Incidence and recurrence of new-onset atrial fibrillation detected during hospitalization for non-cardiac surgery: a systematic review and meta-analysis.

McIntyre, William F; Vadakken, Maria E; Rai, Anand S; Thach, Terry; Syed, Wajahat; Um, Kevin J; Ibrahim, Omar; Dalmia, Shreyash; Bhatnagar, Akash; Mendoza, Pablo A; Benz, Alexander P; Bangdiwala, Shrikant I; Spence, Jessica; McClure, Graham R; Huynh, Jessica T; Zhang, Tianyi; Inami, Toru; Conen, David; Devereaux, P J; Whitlock, Richard P; Healey, Jeff S; Belley-Côté, Emilie P

PURPOSE This systematic review aimed to summarize reports of the incidence and long-term recurrence of new-onset atrial fibrillation (AF) associated with non-cardiac surgery. SOURCES We searched CENTRAL, MEDLINE and EMBASE from inception to November 2019. We included studies that reported on the incidence of new-onset perioperative AF during hospitalization for non-cardiac surgery and/or AF recurrence in such patients following discharge. Reviewers screened articles and abstracted data independently and in duplicate. We assessed study quality by appraising methodology for collecting AF history, incident AF during hospitalization, and AF recurrence after discharge. PRINCIPAL FINDINGS From 39,233 citations screened, 346 studies that enrolled a total of 5,829,758 patients met eligibility criteria. Only 27 studies used prospective, continuous inpatient electrocardiographic (ECG) monitoring to detect incident AF. Overall, the incidence of postoperative AF during hospitalization ranged from 0.004 to 50.3%, with a median [interquartile range] of 8.7 [3.8-15.0]%. Atrial fibrillation incidence varied with type of surgery. Prospective studies using continuous ECG monitoring reported significantly higher incidences of AF than those that did not (13.9% vs 1.9%, respectively; P < 0.001). A total of 13 studies (25,726 patients) with follow-up up to 5.4 years reported on AF recurrence following hospital discharge; only one study used a prospective systematic monitoring protocol. Recurrence rates ranged from 0 to 37.3%. CONCLUSIONS Rates of AF incidence first detected following non-cardiac surgery and long-term AF recurrence vary markedly. Differences in the intensity of ECG monitoring and type of surgery may account for this variation. TRIAL REGISTRATION PROSPERO (CRD42017068055); registered 1 September 2017.