Claudia Berends

The prevalence of (obesity-related) obstructive sleep apnea (OSA) is high (70-94%) in patients undergoing metabolic and bariatric surgery (MBS). Patients with OSA may have an increased risk of postoperative complications, although evidence in the MBS population remains inconclusive. Currently, the best available evidence for the perioperative management of OSA in MBS is a consensus guideline from 2017, which may be outdated as practices such as same-day discharge were not yet widely adopted at that time. Therefore, the aim of this dissertation was to evaluate and improve perioperative strategies for the management of OSA in MBS, in both the inpatient and same-day discharge setting, with particular attention to minimizing excessive OSA diagnostics and treatment. In the Netherlands, there are two strategies that are mainly used: 1. “Sleep test strategy”, in which all patients are preoperatively tested for OSA with poly(somno)graphy, and subsequently treated with continuous positive airway pressure (CPAP) if indicated. 2. “Pulse oximetry strategy”, in which patients are not preoperatively tested for OSA, but are postoperatively continuously monitored with pulse oximetry and receive supplemental oxygen therapy (2L/min) until the next morning. The consensus guideline recommends using the sleep test strategy, while the added value of OSA testing and subsequent treatment is questionable. Considering that surgically induced weight loss will lead to OSA improvement, subsequent CPAP therapy is often applied for only a short treatment period. The pulse oximetry strategy might be a viable alternative, as it minimizes OSA diagnostics and treatment. A large prospective cohort study (POPCORN study, n=1390) compared the sleep test strategy and the pulse oximetry strategy and found the pulse oximetry strategy to be more cost-effective and equally safe. Elaborating on this, in Chapter 2 and Chapter 3 of this dissertation, total weight loss and patient-reported outcomes for sleepiness and health related quality of life were found to be comparable between the sleep test strategy and pulse oximetry strategy in the POPCORN study population until 1 year (Chapter 2) and 5 years (Chapter 3) after MBS. In Chapter 4, the desaturation rate and nursing intervention rate during the first postoperative night was characterized in the pulse oximetry strategy in a prospective observational study (SATBAR study). Though the oxygen administered in the pulse oximetry strategy may improve baseline saturation, it does not prevent collapse of the upper airway in untreated OSA patients. This collapse of the airway could lead to desaturations, which may require nursing interventions and potentially result in a high workload for the nursing staff. The results of the SATBAR study showed a relatively low desaturation rate (22%) and nursing intervention rate (14%), and that male sex, higher age, higher BMI and smoking were risk factors for developing postoperative desaturations. These findings indicate that the pulse oximetry does not impose a high workload for the nursing staff. Furthermore, it can be postulated that desaturations presumably occur more frequently in patients with (undiagnosed) OSA as risk factors for desaturation largely correspond with risk factors for OSA, which is in line with the assumption that oxygen therapy alone does not prevent apneas. To evaluate if omitting supplemental oxygen is non-inferior to administering oxygen during the first night after MBS, a randomized non-inferiority trial (OXYBAR trial) was performed in Chapter 5. The results of the sleep test (WatchPat300®) showed that based on the apnea-hypopnea index (AHI), omitting supplemental oxygen was non-inferior to administering supplemental oxygen therapy. The increased desaturation rate and inferiority of oxygen desaturation index (ODI) appear not to be clinically relevant with low absolute AHI values and similar outcomes for sleep architecture, 30-day complication rate and nursing intervention rate. In Chapter 6, the postoperative course of sleep disordered breathing (including AHI), oxygen saturation and sleep architecture were assessed in same-day discharge after MBS in a population with an unknown OSA status. In order to gain knowledge for evidence-based patient selection and discharge criteria for patients undergoing same-day MBS. This was assessed in a prospective observational study (DAGBAR study), in which patients underwent a sleep test (WatchpatONE®) at home, approximately 1 week prior to MBS and during postoperative night 1 and 3. The DAGBAR study showed that all sleep disordered breathing parameters (AHI, ODI, RDI) and number of desaturations were highest during postoperative night 3, compared to preoperative and postoperative night 1. In line with the hypothesis, the percentage of rapid eye movement sleep decreased during postoperative night one and increased during postoperative night three. The 30-day overall complication rate was 5.3%, of which none were cardiorespiratory related. In conclusion, the pulse oximetry strategy in optimized version without routine postoperative oxygen suppletion is a viable option for perioperative OSA management in MBS. The pulse oximetry strategy minimizes diagnostics and treatment while being more cost-effective and equally safe and with comparable total weight loss and patient-reported outcomes for sleepiness, daily functioning and health-related quality of life. The trend towards same-day discharge in MBS should not pose a barrier to the widespread adoption of the pulse oximetry strategy, as there appears to be no strict reason to consider OSA a contraindication for same-day discharge in the general bariatric population. Future research should be directed toward reducing the postoperative deterioration of OSA in perioperative OSA management strategies without preoperative OSA testing, for example by assessing the effect of opioid free anesthesia and opioid free postoperative pain management.