The prevention of delirium in elderly surgical patients with obstructive sleep apnea (PODESA): a randomized controlled trial.

BACKGROUND: Obstructive sleep apnea (OSA) is associated with neurocognitive impairment - a known risk factor for postoperative delirium. However, it is unclear whether OSA increases the risk of postoperative delirium and whether treatment is protective. The objectives of this study were to identify OSA with a home sleep apnea test (HSAT) and to determine whether auto-titrating positive airway pressure (APAP) reduces postoperative delirium in older adults with newly diagnosed OSA undergoing elective hip or knee arthroplasty. METHODS: We conducted a multi-centre, randomized controlled trial at three academic hospitals in Canada. Research ethics board approval was obtained from the participating sites and informed consent was obtained from participants. Inclusion criteria were patients who were [Formula: see text]0 years and scheduled for elective hip or knee replacement. Patients with a STOP-Bang score of ≥ 3 had a HSAT. Patients were defined as having OSA if the apnea-hypopnea index was ≥ 10/h. These patients were randomized 1:1 to either: 1) APAP for 72 h postoperatively or until discharge, or 2) routine care after surgery. The primary outcome was postoperative delirium, assessed twice daily with the Confusion Assessment Method for 72 h or until discharge or by chart review. The secondary outcome measures included length of stay, and perioperative complications occurring within 30 days after surgery. RESULTS: Of 549 recruited patients, 474 completed a HSAT. A total of 234 patients with newly diagnosed OSA were randomized. The mean age was 68.2 (6.2) years and 58.6% were male. Analysis was performed on 220 patients. In total, 2.7% (6/220) patients developed delirium after surgery: 4.4% (5/114) patients in the routine care group, and 0.9% (1/106) patients in the treatment group (P = 0.21). The mean length of stay for the APAP vs. the routine care group was 2.9 (2.9) days vs. 3.5 (4.5) days (P = 0.24). On postoperative night 1, 53.5% of patients used APAP for 4 h/night or more, this decreased to 43.5% on night 2, and 24.6% on night 3. There was no difference in intraoperative and postoperative complications between the two groups. CONCLUSIONS: We had an unexpectedly low rate of postoperative delirium thus we were unable to determine if postoperative delirium was reduced in older adults with newly diagnosed OSA receiving APAP vs. those who did not receive APAP after elective knee or hip arthroplasty. TRIAL REGISTRATION: This trial was retrospectively registered in clinicaltrials.gov NCT02954224 on 03/11/2016.

Fate of the hypoplastic proximal aortic arch in infants undergoing repair for coarctation of the aorta through a left thoracotomy.

BACKGROUND: Extended end-to-end anastomosis (EEEA) through a left thoracotomy for coarctation of the aorta (CoA) and tubular hypoplasia of the aortic arch (THAA) leaves an unaugmented hypoplastic proximal aortic arch (PAA) segment, which may increase late reintervention for PAA obstruction. We sought to assess PAA growth and reintervention for PAA obstruction after EEEA. METHODS: Preoperative and follow-up echocardiographic images of 140 patients who underwent EEEA for CoA from 2005 to 2012 were reviewed. Patients were divided into two groups on the basis of preoperative PAA z-scores: THAA group, z-score less than -3; non-THAA group, z-score greater than or equal to -3. RESULTS: Eighty (57%) patients were identified as having THAA. There were three surgical reinterventions (PAA in 2 patients and distal aortic arch in 1 patient) and nine catheter reinterventions (all related to anastomotic stenosis) during a median follow-up period of 18 months. Both patients who required PAA reintervention had preoperative PAA z-scores below -8. Freedom from reintervention at 3 years was comparable between the groups (THAA group, 90.0% vs non-THAA group, 87.9%, p = 0.483). Follow-up echocardiography revealed PAA catch-up growth in the THAA group (z-score, preoperative -4.63 vs follow-up -1.17, p < 0.001); however, there was a nonsignificant trend toward smaller PAA in the THAA group (z-score: THAA, -1.17 vs non-THAA, -0.55, p = 0.057). All but 2 patients with preoperative PAA z-scores above -6 did not have any PAA obstruction. CONCLUSIONS: The hypoplastic PAA segment in patients with CoA/THAA grew significantly after EEEA but remained smaller than in those without THAA. Our data support that CoA and PAA with z-scores as small as -6 can be repaired through a thoracotomy approach with a low risk of reintervention.