Post-Liver Transplantation Atrial Fibrillation: Insights into Clinical and ECG Predictors

Introduction

AF is the most common sustained cardiac arrhythmia and contributes significantly to morbidity and mortality burdens worldwide. It is associated with structural and electrophysiological changes in the heart that can be developed in patients with advanced metabolic diseases or undergoing major surgical procedures. The liver transplant recipients have an increased risk for development of POAF due to perioperative hemodynamic stress, increased serum catecholamine levels, intravascular fluid imbalance, and systemic inflammation.

POAF has been associated with serious long-term adverse outcomes such as thromboembolic events, stroke, heart failure, and even graft instability; also, prolonged hospitalization and increased mortality have been noted. Therefore, the study of high-risk patients for POAF is important in surgeries like organ transplantation, since there are ethical responsibilities of surgeons and clinicians to the donor and recipient [1].

Age, HT, and left atrial enlargement are the most studied parameters whose association with POAF was identified in many studies. Furthermore, there is an accumulation of data on the potential value of ECG markers, including PR interval prolongation, QRS fragmentation, and abnormal P-wave duration, as possible predictors of AF in the general population [2].

We assessed potential clinical, biochemical, and ECG predictors of AF in patients in the perioperative period of LT between 2012 and 2024. We aimed to elucidate the risk profile of this special population to enable early intervention, possible prevention, and tailored postoperative care.

Material and Methods

This retrospective study included patients who underwent LT between 2012 and 2024. A total of 161 patients were screened. Patients with a history of pre-existing AF, incomplete ECG data, or lost to follow-up were excluded from the study. After exclusions, data of 110 patients were screened from hospital records. Demographic and clinical variables, including age, sex, body mass index (BMI), hypertension (HT), diabetes mellitus (DM), prior cardiac history, and postoperative data were collected from hospital records.

The following parameters were measured on preoperative ECG at a paper speed of 25 mm/s and a calibration of 10 mm/mV: minimal P-wave duration (ms) (the shortest P-wave duration measured across all 12 leads), maximal P-wave duration (ms) (the longest P-wave duration measured across all 12 leads), PR interval (ms) (the time interval between the onset of the P-wave and the beginning of the QRS complex), QRS duration (ms) (the time duration of the QRS complex measured in the lead with the widest QRS), and QTc interval (ms) (the corrected QT interval calculated by Bazett’s formula). All ECG parameters were measured manually by 3 independent investigators.

Preoperative transthoracic echocardiograms were screened to measure left atrial (LA) diameter, left ventricular dimensions (LVDd, LVS, IVS), left ventricular ejection fraction (LVEF), and systolic pulmonary artery pressure (sPAP). These measurements were performed according to American Society of Echocardiography (ASE) guidelines [3].

All statistical analyses were performed using SPSS version 25.0 (IBM, Armonk, NY, USA). Continuous variables are expressed as means±standard deviations and were compared between groups using the independent t test for normally distributed data or the Mann-Whitney U test for non-normally distributed data. Categorical variables are expressed as frequencies and percentages and were compared using the chi-square test. Univariate analyses were initially conducted to identify variables associated with postoperative AF. Variables with a P value <0.1 in univariate analysis were entered into a multivariate logistic regression model to identify independent predictors of postoperative AF. Statistical significance was defined as a P value <0.05.

Results

The baseline characteristics of the study population are presented in Table 1. The 110 study patients divided into 2 groups: the POAF group and the non-POAF group (34.5%, n: 38 vs 65.4, n: 72, respectively). Basic characteristics, including age, sex, BMI, and smoking, of the 2 groups were similar. Comorbidities, such as DM, HT, chronic obstructive pulmonary disease (COPD), previous percutaneous coronary intervention (PCI), and previous stroke, showed no significant differences between the groups.

Liver-related variables, including ascites, encephalopathy, and variceal bleeding, were not significantly different between the 2 groups. Additionally, MELD and Child-Pugh scores, which are the most common prognostic scores in end-stage liver diseases, were not significantly different between the 2 groups.

Table 2 shows a comparison of ECG, echocardiographic, and postoperative outcomes between the 2 groups. Patients in the POAF group had significantly higher heart rates (81.16±18.62 bpm vs 65.11±12.47 bpm, P<0.01) and longer maximal P-wave durations (169.47±20.41 ms vs 145.06±33.99 ms, P<0.01). Maximal P-wave duration, PR interval, QRS duration, and QTc interval were also significantly longer in the POAF group compared to the non-POAF group. P-wave peak time in lead II (PWPT-II) was significantly longer in the POAF group. There was no significant difference in PWPT-V1 and minimal P-wave duration (P: 0.13 and P: 0.07, respectively).

Parameters of preoperative transthoracic echocardiogram, including LVEF, left ventricular diastolic diameter (LVDd), LA diameter (LA), and sPAP, were similar between the 2 groups.

Postoperatively, renal failure, serum creatinine level, and malignancy rates were comparable between the 2 groups. In-hospital mortality was significantly higher in the POAF group compared to the non-POAF group (38.8% vs 8.3%, P<0.01).

Discussion

In preoperative cardiac assessments before LT, clinicians tend to focus primarily on detecting ischemic heart disease. Consequently, diagnostic tools for ischemia, such as echocardiography, myocardial perfusion scintigraphy (MPS), or dobutamine stress testing, are often prioritized and evaluated with greater precision. However, as our study demonstrates, POAF is not only associated with significant postoperative morbidity but is also closely linked to postoperative mortality.

The incidence of POAF varies depending on the type of surgery and patient characteristics. Incidence rates of approximately 60% have been reported in cardiac and thoracic surgeries, as a result of direct manipulation of the pericardium and stimulation of cardiac innervation [4]. In LT recipients, the incidence is 7.9–10% [5,6]. The incidence of POAF in our study was 34.5%, higher than previously been reported. Most patients with end-stage liver disease have advanced systemic diseases with impairment of autonomic functions and strong inflammatory processes. As a result of the autonomic dysfunction and subsequent dysregulation of the autonomic nervous system, they have increased vulnerability to POAF [7].

Multifactorial mechanisms of POAF development in LT patients are imbalance of autonomic tone, electrolyte shifts, metabolic disturbances, and hemodynamic stress. Perioperative surgical stress, anesthesia, and fluid shifts can worsen this autonomic imbalance. The anhepatic phase of the transplant or postoperative hemodynamic instability increase the demand for inotropic medications to maintain systemic vascular resistance and cardiac output. A heightened adrenergic state has been demonstrated to shorten the atrial refractory period and cause POAF [8]. The introduction of acidotic, hyperkalemic blood into the circulation during reperfusion also further stimulates a sympathetic response [9]. Additionally, reperfusion is associated with significant hemodynamic instability, including rapid falls in both blood pressure and systemic vascular resistance, due to released ischemic by-products and vasodilatory compounds, which can stimulate arrhythmogenic mechanisms. Prolongation of the duration of the atrial action potential due to electrolyte disturbances, including hypocalcemia, hypo- or hyperkalemia, and metabolic acidosis, are common during LT. Similarly, cold preservation solutions administered to the liver graft can contribute to atrial stretching and electrical instability [10].

Time of conduction from the sinoatrial node to the atrioventricular node affects P-waves on ECG. In patients with AF, both atria have structural and electrical remodeling. Therefore, several P-wave parameters, such as P-wave terminal force (PWTF), P-wave duration (PWD), PWPT-II, PWPT-V1, PR duration, and P-wave duration have been investigated in patients diagnosed with AF. All of these parameters have been investigated in small-sample studies, both supporting and not supporting their significance [11–16]. There are few studies investigating PWPT-II’s positive correlation with coronary artery disease and AF. For instance, Öz et al demonstrated that prolonged PWPT-II and PWPT-V1 are associated with paroxysmal AF in patients presenting with acute ischemic stroke [17]. Our study supports that PWPT-II prolongation is significantly associated with increased POAF risk in LT patients.

As observed in the study by Kreimer et al, our post-LT patients demonstrated a significant correlation between POAF and P-wave duration, without a significant relationship between POAF and LA diameter [18]. With these consistent findings, we may consider that changes or abnormalities in P-wave indices are significant indicators of atrial conduction delay, even before the development of LA enlargement.

Our findings emphasize that preoperative evaluation for non-cardiac surgery should not be limited to the assessment of ischemic heart disease or the regulation of medications. For instance, a comprehensive perioperative approach is needed to reduce the risk of POAF after LT, which could, in turn, help minimize major adverse postoperative outcomes, such as mortality. Considering the mechanisms mentioned above, preoperative beta-blocker optimization may suppress sympathetic overactivation and reduce arrhythmogenic risk in this specific population. Although beta-blockers are widely used in cirrhotic patients for portal hypertension, their role in reducing POAF is not well understood and clinicians may lack information on optimizing the dosage. Strict perioperative fluid management is very important to minimize electrolyte and volume imbalances, which are significant triggers for atrial stretching and POAF [5].

Conclusions

LIMITATIONS:

The study has several limitations. Firstly, its small sample size and retrospective design are major limitations. However, given that this study was conducted in a highly specific patient population with limited data, the sample size is acceptable. Nonetheless, larger-scale, prospective studies are needed to confirm these results. Secondly, automated ECG measurements could be better than the manual method, but we attempted to overcome this limitation through manual measurements conducted independently by 3 researchers. Finally, we did not explore the possible association of POAF duration with outcomes.