Sleep-disordered breathing and psychiatric comorbidities in narcolepsy: A 10-year retrospective study from Saudi...

Narcolepsy is a chronic disorder of central hypersomnia that is frequently under-recognized. Diagnostic delay, comorbid sleep-disordered breathing (SDB), and psychiatric comorbidities are well described internationally but remain poorly characterized in Saudi Arabia.

We retrospectively reviewed all consecutive adult patients diagnosed with narcolepsy at the sleep disorders center of King Fahad Medical City (KFMC), a tertiary referral hospital in Riyadh, between January 2015 and April 2026. Diagnosis followed the third edition of the International Classification of Sleep Disorders (ICSD-3). Patients were classified as narcolepsy type 1 (NT1) when cataplexy was present and as narcolepsy type 2 (NT2) when MSLT criteria were met without cataplexy. Sleep breathing disorders and psychiatric conditions were assessed in all patients. Diagnostic delay was defined as the interval between patient-reported symptom onset and the year of formal diagnosis. Predictors of any sleep-disordered breathing (any-SDB: apnea-hypopnea index ≥ 5/h or REM-related OSA) and of diagnostic delay were examined with multivariable logistic and linear regression; the linear regression for diagnostic delay was performed on the natural-logarithm-transformed delay (log[delay + 1]) to satisfy the normality assumption. Cerebrospinal fluid hypocretin testing was not performed.

A total of 113 patients were analyzed (65 NT1 [57.5 %] and 48 NT2 [42.5 %]; 39 women [34.5 %]; median age 27 years [IQR 20-34]). Median body mass index (BMI) was 29.0 kg/m² (IQR 24.9-34.0). Median diagnostic delay was 5 years (IQR 2-10; range 0-28) and increased with age (Kruskal-Wallis H = 10.7, p = 0.030; Pearson r with age = 0.43, p < 0.001). Any-SDB was present in 52/113 (46.0 %); REM-related OSA in 46/110 (40.7 %). After adjustment for narcolepsy type, BMI, and Epworth Sleepiness Scale (ESS), female sex was independently less at risk for any-SDB (adjusted odds ratio [aOR] 0.34, 95 % CI 0.13-0.88, p = 0.026) and older age conferred increased risk (aOR 1.06 per year, 95 % CI 1.01-1.11, p = 0.023). In the multivariable linear model for log-transformed diagnostic delay (adjusted R² = 0.119), age was the only independent predictor (β = +0.020 per year on the log scale, equivalent to a 2.0 % increase in delay per year of age; 95 % CI + 0.002 to +0.038, p = 0.032). ESS independently predicted dichotomized delayed diagnosis (≥5 years) (aOR 1.15 per ESS point, 95 % CI 1.03-1.29, p = 0.013). Cataplexy was, by definition, present in all NT1 patients (100 %) and inabsent of all NT2 patients. Sleep paralysis (61.5 % vs 33.3 %, p = 0.003) and hypnagogic/hypnopompic hallucinations (50.8 % vs 16.7 %, p < 0.001) were significantly more common in NT1. Modafinil was the dominant wake-promoting agent (88.5 %); antidepressant prescribing was concentrated in NT1 (any-antidepressant 73.8 % vs 22.9 %, p < 0.001). Major depressive disorder (16.8 %) and generalized anxiety disorder (9.7 %) showed no subtype difference.

In this Saudi tertiary-center cohort, narcolepsy was associated with a substantial diagnostic delay that increased with age at presentation, and with a high prevalence of comorbid sleep-disordered breathing, in which female sex and younger age were independent predictors of lower risk for SDB. The dataset supports stronger advocacy for early MSLT testing in adult patients with chronic excessive daytime sleepiness, and for systematic screening of psychiatric comorbidity at narcolepsy diagnosis.