Nearly half of all patients with haemorrhagic stroke also experience headache across its acute and chronic phases that could contribute to long-term morbidity, according to a recent review published in Headache.
Yet the prevalence varied substantially across populations and clinical settings, wrote Bradley Ong, MD, adult neurology resident at Neurological Institute, Cleveland Clinic, and colleagues, who suggested the need for clinical trials focused specifically on post-stroke headache treatment, rather than extrapolating from primary headache disorders.
“In clinical practice, headaches after haemorrhagic stroke came up quite often in our clinical practice, but they were rarely addressed,” Ong told Healio.
Most treatment after stroke focuses on motor recovery and preventing its recurrence, he added, with headache treated as an incidental or transient symptom.
“When we looked at the literature, there was no clear, consolidated picture of how common these headaches are or how long they last,” Ong noted. “That gap is what motivated this study.”
He and and colleagues conducted a systematic review and meta-analysis that included 24 peer-reviewed, observational studies from Medline, Embase and CENTRAL with 4 688 adults (mean age, 56.9 years; weighted mean, 58.2% women) with haemorrhagic stroke.
“The most striking finding was how common headaches are,” Ong said. “Nearly half of the patients experience headache, and more than one-third go on to have persistent headaches months or years later.”
Overall, 46.1% (95% CI, 36.3% to 56.1%) of these patients experienced headache after their stroke. Eleven studies (n = 2,481) found that 55.9% of patients (95% CI, 41.1% to 70.1%) experienced acute headache. Thirteen studies (n = 2,207) found that 36.7% of patients (95% CI, 25.6% to 48.5%) had persistent headache.
“This challenges the assumption that headache is mainly an ‘acute’ symptom, especially in haemorrhagic stroke,” the team said.
Specific prevalences of headache included 58.3% (95% CI, 44.4% to 71.6%) for those with subarachnoid haemorrhage (SAH) and 36.1% (95% CI, 26.7% to 46%) for those with intracerebral haemorrhage (ICH).
Prevalence of severe headaches included 42.7% (95% CI, 15.8% to 72.1%) among those whose headaches were acute/subacute and 14.3% (95% CI, 10.4% to 18.7%) among those whose headaches were persistent.
With an overall I2 of 96.7%, the researchers said their findings indicated substantial heterogeneity in these pooled prevalence estimates, with no statistically significant differences based on study design, population, geography, Human Developmental Index or risk for bias.
Further, Ong and colleagues said there were no significant associations between risk for headache and female sex, nor were there any significant associations with history of diabetes mellitus, hypertension, alcoholism or previous headache.
“Another important finding was that headache at stroke onset strongly predicted chronic headache, which gives us an early clinical signal on which we can actually act,” Ong said.
The odds ratio for post-stroke headache among patients with headache at stroke onset was 1.7 (OR = 1.7; 95% CI, 1.4-2.05). Also, the odds ratio for post-stroke headache among patients with lobar ICH was 1.93 (95% CI, 1.08-3.44).
There were no significant associations between headache risk and cortical ICH or delayed cerebral ischaemia. Also, there were no significant associations between headache risk and the presence of an anterior circulation aneurysm among patients with SAH.
Patients with atrial fibrillation had less risk for headache (OR = 0.59; 95% CI, 0.37-0.95), which the researchers attributed to differences in stroke severity and symptom reporting and not to any direct protective effect.
Noting that the prevalence of headache among patients with haemorrhagic stroke exceeds the prevalence of other primary headache disorders among the general population, with substantial variations by population and clinical settings, the researchers called these headaches “common” as well as “persistent and disabling”.
Ong said that clinicians can use these findings to improve outcomes for patients with stroke.
“Clinicians should ask about headache routinely, both in the hospital and during follow-up. Headache should be treated as a meaningful post-stroke complication,” he suggested.
“Patients who report headache early may benefit from closer monitoring and earlier referral to headache care. Even simple steps like education and avoiding unnecessary opioid exposure can improve quality of life.”
Looking ahead, the researchers called for studies with standardised diagnostic criteria, clearly defined populations and detailed headache characteristics into protective therapies and secondary prevention strategies.
“The next step is prospective, longitudinal studies using standardised headache definitions and patient-reported outcomes,” Ong said.
“We also need clinical trials focused specifically on post-stroke headache treatment, rather than extrapolating from primary headache disorders. Ultimately, the goal is to integrate headache care into routine stroke recovery.”
Study details
Headache after haemorrhagic stroke: A systematic review and meta-analysis
Bradley Ong, Jad El Ahdab, Ahmet Günkan, Shervin Badihian, Neil Nero, Marina Vilardo, Lisa Wilson, Nicolas Thompson, Payal Patel Soni.
Published in Headache on 2 December 2025
Abstract
Background/Objective
Haemorrhagic stroke comprises about 20% of all strokes, with intracerebral haemorrhage (ICH) being the most common type. While post-stroke care often focuses on motor and functional recovery, post-stroke headaches remain under-recognised and understudied. This study aimed to summarise and pool the evidence on the prevalence and patterns of headaches after haemorrhagic stroke.
Methods
We conducted a systematic review and meta-analysis of studies published from database inception to December 2024, identifying observational studies that investigated headaches after haemorrhagic stroke. Studies enrolling five or more adult patients were included. Meta-analyses were conducted to estimate pooled prevalence of overall, acute/subacute, persistent, and severe headaches. Acute/subacute headache was defined as onset within 3 months of stroke, while persistent headache as headache lasting more than three months. Severe headache was defined as either a ≥ 7/10 pain intensity or functional impairment.
Results
Twenty-four studies comprising 4688 patients (58.2% female; mean age 56.9) were included. The overall pooled headache prevalence of 46.1% (95% confidence interval [CI]: 36.3%–56.1%; 95% prediction interval [PI]: 4.3%–91.8%; I2 = 96.7%) following haemorrhagic stroke. Stratified analyses showed that the prevalence was 58.3% (95% CI: 44.4%–71.6%, I2 = 97.5%) in patients with subarachnoid haemorrhage (SAH) and 36.1% (95% CI: 26.7%–46.0%, I2 = 93.9%) in those with ICH. Acute/subacute headache occurred in 55.9% (95% CI: 41.1%–70.1%; I2 = 97.6%), while persistent headache occurred in 36.7% (95% CI: 25.6%–48.5%, I2 = 93.1%). Severe headaches were reported in 42.7% (95% CI: 15.8%–72.1%; I2 = 98.0%) of patients with acute/subacute headache and 14.3% (95% CI: 10.4%–18.7%; I2 = 71.5%) with persistent headache. In both SAH and ICH, headaches frequently become chronic. No significant differences were observed by study design, geographic region, Human Development Index, or risk of bias.
Conclusion
Headache is a common but understudied condition that can manifest at or soon after a haemorrhagic stroke and can persist for years, potentially contributing to long-term morbidity. Standardised headache definitions and longitudinal assessments are needed to improve recognition and inform future clinical trials targeting this under-appreciated source of post-stroke morbidity. Further research is essential to better understand the nature of these headaches, which will help shape treatment protocols and enhance patient care.
Healio article – Nearly half of patients with haemorrhagic stroke experience headache
See more from MedicalBrief archives:
Abnormal stroke-related brain blood vessels linked to gut bacteria
Migraine associated with increased CVD risk