- Email: [email protected]
Indications for Tricuspid Valve Detachment in Closure of Ventricular Septal Defect in Children
CARDIOVASCULAR
Indications for Tricuspid Valve Detachment in Closure of Ventricular Septal Defect in Children Lior Sasson, MD, Michael G. Katz, MD, Tiberiu Ezri, MD, Akiva Tamir, MD, Amir Herman, MS, Edward L. Bove, MD, and Arie Schachner, MD Angela & Sami Shamoon Cardiothoracic Department, Department of Anesthesia, Pediatric Cardiology Unit, The Edith Wolfson Medical Center, Holon, Israel, affiliated with the Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel; Save a Child’s Heart, Azur, Israel; Outcomes Research Institute, University of Louisville, Kentucky; Department of Statistics, University of Haifa, Haifa, Israel; Section of Cardiac Surgery, University of Michigan School of Medicine, Ann Arbor, Michigan
Background. Different techniques have been described for tricuspid valve detachment to improve visualization in ventricular septal defect repair. Our hypothesis was that preoperative echocardiographic criteria are important in deciding which patients should undergo ventricular septal defect repair by tricuspid valve detachment, and patients who undergo this procedure may have a better surgical outcome than those who fulfilled the criteria but were actually operated on with the standard surgical approach. Methods. Between January 2000 and December 2004 we prospectively studied 179 patients scheduled for ventricular septal defect repair and criteria for tricuspid valve detachment were established. Of these, 84 patients did not have any criteria for tricuspid valve detachment and were classified as the control group (group 1). Ninety-five patients with at least one criterion for tricuspid valve detachment were intraoperatively divided by patients
who underwent tricuspid valve detachment into group 2 (n ⴝ 41), and those who did not undergo tricuspid valve detachment into group 3 (n ⴝ 53). Results. Surgical complications occurred more frequently in group 3 (26%) as opposed to group 2 (10%) and group 1 (7%). Residual ventricular septal defect and atrioventricular block occurred only in group 3. Tricuspid regurgitation occurred in 15% of group 3 versus 9.8% of group 2 and 7.1% of group 1. Conclusions. Preoperative criteria for tricuspid valve detachment can be established before repair of ventricular septal defect. Patients who had indications for tricuspid valve detachment who actually had detachment performed during repair had fewer postoperative surgical complications as opposed to patients who fulfilled the criteria but did not undergo detachment. (Ann Thorac Surg 2006;82:958 – 63) © 2006 by The Society of Thoracic Surgeons
C
the tricuspid valve on postoperative outcomes after VSD repair in children.
losure of ventricular septal defect (VSD) through a right atriotomy is widely accepted and usually provides good exposure of the defect margins. However, in 10% to 30% of patients this exposure may be inadequate, which may lead to a less than optimal repair and result in residual postoperative shunt or tricuspid valve incompetence [1–3]. To improve visualization of the entire circumference of the VSD, some surgeons use temporary tricuspid valve detachment (TVD) based on a technique that was initially described by Hudspeth and colleagues [4] or one of its numerous modifications [1, 3, 5–7]. During the past few years there have been reports about the usefulness and advantages of these techniques [8 –10]. However, there have been no studies establishing preoperative or intraoperative systematic surgical indications, or both, for detachment of the tricuspid valve in VSD repair. In this study we prospectively evaluated the impact of using the systematic indication criteria for detachment of Accepted for publication March 29, 2006. Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30 –Feb 1, 2006. Address correspondence to Dr Schachner, Angela & Sami Shamoon Cardiothoracic Department, E. Wolfson Medical Center, PO Box 5, Holon, 58100 Israel; e-mail: [email protected].
© 2006 by The Society of Thoracic Surgeons Published by Elsevier Inc
Material and Methods This prospective, observational study was undertaken from January 2000 through December 2004. The Helsinki Committee of our institution retrospectively approved this study and waived the need for patient consent. Three predefined echocardiographic and intraoperative criteria were used to identify patients who would potentially benefit from TVD to facilitate VSD closure: (1) multiple tricuspid valve chordal arrangement obscuring the margins of the defect (Fig 1); (2) tricuspid valve aneurysm that precludes easy access to the defect (Fig 2); and (3) high position of the defect with outlet extension requiring excessive traction on the tricuspid valve leaflet for exposure (Fig 3). One hundred and seventy-nine patients were scheduled for VSD repair, 84 without any criterion for TVD and 95 patients in whom the chosen criteria for TVD were fulfilled. They were subdivided intraoperatively according to the surgeon’s assessment into those in whom detachment was actually performed (41 patients, group 2), and those who in spite of the fact that they fulfilled the 0003-4975/06/$32.00 doi:10.1016/j.athoracsur.2006.03.094
Fig 1. Subcostal long-axis view of multiple chordae from the septal leaflet of the tricuspid valve attached to the intraventricular septum and limiting exposure of the ventricular septal defect. Arrows show chordae attachment. (AO ⫽ aorta; LV ⫽ left ventricle; RV ⫽ right ventricle; SEPTAL ⫽ septal tricuspid valve leaflet; VSD ⫽ ventricular septal defect.)
detachment criteria had a standard, nondetachment procedure (53 patients, group 3). The diagnosis of VSD was established by preoperative transthoracic echocardiography. This was followed by transesophageal echocardiographic studies at the beginning of surgery and after weaning from bypass. At least two follow-up studies were done after surgery, the latest being 1 month after the operation, including echocardiography examination for residual VSD, as well as tricuspid, aortic, and mitral valve function. Left ventricular function was also assessed. Long-term follow-up of these patients is difficult as the children come from developing countries through the Save a Child’s Heart foundation. Follow-up and assessment is being carried out on annual or bi-annual missions to the relevant country. Communication with the local cardiologists reveals successful outcomes. Standard parasternal, apical, subcostal, and suprasternal views were obtained for each patient. Reviewing the echocardiographic Doppler examination was undertaken by a senior pediatric echocardiographer blinded to repair technique. Tricuspid regurgitation (TR) was graded as none, mild, moderate, or severe based on the area and the length of the color jet. Additional surgical procedures were performed on 35 patients. These procedures included closure of the atrial septal defect (8) or patent foramen ovale (6), ligation of the patent ductus arteriosus (11), resection of the subaortic membrane (6), and coarctation of aorta repair (4).
Surgical Technique Standard cardiopulmonary bypass and myocardial protection strategies were used. The right atrium was
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
959
opened parallel to the atrioventricular groove and the edges were held with stay sutures. Fine stay sutures were placed on the septal and anterior tricuspid valve leaflet with traction toward the surgeon for exposure of the VSD. At this point, the echocardiographic and intraoperative findings were compared, and a decision was made by the attending surgeon in all cases, based upon fulfilling at least one of the previously mentioned criteria, whether to proceed with the standard surgery (regular closure of the VSD) or whether to detach the tricuspid valve (by detaching the leaflets or detaching the chordae). Detachment of the septal leaflet commenced at the posteroseptal commissure toward the anteroseptal commissure, approximately 2 to 3 mm from the tricuspid valve annulus. If an extension was needed, the detachment was continued toward the anterior or posterior leaflet. We used marking sutures for the tricuspid valve at the beginning and at the end of the detachment to facilitate accurate positioning of the leaflet to avoid postoperative repair of TR. To close the VSD, a Gore-Tex patch (W.L. Gore & Assoc, Flagstaff, AZ) was sutured using continuous 5-0 Prolene. When deemed necessary, the suture line was reinforced with pledget sutures. The leaflets were reattached to the annulus using 7-0 Prolene continuous suture. After VSD closure, cold saline was injected into the right ventricle to assess the tricuspid
Fig 2. Subcostal long-axis view showing both anterior and septal tricuspid leaflets involved in an aneurysm that partially obstructs the ventricular septal defect. (ANT ⫽ anterior tricuspid valve leaflet; LV ⫽ left ventricle; RV ⫽ right ventricle; SEP ⫽ septal tricuspid valve leaflet; VSD ⫽ ventricular septal defect.)
CARDIOVASCULAR
Ann Thorac Surg 2006;82:958 – 63
CARDIOVASCULAR
960
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
Ann Thorac Surg 2006;82:958 – 63
Modified ultrafiltration was used after termination of cardiopulmonary bypass.
Measurements Demographic data including the number of criteria for tricuspid detachment and postoperative complications were recorded and are listed in Tables 1 and 2.
Statistical Analysis All continuous variables (ie, age, aortic clamp time, bypass time) are presented as mean ⫾ standard deviation, or number of detachment criteria, as median (range). Categorical data (ie, gender, associated heart anomalies, TR pre-surgery) are presented as number (percentage). Comparisons between the 3 patient groups for continuous variables were performed using the Kruskal-Wallis test. Comparisons between patient groups for categorical variables were performed using the 2 test. All p values given are two-sided. Results were considered significant with p ⬍ 0.05. The odds ratios with 95% confidence interval were calculated for assessment of the influence of patient group on complications. A logistic regression was done to evaluate multivariable influence on the probability to have a complication. The stepwise method (include ⫽ 0.05, exclude ⫽ 0.1) was performed for selecting only the influential covariates in the logistic regression. The SPSS 10 software (SPSS, Chicago, IL) was used for data analysis.
Results Fig 3. Parasternal short-axis view showing the aorta and pulmonary artery. The ventricular septal defect is extended to the subpulmonary area. (PV ⫽ pulmonary valve; TV ⫽ tricuspid valve; VSD ⫽ ventricular septal defect.)
valve leaflet coaptation and competence. In cases in which the TR was found, additional sutures were applied to approximate the septal and anterior leaflet close to the commissure. During closure of the right atrium, warm blood was infused into the aortic root, the heart was emptied of air, and the aortic clamp was removed.
Demographic data of the 3 groups are presented in Table 1. Complications of surgery according to treatment groups are presented in Table 2. Demographics were similar among the 3 groups, except for the younger patient age in group 3 (see Table 1). Twenty-one patients from group 2 and 22 from group 3 had more than one criterion. We performed a logistic regression to test the influence of multiple covariates on the risk of having a postoperative complication. The variables that were entered to the regression were age, associated heart anomalies, gender, aortic clamp time, on-pump time, and treatment group
Table 1. Demographics of the Three Groups Characteristic Age (yrs) Aortic clamp time (mins) On pump time (mins) Gender Male Female Number of criteria for tricuspid detachment Associated heart anomalies Tricuspid regurgitation pre-surgery a
Group 1 (n ⫽ 84)
Group 2 (n ⫽ 41)
Group 3 (n ⫽ 54)
p
4.3 (⫾3) 56 (⫾12.9) 80.25 (⫾18.26)
3.36 (⫾3.2) 51.37 (⫾12.1) 75.9 (⫾14.7)
2.11 (⫾2.49) 56.9 (⫾14.16) 81.2 (⫾19.8)
⬍ 0.001 0.099 0.466 0.149
50 (59.5%) 34 (40.5%) 13 (15.5%) 1 (1.2%)
21 (51.2%) 20 (48.8%) 1–3 (2) 13 (31.7%) 2 (4.9%)
Results are expressed as mean ⫾ standard deviation, numbers (percentage), and median (range).
23 (42.6%) 31 (57.4%) 1–3 (2) 9 (16.7%) 1 (1.9%)
0.500 0.081 0.413
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
961
Table 2. Complications of Surgery Surgical Complication
Group 1 (n ⫽ 84)
Group 2 (n ⫽ 41)
Group 3 (n ⫽ 54)
p
0 0 6 (6%) 6 (7.1%)
0 0 4 (9.8%) 4 (9.8%)
2 (3.7%) 4 (7.4%) 8 (14.8%) 14 (25.9%)
0.096 0.009 0.678 0.005
Atrial-ventricular block Residual ventricular septal defect Tricuspid regurgitation post-surgery Total surgical complications a
Results are expressed as absolute numbers (percentage).
(i.e., group 1, 2, or 3). In a stepwise elimination of nonsignificant covariates, the only covariate that was found to be influential was the treatment group. The odds ratio between group 1 and group 2 was 1.4 (95% confidence interval [CI], 0.374 –5.283), whereas the odds ratio between group 1 and group 3 was 4.54 (95% CI, 1.625–12.736). This means that there was a fourfold increase in the risk of complications between the first group and the group of patients that had indications for detachment, but the detachment was not performed (p ⫽ 0.005). All patients were discharged home in stable condition. Two patients in group 3 had postoperative conduction system disturbances develop (complete atrioventricular block and atrioventricular dissociation); one of them needed a permanent DDD-pacemaker implantation. The duration of postoperative hospitalization was 7.43 ⫾ 2.64 days and was not significantly different between the three groups. There were no significant differences between the groups concerning cardiopulmonary bypass time and cross-clamp time (Table 1). Additional sutures for TR repair were added in 17 patients according to in situ findings (4 in group 1, 4 in group 2, and 9 in group 3). Postoperative echocardiographic assessment before discharge revealed 18 patients with new TR (6 from group 1 and 4 from group 2, all with mild TR), and in group 3 there were 6 patients with mild TR and 2 with mild to moderate TR. No patient underwent further surgery to correct the TR in either of the groups. Residual VSD was detected in 4 children. All were in group 3, whereas, despite the presence of a preoperative indication for TVD, it was not carried out. Two patients had small residual VSDs detected by intraoperative transesophageal echocardiography, which were confirmed by an intraoperative QP/QS measurement of less than 1.5:1. They had an uneventful postoperative course and needed no further surgery. The third patient had immediate VSD closure after transesophageal echocardiography revealed a large residual VSD with a QP/QS measurement of 2.3:1 at the end of surgery. The fourth child had a small residual VSD (QP/QS 1.2:1) that suddenly enlarged (QP/QS 2.5:1) on the second postoperative day. This child was successfully reoperated on the same day using the TVD technique. The VSD in the first 2 patients was closed using TVD because of tricuspid valve aneurysm and in the last 2 patients because of the tricuspid valve chordal arrangement.
Comment As outlined in the Results section, the main findings of our study are that a fourfold increase in the risk of complications exists between group 2 (patients who had indications for TVD and detachment was performed) and group 3 (patients who had indications for TVD but detachment was not performed). p value was found to be ⫽ 0.005. Successful transatrial repair of VSD requires adequate exposure of the margins to avoid residual shunt, injury of conduction tissue, and distortion of the leaflets of the tricuspid valve [1, 2, 10]. The use of TVD in our patients did not result in an increased incidence of residual VSD or surgically induced heart block. In group 2, in which tricuspid detachment was indicated according to our criteria and was actually performed, there was no need for reoperation because of residual VSD or pacemaker implantation. This confirms the conclusions of previous studies stating that with multiple chordal attachments crossing over the defect and inserting at the edge of the VSD, it is difficult to examine and reach all the margins of the defect to improve visualization. Leaflet detachment results in more accurate suture placement [2, 11]. The anomalies of the tricuspid valve, which may be acquired secondary to left-to-right shunting, are frequently associated with peri-membranous defects. These anomalies include excess septal or anterior leaflet tissue that can partially or completely occlude the defect and are associated with spontaneous closure of the VSD [12]. This situation alone, with or without chordal attachments crossing the VSD, may make the exposure of the VSD margins difficult, and the surgeon may need to exert traction on the tricuspid valve leaflets that may cause distortion of the tricuspid subvalvular apparatus and TR. As a result, temporary detachment of the tricuspid valve from the annulus is a good option. We carefully performed intraoperative and postoperative assessments of TR, and in 17 cases additional sutures on tricuspid valve leaflets were needed to minimize TR (ie, most of the cases in group 3). When the VSD extends into the outlet components of the interventricular septum, it necessitates severe traction of the tricuspid valve to improve exposure. Detachment of the septal leaflet and a contiguous portion of the anterior leaflet opens the peri-membranous part of the VSD and gives excellent exposure of all the margins [6]. Although this is an observational study, and not a
CARDIOVASCULAR
Ann Thorac Surg 2006;82:958 – 63
CARDIOVASCULAR
962
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
double-blind study, it has been shown that observational studies may yield similar results to clinical trials [13, 14]. We describe three practical criteria for the indication of TVD during VSD repair. When at least one of these criteria was met, performance of VSD repair with tricuspid detachment was associated with significantly lower incidence of postoperative surgery-related complications as compared with performing surgery without tricuspid detachment, despite the presence of at least one criterion for valve detachment. This study demonstrates that in the presence of echocardiographic and surgical indication criteria, the surgeon has to take into consideration that he can perform TVD with good results. There was no correlation between the degree of postoperative TR and the extent of TVD. If detachment was needed, it usually included the entire septal leaflet and part of the anterior leaflet. Group 3 patients had echocardiographic indications for TVD, but did not undergo the technique because intraoperative assessment revealed that VSD exposure was acceptable. In fact, the exposure of the VSD margins was better than in group 2 patients, with fewer obstructive chordae and relatively small aneurysm tissue. Yet these easier cases yielded the worst results. This supports our findings that TVD is safe and effective, and should be used whenever intraoperative exposure is compromised. Among the 3 preoperative indications for TVD, multiple chordal arrangements (indication number 1) has the strongest predictive value.[15, 16] The authors wish to thank Dawn Mizrahi for her editorial assistance and the Save A Child’s Heart Foundation.
References 1. Gaynor JW, Obrien JE Jr, Rychik J, Sanchez GR, DeCampli WM, Spray TL. Outcome following tricuspid valve detachment for ventricular septal defects closure. Eur J Cardiothorac Surg 2001;19:279 – 82. 2. Aeba R, Katogi T, Hashizume K, et al. Liberal use of tricuspid valve detachment for transatrial ventricular septal defect closure. Ann Thorac Surg 2003;76:1073–7.
Ann Thorac Surg 2006;82:958 – 63
3. Bol-Raap G, Weerheim J, Kappetein AP, Witsenburg M, Bogers AJJC. Follow-up after surgical closure of congenital ventricular septal defect. Eur J Cardiothorac Surg 2003;24: 511–5. 4. Hudspeth AS, Cordell AR, Meredith JH, Johnston FR. An improved transatrial approach to the closure of ventricular septal defects. J Thorac Cardiovasc Surg 1962;43:157– 65. 5. Maile S, Kadner A, Pretre R. Detachment of the anterior leaflet of the tricuspid valve to expose perimembranous ventricular septal defects. Ann Thorac Surg 2003;75:944 – 6. 6. Pridjian AK, Pearce FB, Culpepper WS, Williams LC, Van Meter CH, Ochsner JL. Atrioventricular valve competence after takedown to improve exposure during ventricular septal defect repair. J Thorac Cardiovasc Surg 1993;106: 1122–5. 7. Kapoor L, Gan MD, Bandyhopadhyay A, Das MB, Chatterjee S. Improved exposure of isolated perimembranous ventricular septal defects. Ann Thorac Surg 2000;69:291–2. 8. Tatebe S, Miyamira H, Watanabe H, Sugawara M, Educhi S. Closure of isolated ventricular septal defect with detachment of the tricuspid valve. J Card Surg 1995;10:564 – 8. 9. Koshy S, Sunil GS, Anil SR, Dhinakar S, Shivaprakasha K, Rao SG. Tricuspid valve detachment for transatrial closure of ventricular septal defects. Asian Cardiovasc Thorac Ann 2002;10:314 –7. 10. Zhao J, Li J, Wei X, Zhao B, Sun W. Tricuspid valve detachment in closure of congenital ventricular septal defect. Tex Heart Inst J 2003;30:38 – 41. 11. Frenckner BP, Olin CL, Bomfim V, Bjarke B, Wallgren CG, Bjõrk VO. Detachment of the septal tricuspid leaflet during transatrial closure of isolated ventricular defect. J Thorac Cardiovasc Surg 1981;82:773– 8. 12. McDaniel NL, Gutgesell HP. Ventricular septal defects. In: Allen HD, Gutgesell HP, Clark EB, Driscoll DJ, eds. Moss and Adams’ heart disease in infants, children, and adolescents, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2001: 636 –51. 13. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med 2000;342: 1878 – 86. 14. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 2000;342:1887–92. 15. Kouchoukos NT, Blackstone EH, Doty DB, Hanley FL, Karp RB. Ventricular septal defect. In: Kirklin and Barratt-Boyes, Eds. Cardiac surgery: morphology, diagnostic criteria, natural history, technique, results, and indications, 3rd edition. Philadelphia: Churchill Livingstone, 2003:850 –909. 16. de Leval M. Ventricular septal defects. In: Stark J, de Leval M, eds. Surgery for congenital heart defects, 2nd ed. Philadelphia: WB Saunders Co, 1994: 355–71.
DISCUSSION DR CARL L. BACKER (Chicago, IL): You used mostly a circumferential detachment, or was that used in all patients versus a radial opening in the tricuspid valve? DR SASSON: I used a parallel incision parallel, to the annulus. DR BACKER: Because in our practice we very frequently open the septal leaflet with a radial-type incision which unfolds it. I personally find it a little easier to put back together, and I guess it would be interesting to compare the results, although I am impressed with your very low incidence of heart block or tricuspid valve insufficiency after this type of opening of the tricuspid valve.
DR SASSON: At the end of the repair, I usually check the tricuspid valve with saline. And whenever there is regurgitation, I usually take the time and repair it. Sometimes I use stitches to approximate the septal and the anterior leaflet by the commissure. And at times when it seems that one of the papillary muscles of the septal leaflet is being elevated because of the ventricular septal defect (VSD) repair, a shortening of the chorda brings the competence of the valve to be nice. DR CHRISTOPHER KNOTT CRAIG (Oklahoma City, OK): The echocardiography (echo) on which you based your decision whether to detach the tricuspid valve or not, was that done preoperatively or intraoperatively?
DR SASSON: The echo was done preoperatively on two occasions: one, transthoracic echocardiography and the second was preoperatively by the transesophageal echocardiography (TEE). DR KNOTT CRAIG: But is the TEE done when the patient is asleep in the operating room? DR SASSON: Yes. DR KNOTT CRAIG: How often did you find that the preoperative transthoracic echo and the intraoperative TEE, differed significantly with regard to the perceived need to detach the valve leaflet? DR SASSON: I was really impressed by the results, that they were pretty much the same. We have, it seems, good cardiologists and they gave a very good diagnosis preoperatively for the transthoracic. It’s true that the intraoperative TEE could show better if there is more chordae or a septal aneurysm that can make the repair difficult. DR KNOTT CRAIG: Since you had intraoperative TEE, why were the large residual VSDs not addressed at that time? DR SASSON:They were addressed at the time. One patient addressed immediately after coming off bypass, and we saw that there is a VSD so we came in and closed it. And in another patient, the patch dehisced 2 days after the operation and we had to come back and close it again. DR MRINALENDU DAS (Kolkata, India): Regarding your group 3, I understand that the incidence of residual VSD and tricuspid regurgitation is high because of the presence of chordae, and you have not detached the tricuspid valve. But could you explain why the incidence of heart block is so high, 37.7%, which seems to equate pretty high. DR SASSON: The incidence of heart block in the last group is 3.7% compared to none in the other groups. But if you take all 179 patients that were operated on, only one patient had to have a pacemaker implantation. That makes it a very low incidence, even less than 0.5%. This group is the point of this paper. In cases where indications for detachment were met, but detachment was not performed, the struggle for good exposure probably caused the heart block. DR BOHDAN MARUSZEWSKI (Warsaw, Poland): I just want to make sure about the age of your patients, because you don’t mention it in the abstract, but on the presentation I saw it’s 4 years. So the first question is the age of these patients. And it’s not the usual age where we close the VSDs. So do you think this technique could be also easily applicable to the infants of 4 kg when we close the VSD? And the second question is, are you aware of any information about long-term follow-up after detachment of the tricuspid valve leaflet in infants undergoing surgery for VSD?
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
963
DR SASSON: In regard to the first question, our practice in closing the VSD is usually below 2 years of age. And if there is a necessity, we do it even below 1 year of age, according to failure to thrive or congestive heart failure. In our center, we receive most of our most of our patients through the Save A Child’s Heart Foundation, which is a charity organization that brings children from Third World countries to Israel for treatment. Since 1995 we have operated on 1,400 children. So we do not get to choose our patients. And sometimes we get patients with VSDs who may be even 14 years of age. As to if this detachment can be done in younger children, absolutely true. There were a few patients who weighed about 4 kg, where the VSD was difficult to expose, and I had to detach the valve. In those kids, I think that it might be even more justified, because the tricuspid leaflet is so thin, it’s like a butterfly’s wings, and excessive traction of these may cause significant TR postoperatively. So when I detached the valves in those patients and reattached it, I found no residual VSDs and no significant TR. There are a few articles that actually were published in 2003 and 2004 that examined patients for the long term at about 8 years follow-up. And it seems that there weren’t any problems regarding tricuspid regurgitation. There was concern whether the detachment may affect the growth of the leaflet and may cause later tricuspid regurgitation. This was not found. DR DOUGLAS D. PAYNE (Boston, MA): After you’ve made your incision in the tricuspid valve, do you do the whole VSD patch replacement through that hole, or is it partly through that hole and partly through the orifice? The second question is, when you’re working through this hole, I envision it’s rather small. Do you have any trouble with tearing of the valve, particularly the extension of the incision out into the free edge of the leaflet? DR SASSON: Usually I do it through the hole. When I make the incision, I get a good exposure of the margins of the VSD. You need to take care not to take bites too much towards the left ventricle, otherwise you may hit the conduction system. But usually you get a nice exposure of the VSD. There were times that I had to close the VSD or assist a few stitches, not through that opening. And the second question? DR PAYNE: Were there any injuries of the tricuspid valve from traction on that incision in the septal leaflet? DR SASSON: No, no. This is the reason that I just make the incision a little more than 2 mm from the annulus, mark the area of the beginning and the end of it, and exercise very gentle traction on the tricuspid valve, especially in small children. DR BACKER: Why don’t we take a quick poll. How many people here have no problem whatsoever detaching the tricuspid valve during VSD closure? Show of hands. Nearly everyone has raised their hand. Does anyone strongly object to this approach or think this is a bad technique? No one is raising their hand. I think you’ve convinced us. We were convinced before we came and we continue to be convinced.
CARDIOVASCULAR
Ann Thorac Surg 2006;82:958 – 63
Indications for Tricuspid Valve Detachment in Closure of Ventricular Septal Defect in Children Lior Sasson, MD, Michael G. Katz, MD, Tiberiu Ezri, MD, Akiva Tamir, MD, Amir Herman, MS, Edward L. Bove, MD, and Arie Schachner, MD Angela & Sami Shamoon Cardiothoracic Department, Department of Anesthesia, Pediatric Cardiology Unit, The Edith Wolfson Medical Center, Holon, Israel, affiliated with the Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel; Save a Child’s Heart, Azur, Israel; Outcomes Research Institute, University of Louisville, Kentucky; Department of Statistics, University of Haifa, Haifa, Israel; Section of Cardiac Surgery, University of Michigan School of Medicine, Ann Arbor, Michigan
Background. Different techniques have been described for tricuspid valve detachment to improve visualization in ventricular septal defect repair. Our hypothesis was that preoperative echocardiographic criteria are important in deciding which patients should undergo ventricular septal defect repair by tricuspid valve detachment, and patients who undergo this procedure may have a better surgical outcome than those who fulfilled the criteria but were actually operated on with the standard surgical approach. Methods. Between January 2000 and December 2004 we prospectively studied 179 patients scheduled for ventricular septal defect repair and criteria for tricuspid valve detachment were established. Of these, 84 patients did not have any criteria for tricuspid valve detachment and were classified as the control group (group 1). Ninety-five patients with at least one criterion for tricuspid valve detachment were intraoperatively divided by patients
who underwent tricuspid valve detachment into group 2 (n ⴝ 41), and those who did not undergo tricuspid valve detachment into group 3 (n ⴝ 53). Results. Surgical complications occurred more frequently in group 3 (26%) as opposed to group 2 (10%) and group 1 (7%). Residual ventricular septal defect and atrioventricular block occurred only in group 3. Tricuspid regurgitation occurred in 15% of group 3 versus 9.8% of group 2 and 7.1% of group 1. Conclusions. Preoperative criteria for tricuspid valve detachment can be established before repair of ventricular septal defect. Patients who had indications for tricuspid valve detachment who actually had detachment performed during repair had fewer postoperative surgical complications as opposed to patients who fulfilled the criteria but did not undergo detachment. (Ann Thorac Surg 2006;82:958 – 63) © 2006 by The Society of Thoracic Surgeons
C
the tricuspid valve on postoperative outcomes after VSD repair in children.
losure of ventricular septal defect (VSD) through a right atriotomy is widely accepted and usually provides good exposure of the defect margins. However, in 10% to 30% of patients this exposure may be inadequate, which may lead to a less than optimal repair and result in residual postoperative shunt or tricuspid valve incompetence [1–3]. To improve visualization of the entire circumference of the VSD, some surgeons use temporary tricuspid valve detachment (TVD) based on a technique that was initially described by Hudspeth and colleagues [4] or one of its numerous modifications [1, 3, 5–7]. During the past few years there have been reports about the usefulness and advantages of these techniques [8 –10]. However, there have been no studies establishing preoperative or intraoperative systematic surgical indications, or both, for detachment of the tricuspid valve in VSD repair. In this study we prospectively evaluated the impact of using the systematic indication criteria for detachment of Accepted for publication March 29, 2006. Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30 –Feb 1, 2006. Address correspondence to Dr Schachner, Angela & Sami Shamoon Cardiothoracic Department, E. Wolfson Medical Center, PO Box 5, Holon, 58100 Israel; e-mail: [email protected].
© 2006 by The Society of Thoracic Surgeons Published by Elsevier Inc
Material and Methods This prospective, observational study was undertaken from January 2000 through December 2004. The Helsinki Committee of our institution retrospectively approved this study and waived the need for patient consent. Three predefined echocardiographic and intraoperative criteria were used to identify patients who would potentially benefit from TVD to facilitate VSD closure: (1) multiple tricuspid valve chordal arrangement obscuring the margins of the defect (Fig 1); (2) tricuspid valve aneurysm that precludes easy access to the defect (Fig 2); and (3) high position of the defect with outlet extension requiring excessive traction on the tricuspid valve leaflet for exposure (Fig 3). One hundred and seventy-nine patients were scheduled for VSD repair, 84 without any criterion for TVD and 95 patients in whom the chosen criteria for TVD were fulfilled. They were subdivided intraoperatively according to the surgeon’s assessment into those in whom detachment was actually performed (41 patients, group 2), and those who in spite of the fact that they fulfilled the 0003-4975/06/$32.00 doi:10.1016/j.athoracsur.2006.03.094
Fig 1. Subcostal long-axis view of multiple chordae from the septal leaflet of the tricuspid valve attached to the intraventricular septum and limiting exposure of the ventricular septal defect. Arrows show chordae attachment. (AO ⫽ aorta; LV ⫽ left ventricle; RV ⫽ right ventricle; SEPTAL ⫽ septal tricuspid valve leaflet; VSD ⫽ ventricular septal defect.)
detachment criteria had a standard, nondetachment procedure (53 patients, group 3). The diagnosis of VSD was established by preoperative transthoracic echocardiography. This was followed by transesophageal echocardiographic studies at the beginning of surgery and after weaning from bypass. At least two follow-up studies were done after surgery, the latest being 1 month after the operation, including echocardiography examination for residual VSD, as well as tricuspid, aortic, and mitral valve function. Left ventricular function was also assessed. Long-term follow-up of these patients is difficult as the children come from developing countries through the Save a Child’s Heart foundation. Follow-up and assessment is being carried out on annual or bi-annual missions to the relevant country. Communication with the local cardiologists reveals successful outcomes. Standard parasternal, apical, subcostal, and suprasternal views were obtained for each patient. Reviewing the echocardiographic Doppler examination was undertaken by a senior pediatric echocardiographer blinded to repair technique. Tricuspid regurgitation (TR) was graded as none, mild, moderate, or severe based on the area and the length of the color jet. Additional surgical procedures were performed on 35 patients. These procedures included closure of the atrial septal defect (8) or patent foramen ovale (6), ligation of the patent ductus arteriosus (11), resection of the subaortic membrane (6), and coarctation of aorta repair (4).
Surgical Technique Standard cardiopulmonary bypass and myocardial protection strategies were used. The right atrium was
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
959
opened parallel to the atrioventricular groove and the edges were held with stay sutures. Fine stay sutures were placed on the septal and anterior tricuspid valve leaflet with traction toward the surgeon for exposure of the VSD. At this point, the echocardiographic and intraoperative findings were compared, and a decision was made by the attending surgeon in all cases, based upon fulfilling at least one of the previously mentioned criteria, whether to proceed with the standard surgery (regular closure of the VSD) or whether to detach the tricuspid valve (by detaching the leaflets or detaching the chordae). Detachment of the septal leaflet commenced at the posteroseptal commissure toward the anteroseptal commissure, approximately 2 to 3 mm from the tricuspid valve annulus. If an extension was needed, the detachment was continued toward the anterior or posterior leaflet. We used marking sutures for the tricuspid valve at the beginning and at the end of the detachment to facilitate accurate positioning of the leaflet to avoid postoperative repair of TR. To close the VSD, a Gore-Tex patch (W.L. Gore & Assoc, Flagstaff, AZ) was sutured using continuous 5-0 Prolene. When deemed necessary, the suture line was reinforced with pledget sutures. The leaflets were reattached to the annulus using 7-0 Prolene continuous suture. After VSD closure, cold saline was injected into the right ventricle to assess the tricuspid
Fig 2. Subcostal long-axis view showing both anterior and septal tricuspid leaflets involved in an aneurysm that partially obstructs the ventricular septal defect. (ANT ⫽ anterior tricuspid valve leaflet; LV ⫽ left ventricle; RV ⫽ right ventricle; SEP ⫽ septal tricuspid valve leaflet; VSD ⫽ ventricular septal defect.)
CARDIOVASCULAR
Ann Thorac Surg 2006;82:958 – 63
CARDIOVASCULAR
960
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
Ann Thorac Surg 2006;82:958 – 63
Modified ultrafiltration was used after termination of cardiopulmonary bypass.
Measurements Demographic data including the number of criteria for tricuspid detachment and postoperative complications were recorded and are listed in Tables 1 and 2.
Statistical Analysis All continuous variables (ie, age, aortic clamp time, bypass time) are presented as mean ⫾ standard deviation, or number of detachment criteria, as median (range). Categorical data (ie, gender, associated heart anomalies, TR pre-surgery) are presented as number (percentage). Comparisons between the 3 patient groups for continuous variables were performed using the Kruskal-Wallis test. Comparisons between patient groups for categorical variables were performed using the 2 test. All p values given are two-sided. Results were considered significant with p ⬍ 0.05. The odds ratios with 95% confidence interval were calculated for assessment of the influence of patient group on complications. A logistic regression was done to evaluate multivariable influence on the probability to have a complication. The stepwise method (include ⫽ 0.05, exclude ⫽ 0.1) was performed for selecting only the influential covariates in the logistic regression. The SPSS 10 software (SPSS, Chicago, IL) was used for data analysis.
Results Fig 3. Parasternal short-axis view showing the aorta and pulmonary artery. The ventricular septal defect is extended to the subpulmonary area. (PV ⫽ pulmonary valve; TV ⫽ tricuspid valve; VSD ⫽ ventricular septal defect.)
valve leaflet coaptation and competence. In cases in which the TR was found, additional sutures were applied to approximate the septal and anterior leaflet close to the commissure. During closure of the right atrium, warm blood was infused into the aortic root, the heart was emptied of air, and the aortic clamp was removed.
Demographic data of the 3 groups are presented in Table 1. Complications of surgery according to treatment groups are presented in Table 2. Demographics were similar among the 3 groups, except for the younger patient age in group 3 (see Table 1). Twenty-one patients from group 2 and 22 from group 3 had more than one criterion. We performed a logistic regression to test the influence of multiple covariates on the risk of having a postoperative complication. The variables that were entered to the regression were age, associated heart anomalies, gender, aortic clamp time, on-pump time, and treatment group
Table 1. Demographics of the Three Groups Characteristic Age (yrs) Aortic clamp time (mins) On pump time (mins) Gender Male Female Number of criteria for tricuspid detachment Associated heart anomalies Tricuspid regurgitation pre-surgery a
Group 1 (n ⫽ 84)
Group 2 (n ⫽ 41)
Group 3 (n ⫽ 54)
p
4.3 (⫾3) 56 (⫾12.9) 80.25 (⫾18.26)
3.36 (⫾3.2) 51.37 (⫾12.1) 75.9 (⫾14.7)
2.11 (⫾2.49) 56.9 (⫾14.16) 81.2 (⫾19.8)
⬍ 0.001 0.099 0.466 0.149
50 (59.5%) 34 (40.5%) 13 (15.5%) 1 (1.2%)
21 (51.2%) 20 (48.8%) 1–3 (2) 13 (31.7%) 2 (4.9%)
Results are expressed as mean ⫾ standard deviation, numbers (percentage), and median (range).
23 (42.6%) 31 (57.4%) 1–3 (2) 9 (16.7%) 1 (1.9%)
0.500 0.081 0.413
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
961
Table 2. Complications of Surgery Surgical Complication
Group 1 (n ⫽ 84)
Group 2 (n ⫽ 41)
Group 3 (n ⫽ 54)
p
0 0 6 (6%) 6 (7.1%)
0 0 4 (9.8%) 4 (9.8%)
2 (3.7%) 4 (7.4%) 8 (14.8%) 14 (25.9%)
0.096 0.009 0.678 0.005
Atrial-ventricular block Residual ventricular septal defect Tricuspid regurgitation post-surgery Total surgical complications a
Results are expressed as absolute numbers (percentage).
(i.e., group 1, 2, or 3). In a stepwise elimination of nonsignificant covariates, the only covariate that was found to be influential was the treatment group. The odds ratio between group 1 and group 2 was 1.4 (95% confidence interval [CI], 0.374 –5.283), whereas the odds ratio between group 1 and group 3 was 4.54 (95% CI, 1.625–12.736). This means that there was a fourfold increase in the risk of complications between the first group and the group of patients that had indications for detachment, but the detachment was not performed (p ⫽ 0.005). All patients were discharged home in stable condition. Two patients in group 3 had postoperative conduction system disturbances develop (complete atrioventricular block and atrioventricular dissociation); one of them needed a permanent DDD-pacemaker implantation. The duration of postoperative hospitalization was 7.43 ⫾ 2.64 days and was not significantly different between the three groups. There were no significant differences between the groups concerning cardiopulmonary bypass time and cross-clamp time (Table 1). Additional sutures for TR repair were added in 17 patients according to in situ findings (4 in group 1, 4 in group 2, and 9 in group 3). Postoperative echocardiographic assessment before discharge revealed 18 patients with new TR (6 from group 1 and 4 from group 2, all with mild TR), and in group 3 there were 6 patients with mild TR and 2 with mild to moderate TR. No patient underwent further surgery to correct the TR in either of the groups. Residual VSD was detected in 4 children. All were in group 3, whereas, despite the presence of a preoperative indication for TVD, it was not carried out. Two patients had small residual VSDs detected by intraoperative transesophageal echocardiography, which were confirmed by an intraoperative QP/QS measurement of less than 1.5:1. They had an uneventful postoperative course and needed no further surgery. The third patient had immediate VSD closure after transesophageal echocardiography revealed a large residual VSD with a QP/QS measurement of 2.3:1 at the end of surgery. The fourth child had a small residual VSD (QP/QS 1.2:1) that suddenly enlarged (QP/QS 2.5:1) on the second postoperative day. This child was successfully reoperated on the same day using the TVD technique. The VSD in the first 2 patients was closed using TVD because of tricuspid valve aneurysm and in the last 2 patients because of the tricuspid valve chordal arrangement.
Comment As outlined in the Results section, the main findings of our study are that a fourfold increase in the risk of complications exists between group 2 (patients who had indications for TVD and detachment was performed) and group 3 (patients who had indications for TVD but detachment was not performed). p value was found to be ⫽ 0.005. Successful transatrial repair of VSD requires adequate exposure of the margins to avoid residual shunt, injury of conduction tissue, and distortion of the leaflets of the tricuspid valve [1, 2, 10]. The use of TVD in our patients did not result in an increased incidence of residual VSD or surgically induced heart block. In group 2, in which tricuspid detachment was indicated according to our criteria and was actually performed, there was no need for reoperation because of residual VSD or pacemaker implantation. This confirms the conclusions of previous studies stating that with multiple chordal attachments crossing over the defect and inserting at the edge of the VSD, it is difficult to examine and reach all the margins of the defect to improve visualization. Leaflet detachment results in more accurate suture placement [2, 11]. The anomalies of the tricuspid valve, which may be acquired secondary to left-to-right shunting, are frequently associated with peri-membranous defects. These anomalies include excess septal or anterior leaflet tissue that can partially or completely occlude the defect and are associated with spontaneous closure of the VSD [12]. This situation alone, with or without chordal attachments crossing the VSD, may make the exposure of the VSD margins difficult, and the surgeon may need to exert traction on the tricuspid valve leaflets that may cause distortion of the tricuspid subvalvular apparatus and TR. As a result, temporary detachment of the tricuspid valve from the annulus is a good option. We carefully performed intraoperative and postoperative assessments of TR, and in 17 cases additional sutures on tricuspid valve leaflets were needed to minimize TR (ie, most of the cases in group 3). When the VSD extends into the outlet components of the interventricular septum, it necessitates severe traction of the tricuspid valve to improve exposure. Detachment of the septal leaflet and a contiguous portion of the anterior leaflet opens the peri-membranous part of the VSD and gives excellent exposure of all the margins [6]. Although this is an observational study, and not a
CARDIOVASCULAR
Ann Thorac Surg 2006;82:958 – 63
CARDIOVASCULAR
962
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
double-blind study, it has been shown that observational studies may yield similar results to clinical trials [13, 14]. We describe three practical criteria for the indication of TVD during VSD repair. When at least one of these criteria was met, performance of VSD repair with tricuspid detachment was associated with significantly lower incidence of postoperative surgery-related complications as compared with performing surgery without tricuspid detachment, despite the presence of at least one criterion for valve detachment. This study demonstrates that in the presence of echocardiographic and surgical indication criteria, the surgeon has to take into consideration that he can perform TVD with good results. There was no correlation between the degree of postoperative TR and the extent of TVD. If detachment was needed, it usually included the entire septal leaflet and part of the anterior leaflet. Group 3 patients had echocardiographic indications for TVD, but did not undergo the technique because intraoperative assessment revealed that VSD exposure was acceptable. In fact, the exposure of the VSD margins was better than in group 2 patients, with fewer obstructive chordae and relatively small aneurysm tissue. Yet these easier cases yielded the worst results. This supports our findings that TVD is safe and effective, and should be used whenever intraoperative exposure is compromised. Among the 3 preoperative indications for TVD, multiple chordal arrangements (indication number 1) has the strongest predictive value.[15, 16] The authors wish to thank Dawn Mizrahi for her editorial assistance and the Save A Child’s Heart Foundation.
References 1. Gaynor JW, Obrien JE Jr, Rychik J, Sanchez GR, DeCampli WM, Spray TL. Outcome following tricuspid valve detachment for ventricular septal defects closure. Eur J Cardiothorac Surg 2001;19:279 – 82. 2. Aeba R, Katogi T, Hashizume K, et al. Liberal use of tricuspid valve detachment for transatrial ventricular septal defect closure. Ann Thorac Surg 2003;76:1073–7.
Ann Thorac Surg 2006;82:958 – 63
3. Bol-Raap G, Weerheim J, Kappetein AP, Witsenburg M, Bogers AJJC. Follow-up after surgical closure of congenital ventricular septal defect. Eur J Cardiothorac Surg 2003;24: 511–5. 4. Hudspeth AS, Cordell AR, Meredith JH, Johnston FR. An improved transatrial approach to the closure of ventricular septal defects. J Thorac Cardiovasc Surg 1962;43:157– 65. 5. Maile S, Kadner A, Pretre R. Detachment of the anterior leaflet of the tricuspid valve to expose perimembranous ventricular septal defects. Ann Thorac Surg 2003;75:944 – 6. 6. Pridjian AK, Pearce FB, Culpepper WS, Williams LC, Van Meter CH, Ochsner JL. Atrioventricular valve competence after takedown to improve exposure during ventricular septal defect repair. J Thorac Cardiovasc Surg 1993;106: 1122–5. 7. Kapoor L, Gan MD, Bandyhopadhyay A, Das MB, Chatterjee S. Improved exposure of isolated perimembranous ventricular septal defects. Ann Thorac Surg 2000;69:291–2. 8. Tatebe S, Miyamira H, Watanabe H, Sugawara M, Educhi S. Closure of isolated ventricular septal defect with detachment of the tricuspid valve. J Card Surg 1995;10:564 – 8. 9. Koshy S, Sunil GS, Anil SR, Dhinakar S, Shivaprakasha K, Rao SG. Tricuspid valve detachment for transatrial closure of ventricular septal defects. Asian Cardiovasc Thorac Ann 2002;10:314 –7. 10. Zhao J, Li J, Wei X, Zhao B, Sun W. Tricuspid valve detachment in closure of congenital ventricular septal defect. Tex Heart Inst J 2003;30:38 – 41. 11. Frenckner BP, Olin CL, Bomfim V, Bjarke B, Wallgren CG, Bjõrk VO. Detachment of the septal tricuspid leaflet during transatrial closure of isolated ventricular defect. J Thorac Cardiovasc Surg 1981;82:773– 8. 12. McDaniel NL, Gutgesell HP. Ventricular septal defects. In: Allen HD, Gutgesell HP, Clark EB, Driscoll DJ, eds. Moss and Adams’ heart disease in infants, children, and adolescents, 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2001: 636 –51. 13. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med 2000;342: 1878 – 86. 14. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med 2000;342:1887–92. 15. Kouchoukos NT, Blackstone EH, Doty DB, Hanley FL, Karp RB. Ventricular septal defect. In: Kirklin and Barratt-Boyes, Eds. Cardiac surgery: morphology, diagnostic criteria, natural history, technique, results, and indications, 3rd edition. Philadelphia: Churchill Livingstone, 2003:850 –909. 16. de Leval M. Ventricular septal defects. In: Stark J, de Leval M, eds. Surgery for congenital heart defects, 2nd ed. Philadelphia: WB Saunders Co, 1994: 355–71.
DISCUSSION DR CARL L. BACKER (Chicago, IL): You used mostly a circumferential detachment, or was that used in all patients versus a radial opening in the tricuspid valve? DR SASSON: I used a parallel incision parallel, to the annulus. DR BACKER: Because in our practice we very frequently open the septal leaflet with a radial-type incision which unfolds it. I personally find it a little easier to put back together, and I guess it would be interesting to compare the results, although I am impressed with your very low incidence of heart block or tricuspid valve insufficiency after this type of opening of the tricuspid valve.
DR SASSON: At the end of the repair, I usually check the tricuspid valve with saline. And whenever there is regurgitation, I usually take the time and repair it. Sometimes I use stitches to approximate the septal and the anterior leaflet by the commissure. And at times when it seems that one of the papillary muscles of the septal leaflet is being elevated because of the ventricular septal defect (VSD) repair, a shortening of the chorda brings the competence of the valve to be nice. DR CHRISTOPHER KNOTT CRAIG (Oklahoma City, OK): The echocardiography (echo) on which you based your decision whether to detach the tricuspid valve or not, was that done preoperatively or intraoperatively?
DR SASSON: The echo was done preoperatively on two occasions: one, transthoracic echocardiography and the second was preoperatively by the transesophageal echocardiography (TEE). DR KNOTT CRAIG: But is the TEE done when the patient is asleep in the operating room? DR SASSON: Yes. DR KNOTT CRAIG: How often did you find that the preoperative transthoracic echo and the intraoperative TEE, differed significantly with regard to the perceived need to detach the valve leaflet? DR SASSON: I was really impressed by the results, that they were pretty much the same. We have, it seems, good cardiologists and they gave a very good diagnosis preoperatively for the transthoracic. It’s true that the intraoperative TEE could show better if there is more chordae or a septal aneurysm that can make the repair difficult. DR KNOTT CRAIG: Since you had intraoperative TEE, why were the large residual VSDs not addressed at that time? DR SASSON:They were addressed at the time. One patient addressed immediately after coming off bypass, and we saw that there is a VSD so we came in and closed it. And in another patient, the patch dehisced 2 days after the operation and we had to come back and close it again. DR MRINALENDU DAS (Kolkata, India): Regarding your group 3, I understand that the incidence of residual VSD and tricuspid regurgitation is high because of the presence of chordae, and you have not detached the tricuspid valve. But could you explain why the incidence of heart block is so high, 37.7%, which seems to equate pretty high. DR SASSON: The incidence of heart block in the last group is 3.7% compared to none in the other groups. But if you take all 179 patients that were operated on, only one patient had to have a pacemaker implantation. That makes it a very low incidence, even less than 0.5%. This group is the point of this paper. In cases where indications for detachment were met, but detachment was not performed, the struggle for good exposure probably caused the heart block. DR BOHDAN MARUSZEWSKI (Warsaw, Poland): I just want to make sure about the age of your patients, because you don’t mention it in the abstract, but on the presentation I saw it’s 4 years. So the first question is the age of these patients. And it’s not the usual age where we close the VSDs. So do you think this technique could be also easily applicable to the infants of 4 kg when we close the VSD? And the second question is, are you aware of any information about long-term follow-up after detachment of the tricuspid valve leaflet in infants undergoing surgery for VSD?
SASSON ET AL CRITERIA FOR TRICUSPID VALVE DETACHMENT
963
DR SASSON: In regard to the first question, our practice in closing the VSD is usually below 2 years of age. And if there is a necessity, we do it even below 1 year of age, according to failure to thrive or congestive heart failure. In our center, we receive most of our most of our patients through the Save A Child’s Heart Foundation, which is a charity organization that brings children from Third World countries to Israel for treatment. Since 1995 we have operated on 1,400 children. So we do not get to choose our patients. And sometimes we get patients with VSDs who may be even 14 years of age. As to if this detachment can be done in younger children, absolutely true. There were a few patients who weighed about 4 kg, where the VSD was difficult to expose, and I had to detach the valve. In those kids, I think that it might be even more justified, because the tricuspid leaflet is so thin, it’s like a butterfly’s wings, and excessive traction of these may cause significant TR postoperatively. So when I detached the valves in those patients and reattached it, I found no residual VSDs and no significant TR. There are a few articles that actually were published in 2003 and 2004 that examined patients for the long term at about 8 years follow-up. And it seems that there weren’t any problems regarding tricuspid regurgitation. There was concern whether the detachment may affect the growth of the leaflet and may cause later tricuspid regurgitation. This was not found. DR DOUGLAS D. PAYNE (Boston, MA): After you’ve made your incision in the tricuspid valve, do you do the whole VSD patch replacement through that hole, or is it partly through that hole and partly through the orifice? The second question is, when you’re working through this hole, I envision it’s rather small. Do you have any trouble with tearing of the valve, particularly the extension of the incision out into the free edge of the leaflet? DR SASSON: Usually I do it through the hole. When I make the incision, I get a good exposure of the margins of the VSD. You need to take care not to take bites too much towards the left ventricle, otherwise you may hit the conduction system. But usually you get a nice exposure of the VSD. There were times that I had to close the VSD or assist a few stitches, not through that opening. And the second question? DR PAYNE: Were there any injuries of the tricuspid valve from traction on that incision in the septal leaflet? DR SASSON: No, no. This is the reason that I just make the incision a little more than 2 mm from the annulus, mark the area of the beginning and the end of it, and exercise very gentle traction on the tricuspid valve, especially in small children. DR BACKER: Why don’t we take a quick poll. How many people here have no problem whatsoever detaching the tricuspid valve during VSD closure? Show of hands. Nearly everyone has raised their hand. Does anyone strongly object to this approach or think this is a bad technique? No one is raising their hand. I think you’ve convinced us. We were convinced before we came and we continue to be convinced.
CARDIOVASCULAR
Ann Thorac Surg 2006;82:958 – 63