High Failure Rate with Brigance Developmental Screening in 3 Year-Old Children with Sickle Cell Disease

Background: Central nervous system complications of sickle cell disease (SCD) include stroke, silent cerebral infarct, and neurocognitive deficits, but few studies have examined developmental delays in preschool-age children. Using the Brigance developmental screen in 3 year-old children with SCD we found a high frequency of scores below the “normal” cutoff for age, but control data were lacking (Ped. Blood Cancer 2011;56:620). We hypothesized that children with SCD would be more likely to “fail” this screening test than age-matched children from a similar ethnic/socio-economic background.

Methods: This prospective study was approved by the St. Jude Children’s Research Hospital (SJCRH) IRB and informed consent obtained for each subject. 3.0-3.9 year-old SCD subjects were tested in the SJCRH Sickle Cell Clinic. Patients receiving chronic transfusion therapy were excluded. Age-matched and race-matched control subjects with similar socio-economic backgrounds were recruited from 8 Memphis daycare centers. The Brigance Preschool Screen was administered by trained examiners in the clinic or the daycare and required about 15 minutes to complete; the primary caretaker simultaneously completed the demographic form. None of the participants had previous exposure to the test. The proportions of subjects who failed the Brigance test (scoring below the age-related “cutoff”) and test raw scores (higher scores indicating better performance) were compared between SCD and control subjects using the exact Chi-square test and the two-sample t-test, respectively; logistic regression was used to model the associations between the Brigance failure proportions and other covariates for the groups. P-values <0.05 were considered significant.

Results: Testing was performed on 103 SCD subjects not on treatment, an additional 20 HbSS subjects receiving hydroxyurea (HU), and 109 daycare subjects. Seventy-four percent of children with SCD (not on HU) failed the Brigance screen, compared to 61% of controls (p = 0.04). Mean (SD) Brigance raw scores for the non-HU SCD patients and controls were 45.9 (20.8) and 56.1 (22.5), respectively (p < 0.001). In the non-HU SCD group the pass/fail rate was not significantly different in those with more severe (SS, Sβ⁰thal) and milder (SC, Sβ⁺thal) genotypes. Medical factors, including absolute neutrophil count, hemoglobin concentration, %HbF level and O₂saturation by pulse oximetry, were not associated with Brigance results. In contrast, socio-economic/demographic factors had a significant effect on the scores: lower income level, more household children under age 18, and more household residents were associated with increased failure (p </= 0.03). Lower caretaker education level showed a similar trend (p = 0.065). None of the socio-economic/demographic factors were associated with failure in the control group, although there was a trend toward failure with more household children under age 18 (p = 0.07). HU-treated SCD children were not significantly different from untreated patients in the failure rate (p = 0.4) and had no significant associations with medical or socio-economic factors.

Conclusions: Among 3-year-olds tested with the Brigance Developmental Screen, a higher proportion of children with SCD failed compared to matched controls, but both groups showed a surprisingly high rate. Medical factors, such as hemoglobin level and sickle cell genotype, were not associated with the failure rate, nor was treatment with HU in a limited number of subjects. Socio-demographic factors (income level and numbers of household members) were associated with the failure rate in the SCD group, but not in the control group. These data indicate that preschool children with SCD are at very high risk for developmental delay, and underlying environmental factors seem to contribute more than medical factors. More detailed neurocognitive evaluation of preschool children with SCD and effective ways to enhance their development are needed.