Previous studies discussed the possible relationship between migraine, stroke and hypercoagulability syndromes. In this study, a statistically significant difference was found between migraine patients and normal subjects with regard to mean serum level of protein S and anti-thrombin III, being significantly deficient in patients group, which goes in agreement with previous studies [9, 10].
In the current study, there was no statistically significant difference in mean serum level of protein C between migraine patients and healthy controls, which agreed with previous studies done by Bassi et al. , Rajan et al. , D’Amico et al. , Cavestro and Mandrino  who reported that unlike protein S deficiency which has a high prevalence, the main problem of studying protein C deficiency is its very low prevalence in the general population (0.14–0.40%). This usually requires larger sample sizes to detect any significant difference in prevalence between migraine patients and control group.
The present study showed no significant difference between MA and MO patients with regard to mean serum levels of protein C, protein S, or anti-thrombin III. This was in agreement with Martinez-Sanchez et al. , who reported that hypercoagulability had a higher frequency in MA patients (42.4%) in comparison to MO patients (26.7%); however, the difference was not statistically significant.
In the current study, a statistically significant difference was found between patients and controls with regard to the presence of abnormal MRI findings in form of ischemic foci and leukoareosis. This goes in accordance with studies done by Igarashi et al. , Swartz and Kern , and Kruit et al.  which revealed that migraine is associated with sub-clinical brain infarcts in MRI, suggesting that brain infarctions occur more frequently than expected in migraine patients. Three meta-analyses of observational studies also reflect that vascular risk is increased by twofold in migraine patients .
The prevalence of white matter hyper-intense lesions (WMHLs) in migraineurs was found to range from 6 to 40% . This can be explained in many studies by other comorbidities such as age, hypertension, diabetes, or associated demyelinating diseases [18, 19]. On the other hand, Swartz and Kern  demonstrated that migraine patients are close to four times more likely to show these WMHLs changes than age- and sex-matched controls. This increased risk was present even in young individuals who did not have any vascular risk factors or other disease comorbidities.
According to aura presence, this study showed no significant difference between MA and MO patients regarding brain MRI findings. Moreover, the serum protein levels were comparable in the two groups of patients with no significant difference, a finding which go in accordance with previous studies done by Fazekas et al. , Pavese et al. . On the contrary, Ferbert et al. , Gozke and et al.  found a higher prevalence of brain ischemic lesions among MA patients. An increased risk of ischemic stroke was repeatedly reported in young subjects with migraine especially MA. This contradiction may be explained in view of the small sample size in this study.
In MA patients, Gursoy-Ozdemir et al.  suggested that cortical spreading depression (CSD) may cause disruption of the blood–brain barrier (BBB) through a matrix metalloproteinase-9-dependent cascade mechanism, which may result in local tissue damage. Moreover, Pezzini et al.  concluded that MO appears to be associated with hypercoagulability in the whole sample and in the transient ischemic attack (TIA) group, whereas MA is related to hypercoagulability in brain infarction patients under 50 years. One possible explanation was that some TIAs could actually be auras, and this may be acting as a confounding factor.
In this study, a statistically significant correlation was detected between serum protein S deficiency and abnormal brain MRI findings, however, no correlation was found between deficient serum protein C or anti-thrombin III and brain MRI findings. This agreed with D’Amico et al.  who found that 6.4% of patients with ischemic strokes had low levels of protein S, suggesting that protein S deficiency can produce a state of hypercoagulability and induce stroke, at least when it is present in conjunction with other acquired or hereditary factors.
Anzola et al.  also found a relatively high prevalence of protein S deficiency in patients with previous ischemic strokes which highly suggested that the ischemic event might be related to the hypercoagulability caused by the decrease in free protein S levels.
The present study revealed that deficient serum protein C was an independent predictor of the severity of migraine headache. To the best of our knowledge, no previous researches studied this point.