- Email: [email protected]
Modified Single Patch: Are We Still Worried About Subaortic Stenosis?
Carl L. Backer, MD, Osama Eltayeb, MD, Michael C. Mong e, MD, Katherine C. Wurlitzer, BA, Madelaine A. Hack, BA, Lindsay H. Boles, BA, Anne E. Sarwark, MS, John M. Costello, MD, MPH, and Joshua D. Robinson, MD Divisions of Cardiovascular-Thoracic Surgery and Cardiology, Ann & Robert H. Lurie Children’s Hospital of Chicago; and Departments of Surgery and Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, Illinois
Background. When the modified single-patch technique for atrioventricular septal defect (AVSD) repair was introduced by Dr Benson Wilcox, there was concern that these patients might be at risk for late subaortic stenosis and left ventricular outflow tract obstruction (LVOTO). This review evaluated our modified singlepatch population for LVOTO in the postoperative period. Methods. Between January 2000 and 2013, 77 infants underwent AVSD repair with a modified single-patch technique. Median age was 4.2 months, and median weight was 5 kg. Eight patients had a prior repair of coarctation of the aorta via left thoracotomy in the newborn period. Results. The median hospital stay was 10 days. No patient required a pacemaker. The mean and median follow-up times were 4.6 and 3.7 years, respectively. Only 2 patients (2.5%) required reoperation for LVOTO; both
had prior repair of coarctation of the aorta (2 of 8 vs 0 of 69, p [ 0.01). A discrete fibrous subaortic membrane developed in the first patient that required resection at 3 and 7 years after repair. The other patient had LVOTO from accessory chordae of the left atrioventricular valve and required mitral valve replacement 5 months after repair. One early death occurred at 4 months postoperatively due to liver failure related to hyperalimentation. Conclusions. At intermediate term follow-up, LVOTO does not appear to be a significant postoperative issue after modified single-patch repair of AVSD. Coarctation of the aorta was the most significant predictor of late LVOTO after repair of AVSD with the modified single-patch technique.
T
obstruction . . .” [5]. However, there were no reoperations for left ventricular outflow tract obstruction (LVOTO) in the series by Nicholson and colleagues [2]. The potential for LVOTO using the modified singlepatch technique continues to be raised by surgeons using the “classic” single and 2-patch techniques and by some cardiac morphologists [6, 7]. Proponents of the 2patch technique contend that the patch placed to close the ventricular component of the AVSD helps to contribute to the width of the left ventricular outflow tract. They contend that pulling the valve leaflets down to the septum using the modified single-patch technique will encroach upon the left ventricular outflow tract and narrow it. Another issue raised relates to the known fact that LVOTO is more common in partial than complete AVSD [8]. The concern is that using the modified singlepatch technique essentially converts a patient with a complete AVSD to a partial AVSD [9]. This could potentially increase the risk of developing LVOTO. We previously published our results comparing the modified single-patch technique with the 2-patch technique [3]. At that time we reported 26 patients undergoing the modified single-patch technique. No patient in that small group required reoperation for LVOTO at short-term follow-up. We have now tripled the size of that initial series and have follow-up of more than 10 years. Our hypothesis was that the modified
he modified single-patch technique for repair of complete atrioventricular septal defect (AVSD) was independently proposed by Wilcox and colleagues [1] and Nicholson and colleagues [2]. The change in strategy from the classic 2-patch technique was to perform direct closure of the ventricular element, eliminating the patch used for closure of the ventricular component [3]. When Wilcox presented his concept at The Society of Thoracic Surgeons meeting in 1997, the second question during the discussion was, “Did you see any subsequent increased flow velocity in the left ventricular outflow tract indicative of obstruction?” [4]. The answer was: “we find no evidence of that [LVOTO] whatsoever. I think it has less to do with the particular repair that is used than with the basic anatomy. “ When the Nicholson et al [2] paper was presented at the American Association for Thoracic Surgery meeting in 1999, Dr John Brown noted, “it would seem that this [repair] would increase the likelihood of LVOTO
Accepted for publication Jan 6, 2015. Presented at the Sixty-first Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 5–8, 2014. Address correspondence to Dr Backer, Division of CardiovascularThoracic Surgery, 225 E Chicago Ave, MC 22, Chicago, IL 60611; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;99:1671–6) Ó 2015 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.01.032
CONGENITAL HEART
Modified Single Patch: Are We Still Worried About Subaortic Stenosis?
1672
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
CONGENITAL HEART
(simplified) single-patch technique would not be associated with an increased risk of LVOTO (subaortic stenosis).
Material and Methods The Ann & Robert H. Lurie Children’s Hospital of Chicago Institutional Review Board approved this study on June 12, 2012, as a retrospective record analysis and waived the need for patient consent. The congenital cardiac surgery database was analyzed, and all patients undergoing repair of complete AVSD between 2000 and 2013 were included. We excluded AVSD with tetralogy of Fallot, AVSD with double-outlet right ventricle, and unbalanced AVSD. We analyzed the serial postoperative echocardiograms of each patient to assess the degree of LVOTO. Statistical analysis was performed using the Fisher exact test. Seventy-seven patients underwent repair of complete AVSD using a modified single-patch technique at our institution from 2000 to 2013. We previously reported our specific surgical technique [3]. During that period, we transitioned from the 2-patch technique, as shown in Figure 1. Eight patients (10%) had undergone a previous operation for coarctation (CoAo) repair in the newborn period. All CoAo repairs were done via left thoracotomy using resection with extended end-to-end anastomosis.
Results The median age at operation was 4.2 months (mean, 4.8 2.4 months) and the median weight was 5 kg. In this series, 57 patients (74%) had trisomy 21. No patient had a procedure for LVOTO as part of the primary procedure. Rastelli classification was: A in 39, B in 3, C in 16, and unknown in 2. The VSD size was assessed by measuring the distance from the crest of the VSD to the common AV valve in diastole. We considered more than 12 mm to be an indication for a 2-patch repair. Only two patients since 2006 have had a 2-patch repair, both because of a very large ventricular component of 12 and 16 mm. Fig 1. Graph shows the evolution of the 2patch technique (black bars) to the modified single-patch technique (gray bars) at Ann & Robert H. Lurie Children’s Hospital of Chicago during a 13-year period (1997 to 2014). The 2-patch technique has been used in only 2 patients in the past 9 years.
Ann Thorac Surg 2015;99:1671–6
Mean cross-clamp time for the AVSD repair was 103 23 minutes. Mean cardiopulmonary bypass time was 135 33 minutes. Median length of stay was 10 days. No child required a pacemaker in the postoperative period. One early death occurred 4 months postoperatively as a result of liver failure in a child awaiting a liver transplant. Liver failure was secondary to long-term preoperative and postrepair hyperalimentation. At a mean follow-up of 4.2 years, 58 patients with at least 1 year of echocardiographic follow-up had no evidence of LVOTO by serial echocardiography (Fig 2). LVOTO developed in 4 patients, 3 of whom had prior CoAo repair: 3 of 8 (38%) vs 1 of 69 (1%; p ¼ 0.007 by Fisher exact test; Fig 3). Only 2 patients required reoperation for LVOTO, both had prior CoAo repair: 2 of 8 (25%) vs 0 of 69 (p ¼ 0.01 by Fisher exact test). The patient who died of liver failure 4 months postoperatively had mild LVOTO, which was unchanged on serial echocardiograms up until the time of death while awaiting liver transplant. A patient who underwent repair in 2001 had stable mild LVOTO for 11 years, with no change in gradient. That patient had repair of CoAo as an infant. A patient who underwent AVSD repair in 2003 at age 4 months had prior repair of CoAo as a newborn. This child started with mild LVOTO that progressed to moderate to severe obstruction at 3 years of age. At that time, reoperation was performed, and a discrete subaortic membrane was resected. The postoperative LVOTO was initially mild but during the next 5 years progressed to become severe again. A second reoperation was performed, and a recurrence of the fibrous subaortic membrane was resected. The child now has mild LVOTO, which has been stable. The final patient with LVOTO underwent CoAo repair at age 1 week and AVSD repair at age 3 months. After this repair, there was moderate LVOTO from substantial crowding of the LVOTO by accessory chords from the left atrioventricular valve. This progressed over a short period of time, and at 4 months postoperatively, the child underwent attempted resection of chordae but required
Ann Thorac Surg 2015;99:1671–6
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
1673
mitral valve replacement (#17 St. Jude; St. Jude Medical, St. Paul, MN). The child now is 2 years post-repair and has only mild LVOTO, which has been stable. All patients with LVOTO had Rastelli classification type A. In addition, all patients with CoAo and
complete AVSD had Rastelli type A. The age at repair of the patients in whom LVOTO developed was a median of 6 months. The age of repair of the children who did not develop LVOTO was 5 months (p ¼ not significant).
Fig 3. The time course is shown of the 4 patients in whom left ventricular outflow tract obstruction (LVOTO) developed after modified single-patch repair of complete atrioventricular septal defect (AVSD). (COA ¼ coarctation of the aorta; MVR ¼ mitral valve replacement.)
CONGENITAL HEART
Fig 2. This graph shows the number of years of serial echocardiographic follow-up on sequential complete atrioventricular septal defect patients who underwent a modified single-patch repair and had no evidence by echocardiography of left ventricular outflow tract obstruction (LVOTO). Only patients with at least 1 year of postoperative follow-up by echocardiogram were included.
1674
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
CONGENITAL HEART
Comment We have reviewed the outcomes of 77 infants undergoing repair of complete AVSD with a modified (simplified) single-patch technique during a 13-year period. At intermediate follow-up, LVOTO does not appear to be a significant postoperative issue after the modified singlepatch technique. Only 2 patients in this series required reoperation for LVOTO. The most significant predictor of late LVOTO after AVSD repair was the association of CoAo. These patients are known to have small left ventricular outflow tracts even when the intracardiac anatomy is otherwise normal [10]. To answer the question raised in the title of our report, we are no longer particularly worried about subaortic stenosis in patients undergoing the modified single-patch technique. We switched to the modified single-patch technique after many years of what we believed was successful application of the 2-patch technique [3]. Analysis of a comparison of the two techniques indicated that the modified single-patch technique not only had shorter cardiopulmonary bypass and cross-clamp times but resulted in improved outcomes with regard to left atrioventricular valve insufficiency [11]. That has been the Achilles heel of patients with complete AVSD undergoing any kind of repair [12]. Adachi and colleagues [6] from the Royal Brompton recently examined 43 heart specimens with AVSD. They particularly evaluated the left ventricular outflow tract diameter in relationship to the anterosuperior extension of the ventricular component of the AVSD. They concluded that the anterosuperior and skewed scoop could lead to asymmetric configuration of the valvar leaflets and LVOTO if the simplified single-patch technique was applied. They were particularly concerned with the Rastelli type C patients. In an invited commentary regarding this article, the primary limitation noted was a deficiency of clinical material presented to support the theoretical consideration [13]. That commentary combined the results from the Michigan group, [14] Ann & Robert H. Lurie Children’s Hospital [3], the updated series from Nunn [15], and a series published by Jonas and Mora [16]. Combining those series resulted in 200 patients undergoing the modified single-patch technique, with no patients at that point requiring reoperation for LVOTO. The conclusion of the commentary was, “although it may be counterintuitive, the synthetic patch placed in the two-patch technique may actually promote fibrosis that may lead to obstruction of the left ventricular outflow tract because of the increased rigidity to blood flow, and presence of a foreign body in the left ventricular outflow tract.” Of interest, in our series all of the patients developing LVOTO had a Rastelli classification A. Kanani and associates [17] also responded to the Adachi article with a letter to the editor, “Left ventricular outflow tract obstruction after the modified single patch repair of atrioventricular septal defects: teasing fact from fiction.” They said, “we believe this is an assertion taken too far and that it diverts the gaze away from other more pertinent morphologic variants that have proven their
Ann Thorac Surg 2015;99:1671–6
association in the clinical arena.” They cited other factors known to cause LVOTO, such as obstructive subvalvar apparatus, dysplastic leaflet material, and papillary muscles that extend into the left ventricular outflow tract. Myers and colleagues [7] from Children’s Hospital Boston reviewed 35 children with complete AVSD who underwent reoperation for LVOTO. Of these, 27 patients had prior double-patch repair, 6 had prior classic singlepatch repair, and 5 had a prior modified single-patch repair. One of their conclusions was that the modified single-patch repair for AVSD was associated with a higher incidence of reoperation for LVOTO (p ¼ 0.4). After analyzing their data, we believe what it actually shows is that LVOTO develops slightly sooner after the modified single-patch technique than if one of the other techniques is applied. However, because there is no denominator for any of the groups, reaching a conclusion about the relative incidence of LVOTO is hard. In the discussion of that paper, Myers noted that, “the multivariable analysis did not confirm the modified single-patch technique as an independent predictor of reoperation. It was only a predictor in the univariable analysis” [18]. In a review article on reoperations for LVOTO after repair of AVSD, Overman [19] emphasized the importance of the anatomy of the left ventricular outflow tract. The main factors predisposing to LVOTO in AVSD were unwedging of the aortic root (anterior and rightward), disparate inlet and outlet septum lengths (short inlet, long outlet), and deficiency of the muscular interventricular septum (septal “scoop”). Overman also noted the important contribution to LVOTO of the anterolateral muscle bundle of the left ventricle (AML) in AVSD patients. He noted that resection of this hypertrophic AML should be considered for operations for LVOTO after repair of AVSD. The patient in our series who required mitral valve replacement to treat LVOTO is exactly this type of patient. The final conclusion was: “Placement of a straight patch to close a large VSD may unfavorably alter the geometry and flow dynamics within the distal left ventricular outlet tract. It is conceivable that the stiff material used to reconstruct the distal left ventricular outflow tract may provide a substrate for LVOTO.” A technical point that is important in the surgical decision making for the modified single-patch technique is the location of the division of the superior bridging leaflet into the left and right atrioventricular valves. Our policy has been to “steal” from the right and give to the left, as originally recommended by Rastelli and colleagues [20]. The portion of the superior bridging leaflet that is adjacent to the ventricular septum after repair will become the de facto left ventricular outflow tract. Giving as much of this leaflet as possible to the left side will help prevent LVOTO. This relationship was noted by Ebels and coworkers [21] when they stated, “the extent of the attachment between the superior bridging leaflet and the septal crest is of crucial importance in determining the potential for obstruction in the left ventricular outflow tract in AVSD.” The only patients in whom we do not use the modified single-patch technique are those with a ventricular component exceeding 12 mm.
Ann Thorac Surg 2015;99:1671–6
4. 5. 6. 7.
8.
9. 10.
11. 12. 13.
14. 15. 16.
17.
18. 19. 20.
References 1. Wilcox BR, Jones DR, Frantz EG, Brink LW, Henry GW, Mill MR, Anderson RH. Anatomically sound, simplified approach to repair of “complete” atrioventricular septal defect. Ann Thorac Surg 1997;64:487–94; discussion 493–4. 2. Nicholson IA, Nunn GR, Sholler GF, et al. Simplified single patch technique for the repair of atrioventricular septal defect. J Thorac Cardiovasc Surg 1999;118:642–6. 3. Backer CL, Stewart RD, Bailliard F, Kelle AM, Webb CL, Mavroudis C. Complete atrioventricular canal: comparison
21.
22.
1675
of modified single-patch technique with two-patch technique. Ann Thorac Surg 2007;84:2038–46. Ebels T. Discussion of reference 1. Brown J. Discussion of reference 2. Adachi I, Ho SY, McCarthy KP, Uemura H. Ventricular scoop in atrioventricular septal defect: relevance to simplified single-patch method. Ann Thorac Surg 2009;87:198–203. Myers PO, del Nido PJ, Marx GR, et al. Improving left ventricular outflow tract obstruction repair in common atrioventricular canal defects. Ann Thorac Surg 2012;94: 599–605. Piccoli GP, Ho SY, Wilkinson JL, Macartney FJ, Gerlis LM, Anderson RH. Left-sided obstructive lesions in atrioventricular septal defects: an anatomic study. J Thorac Cardiovasc Surg 1982;83:453–60. Stulak JM, Burkhart HM, Dearani JA. Reoperations after repair of partial and complete atrioventricular septal defect. World J Ped Congenit Heart Surg 2010;1:97–104. Kalfa D, Ghez O, Kreitmann B, Metras D. Secondary subaortic stenosis in heart defects without any initial subaortic obstruction: a multifactorial postoperative event. Eur J Cardiothorac Surg 2007;32:582–7. Backer CL, Stewart RD, Mavroudis C. What is the best technique for repair of complete atrioventricular canal? Semin Thorac Cardiovasc Surg 2007;19:249–57. Crawford FA Jr, Stroud MR. Surgical repair of complete atrioventricular septal defect. Ann Thorac Surg 2001;72:1621–9. Backer CL. Invited Commentary for Adachi I, Ho SY, McCarthy KP, Uemura H. Ventricular scoop in atrioventricular septal defect: relevance to simplified single-patch method. Ann Thorac Surg 2009;87:198–203. Suzuki T, Bove EL, Devaney EJ, et al. Results of definitive repair of complete atrioventricular septal defect in neonates and infants. Ann Thorac Surg 2008;86:596–602. Nunn GR. Atrioventricular canal: modified single patch technique. Semin Thorac Cardiovasc Surg Ped Card Surg Annu 2007;10:28–31. Jonas RA, Mora B. Individualized approach to repair of complete atrioventricular canal: selective use of the traditional single-patch technique versus the Australian technique. World J Pediatr Congenit Heart Surg 2010;1:78–86. Kanani M, Cook A, Kostolny M. Left ventricular outflow tract obstruction after the modified single patch repair of atrioventricular septal defects: teasing fact from fiction. Ann Thorac Surg 2010;89:1339–40. Backer CL. Discussion of reference 7. Overman DM. Reoperation for left ventricular outflow tract obstruction after repair of atrioventricular septal. Semin Thorac Cardiovasc Surg Pediatr Surg Annu 2014;17:43–7. Rastelli GC, Ongley PA, McGoon DC. Surgical repair of complete atrioventricular canal with anterior common leaflet undivided and unattached to ventricular septum. Mayo Clin Proc 1969;44; 355–41. Ebels T, Ho SY, Anderson RH, Meijboom EJ, Eijelaar A. The surgical anatomy of the left ventricular outflow tract in atrioventricular septal defect. Ann Thorac Surg 1986;41: 483–8. Tlaskal T, Gebauer R, Gilik J, Tomek V. Experience with the surgical treatment of atrioventricular septal defect with left ventricular outflow tract obstruction. Interact Cardiovasc Thorac Surg 2014;18:789–96.
DISCUSSION DR JORGE D. SALAZAR (Jackson, MS): I would like to congratulate Dr Backer and colleagues for another important contribution to the literature on this subject. I would also like to thank the Association for the opportunity to discuss this paper. I do appreciate having received a copy of the manuscript well in advance to review before your presentation.
As you point out in your presentation and manuscript, patients’ anatomic factors likely contribute to the risk of developing left ventricular outflow tract obstruction more than surgical approach. I would like to suggest, however, that geographical variation may exist in terms of patient anatomy and that patients most at risk for developing left ventricular outflow tract
CONGENITAL HEART
By far the largest series of patients undergoing the modified single-patch technique for AVSD is the series reported by Nunn [15]. He reported 128 patients undergoing the modified single-patch technique. No reoperations for LVOTO were performed in that series. Nunn had the following comment regarding the development of subaortic stenosis, “in actuality the modified single-patch technique may represent an important advance regarding the development of subaortic stenosis.” He had a higher incidence of subaortic stenosis in his series of patients undergoing the twopatch technique. Nunn postulated that the presence of synthetic material in the left ventricular outflow, where a two-patch technique is used, may further encourage fibrotic obstruction in the naturally narrow “goose-neck deformity” (left ventricular outlet). Although our follow-up now extends up to 13 years, we do not have complete follow-up of these patients into adulthood. However, the patients in our series who have required reoperation were all within the first 3 years after their initial procedure. A review by Tlaskal and colleagues [22] of 615 AVSD patients similarly noted a mean age at reoperation for LVOTO of 3.4 years. Our mean follow-up is 4.6 years. There is also a possibility that patients with subaortic stenosis may have had their reoperation at another institution and that we are unaware of this occurrence. In conclusion, our review of the outcomes of 77 infants undergoing repair of complete AVSD with the modified single-patch technique during a 13-year period shows that at intermediate follow-up, LVOTO does not appear to be a significant postoperative issue after repair with the modified single-patch technique. Only 2 patients in this series required reoperation for LVOTO. The most significant predictor of late LVOTO postrepair of complete AVSD in our patients was prior repair of CoAo. We continue to recommend the modified single-patch technique as the procedure of choice for nearly all infants with a balanced complete AVSD, the exception being infants with a very large ventricular component.
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
CONGENITAL HEART
1676
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
obstruction may still benefit from a 2-patch technique. In particular, patients with an already borderline left ventricular outflow tract or a very deep ventricular septal detect, as you allude to in your presentation, may still benefit from a 2-patch technique. Furthermore, patients with a very deep ventricular septal defect appear to be at more risk for mitral regurgitation when the modified single-patch technique is employed. In these instances, we still favor a 2-patch technique. I would appreciate hearing your perspective on the selective use of the 2-patch technique and your experience with mitral regurgitation in the setting of a deep ventricular septal defect. Thank you. DR BACKER: Thanks for that comment. In the discussion this morning we were trying to decide what is a “shallow” ventricular septal defect (VSD) and what is a “deep” VSD. What we have done is look at the 4-chamber view on the echocardiogram where you can see the right and left ventricles and right and left atrium. During diastole you can measure the distance between the crest of the ventricular septum and the closed atrioventricular (AV) valve leaflets. When we did our comparison analysis looking at 2 patch vs modified single patch, the average VSD size was not different between the two groups, 9 vs 10 mm. In my experience, when the VSD is deeper than 12 mm, I get nervous about using the modified single-patch technique. The 2 patients in the past 9 years in whom we elected to use a 2-patch technique had 12 and 16 mm VSDs. Those are the only 2-patch operations in the past 60 patients. I would point out that in Graham Nunn’s series with 128 patients, he used the modified single-patch technique in all patients, and in his series he has had no reoperations for left ventricular outflow tract obstruction. DR SALAZAR: Could you comment also on the incidence of mitral regurgitation in the setting of a deep scallop? It has been our experience that in doing so you pull the leaflet away from its coaptation zone and that the risk of regurgitation is greater in those cases.
Ann Thorac Surg 2015;99:1671–6
DR BACKER: I agree with you. The potential for left AV valve insufficiency has been my main concern about applying the modified single-patch technique to an extraordinarily large ventricular septal defect. DR ROSS M. UNGERLEIDER (Winston-Salem, NC): Carl, excellent. Your illustrations and your discussion make this look so simple. Give us a perspective that can help encourage change. Your data are very convincing, and over time you became more and more confident with the application of the single-patch technique. Left ventricular outflow tract obstruction is one potential complication (and you have demonstrated it is unlikely to occur), but there is mitral insufficiency, which is probably the most common late complication after AV canal repair, along with heart block, ventricular dysfunction postoperatively, and even the occasional patient with mild unbalance that needs adjustment of how we fashion the inter ventricular patch. Should most of us be thinking of doing what you have done and just apply this to all patients except perhaps those with coarctations? Give us some encouragement about helping us shift our practice. DR BACKER: My whole career I have worried about AV block after AVSD repair, and now I have 77 consecutive patients with no AV block! At some point I think I can stop worrying about heart block with this operation. In comparison we had an incidence of heart block of 3% to 4% when we used the 2-patch technique. Our incidence of reoperation for left AV valve insufficiency has also decreased since we adopted the modified singlepatch technique; now 4% vs 8 to 10% with the 2-patch technique. This operation is an improvement for avoiding those two potential complications, and in addition, we do not have a high incidence of left ventricular outflow tract obstruction that everyone was worried about. This is a better operation for infants with complete atrioventricular septal defect. I do agree it is difficult for surgeons to change a strategy that they have been using “successfully” for many years, but I am confident you will get better results with the modified singlepatch technique.
Background. When the modified single-patch technique for atrioventricular septal defect (AVSD) repair was introduced by Dr Benson Wilcox, there was concern that these patients might be at risk for late subaortic stenosis and left ventricular outflow tract obstruction (LVOTO). This review evaluated our modified singlepatch population for LVOTO in the postoperative period. Methods. Between January 2000 and 2013, 77 infants underwent AVSD repair with a modified single-patch technique. Median age was 4.2 months, and median weight was 5 kg. Eight patients had a prior repair of coarctation of the aorta via left thoracotomy in the newborn period. Results. The median hospital stay was 10 days. No patient required a pacemaker. The mean and median follow-up times were 4.6 and 3.7 years, respectively. Only 2 patients (2.5%) required reoperation for LVOTO; both
had prior repair of coarctation of the aorta (2 of 8 vs 0 of 69, p [ 0.01). A discrete fibrous subaortic membrane developed in the first patient that required resection at 3 and 7 years after repair. The other patient had LVOTO from accessory chordae of the left atrioventricular valve and required mitral valve replacement 5 months after repair. One early death occurred at 4 months postoperatively due to liver failure related to hyperalimentation. Conclusions. At intermediate term follow-up, LVOTO does not appear to be a significant postoperative issue after modified single-patch repair of AVSD. Coarctation of the aorta was the most significant predictor of late LVOTO after repair of AVSD with the modified single-patch technique.
T
obstruction . . .” [5]. However, there were no reoperations for left ventricular outflow tract obstruction (LVOTO) in the series by Nicholson and colleagues [2]. The potential for LVOTO using the modified singlepatch technique continues to be raised by surgeons using the “classic” single and 2-patch techniques and by some cardiac morphologists [6, 7]. Proponents of the 2patch technique contend that the patch placed to close the ventricular component of the AVSD helps to contribute to the width of the left ventricular outflow tract. They contend that pulling the valve leaflets down to the septum using the modified single-patch technique will encroach upon the left ventricular outflow tract and narrow it. Another issue raised relates to the known fact that LVOTO is more common in partial than complete AVSD [8]. The concern is that using the modified singlepatch technique essentially converts a patient with a complete AVSD to a partial AVSD [9]. This could potentially increase the risk of developing LVOTO. We previously published our results comparing the modified single-patch technique with the 2-patch technique [3]. At that time we reported 26 patients undergoing the modified single-patch technique. No patient in that small group required reoperation for LVOTO at short-term follow-up. We have now tripled the size of that initial series and have follow-up of more than 10 years. Our hypothesis was that the modified
he modified single-patch technique for repair of complete atrioventricular septal defect (AVSD) was independently proposed by Wilcox and colleagues [1] and Nicholson and colleagues [2]. The change in strategy from the classic 2-patch technique was to perform direct closure of the ventricular element, eliminating the patch used for closure of the ventricular component [3]. When Wilcox presented his concept at The Society of Thoracic Surgeons meeting in 1997, the second question during the discussion was, “Did you see any subsequent increased flow velocity in the left ventricular outflow tract indicative of obstruction?” [4]. The answer was: “we find no evidence of that [LVOTO] whatsoever. I think it has less to do with the particular repair that is used than with the basic anatomy. “ When the Nicholson et al [2] paper was presented at the American Association for Thoracic Surgery meeting in 1999, Dr John Brown noted, “it would seem that this [repair] would increase the likelihood of LVOTO
Accepted for publication Jan 6, 2015. Presented at the Sixty-first Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 5–8, 2014. Address correspondence to Dr Backer, Division of CardiovascularThoracic Surgery, 225 E Chicago Ave, MC 22, Chicago, IL 60611; e-mail: [email protected].
Ó 2015 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2015;99:1671–6) Ó 2015 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.01.032
CONGENITAL HEART
Modified Single Patch: Are We Still Worried About Subaortic Stenosis?
1672
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
CONGENITAL HEART
(simplified) single-patch technique would not be associated with an increased risk of LVOTO (subaortic stenosis).
Material and Methods The Ann & Robert H. Lurie Children’s Hospital of Chicago Institutional Review Board approved this study on June 12, 2012, as a retrospective record analysis and waived the need for patient consent. The congenital cardiac surgery database was analyzed, and all patients undergoing repair of complete AVSD between 2000 and 2013 were included. We excluded AVSD with tetralogy of Fallot, AVSD with double-outlet right ventricle, and unbalanced AVSD. We analyzed the serial postoperative echocardiograms of each patient to assess the degree of LVOTO. Statistical analysis was performed using the Fisher exact test. Seventy-seven patients underwent repair of complete AVSD using a modified single-patch technique at our institution from 2000 to 2013. We previously reported our specific surgical technique [3]. During that period, we transitioned from the 2-patch technique, as shown in Figure 1. Eight patients (10%) had undergone a previous operation for coarctation (CoAo) repair in the newborn period. All CoAo repairs were done via left thoracotomy using resection with extended end-to-end anastomosis.
Results The median age at operation was 4.2 months (mean, 4.8 2.4 months) and the median weight was 5 kg. In this series, 57 patients (74%) had trisomy 21. No patient had a procedure for LVOTO as part of the primary procedure. Rastelli classification was: A in 39, B in 3, C in 16, and unknown in 2. The VSD size was assessed by measuring the distance from the crest of the VSD to the common AV valve in diastole. We considered more than 12 mm to be an indication for a 2-patch repair. Only two patients since 2006 have had a 2-patch repair, both because of a very large ventricular component of 12 and 16 mm. Fig 1. Graph shows the evolution of the 2patch technique (black bars) to the modified single-patch technique (gray bars) at Ann & Robert H. Lurie Children’s Hospital of Chicago during a 13-year period (1997 to 2014). The 2-patch technique has been used in only 2 patients in the past 9 years.
Ann Thorac Surg 2015;99:1671–6
Mean cross-clamp time for the AVSD repair was 103 23 minutes. Mean cardiopulmonary bypass time was 135 33 minutes. Median length of stay was 10 days. No child required a pacemaker in the postoperative period. One early death occurred 4 months postoperatively as a result of liver failure in a child awaiting a liver transplant. Liver failure was secondary to long-term preoperative and postrepair hyperalimentation. At a mean follow-up of 4.2 years, 58 patients with at least 1 year of echocardiographic follow-up had no evidence of LVOTO by serial echocardiography (Fig 2). LVOTO developed in 4 patients, 3 of whom had prior CoAo repair: 3 of 8 (38%) vs 1 of 69 (1%; p ¼ 0.007 by Fisher exact test; Fig 3). Only 2 patients required reoperation for LVOTO, both had prior CoAo repair: 2 of 8 (25%) vs 0 of 69 (p ¼ 0.01 by Fisher exact test). The patient who died of liver failure 4 months postoperatively had mild LVOTO, which was unchanged on serial echocardiograms up until the time of death while awaiting liver transplant. A patient who underwent repair in 2001 had stable mild LVOTO for 11 years, with no change in gradient. That patient had repair of CoAo as an infant. A patient who underwent AVSD repair in 2003 at age 4 months had prior repair of CoAo as a newborn. This child started with mild LVOTO that progressed to moderate to severe obstruction at 3 years of age. At that time, reoperation was performed, and a discrete subaortic membrane was resected. The postoperative LVOTO was initially mild but during the next 5 years progressed to become severe again. A second reoperation was performed, and a recurrence of the fibrous subaortic membrane was resected. The child now has mild LVOTO, which has been stable. The final patient with LVOTO underwent CoAo repair at age 1 week and AVSD repair at age 3 months. After this repair, there was moderate LVOTO from substantial crowding of the LVOTO by accessory chords from the left atrioventricular valve. This progressed over a short period of time, and at 4 months postoperatively, the child underwent attempted resection of chordae but required
Ann Thorac Surg 2015;99:1671–6
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
1673
mitral valve replacement (#17 St. Jude; St. Jude Medical, St. Paul, MN). The child now is 2 years post-repair and has only mild LVOTO, which has been stable. All patients with LVOTO had Rastelli classification type A. In addition, all patients with CoAo and
complete AVSD had Rastelli type A. The age at repair of the patients in whom LVOTO developed was a median of 6 months. The age of repair of the children who did not develop LVOTO was 5 months (p ¼ not significant).
Fig 3. The time course is shown of the 4 patients in whom left ventricular outflow tract obstruction (LVOTO) developed after modified single-patch repair of complete atrioventricular septal defect (AVSD). (COA ¼ coarctation of the aorta; MVR ¼ mitral valve replacement.)
CONGENITAL HEART
Fig 2. This graph shows the number of years of serial echocardiographic follow-up on sequential complete atrioventricular septal defect patients who underwent a modified single-patch repair and had no evidence by echocardiography of left ventricular outflow tract obstruction (LVOTO). Only patients with at least 1 year of postoperative follow-up by echocardiogram were included.
1674
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
CONGENITAL HEART
Comment We have reviewed the outcomes of 77 infants undergoing repair of complete AVSD with a modified (simplified) single-patch technique during a 13-year period. At intermediate follow-up, LVOTO does not appear to be a significant postoperative issue after the modified singlepatch technique. Only 2 patients in this series required reoperation for LVOTO. The most significant predictor of late LVOTO after AVSD repair was the association of CoAo. These patients are known to have small left ventricular outflow tracts even when the intracardiac anatomy is otherwise normal [10]. To answer the question raised in the title of our report, we are no longer particularly worried about subaortic stenosis in patients undergoing the modified single-patch technique. We switched to the modified single-patch technique after many years of what we believed was successful application of the 2-patch technique [3]. Analysis of a comparison of the two techniques indicated that the modified single-patch technique not only had shorter cardiopulmonary bypass and cross-clamp times but resulted in improved outcomes with regard to left atrioventricular valve insufficiency [11]. That has been the Achilles heel of patients with complete AVSD undergoing any kind of repair [12]. Adachi and colleagues [6] from the Royal Brompton recently examined 43 heart specimens with AVSD. They particularly evaluated the left ventricular outflow tract diameter in relationship to the anterosuperior extension of the ventricular component of the AVSD. They concluded that the anterosuperior and skewed scoop could lead to asymmetric configuration of the valvar leaflets and LVOTO if the simplified single-patch technique was applied. They were particularly concerned with the Rastelli type C patients. In an invited commentary regarding this article, the primary limitation noted was a deficiency of clinical material presented to support the theoretical consideration [13]. That commentary combined the results from the Michigan group, [14] Ann & Robert H. Lurie Children’s Hospital [3], the updated series from Nunn [15], and a series published by Jonas and Mora [16]. Combining those series resulted in 200 patients undergoing the modified single-patch technique, with no patients at that point requiring reoperation for LVOTO. The conclusion of the commentary was, “although it may be counterintuitive, the synthetic patch placed in the two-patch technique may actually promote fibrosis that may lead to obstruction of the left ventricular outflow tract because of the increased rigidity to blood flow, and presence of a foreign body in the left ventricular outflow tract.” Of interest, in our series all of the patients developing LVOTO had a Rastelli classification A. Kanani and associates [17] also responded to the Adachi article with a letter to the editor, “Left ventricular outflow tract obstruction after the modified single patch repair of atrioventricular septal defects: teasing fact from fiction.” They said, “we believe this is an assertion taken too far and that it diverts the gaze away from other more pertinent morphologic variants that have proven their
Ann Thorac Surg 2015;99:1671–6
association in the clinical arena.” They cited other factors known to cause LVOTO, such as obstructive subvalvar apparatus, dysplastic leaflet material, and papillary muscles that extend into the left ventricular outflow tract. Myers and colleagues [7] from Children’s Hospital Boston reviewed 35 children with complete AVSD who underwent reoperation for LVOTO. Of these, 27 patients had prior double-patch repair, 6 had prior classic singlepatch repair, and 5 had a prior modified single-patch repair. One of their conclusions was that the modified single-patch repair for AVSD was associated with a higher incidence of reoperation for LVOTO (p ¼ 0.4). After analyzing their data, we believe what it actually shows is that LVOTO develops slightly sooner after the modified single-patch technique than if one of the other techniques is applied. However, because there is no denominator for any of the groups, reaching a conclusion about the relative incidence of LVOTO is hard. In the discussion of that paper, Myers noted that, “the multivariable analysis did not confirm the modified single-patch technique as an independent predictor of reoperation. It was only a predictor in the univariable analysis” [18]. In a review article on reoperations for LVOTO after repair of AVSD, Overman [19] emphasized the importance of the anatomy of the left ventricular outflow tract. The main factors predisposing to LVOTO in AVSD were unwedging of the aortic root (anterior and rightward), disparate inlet and outlet septum lengths (short inlet, long outlet), and deficiency of the muscular interventricular septum (septal “scoop”). Overman also noted the important contribution to LVOTO of the anterolateral muscle bundle of the left ventricle (AML) in AVSD patients. He noted that resection of this hypertrophic AML should be considered for operations for LVOTO after repair of AVSD. The patient in our series who required mitral valve replacement to treat LVOTO is exactly this type of patient. The final conclusion was: “Placement of a straight patch to close a large VSD may unfavorably alter the geometry and flow dynamics within the distal left ventricular outlet tract. It is conceivable that the stiff material used to reconstruct the distal left ventricular outflow tract may provide a substrate for LVOTO.” A technical point that is important in the surgical decision making for the modified single-patch technique is the location of the division of the superior bridging leaflet into the left and right atrioventricular valves. Our policy has been to “steal” from the right and give to the left, as originally recommended by Rastelli and colleagues [20]. The portion of the superior bridging leaflet that is adjacent to the ventricular septum after repair will become the de facto left ventricular outflow tract. Giving as much of this leaflet as possible to the left side will help prevent LVOTO. This relationship was noted by Ebels and coworkers [21] when they stated, “the extent of the attachment between the superior bridging leaflet and the septal crest is of crucial importance in determining the potential for obstruction in the left ventricular outflow tract in AVSD.” The only patients in whom we do not use the modified single-patch technique are those with a ventricular component exceeding 12 mm.
Ann Thorac Surg 2015;99:1671–6
4. 5. 6. 7.
8.
9. 10.
11. 12. 13.
14. 15. 16.
17.
18. 19. 20.
References 1. Wilcox BR, Jones DR, Frantz EG, Brink LW, Henry GW, Mill MR, Anderson RH. Anatomically sound, simplified approach to repair of “complete” atrioventricular septal defect. Ann Thorac Surg 1997;64:487–94; discussion 493–4. 2. Nicholson IA, Nunn GR, Sholler GF, et al. Simplified single patch technique for the repair of atrioventricular septal defect. J Thorac Cardiovasc Surg 1999;118:642–6. 3. Backer CL, Stewart RD, Bailliard F, Kelle AM, Webb CL, Mavroudis C. Complete atrioventricular canal: comparison
21.
22.
1675
of modified single-patch technique with two-patch technique. Ann Thorac Surg 2007;84:2038–46. Ebels T. Discussion of reference 1. Brown J. Discussion of reference 2. Adachi I, Ho SY, McCarthy KP, Uemura H. Ventricular scoop in atrioventricular septal defect: relevance to simplified single-patch method. Ann Thorac Surg 2009;87:198–203. Myers PO, del Nido PJ, Marx GR, et al. Improving left ventricular outflow tract obstruction repair in common atrioventricular canal defects. Ann Thorac Surg 2012;94: 599–605. Piccoli GP, Ho SY, Wilkinson JL, Macartney FJ, Gerlis LM, Anderson RH. Left-sided obstructive lesions in atrioventricular septal defects: an anatomic study. J Thorac Cardiovasc Surg 1982;83:453–60. Stulak JM, Burkhart HM, Dearani JA. Reoperations after repair of partial and complete atrioventricular septal defect. World J Ped Congenit Heart Surg 2010;1:97–104. Kalfa D, Ghez O, Kreitmann B, Metras D. Secondary subaortic stenosis in heart defects without any initial subaortic obstruction: a multifactorial postoperative event. Eur J Cardiothorac Surg 2007;32:582–7. Backer CL, Stewart RD, Mavroudis C. What is the best technique for repair of complete atrioventricular canal? Semin Thorac Cardiovasc Surg 2007;19:249–57. Crawford FA Jr, Stroud MR. Surgical repair of complete atrioventricular septal defect. Ann Thorac Surg 2001;72:1621–9. Backer CL. Invited Commentary for Adachi I, Ho SY, McCarthy KP, Uemura H. Ventricular scoop in atrioventricular septal defect: relevance to simplified single-patch method. Ann Thorac Surg 2009;87:198–203. Suzuki T, Bove EL, Devaney EJ, et al. Results of definitive repair of complete atrioventricular septal defect in neonates and infants. Ann Thorac Surg 2008;86:596–602. Nunn GR. Atrioventricular canal: modified single patch technique. Semin Thorac Cardiovasc Surg Ped Card Surg Annu 2007;10:28–31. Jonas RA, Mora B. Individualized approach to repair of complete atrioventricular canal: selective use of the traditional single-patch technique versus the Australian technique. World J Pediatr Congenit Heart Surg 2010;1:78–86. Kanani M, Cook A, Kostolny M. Left ventricular outflow tract obstruction after the modified single patch repair of atrioventricular septal defects: teasing fact from fiction. Ann Thorac Surg 2010;89:1339–40. Backer CL. Discussion of reference 7. Overman DM. Reoperation for left ventricular outflow tract obstruction after repair of atrioventricular septal. Semin Thorac Cardiovasc Surg Pediatr Surg Annu 2014;17:43–7. Rastelli GC, Ongley PA, McGoon DC. Surgical repair of complete atrioventricular canal with anterior common leaflet undivided and unattached to ventricular septum. Mayo Clin Proc 1969;44; 355–41. Ebels T, Ho SY, Anderson RH, Meijboom EJ, Eijelaar A. The surgical anatomy of the left ventricular outflow tract in atrioventricular septal defect. Ann Thorac Surg 1986;41: 483–8. Tlaskal T, Gebauer R, Gilik J, Tomek V. Experience with the surgical treatment of atrioventricular septal defect with left ventricular outflow tract obstruction. Interact Cardiovasc Thorac Surg 2014;18:789–96.
DISCUSSION DR JORGE D. SALAZAR (Jackson, MS): I would like to congratulate Dr Backer and colleagues for another important contribution to the literature on this subject. I would also like to thank the Association for the opportunity to discuss this paper. I do appreciate having received a copy of the manuscript well in advance to review before your presentation.
As you point out in your presentation and manuscript, patients’ anatomic factors likely contribute to the risk of developing left ventricular outflow tract obstruction more than surgical approach. I would like to suggest, however, that geographical variation may exist in terms of patient anatomy and that patients most at risk for developing left ventricular outflow tract
CONGENITAL HEART
By far the largest series of patients undergoing the modified single-patch technique for AVSD is the series reported by Nunn [15]. He reported 128 patients undergoing the modified single-patch technique. No reoperations for LVOTO were performed in that series. Nunn had the following comment regarding the development of subaortic stenosis, “in actuality the modified single-patch technique may represent an important advance regarding the development of subaortic stenosis.” He had a higher incidence of subaortic stenosis in his series of patients undergoing the twopatch technique. Nunn postulated that the presence of synthetic material in the left ventricular outflow, where a two-patch technique is used, may further encourage fibrotic obstruction in the naturally narrow “goose-neck deformity” (left ventricular outlet). Although our follow-up now extends up to 13 years, we do not have complete follow-up of these patients into adulthood. However, the patients in our series who have required reoperation were all within the first 3 years after their initial procedure. A review by Tlaskal and colleagues [22] of 615 AVSD patients similarly noted a mean age at reoperation for LVOTO of 3.4 years. Our mean follow-up is 4.6 years. There is also a possibility that patients with subaortic stenosis may have had their reoperation at another institution and that we are unaware of this occurrence. In conclusion, our review of the outcomes of 77 infants undergoing repair of complete AVSD with the modified single-patch technique during a 13-year period shows that at intermediate follow-up, LVOTO does not appear to be a significant postoperative issue after repair with the modified single-patch technique. Only 2 patients in this series required reoperation for LVOTO. The most significant predictor of late LVOTO postrepair of complete AVSD in our patients was prior repair of CoAo. We continue to recommend the modified single-patch technique as the procedure of choice for nearly all infants with a balanced complete AVSD, the exception being infants with a very large ventricular component.
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
CONGENITAL HEART
1676
BACKER ET AL LVOTO AFTER MODIFIED SINGLE-PATCH TECHNIQUE
obstruction may still benefit from a 2-patch technique. In particular, patients with an already borderline left ventricular outflow tract or a very deep ventricular septal detect, as you allude to in your presentation, may still benefit from a 2-patch technique. Furthermore, patients with a very deep ventricular septal defect appear to be at more risk for mitral regurgitation when the modified single-patch technique is employed. In these instances, we still favor a 2-patch technique. I would appreciate hearing your perspective on the selective use of the 2-patch technique and your experience with mitral regurgitation in the setting of a deep ventricular septal defect. Thank you. DR BACKER: Thanks for that comment. In the discussion this morning we were trying to decide what is a “shallow” ventricular septal defect (VSD) and what is a “deep” VSD. What we have done is look at the 4-chamber view on the echocardiogram where you can see the right and left ventricles and right and left atrium. During diastole you can measure the distance between the crest of the ventricular septum and the closed atrioventricular (AV) valve leaflets. When we did our comparison analysis looking at 2 patch vs modified single patch, the average VSD size was not different between the two groups, 9 vs 10 mm. In my experience, when the VSD is deeper than 12 mm, I get nervous about using the modified single-patch technique. The 2 patients in the past 9 years in whom we elected to use a 2-patch technique had 12 and 16 mm VSDs. Those are the only 2-patch operations in the past 60 patients. I would point out that in Graham Nunn’s series with 128 patients, he used the modified single-patch technique in all patients, and in his series he has had no reoperations for left ventricular outflow tract obstruction. DR SALAZAR: Could you comment also on the incidence of mitral regurgitation in the setting of a deep scallop? It has been our experience that in doing so you pull the leaflet away from its coaptation zone and that the risk of regurgitation is greater in those cases.
Ann Thorac Surg 2015;99:1671–6
DR BACKER: I agree with you. The potential for left AV valve insufficiency has been my main concern about applying the modified single-patch technique to an extraordinarily large ventricular septal defect. DR ROSS M. UNGERLEIDER (Winston-Salem, NC): Carl, excellent. Your illustrations and your discussion make this look so simple. Give us a perspective that can help encourage change. Your data are very convincing, and over time you became more and more confident with the application of the single-patch technique. Left ventricular outflow tract obstruction is one potential complication (and you have demonstrated it is unlikely to occur), but there is mitral insufficiency, which is probably the most common late complication after AV canal repair, along with heart block, ventricular dysfunction postoperatively, and even the occasional patient with mild unbalance that needs adjustment of how we fashion the inter ventricular patch. Should most of us be thinking of doing what you have done and just apply this to all patients except perhaps those with coarctations? Give us some encouragement about helping us shift our practice. DR BACKER: My whole career I have worried about AV block after AVSD repair, and now I have 77 consecutive patients with no AV block! At some point I think I can stop worrying about heart block with this operation. In comparison we had an incidence of heart block of 3% to 4% when we used the 2-patch technique. Our incidence of reoperation for left AV valve insufficiency has also decreased since we adopted the modified singlepatch technique; now 4% vs 8 to 10% with the 2-patch technique. This operation is an improvement for avoiding those two potential complications, and in addition, we do not have a high incidence of left ventricular outflow tract obstruction that everyone was worried about. This is a better operation for infants with complete atrioventricular septal defect. I do agree it is difficult for surgeons to change a strategy that they have been using “successfully” for many years, but I am confident you will get better results with the modified singlepatch technique.