Article Text


Clinical studies in obstructive sleep apnoea
P207 Obstructive sleep apnoea in patients undergoing bariatric surgery—a London teaching hospital experience
  1. C J Jolley,
  2. S Zimmerman,
  3. S Shah,
  4. N Davison,
  5. S Singh
  1. Chelsea & Westminster Hospital NHS Foundation Trust, London, UK


Introduction and Objectives Obstructive sleep apnoea (OSA) is common in patients scheduled for bariatric surgery and increases the likelihood of peri-operative adverse events. We aimed to describe the prevalence of OSA and risk profile of patients referred for pre-bariatric assessment locally.

Methods A retrospective observational study of patients referred to the sleep clinic for assessment before bariatric surgery between June 2008 and February 2010. Clinical and anthropometric data were collected from the hospital notes and sleep studies. Patient-reported STOP BANG model scores were recorded or derived retrospectively from clinical data. Non-parametric statistics were used due to non-normally distributed data.

Results 140/164 patients referred were seen in clinic. Referral rates increased from 2/month to 15/month in February 2010. The median (range) age was 46.5 (18–68) years, 71% female, weight 135.5 (87.4–180 kg), BMI 48.4 (35.3–84.5) kg/m2), and median ESS was 11 (0–24). STOP BANG scores were reported or could be calculated in 84 patients, in whom the median score was 5/8 (2–8). When incomplete STOP BANG scores were included, 124/130 scored >2. 53% were non-smokers, 15% current smokers and were 37% ex-smokers. Comorbidities included: diabetes/IGT 70%, hypertension 50%, hypercholesterolaemia 39%, ACS/heart failure 9%, CVA 2%, COPD 3%, asthma 19%, and hypothyroidism 19%. Sleep studies were requested for 129 patients, completed in 116 patients. 114 were technically adequate for AHI and 106 for pulse oximetry. The median total AHI was 10.5 (0–111.2)/h, ODI 20.8 (0.2–145.2)/h, and mean SpO2 93.4 (78.3–98.6)%. 27% had AHI ≥5/h, 12% ≥15/h and 25% ≥30/h. Correlations between sleep study outcomes and clinical data are shown in Abstract P207 Table 1. Using STOP BANG >2 to screen for AHI ≥5 had a sensitivity and specificity of 99.0% (95% CI 92.4 to 100%) and 5.6% (0.7–18.7%), respectively.

Abstract P207 Table 1

Correlations between sleep-study measurements and clinical, demographic and anthropometric variables

Conclusions Referral rates for sleep studies pre-bariatric surgery have risen exponentially over the past 20 months, and 37% of patients studied had at least moderate OSA, presenting a burden for sleep services and CPAP provision. 35% of patients studied had no evidence of OSA. Although a sensitive test, STOP BANG alone did not reliably identify these lower-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.