OEF Measured with Diffuse Optical Spectroscopies As a Biomarker of Silent Stroke in Pediatric Sickle Cell Anemia

Abstract: Sickle cell anemia is associated with a high risk of silent cerebral infarct. In children, studies with MRI have demonstrated that regional oxygen extraction fraction (OEF) is elevated in patients with silent infarction compared to those without infarction. Here we use a non-invasive, low-cost optical technique called frequency domain near-infrared spectroscopy (FDNIRS) to demonstrate similar increases in OEF with silent infarct. These results suggest a promising role for FDNIRS in providing a biomarker of silent infarct that merits further neuroimaging in children with sickle cell anemia.

Introduction: Sickle cell anemia is an inherited blood disorder that has a profound effect on the brain, including a high risk of cerebral injury. The most common neurological insult is silent cerebral infarct (SCI), which is associated with progressive cognitive decline resulting in lower IQ, poor school performance, etc. (DeBaun et al., 2020; Schatz et al., 2001). Assessment of these clinically silent injuries requires MRI, which is costly and has limited availability. Thus, current guidelines recommend a single MRI to check for SCI in early-school-age children, along with another MRI in adulthood, resulting in a retrospective, symptom-driven model of monitoring for neurological injury. To move towards a prospective, preventative model of monitoring for neurological injury in these patients, there is a clinical need for a lower cost, point-of-care biomarker of SCI that could guide neuroimaging timing and aid in individualized clinical management of these patients. Recent work with MRI has demonstrated that elevated oxygen extraction fraction (OEF) is a promising biomarker of SCI (Fields et al., 2018). Novel, low-cost optical tools can non-invasively assess OEF at the bedside. These tools, called frequency domain near infrared spectroscopy and diffuse correlation spectroscopy (FDNIRS/DCS), can quantify regional OEF in the frontal cortex as well as other hemodynamic parameters including cerebral blood flow (CBF) and cerebral blood volume (CBV) (Lee et al., 2019, Lee & Brothers et al., 2022, Brothers et al., 2024). Here we test the hypothesis that FDNIRS/DCS are sensitive to alterations in cerebral hemodynamics associated with presence of SCI in pediatric sickle cell patients. Specifically, we hypothesize that OEF measured with FDNIRS/DCS will be elevated with SCI.

Methods: Children ages 5 to 18 y with sickle cell anemia (HbSS or HbSβ⁰-thalassemia) were enrolled at Children’s Healthcare of Atlanta. All participants had a 3T head MRI with 3D FLAIR to assess SCI, followed by optical assessment with FDNIRS/DCS (within 1 day of MRI). Blood hemoglobin concentration was assessed with complete blood count on a venous sample within 1 week of FDNIRS/DCS measurement. Details of FDNIRS/DCS acquisition/analysis is described in (Lee 2022). Silent infarction was determined independently by two blinded, board-certified neuroradiologists based on criteria outlined in (Casella et al., 2010). A Wilcoxon rank-sum test was used to test whether each FDNIRS/DCS measure (OEF, CBF, CBV) was different between patients with and without evidence of SCI.

Results: Thirty subjects were enrolled, with 24 of the 30 enrolled meeting study criteria. Subjects were mostly female (68%), ranging in age from 8 to 18 y, with a median (IQR) hemoglobin of 8.2 (7.4, 8.8) g/dL. Twenty out of 24 subjects (83%) were on hydroxyurea, 3/24 (13%) were on voxelotor, and 1/24 (5%) was on chronic exchange transfusion. As expected, OEF was significantly higher in patients with SCI compared to those without (median (IQR) 56 (54, 58) vs. 47 (44, 55) %, p=0.013). No differences in CBF (p=0.97), CBV (p=0.85), blood hemoglobin (p=0.91), age (p=0.97), hydroxyurea usage (p=0.88), or voxelotor usage (p=0.93) were observed.

Conclusion: Recruitment for this study is ongoing. These preliminary results suggest that FDNIRS/DCS may be sensitive to regional elevations in OEF in patients with silent infarct, indicating a potential role for FDNIRS/DCS in point of care neuromonitoring of these patients.