Abstract 516

Children with sickle cell anemia (SCA) are at increased risk for stroke due to abnormal cerebral blood flow (CBF). The Stroke Prevention (STOP I) study showed that abnormal CBF could be detected using Trans-cranial Doppler (TCD), and may predict the risk for stroke. High TCD may be an indicator of main or branch cerebral artery vascular dysfunction, stenosis or fibrosis. Abnormal CBF is associated with cerebral ischemia and ultimately stroke. Brain derived neurotropic factor (BDNF) is a nerve growth factor with anti-apoptotic and neuro-protective properties. It is released in the central nervous system in response to ischemia. Platelet derived growth factor (PDGF) is an endothelial and smooth muscle mitogen. We hypothesized that circulating BDNF and PDGF levels will be elevated and may be associated with abnormal CBF as indicated by high TCD velocity measurements.

Three groups of children, i.e. 1) SCA with abnormal TCDs (SATCD) - abnormal TCD ≥200 cm/sec, 2) SCA with normal TCDs (SNTCD) -normal TCD ≤170 cm/sec and 3) healthy controls were investigated. Stored, baseline plasma samples from the Stroke Prevention (STOP I) study and from an ongoing study of nutritional effects on inflammation in SCA children (NUTSCD) were analyzed in duplicate, using a multiplex antibody immobilized bead assay. Plasma levels of BDNF, PDGF AA and AB/BB and 10 other pro- and anti-inflammatory cytokines were assessed. A total of 39 samples (11 SATCD, 21 SNTCD and 7 controls) were analyzed. Of the 11 SATCD, 8 were eventually randomized to the standard care (SC), while 3 were in the transfusion (Tx) arm of STOP I. Since some of the SCA patients subsequently developed stroke, the levels of BDNF and PDGF in those who did vs. did not develop stroke were compared. Except where not applicable, values are presented as mean±SD. Chi-square was used to access distribution, while means were compared using t-test and analysis of variance (ANOVA). A Receiver Operator Characteristic (ROC) curve was used to access the sensitivity of BDNF and PDGF in predicting abnormal CBF as indicated by abnormal TCD measurement.

The TCD velocity for SATCD group was 250±32 cm/s. There was no statistically significant difference in age or gender distribution among the groups. The mean body mass index was significantly lower for SCA subjects compared with controls (15±1kg/m2 vs. 23±7 kg/m2, p= 0.026). There was a statistically significant difference in mean BDNF levels, which was highest in SATCD compared with SNTCD (234±53 vs. 166±110 pg/ml, p=0.036) and controls (34±20 pg/ml, p<0.001). Difference in the BDNF levels between SNTCD and controls was also statistically significant, p=0.006. PDGF AA levels were significantly higher in SATCD compared with SNTCD (356±147pg/ml vs. 227±156 pg/ml, p = 0.045) and controls (34±32pg/ml, p<0.001). Also, PDGF AA levels were significantly higher for SNTCD vs. controls (p = 0.014). Six of the 7 patients who developed stroke had abnormal TCD, while one had conditional TCD (186cm/s, F, 3yo). However, all SCA patients who subsequently developed stroke had significantly higher PDGF AA levels than those who did not (350±180pg/ml vs. 249±154pg/ml, p=0.039). The BDNF levels, were not significantly different for SCA who subsequently developed stroke vs. those who did not (201±69pg/ml vs. 186±106pg/ml). Of the 3 patients receiving Tx, one who eventually had stroke, had very high (pre-Tx) PDGF AA level (312pg/ml); almost as high as those on SC who developed stroke (358±197pg/ml, n=6), and higher than the remaining 2 Tx patients who did not develop stroke (233±0.5pg/ml). The ROC curve showed that both BDNF and PDGF were sensitive (AUC=0.8, p<0.05) predictors of abnormal CBF as indicated by high TCDs velocity measurement.

High TCD velocity in children with SCA was associated with elevated BDNF and PDGF AA levels. Moreover, PDGF AA levels were significantly higher in the SCA patients who developed stroke. Further study is warranted to determine whether BDNF and/or PDFG AA, alone or with abnormal TCD, will better predict future stroke risk.
Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution