Abstract 4839

Background:

As children with sickle cell disease (SCD) are at significant risk for neurocognitive complications, an automated and objective measure of neurocognitive functioning would address several challenges facing both clinical and research progress in SCD including longitudinal monitoring of deficits, cross-site comparability of neurocognitive tests in multicenter trials, and limited access to pediatric neuropsychologists. The Cambridge Automated Neuropsychological Testing Automated Battery (CANTAB) is a well-validated computerized test with significant normative data in individuals age 4 to 80 that has been used to monitor disease progression and treatment response in children and adults with a range of disorders but has not been used previously in SCD.

Hypothesis:

We hypothesize that the CANTAB system is a useful and viable tool for the neurocognitive evaluation of pediatric patients with SCD. We expect that the CANTAB testing will be well tolerated by SCD patients and parents, easy to administer in our comprehensive clinic, and generate valid results that correlate with both medical and psychological outcomes.

Methods:

7 CANTAB tests which assess attention, executive function and memory were run on pediatric SCD patients during scheduled clinic visits. Parents completed the child behavior checklist (CBCL) which generates t-scores for children on internalizing, externalizing scales as well as DSM-oriented scales of affective, anxiety, pervasive developmental, attention and oppositional scales. Medical data including SCD genotype, average hemoglobin (hgb), hematocrit (hct), reticulocyte count (rct), lactate dehydrogenase and hospital utilization records (ER visits, # hospital visits in the last year) was collected. Demographic information and a total pain burden assessment were also collected.

Results:

11 children with HbSS SCD were enrolled in the pilot study (table 1). All patients successfully completed the CANTAB testing without difficulty. Hgb and rct were associated with strategy score on spatial working memory and the latency score on the motor screening task. Hgb and rct also correlated with internalizing, externalizing, and total symptoms scores on the CBCL (table 2). Specifically lower hgb and higher rct were associated with increased CBCL scores. A regression model incorporating average hgb and total internalizing scores with spatial working memory as the dependent variable revealed a significant interaction between internalizing scores and hgb and a significant model p=.01 and r2 of 0.89 offering preliminary support for a multi-level model incorporating disease and child specific factors (table 2).

Table 1
N=11Age (sd)HgbHctRctER visitsPainCBCL totalSWM strategySOC mean moves
 8.5(4.0) 8.8 (.59) 26.3(2.2) 9.7(5.9) 3.4(4.6) 6(3.6) 50.7 (12.5) 19.5 (1.5) 2.46 (.57) 
N=11Age (sd)HgbHctRctER visitsPainCBCL totalSWM strategySOC mean moves
 8.5(4.0) 8.8 (.59) 26.3(2.2) 9.7(5.9) 3.4(4.6) 6(3.6) 50.7 (12.5) 19.5 (1.5) 2.46 (.57) 
Table 2
Lab valueInternalizing-CBCLExternalizing CBCLTotal CBCLSwm strategyMot latency
Hgb P=.01 P=.02 P=.0019 P=.01 P=.12 
Hct P=.01 P=.03 P=.002 P=.01 P=.06 
Rct P=.06 P=.08 P=.07 P=.09 P=.04 
Lab valueInternalizing-CBCLExternalizing CBCLTotal CBCLSwm strategyMot latency
Hgb P=.01 P=.02 P=.0019 P=.01 P=.12 
Hct P=.01 P=.03 P=.002 P=.01 P=.06 
Rct P=.06 P=.08 P=.07 P=.09 P=.04 

The total pain burden score correlated error making in several tests including the delayed match to sample test (p=.01), spatial working memory test (p=.06), and Stockings of Cambridge task (p=.0038). The pain burden score was not associated with performance or latency on these tests indicating that pain burden may have a specific association with error making. Pain burden also correlated with the somatic measure on the CBCL (p=.01) indicating cross validation between the two measures.

Conclusion:

This pilot study demonstrates the feasibility and value of the CANTAB system in evaluating neurocognitive deficits in pediatric patients with SCD. These results can be assessed longitudinally following medical interventions. Furthermore, results indicate a multi-level model that includes medical factors, child specific factors, and demographics may be a more appropriate model to utilize in determining the etiology of neurocognitive deficits in SCD. Ongoing studies with an increased sample size will examine the association of neurocognitive function with SCD genotype, MRI, transcranial doppler studies, and family stress.

SWM: Spatial working memory SOC: Stockings of Cambridge MOT: Motor Screening Test

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution