Abstract
BACKGROUND: Obstructive sleep apnea (OSA) is a common clinical condition. Due to its high prevalence, the waiting list for polysomnography is often long. A screening tool is needed to identify individuals at high risk for OSA who should undergo polysomnography. The STOP-Bang questionnaire is a widely used screening tool; however, it may require modification for individuals with a body mass index (BMI) below 35 kg/m(2). This study aimed to evaluate whether the STOP-Bang questionnaire should be modified for patients with a BMI under 35 kg/m(2). METHODS: This retrospective analytical study included adult patients suspected of having OSA who underwent polysomnography. Exclusion criteria included pregnancy and a BMI over 35 kg/m(2) or meeting criteria for bariatric surgery. Patients were categorized into OSA and non-OSA groups. Logistic regression analysis was used to assess the predictive value of STOP-Bang factors for OSA. RESULTS: A total of 188 patients were included, of whom 158 (84.04%) were diagnosed with OSA. Among the eight STOP-Bang criteria, only age was independently associated with OSA after adjustment for other variables (adjusted odds ratio: 1.04; 95% CI [1.02-1.08]). Optimal cut-off points for predicting OSA were identified as age ≥ 40 years (sensitivity: 84.18%), BMI ≥ 23 kg/m(2) (sensitivity: 82.91%), and neck circumference ≥ 35 cm (sensitivity: 86.08%). A modified STOP-Bang score incorporating these cut-offs showed improved sensitivity at a score of 3: 93.0% for apnea-hypopnea index (AHI) ≥ 5 events/hr, 95.9% for AHI ≥ 15 events/hr, and 97.6% for AHI ≥ 30 events/hr. In comparison, the original STOP-Bang score of 3 had sensitivities of 50.0%, 53.6%, and 56.1% for AHI ≥ 5 events/hr, AHI ≥ 15 events/hr, and AHI > 30 events/hr, respectively. CONCLUSIONS: The STOP-Bang questionnaire may require modification for individuals with a BMI below 35 kg/m(2) who are suspected of having OSA. Revised cut-off values for age, neck circumference, and BMI-40 years, 35 cm, and 23 kg/m(2), respectively-may enhance its diagnostic performance.