Cerebral Blood Flow and Metabolic Correlates of Near Infrared Spectroscopy in Patients with Sickle Cell Disease

Introduction Tissue oxygen index (TOI), by near Infrared Spectroscopy (NIRS), is a valuable tool for noninvasive, indirect measurement of oxygen supply-demand balance. Cerebral TOI is decreased in sickle cell disease (SCD), and correlates with disease severity. Previous work suggests that cerebral TOI is inversely correlated with hemoglobin S level and chronic transfusion therapy restores TOI to normal values. Nahavandi et al. have proposed that low cerebral TOI in SCD disease can be attributed to impaired oxygen delivery and/or carrying capacity of sickle blood. Unfortunately, the specificity of cerebral TOI is still an area of active debate. In order to elucidate these mechanisms we measured global cerebral blood flow (CBF),arterial oxygen content (C_aO₂), oxygen delivery (DO₂), arterial and venous oxygen saturation (S_aO₂ and S_vO₂) and oxygen extraction fraction (OEF) and cerebral metabolic rate of oxygen (CMRO₂) in 12 healthy controls and 15 SCD patients using magnetic resonance imaging (MRI).

Methods All patients were recruited with informed consent or assent and this study was approved by the CHLA IRB. Fifteen patients with SCD and 12 healthy ethnicity matched controls (CTL) were studied. MRI compatible NIRS probes were placed on the forehead and TOI was recorded throughout the entire MRI examination. S_aO₂ was measured via peripheral pulse oximetery. Phase Contrast (PC) of the carotid and vertebral arteries was used to measure global CBF. T2 Relaxation Under Spin Tagging (TRUST) was used to measured T2 relaxation of blood within the sagittal sinus. T2 relaxation was converted to S_vO₂ via calibration curves. Blood draw for hemoglobin and electrophoresis was performed. Exclusion criteria included pregnancy, previous stroke, acute chest or pain crisis hospitalization within one month.

Results Table 1 summarizes the relationship between cerebral TOI, age, laboratory values, and hemodynamic variables. Surprisingly, TOI was independent of indices of oxygen supply (S_aO_2, CBF, oxygen delivery) and oxygen demand (CMRO₂); cerebral venous saturation and OEF were the only hemodynamic correlate of TOI. Total hemoglobin and percent sickle hemoglobin were equally and independently correlated with TOI with a combined r² of 0.59 on multivariate regression (p<0.0001).

Discussion This represents the first study comparing TOI to direct measurements of cerebral oxygen supply and consumption in SCD patients. We demonstrate that TOI tracks S_vO₂ and OEF, suggesting that it is weighted toward venous vascular beds. The relationship of TOI and HbS% has been previously described and could either shifting of the oxygen dissociation curve or mechanical disruption of microvascular integrity. TOI's strong dependence on total hemoglobin (after correction for HbS%) is particularly startling given its independence with oxygen delivery, suggesting that total hemoglobin is acting as a surrogate marker of microvascular disease severity in SCD patients.