A Systematic Literature Review of the Burden of Central Nervous System Complications for Patients with Sickle Cell Disease

INTRODUCTION: Sickle cell disease (SCD), a complex genetic blood disorder involving multicellular interactions between blood and endothelial cells, is often accompanied by central nervous system (CNS) complications. Effects range from silent cerebral infarct (SCI) to abnormal blood flow, and consequent overt stroke. This study assessed the humanistic and economic burden associated with CNS complications in patients with SCD and identified patient-reported outcome (PRO) instruments for future research.

METHODS: MEDLINE, Embase, Cochrane CENTRAL/CDSR and 11 congresses were searched to identify English language studies published from January 2000 to May 2020 and screened with predefined criteria by two independent researchers. Clinical trials (CT) or observational studies assessing humanistic burden, economic burden, or instruments used to measure burden in patients with SCD and CNS complications (N≥15) were included. Humanistic burden was broadly defined to include quality of life (QoL), symptoms and function.

RESULTS: Of the 3194 articles identified, 34 were included. Study designs were 29% retrospective observational (10/34), 29% cross-sectional (10), 26% prospective (9), 9% randomized CT (3), and 6% systematic reviews (2). Study size varied widely (16-4,485 patients with SCD and CNS complications). 77% (26) focused on pediatric patients. Separately, 77% (26) were in a US setting. Overt stroke (12), stroke and SCI (11), or SCI only (7) were the most frequently described CNS complications.

Twenty-five studies reported on humanistic burden, 18 of which measured cognitive function using the Wechsler Intelligence Scales. A significant decrease was reported in full scale (FS) (stroke vs. no stroke: 73.5 vs. 84.7; P=.04), verbal (abnormal vs. normal MRI: 74.1 vs. 84.6; P=.02), and performance IQ (stroke vs. no stroke: 69.5 vs. 81.5; P=.02) for patients with overt strokes or SCI compared to non-stroke SCD controls. Greater impairment was reported for overt stroke compared with SCI (multivariate meta-analysis of mean IQ difference: -10.3; P=.0013). In addition to stroke/SCI, socio-environmental factors (i.e., family income level, lack of college education) were significantly associated with a decrease in IQ (P=.005 and P=.023, respectively). Five studies assessed motor function, reporting significantly impaired function for patients with stroke compared to non-stroke SCD controls (Purdue Pegboard both hands: 7.5 vs. 10.1; P=.0001). Among children with SCD who had experienced their first stroke, those receiving hydroxyurea (HU) for prevention of recurrent stroke had significantly less moderate to severe motor disability (physician assessed) than children not receiving HU (23.1 vs. 88.9%; P<.001).

Nine studies reported economic outcomes related to healthcare resource use (HCRU) (8) and direct costs (5). HCRU and cost to manage CNS complications are high, with a median LOS of 5 days (IQR 3-9) and median hospitalization charges of $18,956 (2012 USD) for patients with SCD and stroke. Treatment costs were also substantial in patients with SCI, with annual transfusion plus chelation costs ranging between $18,149 and $67,361/year (2016 USD). No studies reported indirect costs.

While many studies used clinician-administered performance outcome (PerfO) measures such as the Wechsler Scales, only 1 study employed a PRO instrument: the Children's Depression Index. In addition to the Wechsler Scales, PerfO measures of cognitive function included the Woodcock-Johnson Psychoeducational Battery (6 studies), Children's Memory Scale (4), and California Verbal Learning Test for Children (4).

CONCLUSIONS: This systematic review found that SCD patients with CNS complications often experience diminished cognitive/motor function and incur substantial costs. In particular, the mean FSIQ for SCD patients with stroke and SCI was reported to be 'extremely low' to 'low' (65.9 to 83.6) and 'low' to 'average' (77.2 to 95.9), respectively across studies. The Wechsler Scales were the most commonly used PerfO measure of cognitive function, while use of disease-specific PRO instruments was extremely limited. Given that the majority of published data focuses on children, future research is needed to evaluate the burden of disease for SCD adults with CNS complications, and to assess patient perspectives and QoL to better quantify aspects of burden beyond cognitive performance.