This might implicate that patients in western countries are more vulnerable to severe vasoconstrictions, and therefore, more complications than Asian patients. However, whether the results of our study could be generalizable to other ethnic populations deserve further investigation. Since all cases and controls in this study were of the same ethnicity and our observed allele frequencies are consistent with those observed for Han Chinese, population stratification is unlikely to play a major confounding role in our study. Nonetheless, the difference in polymorphism frequency could be a major issue for subsequent replication studies with different ethnic backgrounds. The findings of this study raise the significance of potential biologic actions of BDNF polymorphism on cerebral vasoconstric- tion. If the findings could be validated with consistent and independent replication, we might be able to identify patients at a higher risk of severe Etofibrate vasoconstrictions in the era of personalized medicine. The mechanisms underlying our findings are not known. The followings are some possible explanations. First, earlier study demonstrated that depolarization-induced secretion was reduced in 66Met BDNF-transfected neurons compared with 66Val BDNF analogs. Thus subjects carrying Val allele may have higher intracranial BDNF activity than Met/Met homozygotes. Second, it has been reported that under circumstances of sympathetic overactivity, BDNF could lead to perivascular inflammation and thus cause marked vasoconstriction. BDNF has been shown to dramatically upregulate neuropeptide Y, a vasoconstrictive sympathetic co-transmitter. Since sympathetic overactivity may play an important role in the pathogenesis of RCVS, it is plausible that subjects with higher intracranial BDNF activity, EUK 134 the Val carriers, could have more severe vasoconstriction. Third, BDNF-overexpression was found to modulate vascular tone of small pulmonary arteries through interactions with neurokinin A. Because receptor subtypes of neurokinin A differ between cerebral and pulmonary arteries, whether the study results could be extrapolated to RCVS requires further studies. Ou study had several limitations. First, the case number in this study was not large for a genetic polymorphism study and might have led to false negative results of clinical correlates, such as thunderclap attacks or total duration. However, the positive results with large effect sizes alleviated potential type II errors caused by small sample size. Second, all the patients with VMCA.120 cm/s were Val carriers. Based on our previous study, the risks of PRES or ischemic stroke in these patients should be higher. However, because of limited case number with complications in the present study cohort, we failed to observe this expected finding. One reason for lower percentage of patients with complications could be our improving ability of early recognition and treatment for RCVS. Third, we did not check plasma BDNF levels in patients, and were unable to ensure whether the modulatory effect of this polymorphism came from an altered BDNF level and/or its downstream pathways. Nonetheless, recent studies demonstrated that plasma concentration of BDNF was not associated with the Val66Met variant and there was no significant correlation between serum and regional brain BDNF levels. Furthermore, the possibility that the actual causal SNP was not the Val66Met polymorphism per se but located at a highly linked locus could not be completely excluded.