Postoperative hypoxemia: common, undetected, and unsuspected after bariatric surgery.

BACKGROUND: Patients undergoing gastric bypass are at greater than ordinary risk for postoperative respiratory insufficiency, presumably related to obstructive sleep apnea (OSA) and patient-controlled analgesia (PCA). This study was proposed to quantify the magnitude of the problem.
METHODS: Fifteen patients undergoing gastric bypass had oxygen saturation (SpO(2)) recorded continuously, but not displayed, for 24h postoperatively; eight also had arterial blood analysis every 4h. All received narcotic PCA. SpO(2)<90% lasting more than 10 s was reviewed. Results are mean+/-SEM.
RESULTS: Mean age was 44+/-4 y, and mean BMI was 48+/-2kg/m(2); 77% had OSA. Every patient had more than one episode with SpO(2)<90% for longer than 30s undetected by routine monitoring; most had multiple episodes. Nadir SpO(2) averaged 75% +/- 8%. Mean longest duration of desaturation below 90% averaged 21+/-15min. Mean PaCO(2) was 37+/-3mm Hg; maximum PaCO(2) was 47mm Hg.
CONCLUSIONS: Severe and prolonged episodes of hypoxemia were a consistent finding, despite aggressive preoperative diagnosis and treatment of OSA, including use of CPAP postoperatively. Although some postoperative hypoventilation was expected, the degree and frequency of desaturation were surprising. No patient exhibited arterial PaCO(2) evidence of hypoventilation. No patient experienced cardiopulmonary arrest/instability, in spite of severe, repeated episodes of hypoxemia. In no instance was a significant hypoxemic episode suspected or detected. Continuous pulse oximetry monitoring, with an audible alarm set for a saturation less than 90% for 10 s, would have alerted providers to 100% of significant hypoxemic episodes. Our recommendation is routinely monitoring (with alarm capability enabled) every bariatric surgical patient, to prevent such occurrence.