The perioperative use of continuous positive airway pressure (CPAP) therapy has increased substantially in recent years, particularly in relationship to the treatment of patients with known or suspected obstructive sleep apnea (OSA). OSA is common in the surgical population and is reported as an independent risk factor for postoperative complications, intensive care unit admission, and increased length of hospital stay. A large proportion of OSA patients are undiagnosed at the time of surgery and can therefore not be optimized preoperatively. Nowadays, golden standard treatment of moderate to severe OSA is nightly CPAP at home, often with an autotitration mode. Unfortunately, there are only a handful of randomized clinical trials investigating the effect of preoperative and/or postoperative CPAP treatment in OSA patients, so the perioperative guidelines are based on a combination of randomized clinical trials, observational studies, case studies, and expert opinions. In this review, we have summarized the current evidence regarding the use of perioperative CPAP therapy with an emphasis on patients with OSA. We identified 21 randomized, controlled trials that investigated the effect of CPAP on postoperative physiology and complications in surgical patients. Our review reveals evidence, suggesting that CPAP after surgery improves oxygenation and reduces the need for reintubation and mechanical ventilation after surgery. It is also evident that CPAP reduces apnea and hypopnea frequency and related hypoxemia after surgery. Poor adherence to CPAP in the perioperative setting is a limiting factor in assessing its potential to optimize postoperative cardiorespiratory outcomes. Studies of postoperative outcomes in patients who have previously been prescribed CPAP for OSA and are therefore familiar with its use could help to address this shortcoming, but they are unfortunately lacking. This shortcoming should be addressed in future studies. Furthermore, many of the studies of the postoperative effect of CPAP in OSA patents are small, and therefore, single-center studies and larger randomized, controlled multicenter studies are warranted.