Off-pump epicardial tissue sealing—a novel method for atrioventricular disruption complicating mitral valve procedures Albert Schuetz, Costas Schulze and Stephen M. Wildhirt Ann Thorac Surg 2004;78:569-573
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://ats.ctsnetjournals.org/cgi/content/full/78/2/569
The Annals of Thoracic Surgery is the official journal of The Society of Thoracic Surgeons and the Southern Thoracic Surgical Association. Copyright © 2004 by The Society of Thoracic Surgeons. Print ISSN: 0003-4975; eISSN: 1552-6259.
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
Albert Schuetz, MD, PhD, Costas Schulze, MD, and Stephen M. Wildhirt, MD, PhD Department of Cardiac Surgery, Heart Center Augustinum, University of Munich, Munich, Germany
Background. Atrioventricular disruption (AVD) is a rare (1%–2%) but fatal complication after mitral valve procedures; the intraoperative mortality is more than 50% despite the current standard procedure of surgical closure of the defect. We compared the outcome of 9 patients with intraoperative AV disruption, 4 being surgically treated on-pump and 5 receiving epicardial tissue sealing off-pump. Methods. Between March 1998 and May 2002 a total of 9 patients presented with AV disruption intraoperative. The first 4 patients were treated with surgical repair on-pump by reconstruction of the defects with patch or buttressed suture. The second series of 5 patients were treated with a biodegradable collagen system with fibrinogen-based coating off-pump. Three to six layers were placed over the bleeding site with manual pressure for 30 – 60 minutes on the beating heart until bleeding was stopped. Cell saved blood was retransfused.
Results. In the on-pump surgical repair group 3 patients (75%) died within the first day after repair either because of persistent bleeding or cardiac tamponade. One patient survived at 30 days and 1 year. In the off-pump tissue sealing group 30 days and 1 year survival was 100%. Postoperative echocardiography showed normal left ventricular (LV) function with no regional wall motion abnormalities. Conclusions. Our data show that epicardial tissue sealing off-pump results in successful termination of bleeding from AVD and considerably improves survival when compared with the standard procedure. Because of this tremendous improvement in patient survival we now consider this technique as standard therapy for AV disruption in our center.
A
such as biodegradable scaffolds has been shown to provide excellent hemostasis even in augmented form in both experimental and clinical settings [4, 5]. The devastating results in patients with AVD prompted us to investigate the outcome in a series of 9 patients with documented AVD complicating mitral valve procedures with regard to 1-year morbidity and mortality. The first 4 patients were treated with surgical repair on-pump. The second series of 5 patients were treated off-pump with a method of epicardial tissue sealing in layers. Our data show that off-pump epicardial tissue sealing results in successful termination of bleeding from AVD and considerably improves survival when compared with surgical repair. Because of this tremendous improvement in patient survival we now consider this technique as standard therapy for AV disruption in our center.
trioventricular disruption (AVD) is a rare but often fatal complication following mitral valve procedures. It occurs in approximately 1%–2% of patients and the mortality associated with it is beyond 50% [1, 2]. The complication often results from overly aggressive debridement and decalcification of the posterior mitral valve annulus and subvalvular apparatus or from oversizing the prosthesis [3]. The clinical picture varies from a hematoma of the posterior AV groove to cardiac rupture with massive bleeding. Quite frequently it is followed by compromised blood flow to the circumflex artery with signs of posterolateral wall motion abnormalities or left ventricular pump failure. To date, the standard procedure for this complication is to go back on-pump with cardioplegic arrest, reopen the left atrium, remove the prosthesis, and close the tear with patch or buttress suture. However, even with the attempt of surgical repair the mortality rate remains high because of persistent bleeding as well as prolonged cardiopulmonary bypass (CPB) and aortic cross-clamp time associated with pump-related complications. The development of biological tools (hemostyptics, sealants) Accepted for publication Feb 6, 2004. Address reprint requests to Dr Wildhirt, German Heart Center Munich, Technical University of Munich, Lazarettstr. 36, 80636 Munich, Germany; e-mail:
[email protected].
(Ann Thorac Surg 2004;78:569 –74) © 2004 by The Society of Thoracic Surgeons
Material and Methods Patients Between March 1998 and October 2002 a total of 723 patients received single our combined mitral valve procedures in our center. In 9 patients (1.2%) the complication of atrioventricular disruption (AVD) was diagnosed intraoperative.
© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
0003-4975/04/$30.00 doi:10.1016/j.athoracsur.2004.02.029
CARDIOVASCULAR
Off-Pump Epicardial Tissue Sealing—A Novel Method for Atrioventricular Disruption Complicating Mitral Valve Procedures
570
SCHUETZ ET AL EPICARDIAL TISSUE SEALING FOR ATRIOVENTRICULAR DISRUPTION
Ann Thorac Surg 2004;78:569 –74
Table 1. Patient Demographics
Patient Demographics CARDIOVASCULAR
Age (years) Height (cm) Weight (kg) Sex (% male) Mean CV risk factors Preoperative data Left ventricular function (EF%) NYHA Serum creatinine (mg/dl) Hemoglobin (g/l) Hematocrit (%) WBCs (G/l) Platelets (G/l) Intraoperative data Lowest temperature (°C) Pump time (min) Aortic cross clamp time (min) Re CPB, pump time (min) Re CPB, clamp time (min) Cell saved blood (ml) Postoperative data Intubation time (hours) ICU stay (days) Blood loss (ml/24h) Postoperative EF (%) Need for IABP n, (%) Peak inotropic support Epinephrine (mg/h) Norepinephrine (mg/h) Milrinone (mg/h) Survival (n; %) 30 days Cause of death Acute cardiac tamponade Massive bleeding NYHA class at 1 year follow-up EF (%) at 1 year follow-up
Surgical Closure (n ⫽ 4)
Tissue Sealing (n ⫽ 5)
p Value
68 ⫾ 4 174 ⫾ 12 76 ⫾ 5 25 2.2 ⫾ 1.1
69 ⫾ 7 166 ⫾ 10 67 ⫾ 12 40 2.3 ⫾ 0.8
0.79 0.35 0.25 0.46 0.34
55 ⫾ 5 3.3 1.2 ⫾ 0.2 11.3 ⫾ 2.5 31.1 ⫾ 7.2 6.9 ⫾ 0.1 199 ⫾ 41
56 ⫾ 19 3.2 1.13 ⫾ 0.1 9.6 ⫾ 1.7 26.7 ⫾ 4.2 10.6 ⫾ 6 234 ⫾ 62
0.92 0.72 0.54 0.22 0.34 0.35 0.43
32 135 ⫾ 9 76 ⫾ 19 56 ⫾ 15 38 ⫾ 11
32.3 ⫾ 0.5 125 ⫾ 37 73 ⫾ 20
0.54 0.84 0.79 n.a. n.a. n.a.
1300 ⫾ 900 240 (n ⫽ 1) 93 ⫾ 22 14 (n ⫽ 1) 8.4 ⫾ 4.7 3930 ⫾ 233 1539 ⫾ 121 25 (n ⫽ 1) 58 ⫾ 9 2, (50) 0, (0)
n.a. n.a. ⬍0.001 0.16
0.4 ⫾ 04
0.27 ⫾ 0.25
1.4 ⫾ 0.4 10 ⫾ 2 1; 25
09 ⫾ 0.1 2.5 ⫾ 0.5 5; 100
0.06 0.05 ⬍0.01
2.2 61 ⫾ 13
n.a. n.a. n.a. n.a.
1 2 2 (n ⫽ 1) 55 (n ⫽ 1)
0.62
CPB ⫽ cardiopulmonary bypass; CV ⫽ cardiovascular; EF ⫽ ejection fraction; IABP ⫽ intraaortic balloon pump; n.a. ⫽ not applicable; NYHA ⫽ New York Heart Association; WBCs ⫽ white blood cells.
The first 4 patients (group: surgical repair) received elective surgery for single mitral valve repair (n ⫽ 3) and combined aortic and mitral valve replacement (n ⫽ 1) for postrheumatic disease. Relevant patient demographics are given in Table 1. Surgical repair was performed on-pump with aortic cross clamping, cardioplegic arrest (Bredtschneider’s cardioplegic solution) under moderate systemic hypothermia (30°–32°C). The left atrium was reopened and the previously implanted prosthesis (valve or ring) was removed. A tear in the posterior aspect of the mitral valve annulus was found and repaired with prosthetic patch and buttressed suture (n ⫽ 3). The left atrium was closed and the patients weaned from bypass. In 1 patient addi-
Fig 1. Placement in layers of the collagen system with fibrinogen based coating (Tacho Comb, Nycomed Pharma, Linz, Austria) is shown on the posterolateral aspect of the heart. Each layer is applied by manual compression for 10 –15 minutes starting in the center of the bleeding sit. Between 3– 6 layers may be necessary to achieve hemostasis. The layers are placed in overlapping fashion thereby creating a large area of mechanical resistance against intracardiac bleeding.
tional surgical repair was attempted on-pump using 4-0 Prolene sutures (Ethicon, Somerville, NJ) with pledges placed from the epicardial side. In all patients moderate doses of inotropic agents, phosphodiesterase inhibitors, and vasopressors were required to maintain acceptable hemodynamics. In addition, intraaortic balloon counterpulsation (IABP) was used in 2 patients to maintain stable cardiovascular hemodynamics (Table 1).
Repair Off-Pump With Epicardial Tissue Sealing The other 5 patients (group: tissue sealing) were referred to our hospital for mitral valve surgery because of mitral valve endocarditis (2 patients), mitral valve regurgitation due to chorda rupture (2 patients), and postrheumatic mitral valve stenosis in combination with coronary artery disease (1 patient), (group B; Table 1). When AVD was diagnosed and other bleedings sites had been excluded, these patients were left off-pump and protamine was administered. A cell saver was initiated and cell saved blood was washed and retransfused. Layers of a biodegradable collagen-system with fibrinogen-based coating (Tacho Comb, Nycomed Pharma, Linz, Austria) were carefully placed on the posterolateral epicardium starting from the center of the bleeding site (Fig 1). Manual pressure with a moist gauze for 10 –15 minutes on the posterolateral aspect of the beating heart was applied. This procedure was repeated until hemostasis was achieved. In all patients we observed recurrent bleeding from the edge of each collagen fleece applied. Therefore, it was of utmost importance to spend enough time for the manual pressure procedure of each layer until the collagen-system exerted the best gluing effect on the epicardium. In addition, the manual pressure was performed using moisturized gauze that prevented the collagen system from sticking to the surgeon’s glove rather than the epicardium. We used 3– 6 single layers
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
Ann Thorac Surg 2004;78:569 –74
SCHUETZ ET AL EPICARDIAL TISSUE SEALING FOR ATRIOVENTRICULAR DISRUPTION
571
Statistics Data are presented as mean ⫾ standard deviation (SD). Dichotomous variables were analyzed by the 2 test and the Fisher exact test. The Student t test for unpaired variables was used to compare clinical parameters. A P value of less than 0.05 was considered statistically significant.
Results Early and 1-Year Morbidity and Mortality In the surgical repair group, 1 patient (25%) survived at 30 days and 1 year follow-up (Table 1, Fig 2). In this patient the endocardial injury was in the posterior region 2 (p2) of the mitral valve annulus due to rupture of a suture of the implanted device. It was successfully repaired with two single sutures with pledges. The patient survived at 30 days. At 1 year follow-up she presented with NYHA class II and an ejection fraction (EF) of 55%. The remaining 3 patients in this group died because of continued bleeding and cardiac tamponade within the first 24 hours after the initial mitral valve procedure. In these patients the sites of bleeding were at the region of p2–p3 due to a tear adjacent to the sewing ring (n ⫽ 2) and in a decalcified area (n ⫽ 1) which were repaired on-pump with patch and buttressed sutures. In the off-pump “tissue sealing” group all 5 patients (100%) survived 30 days and 1 year follow-up (Fig 2). Echocardiographic evaluation at 30 days and 1 year follow-up did not reveal any signs for regional wall motion abnormalities. The average EF was 61% ⫾ 13% with an average NYHA class of 2.2. It is important to note that the epicardial bleeding sites with regard to location and severity did not differ among both groups.
CARDIOVASCULAR
per patient to achieve complete hemostasis. With each overlapping layer the epicardial area covered with the collagen fleece increased resulting in decreased bleeding. The total time for the procedure was 45–90 minutes. However, the method was applied off-pump under normothermic conditions avoiding the risk of pump-related complications. Once hemostasis was achieved, manual pressure was released and the heart was left in its regular position for an additional 15–20 minutes to ensure sufficient hemostasis. Thereafter the chest was closed in routine fashion with stainless steel wires and patients were brought to the intensive care unit (ICU) in stable condition with low-to-moderate amounts of inotropic agents. During the early postoperative course echocardiography was used for the evaluation of cardiac function, left ventricular (LV) wall motion, and the presence of intracardiac hematoma or pericardial effusion. The technique described here was adopted from our centers experience in treating other bleeding sites (i.e. smaller epicardial lacerations or injuries to the lungs in redo operations) with this technique in combination with what has been published in the literature [5] Fig 2. Survival at 1-year follow-up is shown. No deaths occurred in the “tissue sealing” group compared with 25% survival in the surgical repair group. *p ⬍ 0.05 versus surgical repair.
Postoperative Inotropic Support and Requirement for Blood Products As shown in Figure 3, patients in the surgical repair group received significantly more units of fresh frozen plasma (FFP) and red blood cells during the intraoperative and postoperative course. In addition, this group presented with a significant increased chest tube loss when compared with the tissue sealing group (Table 1). With regard to cardiovascular hemodynamics, the “surgical repair” group required higher doses of milrinone. However, during the postoperative course the peak doses of epinephrine and norepinephrine administered were not statistically significant (Table 1).
Postoperative Evaluation of Cardiac Function, LV Wall Motion, and Intracardiac Hematoma After Epicardial Tissue Sealing Various echocardiographic evaluations during early follow-up were performed. We focused on global cardiac function, LV wall motion abnormalities (particularly in the posterolateral region), development of LV intracardiac hematoma, and the development of pericardial effusion. Postoperative cardiac function was normal in both groups (Table 1). In addition, no specific regional wall abnormality suggestive of compromised coronary blood flow was found in the epicardial tissue sealing group. Moreover, we did not find specific signs of intracardiac hematoma indicative for prolonged bleeding from the site of AVD. In 1 out of 5 patients in the epicardial tissue sealing group a pericardial effusion of
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
572
SCHUETZ ET AL EPICARDIAL TISSUE SEALING FOR ATRIOVENTRICULAR DISRUPTION
Ann Thorac Surg 2004;78:569 –74
Comment
CARDIOVASCULAR Fig 3. The amounts of blood products given intraoperatively (top) and during the postoperative (bottom) course. In the “surgical repair” group a significantly higher number of units of fresh frozen plasma (FFP) were given intraoperatively and postoperatively as well as more units of red blood cells (RBCs) postoperative when compared with the “tissue sealing” group. *p ⬍ 0.05 versus surgical repair. (White bars ⫽ tissue sealing; black bars ⫽ surgical repair.)
350 mL was drained on postoperative day 10. However, the patient did not present with any signs of cardiac tamponade and the effusion was diagnosed during echocardiographic follow-up.
The complication of AVD after mitral valve procedures is rare but often fatal. The mortality remains beyond 50%. We report here the first series of 5 patients with highly significant improvement of survival after AVD using a method of epicardial tissue sealing off-pump with a biodegradable collagen system with fibrinogen-based coating (Tacho Comb). The major reasons for this tremendous improvement are the successful termination of bleeding and the avoidance of additional pump and clamp time thereby preventing prolonged myocardial ischemia. Because of this tremendous improvement in patient survival we now consider this technique as standard therapy for AV disruption in our center. To date, surgical repair on-pump remains the only therapeutic option for patients with AVD. However, the attempt to close and repair the defect surgically from the endocardial and/or epicardial side remains technically demanding because the tissues are fragile and often severely damaged by the dissecting hematoma. Failure of this procedure remains high because of a variety of reasons. Potential explanations are extended tissue rupture, persistent or even increased bleeding, extensive overall surgery and pump time as well as prolonged myocardial ischemia, and increased amounts of blood products and catecholamines administered [1]. The exact pathologic mechanism for AVD remains speculative. In this regard, three sites of AVD have been described. Type 1 occurs at the AV annulus because of excessive debridement of calcium or leaflet resection, resulting in perforation of the thin attachment of the atrium to the ventricle in the posterior atrioventricular sulcus. Type II describes transverse ventricular disruption secondary to systolic forces applied to thinned-out myocardium when the posterior papillary muscle and chordae are resected. Type III describes a midventricular rupture that may be secondary to overzealous resection of the posterior papillary muscle. In the surgical repair group, 3 out of 4 patients received mitral valve repair and the resulting AVD may have been of type I. In the tissue sealing group, patients received mitral valve replacement. In all patients, both the posterior leaflet and papillary muscle were left in place suggesting a type I AVD. However, because of the nature of the off-pump sealing technique, the underlying cause of the initial endocardial injury cannot be addressed and it may be of importance to describe the anatomic location of the epicardial tear. Our observations in the 9 patients suggest that bleeding occurred primarily in the posterior aspect of the left ventricle spreading either toward the left ventricular apex or the posterior or posterolateral AV groove. In the tissue sealing group, the progression of bleeding depended on the way the initial fleece was placed. When the fleece layers were placed starting more posterior, the area of bleeding moved toward the posterolateral groove and it was easier to apply the following layers. The repair process may require time and the surgeon’s
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
SCHUETZ ET AL EPICARDIAL TISSUE SEALING FOR ATRIOVENTRICULAR DISRUPTION
patience. In 3 out of 5 patients we observed massive bleeding from the posterior aspect of the AV groove. A total of 1300 ⫾ 900 ml of blood was cell saved throughout the repair procedures. In these patients a total of 3– 6 large Tacho Comb layers were placed in overlapping fashion to cover a large area of myocardium with the original bleeding site in its center. Manual compression for up to 10 –15 minutes for each layer may be necessary to achieve hemostasis. The method of off-pump sealing of the epicardial tissue in up to 6 layers provides sufficient mechanical resistance to stabilize the injured transmyocardial tissue over a large area of the heart.
The Concept of Biological Repair of Larger Area Defects The surgical approach for AVD repair remains very difficult or may even be impossible because the dissecting tissue defect is usually quite large. Single or multiple sutures with pledges from the epicardial side are often insufficient to achieve hemostasis. Repair of the thin myocardium from the endocardial side requires additional pump and ischemic time and is often complicated by the massive tissue damage and hematoma. Tissue sealing with biological and synthetic biodegradable scaffolds is especially suitable for managing largerarea defects. With particular focus on the heart as a moving organ, one major aspect in tissue sealing is the adhesive strength of the material (as a suture would provide) necessary to achieve hemostasis but at the same time remaining flexible enough to adapt to the heart’s shape change during every beat. Our results support the observation published in a recent study by Carbon and colleagues; they demonstrated that the ready-to-use collagen system with fibrinogen-based coating (Tacho Comb) provides powerful gluing capacity and withstands high-pressure conditions in patients with recurring pneumothorax [4]. The same group tested various forms of biomaterials with regard to their adhesive strength. Using a biosimulator, it was found that liquid sealants achieve only low (fibrin glue: 4.1 hPa) adhesive strength. Moreover, independent of the fleece material employed, prepare-to-use techniques (fleece ⫹ fibrin glue) achieve moderate adhesive strength (22.3–25.3 hPa). In contrast, the ready-to use systems (Tacho Comb) provide biomechanically relevant adhesive strength (50.2 hPa). The knowledge of the adhesive strength may be of impor-
573
tance in the clinical setting because different tissues (dynamic: heart, lung; static: liver, spleen, kidney) and defects may be treated with different sealants. For example, fibrin glue may be more suitable to seal small tissue defects rather than to provide large area hemostasis on dissected and fragile tissues [6] As shown in our series, sealing large area defects of the human myocardium is successfully achieved by the biodegradable collagen system with fibrinogen-based coating (Tacho Comb) even when applied on the beating heart and under systemic blood pressure conditions. This technique may also be applicable for other large area defects on the heart such as postinfarction ventricular rupture, where ischemic tissues are difficult to suture. This needs to be further evaluated. In summary, the present report shows that off-pump biological tissue repair of the left ventricle is possible even in the presence of large area defects. Because of the excellent long-term outcome in 5 consecutive patients with AVD we have changed the policy for the treatment of AVD at our center. We strongly recommend this procedure for the treatment of patients with AVD or other large area defects on the heart.
References 1. Bjork V, Henze A, Rodriguez L. Left ventricular rupture as a complication of mitral valve replacement. J Thorac Cardiovasc Surg 1977;73:14 –22. 2. Karlson K, Ashraf M, Berger R. Rupture of left ventricle following mitral valve replacement. Ann Thorac Surg 1988;46: 590 –7. 3. Savage D, Garrison R, Brand F, Anderson S, Castelli W, Kannel W, et al. Prevalence and correlates of posterior extra echocardiographic spaces in a free-living population based sample (the Framingham study). Am J Cardiol 1983;51:1207– 12. 4. Carbon R, Baar S, Gusinde J, Huemmer H, Baer K. Tissue sealing concept in minimally invasive surgery in children. Pediatr Endosurg Innovat Tech 2001;5:5–12. 5. Schelling G, Block T, Gokel M, Blanke E, Hammer C, Brendel W. Application of a fibrinogen-thrombin-collagen based hemostyptic agent in experimental injuries of the liver and spleen. J Trauma 1988;28:472–5. 6. Carbon R, Baar S, Kriegelstein S, Huemmer H, Baar K, Simon S. Evaluating the in vitro adhesive strength of biomaterials. Biosimulator for selective leak closure. Biomaterials 2003;24: 1469 –75.
INVITED COMMENTARY Posterior rupture of the left ventricle has been a recognized complication since the origins of mitral valve surgery, and ironically, despite advances in virtually all aspects of the treatment of valvular disease, the outcome of left ventricular rupture has remained exceptionally poor. What is understood about left ventricular rupture after mitral valve surgery can be gained from the many sporadic reports that have been published over the span of more than 30 years. Classification of the ventricular rupture is based on the anatomic site. Ruptures in the
atrioventricular groove (type I) or at the base of the papillary muscles (type II) are thought to be secondary to surgical trauma. The transverse midventricular tear (type III), which occurs midway between the first two, is believed to be the result of dividing the anatomic loop created by the papillary muscle, mitral valve, and posterior ventricular wall. Modification of the technique of mitral valve surgery primarily based on the preservation of the posterior leaflet has been reported to lower the risk of ventricular rupture.
© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
0003-4975/04/$30.00 doi:10.1016/j.athoracsur.2004.03.054
CARDIOVASCULAR
Ann Thorac Surg 2004;78:569 –74
Off-pump epicardial tissue sealing—a novel method for atrioventricular disruption complicating mitral valve procedures Albert Schuetz, Costas Schulze and Stephen M. Wildhirt Ann Thorac Surg 2004;78:569-573 Updated Information & Services
including high-resolution figures, can be found at: http://ats.ctsnetjournals.org/cgi/content/full/78/2/569
References
This article cites 6 articles, 2 of which you can access for free at: http://ats.ctsnetjournals.org/cgi/content/full/78/2/569#BIBL
Citations
This article has been cited by 5 HighWire-hosted articles: http://ats.ctsnetjournals.org/cgi/content/full/78/2/569#otherarticles
Subspecialty Collections
This article, along with others on similar topics, appears in the following collection(s): Cardiac - other http://ats.ctsnetjournals.org/cgi/collection/cardiac_other
Permissions & Licensing
Requests about reproducing this article in parts (figures, tables) or in its entirety should be submitted to: http://www.us.elsevierhealth.com/Licensing/permissions.jsp or email:
[email protected].
Reprints
For information about ordering reprints, please email:
[email protected]
Downloaded from ats.ctsnetjournals.org by on June 8, 2013
