Obstructive sleep apnea (OSA) is a risk factor for significant perioperative complications. This national survey study sought to determine the attitudes of physicians involved in the perioperative care of OSA patients.
Methods: We modified the perioperative survey used by Turner et al. among Canadian anesthesiologists. We mailed the survey to 3,000 US physicians practicing in the following specialties (750 of each specialty): anesthesiology (A), primary care (family practice or internal medicine) (PC), sleep (SM), and general surgery (S). The survey asked questions about attitudes and practice patterns regarding OSA in the perioperative setting.
Results: Of 2,730 eligible subjects, 783 questionnaires (28.7 %) were returned. Overall, 94 % felt OSA was a risk factor for perioperative complications (no difference by specialty) and 90 % felt it was a moderate to major risk factor (A = 91 %, PC = 81 %, SM = 94 %, S = 72 %; p < 0.001). Fifty-two percent reported experience with a patient having an adverse outcome related to OSA in the perioperative setting. Despite this, only 71 % reported regularly screening for OSA preoperatively, mostly by history and physical examination (A = 89 %, PC = 52 %, SM = 88 %, S = 49 %; p < 0.001). If they suspected a patient of having OSA, 32 % would delay surgery pending a sleep study (A = 4 %, PC = 41 %, SM = 54 %, S = 27 %; p < 0.001), while 20 % would proceed with surgery without any special precautions (A = 22 %, PC = 21 %, SM = 5 %, S = 31 %; p < 0.001). Only 27 % of respondents reported that their hospital had a written policy for perioperative care of OSA patients.
Conclusions: The majority of physicians in this survey felt OSA was a significant risk factor for perioperative complications and most reported experience with OSA patients having an adverse outcome. Perioperative management guidelines for OSA are not available at most institutions. Further work is needed to help physicians identify and intervene on patients with OSA in the perioperative setting before adverse events develop.