Diagnostic Delay and Its Predictors in Cluster Headache

Objective: Cluster headache (CH) is a rare, primary headache disorder, characterized of excruciating, strictly one-sided pain attacks and ipsilateral cranial autonomic symptoms. Given the debilitating nature of CH, delayed diagnosis can increase the disease burden. Thus, we aimed to investigate the diagnostic delay, its predictors, and clinical influence among patients with CH. Methods: Data from a prospective multicenter CH registry over a 4-year period were analyzed. CH was diagnosed according to the International Classification of Headache Disorders (ICHD)-3 criteria, and diagnostic delay of CH was assessed as the time interval between the year of the first onset and the year of CH diagnosis. Patients were classified into three groups according to the tertiles of diagnostic delay (1st tertile, <1 year; 2nd tertile, 1–6 years; and 3rd tertile, ≥7 years). Results: Overall, 445 patients were evaluated. The mean duration of diagnosis delay was 5.7 ± 6.7 years, (range, 0–36 years). Regarding the age of onset, majority of young patients (age <20 years) belonged to the third tertile (60%), whereas minority of old patients (>40 years) belonged to the third tertile (9.0%). For year of onset, the proportion of patients in the 3rd tertile was the highest for the groups before the publication year of the ICHD-2 (74.7%) and the lowest for the groups after the publication year of the ICHD-3 beta version (0.5%). Compared with the first CH, episodic CH [multivariable-adjusted odds ratio (aOR) = 5.91, 95% CI = 2.42–14.48], chronic CH (aOR = 8.87, 95% CI = 2.66–29.51), and probable CH (aOR = 4.12, 95% CI = 1.48–11.43) were associated with the tertiles of diagnostic delay. Age of onset (aOR = 0.97, 95% CI = 0.95–0.99) and PHQ-9 score (aOR = 0.96, 95% CI = 0.93–0.99) were inversely associated with the tertile of diagnostic delay. The prevalence of suicidal ideation was highest in the patients of the third tertile. The mean HIT-6 score increased significantly with the diagnostic delay (p = 0.041). Conclusions: Patients with a younger onset of CH have a higher risk of diagnostic delay. Nevertheless, the rate of delayed diagnosis gradually improved over time and with the publication of the ICHD criteria, supporting the clinical significance of diagnostic clinical criteria and headache education to reduce the disease burden of CH.