Assessment of Functional Shunting in Patients with Sickle Cell Disease

Introduction

Silent cerebral infarcts (SCI) are common in patients with sickle cell disease (SCD) and thought to be caused by a mismatch between oxygen delivery and consumption. The mechanism underlying insufficient oxygen utilization is related to severity of anemia, and paradoxically, to the elevated cerebral blood flow (CBF) observed in SCD patients. CBF is elevated as a compensatory mechanism to maintain oxygen delivery, but high CBF levels can result in rapid transit of blood through the brain capillaries, limiting offloading of oxygen to the tissue; a process called arteriovenous shunting. One way to assess functional arteriovenous shunting is to use noncontrast perfusion MRI techniques in which we can assess the signal intensity of an endogenous blood tracer when it reaches the sagittal sinus. This venous signal (VS) reflects the amount of labeled blood that has not exchanged with the brain parenchyma. Under normal physiological conditions, the VS intensity will increase approximately proportionally with CBF as we expect only some of the water to exchange with tissue as it flows by. However, it is unknown whether functional shunting scales with CBF only, or whether other hemodynamic processes play a role in patients with SCD. We hypothesize that, under pathophysiological conditions such as in SCD patients, more labeled blood may pass unexchanged through the capillaries, which results in higher VS. In the present study, we investigated functional shunting by quantifying VS and assessed its association with hemodynamic, demographic and laboratory parameters in both pediatric and adult SCD patients, and controls. In addition, VS-CBF relationship was studied by further increasing CBF after a vasodilatory challenge.

Methods

We included 28 children (mean age 12.7 ± 2.3, 9 F) and 38 adults (mean age 32.1 ± 11.2, 14 F) with SCD (HbSS and HbS), and 10 healthy race-matched adult controls (mean age 36.4 ± 15.9, 4 F). For the CBF and VS measurements, pseudo-continuous arterial spin labelling (pCASL) data were acquired using 3T MRI. We segmented the ASL blood pool in the sagittal sinus to determine a common region of interest for each group. We used these images as masks to calculate average VS. Notably, for the comparison between children and adults the ratio between VS to gray matter CBF was used (VGR) instead of the VS, to take into account higher CBF in children. To get more insight into the oxygen utilization, oxygen extraction fraction (OEF) and cerebral metabolic rate of oxygen (CMRO ₂) were calculated. In adult participants acetazolamide (ACZ) was used as a vasodilatory challenge. The hematologic laboratory parameters hemoglobin (Hb) and LDH were used as markers of anemia and hemolysis, respectively.

Results

VS as a marker of cerebral shunting was higher in both adult and pediatric patients with SCD as compared to controls (p<0.01) and increased after ACZ administration in all groups (p<0.01) (Fig. 1A). VS was significantly associated with CBF both before (R ²=0.59, p<0.001) and after ACZ administration (R ²=0.57, p<0.001) in all groups (Fig. 1B). To test the impact of demographics and hematological parameters on the presence of shunting, VS was corrected for CBF (VS _CBF) and the residuals were used in multiple linear regression analysis against age, sex, blood flow velocity in the brain feeding arteries, participant groups, hydroxyurea, Hb and LDH. Cerebral shunting, reflected by VS _CBF showed significant association with Hb (Fig. 1C). In addition, we added OEF, CMRO ₂ and ACZ condition as additional parameters in adults and used a linear mixed model to accommodate the repeated measures dependencies. A negative association between the level of cerebral shunting (VS _CBF) and CMRO ₂ was found (β=-0.79, p<0.001) in all groups (Fig. 1C), and in adult patients with SCD, CMRO ₂, Hb (β=-14.2 p<0.001) and LDH (β=0.13, p=0.002) were significant predictors of VS _CBF.

Conclusion

Our results show that the VS in the sagittal sinus on ASL images can be used to assess functional arteriovenous shunting in the brain. Given its negative association with CMRO ₂ in combination with the negative association with hemoglobin and positive correlation with LDH, this functional shunting seems to reflect pathophysiologic shunting related to higher disease severity. Future studies will focus on the relation between functional shunting and the prevalence of SCI, investigating its link to aberrant capillary oxygen exchange in SCD.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal