Background Migraine without aura (MwoA) is the most common migraine subtype in children and can cause persistent impairment of quality of life. The International Classification of Headache Disorders, 3rd edition (ICHD-3), requires an attack duration of ≥2 h to diagnose pediatric MwoA; however, this threshold is debated, and several studies suggest that a lower limit may be appropriate in children. Methods In this single-center retrospective study, children with primary headache seen in our neurology clinic from 2018 to 2023 were identified. Baseline demographic and headache characteristics were extracted from records. Patients were grouped by typical attack duration into a short-duration MwoA group (SdMwoA, ≥1 min to 2 h) and an MwoA group. Headache outcomes at 6 months, 12 months and at the last follow-up were collected via clinic visits or structured telephone interviews. Multivariable logistic regression was used to identify factors associated with headache-free status and any improvement at 6 and 12 months. Results Of the 293 screened children, 192 (97 SdMwoA patients and 95 MwoA patients) met the inclusion criteria; follow-up data were available for 67 (69.1%) and 66 (69.5%) patients, respectively. Compared with the SdMwoA group, the MwoA group more often had a headache history ≥1 month before presentation (61.1% vs. 41.2%, p = 0.006) and prior preventive treatment (29.5% vs. 12.4%, p = 0.004), while the other baseline features were similar. At the last follow-up, headaches had remitted or improved in 57/67 (85.1%) SdMwoA patients and 54/66 (81.8%) MwoA patients ( p = 0.613). Typical attack duration was not associated with outcomes according to multivariate regression modeling. Conclusion SdMwoA and MwoA have comparable short- to mid-term outcomes. These findings support a more flexible, child-centered interpretation of the ICHD-3 duration criterion in pediatric migraine, indicating that typical migraine phenotypes with attacks less than 2 h warrant consideration of MwoA when alternative diagnoses have been excluded.
Ma et al. (2026) studied this question.