Article Text


Sleep: clinical studies
P24 Prevalence of obstructive sleep apnoea in patients scheduled for bariatric surgery and validation of the STOP-BANG questionnaire as a screening tool
  1. R Kaiser,
  2. C Hammond,
  3. L Restrick,
  4. H Branley,
  5. S Lock,
  6. P Sufi,
  7. D Heath
  1. Whittington Hospital, London, UK


Background Obstructive Sleep Apnoea (OSA) is common in morbidly obese patients scheduled for bariatric surgery, and many sleep centres perform routine preoperative sleep studies for all. However a significant proportion will not have significant OSA. Epworth Sleepiness Score (ESS) is unreliable in predicting the risk of OSA. A practical screening tool is ideally required. We aimed to define the prevalence of OSA in our bariatric patient population and validate the STOP-BANG questionnaire as a screening tool.

Methods Retrospective review of bariatric patients who had sleep studies over a 3-month period from January to March 2011. Clinical data collected from medical notes and sleep study results. STOP-BANG scores derived retrospectively from clinical data. Questionnaire included 8 yes/no questions, scored 1 for every yes answer (Abstract P24 table 1). A score of 4 or more was considered as high risk for having OSA. STOP-BANG scores were then correlated with sleep study results. Significant OSA (which may require treatment with preoperative CPAP) was defined as a Apnoea-Hypopnoea Index (AHI) of at least >15.

Abstract P24 Table 1

Stop-bang questionnaire (Score 1 for every Yes answer)

Results Patient characteristics (n=61): mean age 45 (24–69), 87.3% female, mean BMI 46.2 (35–67), mean ESS 6.5 (0–20). Sleep study results – 18% had AHI 15–30, 13.1% had AHI >30. 55.7% had STOP-BANG score of = 4, 44.3% had score of = 3. Of patients with significant OSA (AHI>15): mean BMI 47.2, mean neck circumference 41.4 (SD 2.5), Mean ESS 8.5 (SD 4.84), 66.6% had ESS <11, 89.4% were loud snorers. Using STOP-BANG score of = 4 to screen for OSA with AHI >15—Sensitivity 94.7%, Specificity 61%, positive predictive value 52.9%, negative predictive value 96.2%.

Conclusions 31% of patient population studied had at least moderate OSA. ESS poorly predictive of risk of OSA. Using a high risk STOP-BANG score of 4 had a high sensitivity but poor specificity. However, a low risk score of <4 had a high negative predictive value of 96.2% for AHI >15. Therefore STOP-BANG questionnaire using a cut-off risk score of 4 can be used as a screening tool to rule out significant OSA and thus avoiding sleep studies in a significant proportion of low risk patients.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.