ISC 2026 | Post-stroke headache: prevalence, mechanisms, and clinical predictors

Headache is a rather common entity in an acute setting of stroke, but what shocked us the most would be from our study, what we found out was it was very common after hemorrhagic stroke. So in our study, what we did was we pulled in 24 studies and that comprised of around 4,700 patients. We found that the pooled prevalence of headache was around 46 overall. It was higher in subarachnoid hemorrhage and lower after intracerebral bleed, it’s about around 36%. And then we also looked at the timing of the headache, meaning to say we put it in an acute or subacute bucket. It’s less than three months or those that persisted after three months being a chronic type of headache. And what we found out was around 37% of patients would have persistent headache lasting beyond the three-month mark, whether it’s years or even decades after having that stroke. And in terms of the mechanism, I’ll be honest, I don’t think there’s one thing, though it’s one general gestalt to summarize the pathophysiological biology as to why these patients would have headache after the stroke. What we learned through looking at the literature was that in subarachnoid hemorrhage, the biology tends to lean towards more of meningeal inflammation, downstream issues later on after having that stroke in terms of like CSF flow disturbances, intracranial hypertension, versus if you look at an intracerebral bleed, the main pattern there is more so disruption of the central pain networks in our brain, neuroinflammation, especially if the cortex of the brain is involved. So the most consistent clinical predictor that we found for our patients in the study that we pulled together was we found that headache at the onset had around 1.7 times the odds of developing persistent post-stroke headache. In terms of imaging, however, we found that a lower type of bleed, whether it be in a frontal, occipital, temporal area, it’s around twice, is its odds ratio of developing persistent headache. And this fits with the idea that if it’s a cortical involvement, there might be some central pain disruption that occurs that we haven’t discovered as of yet in a molecular basis. A lot of the factors that we also looked into in terms of predictive pooled analyses, we looked into sex, we looked into history of hypertension, history of diabetes, even history of prior headache, and we didn’t see any connection or association with these variables.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.