P276 The value of vital capacity and daytime pulse oximetry to predict hypercapnia in obese patients

Abstract

Introduction: The Health Survey for England reported that 25% of UK adults are obese with a 10% rise over 15 years. Consequently, clinicians are faced with a rising number of obese patients referred for bariatric and non-bariatric surgery. Previous data indicates a 50% incidence of obstructive sleep apnoea in patients with a BMI >40 kg/m 2 with obesity hypoventilation syndrome present in up to a third. These patients have higher risk of peri-operative complications. A screening tool to predict hypercapnic respiratory failure (P a CO 2 6 kPa) based on simple clinic tests would be useful. Correlations were performed to determine which tests may be useful. Methods: Data from all obese patients (BMI >30 kg/m 2) with evidence of sleep-disordered breathing on oximetry initiated on home ventilatory support between August 2005 and December 2010 were obtained from a discharge summary database. Results: 205 patients were included for analysis. The group mean age was 54.9 (SD 14.2) years, daytime clinic oxygen saturations (SpO 2clinic) 91.0% (5.8%), FEV 1 1.8 l (0.96 l), FVC 2.2 l (1.11 l), weight 132.8 kg (28.5 kg), BMI 47.6 kg/m 2 (9.6) and Epworth sleepiness score 8.9 (5.6). Mean daytime P a CO 2 was 6.68 kPa (1.31). Significant correlations were found between P a CO 2 and BMI (r=0.20; p<0.005), FEV 1 % predicted (r=-0.20; p<0.005), FVC% predicted (r=-0.20; p<0.005) and SpO 2clinic (r=-0.52; p<0.005). Receiver operating characteristics (ROC) analysis was used to determine the utility of SpO 2clinic and FVC to predict hypercapnia. The area under the curve (AUC) for SpO 2clinic was 0.81 (p<0.001); a cut-off of SpO 2clinic of <92% demonstrated a sensitivity of 86% and specificity of 52% in predicting hypercapnia. The AUC for FVC was found to be 0.77 (p<0.0001); a cut-off of <1.94 l demonstrated a sensitivity of 77% and specificity of 61% in detecting hypercapnia (see Abstract P276 figure 1). (Figure Presented) Conclusion: These data have significant clinical utility for clinicians involved in providing respiratory support services for obese patients undergoing bariatric and non-bariatric surgery. In particular, it could form the foundations of a screening algorithm including simple measures such as home oximetry, spirometry and clinic pulse oximetry, to identify the highest risk patients that need to be reviewed by sleep and ventilation clinicians.