Prevalence and risk factors of obstructive sleep apnea in patients with unexplained dizziness: A real-world study

Obstructive sleep apnea (OSA) is a globally prevalent sleep disorder linked to a wide range of systemic conditions, including cardiovascular and vestibular dysfunction. Dizziness is frequently encountered in clinical practice but underexplored on causes and relationship with OSA. The present study aimed to assess the prevalence of OSA among patients with unexplained dizziness and identify associated risk factors and potential links between these conditions. A retrospective analytical study was conducted at a tertiary care University Hospital over a two-year period, from November 2022 to November 2024. The inclusion criteria were adult patients (aged ≥18 years) who had unexplained dizziness and underwent overnight polysomnography. Those who were pregnant or had identifiable cause of dizziness such as upper respiratory tract infection, anemia, hypoglycemia, anxiety, depression, postural hypotension, or hypertensive emergency were excluded. Eligible patients were evaluated for OSA by baseline characteristics, comorbidities, STOPBang score and physical examination. The overnight home sleep apnea test was used in all eligible patients. The diagnosis of OSA was defined by an apnea-hypopnea index (AHI) of 5 or more events/h. The primary outcome was the prevalence of OSA in patients with dizziness. Risk factors of OSA were also analyzed by using the multivariable logistic regression analysis. During the study period, 81 patients met the study criteria. Of those, 76 patients (93.83%) diagnosed as OSA with a median AHI of 23 events/h (IQR of 11.5-33.5). Patients with OSA had significantly larger neck circumferences than their non-OSA counterparts (37.63 vs. 33.25 cm; P=0.042). The STOPBang score was also higher in OSA group (3.95 vs. 2.40; P=0.026). There was no significant difference among other factors between both groups including demographic variables, clinical symptoms, other anatomical variants and comorbidities. There were four factors remaining in the model by stepwise logistic regression analysis. Only STOPBang score was independently associated with OSA with an adjusted odds ratio of 2.115 (95% confidence interval of 1.048, 4.268). The cut point of STOPBang score of 3 or over had sensitivity of 82.89% and specificity of 40.00% for OSA. OSA was extremely prevalent in patients with unexplained dizziness. The STOPBang questionnaire may be used as a screening tool to detect OSA in this setting with high sensitivity.