Abstract
Background:
Atrial arrhythmias (AAs) are frequent after lung transplantation (LT) and late postoperatively. Several predictive factors of early postoperative AAs after LT have been identified but those of late AAs remain unknown. Whether AA after LT affects mortality is still being debated. This study assessed in a large cohort of LT patients the incidence of AAs early and late after surgery, their predictive factors and their effect on mortality.
Methods and Results:
We studied 271 consecutive LT patients over 9 years. Mean follow-up was 2.9±2.4 years. 33% patients developed postoperative AAs. Age (odds ratio (OR) 2.35; confidence interval (CI) [1.31–4.24]; P=0.004) and chronic obstructive pulmonary disease (OR 2.13; CI [1.12–4.03]; P=0.02) were independent predictive factors of early AAs. Late AAs occurred 2.2±2.7 years after transplant in 8.8% of the patients. Pretransplant systolic pulmonary arterial pressure (PTsPAP) was the only independent predictive factor of late AA (OR 1.028; CI [1.001–1.056]; P=0.04). Double LT was associated with long-term freedom from atrial fibrillation (AF) but not from atrial flutter (AFL). Early and late AAs after surgery had no effect on mortality. Double LT was associated with better survival.
Conclusions:
Early AA following LT is common in contrast with the low occurrence of late, often organized, AA. Early and late AAs do not affect mortality. PTsPAP is an independent predictor of late AA. Double LT protects against late AF but not AFL.
Atrial arrhythmias (AAs) are commonly observed immediately after thoracic surgery and may occur in up to 20–40% of procedures.1–3
Postoperative AAs are related to age, hypertension, stress, use of inotropes or postoperative pain, and are associated with longer hospital stay and deaths.1,4
The incidence of AA is higher after lung transplantation (LT) than after heart transplantation, in part because of cardiac autonomic denervation.5
LT patients are usually younger with less cardiovascular risk factors than patients undergoing other thoracic surgery.6
AAs are observed early and also late after LT. Double LT is associated with long-term freedom from paroxysmal atrial fibrillation (AF).7 Atrial tachycardias, focal or macroreentrant are probably related to surgical lines around the pulmonary veins (PVs) between the donor and the recipient left atrium. Several predictive factors of postoperative AA after LT have already been identified whereas those of late AA are still unknown. Whether AA after LT has an effect mortality and the optimal strategy of rhythm management are still being debated.7–9
The purpose of our study was to assess in a large cohort of LT patients the incidence, management and outcome of AA early and late after surgery, the predictive factors and effect on mortality.
Methods
Study Design
All consecutive patients who underwent LT between January 2005 and December 2014 were prospectively followed at the Nouvel Hôpital Civil of Strasbourg, France and were included in this study. Heart-lung transplantations were excluded. This study was conducted under a general patient record review protocol approved by the Comité de Protection des Personnes Ouest 1-Tours (reference no. 2009-A00036-51). All patients gave informed written consent to participate in the clinical study.
Data Collection
Preoperative data included patient and donor characteristics, indication for transplantation, cardiovascular risk factors, pretransplant arrhythmias or structural heart disease (coronary artery disease or left ventricular ejection fraction ≤40%), preoperative transthoracic echocardiography, right heart catheterization and coronary angiography, and medications (antiarrhythmics, inhaled bronchodilators, immunosuppressors, anticoagulants). In the postoperative period, the presence of ST elevation, troponin peak, duration of mechanic ventilation, and length of hospital stay were evaluated.
During follow-up, recurrence of any form of AA was recorded from the time of surgery to the end of follow-up. Further outcome events, including graft rejection and thromboembolic events, were recorded.
All patients had continuous cardiac monitoring for at least 30 days postoperatively in the intermediate care unit (ICU). Patients were then followed-up initially once weekly for 3 three months, then on a monthly-basis until 1 year postoperatively and every 3 months thereafter.
Surgery
A single-center transplantation team performed all the interventions. There was no change in surgical technique during the study period. The PVs were removed from the donor heart, including an atrial muscular skirt. The donor right and left PVs were then separated, and excess tissue was trimmed. Each side was sutured on the recipient atrium cross-clamped. The need for cardiopulmonary bypass during surgery depended on surgeon choice. The following operative information was collected: cold ischemic time for each lung, conditions of high emergency allocation, perioperative arrhythmias, need for cardiopulmonary bypass during surgery or extracorporeal membrane oxygenation support (ECMO) after transplantation, need for plasty of the left atrium (because of an excessively short muscular skirt), and use of epidural catheter for pain management.
Definitions
Early AAs were defined as any occurrence of postoperative AF or atrial flutter (AFL) documented by 12-lead ECG or telemetry monitoring during the first postoperative month. Late AAs were defined as AF or AFL occurring after 1 month post-transplantation and documented by 12-lead ECG or ambulatory ECG monitoring.
All ECGs were reviewed by 2 cardiologists, including 1 electrophysiologist, and the rhythm was classified as sinus, AF, typical AFL, or atypical AFL. Typical AFL was defined by organized inverted flutter waves in the inferior leads (with fixed cycle length) and positive in V1. Atypical AFL was defined by organized P-wave activity that did not fulfill the criteria for typical AFL.
Statistical Analysis
Categorical variables are expressed as count and percentages. Continuous variables are reported as mean and SD. Categorical variables were compared with chi-square test or Fisher’s exact test. Continuous variables were compared with the use of the Mann-Whitney test. To determine predictors of early and late AAs, and death, Cox-regression analysis was performed. Variables with P<0.1 in the univariate analysis were entered into the Cox multivariate regression analysis. Survival was graphically displayed according to the Kaplan-Meier method with comparison of cumulative survival rate by the log-rank test. All tests were two-sided. P<0.05 was considered significant. Calculations were performed using SPSS 17.0 (SPSS Inc., Chicago, IL, USA).
Results
Patients’ Characteristics
Over a 9-year period, 271 LT patients were followed for a mean of 2.9±2.4 years; 9 patients died during surgery, and 76 patients died during follow-up. The mean age at time of LT was 49.5±14.5 years and 61% were male.
Double LT was performed in 82% of cases whereas single LT was done in 18% of cases, mostly before 2009. Baseline characteristics are outlined in
Table 1. Indications for transplantation are summarized in
Table 2.
Table 1.
Characteristics of Patients With and Without Early Postoperative AAs
| Characteristics |
All patients
(n=271) |
No early AA
(n=182) |
Early AA
(n=89) |
P value |
| Patient demographics |
| Age (years) |
49±14 |
45±15 |
56±10 |
<0.001* |
| BMI (kg/m2) |
21±4 |
20±4 |
22±4 |
0.003* |
| Thoracic perimeter (cm) |
91±12 |
89±12 |
96±12 |
<0.001* |
| Hypertension |
68 (25.1%) |
34 (18.7%) |
34 (38.2%) |
0.002* |
| Smoking habit |
157 (57.9%) |
92 (50.5%) |
65 (73.0%) |
0.004* |
| Pretransplant AF |
19 (7.0%) |
9 (4.9%) |
10 (11.2%) |
0.095 |
| Structural heart disease |
21 (7.7%) |
9 (4.9%) |
12 (13.5%) |
0.027* |
| Right heart catheterization |
| mPAP (mmHg) |
26±9 |
26±9 |
25±9 |
0.634 |
| RAP (mmHg) |
3.9±3.6 |
3.5±2.6 |
4.6±4.2 |
0.043* |
| Echocardiography |
| LVEF (%) |
64±8 |
63±9 |
64±9 |
0.754 |
| LA size (cm2) |
15±5 |
16±5 |
15±5 |
0.276 |
| sPAP (mmHg) |
42±16 |
41±15 |
46±19 |
0.063 |
| RV dilation (%) |
57 (21.0%) |
35 (19.2%) |
22 (24.7%) |
0.774 |
| Transplantation |
| Double lung transplantation |
223 (82.3%) |
153 (84.0%) |
70 (78.6%) |
0.076 |
| Ischemic time (mn) |
329±77 |
328±81 |
330±70 |
0.304 |
| Peak troponin (μg/L) |
7.4±12 |
8.2±15 |
5.9±5 |
0.137 |
| Postoperative ECMO |
13 (4.8%) |
9 (5.5%) |
4 (4.5%) |
0.754 |
| Epidural catheter |
103 (38.0%) |
64 (35.2%) |
39 (43.8%) |
0.226 |
| Mechanical ventilation duration (days) |
6.4±12 |
7.4±13 |
5.1±10 |
0.142 |
| Hospital length stay (days) |
51±50 |
50±47 |
55±57 |
0.451 |
| Acute lung rejection |
84 (30.9%) |
56 (30.7%) |
28 (31.4%) |
0.750 |
| In-hospital death |
19 (7.0%) |
14 (7.7 %) |
5 (5.6%) |
0.450 |
P<0.05 is significant. AAs, atrial arrhythmias; AF, atrial fibrillation; ECMO, extracorporeal membrane oxygenation support; LVEF, left ventricular ejection fraction; mPAP, mean pulmonary arterial pressure; RAP, right atrial pressure; sPAP, systolic pulmonary arterial pressure; RV, right ventricular.
Table 2.
Indications for LT
| Indication |
Age
(years) |
All patients
(n=271) |
No early AA
(n=182) |
Early AA
(n=89) |
P value |
| COPD |
58±7 |
82 (30.3%) |
46 (25.3%) |
36 (40.4%) |
0.016* |
| Cystic fibrosis |
28±7 |
57 (21.0%) |
50 (27.5%) |
7 (7.9%) |
0.0001* |
| Idiopathic fibrosis |
59±6 |
54 (19.9%) |
33 (18.1%) |
21 (23.6%) |
NS |
| Emphysema |
54±8 |
29 (10.7%) |
16 (8.8%) |
13 (14.6%) |
NS |
| Bronchiectasis |
52±12 |
9 (3.3%) |
6 (3.3%) |
3 (3.4%) |
NS |
| Bronchiolitis obliterans |
34±11 |
6 (2.2%) |
5 (2.7%) |
1 (1.1%) |
NS |
| α1 antitrypsine deficit |
52±12 |
6 (2.2%) |
4 (2.2%) |
2 (2.2%) |
NS |
| Primary pulmonary hypertension |
61±2 |
6 (2.2%) |
4 (2.2%) |
2 (2.2%) |
NS |
| Sarcoidosis |
47±7 |
5 (1.8%) |
4 (2.2%) |
1 (1.1%) |
NS |
| Interstitial pneumonia |
51±4 |
5 (1.8%) |
3 (1.6%) |
2 (2.2%) |
NS |
| Langerhans cell histiocytosis |
45±10 |
4 (1.5%) |
4 (2.2%) |
0 |
NS |
| Hypersensitivity pneumonitis |
40±15 |
3 (1.1%) |
3 (1.6%) |
0 |
NS |
| Retransplantation |
42±11 |
5 (1.8%) |
4 (2.2%) |
1 (1.1%) |
NS |
COPD, chronic obstructive pulmonary disease; LT, lung transplantation.
In the study, 89 (33%) patients developed early postoperative AAs. Of these patients, 79 (89%) presented with AF and 10 (11%) with AFL (53% typical aspect). The mean time to early postoperative AA onset after LT was 7±8 days.
The early AA group was significantly older with higher body mass index (BMI) and thoracic perimeters. Patients with early AAs had more hypertension, structural heart disease, smoking history and higher right atrial pressure. There were age differences among patients with various pulmonary diseases before LT. Cystic fibrosis patients were younger with less early AAs.
Echocardiographic and transplantation data were not significantly different between the 2 groups. Early AF occurred in 28% of double LT and in 35% of single LT patients (P=NS). Postoperative length of hospital stay was not longer in the early AA group.
Of all the patients who presented with early AAs, 92% received treatment for rhythm control. Amiodarone was used in 78% of the patients according to hospital protocol (typically 900 mg intravenous loading dose over 1 day followed by a tapering dose and an oral maintenance dose of 200 mg); 10% of patients underwent electrical cardioversion. One patient underwent cavotricuspid isthmus ablation. Antiarrhythmia treatment was ceased after a mean of 70±12 days. Heparin and vitamin K antagonist (fluindione) were initiated in 48% of the patients for a mean duration of 3.4±2.2 months. No major bleeding complications were recorded. At hospital discharge, all patients except 1 were in sinus rhythm. No pulmonary complications linked to amiodarone use were reported.
On univariate analysis, age, BMI, thoracic perimeter, hypertension, smoking habits, structural heart disease and chronic obstructive pulmonary disease (COPD) were associated with an increased risk of early AA following LT (Table 3).
Table 3.
Predictors of Early AA After LT
| Characteristics |
Univariate analysis |
Multivariate analysis |
| OR |
95% IC |
P value |
OR |
95% IC |
P value |
| Age |
1.057 |
1.035–1.079 |
<0.001* |
2.353 |
1.306–4.240 |
0.004* |
| BMI |
1.070 |
1.022–1.119 |
0.004* |
0.975 |
0.889–1.058 |
NS |
| Thoracic perimeter |
1.038 |
1.018–1.059 |
<0.001* |
1.031 |
0.996–1.068 |
NS |
| Hypertension |
1.922 |
1.256–2.943 |
0.003* |
1.346 |
0.791–2.289 |
NS |
| Smoking habit |
1.963 |
1.211–3.182 |
0.006* |
0.757 |
0.335–1.711 |
NS |
| Pretransplant AA |
1.733 |
0.897–3.348 |
0.102 |
|
|
|
| Acute rejection |
0.920 |
0.443–1.909 |
0.822 |
|
|
|
| LA size (cm2) |
0.965 |
0.868–1.072 |
0.507 |
|
|
|
| Structural heart disease |
1.848 |
1.005–3.399 |
0.048* |
1.274 |
0.616–2.634 |
NS |
| COPD |
1.747 |
1.149–2.656 |
0.009* |
2.126 |
1.121–4.030 |
0.021* |
LA, left atrial. Other abbreviations as in Tables 1,2.
Conversely, pretransplant AA, systolic pulmonary artery pressure (sPAP), left atrial size, troponin peak, acute rejection and double LT were not associated with an increased risk of early AAs. The use of epidural catheter for pain management was not associated with a lower risk of early AAs.
On multivariate analysis, independent predictors of early AAs were age (odds ratio [OR] 2.35; confidence interval [CI] (1.31–4.24); P=0.004) and COPD (OR 2.13; CI (1.12–4.03); P=0.02).
Late AAs
Out of 271 (8.8%) patients, 24 presented with AAs late after transplant. Only 3/24 (12.5%) patients had a history of AF before transplantation; 15/24 (62.5%) patients had also sustained early postoperative AAs. Of these 24 patients, 11 presented with AF and 13 with AFL (4 typical AFL, 6 atypical AFL, 3 that were only documented on ambulatory ECG monitoring and were undifferentiated).
Late AAs occurred 2.2±2.7 years after LT (2.7±2.7 years in AF group, 1.7±1.6 years in AFL group; P=NS). Among the patients presenting with AFL, 11/13 (85%) were double LT patients. Among the patients presenting with AF, 45% (5/11) were single LT, and 55% (6/11) were double LT.
Among all single LT patients, 10.4% presented with late AF vs. 2.7% in the double LT group (P=0.014). There was no difference in the rate of AFL between single LT and double LT patients (4.2% vs. 5%, respectively; P=NS).
The Kaplan-Meier analysis for the development of late AF demonstrated that double LT was associated with long-term freedom from AF compared with single LT (P=0.05), but there was no significant difference for the development of late AFL in the 2 groups of LT patients (Figure 1).
Of the patients who developed late AF, 10 (91%) were treated with antiarrhythmic drugs (8 with amiodarone, 1 with sotalol and 1 with flecainide). In 1 patient cardioversion was required.
Of the patients who developed late AFL, 11 (85%) were treated with antiarrhythmic drugs (6 amiodarone, 1 sotalol, 4 flecainide+β-blockers). Cardioversion was required in 6 cases; 4 patients were referred for an electrophysiology study: in 3 cases, counterclockwise tricuspid isthmus-dependent flutter was diagnosed, with successful ablation, and 1 case of a left-sided flutter for which the patient refused left atrial ablation and subsequent cardioversion was performed.
On univariate analysis, age, BMI, structural heart disease, sPAP, and early AA were associated with an increased risk of AAs late after LT. Conversely, pretransplant AF, single LT or double LT, and acute and chronic transplant rejection were not significantly associated with AAs late after LT.
On multivariate analysis, the only predictor of AAs late after LT was sPAP before transplant (OR 1.028; CI [1.001–1.056]; P=0.04) as measured by transthoracic echography (Table 4).
Table 4.
Predictors of Late AA After LT
| Characteristics |
Univariate analysis |
Multivariate analysis |
| OR |
95% CI |
P value |
OR |
95% CI |
P value |
| Age |
1.061 |
1.019–1.105 |
0.004* |
1.022 |
0.970–1.076 |
NS |
| BMI |
1.093 |
1.010–1.183 |
0.0028* |
1.055 |
0.952–1.168 |
NS |
| Pretransplant AA |
1.950 |
0.580–6.554 |
0.280 |
|
|
|
| Acute rejection |
0.895 |
0.383–2.095 |
0.799 |
|
|
|
| Chronic rejection |
1.931 |
0.828–4.499 |
0.128 |
|
|
|
| Structural heart disease |
3.874 |
1.427–10.514 |
0.008* |
1.226 |
0.359–4.184 |
NS |
| sPAP TTE |
1.038 |
1.013–1.063 |
0.002* |
1.028 |
1.001–1.056 |
0.041 |
| Early AA |
36.463 |
12.025–110.565 |
<0.001* |
1.507 |
0.918–2.473 |
NS |
sPAP TTE, systolic pulmonary arterial pressure with transthoracic echocardiography. Other abbreviations as in Tables 1,2.
In the overall cohort, in-hospital death occurred in 19 patients (7,0%). No increase in in-hospital deaths was observed in the subset of patients with early postoperative AAs (5.6% vs 7.7%; P=0.45).
On univariate analysis, hypertension, sPAP, need of ECMO support following surgery, high emergency transplant and troponin peak after surgery were associated with an increased risk of in-hospital death. On multivariate analysis, sPAP and high emergency transplant were independent predictors of in-hospital death (OR respectively 1.048; 95% CI [1.003–1.094]; P=0.035 and 7.418; 95% CI [1.188–46.325]; P=0.032) (Table 5).
Table 5.
Predictors of In-Hospital and Long-Term Mortality After LT
| Characteristics |
Univariate analysis |
Multivariate analysis |
| OR |
95% CI |
P value |
OR |
95% CI |
P value |
| In-hospital mortality |
| Age |
1.026 |
0.993–1.061 |
0.120 |
|
|
|
| sPAP TTE |
1.032 |
1.003–1.061 |
0.028* |
1.048 |
1.003–1.094 |
0.035* |
| LVEF |
1.099 |
1.033–1.170 |
0.003* |
1.106 |
0.995–1.229 |
NS |
| Double-lung transplantation |
0.548 |
0.216–1.388 |
0.205 |
|
|
|
| Hypertension |
2.889 |
1.272–6.562 |
0.011* |
|
|
|
| Early AA |
0.737 |
0.254–2.138 |
0.575 |
|
|
|
| ECMO support |
18.852 |
5.788–61.404 |
<0.001* |
13.440 |
0.770–234.586 |
NS |
| High emergency |
6.177 |
2.568–14.860 |
<0.001* |
7.418 |
1.188–46.325 |
0.032* |
| Troponine peak |
1.074 |
1.025–1.125 |
0.003* |
1.006 |
0.904–1.120 |
NS |
| Epidural catheter |
0.387 |
0.124–1.209 |
0.102 |
|
|
|
| Ventilation duration |
1.068 |
1.021–1.116 |
0.004* |
|
|
|
| Hospital length of stay |
0.988 |
0.960–1.017 |
0.425 |
|
|
|
| ICU length of stay |
1.004 |
0.996–1.011 |
0.317 |
|
|
|
| Long-term mortality |
| Age |
1.020 |
1.002–1.038 |
0.032* |
1.441 |
0.703–2.955 |
NS |
| BMI |
1.053 |
1.005–1.103 |
0.031* |
1.032 |
0.963–1.105 |
NS |
| Early AA |
1.507 |
0.918–2.473 |
0.105 |
|
|
|
| AA at follow-up |
0.996 |
0.624–1.589 |
0.986 |
|
|
|
| Epidural catheter |
0.506 |
0.266–0.962 |
0.038* |
0.645 |
0.305–1.362 |
NS |
| ECMO support |
5.307 |
2.812–10.016 |
<0.001* |
4.260 |
1.677–10.823 |
0.002* |
| Troponin peak |
1.018 |
1.008–1.029 |
0.001* |
1.014 |
0.975–1.054 |
NS |
| Mechanical ventilation duration |
1.055 |
1.029–1.082 |
<0.001* |
|
|
|
| ICU length of stay |
1.004 |
1.002–1.007 |
0.002* |
|
|
|
| Hospital length of stay |
1.005 |
1.003–1.008 |
<0.001* |
1.011 |
1.002–1.019 |
0.017* |
| Double-lung transplantation |
0.453 |
0.281–0.731 |
0.001* |
0.466 |
0.230–0.943 |
0.034* |
| High emergency |
2.767 |
1.602–4.780 |
<0.001* |
1.803 |
0.868–3.745 |
NS |
ICU, intermediate care unit. Other abbreviations as in Tables 1,4.
Univariate predictors of long-term death during follow-up were age, BMI, ECMO after surgery, troponin peak, mechanical ventilation, ICU length of stay, hospital length of stay and high emergency. In contrast, use of epidural catheter for pain management and double LT were associated with a lower long-term mortality risk (Table 5).
On multivariate analysis, independent predictors of increased long-term mortality were need for ECMO after surgery and hospital length of stay. Conversely, double LT was identified as an independent predictor of long-term survival (OR 0.466; 95% CI [0.230–0.943]; P=0.034) (Table 5). The Kaplan-Meier analysis for survival demonstrates a beneficial effect of double LT (Figure 2).
There was no significant effect of early postoperative AAs and late AAs after surgery on long-term mortality (Figure 3).
Discussion
In the present study, early postoperative AA was observed to occur frequently in up to 33% of cases, and was not associated with adverse outcomes. Postoperative AA was more likely to occur in older patients with COPD. AA late after LT was uncommon (8.8%). Double LT seemed to protect against late AF but not against late AFL recurrence. Pretransplant sPAP was the only predictive factor of late AA after surgery. Early or late AAs after surgery were not associated with increased mortality. Double LT was associated with better survival.
Whether double LT could protect against atrial arrhythmia recurrence was suggested by Lee et al.7
In their study, AF occurred in 5.5% (17/327) of patients 2.9 years after transplant. Organized AA was also identified in addition to AF. In our study, the high incidence of early postoperative AA contrasts with the low occurrence of late AA (8.8%), consistent with previous findings. Our observation substantiates the view of a bimodal distribution of AA following LT, as originally described by See et al2
who also depicted the greater proportion of organized atrial tachycardia compared with AF more than 1 year after transplant.
The surgical technique for LT involves surgical antral isolation of the PVs. PV isolation is the recommended approach for treatment of paroxysmal AF.10
Only one of 200 double LT patients in the study by Lee et al, and 0 of 127 in the study by See et al had AF recurrence. Similarly, we found that double LT was protective against AF recurrence but not against AFL. We observed double LT patients with late AF. This can be explained by AF triggers located outside the PVs, for example, elsewhere in the PV antrum (extent being dependent on surgery) or in the left or right atrium.11
Electrical PV reconnection across anastomotic suture lines or microreentry at the suture sites could also explain the recurrences.12
Among the late AFL recurrences, 85% were in double LT patients. As the PVs are isolated from surgery, macroreentry can occur. Typical and atypical flutters were observed. Only 4 patients underwent electrophysiological study, and 3 CTI-dependent flutters and 1 left AFL were diagnosed. See et al also described 4 patients who underwent electrophysiological study, with 1 CTI, 2 left-sided macroreentrant tachycardia, and 1 focal atrial tachycardia being diagnosed.2
Catheter ablation may indeed be useful in this category of LT patients with recurrent AFL.
Interestingly, in our cohort the only predictive factor for late AA was raised PAP before LT, which theoretically could also predispose to CTI-dependent flutter. In the study by See et al no predictive factors for late AA could be identified. Several studies have shown that pulmonary arterial hypertension induces right atrial remodeling, predisposing to supra-ventricular tachycardia.13,14
In contrast to heart transplantation, where AF is associated with rejection and myocardial dysfunction.5
in LT there is no observable relationship between acute or chronic rejection and AA.
If the incidence of late AA after transplant is low and predictive factors difficult to identify, conversely postoperative AA is more common. Age is regularly identified as a predictive factor.2,4,6
The incidence of early postoperative AA in our LT population was 33%. Nielsen et al reported a higher incidence of 39% in a cohort of similar age with similar proportion of double LT (82% vs. 79%).1
Henri et al6
reported a lower incidence of 29%, which was explained by a younger age and a lower proportion of double LT (56%). Henri et al initially observed that double LT was predictive of postoperative AF.6
After a statistically adjusted model using the year of transplant, they also found that double LT was not any more predictive of postoperative AF on multivariate analysis. In our study, double LT was not associated with postoperative AA. Lee et al also found that early postoperative AF occurred in equal proportions in single LT and double LT populations.7
Arrhythmia pretransplant was also found to be predictive of postoperative AA.2,4
For Henri et al, these findings are dependent on the statistical model.6
It has been suggested that mechanisms favoring postoperative AF are multifactorial, and not dependent on PV triggers.2,15,16
Inflammation has been associated with AF17
and its implications following thoracic surgery has been extensively studied.18
Pericardial inflammation after surgery, variation in left and right atrial pressures, changes in oxygenation, pain and stress may influence atrial conduction and refractory periods, predisposing to AF. The factors related to thoracic surgery may decrease during the recovery phase, usually during the first month, and may relate to atrial electrical reverse remodeling. Postoperative AAs are mostly transient and 99% of the patients return to sinus rhythm before discharge. In our study, length of stay and in-hospital death were not increased in the early AA group. We found that early postoperative AA was neither a risk factor for in-hospital death nor for overall long-term mortality. These findings are corroborated by Henri et al who also found that postoperative AF had no impact on mortality. Similar to our study, the strategy of rhythm control was most frequently chosen (97%), with nearly all patients in sinus rhythm upon discharge. No amiodarone-related lung toxicity was noted. Conversely, Orrego et al4
chose a rhythm strategy in only 50% of patients and found a higher mortality in the rhythm control group (amiodarone/propafenone) compared with the rate control group. Isiadinso et al observed a negative effect of amiodarone on short-term mortality in the postoperative AF group, but the cohort comprised heart and lung transplant patients.9
Postoperative AF in heart transplant patients carries different clinical implications and is related to death and rejection.19
Recently, Gillinov et al compared rate vs. rhythm control for AF after cardiac surgery and showed no net clinical advantage of 1 treatment strategy over the other.20
In our experience, rhythm control strategy was used in the majority of patients without significant complications and could constitute a reasonable first-line approach for these patients.
In our study, early postoperative AA had no effect on in-hospital and long-term mortality. Late AA after LT was also not associated with increased mortality. Predictive factors for in-hospital death were PAP and high emergency transplantation, reflecting the more serious clinical preoperative state of the patient. These factors are not frequently reported in the literature. In addition, Henri et al found that postoperative vasopressor use was predictive of death. In our study, hospital length of stay following transplantation and ECMO support were predictive of overall long-term mortality. Sabashnikov et al recently identified that intraoperative conversion to cardiopulmonary bypass and ECMO support was strongly associated with increased mortality.21
In our study, double LT was predictive of increased survival. Schaffer et al showed that double LT was associated with better graft survival than single LT in patients with idiopathic fibrosis, whereas in the COPD population, no survival difference between single LT and double LT recipients at 5 years could be evidenced.22
An ongoing debate remains between proponents of single LT and double LT for end-stage pulmonary disease.23
Study Limitations
There are several limitations that should be considered. This was a retrospective single-center study of patients who had uniform surgical procedures and postoperative management. Although the cohort was large, the retrospective analysis may have underestimated the prevalence of AAs. Subclinical episodes of late AA may have been missed even if particular care concerning arrhythmias were undertaken during the follow-up period.
Systolic PAP measured by TTE was predictive of late AA after transplantation but right heart catheterization parameters were not. The delay between catheterization, often done prior to registration of a patient on the LT waiting list, and transplantation can be long. TTE systematically done before transplantation may reflect the deterioration in patient state with the highest PAP recorded before transplantation.
Conclusions
Early postoperative AAs following LT are common and often transient, in contrast to the low occurrence of late and often organized AAs. Early and late AAs do not affect mortality following LT. Double LT compared with single LT, analogous to PV isolation, protected against late AF but not AFL, underlying the role of the PVs in the genesis of AF. Pretransplant sPAP was an independent predictor of late AA after surgery.
Disclosures
The authors do not have any relevant financial disclosures that might pose a conflict of interest in connection with the submitted article.
References
- 1.
Nielsen TD, Bahnson T, Davis RD, Palmer SM. Atrial fibrillation after pulmonary transplant. Chest 2004; 126: 496–500.
- 2.
See VY, Roberts-Thomson KC, Stevenson WG, Camp PC, Koplan BA. Atrial arrhythmias after lung transplantation: Epidemiology, mechanisms at electrophysiology study, and outcomes. Circ Arrhythm Electrophysiol 2009; 2: 504–510.
- 3.
Mason DP, Marsh DH, Alster JM, Murthy SC, McNeill AM, Budev MM, et al. Atrial fibrillation after lung transplantation: Timing, risk factors, and treatment. Ann Thorac Surg 2007; 84: 1878–1884.
- 4.
Orrego CM, Cordero-Reyes AM, Estep JD, Seethamraju H, Scheinin S, Loebe M, et al. Atrial arrhythmias after lung transplant: Underlying mechanisms, risk factors, and prognosis. J Heart Lung Transplant 2014; 33: 734–740.
- 5.
Dizon JM, Chen K, Bacchetta M, Argenziano M, Mancini D, Biviano A, et al. A comparison of atrial arrhythmias after heart or double-lung transplantation at a single center: Insights into the mechanism of post-operative atrial fibrillation. J Am Coll Cardiol 2009; 54: 2043–2048.
- 6.
Henri C, Giraldeau G, Dorais M, Cloutier AS, Girard F, Noiseux N, et al. Atrial fibrillation after pulmonary transplantation: Incidence, impact on mortality, treatment effectiveness, and risk factors. Circ Arrhythm Electrophysiol 2012; 5: 61–67.
- 7.
Lee G, Wu H, Kalman JM, Esmore D, Williams T, Snell G, et al. Atrial fibrillation following lung transplantation: Double but not single lung transplant is associated with long-term freedom from paroxysmal atrial fibrillation. Eur Heart J 2010; 31: 2774–2782.
- 8.
Malik A, Hsu JC, Hoopes C, Itinarelli G, Marcus GM. Elevated pulmonary artery systolic pressures are associated with a lower risk of atrial fibrillation following lung transplantation. J Electrocardiol 2013; 46: 38–42.
- 9.
Isiadinso I, Meshkov AB, Gaughan J, Sandhu P, Lim S, Cordova F, et al. Atrial arrhythmias after lung and heart-lung transplant: Effects on short-term mortality and the influence of amiodarone. J Heart Lung Transplant 2011; 30: 37–44.
- 10.
Haïssaguerre M, Jaïs P, Shah DC, Takahashi A, Hocini M, Quiniou G, et al. Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N Engl J Med 1998; 339: 659–666.
- 11.
Haissaguerre M, Hocini M, Denis A, Shah AJ, Komatsu Y, Yamashita S, et al. Driver domains in persistent atrial fibrillation. Circulation 2014; 130: 530–538.
- 12.
Roten L, Pedersen M, Pascale P, Shah A, Eliautou S, Scherr D, et al. Noninvasive electrocardiographic mapping for prediction of tachycardia mechanism and origin of atrial tachycardia following bilateral pulmonary transplantation. J Cardiovasc Electrophysiol 2012; 23: 553–555.
- 13.
Zhao Q, Deng H, Jiang X, Dai Z, Wang X, Wang X, et al. Effects of intrinsic and extrinsic cardiac nerves on atrial arrhythmia in experimental pulmonary artery hypertension. Hypertension 2015; 66: 1042–1049.
- 14.
Wen L, Sun ML, An P, Jiang X, Sun K, Zheng L, et al. Frequency of supraventricular arrhythmias in patients with idiopathic pulmonary arterial hypertension. Am J Cardiol 2014; 114: 1420–1425.
- 15.
Gokalp O, Eygi B, Besir Y, Gokalp G, Gurbuz A. Intraoperative factors cause postoperative atrial fibrillation? Circ J 2016; 80: 1052.
- 16.
Takahashi S, Sueda T. Intraoperative factors cause postoperative atrial fibrillation?: Reply. Circ J 2016; 80: 1053.
- 17.
Aviles RJ, Martin DO, Apperson-Hansen C, Houghtaling PL, Rautaharju P, Kronmal RA, et al. Inflammation as a risk factor for atrial fibrillation. Circulation 2003; 108: 3006–3010.
- 18.
Jacob KA, Nathoe HM, Dieleman JM, van Osch D, Kluin J, van Dijk D. Inflammation in new-onset atrial fibrillation after cardiac surgery: A systematic review. Eur J Clin Invest 2014; 44: 402–428.
- 19.
Ahmari SAL, Bunch TJ, Chandra A, Chandra V, Ujino K, Daly RC, et al. Prevalence, pathophysiology, and clinical significance of post-heart transplant atrial fibrillation and atrial flutter. J Heart Lung Transplant 2006; 25: 53–60.
- 20.
Gillinov AM, Bagiella E, Moskowitz AJ, Raiten JM, Groh MA, Bowdish ME, et al. Rate control versus rhythm control for atrial fibrillation after cardiac surgery. N Engl J Med 2016; 374: 1911–1921.
- 21.
Sabashnikov A, Weymann A, Mohite PN, Zych B, Patil N, Garcia Saez D, et al. Risk factors predictive of one-year mortality after lung transplantation. Eur J Cardio-Thorac Surg 2014; 46: e82–e88.
- 22.
Schaffer JM, Singh SK, Reitz BA, Zamanian RT, Mallidi HR. Single- vs double-lung transplantation in patients with chronic obstructive pulmonary disease and idiopathic pulmonary fibrosis since the implementation of lung allocation based on medical need. JAMA 2015; 313: 936–948.
- 23.
Puri V, Patterson GA, Meyers BF. Single versus bilateral lung transplantation: Do guidelines exist? Thorac Surg Clin 2015; 25: 47–54.
