 per 1000 children and adolescents between 2000 and 2013. Essential hypertension was most commonly coded (90.6%), which was much more than secondary hypertension (14.3%). Children and adolescents with hypertension were often associated with DM, hyperlipidemia, and obesity, with the odds ratios as 14.05 (95% confidence interval (CI) 11.74–16.81, p < 0.001), 10.65 (95% CI 9.48–11.97, p < 0.001), and 19.08 (95% CI 15.65–23.26, p < 0.001), respectively. To improve lifelong cardiovascular health, our results emphasize the importance of early proper recognition and suitable management of hypertension, as well as metabolic syndrome-related diseases, among children and adolescents.

www.nature.com/scientificreports/ An analysis of the distribution in subcodes of each hypertension, based on the classification of ICD-9-CM, is demonstrated in Table 4. To protect personal information, the NHIRD was only available for on-site analysis after 2016 21 ; additionally, if the case number was fewer than three, it may increase the risk of personal information being identified and the data were not allowed to be taken out. Thus, some subtypes with fewer than three cases are not presented.
Among the subcodes of essential hypertension, unspecified essential hypertension was most commonly coded. The proportion of overall essential hypertension in male children and adolescents was significantly higher than that in females (p < 0.001); the same conditions were also seen in benign essential hypertension and unspecified essential hypertension (p < 0.001 each).
The major subcode of hypertensive heart disease was unspecified hypertensive heart disease. The proportion of overall hypertensive heart disease in male children and adolescents was significantly higher than that in females (p < 0.05); the same condition was also seen in benign hypertensive heart disease (p < 0.001).
In contrast to essential hypertension and hypertensive heart disease, the proportion of overall hypertensive CKD in female children and adolescents was significantly higher than that in males (p < 0.001); the same conditions were also seen in benign hypertensive renal disease and unspecified hypertensive renal disease (p < 0.05 each). Among the subcodes of hypertensive CKD, the most common coded diagnosis was unspecified hypertensive renal disease.
The most common subcode among hypertensive heart and CKD was unspecified hypertensive heart and renal disease. Proportions of each subcode of hypertensive heart and CKD between males and females did not significantly differ.
Proportions of each subcode of secondary hypertension did not significantly differ between males and females. Among the subcodes of secondary hypertension, unspecified secondary hypertension was most common.

Discussion
This is the first nationwide population-based study to explore the prevalence of hypertension and the associated risk of metabolic syndrome-related diseases among children and adolescents. Additionally, instead of conducting a health survey, which relies on a physiological measurement of blood pressure, hypertensive children and adolescents in our study were identified from the NHIRD; we believe such a research approach is more representative of the current situation of the entire pediatric group in Taiwan. There are several important points that can be drawn from the study. First, the prevalence of pediatric hypertension in Taiwan ranged from 0.19 to 0.38 per 1000 children and adolescents and it was increasing gradually during the entire study period. Second, in comparison to the non-hypertension group, children and adolescents with hypertension were found to be www.nature.com/scientificreports/ at higher risks for metabolic syndrome-related diseases, including DM, hyperlipidemia, and obesity. Third, according to ICD-9-CM, the most common code of hypertension among children and adolescents was essential hypertension. Forth, the proportion of malignant hypertension among each code of hypertension was very small; further studies are undertaken to investigate the long-term outcome and the association between pediatric and adult malignant hypertension. Although the prevalence of pediatric hypertension was not high in Taiwan, the present study did reveal the medical claims by the physicians; additionally, it may also reflect the current state of healthcare providers' awareness of these issues. Among the study population, there were 3472 (33.5%) pediatric patients with both hypertension and hyperlipidemia. In the past, lipid measurements were not frequently conducted in pediatrics 23 . However, dyslipidemia in children and adolescents was found to be linked with several comorbidities, including being a predictor of thickening of the carotid intima-media 23,24 , which is one of the risk factors for future cardiovascular events 25 . Therefore, lipid profiles should be carefully monitored in children and adolescents, especially in pediatric patients with other cardiovascular risks, such as hypertension. Once dyslipidemia is found, management of these children includes lifestyle modifications, such as weight loss, diet control, exercise, etc.; pharmacology for lowering lipid levels can be used in some conditions 26,27 .
With the continuous increase in prevalence, obesity was proposed as being a global health challenge 28 . Recent studies showed that the prevalence of hypertension is higher in obese children and adolescents [29][30][31] . Nevertheless, Table 2. Paired comparison of metabolic syndrome-related diseases in children and adolescents with hypertension and non-hypertension. CI confidence interval, DM diabetes mellitus, OR odds ratio, SD standard deviation. *p < 0.001.  www.nature.com/scientificreports/ there are also studies which found that a correlation between secular changes in hypertension and obesity in children was not very strong 17,32 . Such results may reflect conclusions by Cheung et al., although obesity is recognized as a strong predictor of early hypertension, the association still varies among different ethnicities 33 . In our study, the occurrence of obesity in hypertensive children and adolescents was 19.2%, which was significantly higher than that of the non-hypertension group. However, because the information regarding body-mass index (BMI) was unavailable in the database, the patients with obesity were identified based on the claims data in NHIRD, instead of BMI. Therefore, we could not further investigate the association between pediatric hypertension and BMI. Further study is undertaken to evaluate the relationship between hypertension and obesity among children and adolescents in Taiwan. Both type 1 and type 2 DM are known to be associated with hypertension in children and adolescents; additionally, the rate of developing hypertension was higher in patients with type 2 DM than those with type 1 DM 34,35 . In our study, we also found that children and adolescents with hypertension were associated with higher risks of DM than those without hypertension, regardless of the type of DM. Nevertheless, inadequate awareness of hypertension and unsatisfactory outcomes of anti-hypertensive treatment among children and adolescents with either type 1 or type 2 DM were reported 34,36 . Further analysis revealed type 2 DM was predominant in our study group; the result was compatible with the findings by Zeitler et al., they noted the prevalence of youth-onset type 2 DM was much higher in non-Caucasian descendants 37 . Therefore, our findings emphasized again that children and adolescents with DM need to closely monitor their BP, similarly, children and adolescents with hypertension should also closely monitor their blood glucose. We believe the recommendation were also applicable to pediatric patients with hyperlipidemia and obesity to improve their cardiovascular health.

Number (%)
In the past, secondary hypertension was generally thought to be more common in children and adolescents than in adults 38 ; however, in some more-recent studies on pediatric hypertension, patients with primary hypertension were more than secondary hypertension 39,40 . Similarly, the present study also found that primary hypertension among children and adolescents in Taiwan was more common than secondary hypertension. The analysis by age groups also revealed that the age group 12-18 years old accounted for the largest proportion and followed by the age group 6-12 years old. The results revealed that, among the children and adolescents with hypertension in Taiwan, more than 87% of these patients were diagnosed after the age of 6 years. Because relatively older age (more than 6 years old) is one of the general characteristics of children with primary hypertension, extensive assessment of other secondary causes could be reserved for younger children (0-6 years old) and those with other risk factors 19 . Once children and adolescents were diagnosed with hypertension, nonpharmacologic intervention such as lifestyle modification are still preferred initially; however, in patients with advanced conditions or other secondary causes, pharmacologic therapy and specific intervention for underlying diseases should be considered 19 . Interestingly, although the association between pediatric hypertension and CKD is well-established and the reported prevalence of hypertension among children and adolescents with CKD could be as high as 54% [41][42][43] ; our data revealed that the prevalence of hypertension among children and adolescents with CKD was only 9.6%. There are two possible reasons for this discrepancy: (1) the main cause of hypertension www.nature.com/scientificreports/ in our study group was essential hypertension; thus the ratio of other causes was decreased; (2) the result may imply that hypertension was not strongly correlated with renal diseases in Taiwanese children and adolescents, probably because pediatric patients with renal diseases can receive timely and suitable medical therapy via the easily accessible medical care system in Taiwan. Further studies are necessary to clarify this point. Compared to reports from the published literatures, the rate of pediatric hypertension in Taiwan was relatively low [12][13][14][15][16][17][18][19]44 . Possible reasons for this might be multifactorial. First, according to the guidelines, both pediatric and adult patients, BP should be measured repeatedly on separate occasions to confirm the diagnosis of hypertension 19,20,[45][46][47] . However, most of the reported prevalence were estimated based on the health survey in selected groups and the BP was measured with different intervals at the same visit. The present study was conducted through a nationwide database, in which the diagnoses were required to be confirmed by the attending physicians; additionally, we excluded children and adolescents who had an outpatient clinic visit under the diagnosis of hypertension less than twice to avoid including patients with tentative diagnoses. We believe this approach could avoid the limitation of a single health survey and would be more compatible with the definitions of current guidelines 19,20 . Second, if children and adolescents have a known primary disease, it would have been coded as the main diagnosis; however, if a patient was simultaneously found to have hypertension, the coding of hypertension might be missed. Frequent under-diagnosis of hypertension in pediatric patients was previously demonstrated 44,48 . Although the medical care system in Taiwan is efficient and easily accessible, because of a lack of early symptoms and signs in hypertension, children and adolescents with hypertension might not seek medical help. In addition, young children usually find it difficult to cooperate during BP measurements. Nevertheless, early and regular BP evaluation during each healthcare encounter in pediatric patients are strongly encouraged; this is also recommended by recent guidelines 19 . Third, there could be ethnicity/genetic factors in play. Cheung et al. performed a school-based BP screening in over 20,000 adolescents and demonstrated that although obesity is recognized as a strong predictor of the early development of hypertension, the strength of the association varies from race to race 33 . Fourth, there are other environmental factors. A national survey in China found that the prevalence of hypertension in children and adolescents from different regions significantly differed 16 . The authors postulated that the socioeconomic status, especially dietary habits, may have played an important role.
There are several significant changes in newly published guidelines for high BP in children and adolescents, including a new percentile reference, replacement of the term "prehypertension" with a new term "elevated BP", a simplified screening table for recognizing abnormal BP, updated definitions of BP categories and stages, etc. 19 . Although hypertension and metabolic syndrome-related diseases are not major causes of death in children and adolescents 49,50 , as we mentioned earlier, adult hypertension is a leading cause of mortality 1 . Abundant evidence has demonstrated that adult hypertension originates in childhood 4,5 . As a result, to promote lifelong cardiovascular health, efforts to improve the diagnosis and management of pediatric hypertension are crucial.
The present study had several limitations. First, the research was based on the claims data recorded in the NHIRD, and the diagnoses relied only on ICD-9-CM codes. Therefore, some information was not available for evaluation, including the clinical severity of hypertension and metabolic syndrome-related diseases. Second, the diagnosis and coding of hypertension and metabolic syndrome-related diseases may have been inadequate because BP is rarely measured in children if there are no significant symptoms or signs. In addition, if pediatric patients seek medical care due to other known diseases, such as renal disease, inadequate coding of hypertension may be possible because it may be considered not significant. Nevertheless, we suppose the clinicians would be competent to make appropriate diagnoses based on the clinical practice guideline after adequate training and mandatory continued medical education. As a result, we believe the present study still offers a useful summary of hypertension and metabolic syndrome-related diseases among children and adolescents in Taiwan using our national health database. Additionally, the results may also remind healthcare providers to pay more attention to these issues. Third, we did not use the information of medications to improve the certainty of hypertension diagnosis because drug therapy in children and adolescents with hypertension were not the primary option 19,45 . Forth, owing to the limitation of the ICD-9-CM codes, it was possible if patients were noted to have cardiac problems, patients with primary or secondary hypertension would both be coded as hypertensive heart disease. This situation may also occur in patients coded as hypertensive CKD, and hypertensive heart and CKD. As a result, further research to clarify the natural history of these patients is warranted. Fifth, other factors, such as the BMI, genetic, occupation, stress events, and environmental factors, were not available for further investigation in the NHIRD. However, because more than 99% of the Taiwanese population participates in the NHI program, the data should be representative without institutional or cohort bias.
In conclusion, although the prevalence of hypertension among children and adolescents was low, its number was increasing. The patients who had an outpatient clinic visit under the diagnosis of hypertension less than twice would be excluded; we believed the approach will be more compatible with current guidelines. Most of the identified children and adolescents with hypertension were diagnosed with primary hypertension, rather than secondary hypertension. To the best of our knowledge, this is the first study to evaluate the prevalence and associated risks of metabolic syndrome-related disease among children and adolescents with hypertension through a nationwide database. We believe these findings can offer healthcare providers more-comprehensive insights into these issues and help establish better screening and therapeutic policies to further improve lifelong cardiovascular health. Statistical analyses. All statistical analyses were performed using the SAS Statistical Software, vers. 9.4

Methods
(SAS Institute, Cary, NC, USA) and R Statistical Software, vers. 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria; http://www.r-proje ct.org). Continuous variables are presented as the mean ± standard deviation (SD) and were evaluated by a two-sided independent Student's t-test. Categorical data are presented as sample proportions and were evaluated by a Chi-squared test. Conditional logistic regression analysis was conducted to calculate the association of metabolic syndrome-related diseases between the hypertension and non-hypertension group; the results were presented as odds ratios (ORs) with 95% confidence intervals (CIs). Statistical significance was set at p < 0.05.

Data availability
The data analyzed in this study were from Taiwan's NHIRD. www.nature.com/scientificreports/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.