Evaluating the efficacy of brief neurocognitive assessments in adults and adolescents with sickle cell disease

Abstract

Abstract Introduction: Sickle cell disease (SCD) is a common, life-threatening inherited hemoglobin disorder characterized by both acute and chronic complications. Neurologic complications are common among individuals with SCD, ranging from overt stroke to subtle neurocognitive impairments. Screening for these complications includes transcranial doppler (TCD) and brain MRI studies to assess for overt stroke risk or “silent” infarctions, but these studies are often deceptively normal. In addition to gross neurologic deficits due to overt stroke, individuals with SCD are at increased risk for deficits from smaller hypoxic episodes in sensitive brain tissue. These impairments have lifelong implications affecting both academic and vocational performance. Screening is difficult to access due to limited neurocognitive specialists, long waitlists for appointments, and the time-consuming nature of formal assessment and interpretation. This study prospectively evaluates the feasibility and utility of three streamlined neurocognitive screening tools administered in a comprehensive sickle cell clinic. The primary objective is to identify an evidence-based strategy for effective universal screening for adolescents and adults with SCD to enable timely referral for comprehensive neurocognitive evaluation and to inform educational and vocational directions. Methods: Participants were recruited during routine study visits or scheduled blood transfusions at the Brown University Health Comprehensive Sickle Cell Clinic. Clinical, laboratory and demographic information was extracted from the medical record. Exclusion criteria includes a history of overt stroke or other known neurological disorder that could interfere with test completion. Upon providing informed consent, participants complete three neurocognitive assessment tools: 1) the NIH Toolbox, a tablet-based tool (~30 minutes) assessing executive function, language, processing speed, episodic memory, working memory and attention; 2) the Montreal Cognitive Assessment (MoCA), a 30-point paper-based tool (~10 minutes) examining short term memory, visuospatial abilities, executive functions, attention, working memory, language, and orientation; 3) the Rowland Universal Dementia Assessment Scale (RUDAS), a 6-item paper-based tool (~5 minutes) covering memory, praxis, drawing, judgement, and language. We hypothesize these tools will capture more patients than traditional referral patterns and uncover previously unrecognized neurocognitive challenges. Phase 2 of the study will also include matched controls to compare the extent of cognitive challenges in adults with SCD. Statistical analysis will evaluate the feasibility and effectiveness of in-clinic screenings, as well as comparison of results from each tool. We will also investigate associations between neurocognitive performance and factors such as age, genotype, lab values (ie hemoglobin, mean corpuscular volume, etc), disease severity, and medications. Tool practicality will be evaluated by participant feedback on a 1-5 Likert Scale (Very Difficult to Very Easy). Results: To date, a total of 25 participants were enrolled (mean age 32.8 ± 12.2 years; range 18-68). Genotypes included HbSS (17), HbSC (2), HbSβ+ (2), HbS-HPFH (3), and HbSD (1). Average completion time varied significantly: NIH Toolbox 37.3 ± 5.3 minutes, MoCA 11.7 ± 3.5, RUDAS 5.6 ± 2.3). The NIH Toolbox defines mild impairment as an age-corrected t-score one SD=15 from the x̄=100 and the MoCA and RUDAS define it as a total score ≤ 26/30 and 22/30 respectively. Impairments were identified in 55% of subjects via the NIH Toolbox (x̄=83.9, SD=15.3), 60% via MoCA (x̄=24.3/30, SD=4.2) and 8% via RUDAS (x̄=27.3/30, SD=2.8). One participant was flagged by all three tools; four showed no impairment on any. NIH Toolbox and MoCA results agreed in 11 cases and diverged in 9. Patient feedback rated the NIH Toolbox as the most difficult (6 “Hard”, 10 “Moderate”, 4 “Easy”) but also the most useful for clinical settings (55%).Conclusion: Clinic-based neurocognitive screening is feasible and can take as little as 10 minutes to conduct. Initial data suggests the NIH Toolbox, while more challenging, may be the most effective for identifying deficits and was the patients' preferred performance metric. Continued enrollment will support a more robust comparison and evaluation of these three tools, including comparison of performance of SCD patients compared to matched controls.