Background: Postoperative hypoxemia is a common complication in the anesthesia recovery room (ARR), which is predominantly based on the development of atelectasis, excessive intraoperative fluid shift and insufficient ventilation. The goal of this prospective observational study was to compare the effect of standard oxygen administration via a face mask with oxygen administration using the EzPAP® system, a device which additionally provides a positive end-expiratory pressure (PEEP)., Methods: This study included 210 patients with postoperative hypoxemia (S(p)O(2) < 93%) subdivided into the control group (105 patients) and the EzPAP group (105 patients). Postoperative residual paralysis was excluded using relaxometry and a train of four (TOF) ratio of 0.9 was assumed to ensure sufficient recovery of respiratory function from neuromuscular blockade. Patients who received a reversal of neuromuscular blockade were excluded. In cases of hypoxemia (S(p)O(2) < 93%) control patients were treated with oxygen (6 l/min) using a face mask, whereas the EzPAP group received oxygen using the EzPAP® system. In order to adjust the PEEP in the EzPAP group, the O(2) flow was verified and measured by a manometer. After 1 h of oxygen therapy, the oxygen supply was stopped. In cases of reoccurring hypoxemia (S(p)O(2) < 93%, persistence > 5 min), the oxygen therapy was restarted in both groups via a facemask. Both groups were compared using repeat measurement analysis of variance (ANOVA), the unpaired t-test, the Mann-Whitney U-test, Fisher's exact test and the χ(2)-test. The correlation of O(2) flow and PEEP was evaluated by regression analysis and p < 0.05 was considered to be statistically significant. Apart from this a subgroup analysis was performed depending on body-mass index (BMI), American Society of Anesthesiologists (ASA) classification, intraoperative airway management, the use of neuromuscular blocking agents and co-existing disorders, e.g. chronic obstructive lung disease (COLD), obesity and chronic heart failure., Results: All patients were equally distributed between both groups with respect to demographic data, ASA classification, BMI, co-existing disorders and surgical procedures. The S(p)O(2) values did not differ between the EzPAP patients and the control group, except for 0.5 min after initiation of oxygen therapy: EzPAP group 96 ± 3.7% (mean ± standard deviation) versus the control group 93.8 ± 4.4% (p < 0.001). However, restarting oxygen therapy was less common in the EzPAP group (EzPAP group 25 versus control group 41, p = 0.03), as well as the occurrence of postoperative complications (EzPAP group 13 versus control group 25, p = 0.02), e.g. nosocomial pneumonia (0 versus 4) and wound infections (2 versus 3). Furthermore, patients with obesity and pulmonary disorders, such as COLD had a benefit from oxygen administration using the EzPAP device and showed higher postoperative than preoperative S(p)O(2) values. In contrast, the subgroup analysis of patients with heart failure did not reveal any differences between both groups and both groups did not differ in terms of time spent in the recovery room (EzPAP group 113 min versus control group 174.8 min, p = 0.2)., Conclusions: In this observational study oxygen supply using the EzPAP® system appeared to be at least equally as effective in the therapy of postoperative hypoxemia compared to standard oxygen supply using a face mask. In patients with a high risk of postoperative hypoxemia, such as patients with obesity and/or pulmonary disorders, oxygen administration using the EzPAP® system possibly improves pulmonary oxygenation more effectively and is longer lasting compared to standard oxygen supply via a face mask. Hence, the EzPAP® system represents a well-tolerated, effective, cost-effective and easily operated tool to improve postoperative oxygenation. In order to investigate the possibilities of this promising tool more intensively, randomized clinical trials are warranted.