Introduction 1 2 3 4 5 6 7 7 6 4 One thing to note, however, is that most studies to date have focused primarily on children with ESRD, leaving unanswered whether children with mild to moderate CKD also suffer this same vulnerability. Cognitive deficiencies in children with CKD could arise through a gradual process, proportional to the level of kidney dysfunction, or may develop once a filtration threshold has been passed. It seems logical that the severity of CKD might be proportional to the degree of cognitive impairment, but the data are not yet available to support this. The stage of neurologic development at the time of disease onset or the cumulative time children spend with CKD may also impact on the degree or type of cognitive impairment experienced. To our knowledge, no published study has yet focused specifically on isolating risk factors to identify which children with CKD are most at risk for cognitive decline. The primary goal of this study was to explore associations between clinical aspects describing CKD and selected neurocognitive test scores in a sample of children and adolescents with CKD. In addition, this study should provide the foundation for identifying specific clinical risk factors for subsequent neurodevelopmental dysfunction. Methods Participants 8 2 2 1 Table 1 PD HD Study sample characteristics n a Range Age (years) 12.5 (3.2) 7–19 Male (%) 52  Caucasian (%) 52  2 32 (29) 4–89 ESRD (13 PD, 1 HD) (%) 48  Age at disease onset (years) 4.4 (5.9) 0–16 Duration of CKD (years) 6.7 (4.7) 0.2–15 Percent of life with CKD 69 (39) 1–100 Hypertensive (%) 59  Hemoglobin (mg/dl) 12.7 (1.5) 10–17 a Neurocognitive instruments 9 10 11 9 10 11 Clinical measures Disease severity, age of onset, and duration of renal disease, coded as time in years and as percent of life with CKD, as well as the presence of anemia or hypertension were explored as potential risk factors of interest, given their association with detrimental cognitive outcomes in the CKD literature. Clinical data were collected through physical examinations and laboratory tests performed within 2 weeks of cognitive testing. Age of disease onset, as well as duration and percent of life with CKD, were determined through record reviews and were based on the date of primary disease diagnosis. 8 12 2 12 2 13 14 Data analysis t t Results An initial inspection of the neurocognitive test scores showed that the sample of children and adolescents with CKD was mildly variable, with most of the participants falling within the low average to average range on most tasks. For the full study sample, the mean IQ score was 91 (range 61 to 117, SD 16), and the mean memory score was 88 (range 65 to 128, SD 16), both falling within the low average to average range when compared with normative expectations. The mean attention score of 96 fell well within the average range but did show much more variability (range 37 to 120, SD 23). Disease severity r P r P 1 2 P P P 3 r P Fig. 1 Relationship between IQ and estimated renal function  Fig. 2 Relationship between memory and estimated renal function  Fig. 3 Mean cognitive scores by level of renal disease  Age at disease onset and CKD duration r P n 2 r P r P Hypertension and anemia 2 Table 2 Relationships between hypertension, anemia and cognitive measures n a a a Total cohort 91 (16) 88 (16) 96 (23) Non-hypertensive 90 (18) 87 (16) 89 (24) Hypertensive 92 (15) 89 (17) 100 (21) Non-anemic 91 (16) 89 (16) 98 (22) Anemic (<11 mg/dl) 92 (17) 81 (23) 76 (20) a t r P P We should also acknowledge that there is a high degree of correlation among the results of the individual cognitive domains tested. Among the children included in our study, 12 (41%) had no cognitive scores greater than one standard deviation below the normative mean of 100, eight (28%) deviated in this fashion in a single cognitive domain, three (10%) did so in two areas, and six (21%) did so in all three areas. The level of overlap is, in part, likely due to the nature of cognitive testing, whereby testing in one area is, to some degree, dependent upon function in another area. The fact that there is not complete overlap, however, indicates that cognitive testing can distinguish among these deficits with some level of specificity. Discussion Neurocognitive deficits among children with ESRD have been demonstrated in the literature, with some suggestion that key clinical variables can be associated with the level and, perhaps, the pattern of these deficits. The primary purpose of this study was to examine the relationship between targeted clinical variables, based on the available literature, and selected neurocognitive functions as defined by IQ, memory, and attention. Further, the current study examines these relationships across the full spectrum of CKD, including the mild to moderate stages of CKD. While it is recognized that some of the etiologies leading to CKD in our patients, namely cystinosis, calcineurin inhibitor toxicity, and ischemic injury, could potentially be independently associated with some level of cognitive dysfunction, in all three cases the children did not demonstrate evidence of other neurologic signs that would indicate central nervous system damage from their underlying disorders. Additionally, these children are representative of a typical pediatric CKD population, and any contribution from the presence of their comorbid conditions would not compromise the validity of our findings. 3 15 15 3 1 2 3 The literature has been more ambiguous with regard to attention, and our findings did not demonstrate a significant deficit in attention. Our sample did, however, demonstrate significantly more variation in the attention measure among children with CKD than one would anticipate in the general population, as exhibited by the large standard deviation in our study sample. This suggests the possibility that there may be a subset of children with CKD who are more vulnerable to attention-related problems. More extensive exploration of the various components of attention may be required to explore this observation further. 16 16 17 18 17 18 This is the first study to date that focuses specifically on identifying potential risk factors for neurocognitive decline in children and adolescents with CKD. Its value is further enhanced by the inclusion of children with mild to moderate CKD, thus extending findings to a wider range of pediatric CKD patients. Increased disease severity, duration of CKD, and younger age at CKD onset were identified as potential risk factors for the targeted neuropsychological functions of IQ and memory. Our results further suggest a significant linear relationship with disease severity, with IQ scores continuously declining as disease severity worsens. Not only does this finding argue against a threshold effect, it implicates disease severity as an important risk factor for neuropsychological dysfunction. These findings also suggest the possible utility of these variables in a cumulative risk model to predict neurocognitive dysfunction in CKD and lay the foundation for exploration of such a model. Finally, these findings emphasize the importance of conducting further research in neurocognitive development among children with CKD. This includes research both to more clearly predict which children with CKD are most vulnerable to neuropsychological dysfunction and to identify specific correlates of this dysfunction. While there have been great strides in improving quality and length of life among children with CKD, a better understanding of those processes that interfere with normal neurodevelopment is a critical element of providing optimal care to these children.