- Email: [email protected]
Neurocognitive Dysfunction in Children, Adolescents, and Young Adults With CKD
Original Investigation Neurocognitive Dysfunction in Children, Adolescents, and Young Adults With CKD Rebecca L. Ruebner, MD, MSCE,1 Nina Laney, BA,1 Ji Young Kim, PhD,2 Erum A. Hartung, MD,1 Stephen R. Hooper, PhD,3 Jerilynn Radcliffe, PhD,4 and Susan L. Furth, MD, PhD1 Background: Neurocognitive dysfunction is a known complication in children with chronic kidney disease (CKD). However, less is known about putative mechanisms or modifiable risk factors. The objective of this study was to characterize and determine risk factors for cognitive dysfunction in children, adolescents, and young adults with CKD compared with controls. Study Design: Cross-sectional study. Setting & Participants: The Neurocognitive Assessment and Magnetic Resonance Imaging Analysis of Children and Young Adults With Chronic Kidney Disease (NiCK) Study included 90 individuals aged 8 to 25 years with CKD compared with 70 controls. Predictors: CKD versus control, estimated glomerular filtration rate (eGFR), ambulatory blood pressure. Outcomes: Performance on neurocognitive assessment with relevant tests grouped into 11 domains defined a priori by expert opinion. Results of tests were converted to age-normalized z scores. Measurements: Each neurocognitive domain was analyzed through linear regression, adjusting for eGFR and demographic and clinical variables. For domains defined by multiple tests, the median z score of tests in that domain was used. Results: We found significantly poorer performance in multiple areas of neurocognitive function among individuals with CKD compared with controls. Particular deficits were seen in domains related to attention, memory, and inhibitory control. Adjusted for demographic and clinical factors, we found lower performance in multiple domains with decreasing eGFRs (attention: b 5 0.053, P 5 0.02; visual spatial: b 5 0.062, P 5 0.02; and visual working memory: b 5 0.069, P 5 0.04). Increased diastolic load and decreased diastolic nocturnal dipping on ambulatory blood pressure monitoring were independently associated with impairments in neurocognitive performance. Limitations: Unable to assess changes in neurocognitive function over time, and neurocognitive tests were grouped into predetermined neurocognitive domains. Conclusions: Lower eGFR in children, adolescents, and young adults is associated with poorer neurocognitive performance, particularly in areas of attention, memory, and inhibitory control. Hypertension identified on ambulatory blood pressure monitoring may be an important risk factor, illustrating that neurocognitive function is an area of target-organ damage in CKD. Am J Kidney Dis. 67(4):567-575. ª 2016 by the National Kidney Foundation, Inc. INDEX WORDS: Neurocognitive dysfunction; cognitive deficit; chronic kidney disease (CKD); children; adolescents; young adult; pediatric; attention; memory; inhibitory control; renal function; estimated glomerular filtration rate (eGFR); blood pressure; hypertension.
C
ognitive dysfunction is a known complication of chronic kidney disease (CKD) in adults and children. Impairments have been described across a wide range of cognitive functions, including IQ, verbal and visual skills, memory, attention, processing speed, and executive function.1-5 Psychosocial factors, such as depression, stress, or school absences due to medical treatments, as well as malnutrition,
may partially be responsible for these deficits.3,6 However, cardiovascular disease risk factors have also been described as independent risk factors for cognitive dysfunction, both in patients with CKD and in the general population.7-12 Because cardiovascular disease risk factors are common in the context of CKD7,13-16 and many can be modified with medical management, early identification and aggressive
From the 1Division of Nephrology, Department of Pediatrics, and 2Biostatistics Core, Clinical and Translational Research Center, Children’s Hospital of Philadelphia, Philadelphia, PA; 3Department of Allied Health Sciences, University of North Carolina School of Medicine, Chapel Hill, NC; and 4Department of Clinical Psychology in Pediatrics, Clinical and Translational Research Center, Children’s Hospital of Philadelphia, University of Pennsylvania, Philadelphia, PA.
Received February 25, 2015. Accepted in revised form August 15, 2015. Originally published online October 15, 2015. Address correspondence to Susan L. Furth, MD, PhD, Division of Nephrology, Department of Pediatrics, Children’s Hospital of Philadelphia, University of Pennsylvania, 34th and Civic Center Blvd, Philadelphia, PA 19104. E-mail: [email protected] 2016 by the National Kidney Foundation, Inc. 0272-6386 http://dx.doi.org/10.1053/j.ajkd.2015.08.025
Am J Kidney Dis. 2016;67(4):567-575
567
Ruebner et al
treatment of these risk factors may be imperative in helping to preserve cognitive function in patients with CKD, especially in the critical stages of development during childhood and adolescence. Although cognitive dysfunction in adults with CKD has been demonstrated in a multitude of studies,4,17-19 less is known about cognitive dysfunction in children, adolescents, and young adults with CKD, and previous studies have been limited by small sample sizes or lack of a direct comparison group. To address these limitations and comprehensively assess cognitive function in children, adolescents, and young adults with CKD, we performed a cross-sectional study of a cohort of 8- to 25-year-olds with CKD and controls to: (1) examine group differences in performance on a neurocognitive assessment and (2) elucidate the underlying risk factors that contribute to cognitive dysfunction in CKD.
METHODS Study Design The Neurocognitive Assessment and Magnetic Resonance Imaging Analysis of Children and Young Adults With Chronic Kidney Disease (NiCK) Study is a cross-sectional investigation of child, adolescent, and young adult individuals with CKD stages 2 to 5 compared with controls of similar age, sex, and insurance status. The Institutional Review Board at the Children’s Hospital of Philadelphia approved this study (10-007919), and informed consent was obtained from all participants.
Study Population Participants were aged 8 to 25 years. The lower age limit of 8 years was used to ensure the ability to follow imaging and neurocognitive testing procedures. Because many neurocognitive measures are standardized only in English, participants were required to have English as their primary language. All patients seen in the nephrology clinic during the recruitment period were screened for inclusion. Patients were included in the CKD group if they had a known diagnosis of kidney disease with at least 2 estimated glomerular filtration rates (eGFRs) , 90 mL/ min/1.73 m2 over at least 6 months, including individuals who were currently or previously on dialysis therapy or had a functioning transplant. eGFR was calculated using the bedside CKD in Children (CKiD) equation20 for individuals aged 8 to 18 years and the 4-variable MDRD (Modification of Diet in Renal Disease) Study equation21 for those older than 18 years. Control participants were siblings or individuals of similar age, sex, and insurance status. Controls were recruited from within the Children’s Hospital of Philadelphia general pediatric practices. Control recruitment was intentionally lagged in order to collect demographic data on the first CKD participants. Criteria for control recruitment were frequently updated based on CKD participant demographics. Although demographic criteria for control recruitment were frequently updated, both groups were recruited simultaneously after the initial lag. Therefore, the 2 groups were not exactly matched on all demographic criteria. Individuals with a number of comorbid conditions that independently affect brain function or the ability to complete test measures were excluded from participating, including auditory impairment, history of traumatic brain injury, significant medical or neurologic abnormality affecting motor or higher cortical functioning (eg, seizure disorder; genetic syndromes; systemic 568
diseases that can affect the brain such as sickle cell disease, cerebral lupus, and spina bifida; gestational age , 32 weeks; or perinatal injury), profound developmental disability or sensory-motor difficulties that would preclude valid use of diagnostic instruments or scanning procedures, a severe Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition, Text Revision) axis I disorder or other psychiatric symptoms that would interfere with the participant’s ability to participate in the study, known drug or alcohol use within 24 hours of any assessment, or pregnancy.
Study Procedures Baseline demographic data were collected, including age at visit, sex, race, age at CKD diagnosis, cause of CKD, maternal education, insurance status, medical history, family history, and current medications. All participants underwent clinical, neurocognitive, and imaging assessments (imaging not discussed in this article).
Clinical Evaluations Participants had a physical examination including measurement of manual blood pressure (BP), heart rate, respiratory rate, height, weight, and body mass index. Laboratory data were collected, including complete blood cell count, comprehensive metabolic panel, calcium, phosphate, lipid panel, and urine studies for total protein, albumin, and creatinine. Urine pregnancy testing was performed in postpubertal females prior to magnetic resonance imaging. All participants underwent 24-hour ambulatory BP monitoring (ABPM) using a SpaceLabs 90217A-1 oscillometric device (SpaceLabs Healthcare). Data generated by ABPM included measures for systolic and diastolic BP and heart rate divided into waking hours, sleeping hours, and total 24 hours. Variables included mean BP, mean BP index (BP normalized for age, height, and sex), BP load (proportion of readings . 95th percentile for age, height, and sex), nocturnal BP dipping (percent reduction in mean BP from waking to sleeping), and BP and heart rate variability (coefficient of variation of BP and heart rate readings). Normative data for 95th percentiles were based on height and sex for participants younger than 18 years22 and American Heart Association2recommended limits for individuals 18 years or older.23 ABPM data were only included in the analysis if the monitor was worn for at least 21 hours, with at least 18 hours having one successful reading per hour and at least 75% successful waking and sleeping readings.
Neurocognitive Assessments A battery of age-specific standardized neurocognitive assessments was performed to assess targeted areas of neurocognition, including executive functioning, attention, memory, and visual spatial processing. All tests were administered by a trained examiner supervised by a licensed psychologist. To counterbalance the effects of fatigue and attention loss on test performance, the order in which these tests were administered to each study participant was systematically alternated. A behavior coding system was used by the examiner to provide perception of the validity of the test data collected.
Analytic Approach We assessed baseline demographic and clinical characteristics of the study population using mean 6 standard deviation for continuous variables and distributions for categorical variables. Because we were interested in assessing differences in cognitive domains as opposed to performance on individual tests, relevant tests were grouped a priori into 11 neurocognitive domains based on agreement by content area experts: attention, language, verbal memory, verbal working memory, visual memory, visual working memory, visual spatial, ratings of executive function, inhibitory control, planned problem solving, and set shifting. Tests used to define each Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD neurocognitive domain are summarized in Table S1 (provided as online supplementary material). Results of tests were converted to age-normalized z scores. For domains defined by multiple tests, the median z score of tests in that domain was used. In the primary analysis, we compared performance on the neurocognitive assessment between individuals with CKD and controls. Each domain was analyzed in a linear regression model with the median z score of tests in a domain as the dependent variable and CKD versus control as the main explanatory variable, adjusted for age, race, and maternal education level. In a secondary analysis, to determine the effect of level of kidney function independent of other demographic and clinical variables that may adversely affect cognitive performance, eGFR was used as the main explanatory variable. In this analysis, only the subgroup of individuals with complete ABPM data was included. We identified a priori a set of covariates to include age, race, sex, maternal education, BP control, hemoglobin level, calcium-phosphate balance, and dyslipidemia (cholesterol or triglyceride level). Maternal education was included because this has previously been shown to be an important marker of socioeconomic status associated with neurocognitive performance in children with CKD, with effects on verbal IQ, performance IQ, and full-scale IQ.1 Because ABPM yields multiple parameters that are highly correlated, we focused on a subset of parameters that have been reported to be associated with risk for cardiovascular disease and stroke and also were not highly correlated with each other. We initially assessed systolic and diastolic index, load, and dip and included in the final model only diastolic load and diastolic dip. Diastolic load was square root transformed. All other parameters were in the original scale. Stepwise regressions using the Akaike information criterion were performed to assess model performance. The single final model, which regressed neurocognitive domains on eGFR, included adjustment for age, race, sex, maternal education, hemoglobin level, cholesterol level, phosphate level, diastolic load, and nocturnal diastolic dipping. Because correlation between total cholesterol and triglyceride levels was high at r 5 0.6, only cholesterol level was included in the final models. No interaction term was included in the model because most interaction terms were dropped in the stepwise regression. Analyses were conducted using R, version 3.1.0 (R Foundation for Statistical Computing). All reported P values are 2 sided, and P , 0.05 was the threshold for statistical significance. We performed Bonferroni corrections to evaluate potential issues with multiple comparisons. Bonferroni corrections are reported in addition to uncorrected P values because the Bonferroni correction may be too conservative given that the neurocognitive domains are likely correlated.
RESULTS Demographic and Clinical Characteristics A total of 2,628 patients seen in nephrology clinics were screened for inclusion in the CKD group. Of these, 184 met inclusion criteria; most individuals were excluded because they had eGFRs $ 90 mL/ min/1.73 m2, did not have a second eGFR within 6 months, or were being followed up by a nephrologist for a diagnosis other than CKD. Of eligible individuals, 92 individuals with CKD were enrolled. There were 70 controls; 15 were siblings of individuals with CKD and 55 were unrelated controls. Baseline demographic features of the cohort are summarized in Table 1. There was similar age distribution between individuals with CKD and controls; Am J Kidney Dis. 2016;67(4):567-575
Table 1. Demographic Features of Patients With CKD and Controls
Characteristic
Age, y Male sex Black race Insurance Private Combined private and Medicaid Medicaid No insurance Missing Income ,$30,000 $30,000-$75,000 .$75,000 Missing Maternal education, y Duration of CKD, mo Cause of CKD CAKUTa FSGS Other glomerular disordersb Other nonglomerular disordersc Ever on dialysis Currently on dialysis Ever received a transplant Functioning transplant
CKD (n 5 92)
Control (n 5 70)
16.3 6 3.94 60 (65) 25 (27)
15.9 6 3.93 39 (56) 27 (39)
42 20 27 2 1
(46) (22) (29) (2) (1)
44 3 16 1 6
(63) (4) (23) (1) (9)
32 23 34 3
(35) (25) (37) (3)
18 24 27 1
(26) (34) (39) (1)
13.6 6 2.7 115.9 6 74.0 44 14 17 17
(48) (15) (18) (18)
19 3 24 21
(21) (3) (26) (23)
14.7 6 2.9
Note: Values for categorical variables are given as number (percentage); for continuous variables, as mean 6 standard deviation. Abbreviations: CAKUT, congenital anomalies of the kidney and urinary tract; CKD, chronic kidney disease; FSGS, focal segmental glomerulosclerosis. a Including renal dysplasia, posterior urethral valves, and other obstructive uropathies. b Including lupus nephritis, membranoproliferative glomerulonephritis, membranous nephropathy, Alport syndrome, immunoglobulin A nephropathy, Goodpasture syndrome, and granulomatosis with polyangiitis. c Including juvenile nephronophthisis, autosomal recessive polycystic kidney disease, chronic tubulointerstitial nephritis, cystinosis, and other.
individuals with CKD were aged 8.6 to 25.7 (mean, 16.3 6 3.9 [standard deviation]) years and controls were aged 9.0 to 25.0 (mean, 15.9 6 3.9) years. There was a slightly higher proportion of males in the CKD group (65% CKD vs 56% controls) and individuals of black race in the control group (27% CKD vs 39% controls). There were higher proportions of controls with private insurance (46% CKD vs 63% controls) and individuals with CKD in the lowest income group (,$30,000; 35% CKD vs 26% controls). There was similar distribution of maternal education (mean values of 13.3 years for CKD and 14.7 years for controls). For individuals with CKD, the mean duration of kidney disease was close to 569
Ruebner et al Table 2. Clinical Data for Patients With CKD and Controls Characteristic
Laboratory valuesb eGFR, mL/min/1.73 m2 eGFR category $60 mL/min/1.73 m2 45-,60 mL/min/1.73 m2 30-,45 mL/min/1.73 m2 15-,30 mL/min/1.73 m2 ,15 mL/min/1.73 m2 or dialysis Serum urea nitrogen, mg/dL Urine protein-creatinine ratio Calcium, mg/dL Phosphorus, mg/dL Albumin, g/dL Hemoglobin, g/dL Cholesterol Total, mg/dL HDL, mg/dL LDL, mg/dL Triglycerides, mg/dL ABPM parametersc Total systolic index Waking Sleeping Total diastolic index Waking Sleeping Total systolic load Waking Sleeping Total diastolic load Waking Sleeping Systolic dipping Diastolic dipping
CKD
Control
Pa
47.8 6 24.4
97.9 6 19.8
,0.001
30 (33) 19 (21) 18 (19) 18 (19) 7 (8) 29.0 6 17.5 1.3 6 2.3 9.4 6 0.5 4.6 6 1.0 4.1 6 0.6 12.6 6 1.9
12.7 6 3.2 0.08 6 0.3 9.5 6 0.3 4.4 6 0.6 4.4 6 0.3 14.0 6 1.4
,0.001 ,0.001 0.8 0.6 ,0.001 ,0.001
174.6 6 42.1 48.5 6 14.7 97.6 6 32.1 133.2 6 69.3
148.7 6 25.7 49.2 6 13.4 80.7 6 23.5 84.2 6 41.5
,0.001 0.5 0.002 ,0.001
0.92 6 0.08 0.92 6 0.08 0.93 6 0.10 0.88 6 0.09 0.87 6 0.09 0.89 6 0.12 24% 6 24% 23% 6 23% 25% 6 29% 19% 6 19% 17% 6 17% 23% 6 24% 11% 6 6% 18% 6 7%
0.89 6 0.08 0.89 6 0.08 0.89 6 0.08 0.84 6 0.08 0.84 6 0.08 0.83 6 0.09 15% 6 19% 16% 6 19% 14% 6 19% 11% 6 11% 11% 6 13% 11% 6 14% 13% 6 5% 21% 6 6%
0.04 0.07 0.03 0.009 0.05 0.001 0.03 0.05 0.01 0.01 0.04 ,0.001 0.2 0.007
Note: Values for categorical variables are given as number (percentage); for continuous variables, as mean 6 standard deviation. Abbreviations: ABPM, ambulatory blood pressure monitoring; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein. a P values are based on Wilcoxon rank sum test because some variables showed considerable departures from normality. b Based on 92 participants in the CKD group and 70 in the control group. c Based on 80 participants in the CKD group and 56 in the control group.
10 years. Of individuals with CKD, 21% had been on dialysis therapy at some point in their illness, and 26% had received a kidney transplant. Table 2 summarizes laboratory and ABPM data for CKD and control participants. The mean eGFR of individuals with CKD was 47.8 mL/min/1.73 m2 compared to 97.9 mL/min/1.73 m2 for controls. Individuals with CKD had lower mean hemoglobin (12.6 vs 14.0 g/dL), higher total cholesterol (174.6 vs 148.7 mg/dL), and higher triglyceride levels (133.2 vs 84.2 mg/dL). There were 80 CKD and 56 control participants who had complete ABPM data and were included in secondary analyses. These participants had similar demographic features to the overall cohort. In the subcohort with ABPM data, individuals with CKD tended to have a higher systolic and diastolic index, higher systolic and diastolic load, and 570
lesser diastolic nocturnal dipping compared with controls. Performance on Neurocognitive Assessment Figure 1 shows results of the primary analysis comparing the median difference in z scores between controls and individuals with CKD in the 11 neurocognitive domains, adjusted for age, race, sex, and maternal education level. Individuals with CKD tended to have lower performance in all neurocognitive domains, with statistically significant differences in attention; memory, including verbal and visual memory and short-term and working memory; visual spatial; and inhibitory control. Adjusting for multiple comparisons, attention, verbal memory, verbal working memory, visual spatial, and inhibitory control remained statistically significant. Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD
Figure 1. Difference in z scores between controls and individuals with chronic kidney disease (CKD) in 11 neurocognitive domains. Circles represent the difference in z scores, and bars represent 95% confidence intervals. Difference in z scores above zero indicates better performance among controls compared with individuals with CKD. Adjusted for age, race, sex, and maternal education. *Statistically significant difference in z scores between controls and individuals with CKD. All remained signficant after accounting for multiple comparisons except for visual memory.
In secondary analyses, we assessed the effect of level of kidney function on cognitive performance, adjusted for other demographic and clinical variables that may affect neurocognitive function. In this analysis, only individuals with complete ABPM data were included. In these models, eGFR was used as the main explanatory variable, adjusting for age, sex, race, maternal education, hemoglobin level, cholesterol level, phosphate level, and BP parameters, including diastolic load and nocturnal diastolic dipping. There was a trend toward better performance in
all domains with higher eGFRs, with statistically significant differences seen in the domains of attention, visual spatial, and visual working memory (Fig 2; Table 3). In addition to eGFR, metrics of BP control had significant effects in multiple domains (Table 3). Higher diastolic load was associated with poorer performance in language and verbal memory. Lower nocturnal diastolic dipping was associated with poorer performance in attention. Adjusted for multiple comparisons, these analyses did not achieve
Figure 2. Increase in z scores per 10-unit higher estimated glomerular filtration rate (eGFR). Circles represent the increase in z scores, and bars represent 95% confidence intervals. Models are adjusted for age, sex, race, maternal education, hemoglobin level, cholesterol level, phosphate level, diastolic load (square root transformed), and diastolic dipping. *Statistically significant difference in z scores per 10-unit higher eGFR. None remained significant after accounting for multiple comparisons. Am J Kidney Dis. 2016;67(4):567-575
571
Ruebner et al Table 3. Multivariable Linear Regression Analysis eGFRa
Domain
Attention Language Verbal memory Verbal working memory Visual memory Visual spatial Visual working memory Executive function Inhibitory control Problem solving Set shift
0.053 0.041 0.054 0.063 0.039 0.062 0.069 0.079 0.034 0.036 0.005
(0.008 to 0.098); 0.02d (20.013 to 0.096); 0.1 (20.048 to 0.156); 0.3 (20.030 to 0.156); 0.2 (20.033 to 0.110); 0.3 (0.008 to 0.115); 0.02d (0.003 to 0.136); 0.04d (20.009 to 0.167); 0.08 (20.005 to 0.073); 0.08 (20.001 to 0.074); 0.06 (20.054 to 0.064); 0.9
% Diastolic Loadb
20.003 20.076 20.166 20.056 20.069 20.055 20.024 0.003 0.001 20.046 0.058
(20.066 (20.152 (20.308 (20.186 (20.168 (20.130 (20.117 (20.121 (20.053 (20.098 (20.022
to to to to to to to to to to to
% Diastolic Dippingc
0.060); 0.9 20.001); 0.05d 20.023); 0.02d 0.074); 0.4 0.031); 0.2 0.020); 0.2 0.069); 0.6 0.127); 0.9 0.055); 0.9 0.006); 0.08 0.139); 0.2
0.017 0.007 20.005 0.003 20.003 0 0.005 0.003 0.012 20.007 20.013
(0.001 to 0.034); 0.04d (20.013 to 0.027); 0.5 (20.043 to 0.032); 0.8 (20.031 to 0.037); 0.9 (20.029 to 0.023); 0.8 (20.020 to 0.019); 0.9 (20.019 to 0.029); 0.7 (20.029 to 0.035); 0.8 (20.003 to 0.026); 0.1 (20.020 to 0.007); 0.3 (20.034 to 0.008); 0.2
Note: Values are given as b coefficient (95% confidence interval); P value. Analysis adjusted for age, race, maternal education, sex, hemoglobin level, cholesterol level, and phosphate level. Diastolic load variable was square root transformed. Abbreviation: eGFR, estimated glomerular filtration rate. a b coefficient indicates difference in domain z score per 10-unit higher eGFR. b b coefficient indicates difference in domain z score per 1% higher diastolic load. c b coefficient indicates difference in domain z score per 1% greater diastolic dip. d P ,0.05.
statistical significance but still tended to have nominally poorer performance with lower eGFRs.
DISCUSSION In this study of children, adolescents, and young adults with CKD, we found significantly lower performance in multiple areas of neurocognitive function among individuals with CKD compared with controls. Particular deficits were seen in an array of functions that capture attention, memory, and inhibitory control. After adjustment for other factors that may contribute to cognitive performance (including age, sex, race, maternal education, hypertension, anemia, hyperlipidemia, and hyperphosphatemia), we found lower neurocognitive performance in multiple domains with decreasing eGFRs. We also found that markers of BP control, including increased diastolic load and decreased nocturnal diastolic dipping, were associated with impairments in neurocognitive performance. This study characterizes the nature of cognitive dysfunction in children, adolescents, and young adults with CKD. In particular, dysfunction in areas related to attention and memory may have important implications for the delivery of health information to patients with CKD. Ancillary academic and cognitive support services may improve overall social functioning for children and young adults with kidney disease in academic and functional contexts.3 In addition, this study highlights important potentially modifiable cardiovascular disease risk factors for neurocognitive impairment, particularly hypertension. Hypertension is a well-described complication in CKD,16,24,25 and aggressive BP control has been associated with delayed progression to end-stage kidney disease.26 Our finding that higher diastolic load and decreased nocturnal dipping are independently associated with 572
poorer cognitive performance, even after controlling for socioeconomic and clinical factors, provides another important reason to screen for and treat hypertension in the CKD population because problems with cognition may represent another marker of targetorgan damage. Neurocognitive dysfunction is a well-described complication in patients with CKD. Numerous studies of adults with CKD have shown neurocognitive deficits in a range of areas, including learning, concentration, and attention.18,27 In a large cohort of 855 adults with CKD, lower eGFR was associated with lower scores in most cognitive domains, and individuals with advanced CKD (eGFR , 30 mL/min/1.73 m2) were more likely to have cognitive dysfunction compared with those with more mild disease.19 Among patients with kidney disease, stage of CKD, lower hemoglobin level, and increased parathyroid hormone level have been associated with cognitive dysfunction.8 Studies of adult dialysis patients have also shown that neurocognitive deficits are common, with abnormalities in areas of attention, processing speed, executive function, language, and memory.17,28 Patients who remain on dialysis therapy may have progressive declines in neurocognitive function, whereas kidney transplantation may lead to improved performance among patients previously on dialysis therapy.29,30 Among children with CKD, increased disease severity and longer duration of disease have been associated with neurocognitive deficits.1,31 In a study comparing 20 children with CKD with healthy controls, individuals with CKD were at risk for a lower IQ and deficits in executive functioning.32 In a study of 50 pediatric kidney transplant recipients, those with a kidney transplant had lower performance in verbal Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD
and visuospatial domains compared with controls.33 Our results are consistent with these findings. Hypertension has been associated with neurocognitive dysfunction in adults34-36 and children9,10,37 with and without CKD. In a community-based study of 337 adults without kidney disease, there was an inverse linear relationship between systolic BP and global cognitive performance, even within the normotensive range.34 In a population-based study of otherwise healthy school-aged children in the United States, children with systolic or diastolic BP above the 90th percentile for age, sex, and height had lower performance in areas of memory, attention, and concentration.37 Previous studies of children with CKD have not shown definitive relationships between hypertension and neurocognitive performance.1,31 In a large study of 383 children with mild to moderate CKD from the CKiD multicenter longitudinal study, high BP was associated with lower performance IQ score, even after adjustment for demographic and clinical variables that may affect neurocognitive function. However, high BP was not associated with measures of attention, verbal IQ, academic achievement, or parental ratings of executive function.10 Although previous studies have shown an association between hypertension and neurocognitive dysfunction, our study was unique in the use of ABPM. We found that abnormalities including increased diastolic load and decreased nocturnal diastolic dipping were associated with impairments in neurocognitive performance. Casual assessment of BP does not necessarily identify all patients with hypertension. In a study of 198 children from the CKiD cohort, 38% of children had masked hypertension, defined as normal casual BP but increased ambulatory BP. 24 Therefore, ABPM may provide an important tool in the assessment of children with CKD, not only to guide BP management, but also as a screen to assess risk for cognitive dysfunction. The mechanism of hypertension leading to neurocognitive deficits remains unclear. Possible causes include vascular remodeling, altered cerebrovascular reactivity, or direct neuronal effects,38-40 but there are few studies that directly explore the neurophysiologic impact of hypertension. In this study, we were not able to identify why elevated diastolic load and decreased nocturnal dipping are associated with cognitive deficits, although we postulate that decreased nocturnal dipping may suggest decreased cerebral blood flow autoregulation, which may be associated with decreased cerebral perfusion. Additional research is needed to understand the interplay of hypertension, anemia, and cerebral autoregulation and their contribution to neurocognitive function. Am J Kidney Dis. 2016;67(4):567-575
This study has several limitations. First, this was a cross-sectional study so we were unable to assess changes in neurocognitive function over time or alterations in neurocognition with changes in kidney function, dialysis, or transplantation status. There was a relatively small sample size in the CKD and control groups, potentially decreasing the ability to detect effects of variables such as eGFR, hypertension, and nocturnal dipping across all cognitive domains. In addition, this study included a relatively heterogeneous population of individuals with CKD, including a wide range of eGFRs, as well as both dialysis patients and transplant recipients. In secondary analyses, after adjusting for eGFR, we found a trend toward poorer performance in multiple domains among individuals who ever had a transplant, ever had dialysis, or required first renal replacement therapy before the age of 5 years. However, the number of participants in each of these subgroups was small, limiting our ability to make statistically significant conclusions about the effect of CKD severity on cognitive performance. When we adjusted for multiple comparisons, in the primary analysis, there was still a statistically significant difference in neurocognitive performance between individuals with CKD and controls in multiple domains. In secondary analyses assessing the effect of eGFR on neurocognitive performance, the effect of eGFR no longer reached statistical significance after adjusting for multiple comparisons. However, the nominal trend toward poorer neurocognitive performance with declining eGFR still generates important hypotheses that should be analyzed further in future larger studies. In addition, the control group consisted of unrelated individuals as well as a smaller proportion who were siblings of participants with decreased eGFR; the inclusion of siblings may present a different sample than individuals from the general population. Finally, we grouped individual neurocognitive tests into predetermined neurocognitive domains and selected the median test score for each domain as representative of that domain. There was overlap in individual tests across different domains. While this process is a conceptually viable approach to data reduction, it is unclear whether different results may have been obtained using a factor-based approach to data reduction, a strategy that would require a larger sample size, but would have eliminated potential redundancy across the targeted neurocognitive domains. In summary, we found significant neurocognitive differences among children and young adults with CKD compared with controls. Abnormalities in ambulatory BP may be an important risk factor for 573
Ruebner et al
cognitive dysfunction in CKD. The next steps of the NiCK Study will be to correlate neurocognitive test results with findings on structural and functional brain imaging to determine whether specific neurocognitive deficits correlate with changes in brain structure, connectivity, or blood flow. In addition, we will explore whether structural abnormalities are related to cardiovascular disease risk factors, including hypertension, anemia, dyslipidemia, and hyperphosphatemia. Examining these relationships will further the understanding of the biological underpinnings of neurocognitive abnormalities in CKD and may help determine targets for intervention to prevent cognitive dysfunction.
ACKNOWLEDGEMENTS Support: This project is funded in part under a Commonwealth Universal Research Enhancement grant with the Pennsylvania Department of Health (#SAP 4100054843). The Department specifically disclaims responsibility for any analyses, interpretations or conclusions. The funders did not have any role in study design, data collection, data analysis, writing of the report, or the decision to submit the report for publication. Financial Disclosure: The authors declare that they have no other relevant financial interests. Contributions: Research idea and study design: SRH, JR, SLF; data acquisition: NL, RLR; data analysis and interpretation: RLR, NL, JYK, EAH, SRH, JR, SLF; statistical analysis: JYK; mentorship: SLF. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved. RLR takes responsibility that this this study has been reported honestly, accurately, and transparently; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.
SUPPLEMENTARY MATERIAL Table S1: Neurocognitive battery and associated domains. Note: The supplementary material accompanying this article (http://dx.doi.org/10.1053/j.ajkd.2015.08.025) is available at www.ajkd.org
REFERENCES 1. Hooper SR, Gerson AC, Butler RW, et al. Neurocognitive functioning of children and adolescents with mild-tomoderate chronic kidney disease. Clin J Am Soc Nephrol. 2011;6(8):1824-1830. 2. Gipson DS, Wetherington CE, Duquette PJ, Hooper SR. The nervous system and chronic kidney disease in children. Pediatr Nephrol. 2004;19(8):832-839. 3. Gerson AC, Butler R, Moxey-Mims M, et al. Neurocognitive outcomes in children with chronic kidney disease: current findings and contemporary endeavors. Ment Retard Dev Disabil Res Rev. 2006;12(3):208-215. 4. Bugnicourt JM, Godefroy O, Chillon JM, Choukroun G, Massy ZA. Cognitive disorders and dementia in CKD: the neglected kidney-brain axis. J Am Soc Nephrol. 2013;24(3):353-363. 5. Duquette PJ, Hooper SR, Icard PF, et al. Neurodevelopmental status and adaptive behaviors in preschool children with chronic kidney disease. J Special Educ. 2009;43(1):45-51. 574
6. Brouhard BH, Donaldson LA, Lawry KW, et al. Cognitive functioning in children on dialysis and post-transplantation. Pediatr Transplant. 2000;4(4):261-267. 7. Drew DA, Weiner DE. Cognitive impairment in chronic kidney disease: keep vascular disease in mind. Kidney Int. 2014;85(3):505-507. 8. Kalaitzidis RG, Karasavvidou D, Tatsioni A, et al. Risk factors for cognitive dysfunction in CKD and hypertensive subjects. Int Urol Nephrol. 2013;45(6):1637-1646. 9. Kupferman JC, Lande MB, Adams HR, Pavlakis SG. Primary hypertension and neurocognitive and executive functioning in school-age children. Pediatr Nephrol. 2013;28(3):401-408. 10. Lande MB, Gerson AC, Hooper SR, et al. Casual blood pressure and neurocognitive function in children with chronic kidney disease: a report of the children with chronic kidney disease cohort study. Clin J Am Soc Nephrol. 2011;6(8):1831-1837. 11. Nash DT, Fillit H. Cardiovascular disease risk factors and cognitive impairment. Am J Cardiol. 2006;97(8):1262-1265. 12. Wong LJ, Kupferman JC, Prohovnik I, et al. Hypertension impairs vascular reactivity in the pediatric brain. Stroke. 2011;42(7):1834-1838. 13. Cachofeiro V, Goicochea M, de Vinuesa SG, Oubina P, Lahera V, Luno J. Oxidative stress and inflammation, a link between chronic kidney disease and cardiovascular disease. Kidney Int Suppl. 2008;111:S4-S9. 14. Menon V, Gul A, Sarnak MJ. Cardiovascular risk factors in chronic kidney disease. Kidney Int. 2005;68(4):1413-1418. 15. Mitsnefes MM. Cardiovascular disease in children with chronic kidney disease. J Am Soc Nephrol. 2012;23(4):578-585. 16. Wilson AC, Schneider MF, Cox C, et al. Prevalence and correlates of multiple cardiovascular risk factors in children with chronic kidney disease. Clin J Am Soc Nephrol. 2011;6:27592765. 17. Kalirao P, Pederson S, Foley RN, et al. Cognitive impairment in peritoneal dialysis patients. Am J Kidney Dis. 2011;57(4): 612-620. 18. Lee JJ, Chin HJ, Byun MS, et al. Impaired frontal executive function and predialytic chronic kidney disease. J Am Geriatr Soc. 2011;59(9):1628-1635. 19. Yaffe K, Ackerson L, Kurella Tamura M, et al. Chronic kidney disease and cognitive function in older adults: findings from the Chronic Renal Insufficiency Cohort cognitive study. J Am Geriatr Soc. 2010;58(2):338-345. 20. Schwartz GJ, Munoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20(3):629-637. 21. Levey AS, Greene T, Kusek JW, Beck GL, MDRD Study Group. A simplified equation to predict glomerular filtration rate from serum creatinine (abstract). J Am Soc Nephrol. 2000;11: 155A. 22. Wuhl E, Witte K, Soergel M, Mehls O, Schaefer F. Distribution of 24-h ambulatory blood pressure in children: normalized reference values and role of body dimensions. J Hypertens. 2002;20(10):1995-2007. 23. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Hypertension. 2005;45(1): 142-161. 24. Mitsnefes M, Flynn J, Cohn S, et al. Masked hypertension associates with left ventricular hypertrophy in children with CKD. J Am Soc Nephrol. 2010;21(1):137-144. Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD 25. Wong H, Mylrea K, Feber J, Drukker A, Filler G. Prevalence of complications in children with chronic kidney disease according to KDOQI. Kidney Int. 2006;70(3):585-590. 26. Wuhl E, Trivelli A, Picca S, et al. Strict blood-pressure control and progression of renal failure in children. N Engl J Med. 2009;361(17):1639-1650. 27. Hailpern SM, Melamed ML, Cohen HW, Hostetter TH. Moderate chronic kidney disease and cognitive function in adults 20 to 59 years of age: Third National Health and Nutrition Examination Survey (NHANES III). J Am Soc Nephrol. 2007;18(7): 2205-2213. 28. Post JB, Morin KG, Sano M, Jegede AB, Langhoff E, Spungen AM. Increased presence of cognitive impairment in hemodialysis patients in the absence of neurological events. Am J Nephrol. 2012;35(2):120-126. 29. Harciarek M, Biedunkiewicz B, LichodziejewskaNiemierko M, Debska-Slizien A, Rutkowski B. Continuous cognitive improvement 1 year following successful kidney transplant. Kidney Int. 2011;79(12):1353-1360. 30. Harciarek M, Williamson JB, Biedunkiewicz B, Lichodziejewska-Niemierko M, Debska-Slizien A, Rutkowski B. Risk factors for selective cognitive decline in dialyzed patients with end-stage renal disease: evidence from verbal fluency analysis. J Int Neuropsychol Soc. 2012;18(1):162-167. 31. Slickers J, Duquette P, Hooper S, Gipson D. Clinical predictors of neurocognitive deficits in children with chronic kidney disease. Pediatr Nephrol. 2007;22(4):565-572.
Am J Kidney Dis. 2016;67(4):567-575
32. Gipson DS, Hooper SR, Duquette PJ, et al. Memory and executive functions in pediatric chronic kidney disease. Child Neuropsychol. 2006;12(6):391-405. 33. Haavisto A, Korkman M, Holmberg C, Jalanko H, Qvist E. Neuropsychological profile of children with kidney transplants. Nephrol Dial Transplant. 2012;27(6):2594-2601. 34. Knecht S, Wersching H, Lohmann H, et al. High-normal blood pressure is associated with poor cognitive performance. Hypertension. 2008;51(3):663-668. 35. Launer LJ, Ross GW, Petrovitch H, et al. Midlife blood pressure and dementia: the Honolulu-Asia Aging Study. Neurobiol Aging. 2000;21(1):49-55. 36. Vicario A, Martinez CD, Baretto D, Diaz Casale A, Nicolosi L. Hypertension and cognitive decline: impact on executive function. J Clin Hypertens (Greenwich). 2005;7(10):598-604. 37. Lande MB, Kaczorowski JM, Auinger P, Schwartz GJ, Weitzman M. Elevated blood pressure and decreased cognitive function among school-age children and adolescents in the United States. J Pediatr. 2003;143(6):720-724. 38. Ostrovskaya MA, Rojas M, Kupferman JC, et al. Executive function and cerebrovascular reactivity in pediatric hypertension. J Child Neurol. 2015;30(5):543-546. 39. Jennings JR, Zanstra Y. Is the brain the essential in hypertension? Neuroimage. 2009;47(3):914-921. 40. Jennings JR. Autoregulation of blood pressure and thought: preliminary results of an application of brain imaging to psychosomatic medicine. Psychosom Med. 2003;65(3):384-395.
575
C
ognitive dysfunction is a known complication of chronic kidney disease (CKD) in adults and children. Impairments have been described across a wide range of cognitive functions, including IQ, verbal and visual skills, memory, attention, processing speed, and executive function.1-5 Psychosocial factors, such as depression, stress, or school absences due to medical treatments, as well as malnutrition,
may partially be responsible for these deficits.3,6 However, cardiovascular disease risk factors have also been described as independent risk factors for cognitive dysfunction, both in patients with CKD and in the general population.7-12 Because cardiovascular disease risk factors are common in the context of CKD7,13-16 and many can be modified with medical management, early identification and aggressive
From the 1Division of Nephrology, Department of Pediatrics, and 2Biostatistics Core, Clinical and Translational Research Center, Children’s Hospital of Philadelphia, Philadelphia, PA; 3Department of Allied Health Sciences, University of North Carolina School of Medicine, Chapel Hill, NC; and 4Department of Clinical Psychology in Pediatrics, Clinical and Translational Research Center, Children’s Hospital of Philadelphia, University of Pennsylvania, Philadelphia, PA.
Received February 25, 2015. Accepted in revised form August 15, 2015. Originally published online October 15, 2015. Address correspondence to Susan L. Furth, MD, PhD, Division of Nephrology, Department of Pediatrics, Children’s Hospital of Philadelphia, University of Pennsylvania, 34th and Civic Center Blvd, Philadelphia, PA 19104. E-mail: [email protected] 2016 by the National Kidney Foundation, Inc. 0272-6386 http://dx.doi.org/10.1053/j.ajkd.2015.08.025
Am J Kidney Dis. 2016;67(4):567-575
567
Ruebner et al
treatment of these risk factors may be imperative in helping to preserve cognitive function in patients with CKD, especially in the critical stages of development during childhood and adolescence. Although cognitive dysfunction in adults with CKD has been demonstrated in a multitude of studies,4,17-19 less is known about cognitive dysfunction in children, adolescents, and young adults with CKD, and previous studies have been limited by small sample sizes or lack of a direct comparison group. To address these limitations and comprehensively assess cognitive function in children, adolescents, and young adults with CKD, we performed a cross-sectional study of a cohort of 8- to 25-year-olds with CKD and controls to: (1) examine group differences in performance on a neurocognitive assessment and (2) elucidate the underlying risk factors that contribute to cognitive dysfunction in CKD.
METHODS Study Design The Neurocognitive Assessment and Magnetic Resonance Imaging Analysis of Children and Young Adults With Chronic Kidney Disease (NiCK) Study is a cross-sectional investigation of child, adolescent, and young adult individuals with CKD stages 2 to 5 compared with controls of similar age, sex, and insurance status. The Institutional Review Board at the Children’s Hospital of Philadelphia approved this study (10-007919), and informed consent was obtained from all participants.
Study Population Participants were aged 8 to 25 years. The lower age limit of 8 years was used to ensure the ability to follow imaging and neurocognitive testing procedures. Because many neurocognitive measures are standardized only in English, participants were required to have English as their primary language. All patients seen in the nephrology clinic during the recruitment period were screened for inclusion. Patients were included in the CKD group if they had a known diagnosis of kidney disease with at least 2 estimated glomerular filtration rates (eGFRs) , 90 mL/ min/1.73 m2 over at least 6 months, including individuals who were currently or previously on dialysis therapy or had a functioning transplant. eGFR was calculated using the bedside CKD in Children (CKiD) equation20 for individuals aged 8 to 18 years and the 4-variable MDRD (Modification of Diet in Renal Disease) Study equation21 for those older than 18 years. Control participants were siblings or individuals of similar age, sex, and insurance status. Controls were recruited from within the Children’s Hospital of Philadelphia general pediatric practices. Control recruitment was intentionally lagged in order to collect demographic data on the first CKD participants. Criteria for control recruitment were frequently updated based on CKD participant demographics. Although demographic criteria for control recruitment were frequently updated, both groups were recruited simultaneously after the initial lag. Therefore, the 2 groups were not exactly matched on all demographic criteria. Individuals with a number of comorbid conditions that independently affect brain function or the ability to complete test measures were excluded from participating, including auditory impairment, history of traumatic brain injury, significant medical or neurologic abnormality affecting motor or higher cortical functioning (eg, seizure disorder; genetic syndromes; systemic 568
diseases that can affect the brain such as sickle cell disease, cerebral lupus, and spina bifida; gestational age , 32 weeks; or perinatal injury), profound developmental disability or sensory-motor difficulties that would preclude valid use of diagnostic instruments or scanning procedures, a severe Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition, Text Revision) axis I disorder or other psychiatric symptoms that would interfere with the participant’s ability to participate in the study, known drug or alcohol use within 24 hours of any assessment, or pregnancy.
Study Procedures Baseline demographic data were collected, including age at visit, sex, race, age at CKD diagnosis, cause of CKD, maternal education, insurance status, medical history, family history, and current medications. All participants underwent clinical, neurocognitive, and imaging assessments (imaging not discussed in this article).
Clinical Evaluations Participants had a physical examination including measurement of manual blood pressure (BP), heart rate, respiratory rate, height, weight, and body mass index. Laboratory data were collected, including complete blood cell count, comprehensive metabolic panel, calcium, phosphate, lipid panel, and urine studies for total protein, albumin, and creatinine. Urine pregnancy testing was performed in postpubertal females prior to magnetic resonance imaging. All participants underwent 24-hour ambulatory BP monitoring (ABPM) using a SpaceLabs 90217A-1 oscillometric device (SpaceLabs Healthcare). Data generated by ABPM included measures for systolic and diastolic BP and heart rate divided into waking hours, sleeping hours, and total 24 hours. Variables included mean BP, mean BP index (BP normalized for age, height, and sex), BP load (proportion of readings . 95th percentile for age, height, and sex), nocturnal BP dipping (percent reduction in mean BP from waking to sleeping), and BP and heart rate variability (coefficient of variation of BP and heart rate readings). Normative data for 95th percentiles were based on height and sex for participants younger than 18 years22 and American Heart Association2recommended limits for individuals 18 years or older.23 ABPM data were only included in the analysis if the monitor was worn for at least 21 hours, with at least 18 hours having one successful reading per hour and at least 75% successful waking and sleeping readings.
Neurocognitive Assessments A battery of age-specific standardized neurocognitive assessments was performed to assess targeted areas of neurocognition, including executive functioning, attention, memory, and visual spatial processing. All tests were administered by a trained examiner supervised by a licensed psychologist. To counterbalance the effects of fatigue and attention loss on test performance, the order in which these tests were administered to each study participant was systematically alternated. A behavior coding system was used by the examiner to provide perception of the validity of the test data collected.
Analytic Approach We assessed baseline demographic and clinical characteristics of the study population using mean 6 standard deviation for continuous variables and distributions for categorical variables. Because we were interested in assessing differences in cognitive domains as opposed to performance on individual tests, relevant tests were grouped a priori into 11 neurocognitive domains based on agreement by content area experts: attention, language, verbal memory, verbal working memory, visual memory, visual working memory, visual spatial, ratings of executive function, inhibitory control, planned problem solving, and set shifting. Tests used to define each Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD neurocognitive domain are summarized in Table S1 (provided as online supplementary material). Results of tests were converted to age-normalized z scores. For domains defined by multiple tests, the median z score of tests in that domain was used. In the primary analysis, we compared performance on the neurocognitive assessment between individuals with CKD and controls. Each domain was analyzed in a linear regression model with the median z score of tests in a domain as the dependent variable and CKD versus control as the main explanatory variable, adjusted for age, race, and maternal education level. In a secondary analysis, to determine the effect of level of kidney function independent of other demographic and clinical variables that may adversely affect cognitive performance, eGFR was used as the main explanatory variable. In this analysis, only the subgroup of individuals with complete ABPM data was included. We identified a priori a set of covariates to include age, race, sex, maternal education, BP control, hemoglobin level, calcium-phosphate balance, and dyslipidemia (cholesterol or triglyceride level). Maternal education was included because this has previously been shown to be an important marker of socioeconomic status associated with neurocognitive performance in children with CKD, with effects on verbal IQ, performance IQ, and full-scale IQ.1 Because ABPM yields multiple parameters that are highly correlated, we focused on a subset of parameters that have been reported to be associated with risk for cardiovascular disease and stroke and also were not highly correlated with each other. We initially assessed systolic and diastolic index, load, and dip and included in the final model only diastolic load and diastolic dip. Diastolic load was square root transformed. All other parameters were in the original scale. Stepwise regressions using the Akaike information criterion were performed to assess model performance. The single final model, which regressed neurocognitive domains on eGFR, included adjustment for age, race, sex, maternal education, hemoglobin level, cholesterol level, phosphate level, diastolic load, and nocturnal diastolic dipping. Because correlation between total cholesterol and triglyceride levels was high at r 5 0.6, only cholesterol level was included in the final models. No interaction term was included in the model because most interaction terms were dropped in the stepwise regression. Analyses were conducted using R, version 3.1.0 (R Foundation for Statistical Computing). All reported P values are 2 sided, and P , 0.05 was the threshold for statistical significance. We performed Bonferroni corrections to evaluate potential issues with multiple comparisons. Bonferroni corrections are reported in addition to uncorrected P values because the Bonferroni correction may be too conservative given that the neurocognitive domains are likely correlated.
RESULTS Demographic and Clinical Characteristics A total of 2,628 patients seen in nephrology clinics were screened for inclusion in the CKD group. Of these, 184 met inclusion criteria; most individuals were excluded because they had eGFRs $ 90 mL/ min/1.73 m2, did not have a second eGFR within 6 months, or were being followed up by a nephrologist for a diagnosis other than CKD. Of eligible individuals, 92 individuals with CKD were enrolled. There were 70 controls; 15 were siblings of individuals with CKD and 55 were unrelated controls. Baseline demographic features of the cohort are summarized in Table 1. There was similar age distribution between individuals with CKD and controls; Am J Kidney Dis. 2016;67(4):567-575
Table 1. Demographic Features of Patients With CKD and Controls
Characteristic
Age, y Male sex Black race Insurance Private Combined private and Medicaid Medicaid No insurance Missing Income ,$30,000 $30,000-$75,000 .$75,000 Missing Maternal education, y Duration of CKD, mo Cause of CKD CAKUTa FSGS Other glomerular disordersb Other nonglomerular disordersc Ever on dialysis Currently on dialysis Ever received a transplant Functioning transplant
CKD (n 5 92)
Control (n 5 70)
16.3 6 3.94 60 (65) 25 (27)
15.9 6 3.93 39 (56) 27 (39)
42 20 27 2 1
(46) (22) (29) (2) (1)
44 3 16 1 6
(63) (4) (23) (1) (9)
32 23 34 3
(35) (25) (37) (3)
18 24 27 1
(26) (34) (39) (1)
13.6 6 2.7 115.9 6 74.0 44 14 17 17
(48) (15) (18) (18)
19 3 24 21
(21) (3) (26) (23)
14.7 6 2.9
Note: Values for categorical variables are given as number (percentage); for continuous variables, as mean 6 standard deviation. Abbreviations: CAKUT, congenital anomalies of the kidney and urinary tract; CKD, chronic kidney disease; FSGS, focal segmental glomerulosclerosis. a Including renal dysplasia, posterior urethral valves, and other obstructive uropathies. b Including lupus nephritis, membranoproliferative glomerulonephritis, membranous nephropathy, Alport syndrome, immunoglobulin A nephropathy, Goodpasture syndrome, and granulomatosis with polyangiitis. c Including juvenile nephronophthisis, autosomal recessive polycystic kidney disease, chronic tubulointerstitial nephritis, cystinosis, and other.
individuals with CKD were aged 8.6 to 25.7 (mean, 16.3 6 3.9 [standard deviation]) years and controls were aged 9.0 to 25.0 (mean, 15.9 6 3.9) years. There was a slightly higher proportion of males in the CKD group (65% CKD vs 56% controls) and individuals of black race in the control group (27% CKD vs 39% controls). There were higher proportions of controls with private insurance (46% CKD vs 63% controls) and individuals with CKD in the lowest income group (,$30,000; 35% CKD vs 26% controls). There was similar distribution of maternal education (mean values of 13.3 years for CKD and 14.7 years for controls). For individuals with CKD, the mean duration of kidney disease was close to 569
Ruebner et al Table 2. Clinical Data for Patients With CKD and Controls Characteristic
Laboratory valuesb eGFR, mL/min/1.73 m2 eGFR category $60 mL/min/1.73 m2 45-,60 mL/min/1.73 m2 30-,45 mL/min/1.73 m2 15-,30 mL/min/1.73 m2 ,15 mL/min/1.73 m2 or dialysis Serum urea nitrogen, mg/dL Urine protein-creatinine ratio Calcium, mg/dL Phosphorus, mg/dL Albumin, g/dL Hemoglobin, g/dL Cholesterol Total, mg/dL HDL, mg/dL LDL, mg/dL Triglycerides, mg/dL ABPM parametersc Total systolic index Waking Sleeping Total diastolic index Waking Sleeping Total systolic load Waking Sleeping Total diastolic load Waking Sleeping Systolic dipping Diastolic dipping
CKD
Control
Pa
47.8 6 24.4
97.9 6 19.8
,0.001
30 (33) 19 (21) 18 (19) 18 (19) 7 (8) 29.0 6 17.5 1.3 6 2.3 9.4 6 0.5 4.6 6 1.0 4.1 6 0.6 12.6 6 1.9
12.7 6 3.2 0.08 6 0.3 9.5 6 0.3 4.4 6 0.6 4.4 6 0.3 14.0 6 1.4
,0.001 ,0.001 0.8 0.6 ,0.001 ,0.001
174.6 6 42.1 48.5 6 14.7 97.6 6 32.1 133.2 6 69.3
148.7 6 25.7 49.2 6 13.4 80.7 6 23.5 84.2 6 41.5
,0.001 0.5 0.002 ,0.001
0.92 6 0.08 0.92 6 0.08 0.93 6 0.10 0.88 6 0.09 0.87 6 0.09 0.89 6 0.12 24% 6 24% 23% 6 23% 25% 6 29% 19% 6 19% 17% 6 17% 23% 6 24% 11% 6 6% 18% 6 7%
0.89 6 0.08 0.89 6 0.08 0.89 6 0.08 0.84 6 0.08 0.84 6 0.08 0.83 6 0.09 15% 6 19% 16% 6 19% 14% 6 19% 11% 6 11% 11% 6 13% 11% 6 14% 13% 6 5% 21% 6 6%
0.04 0.07 0.03 0.009 0.05 0.001 0.03 0.05 0.01 0.01 0.04 ,0.001 0.2 0.007
Note: Values for categorical variables are given as number (percentage); for continuous variables, as mean 6 standard deviation. Abbreviations: ABPM, ambulatory blood pressure monitoring; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein. a P values are based on Wilcoxon rank sum test because some variables showed considerable departures from normality. b Based on 92 participants in the CKD group and 70 in the control group. c Based on 80 participants in the CKD group and 56 in the control group.
10 years. Of individuals with CKD, 21% had been on dialysis therapy at some point in their illness, and 26% had received a kidney transplant. Table 2 summarizes laboratory and ABPM data for CKD and control participants. The mean eGFR of individuals with CKD was 47.8 mL/min/1.73 m2 compared to 97.9 mL/min/1.73 m2 for controls. Individuals with CKD had lower mean hemoglobin (12.6 vs 14.0 g/dL), higher total cholesterol (174.6 vs 148.7 mg/dL), and higher triglyceride levels (133.2 vs 84.2 mg/dL). There were 80 CKD and 56 control participants who had complete ABPM data and were included in secondary analyses. These participants had similar demographic features to the overall cohort. In the subcohort with ABPM data, individuals with CKD tended to have a higher systolic and diastolic index, higher systolic and diastolic load, and 570
lesser diastolic nocturnal dipping compared with controls. Performance on Neurocognitive Assessment Figure 1 shows results of the primary analysis comparing the median difference in z scores between controls and individuals with CKD in the 11 neurocognitive domains, adjusted for age, race, sex, and maternal education level. Individuals with CKD tended to have lower performance in all neurocognitive domains, with statistically significant differences in attention; memory, including verbal and visual memory and short-term and working memory; visual spatial; and inhibitory control. Adjusting for multiple comparisons, attention, verbal memory, verbal working memory, visual spatial, and inhibitory control remained statistically significant. Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD
Figure 1. Difference in z scores between controls and individuals with chronic kidney disease (CKD) in 11 neurocognitive domains. Circles represent the difference in z scores, and bars represent 95% confidence intervals. Difference in z scores above zero indicates better performance among controls compared with individuals with CKD. Adjusted for age, race, sex, and maternal education. *Statistically significant difference in z scores between controls and individuals with CKD. All remained signficant after accounting for multiple comparisons except for visual memory.
In secondary analyses, we assessed the effect of level of kidney function on cognitive performance, adjusted for other demographic and clinical variables that may affect neurocognitive function. In this analysis, only individuals with complete ABPM data were included. In these models, eGFR was used as the main explanatory variable, adjusting for age, sex, race, maternal education, hemoglobin level, cholesterol level, phosphate level, and BP parameters, including diastolic load and nocturnal diastolic dipping. There was a trend toward better performance in
all domains with higher eGFRs, with statistically significant differences seen in the domains of attention, visual spatial, and visual working memory (Fig 2; Table 3). In addition to eGFR, metrics of BP control had significant effects in multiple domains (Table 3). Higher diastolic load was associated with poorer performance in language and verbal memory. Lower nocturnal diastolic dipping was associated with poorer performance in attention. Adjusted for multiple comparisons, these analyses did not achieve
Figure 2. Increase in z scores per 10-unit higher estimated glomerular filtration rate (eGFR). Circles represent the increase in z scores, and bars represent 95% confidence intervals. Models are adjusted for age, sex, race, maternal education, hemoglobin level, cholesterol level, phosphate level, diastolic load (square root transformed), and diastolic dipping. *Statistically significant difference in z scores per 10-unit higher eGFR. None remained significant after accounting for multiple comparisons. Am J Kidney Dis. 2016;67(4):567-575
571
Ruebner et al Table 3. Multivariable Linear Regression Analysis eGFRa
Domain
Attention Language Verbal memory Verbal working memory Visual memory Visual spatial Visual working memory Executive function Inhibitory control Problem solving Set shift
0.053 0.041 0.054 0.063 0.039 0.062 0.069 0.079 0.034 0.036 0.005
(0.008 to 0.098); 0.02d (20.013 to 0.096); 0.1 (20.048 to 0.156); 0.3 (20.030 to 0.156); 0.2 (20.033 to 0.110); 0.3 (0.008 to 0.115); 0.02d (0.003 to 0.136); 0.04d (20.009 to 0.167); 0.08 (20.005 to 0.073); 0.08 (20.001 to 0.074); 0.06 (20.054 to 0.064); 0.9
% Diastolic Loadb
20.003 20.076 20.166 20.056 20.069 20.055 20.024 0.003 0.001 20.046 0.058
(20.066 (20.152 (20.308 (20.186 (20.168 (20.130 (20.117 (20.121 (20.053 (20.098 (20.022
to to to to to to to to to to to
% Diastolic Dippingc
0.060); 0.9 20.001); 0.05d 20.023); 0.02d 0.074); 0.4 0.031); 0.2 0.020); 0.2 0.069); 0.6 0.127); 0.9 0.055); 0.9 0.006); 0.08 0.139); 0.2
0.017 0.007 20.005 0.003 20.003 0 0.005 0.003 0.012 20.007 20.013
(0.001 to 0.034); 0.04d (20.013 to 0.027); 0.5 (20.043 to 0.032); 0.8 (20.031 to 0.037); 0.9 (20.029 to 0.023); 0.8 (20.020 to 0.019); 0.9 (20.019 to 0.029); 0.7 (20.029 to 0.035); 0.8 (20.003 to 0.026); 0.1 (20.020 to 0.007); 0.3 (20.034 to 0.008); 0.2
Note: Values are given as b coefficient (95% confidence interval); P value. Analysis adjusted for age, race, maternal education, sex, hemoglobin level, cholesterol level, and phosphate level. Diastolic load variable was square root transformed. Abbreviation: eGFR, estimated glomerular filtration rate. a b coefficient indicates difference in domain z score per 10-unit higher eGFR. b b coefficient indicates difference in domain z score per 1% higher diastolic load. c b coefficient indicates difference in domain z score per 1% greater diastolic dip. d P ,0.05.
statistical significance but still tended to have nominally poorer performance with lower eGFRs.
DISCUSSION In this study of children, adolescents, and young adults with CKD, we found significantly lower performance in multiple areas of neurocognitive function among individuals with CKD compared with controls. Particular deficits were seen in an array of functions that capture attention, memory, and inhibitory control. After adjustment for other factors that may contribute to cognitive performance (including age, sex, race, maternal education, hypertension, anemia, hyperlipidemia, and hyperphosphatemia), we found lower neurocognitive performance in multiple domains with decreasing eGFRs. We also found that markers of BP control, including increased diastolic load and decreased nocturnal diastolic dipping, were associated with impairments in neurocognitive performance. This study characterizes the nature of cognitive dysfunction in children, adolescents, and young adults with CKD. In particular, dysfunction in areas related to attention and memory may have important implications for the delivery of health information to patients with CKD. Ancillary academic and cognitive support services may improve overall social functioning for children and young adults with kidney disease in academic and functional contexts.3 In addition, this study highlights important potentially modifiable cardiovascular disease risk factors for neurocognitive impairment, particularly hypertension. Hypertension is a well-described complication in CKD,16,24,25 and aggressive BP control has been associated with delayed progression to end-stage kidney disease.26 Our finding that higher diastolic load and decreased nocturnal dipping are independently associated with 572
poorer cognitive performance, even after controlling for socioeconomic and clinical factors, provides another important reason to screen for and treat hypertension in the CKD population because problems with cognition may represent another marker of targetorgan damage. Neurocognitive dysfunction is a well-described complication in patients with CKD. Numerous studies of adults with CKD have shown neurocognitive deficits in a range of areas, including learning, concentration, and attention.18,27 In a large cohort of 855 adults with CKD, lower eGFR was associated with lower scores in most cognitive domains, and individuals with advanced CKD (eGFR , 30 mL/min/1.73 m2) were more likely to have cognitive dysfunction compared with those with more mild disease.19 Among patients with kidney disease, stage of CKD, lower hemoglobin level, and increased parathyroid hormone level have been associated with cognitive dysfunction.8 Studies of adult dialysis patients have also shown that neurocognitive deficits are common, with abnormalities in areas of attention, processing speed, executive function, language, and memory.17,28 Patients who remain on dialysis therapy may have progressive declines in neurocognitive function, whereas kidney transplantation may lead to improved performance among patients previously on dialysis therapy.29,30 Among children with CKD, increased disease severity and longer duration of disease have been associated with neurocognitive deficits.1,31 In a study comparing 20 children with CKD with healthy controls, individuals with CKD were at risk for a lower IQ and deficits in executive functioning.32 In a study of 50 pediatric kidney transplant recipients, those with a kidney transplant had lower performance in verbal Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD
and visuospatial domains compared with controls.33 Our results are consistent with these findings. Hypertension has been associated with neurocognitive dysfunction in adults34-36 and children9,10,37 with and without CKD. In a community-based study of 337 adults without kidney disease, there was an inverse linear relationship between systolic BP and global cognitive performance, even within the normotensive range.34 In a population-based study of otherwise healthy school-aged children in the United States, children with systolic or diastolic BP above the 90th percentile for age, sex, and height had lower performance in areas of memory, attention, and concentration.37 Previous studies of children with CKD have not shown definitive relationships between hypertension and neurocognitive performance.1,31 In a large study of 383 children with mild to moderate CKD from the CKiD multicenter longitudinal study, high BP was associated with lower performance IQ score, even after adjustment for demographic and clinical variables that may affect neurocognitive function. However, high BP was not associated with measures of attention, verbal IQ, academic achievement, or parental ratings of executive function.10 Although previous studies have shown an association between hypertension and neurocognitive dysfunction, our study was unique in the use of ABPM. We found that abnormalities including increased diastolic load and decreased nocturnal diastolic dipping were associated with impairments in neurocognitive performance. Casual assessment of BP does not necessarily identify all patients with hypertension. In a study of 198 children from the CKiD cohort, 38% of children had masked hypertension, defined as normal casual BP but increased ambulatory BP. 24 Therefore, ABPM may provide an important tool in the assessment of children with CKD, not only to guide BP management, but also as a screen to assess risk for cognitive dysfunction. The mechanism of hypertension leading to neurocognitive deficits remains unclear. Possible causes include vascular remodeling, altered cerebrovascular reactivity, or direct neuronal effects,38-40 but there are few studies that directly explore the neurophysiologic impact of hypertension. In this study, we were not able to identify why elevated diastolic load and decreased nocturnal dipping are associated with cognitive deficits, although we postulate that decreased nocturnal dipping may suggest decreased cerebral blood flow autoregulation, which may be associated with decreased cerebral perfusion. Additional research is needed to understand the interplay of hypertension, anemia, and cerebral autoregulation and their contribution to neurocognitive function. Am J Kidney Dis. 2016;67(4):567-575
This study has several limitations. First, this was a cross-sectional study so we were unable to assess changes in neurocognitive function over time or alterations in neurocognition with changes in kidney function, dialysis, or transplantation status. There was a relatively small sample size in the CKD and control groups, potentially decreasing the ability to detect effects of variables such as eGFR, hypertension, and nocturnal dipping across all cognitive domains. In addition, this study included a relatively heterogeneous population of individuals with CKD, including a wide range of eGFRs, as well as both dialysis patients and transplant recipients. In secondary analyses, after adjusting for eGFR, we found a trend toward poorer performance in multiple domains among individuals who ever had a transplant, ever had dialysis, or required first renal replacement therapy before the age of 5 years. However, the number of participants in each of these subgroups was small, limiting our ability to make statistically significant conclusions about the effect of CKD severity on cognitive performance. When we adjusted for multiple comparisons, in the primary analysis, there was still a statistically significant difference in neurocognitive performance between individuals with CKD and controls in multiple domains. In secondary analyses assessing the effect of eGFR on neurocognitive performance, the effect of eGFR no longer reached statistical significance after adjusting for multiple comparisons. However, the nominal trend toward poorer neurocognitive performance with declining eGFR still generates important hypotheses that should be analyzed further in future larger studies. In addition, the control group consisted of unrelated individuals as well as a smaller proportion who were siblings of participants with decreased eGFR; the inclusion of siblings may present a different sample than individuals from the general population. Finally, we grouped individual neurocognitive tests into predetermined neurocognitive domains and selected the median test score for each domain as representative of that domain. There was overlap in individual tests across different domains. While this process is a conceptually viable approach to data reduction, it is unclear whether different results may have been obtained using a factor-based approach to data reduction, a strategy that would require a larger sample size, but would have eliminated potential redundancy across the targeted neurocognitive domains. In summary, we found significant neurocognitive differences among children and young adults with CKD compared with controls. Abnormalities in ambulatory BP may be an important risk factor for 573
Ruebner et al
cognitive dysfunction in CKD. The next steps of the NiCK Study will be to correlate neurocognitive test results with findings on structural and functional brain imaging to determine whether specific neurocognitive deficits correlate with changes in brain structure, connectivity, or blood flow. In addition, we will explore whether structural abnormalities are related to cardiovascular disease risk factors, including hypertension, anemia, dyslipidemia, and hyperphosphatemia. Examining these relationships will further the understanding of the biological underpinnings of neurocognitive abnormalities in CKD and may help determine targets for intervention to prevent cognitive dysfunction.
ACKNOWLEDGEMENTS Support: This project is funded in part under a Commonwealth Universal Research Enhancement grant with the Pennsylvania Department of Health (#SAP 4100054843). The Department specifically disclaims responsibility for any analyses, interpretations or conclusions. The funders did not have any role in study design, data collection, data analysis, writing of the report, or the decision to submit the report for publication. Financial Disclosure: The authors declare that they have no other relevant financial interests. Contributions: Research idea and study design: SRH, JR, SLF; data acquisition: NL, RLR; data analysis and interpretation: RLR, NL, JYK, EAH, SRH, JR, SLF; statistical analysis: JYK; mentorship: SLF. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved. RLR takes responsibility that this this study has been reported honestly, accurately, and transparently; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.
SUPPLEMENTARY MATERIAL Table S1: Neurocognitive battery and associated domains. Note: The supplementary material accompanying this article (http://dx.doi.org/10.1053/j.ajkd.2015.08.025) is available at www.ajkd.org
REFERENCES 1. Hooper SR, Gerson AC, Butler RW, et al. Neurocognitive functioning of children and adolescents with mild-tomoderate chronic kidney disease. Clin J Am Soc Nephrol. 2011;6(8):1824-1830. 2. Gipson DS, Wetherington CE, Duquette PJ, Hooper SR. The nervous system and chronic kidney disease in children. Pediatr Nephrol. 2004;19(8):832-839. 3. Gerson AC, Butler R, Moxey-Mims M, et al. Neurocognitive outcomes in children with chronic kidney disease: current findings and contemporary endeavors. Ment Retard Dev Disabil Res Rev. 2006;12(3):208-215. 4. Bugnicourt JM, Godefroy O, Chillon JM, Choukroun G, Massy ZA. Cognitive disorders and dementia in CKD: the neglected kidney-brain axis. J Am Soc Nephrol. 2013;24(3):353-363. 5. Duquette PJ, Hooper SR, Icard PF, et al. Neurodevelopmental status and adaptive behaviors in preschool children with chronic kidney disease. J Special Educ. 2009;43(1):45-51. 574
6. Brouhard BH, Donaldson LA, Lawry KW, et al. Cognitive functioning in children on dialysis and post-transplantation. Pediatr Transplant. 2000;4(4):261-267. 7. Drew DA, Weiner DE. Cognitive impairment in chronic kidney disease: keep vascular disease in mind. Kidney Int. 2014;85(3):505-507. 8. Kalaitzidis RG, Karasavvidou D, Tatsioni A, et al. Risk factors for cognitive dysfunction in CKD and hypertensive subjects. Int Urol Nephrol. 2013;45(6):1637-1646. 9. Kupferman JC, Lande MB, Adams HR, Pavlakis SG. Primary hypertension and neurocognitive and executive functioning in school-age children. Pediatr Nephrol. 2013;28(3):401-408. 10. Lande MB, Gerson AC, Hooper SR, et al. Casual blood pressure and neurocognitive function in children with chronic kidney disease: a report of the children with chronic kidney disease cohort study. Clin J Am Soc Nephrol. 2011;6(8):1831-1837. 11. Nash DT, Fillit H. Cardiovascular disease risk factors and cognitive impairment. Am J Cardiol. 2006;97(8):1262-1265. 12. Wong LJ, Kupferman JC, Prohovnik I, et al. Hypertension impairs vascular reactivity in the pediatric brain. Stroke. 2011;42(7):1834-1838. 13. Cachofeiro V, Goicochea M, de Vinuesa SG, Oubina P, Lahera V, Luno J. Oxidative stress and inflammation, a link between chronic kidney disease and cardiovascular disease. Kidney Int Suppl. 2008;111:S4-S9. 14. Menon V, Gul A, Sarnak MJ. Cardiovascular risk factors in chronic kidney disease. Kidney Int. 2005;68(4):1413-1418. 15. Mitsnefes MM. Cardiovascular disease in children with chronic kidney disease. J Am Soc Nephrol. 2012;23(4):578-585. 16. Wilson AC, Schneider MF, Cox C, et al. Prevalence and correlates of multiple cardiovascular risk factors in children with chronic kidney disease. Clin J Am Soc Nephrol. 2011;6:27592765. 17. Kalirao P, Pederson S, Foley RN, et al. Cognitive impairment in peritoneal dialysis patients. Am J Kidney Dis. 2011;57(4): 612-620. 18. Lee JJ, Chin HJ, Byun MS, et al. Impaired frontal executive function and predialytic chronic kidney disease. J Am Geriatr Soc. 2011;59(9):1628-1635. 19. Yaffe K, Ackerson L, Kurella Tamura M, et al. Chronic kidney disease and cognitive function in older adults: findings from the Chronic Renal Insufficiency Cohort cognitive study. J Am Geriatr Soc. 2010;58(2):338-345. 20. Schwartz GJ, Munoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol. 2009;20(3):629-637. 21. Levey AS, Greene T, Kusek JW, Beck GL, MDRD Study Group. A simplified equation to predict glomerular filtration rate from serum creatinine (abstract). J Am Soc Nephrol. 2000;11: 155A. 22. Wuhl E, Witte K, Soergel M, Mehls O, Schaefer F. Distribution of 24-h ambulatory blood pressure in children: normalized reference values and role of body dimensions. J Hypertens. 2002;20(10):1995-2007. 23. Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals: part 1: blood pressure measurement in humans: a statement for professionals from the Subcommittee of Professional and Public Education of the American Heart Association Council on High Blood Pressure Research. Hypertension. 2005;45(1): 142-161. 24. Mitsnefes M, Flynn J, Cohn S, et al. Masked hypertension associates with left ventricular hypertrophy in children with CKD. J Am Soc Nephrol. 2010;21(1):137-144. Am J Kidney Dis. 2016;67(4):567-575
Neurocognitive Dysfunction in CKD 25. Wong H, Mylrea K, Feber J, Drukker A, Filler G. Prevalence of complications in children with chronic kidney disease according to KDOQI. Kidney Int. 2006;70(3):585-590. 26. Wuhl E, Trivelli A, Picca S, et al. Strict blood-pressure control and progression of renal failure in children. N Engl J Med. 2009;361(17):1639-1650. 27. Hailpern SM, Melamed ML, Cohen HW, Hostetter TH. Moderate chronic kidney disease and cognitive function in adults 20 to 59 years of age: Third National Health and Nutrition Examination Survey (NHANES III). J Am Soc Nephrol. 2007;18(7): 2205-2213. 28. Post JB, Morin KG, Sano M, Jegede AB, Langhoff E, Spungen AM. Increased presence of cognitive impairment in hemodialysis patients in the absence of neurological events. Am J Nephrol. 2012;35(2):120-126. 29. Harciarek M, Biedunkiewicz B, LichodziejewskaNiemierko M, Debska-Slizien A, Rutkowski B. Continuous cognitive improvement 1 year following successful kidney transplant. Kidney Int. 2011;79(12):1353-1360. 30. Harciarek M, Williamson JB, Biedunkiewicz B, Lichodziejewska-Niemierko M, Debska-Slizien A, Rutkowski B. Risk factors for selective cognitive decline in dialyzed patients with end-stage renal disease: evidence from verbal fluency analysis. J Int Neuropsychol Soc. 2012;18(1):162-167. 31. Slickers J, Duquette P, Hooper S, Gipson D. Clinical predictors of neurocognitive deficits in children with chronic kidney disease. Pediatr Nephrol. 2007;22(4):565-572.
Am J Kidney Dis. 2016;67(4):567-575
32. Gipson DS, Hooper SR, Duquette PJ, et al. Memory and executive functions in pediatric chronic kidney disease. Child Neuropsychol. 2006;12(6):391-405. 33. Haavisto A, Korkman M, Holmberg C, Jalanko H, Qvist E. Neuropsychological profile of children with kidney transplants. Nephrol Dial Transplant. 2012;27(6):2594-2601. 34. Knecht S, Wersching H, Lohmann H, et al. High-normal blood pressure is associated with poor cognitive performance. Hypertension. 2008;51(3):663-668. 35. Launer LJ, Ross GW, Petrovitch H, et al. Midlife blood pressure and dementia: the Honolulu-Asia Aging Study. Neurobiol Aging. 2000;21(1):49-55. 36. Vicario A, Martinez CD, Baretto D, Diaz Casale A, Nicolosi L. Hypertension and cognitive decline: impact on executive function. J Clin Hypertens (Greenwich). 2005;7(10):598-604. 37. Lande MB, Kaczorowski JM, Auinger P, Schwartz GJ, Weitzman M. Elevated blood pressure and decreased cognitive function among school-age children and adolescents in the United States. J Pediatr. 2003;143(6):720-724. 38. Ostrovskaya MA, Rojas M, Kupferman JC, et al. Executive function and cerebrovascular reactivity in pediatric hypertension. J Child Neurol. 2015;30(5):543-546. 39. Jennings JR, Zanstra Y. Is the brain the essential in hypertension? Neuroimage. 2009;47(3):914-921. 40. Jennings JR. Autoregulation of blood pressure and thought: preliminary results of an application of brain imaging to psychosomatic medicine. Psychosom Med. 2003;65(3):384-395.
575