Learning and Attention Problems Among Children With Pediatric Primary Hypertension

Abstract

OBJECTIVE

The objective was to determine whether children with sustained primary hypertension are at increased risk for learning disabilities (LDs), as a school-related manifestation of neurocognitive problems.

METHODS

A total of 201 children 10 to 18 years of age who were referred because of elevated blood pressure (BP) were included. Patients were categorized as having or not having hypertension, on the basis of BP evaluation at the initial hypertension clinic visit and subsequent confirmation of sustained elevated BP outside the clinic setting. Parents reported whether their child had a provider-confirmed LD or attention-deficit/hyperactivity disorder (ADHD).

RESULTS

A total of 101 children without hypertension and 100 children with hypertension were evaluated; 18% of the children (n = 37) had LDs. In comparison with children without hypertension, children with hypertension were significantly more likely to have LDs (18% vs 9%; P < .001), irrespective of comorbid ADHD. With adjustment for demographic variables and obesity, the odds of having LDs were elevated for subjects with hypertension, in comparison with subjects without hypertension (odds ratio: 4.1 [95% confidence interval: 1.8 –9.4]).

CONCLUSIONS

The rate of LDs, with or without ADHD treatment, was significantly higher for children with sustained primary hypertension, compared with children without hypertension. These findings add to the growing evidence for an association between primary hypertension and cognitive function and may inform treatment and monitoring decisions for these children who may be at risk for learning problems.

Adults with hypertension exhibit reduced performance in neurocognitive testing, compared with normotensive control subjects. Some of the largest and most consistent associations are noted within the domains of executive function and working memory.¹ This difference in cognition between subjects with and without hypertension has been found to be more pronounced among young adults, compared with middle-aged or older adults,² which lends biological plausibility to the possibility of a similar hypertension-cognition link for children.

In an analysis of a large, nationally representative database, our group showed that children with elevated systolic blood pressure (BP) demonstrated decreased performance on the digit span test, a measure of attention and working memory.³ We showed subsequently that children with sustained primary hypertension had decreased parental ratings of executive function, compared with normotensive control subjects.⁴ The latter results suggest that childhood hypertension might lead to observable deficits in the performance of daily activities, outside the isolated setting of formal neuropsychological examinations. However, our previous investigations intentionally excluded children with a history of learning disabilities (LDs) or attention-deficit/hyperactivity disorder (ADHD), to limit enrollment of subjects with pathophysiologic processes other than hypertension that might influence performance in neurocognitive testing. As a consequence, our previous studies did not allow us to investigate the association between primary hypertension and school-related learning difficulties. The objective of the present study was to determine whether children with primary hypertension are at increased risk for LDs, as a school-related manifestation of neurocognitive problems.

METHODS

We performed a retrospective review of data for patients referred to the pediatric hypertension clinic at the University of Rochester Medical Center over a 3-year period. All patients had been referred to the clinic because of elevated BP measurements obtained in their primary physician’s office. We included all patients 10 to 18 years of age who were not receiving antihypertensive medication at the time of referral. Patients with secondary hypertension or known developmental delay were excluded. For comparison, subjects were categorized as either having or not having hypertension. To be considered as having hypertension, subjects were required to have systolic and/or diastolic BP measurements of ≥95th percentile at the initial hypertension clinic visit and subsequent confirmation of sustained hypertension outside the clinic setting, either through 24-hour ambulatory BP monitoring or through elevated BP readings on ≥3 occasions, as measured by the school nurse or through home self-monitoring. All other subjects were considered to have prehypertension, were categorized as not having hypertension, and constituted the comparison group for this study.

During the study period, all referred patients were asked during the initial hypertension clinic visit about the presence of a LD and whether their child was receiving medication for ADHD. LDs were strictly defined as having a current individualized education plan or section 504 plan, both formal indicators of a student’s need for services to address an educational problem. Because students can receive these plans for various reasons, we required the educational plan to address a LD specifically, rather than a behavioral, sensory, medical, or other concern. Receiving medications for ADHD was defined as currently receiving a psychostimulant or other medication indicated for the management of ADHD symptoms. We chose this definition (ie, formal treatment for the diagnosis of ADHD) because of potential confusion or misuse of the diagnostic term ADHD.

For this study, the independent variable was hypertension status. The dependent variables were LDs and ADHD, as defined above. As covariates, we included the following demographic and medical variables available from our retrospective review: age, gender, low socioeconomic status (SES), and obesity. Low SES was defined as Census tract median income in the lowest quartile and/or residing in the inner city of Rochester, New York (defined as Census tracts within city limits where >50% of the children were enrolled in Medicaid). Obesity was defined as BMI of ≥95th percentile according to age and gender. SES was selected as a covariate because of the close link between SES and health care utilization.⁵ Obesity was included as a covariate because of the known interactive effect on cognition with hypertension.⁶

We performed bivariate analyses to compare the groups with and without hypertension with respect to demographic and dependent variables. We used Fisher’s exact test for proportions and 2-tailed Student’s t test or the Mann-Whitney U test, as indicated, for continuous variables. We then performed multivariate logistic regression analysis to examine the relationship between hypertension and LDs, with adjustment for age, gender, obesity, and low SES. All analyses were performed with SAS 9.2 (SAS Institute, Cary, NC). The significance level of the data analysis was set at .05. The study protocol was approved by the research subjects review board at the University of Rochester Medical Center.

RESULTS

Two hundred one consecutive referred patients met the inclusion criteria. Of those patients, 101 were classified as not having hypertension and 100 were classified as having hypertension. There were no significant differences between the groups with and without hypertension with respect to gender, age, low SES, BMI percentile, or obesity (Table 1).

TABLE 1.

Baseline Demographic Characteristics of Subjects With and Without Hypertension

	No Hypertension (N = 101)	Hypertension (N = 100)	P
Male/female, %	71/30	78/22	.26
Age, median (IQR), y	15 (13–16)	15 (13–17)	.56
Low SES, %	24	26	.74
BMI percentile, median (IQR)	97 (82–99)	98 (87.5–99)	.09
Obese, %	57	66	.24

IQR indicates interquartile range.

A total of 37 (18%) of 201 children had LDs, and 27 children (13%) were receiving treatment for ADHD. Among the 37 children with LDs, 10 (27%) were receiving ADHD treatment; among children without LDs, 17 (10%) were receiving ADHD treatment. In bivariate analysis, subjects with hypertension were more likely to have ADHD and LDs, compared with subjects without hypertension. The results for LDs were similar regardless of whether the subject was also receiving ADHD medications (Table 2).

TABLE 2.

Prevalence of ADHD and LDs According to BP Group

Diagnosis	n (%)			P
	Total (N = 201)	No Hypertension (N = 101)	Hypertension (N = 100)
ADHD	27 (13)	7 (7)	20 (20)	.007
LD
All subjects	37 (18)	9 (9)	28 (28)	<.001
Without ADHD	27 (13)	7 (7)	20 (20)	.002
With ADHD	10 (5)	2 (2)	8 (8)	.67

Univariate analysis and Fisher’s exact test were used.

Multivariate analysis showed that the odds of having LDs were elevated for subjects with hypertension, compared with subjects without hypertension, with adjustment for age, gender, obesity, and low SES (odds ratio [OR]: 4.1 [95% confidence interval [CI]: 1.8 –9.4]). The results were similar regardless of whether the subjects were receiving ADHD medications (LD without ADHD, OR: 4.0 [95% CI: 1.6 –10.2]). Crude and adjusted ORs and 95% CIs for LDs and hypertension, with and without ADHD, are presented in Table 3.

TABLE 3.

Multivariate Logistic Regression Analysis of Association Between Hypertension and LD Diagnosis

	All Subjects		Without ADHD
	β, Estimate ± SE	OR (95% CI)	β, Estimate ± SD	OR (95% CI)
Crude	1.38 ± 0.41	4.0 (1.8–9.0)	1.42 ± 0.47	4.1 (1.6–10.4)
Adjusteda	1.42 ± 0.42	4.1 (1.8–9.4)	1.39 ± 0.48	4.0 (1.6–10.2)

^aAdjusted for age, gender, proportion with obesity, and proportion with low SES.

DISCUSSION

We found that children with primary hypertension had higher LD rates, in comparison with a control group without hypertension. Regardless of ADHD medication history, children with hypertension had fourfold higher odds of LDs, compared with children without hypertension. In fact, 28% of the subjects with hypertension had a diagnosis of a LD, a proportion far in excess of the general population estimate for LDs of 5% among all school-aged children.⁷ The fact that the prevalence of LDs among subjects with hypertension was similar even among subjects who were not receiving medication for ADHD suggests that the association between LDs and hypertension was not an artifact of stimulant medication increasing BP and thereby biasing the results toward hypertension for the subjects with LDs.

Both ADHD treatment (ie, stimulant medications) and anxiety may confound the assessment of hypertension in children. Stimulants prescribed for ADHD can increase BP during active treatment, an observation that might have resulted in the increased rate of ADHD in the group with hypertension, because we defined the diagnosis of ADHD as receiving medication because of inattention.^8,9 ADHD and LDs are highly comorbid, however; ≥50% of subjects in ADHD samples may have a formal diagnosis of a LD or some problem with learning, and at least one-half of children in groups with LDs either meet the diagnostic criteria for ADHD or may have some attention problems.^10–12 Although it is possible that some of the children with ADHD exhibited hypertension because they were receiving stimulants for treatment of ADHD, it also is possible that the increased prevalence of ADHD in the hypertensive group is in part a reflection of neurocognitive difficulties among children with hypertension. We also note that anxiety is both an independent and comorbid disorder among children with ADHD and LDs; ~25% of children with ADHD have a comorbid anxiety disorder.¹³ Furthermore, research involving adults and children suggests an interaction between hypertension and internalizing mood problems (anxiety and depression).⁴ These observations suggest a potential interplay among primary hypertension, anxiety, and LDs in children.

Finally, the prevalence of primary hypertension in children is now estimated to be nearly 4%,¹⁴ and many of these children may fail to be identified,^15,16 a remarkable number that underscores the importance of understanding the potential impact of hypertension on neurocognition in childhood. Hypertension may lead to reversible neurocognitive problems in childhood.¹⁷ If so, then better identification and treatment of pediatric primary hypertension potentially could reduce the burden of LDs.

The reason for neurocognitive problems in children with hypertension is not known. However, even mild/moderate hypertension may be associated with central nervous system abnormalities.^18,19 Although this is not yet established anatomically in children, adults with chronic essential hypertension have a greater likelihood of regional changes in frontal lobe integrity (reduced prefrontal cortex and increased white matter hyperintensities), in comparison with matched control subjects with normotension, and the extent of these changes may correlate positively with disease duration.²⁰ What remains equivocal is whether there is a primary relationship between the presumed hypertension-related brain abnormalities and neurocognitive test performance.²⁰

The current study has several limitations. First, because of the retrospective nature of the study design, the clinical evidence for sustained hypertension versus no hypertension was not based on uniform criteria; some subjects had ambulatory BP monitoring data and others had less-rigorous evidence such as school nurse or home self-monitoring data. Therefore, it is possible that there were subjects in the hypertensive group who actually did not have hypertension and subjects in the nonhypertensive group who had hypertension. However, this limitation would bias the results toward no difference between the groups. Second, as noted above, the definition of ADHD was based solely on medication history, which would result in the inclusion of children taking stimulant medications that are known to increase BP.⁹ Last, the cross-sectional nature of the study prevents inferences regarding cause and effect between hypertension and LDs. For example, we cannot discern whether children with ADHD are more likely to have hypertension or whether having hypertension somehow predisposes children to ADHD. In addition, it is possible that some other factors resulted in both hypertension and LDs.

CONCLUSIONS

We report the novel finding of increased prevalence of learning problems in children diagnosed as having sustained primary hypertension. These results provide additional evidence of the emerging association between primary hypertension and neurocognitive problems in children. Future studies on the mechanism of this association and the impact of treatment of hypertension on learning and other cognitive skills in children are warranted.

WHAT’S KNOWN ON THIS SUBJECT

Children with sustained primary hypertension exhibit neuropsychological impairments in attention, working memory, and parent ratings of executive function skills. Previous studies excluded children with learning disabilities or attention-deficit/hyperactivity disorder, which precluded examination of associations between hypertension and academic problems.

WHAT THIS STUDY ADDS

This study found increased prevalence of learning problems among children with sustained primary hypertension, irrespective of attention-deficit/hyperactivity disorder history, in comparison with children without hypertension.

ABBREVIATIONS

ADHD

attention-deficit/hyperactivity disorder

BP

blood pressure

LD

learning disability

SES

socioeconomic status

OR

odds ratio

CI

confidence interval