Document resource
Introduction Worcester Acute Hospitals Trust operates a physiologist-led sleep apnoea service with over 4000 patients. For individuals with suspected hypercapnia (pCO2>6.0 kPa), a capillary blood gas (CBG) test is required to confirm pCO2 status, ensuring accurate diagnosis and treatment pathway. If patients exhibit SpO2<90% for over 30% of total sleep time (SpO2<90≥30%TST), they meet criteria for a CBG test to screen for daytime hypercapnia. Anecdotally, staff observed that some patients who met these criteria were not receiving CBG tests, and many patients with a pre-CPAP CBG did not have hypercapnia.This study aimed to:Evaluate sleep service compliance for CBG testing compared to local guidance, and whether local guidance is appropriate.Evaluate whether other sleep study parameters may be better predictors of daytime hypercapnia, and can these be used to streamline the service.Methods Data was collected retrospectively from 938 patients who had a CPAP trial between September 2022–2023. Of these, 214 patients met the criteria for a CBG, and their outcomes were analysed. Compliance with local guidance was assessed, and CO2 status of patients who received a CBG was determined. Various sleep parameters were also evaluated for their ability to differentiate between eucapnic and hypercapnic patients. Pearson correlation coefficients, multivariate logistic regression, and receiver operating curves (ROC) were applied to assess model performance.Results 44.4% of patients (n=95) who met criteria for a CBG, did not have a CBG pre-CPAP trial. Of those with a CBG (n=119), 25.2% (n=30) were hypercapnic. SpO2<90 correlated with pCO2 (r=0.310), while mean nocturnal SpO2 showed a stronger correlation with pCO2 (r=-0.407). Regarding discriminatory performance, SpO2<90 had an AUC of 0.656. A model incorporating mean nocturnal SpO2 and BMI demonstrated superior discriminatory power with an AUC of 0.758, which was significantly better than SpO2<90 at distinguishing between hypercapnic and eucapnic patients (P=0.0253). ( Figure 1)Abstract P30 Figure 1Conclusions The sleep service is not meeting local guidance, with 44.4% of eligible patients not receiving a CBG pre-CPAP. A model incorporating mean nocturnal SpO2 and BMI may provide improved discriminatory power for identifying patients who require a CBG, potentially improving service efficiency.