Article Text

Download PDFPDF

P69 Evaluation of the STOPBANG threshold in the pre-operative screening for obstructive sleep apnoea at sherwood forest hospitals foundation trust
  1. A Reynor,
  2. AW Molyneux,
  3. SD Tilbrook,
  4. RB Dean,
  5. E Crookes,
  6. J Tansley,
  7. NJ Ali
  1. Sherwood Forest Hospitals Foundation Trust, Mansfield, UK


Background Undiagnosed Obstructive Sleep Apnoea (OSA) has been associated with a higher perioperative morbidity and mortality. The aim of this study is to prepare a comprehensive review of the STOPBANG score, a pre-operative screening tool for patients with possible OSA. This study investigates if the current STOPBANG threshold of ≥3/8 is appropriate, or if it should be increased to ≥5/8 or ≥4/8 (with ¾ in the STOP category) to avoid the unnecessary cancellation/postponement of surgery and inappropriate referrals into the sleep service.

Methods This was a retrospective study of patients referred to the Sleep Service following a positive STOPBANG score (≥3). The Research and Development Department of the hospital deemed the study did not require ethical approval. 84 patients were included in the study. The selected patients’ case notes were used to review their STOPBANG score, Epworth Sleepiness Score, type of sleep study performed, Oxygen Desaturation Index (ODI), diagnosis and treatment. If the patient had an ODI ≥ 15, or was successfully started on treatment with a borderline ODI > 5 < 15, this was considered an appropriate referral for that threshold.

The sensitivity and specificity of the different STOPBANG thresholds were calculated to assess if the threshold score of STOPBANG ≥3 is appropriate, or if this should be adjusted to more appropriately identify those patients with OSA.

Results For a threshold of ≥3, the sensitivity is very high (100%). The sensitivity is decreased for the threshold of ≥5 (71%), and further decreased for the ≥4/8 (3 from STOP) threshold (53%). For the ≥3 threshold, the specificity is 0%. The specificity is dramatically increased for the ≥5 threshold (70%), and highest in the ≥4/8 (3 from STOP) category (85%).

Conclusion The statistical analysis confirms that a threshold of ≥3 has a very high sensitivity, but very low specificity. A threshold of ≥5 has a lower sensitivity however is much more specific and may be a more useful way of identifying the high risk of OSA surgical patient. A change in protocol to ≥5 aims to reduce the unnecessary delay or cancellation of surgery and avoid inappropriate referrals into the Respiratory and Sleep department.

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.