The prevalence of hypertension in children and adolescents under 19 years of age has been reported to be 4.0% and has been increasing over the past two decades. Reference Song, Zhang and Yu1 This secular increase in the prevalence of hypertension has coincided with the concurrent obesity epidemic in children in the United States. Reference McNiece, Poffenbarger, Turner, Franco, Sorof and Portman2 Along with obesity, hypertension in children is reported to be more prevalent in African American and Hispanic populations. A national survey in the United States from 1963 to 2002 revealed that the prevalence of hypertension in children was the highest in Mexican American (4.6%), followed by in African American (4.2%), and in Caucasian (3.3%) populations. Reference Din-Dzietham, Liu, Bielo and Shamsa3 Similar observations were reported in a Texas school-based screening programme from 2000 to 2015 revealing a prevalence of hypertension of 3.1% in Hispanic, 2.7% in African American, 2.6% in Caucasian, and 1.7% in Asian adolescents (aged between 10 and 19 years). Reference Cheung, Bell, Samuel, Poffenbarger, Redwine and Samuels4
Left ventricle hypertrophy occurs in systemic hypertension as a remodelling adaptation to maintain contractile forces and counteract increased ventricular wall stress due to altered loading conditions. Reference Nadruz5 However, increased wall stress may cause activation of cytokines and other molecular pathways Reference Jia, Aroor, Hill and Sowers6 leading to apoptosis, myocardial fibrosis, and architectural changes. Therefore, left ventricular hypertrophy could have counterintuitive effects and may result in a decline in systolic reserve capacity and diastolic function. Reference Marketou, Parthenakis and Vardas7 Therefore, according to the American College of Cardiology/American Heart Association heart failure classification, left ventricular hypertrophy constitutes stage A (at high risk for developing heart failure without structural heart disease) of heart failure with potential to progress to stage B (structural heart disease but without symptoms of heart failure) and stage C (structural heart disease with prior or current symptoms of heart failure). Reference Hunt, Abraham and Chin8
It has been increasingly recognised that children with hypertension can develop left ventricular hypertrophy in their youth. As the presence of left ventricular hypertrophy in children with hypertension suggests end-organ damage, echocardiograms are regularly performed to monitor the development of left ventricular hypertrophy. The reported prevalence of left ventricular hypertrophy among children with hypertension varies from 14 to 41%. Reference Hanevold, Waller, Daniels, Portman and Sorof9–Reference Pruette, Fivush, Flynn and Brady13 However, data on racial differences in the prevalence of left ventricular hypertrophy in children with essential hypertension is lacking. Since blood pressure in childhood is highly predictive of BP in adulthood (the tracking of BP), Reference Du, Fernandez, Barshop, Chen, Urbina and Bazzano14,Reference Chen and Wang15 the presence of hypertension in childhood could persist in adulthood. Therefore, knowledge of racial prevalence of left ventricular hypertrophy in children constitutes an important public health need.
The primary aim of this study is to evaluate the impact of race on the prevalence of left ventricular hypertrophy among children with essential hypertension. The secondary aim is to evaluate if there are any racial differences in left ventricular remodelling in children with essential hypertension. The further secondary aim is to evaluate the impact of confounding factor of obesity on the development of left ventricular hypertrophy and left ventricular adaptation.
Methods
Study design
This study was a retrospective, single-center study approved by the Central Michigan University Institutional Review Board. The study included patients aged 3–18 years who were diagnosed with essential hypertension, according to the defined criteria (given under Definition section), in an outpatient setting between 2010 and 2020.
Patients were identified by searching diagnostic codes (ICD-9; 401.1 or 401.9, ICD-10; I-10) in our hospital medical records. Patients who had been discharged from the emergency department or inpatient settings with these codes were excluded, as patients with hypertension in the setting of acute illness might be due to secondary hypertension. Following the screening with diagnostic codes, individual medical records were reviewed. Our exclusion criteria included patients diagnosed with (1) secondary hypertension, (2) any congenital heart diseases, (3) congenital genitourinary anomalies, (4) renal diseases and chronic renal insufficiency, (5) unknown self-reported race, (6) lack of availability of echocardiogram shortly before or after the diagnosis of essential hypertension, and (7) patients < 3 years of age or > 18 years at the time of diagnosis of essential hypertension.
The demographic data collected included age at diagnosis, age at echocardiogram, gender, height and weight, and self-reported race. The first echocardiogram at the time of diagnosis was evaluated for various left ventricular measurements to calculate the left ventricle mass and left ventricular hypertrophy. Screening laboratory findings were also collected if available.
Definition
Blood pressure measurements were taken using a DINAMAP PROCARE 400 monitor (GE Healthcare, UK) after the patients rested for three to five minutes in a sitting position, using an appropriate cuff size that covered 80–100% of the arm’s circumference and a width that covered at least 40%. Reference Flynn, Kaelber and Baker-Smith16 The patients’ blood pressure was categorised according to the 2017 American Academy of Pediatrics guideline, with normal blood pressure defined as < 90th percentile for children aged < 13 years and < 120/80 mmHg for children aged ≥ 13 years. Reference Flynn, Kaelber and Baker-Smith16 Hypertension was defined as any elevation in blood pressure more than 90th percentile for age, gender, and height for children aged < 13 years and ≥ 120/80 mmHg for children aged ≥ 13 years. Essential hypertension was defined as hypertension in which secondary causes such as renovascular disease, renal failure, or other causes of secondary hypertension were excluded. Reference Carretero and Oparil17 In our institution, screening laboratory workup, including chemistry panel (electrolytes, blood urea nitrogen, and creatinine), renal ultrasound, urinalysis, and lipid profile, was performed in all patients with hypertension to exclude secondary hypertension. For other potential causes of hypertension, such as endocrine-related factors, further laboratory tests are conducted on a case-by-case basis. The age at diagnosis of hypertension was determined as the age when the patients were first presented to our nephrology or cardiology clinic after being referred by their primary care physician following confirmation of multiple abnormal blood pressure measurements at their office.
Body mass index was calculated as kg/m2, where kg represents the patient’s weight and m2 is height in meters squared. The body mass index percentile was determined based on the Centers for Disease Control and Prevention growth charts for children and teens ages 2 through 19 years. 18
Self-reported race was collected by asking the patients about their race during their first hospital visit, with options including African American, Asian, Caucasian, Middle Eastern, Hispanic/Latino, interracial, or not known. The answer was recorded in the patient’s medical chart as personal information.
Echocardiogram
A single reader (DT) who was blinded to the clinical and demographic data, evaluated all eligible patients’ echocardiograms at the time of their hypertension diagnosis. The echocardiograms were performed by certified and experienced cardiac sonographers using phase-arrayed appropriate probe for the patient’s age and body habitus interfaced with Philips Sono IE 33 echocardiographic equipment (Philips, Andover MA), and stored for offline analysis. Echocardiograms at our centre are performed in accordance with the American Society of Echocardiography guidelines. Reference Lopez, Colan and Frommelt19 To assess of left ventricular mass and relative wall thickness, left ventricular end-diastolic linear dimensions (left ventricular end-diastolic dimension, left ventricular wall thickness, and interventricular septal thickness) were measured using M-mode echocardiography from the parasternal short-axis view below the mitral valve leaflets. The measurements were performed using leading edge-to-leading edge method and were averaged from three measurements. To assess the intra- and inter-observer variability, 20 echocardiograms were re-analyzed by the same investigator and a second investigator (SA).
Left ventricular mass
The left ventricular mass was calculated using the standard formula: left ventricular mass (g) = 0.8 * 1.04 (IVSD + LVIDD + PWTD) Reference Din-Dzietham, Liu, Bielo and Shamsa3 - LVIDD Reference Din-Dzietham, Liu, Bielo and Shamsa3 ] + 0.6g, where IVSD is interventricular septum, LVIDD is left ventricular internal dimensions, and PWTD is posterior wall thickness. All measurements were performed at end-diastole. Reference Lang, Badano and Mor-Avi20 Left ventricular mass index was calculated as left ventricular mass/height (m)2.7. Left ventricular hypertrophy was defined as left ventricular mass index > 95th percentile for gender and age. For children ≥ 9 years old, the 95th percentile for left ventricular mass index was 40 g/m2.7 for females and 45 g/m2.7 for males. Reference Khoury, Mitsnefes, Daniels and Kimball21 For children < 9 years old, age and gender-specific 95th percentile cut-off values were used, as reported by Khoury et al. Reference Khoury, Mitsnefes, Daniels and Kimball21
Left ventricle relative wall thickness
Left ventricular relative wall thickness is a measure used to assess various geometric adaptations or remodelling of the left ventricle in response to hypertension. It was calculated from the standard formula: LVRWT = 2 * PWTD / LVIDD. An left ventricular relative wall thickness > 0.42 was considered abnormal. Reference Lang, Badano and Mor-Avi20 Different left ventricular geometry responses to hypertension were defined as follows: concentric hypertrophy (left ventricular mass index > 95th percentile and left ventricular relative wall thickness > 0.42), concentric re-modeling (left ventricular mass index ≤ 95th percentile and left ventricular relative wall thickness > 0.42), eccentric re-modeling (left ventricular mass index > 95th and left ventricular relative wall thickness ≤ 0.42), and normal left ventricular geometry when both left ventricular mass index and left ventricular relative wall thickness were normal (Supplemental Figure 1).
Statistical analysis
The statistical analysis was performed using SPSS version 28 (IBM SPSS Inc., Chicago, IL). Based on normality tests (Kolmogorov-Smirnov and Shapiro-Wilk), it was determined that there were no normally distributed continuous variables. Therefore, continuous variables were reported as median and interquartile range, while categorical variables were denoted by number and percentage. Due to the small sample size of other races, only African American and Caucasian races were included in the analysis. All pre-specified demographic and echocardiographic parameters between the groups were analysed using the Mann-Whitney U test and Chi-Square test, as appropriate. Univariate and multivariate logistic regression analysis were used to identify the significant parameters to predict left ventricular mass index > 95th percentile and left ventricular relative wall thickness > 0.42. A p < 0.05 was considered statistically significant. Inter-rater and intra-rater reliability were assessed by means of the Shrout-Fleiss intra-class correlation coefficient to determine consistency among raters. Reference Shrout and Fleiss22
Results
Patient population
A total of 422 patients, including 297 males (70.4%), were enrolled in the study, comprising 289 African American (68.5%) and 133 Caucasians (31.5%) (Supplemental Figure 2). The demographics and clinical data of the entire cohort are represented in Table 1. For the entire cohort, the median body mass index and body mass index percentile were 29.4 (22.7–37.2) kg/m2 and 98th (85–99) percentile, respectively. Laboratory tests were performed on 322 patients (76.3%) at the time of diagnosis, and the median values of serum haemoglobin A1c, total cholesterol, triglyceride, high-density lipoprotein, and low-density lipoprotein were all within normal limits (Table 1).
AA = African American; BMI = body mass index; HbA1c = haemoglobin A1c; HDL = high-density lipoprotein; HTN = hypertension; LDL = low-density lipoprotein; LV = left ventricular, LVRWT = left ventricle relative wall thickness.
* HbA1c level was available in 75 out of 422 patients only.
¶ Severe obesity was defined by body mass index≥120% of 95th percentile or body mass index≥35.
When comparing demographic data between two races, height (170 versus 166, p = 0.02), weight (86 versus 76, p = 0.002), and body mass index (31.2 versus 27.9, p = 0.026) were significantly higher among African American than Caucasian patients. Serum creatinine level and haemoglobin A1c were significantly higher in African American patients than Caucasians (creatinine; 0.77 versus 0.67, p < 0.001, haemoglobin A1c; 5.8 versus 5.4, p < 0.001).
Diagnosis of hypertension
The median age at the diagnosis of hypertension in the study population was 14.6 (12.1–16.3) years. The median systolic blood pressure and diastolic blood pressure for the entire cohort were 140 (131–149) mmHg and 76 (68–83) mmHg, respectively. Of the total patients, 54 (12.8%) had elevated blood pressure, 135 (32.0%) had Stage 1 hypertension, and 233 (55.2%) stage 2 (Supplemental Table 1). The systolic and diastolic blood pressure were not significantly different between African American and Caucasian patients (systolic blood pressure; 140 versus 140, p = 0.485, diastolic blood pressure; 76 versus 74, p = 0.290) (Table 1). The relationship between blood pressure and age, gender, race, and body mass index is shown in Figure 1. Most patients had an echocardiogram performed before starting anti-hypertensive medication, but 10 patients (2.4%) were already on medication at the time of diagnosis by their primary care physician. The median age at echocardiogram was 14.8 (12.5–16.4) years, and the median interval between the diagnosis of hypertension and the initial echocardiogram was 18 (4–48) days.
Left ventricular mass index, left ventricular relative wall thickness (Table 2)
The median left ventricular mass for the entire group was 138 (107–181) g, and the median left ventricular mass index was 35.2 (29.1–41.9) g/m2.7. Left ventricular hypertrophy, as defined by left ventricular mass index > 95th percentile, was present in 88 patients (20.9%). The median left ventricular relative wall thickness was 0.35 (0.29–0.43), and 114 patients (27.0%) had left ventricular relative wall thickness > 0.42. The left ventricular adaptation to essential hypertension was characterised as normal, concentric hypertrophy, concentric remodelling, and eccentric remodelling patterns in 259 (61.4%), 39 (9.2%), 75 (17.8%), 49 (11.6%) patients, respectively.
AA = African American; BMI = body mass index; LV = left ventricular; LVH = left ventricle hypertrophy; LVMI = LV mass index; LVRWT = left ventricle relative wall thickness.
Although there was no statistically significant difference in left ventricular mass index (g/m2.7) between the African American and Caucasian groups (35.4 versus 34.6, p = 0.24), African American patients had a higher left ventricular relative wall thickness (0.37 versus 0.32, p < 0.001) and a higher percentage presence of left ventricular relative wall thickness > 0.42 (30.1% versus 20.3%, p = 0.035) than Caucasians. Therefore, concentric adaptation (concentric remodelling and concentric hypertrophy) was more prevalent among African Americans than Caucasians. The presence of left ventricular mass index > 95th percentile was not significantly different between the two groups (22.5% versus 17.3%, p = 0.222).
Comparison between two races with stratification according to body mass index (Supplemental Table 2, Figure 2, Supplemental Figure 3)
The study further examined the potential impact of overweight and obesity on left ventricular mass index and left ventricular relative wall thickness by dividing the entire cohort into two groups: a group with normal body mass index and a group with overweight/obesity. The analysis of left ventricular mass, left ventricular mass index, and left ventricular relative wall thickness for subgroups of African American and Caucasian patients is presented in Supplemental Table 2 . The results showed no significant difference in left ventricular mass index between African American and Caucasian patients with normal body mass index (31.2 versus 29.3, p = 0.617) or with overweight/obesity (37.1 versus 36.1, p = 0.243). However, left ventricular relative wall thickness was significantly higher among African American patients than Caucasian patients for both with normal body mass index (0.37 versus 0.29, p < 0.01) and the overweight/obesity group (0.37 versus 0.33, p = 0.006). No significant difference was observed in left ventricular geometry response to essential hypertension between African American and Caucasian patients in either the normal body mass index or overweight/obesity groups. In addition, the study performed gender stratification in combination with body mass index, and the results remained consistent in all the echocardiographic parameters except for male patients in the overweight/obesity group (Supplemental Table 3).
Univariate logistic regression model to predict left ventricular hypertrophy (Supplemental Table 4)
The study conducted univariate logistic regression analysis to predict left ventricular mass index > 95th percentile and left ventricular relative wall thickness > 0.42. For left ventricular mass index > 95th percentile, the analysis identified body mass index percentile as the only significant variable (Odds ratio 1.04, p < 0.001). For left ventricular relative wall thickness > 0.42, the analysis identified African American (Odds ratio 1.69, p = 0.036) and age at diagnosis of hypertension (Odds ratio 1.10, p = 0.012) as significant variables in univariate regression. The final multivariable model confirmed that the significant predictors for left ventricular relative wall thickness > 0.42 were African American (Odds ratio 1.614, p = 0.059) and age at diagnosis of hypertension (Odds ratio 1.098, p = 0.016).
Reproducibility
The inter-rater reliability between rater 1 (DT) and 2 (SA) was assessed using six variables in 20 randomly selected echocardiograms, resulting in a total of 120 measurements. The two-way random model analysis showed an average intra-class correlation coefficient of 0.970. Similarly, the intra-rater reliability of rater 1 (DT) was assessed using the same six variables in 20 randomly selected echocardiograms, also resulting in 120 measurements. The one-way random model analysis revealed an average intra-class correlation coefficient of 0.979.
Discussion
Our study did not find any differences in the prevalence of left ventricular hypertrophy at the time of essential hypertension diagnosis between African American and Caucasian paediatric populations. However, we did observe a significantly higher prevalence of concentric adaptation of left ventricle in African Americans compared to Caucasians with essential hypertension.
We found that nearly 20% of children with essential hypertension had left ventricular hypertrophy at the time of initial presentation, emphasising the importance of obtaining an echocardiogram to manage essential hypertension. The prevalence of left ventricular hypertrophy among children with hypertension ranges from 14 to 41%, depending on the method used to diagnose left ventricular hypertrophy. Reference Hanevold, Waller, Daniels, Portman and Sorof9–Reference Pruette, Fivush, Flynn and Brady13 Monitoring for left ventricular hypertrophy is crucial as patients with left ventricular hypertrophy are at a higher risk for developing heart failure. Reference Hunt, Abraham and Chin8
Previous studies have reported a higher prevalence of left ventricular hypertrophy in African Americans than in Caucasians, Reference Drazner, Dries and Peshock23,Reference Levy, Anderson, Savage, Kannel, Christiansen and Castelli24 with higher prevalence observed in paediatric patients with hypertension. Reference Hanevold, Waller, Daniels, Portman and Sorof9,Reference Brady, Fivush, Flynn and Parekh10,Reference Daniels, Loggie, Khoury and Kimball12,Reference Pruette, Fivush, Flynn and Brady13,Reference Mendizabal, Khoury, Woo and Urbina25–Reference Daniels, Meyer, Liang and Bove31 However, these studies included patients with various significant confounding factors for left ventricular hypertrophy such as obesity, renal diseases, congenital heart diseases, or other causes of secondary hypertension. Our study is the first to focus on patients with essential hypertension.
Although we found no difference in the prevalence of obesity and overweight in the two races in our cohort, we conducted a subgroup analysis to determine the impact of race on left ventricular mass index and left ventricular relative wall thickness by stratifications according to the presence of overweight and obesity. We found that being African American was not associated with an increased left ventricular mass index after stratification according to body mass index percentile, but obesity or overweight (body mass index percentile) was an independent predictor of increased left ventricular mass index regardless of race.
The question remains whether the African American population has a higher left ventricular relative wall thickness (“left ventricular concentric adaptation”) compared to other races, regardless of the presence of hypertension. No previous studies compared left ventricular relative wall thickness among children with normal blood pressure between African Americans and other races. In adults, a study reported that concentric left ventricular hypertrophy and increased electrocardiographic voltage were associated with genetically determined African ancestry. Reference Alame, Garg and Kozlitina32 It is possible that African American patients need a different reference range for left ventricular relative wall thickness if they normally have higher left ventricular relative wall thickness compared with other races. However, another study comparing African American and Caucasian patients with hypertension found that African Americans had a higher prevalence of concentric adaptation compared to Caucasian patients, while there was no difference in left ventricular mass index. Reference Ejiofor, Di Nora and Cervesato33 This finding supports our observations, suggesting that the increased prevalence of concentric adaptation with a similar prevalence of left ventricular hypertrophy in response to an exaggerated burden of hypertension in African Americans compared to Caucasians in paediatric populations has a genetic underpinning. This genetic factor may play a significant role in cardiac remodelling in response to increased afterload. Reference Kizer, Arnett and Bella34 Further studies are needed to determine if the African American population has “thicker posterior wall” regardless of the presence of hypertension.
Limitation
This single-centre, retrospective study has a few limitations. First, the actual duration of hypertension is not clear due to our study design. Although we attempted to identify patients who underwent an echocardiogram soon after being diagnosed with hypertension, some patients may have long-standing hypertension before being identified by their primary care physician’s office. However, even in a prospective study, it would be difficult to determine the duration of hypertension. Second, we relied on self-reported race and were unable to collect data on genetically determined ancestry. Third, we assessed overweight and obesity using body mass index and were unable to use body fat percentage, which may be more effective in distinguishing patients with obesity from athletes. Fourth, as American Academy of Pediatrics 2017 guideline points out, Reference Flynn, Kaelber and Baker-Smith16 accurate blood pressure measurement can be challenging in individuals with obesity associated with appropriate cuff size selection. Moreover, obese or overweight children have been excluded from the most recent nomogram of blood pressure. These factors potentially influence accurate diagnosis of hypertension in this population. Lastly, 10 patients were taking anti-hypertensive medication when they underwent echocardiograms at our institution. Their primary care physicians had prescribed these medications upon diagnosing hypertension, prior to the evaluation at our facility. While the use of anti-hypertensive medication has the potential to affect left ventricular hypertrophy and left ventricular remodelling, we believe this had a relatively small impact on our results. These patients were referred to our centre shortly after receiving their hypertension diagnosis and starting their anti-hypertensive treatment at their primary care physician’s office.
Conclusions
A significant (20%) of children with essential hypertension had left ventricular hypertrophy even at the time of initial diagnosis. There were no differences in the prevalence of left ventricular hypertrophy in children with essential hypertension between African American and Caucasian populations when the rates of obesity and overweight were similar. It is more likely that the presence of obesity and overweight, rather than race, is associated with left ventricular hypertrophy. The major finding of our study is that children with African American race with essential hypertension had a significantly higher prevalence of concentric adaptation compared to Caucasians. Further studies are warranted to investigate the impact of differences in cardiac remodelling among races in response to hypertension.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951123003840.
Acknowledgements
None.
Author contributions
Dr Takajo, Mr. Przybycien, Dr Natarajan, and Dr Aggarwal conceptualised and designed the study, designed the data collected instruments, collected data, drafted the initial manuscript, and reviewed and revised the manuscript.
Dr Balakrishnan and Dr Singh critically reviewed the manuscript for important intellectual content.
All authors approved the final manuscript as submitted and agreed to be accountable for all aspects of the work.
Financial support
This research project received the financial support from American Academy of Pediatrics through the American Academy of Pediatrics Resident Research Grant Program.
Competing interests
None.
Ethical approval
This retrospective chart review study involving human participants was in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The Human Investigation Committee (IRB) of Central Michigan University approved this study.