Article Text
PDF
Abstract
Introduction and objectives Obstructive sleep apnoea (OSA) is common and can be associated with adverse health outcomes. There are conflicting data for the impact of undiagnosed OSA on the outcome of surgical procedures but at least some results suggest an association with worse outcomes. EuroSCORE risk model was developed to calculate the risk of mortality after cardiac surgery. We evaluated the prevalence and impact of undiagnosed sleep apnoea (SA) on postoperative outcomes in cardiac surgery.
Methods Patients undergoing coronary artery bypass grafting with or without cardiac valve surgery were screened for the presence of SA, prior to surgery, with the STOPBANG questionnaire and overnight oximetry. SA was defined as a 4% oxygen desaturation index (ODI) of ≥5/hr. A Weibull model was used to analyse lengths of stay (LoS) in intensive care unit (ICU). Complications in ICU were dichotomised and analysed with binary logistic regressions. Parsimonious models were obtained using a combination of step-wise regression and manually removing predictors that did not reach the 5% significance level.
Results 122 subjects were included in final analysis of which 57 (47%) had a new diagnosis of SA. Of those, 45 (79%) had mild SA and 12 (21%) had moderate/severe SA. There was no simple relationship between OSA as measured by ODI and LoS in ICU. The most significant predictor for ICU LoS was developing complications at ICU (p < 0.001). The independent predictors associated with increasing likelihood of developing major organ complications following cardiac surgery were EuroSCORE, ODI and intravenous opioid analgesia (IOA). When patients with mild and moderate SA received IOA, predicted probability of complications rose 2.4 and 1.4 times respectively (Figure 1).
Predicted probabilities and 95% CI of suffering a complication at ICU as ODI increases for individuals with average EuroSCORE (5) and with or without IOA
Conclusion We found a high prevalence of undiagnosed sleep apnoea in our cohort. EuroSCORE, SA and the administration of intravenous morphine were found to be independent risk factors for developing post-operative complications. This risk has increased when patients with SA received intravenous morphine.