Predictor of atrial fibrillation recurrence in patients who underwent a tricuspid valve operation with modified Cox maze procedure

Recurrence of any atrial arrhythmia after surgical ablation is known as a negative predictor of cardiovascular events and total mortality. However, there have been no focused studies for atrial fibrillation (AF) recurrence prediction in patients with significant tricuspid regurgitation (TR), and the risk‐benefit estimation of surgical ablation in tricuspid valve (TV) surgery is not fully established.


INTRODUCTION
Ablation of atrial fibrillation (AF) can reduce the risk of cardiovascular events compared to medical treatment, and recurrence of any atrial arrhythmia after ablation can be a predictor of cardiovascular events and total mortality. 1 The benefits of surgical ablation in structural heart disease (SHD) are well recognized, 2 and more than half of the patients with AF undergoing open-heart surgery have concomitant AF surgery these days. 3 Despite ablation, the risk of AF recurrence still remains, and AF recurrence can have negative effects on survival, heart failure exacerbation, and stroke risk. 4 AF recurrence risk estimation after ablation is not fully established yet as much as its importance. AF recurrence risk estimation is important in respect to decision to additional procedure or decision to continue the anticoagulation. For now, analysis of AF recurrence is usually conducted in patients who undergo general cardiac surgery or mitral valve surgery. 5,6 However, as far as my knowledge, there is no concerted study for patients with significant tricuspid valve (TV) disease.
The Maze operation is controversial in patients with significant tricuspid regurgitation (TR) resulting from SHD due to concern of recurrence of AF regardless of the Maze operation. There is no established risk benefit estimation for Maze operation in TV surgery. Therefore, in this study, we analyzed predictors of AF recurrence after a modified Cox Maze (CM) procedure in patients with AF who underwent a TR operation and attempted to identify clear cutoffs for the benefits and safety of the CM procedure in TV surgery.

Study subject and data collection
We retrospectively investigated 385 patients who underwent a TV operation between 2001 and 2017 at Samsung Medical Center in Seoul, South Korea. We enrolled 158 patients who had significant TR with AF who underwent TV repair or replacement and CM procedure.
Patients who did not have the CM procedure, did not have a followup ECG after 3 months post-op, or had inadequate echocardiographic measurements were not included. Patients were divided by recurrence of AF within 10 years after the TV operation with CM procedure ( Figure 1).
The primary endpoint was AF recurrence. We analyzed the difference between the AF recurrence group versus no recurrence group and analyzed the AF recurrence factor in terms of clinical and echocardiographic risk factors. After that, subgroup analysis was done between types of SHD, and between isolated TR and non-isolated TR.
The medical records of the enrolled patients were carefully reviewed by research coordinators. Mortality data for patients who were lost to follow-up were confirmed by National Death Records. The study protocol was approved and the requirement for informed con-

Surgical technique
Detailed techniques of the cryo-maze procedure were described in our previous report. 7 The cryo-maze procedure was performed with an N2O-based cryoprobe or an argon-based cryoprobe according to sur- The opening LA appendage was internally obliterated without an excision using a running 3-0 monofilament suture.

Definitions
AF recurrence was defined as restoration of AF rhythm at least one time in follow up electrocardiography (ECG) more than 3 months after the operation. Patients who never returned to sinus rhythm were counted in the AF recurrence group and recurrence day was set as zero. All of the ECG taken in outpatient clinic and admission period were included in analysis of AF recurrence after the index procedure.
Median follow up duration was 7.6 years.
SHD in this study was defined as more than a moderate degree of valve disease, previous cardiac operation history, or congenital heart such as like atrial septal defects (ASD). Isolated TR without SHD indicates secondary TR caused by AF. Significant VHD indicates more than a moderate degree of valve disease.
Chronic kidney disease was defined as a glomerular filtration rate (GFR) < 60 ml/min/1.73 m 2 over 3 months and a disease code in medical record. Coronary artery disease was defined as over 50% stenosis in at least one coronary artery on computed tomography (CT) angiography or coronary angiography. TR recurrence was defined as reappearance of more than a moderate degree of TR after restoration to a minimal or mild degree at early after surgery. Stroke was defined as a neurological deficit of abrupt onset caused by ischemia or hemorrhage within the brain.

Statistical analysis
All data are presented as mean ± SD or median (IQR) for continuous variables and as number and percentage for categorical variables. The C index or concordance C was considered an overall measure of discrimination in survival analysis, and we tested whether there was a difference between two correlated overall C indices. The optimal cutoff values of RA diameter for predicting AF recurrence was calculated to maximize the product of sensitivity and specificity using receiver operating characteristic (ROC) curves.
All p-values were two-sided, and p-values < .05 was considered statistically significant. Statistical analyses were performed using R Statistical Software (version 3.6.0; R Foundation for Statistical Computing, Vienna, Austria) and SPSS statistics 20 (SPSS Inc., Chicago, IL).  Table 3). The C-index value of age for predicting AF recurrence was 59.6%. When this prediction value of age was combined with significant echocardiographic parameters, only RAD had an additive value to the C index, at 69.2% (p = .008, when compared to the C-index of age alone, which was 59.6%) (Figure 2).  (Figure 3) When comparing the C index value of solitary echocardiographic parameters for AF recurrence prediction, RAD showed superior predictive value over LAVI (p value = .043, Figure S1).

Predictors of AF recurrence
In subgroup analysis according to valve disease, LAVI and RAD were higher in AF recurrence group compared to non-recurrence group in every subgroup with or without statistical significance (Table S1). In comparison of C index value of LAVI and RAD, C index of RAD was numerically higher than C index of LAVI in every subgroup and statistical significance was prominent in significant mitral regurgitant group ( Comparing right heart-related echocardiographic parameters between the AF recurrence group and the no AF recurrence group, only IVC diameter showed a significant difference. The IVC diameter was higher in the AF recurrence group, although the RV dysfunction incidence estimated by TAPSE and TVs' did not differ significantly. Differences of RV size and TV annulus size were also insignificant ( had an addictive prognostic value to age ( Figure S2). For 145 patients, the optimal cutoff point for AF recurrence was 49 mm, and the probability of AF recurrence was 4.1 times in those with a preoperative RAD diameter of more than 49 mm ( Figure S3).

Clinical outcome
In  Figure 4).

Surgical outcome
During TV operation with CM procedure, there was no mortality case. There was one redo MVR (mitral valve replacement) case for bleeding control and three reoperation cases (1.9%) due to hemothorax. Sick sinus syndrome requiring PPM insertion occurred at six patients (3.8%) and AV block was not observed. There was no brain infarction and there were four cases of brain hemorrhage (2.5%) without neurologic sequela. Among four cases of brain hemorrhage, one patient needed burr hole operation during hospitalization period.
CRRT needed at three patients who had chronic kidney disease before operation. Two patients needed MCS after operation for short period (Table 6).

DISCUSSION
In this study, we investigated predictors of AF recurrence and the safety and efficacy of a modified Cox maze procedure in patients with AF who underwent a TV operation. The major findings of this study were as follows: 1. RA enlargement is key predictor of AF recurrence after a TV operation with CM procedure in patients with significant TR, 2. AF recurred more than four times more commonly in patients with RAD over 49.2 mm.

Predictors of AF recurrence
AF recurrence analysis based on echocardiographic parameters has been well-studied recently. Risk models like the APPLE score 9 and

Relation of PPM insertion and AF recurrence
AF is commonly associated with sick sinus syndrome. Histopathologic change damage to the sinus node, the perinodal tissue and the sinus node artery is associated with persistent AF. 29 Atrial myocardial structure relating to sinus node dysfunction (SND) is primarily in the RA 30 and RA stretching is known to cause SND. 31 Electrical remodeling secondary to AF also could contribute to SND. 32,33 In this study, AF recurrence group has higher rate of PPM insertion during follow up period. The association between AF recurrence and high incidence of sick sinus syndrome after maze was addressed in previous study. 34 Electrical remodeling by AF and histopathological change of atrium could contribute to higher incidence of sick sinus syndrome requiring PPM insertion in AF recurrence group.

Safety and efficacy of modified Cox maze procedure in TR operation
The pathophysiology of AF is reentry and atrial ectopic activity. Ablation for arrhythmia targets rhythm control by isolation of focal atrial ectopic activity. Concomitant surgical ablation of AF is recommended with level of evidence grade I in open surgical ablation and level II evidence in closed surgical ablation. 35 However, the beneficial effect of the maze procedure in TV surgery has not been discussed. In this study, the success rate of the CM procedure in TV surgery was 58.9% (93/158) without mortality. Although there were cases of brain hemorrhage, acute renal injury requiring renal replacement therapy in few patients, there was no major sequela in these patients which need permanent dialysis or neurologic rehabilitation. A total 41.1% recurrence rate of AF at 10 years is comparable to other results of CM procedure during mitral valve surgery 36 and catheter ablation. 37 This study result proved safety and efficacy of CM procedure in TV surgery.

Limitations
The study has several limitations that need to be addressed. First, this is single center study, so center characteristics could be reflected here.
However, a relatively long follow-up duration after surgery (median 7.9 years) is strength of this study. We need external validation moving forward. Second, RAD could be affected by volume status, but there was no measurable value that reflects volume status of patients in our data. We should assume that all patients have euvolemic volume status before surgery. Third, because of the limitations of a retrospective study, we could not identify the value of the RA volume index (RAVI).
Since RA size has fewer clinical implications than LA size, RAVI was not included in routine protocol. We could include RAVI measurements in a prospective study later. Fourth, there was a limitation in subgroup analysis in patients with isolated TR (n = 22) due to the relatively small number of patients. Although we could not reach valuable conclusions in the isolated TR subgroup, it is meaningful that AF recurrence factors are analyzed in end-stage SHD accompanied by significant TR, and this differs from the common-sense view that LA size is most important among echocardiographic parameters in AF recurrence prediction.

CONCLUSION
RA enlargement is key predictor of AF recurrence after a modified Cox Maze procedure in patients with significant TR. We can use RA size in clinical risk prediction models to create a risk-benefit estimation of the CM procedure in patients with significant TR and AF.

CONFLICTS OF INTEREST
None of the authors have any conflicts of interest to declare.