ORIGINAL ARTICLE
Reduced Incidence of Thromboembolic Events After Surgical Closure of Left Atrial Appendage in Patients With Atrial Fibrillation Manuel Wilbring, MD,* Friedrich Jung, MD,Þ Christoph Weber, MD,* Klaus Matschke, MD,* and Michael Knaut, MD*
Objective: Most of the detected thrombi in patients with atrial fibrillation (AF) can be found in the left atrial appendage (LAA). Interventional LAA closure recently proved to be noninferior to warfarin therapy. Whether these results can be fully translated into surgical LAA closure remains unclear. Corresponding data are still lacking. The present observational study evaluated the impact of surgical LAA closure in patients with AF undergoing cardiac surgery on postoperative thromboembolic events. Methods: A prospective registry enrolled 398 patients with permanent AF undergoing cardiac surgery. Concomitant procedures were isolated surgical ablation (group I, n = 71), isolated LAA closure (group II, n = 44), and combined surgical ablation and LAA closure (group III, n = 196). The control group consisted of 87 patients without concomitant surgical ablation or LAA closure. One-year follow-up was completed in all patients. End points were thromboembolic events and death from any cause. Results: Clinical baseline characteristics were comparable among the groups. General hospital mortality was 5.5% and likewise differed not significantly. Postoperatively, mean (SD) CHAD2S2-VASc score of 3.5 (1.3) differed not significantly among the groups, indicating comparable thromboembolic risk. Follow-up referred to all hospital survivors (n = 376). Herein, overall incidence of thromboembolic events was 9.8% (n = 37), with an associated mortality of 41.0%. Patients with LAA closure alone or in combination with surgical ablation had a significantly reduced incidence of thromboembolic events (6.6% vs 20.5%, P G 0.01) and consecutively improved survival after 1 year of follow-up (7.0% vs 17.1%, P G 0.01). Conclusions: Left atrial appendage closure alone or in combination with surgical ablation was associated with a significantly reduced Accepted for publication October 31, 2015. From the *Department of Cardiac Surgery, University Heart Center Dresden, Dresden, Germany; and †Institute for Polymer Research and BerlinBrandenburg Centre for Regenerative Therapies, Helmholtz-Zentrum Geesthacht, Berlin and Teltow, Germany. Presented at the Annual Scientific Meeting of the International Society for Minimally Invasive Cardiothoracic Surgery, May 30YJune 2, 2012, in Los Angeles, CA USA. Disclosure: The authors declare no conflicts of interest. Address correspondence and reprint requests to Manuel Wilbring, MD, Department of Cardiac Surgery, University Heart Center Dresden, Fetscherstrasse 76, 01307 Dresden, Germany. E-mail: [email protected]. Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery ISSN: 1556-9845/16/1101-0024
24
rate of thromboembolic events and consecutively improved survival after 1 year of follow-up. Key Words: Left atrial appendage, Atrial fibrillation, Surgical closure, Stroke, Surgical ablation. (Innovations 2016;11:24Y30)
A
trial fibrillation (AF) is the most common cardiac arrhythmia, which affects up to 2% of the general population.1,2 The most threatening and most disabling complication of AF is the ischemic stroke. Approximately 20% of all ischemic strokes are considered to be associated with AF.1 These data underline the general importance of AF. Actually, permanent oral anticoagulation is recommended in patients with a CHAD2S2-VASc score of 2 or greater.3 Nonetheless, permanent oral anticoagulation is associated with several problems including intolerance, noncompliance, and bleeding complications. Recent focus was set on the left atrial appendage (LAA), where up to 90% of the detected thrombi in patients with AF can be found.1,4 This gave the LAA the reputation to be ‘‘the most lethal human attachment.’’3 Therefore, closure of the LAA potentially could be the solution of the problem of AF and cardiac embolism. Recent studies tried to determine the effect of mechanical LAA closure on the incidence of stroke. The LAAOS trial, published in 2005, was the first randomized trial comparing different methods of LAA closureV epicardial suture or staplingVin patients undergoing coronary artery bypass grafting.5 Unfortunately, the LAAOS trial could not answer the intended question but revealed significant problems with the closure methods.5 Epicardial suture was only effective in 43% and stapling in 72%.5 The remaining patients were considered as treatment failures because of LAA reconnection or remnants greater than 1 cm.5 For those reasons, an optimized sequel study was initiated in 2013V the LAAOS II trial.6 This trial actually is still enrolling patients and hopefully will provide meaningful data to support further decision making. Until these data will become available, smaller monocentric series might remain the only source of knowledge. The proof of concept of LAA closure was delivered by interventional cardiologists. The recently published PROTECTAF trial was a prospective randomized trial, showing noninferiority Innovations & Volume 11, Number 1, January/February 2016
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
of the WATCHMAN device to warfarin therapy in the prevention of stroke.7 The most recent study, the PREVAIL trial, confirmed these findings and additionally revealed an improved procedural safety.8 These well-conducted studies implicated the pathogenesis of stroke in AF and confirmed mechanical closure of the LAA to be an effective approach. Nonetheless, these data cannot completely be translated into patients undergoing cardiac surgery, and data dealing with surgical patients are still scarce. For those reasons, the present observational study was initiated. With the present study, we sought to determine the clinical impact of surgical LAA closure on the incidence of stroke in a prospective, nonrandomized registry of patients with AF undergoing any cardiac surgery.
PATIENTS AND METHODS Patients and Study Design The study was designed as a prospective registry, which included all patients with permanent AF undergoing any elective cardiac surgery. Urgent procedures or emergencies were excluded. Preoperative, intraoperative, and postoperative data were extracted from the hospital’s database for subsequent analysis. Finally, a total of 398 patients were enrolled. The enrollment dated back to the period between 2005 and 2010. Because in this period, there existed no standardized protocol for patients with permanent AF, the decision if the patient received either LAA closure, surgical ablation, both, or none of these depended on the operating surgeon’s preference. Ablation was performed using microwave energy (until the Guidant devices were taken off the market) and subsequently cryoenergy. The modified left atrial line concept was applied as reported previously.9 The patients were divided into four groups depending on concomitant procedures as follows: Group I: only surgical ablation (SA, n = 71). Group II: only LAA closure (LAA, n = 44). Group III: combined surgical ablation and LAA closure (SA + LAA, n = 196). Control: no concomitant surgical ablation or LAA closure (control, n = 87). All patients were discharged with oral anticoagulation (warfarin), independently from the heart rhythm. Follow-up visits including 72-hour Holter-ECG were performed after 6 and 12 months. Standard antiarrhythmic treatment consisted of metoprolol. One-year follow-up was completed in all surviving patients. Primary study end points were incidence of thromboembolic events, death from any cause, and cardiac rhythm after 1 year of follow-up. Follow-up included direct interview of the patients and their cardiologist. Stroke and other thromboembolic events were defined as clinical apparent events, being revealed by computed angiography or other angiography in general. Comparative analyses were performed between groups I to III and control as well as between all patients receiving LAA closure (groups I and III) and the control group. Primary study end points were analyzed in reference to the hospital survivors. Accordingly, patients dying during
Thromboembolic Events After Surgical LAA Closure
their primary hospital stay were excluded from further followup and analyses. All patients gave informed consent. The study was reviewed and approved by the institutional review board at Medical Faculty ‘‘Carl Gustav Carus’’ at Technical University of Dresden.
Statistical Analysis Continuous data are expressed as means and SDs as well as absolute numbers and percentages for categorical variables. For analyzing continuous variables, unpaired t test was used; for comparing categorical variables, Fisher exact test was used. Time-to-event analyses were performed by Kaplan-Meier analysis and log-rank test. A P value of less than 0.05 was considered significant. Statistical analysis was performed using JMP 9.0 Software (SAS Institute Corp, Cary, NC USA). Power analysis was performed.
RESULTS Clinical Baseline Characteristics and Comparability of the Groups Mean (SD) age of all patients was 68 (9) years with a balanced sex distribution (male, n = 201, 50.5%). Risk factors for embolization according to the CHAD2S2-VASc score were balanced in between the groups. Likewise, the general surgical risk profile as predicted by additive EuroSCORE differed not significantly between all groups. Clinical baseline characteristics mainly showed no significant differences. Significant differences were recorded concerning the presence of coronary artery disease, being more frequent in the control group, and the presence of chronic kidney disease, which was more frequent in group II (only LAA closure). All remaining baseline characteristics did not differ significantly. The clinical baseline characteristics are summarized in Table 1.
Intraoperative Data and Hospital Outcome Mitral valve procedures were performed in 250 patients (62.8%), with the majority of patients receiving valve repair (54.8%, n = 137). Aortic valve replacement was performed in 26% of the patients (n = 103) and tricuspid valve repair in 18.3% (n = 73). Mechanical substitutes were used in 73.7% (n = 101) of all mitral and in 51.5% (n = 53) of all aortic valve replacements. Coronary artery bypass grafting was performed in 163 patients (41.0%). A total of 116 (29.1%) procedures were combined procedures, with the majority being a combination of mitral and tricuspid valve surgery (n = 73) or aortic and mitral valve surgery (n = 30). The remaining 13 combined procedures consisted of either mitral or aortic valve surgery combined with coronary artery bypass grafting. In addition, 43 patients (10.8%) received direct closure of an atrial septal defect. Concomitant surgical ablation was performed in 267 patients (67.1%) and LAA closure in 240 patients (60.3%). Closure of the LAA was performed in 65.8% by ligation and in the remaining 34.2% by amputation. Effective LAA closure was verified by transesophageal echocardiography at the end of the procedure. No significant differences were recorded concerning the intraoperative data between the groups.
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
25
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Wilbring et al
TABLE 1. Clinical Baseline Characteristics All Patients n n (all) Add EuroSCORE CHAD2S2-VASc score EF, % Age, y aHT Diabetes History of embolism Male Extracardiac arteropathy CAD COPD CKD
%
398 6.8 3.5 55 68 303 148 152 201 34 205 34 61
100.0 (2.3) (1.3) (13) (9) 76.1 37.2 38.2 50.5 8.5 51.5 8.5 15.3
n 71
55 27 26 35 5 19 5 7
Group I
Group II
Group III
SA
LAA Closure
SA + LAA
% 17.8 6.0 (2.2) 3.5 (1.1) 55 (11) 65 (8) 77.5 38.0 36.6 49.3 7.0 26.8 7.0 9.8
n 44
33 16 18 25 4 18 4 10
% 11.0 7.2 (3.6) 3.6 (1.2) 52 (14) 66 (13) 75.0 36.4 40.9 56.8 9.1 40.9 9.1 22.7
n 196
%
49.2 7.1 (2.7) 3.4 (1.0) 55 (12) 69 (6) 146 74.5 71 36.2 73 37.2 99 50.5 16 8.1 103 52.6 17 8.7 32 16.3
Control n 87
69 34 35 42 9 65 8 12
% 21.9 6.7 (3.0) 3.6 (1.3) 54 (14) 70 (9) 79.3 39.1 40.2 48.3 10.3 74.7 9.2 13.8
P n/a n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
aHT, arterial hypertension; CAD, coronary artery disease; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; EF, ejection fraction; LAA, left atrial appendage; n/a, not applicable; SA, surgical ablation.
Thirty-day mortality was 5.5% (n = 22) and did not differ significantly between the groups (group I, SA, 5.6%; group II, LAA closure, 4.5%; group III, LAA + SA, 5.6%; control, 5.7%; P = 0.760). Furthermore, the LAA closure technique (amputation/ ligation) had no impact on the main hospital outcomes. Finally, 376 patients survived hospital stay and were included in further follow-up analyses of primary study end points.
Primary Study End Points: Incidence of Thromboembolic Events at 1 Year of Follow-up Figure 1 depicts the incidence of thromboembolic events within the particular groups during 1 year of follow-up in reference to the hospital survivors. In 37 patients (9.8%), a thromboembolic event occurred during the first year of follow-up. Thromboembolic events included stroke (n = 20/37, 54.1%) or
FIGURE 1. Incidence of thromboembolic events during 1 year of follow-up of the hospital survivors. Thromboembolic events are depicted as the sum of stroke and PAE. The control groupVreceiving neither LAA closure nor surgical ablationVexperiences significantly higher incidence of thromboembolic events. Generally, patients with LAA closure had lower incidence of thromboembolic events compared with the control group. LAA, surgical left atrial appendage occlusion; PAE, peripheral arterial embolism; SA, surgical ablation.
26
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
peripheral arterial embolism (n = 17/37, 45.9%). The distribution of stroke and peripheral arterial embolism was counterbalanced among all groups [P = nonsignificant (n.s.)]. The particular mortality of patients experiencing such an embolic event was 41% (n = 15) and significantly increased compared with the remaining patients (P G 0.01). The highest incidence of thromboembolic events was recorded in patients of the control group (20.5%, n = 17/82). The corresponding incidence was significantly higher in comparison with all three groups (P = 0.01). Particularly, patients receiving LAA closure either lone standing or combined with surgical ablation had a significantly lower incidence of thromboembolic events compared with the control group (6.6% vs 20.5%, P = 0.01). No significant differences were recorded among the groups I, II, and III (P = n.s.). In addition, the LAA closure technique (amputation/ligation) had no impact on the incidence of thromboembolic events.
Coherence Between Rhythm and Incidence of Thromboembolic Event After 1 year of follow-up, sinus rhythm was documented in 38.2% of the survivors (n = 131/343). Hereby, patients of the control group and group II (LAA only) had significantly less frequent sinus rhythm (P = 0.01). In 41.2% of the patients experiencing a thromboembolic event persistent AF and in 23.5% a sinus rhythm was documented. In the remaining 35.3%, the cardiac rhythm could not be evaluated mainly because the patient usually died before the final follow-up date, and thus, the associated decided rhythm diagnostics was not performed. The presence of sinus rhythm had no statistically significant impact on
Thromboembolic Events After Surgical LAA Closure
the incidence of thromboembolic events after 1 year of follow-up (P = n.s.). Figures 2 and 3 summarize these findings.
Primary Study End Points: Survival After 1 Year of Follow-up The overall mortality (hospital plus follow-up mortality) was 13.8%, with 55 patients dying from any cause. When excluding the patients dying during primary hospital stay (n = 22, 5.5%), the mortality rate of the hospital survivors was 8.8% (n = 33) during 1 year of follow-up. Hereby, the highest mortality was observed with 17.1% (n = 14) in the control group. This was significantly higher compared with the other groups (P = 0.01). Particularly, all patients with LAA closure either lone standing or combined with surgical ablation had a significantly better survival compared with the patients of the control group (93.0% vs 82.9%, P = 0.01). No significant differences were recorded among the groups I, II, and III (P = n.s.). Figure 4 summarizes the mortality rates of the hospital survivors during 1 year of follow-up.
DISCUSSION Atrial fibrillation is the most common arrhythmia.1,2 Aging of the general population and improved survival from cardiac diseases have furthermore led to increasing incidence and prevalence of AF.10 The most devastating complication is the formation of intracardiac thrombi and the development of an ischemic stroke. Today, approximately 20% of all ischemic strokes are associated with AF.1 Several points of action are discussed for minimizing the risk of stroke in patients with AF undergoing cardiac surgery.
FIGURE 2. Cardiac rhythm of the survivors after 1 year of follow-up. Patients of the control group and patients of group II had significantly less frequent sinus rhythm (P = 0.01 for both). The presence of sinus rhythm had no statistically significant impact on the incidence of thromboembolic events or on mortality after 1 year of follow-up (P = n.s.). LAA, surgical left atrial appendage occlusion; SA, surgical ablation. Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
27
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Wilbring et al
FIGURE 3. Cardiac rhythm found in patients experiencing an thromboembolic event. No statistical significant association between rhythm and thromboembolic event was found (P = n.s.). Most of the patients experiencing a thromboembolic event died before decided rhythm diagnostics could be performed. Thus, a high percentage of unclear rhythm had to be documented.
Oral anticoagulative treatment with warfarin is the grounding for prevention of cardiac embolism in the presence of AF, but despite its undoubted importance, several deficiencies such as intolerance, noncompliance, or bleeding complications are reported.1,2 For instance, Humphries et al11,12 demonstrated that only half of the patients with AF being hospitalized took warfarin as prescribed. Prevention of cardiac embolism by restoring sinus rhythm is the main goal of surgical ablation. Presently, several energy sources and lesion sets are discussed, and many studies report effectiveness of surgical ablation.9,13,14 Reported success rates range in between 40% and 90%.10,15,16 This high variability reveals a meaningful number of patients, in which surgical ablation does not represent an effective treatment strategy. Because the LAA was identified to be the origin of 90% of all intracardiac thrombi, the LAA was brought back into the cardiac surgeon’s focus.1,4 Closure of the LAA by interventional devices proved to be equally effective for prevention of embolic strokes compared with warfarin therapy during the most recent trials PROTECT-AF and PREVAIL.7,8 For those reasons, the LAA often is concomitantly closed during cardiac surgery. Nonetheless, available evidence concerning surgical LAA closure and prevention of stroke is still scarce. A recently published meta-analysis by Tsai et al12 identified seven relevant studies, concertedly reporting approximately 1700 patients receiving concomitant LAA closure. Despite the heterogeneity of the study populations and applied closing techniques, Tsai et al could demonstrate a trend toward
28
significantly reduced incidence of stroke at 30 days of follow-up as well at the latest follow-up, which nearby was not clearly defined. Likewise, the LAA occlusion group had a significantly decreased mortality compared with the nonocclusion group.12 In line with these findings, the present registry revealed a nearly two-third reduction in the incidence of thromboembolic events in patients receiving LAA closure, surgical ablation, or combination of both after 1 year of follow-up. Based on the high observed mortality of 41% in patients experiencing a thromboembolic event, the reduced incidence of these events had a significant impact on survival rates. Accordingly, the observed mortality of the hospital survivors after 1 year of follow-up was more than twice as high as in the control group. In this context, it has to be emphasized that corresponding to our institution’s recommendation, all patients with AF postoperatively received continuation of warfarin therapy during the first year after surgery, independently from the recorded heart rhythm in the meantime. The results of AF ablation could be seen as poor. However, the interpretation of AF ablation results is part of a lively debate. In the present series, we looked for stable sinus rhythm and not the absence of AF, performing multiple 72-hour ECG. Study results often vary widely because of the mode of follow-up and corresponding differing detection ratesVsingle ECG (like the older series) or event recorder (which should be the criterion standard). In addition, it has to be kept in mind that the study only included long-standing persistent/permanent AFVnot mixed up with patients with paroxysmal AF.
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Thromboembolic Events After Surgical LAA Closure
FIGURE 4. Mortality rates of the hospital survivors after 1 year of follow-up. Patients of the control group had significantly higher mortality compared with all other groups. Particularly, patients receiving LAA closure (alone or combined with surgical ablation) had a significantly better survival after 1 year of follow-up compared with the control group (P = 0.01). LAA, surgical left atrial appendage occlusion; SA, surgical ablation.
As expected, the percentage of sinus rhythm after 1 year of follow-up was the lowest in the control group and significantly higher in patients receiving concomitant surgical ablationValone or combined with LAA closure. Solely, the group of patients receiving only LAA closure had comparably low rates of sinus rhythm after 1 year of follow-up. With regard to these findings, the bias of the cardiac rhythm on the primary study end points needs to be discussed. Univariate analysis detected no impact of rhythm on the primary study end points. A wise multivariate analysis was not possible because of the small sample size, the different group sizes, and the limited number of events. In addition, most of the patients experiencing a thromboembolic event died before the rhythm could be clarified. For those reasons, more than half of the patients experiencing a thromboembolic event had an unclear or sinus rhythm (Fig. 3). In addition, even patients receiving only LAA occlusion without concomitant surgical ablation had a significantly lower incidence of thromboembolic events compared with the control group. This relativizes the discussed potential bias by the heart rhythm. Generally, observed rates of events seemed to be quite high. On the one hand, this not only might have resulted from the high surgical risk group with an additionally high thromboembolic risk but also, on the other hand, can be an effect of a captious documentation of any thromboembolic events (not only stroke) or even as a result of potentially insufficient warfarin therapyVwhat about, we particularly have no detailed information. Interestingly, surgical ablation and LAA closure had no significant additive effect concerning the incidence of thromboembolic events (Fig. 1). This might be an important finding and needs to be evaluated in further studies. If it comes true
that LAA closure would suffice for preventing thromboembolic events, elaborate surgical ablation potentially could be omitted in selected high-risk constellations. Like the other series, the present study is characterized by a heterogeneous patient population receiving various types of cardiac surgery as well as different types of LAA closure.12 Despite this particular limitation, patients of all groups had a comparable risk for thromboembolic events, as predicted by the CHAD2S2-VASc score. Effectiveness of LAA closure was assessed by intraoperative echocardiography, but the design of the present study did not allow to confirm the stability of LAA occlusion. Since the LAAOS trial, it is known that a relevant percentage of LAA closures by suture or stapling is ineffective.5 Thus, it has to be assumed that some of the LAAs in the present study likewise reconnected during follow-up. Anyhow, despite this remaining uncertainty, patients receiving LAA occlusion alone or in combination with surgical ablation had a significantly decreased incidence of thromboembolic events and an improved survival after 1 year of follow-up.
Limitations The main limitation of the present study of course is its nonrandomized design. Selection of the patients for each group partially also related to surgeon’s discretion, which suggests significant selection bias. Therefore, causalities cannot be claimed. Despite an undeniable bias, the bottom line remains less strokes after LAA closure after 1-year of follow-up in patients with chronic AF and comparable CHAD2S2-VASc scores. A further limitation is that the impact of conversion to sinus rhythm in patients additionally receiving surgical ablation was not clearly determined because of patchy diagnostics. An additional limitation is the
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
29
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Wilbring et al
lacking echocardiographic control during follow-up and the remaining final uncertainty concerning the stability of complete LAA occlusion without reconnection or larger remnants.
CONCLUSIONS Despite the mentioned limitations, the present study in accordance to other available trials gives evidence that the results yielded from the PROTECT-AF or PREVAIL trial can be translated into surgical occlusion of the LAA. Thus, surgical LAA occlusion seems to be a safe, convenient-to-perform, and capable method to prevent thromboembolic events in patients with AF undergoing cardiac surgery. The presently ongoing randomized trial will further add important knowledge to this important discussion.
REFERENCES 1. Cox JL. Mechanical closure of the left atrial appendage: is it time to be more aggressive? J Thorac Cardiovasc Surg. 2013;146:1018Y1027. 2. Barekatain A, Rasekh A, Massumi A. Exclusion of the left atrial appendage to prevent stroke in cases of atrial fibrillation. Tex Heart Inst J. 2012;39:535Y537. 3. Johnson WD, Ganjoo AK, Stone CD, Srivyas RC, Howard M. The left atrial appendage: our most lethal human attachment! Surgical implications. Eur J Cardiothorac Surg. 2000;17:718Y722. 4. Fuster V, Ryde´n LE, Cannom DS, et al. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol. 2011;57:e101Ye198. 5. Healey JS, Crystal E, Lamy A, et al. Left Atrial Appendage Occlusion Study (LAAOS): results of a randomized controlled pilot study of left atrial appendage occlusion during coronary bypass surgery in patients at risk for stroke. Am Heart J. 2005;150:288Y293.
6. Whitlock RP, Vincent J, Blackall MH, et al. Left Atrial Appendage Occlusion Study II (LAAOS II). Can J Cardiol. 2013;29:1443Y1447. 7. Reddy VY, Doshi SK, Sievert H, et al. PROTECT AF Investigators. Percutaneous left atrial appendage closure for stroke prophylaxis in patients with atrial fibrillation: 2.3-year follow-up of the PROTECT AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients with Atrial Fibrillation) Trial. Circulation. 2013;127:720Y729. 8. Holmes DR Jr, Kar S, Price MJ, et al. Prospective randomized evaluation of the Watchman Left Atrial Appendage Closure device in patients with atrial fibrillation versus long-term warfarin therapy: the PREVAIL trial. J Am Coll Cardiol. 2014;64:1Y12. 9. Knaut M, Tugtekin SM, Jung F, Matschke K. Microwave ablation for the surgical treatment of permanent atrial fibrillationVa single centre experience. Eur J Cardiothorac Surg. 2004;26:742Y746. 10. Pinho-Gomes AC, Amorim MJ, Oliveira SM, Leite-Moreira AF. Surgical treatment of atrial fibrillation: an updated review. Eur J Cardiothorac Surg. 2014;46:167Y178. 11. Humphries KH, Jackevicius C, Gong Y, et al. Canadian Cardiovascular Outcomes Research Team. Population rates of hospitalization for atrial fibrillation/flutter in Canada. Can J Cardiol. 2004;20:869Y876. 12. Tsai YC, Phan K, Munkholm-Larsen S, Tian DH, La Meir M, Yan TD. Surgical left atrial appendage occlusion during cardiac surgery for patients with atrial fibrillation: a meta-analysis. Eur J Cardiothorac Surg. 2014;47: 847Y854. 13. Saint LL, Damiano RJ Jr. Surgical treatment of atrial fibrillation. Mo Med. 2012;109:281Y287. 14. Damiano RJ Jr, Badhwar V, Acker MA, et al. The CURE-AF trial: a prospective, multicenter trial of irrigated radiofrequency ablation for the treatment of persistent atrial fibrillation during concomitant cardiac surgery. Heart Rhythm. 2014;11:39Y45. 15. Wolf RK, Schneeberger EW, Osterday R, et al. Video-assisted bilateral pulmonary vein isolation and left atrial appendage exclusion for atrial fibrillation. J Thorac Cardiovasc Surg. 2005;130:797Y802. 16. Gelsomino S, La Meir M, Luca` F, et al. Treatment of lone atrial fibrillation: a look at the past, a view of the present and a glance at the future. Eur J Cardiothorac Surg. 2012;41:1284Y1294.
CLINICAL PERSPECTIVE This prospective registry enrolled 398 patients with permanent atrial fibrillation undergoing cardiac surgery. Concomitant surgical procedures were isolated surgical ablation, isolated left atrial appendage closure, and combined surgical ablation and left atrial appendage closure. The control group consisted of 87 patients without concomitant surgical ablation or appendage closure. One-year follow-up was completed in all patients. Patients with left atrial appendage closure alone or in combination with surgical ablation had a significantly reduced incidence of thromboembolic events (6.6% vs 20.5%) and an improved survival after 1 year of follow-up. This is an interesting study and suggests that managing the left atrial appendage has benefits both in terms of stroke reduction and survival in patients with permanent atrial fibrillation. However, the study has significant limitations. First of all, it was nonrandomized and thus subject to significant selection bias. Moreover, the control group had a very high rate of thromboembolic events in the first year (21%), which is much higher than expected in anticoagulated patients with atrial fibrillation, particularly with a CHAD2S2-VASc score of 3.6 in this group. There were no data given regarding compliance with anticoagulation in the different groups. Thus, it is difficult to draw any conclusions from this report. A prospective, randomized trial would be needed to precisely define the benefits of left atrial appendage occlusion.
30
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Reduced Incidence of Thromboembolic Events After Surgical Closure of Left Atrial Appendage in Patients With Atrial Fibrillation Manuel Wilbring, MD,* Friedrich Jung, MD,Þ Christoph Weber, MD,* Klaus Matschke, MD,* and Michael Knaut, MD*
Objective: Most of the detected thrombi in patients with atrial fibrillation (AF) can be found in the left atrial appendage (LAA). Interventional LAA closure recently proved to be noninferior to warfarin therapy. Whether these results can be fully translated into surgical LAA closure remains unclear. Corresponding data are still lacking. The present observational study evaluated the impact of surgical LAA closure in patients with AF undergoing cardiac surgery on postoperative thromboembolic events. Methods: A prospective registry enrolled 398 patients with permanent AF undergoing cardiac surgery. Concomitant procedures were isolated surgical ablation (group I, n = 71), isolated LAA closure (group II, n = 44), and combined surgical ablation and LAA closure (group III, n = 196). The control group consisted of 87 patients without concomitant surgical ablation or LAA closure. One-year follow-up was completed in all patients. End points were thromboembolic events and death from any cause. Results: Clinical baseline characteristics were comparable among the groups. General hospital mortality was 5.5% and likewise differed not significantly. Postoperatively, mean (SD) CHAD2S2-VASc score of 3.5 (1.3) differed not significantly among the groups, indicating comparable thromboembolic risk. Follow-up referred to all hospital survivors (n = 376). Herein, overall incidence of thromboembolic events was 9.8% (n = 37), with an associated mortality of 41.0%. Patients with LAA closure alone or in combination with surgical ablation had a significantly reduced incidence of thromboembolic events (6.6% vs 20.5%, P G 0.01) and consecutively improved survival after 1 year of follow-up (7.0% vs 17.1%, P G 0.01). Conclusions: Left atrial appendage closure alone or in combination with surgical ablation was associated with a significantly reduced Accepted for publication October 31, 2015. From the *Department of Cardiac Surgery, University Heart Center Dresden, Dresden, Germany; and †Institute for Polymer Research and BerlinBrandenburg Centre for Regenerative Therapies, Helmholtz-Zentrum Geesthacht, Berlin and Teltow, Germany. Presented at the Annual Scientific Meeting of the International Society for Minimally Invasive Cardiothoracic Surgery, May 30YJune 2, 2012, in Los Angeles, CA USA. Disclosure: The authors declare no conflicts of interest. Address correspondence and reprint requests to Manuel Wilbring, MD, Department of Cardiac Surgery, University Heart Center Dresden, Fetscherstrasse 76, 01307 Dresden, Germany. E-mail: [email protected]. Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery ISSN: 1556-9845/16/1101-0024
24
rate of thromboembolic events and consecutively improved survival after 1 year of follow-up. Key Words: Left atrial appendage, Atrial fibrillation, Surgical closure, Stroke, Surgical ablation. (Innovations 2016;11:24Y30)
A
trial fibrillation (AF) is the most common cardiac arrhythmia, which affects up to 2% of the general population.1,2 The most threatening and most disabling complication of AF is the ischemic stroke. Approximately 20% of all ischemic strokes are considered to be associated with AF.1 These data underline the general importance of AF. Actually, permanent oral anticoagulation is recommended in patients with a CHAD2S2-VASc score of 2 or greater.3 Nonetheless, permanent oral anticoagulation is associated with several problems including intolerance, noncompliance, and bleeding complications. Recent focus was set on the left atrial appendage (LAA), where up to 90% of the detected thrombi in patients with AF can be found.1,4 This gave the LAA the reputation to be ‘‘the most lethal human attachment.’’3 Therefore, closure of the LAA potentially could be the solution of the problem of AF and cardiac embolism. Recent studies tried to determine the effect of mechanical LAA closure on the incidence of stroke. The LAAOS trial, published in 2005, was the first randomized trial comparing different methods of LAA closureV epicardial suture or staplingVin patients undergoing coronary artery bypass grafting.5 Unfortunately, the LAAOS trial could not answer the intended question but revealed significant problems with the closure methods.5 Epicardial suture was only effective in 43% and stapling in 72%.5 The remaining patients were considered as treatment failures because of LAA reconnection or remnants greater than 1 cm.5 For those reasons, an optimized sequel study was initiated in 2013V the LAAOS II trial.6 This trial actually is still enrolling patients and hopefully will provide meaningful data to support further decision making. Until these data will become available, smaller monocentric series might remain the only source of knowledge. The proof of concept of LAA closure was delivered by interventional cardiologists. The recently published PROTECTAF trial was a prospective randomized trial, showing noninferiority Innovations & Volume 11, Number 1, January/February 2016
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
of the WATCHMAN device to warfarin therapy in the prevention of stroke.7 The most recent study, the PREVAIL trial, confirmed these findings and additionally revealed an improved procedural safety.8 These well-conducted studies implicated the pathogenesis of stroke in AF and confirmed mechanical closure of the LAA to be an effective approach. Nonetheless, these data cannot completely be translated into patients undergoing cardiac surgery, and data dealing with surgical patients are still scarce. For those reasons, the present observational study was initiated. With the present study, we sought to determine the clinical impact of surgical LAA closure on the incidence of stroke in a prospective, nonrandomized registry of patients with AF undergoing any cardiac surgery.
PATIENTS AND METHODS Patients and Study Design The study was designed as a prospective registry, which included all patients with permanent AF undergoing any elective cardiac surgery. Urgent procedures or emergencies were excluded. Preoperative, intraoperative, and postoperative data were extracted from the hospital’s database for subsequent analysis. Finally, a total of 398 patients were enrolled. The enrollment dated back to the period between 2005 and 2010. Because in this period, there existed no standardized protocol for patients with permanent AF, the decision if the patient received either LAA closure, surgical ablation, both, or none of these depended on the operating surgeon’s preference. Ablation was performed using microwave energy (until the Guidant devices were taken off the market) and subsequently cryoenergy. The modified left atrial line concept was applied as reported previously.9 The patients were divided into four groups depending on concomitant procedures as follows: Group I: only surgical ablation (SA, n = 71). Group II: only LAA closure (LAA, n = 44). Group III: combined surgical ablation and LAA closure (SA + LAA, n = 196). Control: no concomitant surgical ablation or LAA closure (control, n = 87). All patients were discharged with oral anticoagulation (warfarin), independently from the heart rhythm. Follow-up visits including 72-hour Holter-ECG were performed after 6 and 12 months. Standard antiarrhythmic treatment consisted of metoprolol. One-year follow-up was completed in all surviving patients. Primary study end points were incidence of thromboembolic events, death from any cause, and cardiac rhythm after 1 year of follow-up. Follow-up included direct interview of the patients and their cardiologist. Stroke and other thromboembolic events were defined as clinical apparent events, being revealed by computed angiography or other angiography in general. Comparative analyses were performed between groups I to III and control as well as between all patients receiving LAA closure (groups I and III) and the control group. Primary study end points were analyzed in reference to the hospital survivors. Accordingly, patients dying during
Thromboembolic Events After Surgical LAA Closure
their primary hospital stay were excluded from further followup and analyses. All patients gave informed consent. The study was reviewed and approved by the institutional review board at Medical Faculty ‘‘Carl Gustav Carus’’ at Technical University of Dresden.
Statistical Analysis Continuous data are expressed as means and SDs as well as absolute numbers and percentages for categorical variables. For analyzing continuous variables, unpaired t test was used; for comparing categorical variables, Fisher exact test was used. Time-to-event analyses were performed by Kaplan-Meier analysis and log-rank test. A P value of less than 0.05 was considered significant. Statistical analysis was performed using JMP 9.0 Software (SAS Institute Corp, Cary, NC USA). Power analysis was performed.
RESULTS Clinical Baseline Characteristics and Comparability of the Groups Mean (SD) age of all patients was 68 (9) years with a balanced sex distribution (male, n = 201, 50.5%). Risk factors for embolization according to the CHAD2S2-VASc score were balanced in between the groups. Likewise, the general surgical risk profile as predicted by additive EuroSCORE differed not significantly between all groups. Clinical baseline characteristics mainly showed no significant differences. Significant differences were recorded concerning the presence of coronary artery disease, being more frequent in the control group, and the presence of chronic kidney disease, which was more frequent in group II (only LAA closure). All remaining baseline characteristics did not differ significantly. The clinical baseline characteristics are summarized in Table 1.
Intraoperative Data and Hospital Outcome Mitral valve procedures were performed in 250 patients (62.8%), with the majority of patients receiving valve repair (54.8%, n = 137). Aortic valve replacement was performed in 26% of the patients (n = 103) and tricuspid valve repair in 18.3% (n = 73). Mechanical substitutes were used in 73.7% (n = 101) of all mitral and in 51.5% (n = 53) of all aortic valve replacements. Coronary artery bypass grafting was performed in 163 patients (41.0%). A total of 116 (29.1%) procedures were combined procedures, with the majority being a combination of mitral and tricuspid valve surgery (n = 73) or aortic and mitral valve surgery (n = 30). The remaining 13 combined procedures consisted of either mitral or aortic valve surgery combined with coronary artery bypass grafting. In addition, 43 patients (10.8%) received direct closure of an atrial septal defect. Concomitant surgical ablation was performed in 267 patients (67.1%) and LAA closure in 240 patients (60.3%). Closure of the LAA was performed in 65.8% by ligation and in the remaining 34.2% by amputation. Effective LAA closure was verified by transesophageal echocardiography at the end of the procedure. No significant differences were recorded concerning the intraoperative data between the groups.
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
25
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Wilbring et al
TABLE 1. Clinical Baseline Characteristics All Patients n n (all) Add EuroSCORE CHAD2S2-VASc score EF, % Age, y aHT Diabetes History of embolism Male Extracardiac arteropathy CAD COPD CKD
%
398 6.8 3.5 55 68 303 148 152 201 34 205 34 61
100.0 (2.3) (1.3) (13) (9) 76.1 37.2 38.2 50.5 8.5 51.5 8.5 15.3
n 71
55 27 26 35 5 19 5 7
Group I
Group II
Group III
SA
LAA Closure
SA + LAA
% 17.8 6.0 (2.2) 3.5 (1.1) 55 (11) 65 (8) 77.5 38.0 36.6 49.3 7.0 26.8 7.0 9.8
n 44
33 16 18 25 4 18 4 10
% 11.0 7.2 (3.6) 3.6 (1.2) 52 (14) 66 (13) 75.0 36.4 40.9 56.8 9.1 40.9 9.1 22.7
n 196
%
49.2 7.1 (2.7) 3.4 (1.0) 55 (12) 69 (6) 146 74.5 71 36.2 73 37.2 99 50.5 16 8.1 103 52.6 17 8.7 32 16.3
Control n 87
69 34 35 42 9 65 8 12
% 21.9 6.7 (3.0) 3.6 (1.3) 54 (14) 70 (9) 79.3 39.1 40.2 48.3 10.3 74.7 9.2 13.8
P n/a n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
aHT, arterial hypertension; CAD, coronary artery disease; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; EF, ejection fraction; LAA, left atrial appendage; n/a, not applicable; SA, surgical ablation.
Thirty-day mortality was 5.5% (n = 22) and did not differ significantly between the groups (group I, SA, 5.6%; group II, LAA closure, 4.5%; group III, LAA + SA, 5.6%; control, 5.7%; P = 0.760). Furthermore, the LAA closure technique (amputation/ ligation) had no impact on the main hospital outcomes. Finally, 376 patients survived hospital stay and were included in further follow-up analyses of primary study end points.
Primary Study End Points: Incidence of Thromboembolic Events at 1 Year of Follow-up Figure 1 depicts the incidence of thromboembolic events within the particular groups during 1 year of follow-up in reference to the hospital survivors. In 37 patients (9.8%), a thromboembolic event occurred during the first year of follow-up. Thromboembolic events included stroke (n = 20/37, 54.1%) or
FIGURE 1. Incidence of thromboembolic events during 1 year of follow-up of the hospital survivors. Thromboembolic events are depicted as the sum of stroke and PAE. The control groupVreceiving neither LAA closure nor surgical ablationVexperiences significantly higher incidence of thromboembolic events. Generally, patients with LAA closure had lower incidence of thromboembolic events compared with the control group. LAA, surgical left atrial appendage occlusion; PAE, peripheral arterial embolism; SA, surgical ablation.
26
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
peripheral arterial embolism (n = 17/37, 45.9%). The distribution of stroke and peripheral arterial embolism was counterbalanced among all groups [P = nonsignificant (n.s.)]. The particular mortality of patients experiencing such an embolic event was 41% (n = 15) and significantly increased compared with the remaining patients (P G 0.01). The highest incidence of thromboembolic events was recorded in patients of the control group (20.5%, n = 17/82). The corresponding incidence was significantly higher in comparison with all three groups (P = 0.01). Particularly, patients receiving LAA closure either lone standing or combined with surgical ablation had a significantly lower incidence of thromboembolic events compared with the control group (6.6% vs 20.5%, P = 0.01). No significant differences were recorded among the groups I, II, and III (P = n.s.). In addition, the LAA closure technique (amputation/ligation) had no impact on the incidence of thromboembolic events.
Coherence Between Rhythm and Incidence of Thromboembolic Event After 1 year of follow-up, sinus rhythm was documented in 38.2% of the survivors (n = 131/343). Hereby, patients of the control group and group II (LAA only) had significantly less frequent sinus rhythm (P = 0.01). In 41.2% of the patients experiencing a thromboembolic event persistent AF and in 23.5% a sinus rhythm was documented. In the remaining 35.3%, the cardiac rhythm could not be evaluated mainly because the patient usually died before the final follow-up date, and thus, the associated decided rhythm diagnostics was not performed. The presence of sinus rhythm had no statistically significant impact on
Thromboembolic Events After Surgical LAA Closure
the incidence of thromboembolic events after 1 year of follow-up (P = n.s.). Figures 2 and 3 summarize these findings.
Primary Study End Points: Survival After 1 Year of Follow-up The overall mortality (hospital plus follow-up mortality) was 13.8%, with 55 patients dying from any cause. When excluding the patients dying during primary hospital stay (n = 22, 5.5%), the mortality rate of the hospital survivors was 8.8% (n = 33) during 1 year of follow-up. Hereby, the highest mortality was observed with 17.1% (n = 14) in the control group. This was significantly higher compared with the other groups (P = 0.01). Particularly, all patients with LAA closure either lone standing or combined with surgical ablation had a significantly better survival compared with the patients of the control group (93.0% vs 82.9%, P = 0.01). No significant differences were recorded among the groups I, II, and III (P = n.s.). Figure 4 summarizes the mortality rates of the hospital survivors during 1 year of follow-up.
DISCUSSION Atrial fibrillation is the most common arrhythmia.1,2 Aging of the general population and improved survival from cardiac diseases have furthermore led to increasing incidence and prevalence of AF.10 The most devastating complication is the formation of intracardiac thrombi and the development of an ischemic stroke. Today, approximately 20% of all ischemic strokes are associated with AF.1 Several points of action are discussed for minimizing the risk of stroke in patients with AF undergoing cardiac surgery.
FIGURE 2. Cardiac rhythm of the survivors after 1 year of follow-up. Patients of the control group and patients of group II had significantly less frequent sinus rhythm (P = 0.01 for both). The presence of sinus rhythm had no statistically significant impact on the incidence of thromboembolic events or on mortality after 1 year of follow-up (P = n.s.). LAA, surgical left atrial appendage occlusion; SA, surgical ablation. Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
27
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Wilbring et al
FIGURE 3. Cardiac rhythm found in patients experiencing an thromboembolic event. No statistical significant association between rhythm and thromboembolic event was found (P = n.s.). Most of the patients experiencing a thromboembolic event died before decided rhythm diagnostics could be performed. Thus, a high percentage of unclear rhythm had to be documented.
Oral anticoagulative treatment with warfarin is the grounding for prevention of cardiac embolism in the presence of AF, but despite its undoubted importance, several deficiencies such as intolerance, noncompliance, or bleeding complications are reported.1,2 For instance, Humphries et al11,12 demonstrated that only half of the patients with AF being hospitalized took warfarin as prescribed. Prevention of cardiac embolism by restoring sinus rhythm is the main goal of surgical ablation. Presently, several energy sources and lesion sets are discussed, and many studies report effectiveness of surgical ablation.9,13,14 Reported success rates range in between 40% and 90%.10,15,16 This high variability reveals a meaningful number of patients, in which surgical ablation does not represent an effective treatment strategy. Because the LAA was identified to be the origin of 90% of all intracardiac thrombi, the LAA was brought back into the cardiac surgeon’s focus.1,4 Closure of the LAA by interventional devices proved to be equally effective for prevention of embolic strokes compared with warfarin therapy during the most recent trials PROTECT-AF and PREVAIL.7,8 For those reasons, the LAA often is concomitantly closed during cardiac surgery. Nonetheless, available evidence concerning surgical LAA closure and prevention of stroke is still scarce. A recently published meta-analysis by Tsai et al12 identified seven relevant studies, concertedly reporting approximately 1700 patients receiving concomitant LAA closure. Despite the heterogeneity of the study populations and applied closing techniques, Tsai et al could demonstrate a trend toward
28
significantly reduced incidence of stroke at 30 days of follow-up as well at the latest follow-up, which nearby was not clearly defined. Likewise, the LAA occlusion group had a significantly decreased mortality compared with the nonocclusion group.12 In line with these findings, the present registry revealed a nearly two-third reduction in the incidence of thromboembolic events in patients receiving LAA closure, surgical ablation, or combination of both after 1 year of follow-up. Based on the high observed mortality of 41% in patients experiencing a thromboembolic event, the reduced incidence of these events had a significant impact on survival rates. Accordingly, the observed mortality of the hospital survivors after 1 year of follow-up was more than twice as high as in the control group. In this context, it has to be emphasized that corresponding to our institution’s recommendation, all patients with AF postoperatively received continuation of warfarin therapy during the first year after surgery, independently from the recorded heart rhythm in the meantime. The results of AF ablation could be seen as poor. However, the interpretation of AF ablation results is part of a lively debate. In the present series, we looked for stable sinus rhythm and not the absence of AF, performing multiple 72-hour ECG. Study results often vary widely because of the mode of follow-up and corresponding differing detection ratesVsingle ECG (like the older series) or event recorder (which should be the criterion standard). In addition, it has to be kept in mind that the study only included long-standing persistent/permanent AFVnot mixed up with patients with paroxysmal AF.
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Thromboembolic Events After Surgical LAA Closure
FIGURE 4. Mortality rates of the hospital survivors after 1 year of follow-up. Patients of the control group had significantly higher mortality compared with all other groups. Particularly, patients receiving LAA closure (alone or combined with surgical ablation) had a significantly better survival after 1 year of follow-up compared with the control group (P = 0.01). LAA, surgical left atrial appendage occlusion; SA, surgical ablation.
As expected, the percentage of sinus rhythm after 1 year of follow-up was the lowest in the control group and significantly higher in patients receiving concomitant surgical ablationValone or combined with LAA closure. Solely, the group of patients receiving only LAA closure had comparably low rates of sinus rhythm after 1 year of follow-up. With regard to these findings, the bias of the cardiac rhythm on the primary study end points needs to be discussed. Univariate analysis detected no impact of rhythm on the primary study end points. A wise multivariate analysis was not possible because of the small sample size, the different group sizes, and the limited number of events. In addition, most of the patients experiencing a thromboembolic event died before the rhythm could be clarified. For those reasons, more than half of the patients experiencing a thromboembolic event had an unclear or sinus rhythm (Fig. 3). In addition, even patients receiving only LAA occlusion without concomitant surgical ablation had a significantly lower incidence of thromboembolic events compared with the control group. This relativizes the discussed potential bias by the heart rhythm. Generally, observed rates of events seemed to be quite high. On the one hand, this not only might have resulted from the high surgical risk group with an additionally high thromboembolic risk but also, on the other hand, can be an effect of a captious documentation of any thromboembolic events (not only stroke) or even as a result of potentially insufficient warfarin therapyVwhat about, we particularly have no detailed information. Interestingly, surgical ablation and LAA closure had no significant additive effect concerning the incidence of thromboembolic events (Fig. 1). This might be an important finding and needs to be evaluated in further studies. If it comes true
that LAA closure would suffice for preventing thromboembolic events, elaborate surgical ablation potentially could be omitted in selected high-risk constellations. Like the other series, the present study is characterized by a heterogeneous patient population receiving various types of cardiac surgery as well as different types of LAA closure.12 Despite this particular limitation, patients of all groups had a comparable risk for thromboembolic events, as predicted by the CHAD2S2-VASc score. Effectiveness of LAA closure was assessed by intraoperative echocardiography, but the design of the present study did not allow to confirm the stability of LAA occlusion. Since the LAAOS trial, it is known that a relevant percentage of LAA closures by suture or stapling is ineffective.5 Thus, it has to be assumed that some of the LAAs in the present study likewise reconnected during follow-up. Anyhow, despite this remaining uncertainty, patients receiving LAA occlusion alone or in combination with surgical ablation had a significantly decreased incidence of thromboembolic events and an improved survival after 1 year of follow-up.
Limitations The main limitation of the present study of course is its nonrandomized design. Selection of the patients for each group partially also related to surgeon’s discretion, which suggests significant selection bias. Therefore, causalities cannot be claimed. Despite an undeniable bias, the bottom line remains less strokes after LAA closure after 1-year of follow-up in patients with chronic AF and comparable CHAD2S2-VASc scores. A further limitation is that the impact of conversion to sinus rhythm in patients additionally receiving surgical ablation was not clearly determined because of patchy diagnostics. An additional limitation is the
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
29
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
Innovations & Volume 11, Number 1, January/February 2016
Wilbring et al
lacking echocardiographic control during follow-up and the remaining final uncertainty concerning the stability of complete LAA occlusion without reconnection or larger remnants.
CONCLUSIONS Despite the mentioned limitations, the present study in accordance to other available trials gives evidence that the results yielded from the PROTECT-AF or PREVAIL trial can be translated into surgical occlusion of the LAA. Thus, surgical LAA occlusion seems to be a safe, convenient-to-perform, and capable method to prevent thromboembolic events in patients with AF undergoing cardiac surgery. The presently ongoing randomized trial will further add important knowledge to this important discussion.
REFERENCES 1. Cox JL. Mechanical closure of the left atrial appendage: is it time to be more aggressive? J Thorac Cardiovasc Surg. 2013;146:1018Y1027. 2. Barekatain A, Rasekh A, Massumi A. Exclusion of the left atrial appendage to prevent stroke in cases of atrial fibrillation. Tex Heart Inst J. 2012;39:535Y537. 3. Johnson WD, Ganjoo AK, Stone CD, Srivyas RC, Howard M. The left atrial appendage: our most lethal human attachment! Surgical implications. Eur J Cardiothorac Surg. 2000;17:718Y722. 4. Fuster V, Ryde´n LE, Cannom DS, et al. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol. 2011;57:e101Ye198. 5. Healey JS, Crystal E, Lamy A, et al. Left Atrial Appendage Occlusion Study (LAAOS): results of a randomized controlled pilot study of left atrial appendage occlusion during coronary bypass surgery in patients at risk for stroke. Am Heart J. 2005;150:288Y293.
6. Whitlock RP, Vincent J, Blackall MH, et al. Left Atrial Appendage Occlusion Study II (LAAOS II). Can J Cardiol. 2013;29:1443Y1447. 7. Reddy VY, Doshi SK, Sievert H, et al. PROTECT AF Investigators. Percutaneous left atrial appendage closure for stroke prophylaxis in patients with atrial fibrillation: 2.3-year follow-up of the PROTECT AF (Watchman Left Atrial Appendage System for Embolic Protection in Patients with Atrial Fibrillation) Trial. Circulation. 2013;127:720Y729. 8. Holmes DR Jr, Kar S, Price MJ, et al. Prospective randomized evaluation of the Watchman Left Atrial Appendage Closure device in patients with atrial fibrillation versus long-term warfarin therapy: the PREVAIL trial. J Am Coll Cardiol. 2014;64:1Y12. 9. Knaut M, Tugtekin SM, Jung F, Matschke K. Microwave ablation for the surgical treatment of permanent atrial fibrillationVa single centre experience. Eur J Cardiothorac Surg. 2004;26:742Y746. 10. Pinho-Gomes AC, Amorim MJ, Oliveira SM, Leite-Moreira AF. Surgical treatment of atrial fibrillation: an updated review. Eur J Cardiothorac Surg. 2014;46:167Y178. 11. Humphries KH, Jackevicius C, Gong Y, et al. Canadian Cardiovascular Outcomes Research Team. Population rates of hospitalization for atrial fibrillation/flutter in Canada. Can J Cardiol. 2004;20:869Y876. 12. Tsai YC, Phan K, Munkholm-Larsen S, Tian DH, La Meir M, Yan TD. Surgical left atrial appendage occlusion during cardiac surgery for patients with atrial fibrillation: a meta-analysis. Eur J Cardiothorac Surg. 2014;47: 847Y854. 13. Saint LL, Damiano RJ Jr. Surgical treatment of atrial fibrillation. Mo Med. 2012;109:281Y287. 14. Damiano RJ Jr, Badhwar V, Acker MA, et al. The CURE-AF trial: a prospective, multicenter trial of irrigated radiofrequency ablation for the treatment of persistent atrial fibrillation during concomitant cardiac surgery. Heart Rhythm. 2014;11:39Y45. 15. Wolf RK, Schneeberger EW, Osterday R, et al. Video-assisted bilateral pulmonary vein isolation and left atrial appendage exclusion for atrial fibrillation. J Thorac Cardiovasc Surg. 2005;130:797Y802. 16. Gelsomino S, La Meir M, Luca` F, et al. Treatment of lone atrial fibrillation: a look at the past, a view of the present and a glance at the future. Eur J Cardiothorac Surg. 2012;41:1284Y1294.
CLINICAL PERSPECTIVE This prospective registry enrolled 398 patients with permanent atrial fibrillation undergoing cardiac surgery. Concomitant surgical procedures were isolated surgical ablation, isolated left atrial appendage closure, and combined surgical ablation and left atrial appendage closure. The control group consisted of 87 patients without concomitant surgical ablation or appendage closure. One-year follow-up was completed in all patients. Patients with left atrial appendage closure alone or in combination with surgical ablation had a significantly reduced incidence of thromboembolic events (6.6% vs 20.5%) and an improved survival after 1 year of follow-up. This is an interesting study and suggests that managing the left atrial appendage has benefits both in terms of stroke reduction and survival in patients with permanent atrial fibrillation. However, the study has significant limitations. First of all, it was nonrandomized and thus subject to significant selection bias. Moreover, the control group had a very high rate of thromboembolic events in the first year (21%), which is much higher than expected in anticoagulated patients with atrial fibrillation, particularly with a CHAD2S2-VASc score of 3.6 in this group. There were no data given regarding compliance with anticoagulation in the different groups. Thus, it is difficult to draw any conclusions from this report. A prospective, randomized trial would be needed to precisely define the benefits of left atrial appendage occlusion.
30
Copyright * 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery
Copyright © 2016 by the International Society for Minimally Invasive Cardiothoracic Surgery. Unauthorized reproduction of this article is prohibited.
