MATERIAL AND METHODS
This was a single-center cohort study in which all data were collected prospectively for surgery occurring between January 2011 and December 2017. Patients with pre-operative AF, who were above 70 years and underwent 2 cardiac procedures with or without additional AF procedures were included in the study. Patients were divided into two groups based on how the AF was addressed: (1) Cox-maze IV procedure (2) Nil surgical AF treatment group. Patients undergoing redo procedures or who had isolated Pulmonary vein isolation (PVI) or left atrial appendage occlusion (LAAO) to address their AF were excluded from the study. History of preoperative atrial fibrillation was determined through our local database and type of atrial fibrillation was determined according to Heart Rhythm Society guidelines. The database was also used to gain additional preoperative characteristics and perioperative outcomes. Detailed follow-up data was collected for patients through telephone consultations and medical record review. In addition, reports from primary care physicians and cardiologists from referring centres were obtained if required. Rhythm status for patients who underwent a surgical ablation procedure was determined according to the Heart Rhythm Society guidelines and verified by electrocardiogram and Holter monitor. The Heart Rhythm Society definition of success (ie, all documented atrial Arrhythmias > 30 seconds are considered a failure) was used to determine the return to sinus rhythm rate at first follow (usually 6 weeks), annual follow and long-term follow-up [15]. Anticoagulation status was also collected at the follow-up time points. Operative mortality was defined as death occurring within 30 days of operation or at any time point during the index hospitalization.
Operative Approach
Multiple surgeons performed the complete Cox-maze IV lesion set in a standard fashion as described previously. This consisted of a bilateral PVI, roof and connecting lesions between the right and left pulmonary veins, lesion to the left atrial appendage, mitral isthmus lesion, right intercaval lesion, right appendage lesion, right free wall lesion to the tricuspid annulus lesion and the coronary sinus lesion. The energy source used was cryothermia and bipolar radiofrequency (Medtronic, Minneapolis, Minn; AtriCure Inc, West Chester, Ohio). The left atrial appendage was occluded in all patients who had Cox-Maze IV. This was performed using the Atriclip device (AtriCure Inc, West Chester, Ohio). The patients in the “Nil procedure” group only had two cardiac procedures and served as our primary control group. The decision of whether to add the Cox Maze procedure to a specific surgical procedure was left to the discretion of the surgeon.
Statistical Analysis
Continuous data are presented as mean +/- standard deviation or Median +/- Interquartile range. Categorical data is presented as frequency (+/-percent) unless otherwise noted. Patient groups were compared usingc 2 or Fisher exact test for preoperative and postoperative categorical variables and student independent samples t test or Mann-Whitney U test for continuous measures as appropriate based on parametric test assumptions. Statistical significance was considered p < 0.05, 2-tailed. Kaplan-Meier analysis was used to compare the groups on cumulative survival, freedom from AF, NYHA 1 status, freedom from permanent pacemaker insertion and freedom from stroke. Gehan-Behan-Wilcoxon test was used to assess significance of these survival analyses. All analyses were conducted using SPSS version 17.0 (SPSS Inc, Chicago, Ill) or GraphPad Prism, Version 6.00 for Mac (GraphPad Software, La Jolla, CA, USA).