RRT indicates renal replacement therapy.
ESRD indicates end-stage renal disease.
eFigure 1. The Rate of New ESRD Cases per 100,000 Participants by Age During 2 Follow-up Periods: 1967-2013 and 1990-2013
eFigure 2. Survival Diagrams by Examination Periods: 1977-2013 and 1967-1984
eFigure 3. Survival Diagrams Stratified by BMI
eTable 1. Models’ Results of Subanalysis of the Participants Enrolled During 1977-2013
eTable 2. Models’ Results of Subanalysis of the Participants Enrolled During 1967-1984
eTable 3. Stratification of the Population by BMI: Non-overweight and Overweight
eTable 4. Models’ Results of Subanalysis of the Overweight Population
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Leiba A, Fishman B, Twig G, et al. Association of Adolescent Hypertension With Future End-stage Renal Disease. JAMA Intern Med. 2019;179(4):517–523. doi:10.1001/jamainternmed.2018.7632
Is established hypertension among otherwise healthy adolescents associated with increased risk for future end-stage renal disease?
In this cohort study of 2.65 million adolescent candidates for military service in Israel, a very small percentage of participants (0.3%) had a well-established hypertension diagnosis and end-stage renal disease developed in some participants after an extensive period.
Well-established hypertension during adolescence appears to double the risk for future end-stage renal disease, irrespective of overweight condition or severity of hypertension.
Hypertension is a leading risk factor of cardiovascular morbidity and mortality. The role of nonmalignant hypertension as the sole initiating factor of end-stage renal disease (ESRD) in non–African American populations has recently been questioned.
To investigate the association between hypertension and future ESRD in otherwise healthy adolescents.
Design, Setting, and Participants
This retrospective cohort study examined the data of 16- to 19-year-old healthy candidates for military service in the Israel Defense Forces between January 1, 1967, and December 31, 2013. Data were obtained from the central conscription registry of the Israel Defense Forces and the ESRD registry of the Israel Ministry of Health. Participants underwent a comprehensive medical assessment prior to their military service. Individuals with evidence of renal damage or kidney-related risk factors were excluded. The data analysis was conducted from February 12, 2017, to October 16, 2018.
Main Outcomes and Measures
End-stage renal disease as recorded by the Israeli ESRD registry, including hemodialysis, peritoneal dialysis, renal transplant diagnosed between January 1, 1990, and December 31, 2014.
The cohort included 2 658 238 adolescents (1 596 709 [60.1%] male with a mean [SD] age of 17.4 [0.5] years), of whom 7997 (0.3%) had an established hypertension diagnosis. Half of the individuals in the hypertensive group were overweight (1559 [20.1%]) or obese (2243 [28.9%]), and most (7235 [90.5%]) were male. During a median follow-up of 19.6 years (52 287 945 person-years), 2189 individuals developed ESRD, with an incidence rate of 3.9 per 100 000 person-years. Adolescent hypertension was found to be associated with future ESRD (crude hazard ratio [HR], 5.07; 95% CI, 3.73-6.88). In a multivariable model adjusted for sex, age, years of education, body mass index, and other sociodemographic variables, the HR was 1.98 (95% CI, 1.42-2.77). When excluding participants with severe hypertension, the association with ESRD remained statistically significant (HR, 1.93; 95% CI, 1.37-2.70). In the subanalysis of nonoverweight adolescents, the association between hypertension and ESRD was statistically significant as well (HR, 2.11; 95% CI, 1.05-4.24).
Conclusions and Relevance
Hypertension appears to be associated with a doubling of the risk of future ESRD in an otherwise healthy adolescent population.
Hypertension is the most important contributor to the burden of disease and the leading cause of mortality worldwide, estimated to be responsible for almost 13% of all deaths.1-3 Although hypertension is much more common in adults and elderly people, it has increasingly been observed in adolescents and young adults during the past 30 years.4,5 Studies have shown that hypertension, when manifested in this age group, is a risk factor for all-cause mortality, specifically for cardiovascular mortality and, in some other studies, cerebrovascular mortality.6-9
End-stage renal disease (ESRD) is a major cause of mortality as well as a leading risk factor for other fatal diseases, including cardiovascular disease. Worldwide, the number of individuals with ESRD in 2010 was estimated at 5 to 10 million and is expected to rise, owing to the increasing prevalence of hypertension, diabetes, population aging, and renal replacement therapies that prolong the lives of patients with ESRD.10 Hypertension in adulthood has been recognized as 1 of the 2 leading causes of ESRD, responsible for more than 25% of cases, second only to diabetes (responsible for 44%), in the United States.11 However, the role of nonmalignant hypertension as the sole initiating factor of ESRD in non–African American populations is debatable at present.12,13 In this study, we investigated whether well-established essential hypertension, presenting between ages 16 and 19 years without concurrent kidney morbidities, is a substantial initiating risk factor for future ESRD.
We conducted a large population-based retrospective cohort study that included all eligible adolescents, males and females, between 16 and 19 years of age who were called up to the Israel Defense Forces regional recruitment centers for a medical health assessment approximately 1 year prior to their conscription into mandatory military service. The study was approved by the Israel Defense Forces Medical Corps Institutional Review Board, which waived the requirement for informed consent because the data used were obtained from medical records without patient participation. The data analysis was conducted from February 12, 2017, to October 16, 2018.
Arab people, Jewish Orthodox females, and Jewish ultra-Orthodox males do not serve in the Israeli army and therefore were not included in this study. All candidates for military service between January 1, 1967, and December 31, 2013, were examined. The cohort was created by linking the data obtained from the central conscription registry of the Israel Defense Forces with the data from ESRD registry of the Israel Ministry of Health. Excluded were those who were deceased before the initiation of the ESRD registry (January 1, 1990). Also excluded were individuals who, on recruitment, received a diagnosis other than hypertension that is believed to affect renal function such as diabetes, systemic lupus erythematosus, Familial Mediterranean Fever, all types of vasculitis, persistent hematuria (including microhematuria), proteinuria, chronic nephrolithiasis, recurrent urinary tract infection, cystic kidney disease, acute kidney injury and chronic kidney disease, congenital and acquired anomalies of the kidney, and present and resolved glomerulonephritis (Figure 1). Individuals who already received an ESRD diagnosis at the time of the medical examination were also excluded from the study.
As part of their medical assessment at the regional recruitment center, all individuals were required to fill out a medical status form and provide a summary of their medical status furnished by their primary care physician. During the assessment, a detailed medical history was taken and a physical examination was performed by a trained physician. In addition, height and weight were measured, and mean systolic and diastolic blood pressure (BP) (based on the measurement acquired while sitting) were recorded. Urinalysis was performed using a urine dipstick. If any information was missing or abnormal during the assessment, the individuals were referred for further evaluation.
When the systolic and diastolic BP values were higher than 140/90 mm Hg, the individuals were referred to their primary care physician for an additional 10 consecutive BP measurements over a period of at least 3 weeks. If the urinalysis results were abnormal, additional tests such as a 24-hour urine collection were performed and the individual was further examined by a nephrologist. As with other medical diagnoses, the diagnosis of essential hypertension was determined only after the Israel Defense Forces medical board had completed the evaluations needed to render the diagnosis and assign a relevant code. The final diagnosis of hypertension was determined when the mean of the outpatient BP measurements was higher than 140/90 mm Hg and at least 50% of these measurements were above this level. All individuals with hypertension underwent a full evaluation to exclude secondary hypertension. Severe hypertension was defined when the mean of 10 measurements taken in the outpatient clinic was higher than 160/100 mm Hg and at least 50% of the measurements were above this level or when left ventricular hypertrophy or grade 2 retinopathy was present. Individuals with severe hypertension were released from military service. If a diagnosis of essential hypertension was verified, the individual received a code that represented the diagnosis and determined their classification in the army. Individuals with diagnosis codes that represented essential hypertension composed the hypertensive group.
In addition to the medical evaluation at the recruitment center, all candidates for military service completed a psychometric examination and were interviewed with regard to years of education, place of residence, and country of origin. Country of origin refers to the country of the person’s birth or, in cases of Israel born, the father’s or paternal grandfather’s country of origin; the mother's country of origin was unavailable in this cohort.
The primary outcome of this study was ESRD as recorded by the Israeli ESRD registry. The registry is a national administrative database founded in 1990 and maintained by the Israel Center for Disease Control. All nephrology dialysis units in Israel are obligated to report new patients undergoing renal replacement therapy (ie, hemodialysis, peritoneal dialysis, and renal transplant) to the Israel Center for Disease Control. The database contains demographic data, type of renal replacement therapy, and ESRD onset date (the date of renal replacement therapy initiation). All ESRD cases from January 1, 1990, to December 31, 2014, were included in this study.
The SPSS software, version 23 (IBM), was used to conduct an unadjusted survival analysis to determine the cumulative incidence of ESRD. The Cox proportional hazards regression models estimated the crude hazard ratio (HR) and 95% CI for ESRD. Follow-up began on the day of the original examination at the recruitment center and concluded when data were conveyed to the ESRD registry, when death had occurred, or on December 31, 2014, whichever came first.
We included in the multivariable model the variables that are known from previous studies to have an association with the outcome and variables that were associated with the primary outcome in the univariate analyses. The final multivariable model was adjusted for the following variables: sex, age at the medical examination, year of birth, country of origin (ie, father’s or paternal grandfather’s country of birth), and socioeconomic status divided into 3 groups according to place of residence (using a scale provided by the Israel Central Bureau of Statistics). Educational status was divided into 4 groups (9, 10, 11, or 12 or more years of education). Body mass index (calculated as weight in kilograms divided by height in meters squared) for age was divided into 4 percentile groups, as defined by the US Centers for Disease Control and Prevention for teenagers: underweight (<5% percentile), normal weight (5%-85% percentile), overweight (85%-95% percentile), and obese (>95% percentile).
To overcome the lack of an ESRD registry between January 1, 1967, and December 31, 1989, and to estimate the possible association between this deficit and our findings, we estimated the rate of ESRD across different ages of the participants who had enrolled in the study between January 1, 1967, and December 31, 2013, and among those who had enrolled between January 1, 1990, and December 31, 2013. An additional subanalysis was performed for various enrollment periods. To minimize the period without ESRD data, we performed a sensitivity analysis of the participants who had enrolled in the study between January 1, 1977, and December 31, 2013. A competing risk analysis for mortality of any cause, for those who were alive at the establishment of the ESRD registry (January 1, 1990), was also performed. We performed a subanalysis of participants with severe hypertension and nonoverweight participants, and we conducted a separate analysis of the overweight and obese participants. Details of these extensive sensitivity analyses are provided in the Supplement.
The final cohort included 2 658 238 individuals, of whom 7997 (0.3%) had a hypertension diagnosis (Figure 1) and 1 596 709 (60.1%) were male with a mean (SD) age of 17.4 (0.5) years. In the hypertensive group (n = 7997), 7235 (90.5%) were male, representing 60.1% (1 596 709) of the general cohort. Nearly half (49.0%) of the individuals in the hypertensive group were overweight (1559 [20.1%]) or obese (2243 [28.9%]) at the time of their examination, compared with the 12.9% of those in the nonhypertensive population (n = 2 650 241) who were either overweight (225 266 [8.8%]) or obese (104 023 [4.1%]). Most (63.7%) of the individuals in the hypertensive group originated from North America and Europe (3335 [42.4%]) as well as the former Soviet Union (1678 [21.3%]), although their percentage (43.7%) in the general cohort was lower (758 776 [29.3%] vs 373 314 [14.4%]). Baseline characteristics of the cohort are shown in Table 1.
During a median (interquartile range [IQR]) follow-up of 19.6 (10.4-31.2) years, within a total of 56 287 945 person-years, 2189 cases of ESRD with a crude incidence rate of 3.9 per 100 000 person-years were found among the nonhypertensive group (Table 2). Forty-two individuals (0.5%) who had a diagnosis of established hypertension at conscription later received a diagnosis of ESRD with a crude incidence rate of 20.2 per 100 000 person-years. Figure 2 describes the cumulative ESRD rate during follow-up. The rate of ESRD in individuals younger than 40 years was very low (eFigure 1 in the Supplement).
In an unadjusted Cox regression model (model 1), established hypertension was found to be associated with a 5-fold increased risk of ESRD (HR, 5.07; 95% CI, 3.73-6.88). In a multivariable model adjusted for year of birth, age at examination at a recruitment center, and sex (model 2), the association was attenuated but still statistically significant (HR, 3.33; 95% CI, 2.45-4.52). In the final multivariable model (model 3), also adjusted for body mass index, country of origin, years of education, and socioeconomic status, hypertension was found to almost double the risk of ESRD (HR, 1.98; 95% CI, 1.42-2.77). In a subanalysis excluding 95 adolescents with a severe hypertension diagnosis, the results of the multivariable analyses were similar to the results found in the entire cohort (HR, 1.93; 95% CI, 1.37-2.70). Additional subanalyses of various periods of enrollment (1977-2013 and 1967-1984) found similar results as findings of the analysis of the entire cohort population.
The results of the models are described extensively in eTables 1 and 2 and eFigure 2 in the Supplement. In a competing risk analysis, the association between hypertension and ESRD remained the same with an HR of 1.96 (95% CI, 1.40-2.75).
We conducted a sensitivity analysis consisting of only individuals who were not overweight or obese as defined by the Centers for Disease Control and Prevention. Overall, they comprised 87.1% (n = 2 241 177) of the original cohort. Nine cases (0.2%) of ESRD were observed among the hypertensive group. The unadjusted HR for ESRD was 2.59 (95% CI, 1.34-4.98). In the multivariable model adjusted for all of the above variables, hypertension among individuals was found associated with ESRD (HR, 2.11; 95% CI, 1.05-4.24). Hypertension in overweight and obese individuals was also found to be associated with ESRD, with an adjusted HR of 1.90 (95% CI, 1.29-2.78) (eTables 3 and 4 and eFigure 3 in the Supplement).
Adolescent hypertension is a public health issue with a growing prevalence, especially in the developed world, which can probably be explained by the increasing rates of overweight condition and obesity among adolescents worldwide as well as by the increasing awareness of hypertension at this age, leading to higher rates of diagnosis.14 However, the potential advantages and harms of essential hypertension screening among asymptomatic children and adolescents are still unclear, as stated by the US Preventive Services Task Force 2013 recommendations.15 The rate of hypertension among the adolescents in this study was lower (0.3%) than the rate previously reported worldwide (1%-5%).15,16 This difference may stem from the study population; that is, we excluded all individuals with morbidities known to be associated with hypertension such as diabetes and present and resolved glomerulonephritis. Moreover, most previous studies based their diagnosis of hypertension on a single BP measurement, whereas we labeled participants as hypertensive only if their established hypertension was confirmed by repeated elevated BP measurements and a thorough evaluation. We found that hypertension was much more prevalent among individuals who originated from or were descendants of European and North American countries, compared with other regions such as Asia and Africa.
The main finding of this study is that hypertension in late adolescence is associated with an almost 2-fold increased risk for future ESRD. Numerous previous studies have shown that hypertension during adolescence is an independent risk factor for future morbidity and mortality, especially after cardiovascular and cerebrovascular events.6-9 Leiba et al9 demonstrated that adolescents with established hypertension have a 3-fold risk of developing stroke-related mortality in midlife. The Falkstedt et al7 cohort study of Swedish military conscripts found that elevated BP was associated with cardiovascular morbidity. The Gray et al6 study of undergraduate 18-year-old males found that elevated BP was associated with future all-cause death, particularly cardiovascular death. Their study was unique in that their data related to middle-aged hypertension; thus, they used this information in their statistical models. Sundin et al17 found in their study of Swedish late-adolescent male conscripts that hypertension was associated with future ESRD. However, in their study, as well as in most other studies, the diagnosis of hypertension was based on only 1 BP measurement, whereas in this present study, the diagnosis was based on several elevated measurements.
Obesity during childhood and adolescence has been shown to be an independent risk factor for future ESRD.18,19 Obesity is associated with hypertension. This association was mainly studied in adults; however, in recent years, it has also been observed in adolescents, emphasizing the role of obesity in the growing prevalence of hypertension.14,20 To separate the association between essential hypertension in adolescents and obesity in future ESRD, we performed a subgroup analysis of nonoverweight adolescents as defined by the Centers for Disease Control and Prevention. Albeit the total number of ESRD cases in nonoverweight participants was much lower, the association remained statistically significant with a nearly 2-fold increased risk (HR, 2.11; 95% CI, 1.05-4.24). This subanalysis suggests that established hypertension in adolescents is independently associated with ESRD (unrelated to obesity).
A previous study using the same cohort found that persistent asymptomatic isolated microscopic hematuria increased the risk of future ESRD by 18.5-fold.21 Hematuria may suggest an underlying renal disease; therefore, the risk of developing ESRD is very high. In the present study, individuals with hematuria were excluded, and only healthy individuals with hypertension were included. We found that even mild to moderate hypertension doubles the risk of future ESRD. Another previous study showed that overweight condition (HR of 3.00) and obesity (HR of 6.89) were associated with ESRD.19
The probable risk for adolescents with hypertension to develop ESRD is relatively modest and transpires over an extensive period, but it is of major importance. This finding may suggest that nonmalignant hypertension, while being a close surrogate and strong promoter of chronic kidney disease progression, is a relatively modest initiator of the disease.12,13 That the pathogenesis of hypertension involves a balanced afferent and efferent arteriolar constriction, maintaining a relatively preserved glomerular filtration rate despite reduced blood flow, can explain why hypertension, although a major risk factor of cerebrovascular and cardiac disease, is only a modest risk factor for chronic kidney disease progression and ESRD.13
The association between hypertension and ESRD does not necessarily signify that hypertension is the cause of future ESRD. Although most data, especially from epidemiologic studies, support the belief that hypertension is a major cause of ESRD, some studies challenge this point of view, claiming that often the diagnosis of hypertensive nephrosclerosis, as the cause of ESRD, is based merely on the clinical judgment of the treating physician after excluding other known causes.13,22
Studies based on pathologic report findings of kidney biopsies and autopsies found that many cases of ESRD are believed to be caused by hypertension and had other etiologies.23,24 Marcantoni et al23 assessed the kidney biopsies of African American and white patients without diabetes to find the association between elevated BP, proteinuria, and the pathologic morphologic structure of renal lesions. They concluded that vascular sclerosis lesions caused by hypertension were insufficient to account for clinical renal failure. Hsu’s25 meta-analysis of studies including patients with hypertension treated for at least 1 year with antihypertensive medications endeavored to explore this controversy. He showed that antihypertensive treatment did not reduce the risk of renal dysfunction, questioning again the role of nonmalignant essential hypertension in chronic kidney disease. It has been shown that the higher rates of ESRD attributable to hypertension in African Americans, compared with white individuals, were partly explained by the genetic polymorphism of the APOL1 (Gen Bank 8542) gene rather than by hypertension alone.26 The results of our study suggest that adolescents with established hypertension are at a greater risk for future ESRD, however, it is still ambiguous whether hypertension is the cause of ESRD.
This study has several limitations. A noteworthy limitation is that we did not obtain any relevant clinical information such as BP levels during the follow-up period that might have affected the risk for future ESRD. However, Gray et al6 found that a hypertension diagnosis in middle-aged individuals only slightly attenuated the association between late-adolescence hypertension and cardiovascular mortality, emphasizing the potential role of adolescent hypertension itself. Another possible limitation was the lack of information as to the estimated glomerular filtration rate at the time of the medical examination. Nevertheless, all individuals with hypertension were evaluated to exclude renal disease. We excluded individuals with abnormal urinalysis findings and kidney-associated comorbidities, thus minimizing the probability that the study’s participants had an abnormal glomerular filtration rate. Furthermore, the outcome of the study comprised only individuals with ESRD without incorporating other levels of chronic kidney disease. In light of our findings, we would assume that this has yielded an underestimation of the association between hypertension and ESRD. Similar results were observed when only individuals enrolled from 1977 were included. In this subanalysis, the oldest participant would have been 30 years of age at the beginning of the registry with an extremely low rate of possibly missed ESRD cases.
This study has several strengths. It is a large population-based cohort including males and females from different racial/ethnic and socioeconomic backgrounds. The detailed data derived from the recruitment centers and the ESRD registry enabled us to adjust for most of the known confounders of hypertension. This study links well-established late-adolescence hypertension and ESRD; other studies have defined hypertension based only on 1 BP measurement. This definition of hypertension is known to be biased, especially in this population, owing to the stressful situation of the conscription examination itself and to the high rates of white-coat hypertension syndrome in this age group.27 Although up to 75% of hypertension cases in adolescents are essential,28 to exclude cases of secondary hypertension, we labeled individuals as hypertensive only if they were designated with diagnostic codes of essential hypertension. We excluded all individuals with known kidney disease or known to have morbidities that may reflect kidney disease such as microhematuria. Furthermore, we emphasized the independent role of essential hypertension unrelated to obesity. Finally, we showed that the association of hypertension with future ESRD was maintained among those who had received only a mild or moderate severity hypertension diagnosis.
Late-adolescence well-established essential hypertension appears to be associated with future ESRD. This association was found regardless of overweight condition and severity of the hypertension.
Accepted for Publication: November 9, 2018.
Corresponding Author: Ehud Grossman, MD, Internal Medicine D and Hypertension Unit, The Chaim Sheba Medical Center, 2 Derech Sheba, Migdal Ishpuz, 1st Floor, Ramat Gan 5266202, Israel (firstname.lastname@example.org).
Published Online: February 25, 2019. doi:10.1001/jamainternmed.2018.7632
Author Contributions: Drs Leiba and Fishman contributed equally to this manuscript. Drs Lieba and Fishman had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Leiba, Fishman, Gilad, Derazne, Shamiss, Shohat, Grossman.
Acquisition, analysis, or interpretation of data: Leiba, Fishman, Twig, Derazne, Shohat, Ron, Grossman.
Drafting of the manuscript: Leiba, Fishman, Twig, Gilad, Ron, Grossman.
Critical revision of the manuscript for important intellectual content: Leiba, Fishman, Twig, Gilad, Derazne, Shamiss, Shohat, Grossman.
Statistical analysis: Leiba, Fishman, Twig, Derazne.
Administrative, technical, or material support: Fishman, Twig, Shohat, Ron.
Conflict of Interest Disclosures: None reported.
Additional Contributions: The authors thank Mrs Phyllis Curchack Kornspan for her editorial services; she received compensation for her contribution.