Abstract

Introduction: The pathophysiology of overt stroke in children with sickle cell anemia (SCA) is not well understood. However, a compromised cerebral hemodynamic environment, marked by elevated cerebral blood flow (CBF) and impaired cerebrovascular reserve are thought to play a critical role. Currently, stroke prevention is based on chronic transfusion therapy (Tx). However, Tx suffers from several complications including; iron overload, risk of infection, and hyperhemolysis reactions. Hydroxyurea (HU) is a well-tolerated myelosupressive agent, which increases fetal hemoglobin (HbF) and is being tested as an alternative to Tx for primary stroke prevention in children with SCA. However, it is currently not clear how either Tx or HU treatment compare with respect to their effects on the precarious hemodynamic environment in children with SCA. Therefore, we assessed cerebrovascular reserve by obtaining MR-based measures of cerebrovascular reactivity (CVR), defined as the change in blood flow in response to a vasoactive stimulus. Considering that HU and Tx both act to dilute HbS and increase oxygen carrying capacity, we hypothesized that SCA children treated with HU and Tx will exhibit increased CVR and decreased CBF compared to age-matched, non-treated, SCA children.

Methods: 21 SCA (HbSS) untreated SCA children (5M/16F; avg age 14.1 ± 2.42), 16 SCA children treated with HU (10M/6F; avg age 14.0 ± 2.49) and 7 patients on Tx (3M/4F ; avg age 13.0 ± 1.81) were recruited from the Haematology Department at The Hospital for Sick Children. Exclusion criteria were; history of overt stroke, vasoocclusive crises, and febrile illness within 30 days of the scan. CVR was obtained using a blood-oxygen level-dependent (BOLD) MRI sequence in combination with a CO2 breathing challenge. CO2 was administered via a re-breathing mask in 4 alternating cycles of 40 mmHg PETCO2 for 60 seconds and 45 mmHg PETCO2 for 45 seconds. BOLD MRI signal changes were aligned and regressed against the CO2 waveform on a voxel-wise bases using FSLv4.1 to yield the raw CVR maps. The CVR maps were then co-registered to anatomical images. CBF measurements were obtained using a pulsed arterial spin labeling (pASL) sequence and were quantified from the mean signal difference between the ASL tag and control images using a single compartment kinetic model. The Student's t-test was used to test for significant differences between CVR, CBF, and blood parameters between the HU, Tx, and No Treatment groups. Blood parameters were obtained from complete blood counts.

Results: Blood parameters are shown in Table 1.HU-treated SCA children exhibited significantly higher Hct and CVR in both the gray and white matter compared to the No Treatment group, with no significant differences in CBF. The Tx-treated SCA group exhibited higher Hct, CVR in the gray matter, and decreased CBF in both the grey and white matter compared to the No Treatment group (see Figure 1).

Conclusions: Both the HU and Tx-treated SCA groups demonstrated significantly improved cerebrovascular reserve compared to the No Treatment group. The increase vascular reserve may be a result of HU and Tx decreasing the severity of anemia thereby reducing the baseline vessel tone necessary to sufficiently perfuse the cerebral tissue. CBF was significantly reduced in the Tx-treated group compared to the No Treatment group, which may be due to the higher oxygen content provided by Tx. Interestingly, CBF was not significantly different between the HU-treated and No Treatment group despite the HU-treated group having significantly greater hematocrit. This was surprising as a previous study using Doppler ultrasound (TCD) demonstrated that HU significantly reduces blood velocities (TCDv). However, unlike TCDv which are velocity measurements obtained at the large feeding arteries, pASL CBF values reflect perfusion at the tissue level. Further studies are required to fully compare TCDv and pASL measures of CBF as a means to monitor treatment effects in SCA patients.

Table 1.
 SCA: No Treatment SCA: Hydroxyurea SCA: Transfusion 
Sample Size (N) 21 16 
Gender (M/F) 5/16 10/6* 3/4 
Age (mean ± SD) 14.1 ± 2.42 14.0 ± 2.49 13.0 ±1.81 
Hematocrit 0.253 ± 0.037 0.289 ± 0.034* 0.308 ± 0.034¥ 
Hemoglobin (g/L) 86.4 ± 10.9 101 ± 12.1* 105 ± 11.2¥ 
Hemoglobin F (%) 15.8 ± 8.07 3.01 ± 32.2 
Reticulocyte (%) 9.4 ± 6.1 7.2 ± 3.6 
MCV (fL) 80.3 ± 10.8 94.9 ± 12.3* 
SaO2 0.97 ± 0.03 0.99 ± 0.01 
Neutrophils (K/uL) 6.1 ± 3.1 4.2 ± 1.6* 
 SCA: No Treatment SCA: Hydroxyurea SCA: Transfusion 
Sample Size (N) 21 16 
Gender (M/F) 5/16 10/6* 3/4 
Age (mean ± SD) 14.1 ± 2.42 14.0 ± 2.49 13.0 ±1.81 
Hematocrit 0.253 ± 0.037 0.289 ± 0.034* 0.308 ± 0.034¥ 
Hemoglobin (g/L) 86.4 ± 10.9 101 ± 12.1* 105 ± 11.2¥ 
Hemoglobin F (%) 15.8 ± 8.07 3.01 ± 32.2 
Reticulocyte (%) 9.4 ± 6.1 7.2 ± 3.6 
MCV (fL) 80.3 ± 10.8 94.9 ± 12.3* 
SaO2 0.97 ± 0.03 0.99 ± 0.01 
Neutrophils (K/uL) 6.1 ± 3.1 4.2 ± 1.6* 

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.