Cerebral Metabolic Stress Is Increased in Children with Sickle Cell Anemia Compared to Anemic Controls

Children with sickle cell anemia (SCA) experience a host of neurologic complications, including cognitive dysfunction, silent and overt stroke. Oxygen extraction fraction (OEF), the percent of oxygen removed from the blood into the brain tissue, can increase to compensate for decreased oxygen delivery and maintain the cerebral metabolic rate of oxygen utilization (CMRO₂), as stroke occurs with a decrease in CMRO₂. Elevated OEF is an independent predictor for ischemic injury in adults without SCA. We have previously shown that OEF is increased in children with SCA compared to healthy controls. Furthermore, the region of the brain with greatest elevation in OEF in SCA co-localizes with the regions at greatest risk for stroke, and OEF indirectly correlates with hemoglobin. However, it remains unclear if OEF is only elevated due to decreased arterial oxygen content with severe anemia in SCA, or if other factors contribute to this ongoing cerebral metabolic stress. We prospectively obtained brain MRIs in children with SCA, children with non-sickle cell anemia (ACTL) and healthy controls (CTL) to test our hypothesis that all children with anemia would have elevated OEF compared to controls, but those with SCA would have more elevated OEF than children with non-sickle cell anemia.

Brain MRI measured voxel-wise OEF with an asymmetric spin echo sequence. OEF processing is described in Fields et al. Blood. 2019; 133(22):2436-2444. Group comparisons were made with a Mann-WhitneyUor chi-square tests. General linear models were used to understand the difference in whole brain (WB), gray matter (GM) and white matter (WM) OEF between cohorts while controlling for hemoglobin.

Table 1 describes the 78 participants (20 CTL, 17 ACTL, 41 SCA). The CTL cohort includes 11 HbAA and 9 HbAS participants; the ACTL cohort includes 4 hereditary spherocytosis, 1 congenital dyserythropoietic anemia, 1 beta thalassemia intermedia, 6 iron deficiency, 4 aplastic anemia and 1 loxoscelism participant; the SCA cohort includes 36 HbSS and 5 HbS beta thalassemia null participants. Hemoglobin was lower in the ACTL (p < 0.001) and SCA (p < 0.001) cohorts compared to CTL, but was not different between the ACTL and SCA cohorts (p = 0.682). Ninety percent of the SCA cohort was receiving hydroxyurea at a median dose of 29.1 [21.5-33.1] mg/kg/day, and none were receiving chronic transfusions. Thirty-four percent of the SCA cohort had a history of silent cerebral infarct, while none had a history of overt stroke or vasculopathy identified on screening MRA.

OEF was significantly elevated in the ACTL cohort compared to the CTL cohort, but significantly lower than the SCA cohort in the WB, GM and WM (Table 1, Figure 1). Final general linear models predicting WB, GM and WM OEF used a quadratic fit for hemoglobin, and included hemoglobin, hemoglobin², and cohort as covariates. After accounting for hemoglobin and hemoglobin²(indicating that the relationship between hemoglobin and OEF is not linear), OEF is significantly higher in the SCA cohort compared to the ACTL cohort in WB, GM and WM, while there was not a significant difference in OEF between the ACTL and CTL cohorts (Table 2, Figure 2). The volume of WM with elevated OEF was significantly higher in the ACTL cohort compared to the CTL cohort (OEF 35%: p < 0.001, OEF 37%: p < 0.001, OEF 39%: p < 0.001), but lower in ACTL cohort compared to the SCA cohort (OEF 35%: p = 0.032, OEF 37%: p = 0.028, OEF 39%: p = 0.029).

We conclude that OEF is significantly higher in children with SCA compared to age- and sex-matched children that are equally anemic for reasons other than SCA. While severity of anemia was a significant predictor of OEF, cohort (SCA vs. ACTL) remained an independent predictor of OEF across the WB, GM and WM while accounting for hemoglobin. These data suggest that further investigation is required to understand additional covariates contributing to increased cerebral metabolic stress, as measured by OEF, in children with SCA, as these may be clinical targets to reduce OEF and provide further neuroprotection in this vulnerable population.