Diagnostic performance of one visual aura image in identifying migraine with aura

Accurate diagnosis of migraine with aura is crucial. This study compared diagnostic accuracy of a visual aura image alone and the same image combined with additional questions, using the clinical diagnosis of migraine with or without aura as the reference standard.

In this cross-sectional diagnostic accuracy study conducted by the Leiden Migraine Group between September 2024 and January 2025, patients with diagnosed migraine were recruited. Upon enrolment, participants completed a survey followed by a questionnaire featuring an image depicting a typical visual aura disturbance. Only one image was used, as presenting multiple images may be impractical in clinical settings or large epidemiological studies.

A total of 481 migraine patients were included, 171 with aura and 310 without aura. The visual aura image alone demonstrated a sensitivity of 0.89 (95%CI 0.84, 0.94), a specificity of 0.65 (0.60, 0.70), a positive predictive value (PPV) of 0.59 (95%CI 0.52, 0.65), and a negative predictive value (NPV) of 0.92 (95%CI 0.87, 0.95). Adding a validated extensive questionnaire reduced sensitivity to 0.53 (95%CI 0.45, 0.60), but increased specificity to 0.90 (95%CI 0.86, 0.93), with a PPV of 0.74 (95%CI 0.65, 0.81) and NPV of 0.77 (95%CI 0.73, 0.82).

A single visual aura image identifies most migraine with aura cases but shows limited specificity, frequently misclassifying patients without aura. Adding structured questions about visual symptom spread, positive phenomena, lateralization, duration, and timing relative to headache improves diagnostic accuracy, indicating that a visual aura image alone has limited clinical utility.