OBSTRUCTIVE sleep apnea affected nearly half of referred patients with cluster headache undergoing polysomnography in a retrospective study.
Researchers evaluated 52 adults with cluster headache who underwent polysomnography because of suspected sleep-disordered breathing or clinical judgment. Of these, 41 patients, or 78.8%, met criteria for obstructive sleep apnea using an apnea hypopnea index of at least 5 events per hour, while 25, or 48.1%, had treatment relevant obstructive sleep apnea, defined in the study as an index of at least 10 events per hour.
The findings highlight a potential role for more targeted obstructive sleep apnea screening in patients with cluster headache, although the study population represented a selected group referred for sleep testing rather than the broader cluster headache population.
The conventional STOP-Bang questionnaire showed modest performance for identifying treatment relevant obstructive sleep apnea. At a cutoff score of at least 3, sensitivity was 76.0% and specificity was 55.6%.
Patients with treatment relevant obstructive sleep apnea had a significantly later median age at cluster headache onset than those without the condition, at 32 years compared with 21 years.
Researchers therefore added cluster headache onset at age 30 or older to the STOP-Bang score. At the same cutoff of at least 3, sensitivity increased to 84.0%, while specificity declined to 48.1%. The modified model achieved an area under the curve of 0.804 compared with 0.767 for conventional STOP-Bang, although this difference was not statistically significant.
However, cluster headache onset at age 30 or later provided significant additional predictive information beyond STOP-Bang in likelihood ratio testing. Increasing the modified score cutoff to at least 4 improved specificity to 77.8%.
Polysomnography was performed during an active cluster bout in 28 patients and during remission in 24. Apnea hypopnea index, STOP-Bang scores, and treatment relevant obstructive sleep apnea prevalence were comparable between the groups.
Adding age at cluster headache onset numerically increased screening sensitivity during both active bouts and remission, suggesting that the fixed characteristic could offer a practical advantage regardless of when sleep evaluation occurs.
Circadian, nocturnal, morning, and seasonal headache patterns did not improve obstructive sleep apnea identification beyond conventional STOP-Bang.
The findings suggest clinicians may benefit from maintaining awareness of coexisting sleep-disordered breathing when evaluating patients with cluster headache, particularly when other clinical features raise suspicion. However, the modified score was developed in a small, selected cohort and requires prospective external validation before clinical implementation.
Reference
Kang MK et al. STOP-Bang Modification for Screening Treatment-Relevant Obstructive Sleep Apnea in Cluster Headache: A Retrospective Cohort Study. J Clin Med. 2026;15(15):5952.
Featured Image: kittyfly on Adobe Stock.
Each article is made available under the terms of the Creative Commons Attribution-Non Commercial 4.0 License.
