Flap Valve Double Patch Closure of Ventricular Septal Defects in Children With Increased Pulmonary Vascular Resistance William M. Novick, MD, Nestor Sandoval, MD, Vasiliy V. Lazorhysynets, MD, PhD, Victor Castillo, MD, Alexander Baskevitch, MD, Xiomung Mo, MD, Robert W. Reid, MD, Branko Marinovic, MD, PhD, and Thomas G. Di Sessa, MD Departments of Surgery and Pediatrics, University of Tennessee Health Sciences Center, Memphis, Tennessee; Department of Cardiac Surgery, Fundacion Abood Shaio, Bogotá, Colombia; Department of Congenital Surgery, Amosov Institute of Cardiovascular Surgery, Kyiv, Ukraine; Instituto del Corazon, Santander, Colombia; Department of Cardiac Surgery, Children’s Surgical Center of Minsk, Minsk, Belarus; Department of Cardiac Surgery, Nanjing Children’s Hospital, Nanjing, China; International Children’s Heart Foundation, Memphis, Tennessee; Department of Pediatrics, Clinical Hospital Center of Zagreb, Zagreb, Croatia; and Department of Pediatrics, University of Kentucky, Lexington, Kentucky
Background. Closure of a large ventricular septal defect (VSD) in children with elevated pulmonary vascular resistance (PVR) is associated with significant morbidity and mortality. Sophisticated medications and circulatory assist devices may not be available to assist in the care of children with elevated PVR undergoing VSD closure. We designed a fenestrated flap valve double VSD patch to decrease the morbidity and mortality associated with the closure of a large VSD in this high-risk group. Methods. Ninety-one children (median age 4.0 ⴞ 3.1 years) with a large VSD and elevated PVR (10.5 ⴞ 4.9 Wood units) underwent double patch VSD closure. The routine VSD patch was fenestrated (4 to 8 mm), and on the left ventricular side of the patch, a second smaller patch was attached to the upper third of the fenestration before VSD patch placement.
T
he closure of a large ventricular septal defect (VSD) is usually performed, in most industrialized countries, at an early age before the onset of elevated pulmonary vascular resistance (PVR). Although the occurrence of a pulmonary hypertensive crisis can contribute to morbidity and mortality, most children undergoing closure of a large VSD in infancy can expect excellent long-term results. In most medically sophisticated countries of the world, these pulmonary hypertensive events can be managed with either sophisticated pharmacologic agents such as nitric oxide or circulatory assist devices such as extracorporeal membrane oxygenation (ECMO) [1, 2]. Although children with a large VSD and elevated PVR are rarely seen in surgically advanced countries, they continue to represent a significant proportion of the congenital heart disease population in the remainder of the world. The prognosis for children undergoing surgical closure of a
Results. Fifty-six children with a VSD as the primary lesion, 16 with complete atrioventricular canal, 10 with double outlet right ventricle/VSD, 2 with interrupted aortic arch/VSD, 2 with truncus arteriosus, and 1 each with transposition/VSD, corrected transposition/VSD, total anomalous pulmonary venous connection/VSD, VSD/left pulmonary artery atresia, and aortopulmonary window underwent operation; the overall early mortality rate was 7.7% (7 of 91). There have been 7 late deaths: 2 VSD and 5 complex defects. Conclusions. Closure of a large VSD with elevated PVR can be performed with reasonable mortality and morbidity. (Ann Thorac Surg 2005;79:21– 8) © 2005 by The Society of Thoracic Surgeons
large VSD with elevated PVR is dependent upon the age and PVR at presentation [3]. These children are at an increased risk for significant morbidity and mortality even when closure is performed in infancy [4]. Confronted with this problem, we designed a simple fenestrated flap valve VSD closure patch to reduce the morbidity and mortality associated with surgery on children with a large VSD and increased PVR. This report provides the intermediate results of this innovative surgical modification.
Patients and Methods Patient Demographics and Sites of Operation Ninety-one patients with a large VSD and pulmonary hypertension underwent VSD closure between May 1996
Accepted for publication June 25, 2004. Presented at the Fortieth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 26 –28, 2004. Address reprint requests to Dr Novick, 1750 Madison Ave, Suite 100, Memphis, TN 38104; e-mail:
[email protected].
© 2005 by The Society of Thoracic Surgeons Published by Elsevier Inc
This article has been selected for the open discussion forum on the CTSNet Web site: http://www.ctsnet.org.discuss
0003-4975/05/$30.00 doi:10.1016/j.athoracsur.2004.06.107
CARDIOVASCULAR
ORIGINAL ARTICLES: CARDIOVASCULAR
CARDIOVASCULAR
22
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
Abbreviations AV ⫽ CPB ⫽ ECMO ⫽ ICU ⫽ LV ⫽ PA ⫽ PAs/Aos ⫽ PVR Qp/Qs
⫽ ⫽
SpO2 VSD
⫽ ⫽
and Acronyms atrioventricular cardiopulmonary bypass extracorporeal membrane oxygenation intensive care unit left ventricle pulmonary artery ratio of pulmonary artery systolic pressure to aortic systolic pressure pulmonary vascular resistance ratio of pulmonary blood flow to systemic blood flow arterial hemoglobin oxygen saturation ventricular septal defect
and February 2003. Fifty-two patients (57%) were female. The age of the patients ranged from 5 months to 17 years with a median age of 4.0 ⫾ 3.1 years. The cities in which Table 1. Preoperative Diagnosis and Operation Performed in Addition to DFV/VSD Closure Defect VSD VSDm VSD/ASD VSD/ASD/PDA
Number of Patients Additional Procedure Performed 40 2 8 1
VSD/PDA VSD/PAB AVC AVC/PAB
2 3 7 2
AVC/PDA
7
DORV DORVm
6 2
DORV/PDA DORV/PAB Truncus arteriosus TGA/VSD IAA/VSD CTGA/VSD TAPVC/VSD APW LPA At/VSD
1 1 2
None Routine patch closure 2nd VSD Patch closure ASD ASD suture closure, PDA ligation PDA ligation PA reconstruction Two patch AVC repair Two patch AVC repair, PA reconstruction Two patch AVC repair, PDA ligation IVT DORV repair IVT DORV repair, Routine patch closure 2nd VSD IVT DORV repair, PDA ligation ASO RV/PA conduit
1 2 1 1 1 1
ASO IAA repair None TAPVC repair DFV/APW repair, transaortic None
APW ⫽ aortopulmonary window; ASD ⫽ atrial septal defect; ASO ⫽ arterial switch operation; At ⫽ atresia; AVC ⫽ complete atrioventricular canal defect; CTGA ⫽ corrected transposition of the great arteries; DFV ⫽ double patch flap valve; DORV ⫽ double outlet right ventricle; IAA ⫽ interrupted aortic arch; IVT ⫽ intraventricular tunnel; lig ⫽ ligation; LPA ⫽ left pulmonary artery; m⫽ multiple VSD’s; PA ⫽ pulmonary artery; PAB ⫽ previously placed pulmonary artery band; PDA ⫽ patent ductus arteriosus; reconst ⫽ reconstruction; rep ⫽ repair; RV/PA ⫽ right ventricular to pulmonary artery; TAPVC ⫽ total anomalous pulmonary venous connection; VSD ⫽ ventricular septal defect.
Ann Thorac Surg 2005;79:21– 8
the operations were performed were Kyiv, Ukraine (21), Minsk, Belarus (15), Zagreb, Croatia (13), Santander, Colombia (11), Bogotá, Columbia (9), Lima, Peru (6), Nanjing, China (4), Memphis, Tennessee (4), Managua, Nicaragua (2), Maracaibo, Venezuela (2), Jerusalem, Israel (2), and Shanghai, China (2). The children who received operations in Memphis were from Bosnia (2), Ukraine (1), and Bangladesh (1).
Preoperative Evaluation All children had a preoperative two-dimensional echocardiographic Doppler evaluation, and in 12 this was the sole preoperative diagnostic test. Seventy-nine children underwent cardiac catheterization. All catheterizations were performed in the patient’s country of origin, and maneuvers to manipulate the PVR, including the use of 100% oxygen, were not routinely employed. Preoperative diagnoses are listed in Table 1. For the sake of evaluating outcome, the patients were divided into two groups: those requiring surgery primarily for a VSD (simple), and those requiring surgery for more complicated defects (complex). All patients listed underwent operative correction. Patients were evaluated independently at each site and suitability for surgery was determined locally; no data were kept on those patients refusing surgery or those deemed inoperable by the local team.
Operative Management All surgical procedures were performed by one of the authors (W.M.N., N.S., V.V.L., V.C., A.B., X.M.). Routine cardiopulmonary bypass (CPB) with normothermia was employed for all cases of simple VSD closure from 1999 onward. Moderate hypothermia (28°C) was used for all complex cases except those requiring an arterial switch procedure or arch reconstruction. Cold cardioplegic arrest with either blood based or crystalloid cardioplegia was used in all cases. The flap valve fenestrated VSD
Fig 1. Illustration of double flap valve patch in place.
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
23
Table 2. Criteria for Early Extubation 1. Must be adequately recovered from anesthesia with spontaneous respirations. 2. Adequate hemostasis (bleeding ⬍ 5 mL/kg). 3. Normal acid-base balance. 4. Temperature must be ⱖ 36.5°C. 5. Adequate cardiopulmonary performance (ie, no significant arrhythmias, adequate perfusion, acceptable systemic blood pressure, filling pressures and 0 to 2 vasoactive medications for cardiac support). 6. Agreement between anesthesiologist and surgeon for extubation in the operating room and surgeon and intensivist in the intensive care unit.
tain moderate hypocarbia while in the operating room were employed for all patients.
Postoperative Management Fig 2. Illustration of double flap valve patch in profile with open valve.
closure patch was constructed of Sauvage Dacron (C. R. Bard, Murray Hill, New Jersey) and Gore-Tex (W. L. Gore & Associates, Newark, Delaware) in 3 of the patients early in our experience and exclusively of Gore-Tex in the remainder. The VSD patch was tailored after inspection of the defect, and the fenestration was then made in the center of the patch for all defects except atrioventricular canal defects. The fenestration was placed in the lower third of the patch when correcting atrioventricular (AV) canal defects to prevent the flap valve from interfering with left AV valve excursion. The size of the fenestration was determined for each child at the time of operation. Fenestrations were sized according to the expected aortic annulus size for each child. Once the size was determined, a fenestration that was one half of the expected aortic annulus diameter was made in the patch. A separate flap patch at least 4 mm larger than the fenestration was then constructed and sewn onto the superior margin of the fenestration along one third of the circumference. A separate tethering stitch was placed at the inferior apex of the flap valve and tied loosely over a Hegar dilator that was the same size as the fenestration. Thus, the tethering stitch length approximated the diameter of the fenestration. The VSD patch was then sewn into place orienting the patch so that the flap valve was placed on the left ventricular (LV) side and directing the flap so that it would open toward the LV apex (Figs 1 and 2). No patient was intentionally left with an atrial level communication. Cardiac support medications and vasodilators were started according to local preferences before weaning from CPB. Additionally, the use of modified ultrafiltration was dictated by local customs and based upon economic ability to acquire the modified ultrafiltration circuit. The use of a transthoracic pulmonary artery catheter and the performance of an open lung biopsy were determined at each site. After CPB, efforts to main-
Patients were allowed to awaken from anesthesia after operation, and no attempts were made to routinely manage patients with pulmonary hypertension using neuromuscular blockade or sedation. Patients were extubated if they fulfilled criteria (Table 2) whether they were in the operating room or intensive care unit (ICU). Early extubation was aggressively pursued in all patients who exhibited adequate criteria [5]. Discontinuation of inotropic drugs and intravenous vasodilators was accomplished as quickly as the clinical status allowed. No patient was maintained on a regimen of aspirin or anticoagulation.
Follow-Up Evaluation Follow-up was conducted in the countries of origin. All patients were asked to return to the hospital for an echocardiogram and clinical evaluation. A total of 6 operative survivors were lost to follow-up. The remaining survivors were located. Forty-three patients returned for an echocardiogram postoperatively. The echocardiographic follow-up data reported herein were from studies performed within 6 months of the closure of patient enrollment.
Statistical Analysis Data were evaluated for normality with the KolmogorovSmirnov test. Those data with normal distribution (ratio Table 3. Preoperative Catheterization Data
PAp/Aop PVR (Wood units) Qp/Qs SpO2 (%)
Simple Defects
Complex Defects
0.99 ⫾ 0.10 11.7 ⫾ 5.4 1.5 ⫾ 0.6 90 ⫾ 3
0.92 ⫾ 0.13 (p ⫽ 0.008) 9.0 ⫾ 3.7 (p ⫽ 0.02) 1.7 ⫾ 1.0 (NS) 86 ⫾ 9 (p ⫽ 0.03)
Values are expressed as mean ⫾ SD (PAp/Aop, PVR, Qp/Qs) and median ⫾ interquartile deviation (SpO2). NS ⫽ not significant; PAp/Aop ⫽ peak systolic pulmonary artery pressure/peak systolic aortic pressure; PVR ⫽ pulmonary vascular resisSpO2 ⫽ tance; QP/QS ⫽ pulmonary blood flow/systemic blood flow; systemic arterial saturation.
CARDIOVASCULAR
Ann Thorac Surg 2005;79:21– 8
CARDIOVASCULAR
24
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
Ann Thorac Surg 2005;79:21– 8
Fig 3. Kaplan-Meier estimates of overall survival among patients with simple defects (top curve) versus complex defects (bottom curve). Survival in these two groups was significantly different (p ⫽ 0.003, complex versus simple, by log-rank method). A survival curve for the full study group is included for reference (middle curve).
Fig 5. Kaplan-Meier survival estimates of overall survival among children with postoperative echocardiographic pulmonary artery systolic pressure less than 40 torr (solid line) versus greater than 40 torr (dashed line). Survival between these two groups is not statistically different (p ⫽ not significant).
of pulmonary artery systolic to systemic systolic pressure [PAs/Aos], pulmonary vascular resistance, and the ratio of pulmonary blood flow to systemic blood flow [Qp/Qs])
are presented as mean ⫾ standard deviation and were analyzed with the Student t test and confirmed with the use of the Mann-Whitney rank sum test. Those data with nonnormal distribution (age, weight, arterial hemoglobin oxygen saturation [SpO2], CPB time, ischemic time, and time to extubation) are presented as median ⫾ interquartile deviation and were analyzed with the nonparametric Wilcoxon rank sum test. Time to extubation is defined as the time interval between transport from operating room until extubation in the intensive care unit. Categorical data were evaluated using the 2 test or Fisher’s exact test. The correlation between age, PVR, and survival time was analyzed with linear regression and reported with the R2 adjusted value. A p value of 0.05 or lower on any test was considered to indicate statistical significance. All tests were two-tailed. Survival between those children with simple and complex defects was analyzed with the Kaplan-Meier method and statistical significance was determined by log-rank method. Survival times are expressed as mean ⫾ standard error.
Results Preoperative Data
Fig 4. Kaplan-Meier survival estimates of overall survival among children with preoperative pulmonary vascular resistance less than 10 Wood units (dark line) versus greater than 10 Wood units (light line). Survival between these two groups is not statistically different (p ⫽ not significant).
No significant difference was found in the sex mix between the simple and the complex groups. The median age for the simple group was 5.1 ⫾ 3.4 and for the complex group, 2.3 ⫾ 2.2 (p ⫽ 0.005). Median weight similarly differed significantly between the two groups (simple 16.0 ⫾ 5.9, complex 10.7 ⫾ 4.5; p ⬍ 0.03). For the entire study population, the preoperative room
air SpO2 obtained at cardiac catheterization was 90% ⫾ 4%, and the ratio of pulmonary artery systolic to systemic systolic pressure was 0.96 ⫾ 0.12. In addition, the PVR and Qp/Qs were 10.5 ⫾ 4.9 and 1.6 ⫾ 0.8, respectively. Comparisons of the preoperative values for the simple group versus the complex group are presented in Table 3. Significant differences between the two groups were apparent in PVR, SpO2, and PAs/Aos.
Operative Procedures All VSD were closed using the flap valve double patch. The operative procedures performed on the patients are listed in Table 1. All patients were weaned from CPB. Two patients had acute cardiovascular collapse after CPB, 1 upon completion of modified ultrafiltration and 1 after administration of protamine. Both patients were in the complex group (AV canal, modified ultrafiltration) and subsequently died in the operating room after prolonged attempts at resuscitation, including return to CPB. A total of 47 patients underwent modified ultrafiltration after CPB. A significant difference was noted in the percentage of patients undergoing modified ultrafiltration when the simple (41%) and complex groups (76%) were compared (p ⬍ 0.003).
Postoperative Data Except for 1 child with pneumonia who remained intubated for 12 days, all patients surviving to hospital discharge were extubated either in the operating room or within the first 65 hours in the ICU. The time to extubation for the remaining children was 8 ⫾ 7.5 hours. No significant difference was noted when extubation time for the simple group (7.0 ⫾ 7.0) was compared with the complex group (11.0 ⫾ 8.5). Early extubation was defined as extubation in the operating room or within four hours after arrival in the ICU. No difference was noted in the percentage of patients undergoing early extubation between the simple (36%) and complex groups (26%). We did find a significant difference (p ⫽ 0.004) in the early extubation rate between those patients who received modified ultrafiltration (46%) and those who did not (14%). Five of 89 patients admitted to the ICU died. Two patients died because of technical problems, 1 in each group. One patient, in the simple group, had a flap valve patch that had been inserted such that when it opened it did so into the LV outflow tract, causing LV outflow tract obstruction (observed on transthoracic echocardiography) during a pulmonary hypertensive event. No other complications were noted with patch placement. The other child (complex; modified ultrafiltration) died when the PA transthoracic line was removed in the ICU on postoperative day 2. This patient exsanguinated upon removal of the line. Two children died (1 simple and 1 complex; truncus) on postoperative day 4 and 8 secondary to hypoxemia resulting from pneumonia. One child (complex; AV canal) died secondary to the development of a dysrrhythmia on postoperative day 2.
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
25
Postoperative Echocardiograms Forty-three patients were evaluated in follow-up. Pulmonary hypertension estimated by Doppler (mean PA pressure greater than 25 mm Hg) was present in 26 and normal pulmonary pressure in 17. Eight patients in the pulmonary hypertensive group displayed right to left shunting across the flap valve. There was no correlation between the preoperative PVR and the postoperative Doppler estimation of pulmonary artery pressure.
Early and Late Morbidity and Mortality Ninety-one children were operated on using the flap valve double patch VSD closure method. There were 7 hospital deaths (7.7%). Hospital survival was significantly different (p ⬍ 0.05) when the simple group (54 of 56, 96.4%) was compared with the complex group (30 of 35, 85.7%). Follow-up is available for 78 of the 84 hospital survivors (93%). All children with complex defects had follow-up, and 6 with simple VSD were lost to follow-up. There have been 7 late deaths (7 of 78, 8.9%): 2 in the simple group and 5 in the complex group. No thromboembolic events have occurred to date, and other than the 1 patient mentioned above, no flap-valve dysfunction has been observed. Mean survival time in the simple group was 7.1 ⫾ 0.3 years (range, 0 to 6.9); and in the complex group it was 4.9 ⫾ 0.6 years (range, 1 month to 7.6 years). Survival for the simple group was significantly better than that for the complex group (p ⫽ 0.003; Fig 3). Moreover, survival was not dependent upon preoperative PVR, as there was no statistical significance in survival when patients with a preoperative PVR less than 10 Wood units were compared with patients with a PVR greater than 10 Wood units (Fig 4). In addition, there was no statistical difference in survival observed when we compared patients with a postoperative echocardiographic PA systolic pressure less than 40 torr and those with a PA systolic pressure greater than 40 torr (Fig 5). One child required reoperation (4 years postoperatively) for progressive mitral valve regurgitation after AV canal repair and another child needed reoperation (5 years postoperatively) for subaortic fibrous ridge after simple VSD closure. A residual left-to-right shunt was present in 4 children, and all were trivial by echocardiography.
Comment Pulmonary hypertension after closure of a large VSD continues to cause significant morbidity and mortality even in industrialized countries [6]. The use of nitric oxide and ECMO rescue has reduced the mortality, but with significant cost and patient morbidity. These expensive and sophisticated modalities are not available in many countries throughout the world. Even the additional expense of a hemoconcentrator precluded the use of the relatively inexpensive device in some of the sites represented in this study. We previously described the use of the flap valve double patch VSD closure technique as a method which could provide surgeons with an inexpensive option for the management of children with
CARDIOVASCULAR
Ann Thorac Surg 2005;79:21– 8
CARDIOVASCULAR
26
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
pulmonary hypertension and an elevated PVR [7]. The double patch technique provides a simple physiologic mechanism for unloading the right ventricle during periods of severe pulmonary hypertension whether acute and transient or sustained. Although children with a large VSD and elevated PVR are rarely seen in industrialized countries, they remain a very significant population of patients seen in the remainder of the world. Kannan and colleagues [8] reported their results on the closure of a large VSD in patients with elevated PVR and concluded that closure was warranted. The early mortality rate in their series using a nonfenestrated patch was 13.1%, which is considerably higher than the 3.6% observed in the simple group in our series. The children in the Kannan study were older (7.5 years) than those in this study (4.0 years). However, the PVR of patients in the former study was less (7.6 ⫾ 1.8 Wood units) than the PVR presented herein (10.5 ⫾ 4.9 Wood units). We believe that this difference supports our hypothesis that the double flap valve VSD closure technique provides for a lower mortality rate among children with pulmonary hypertension and an elevated PVR. The development of pulmonary hypertensive events after closure of a large left to right shunt is known to delay extubation [9] and prolong length of stay in the ICU after operation [10]. Utilizing the double patch flap, valve we were able to extubate the children rapidly and thereby diminish the need for prolonged mechanical ventilation. In addition, prolonged stay in the ICU is associated with an increase in septic complications [11] and other adverse events. Moreover, early extubation reduces the use of resources that are already limited in a number of developing countries [12]. The results of closure of a large VSD in patients with an elevated PVR are well documented [13, 14]. The closure of VSD in patients with a PVR of 10 Wood units or greater is considered to be contraindicated [15] owing to high operative mortality and debatable long-term results. We would agree with others that the ideal situation is to operate on children before the development of an elevated PVR. However, owing to delays in diagnosis in many underserved regions of the world, [16] this is not practical. The strategy outlined by Wessel [17] for children in industrialized countries who present with an elevated PVR is costly and logistically impractical in many areas of the world. Pediatric lung or heart-lung transplantation is simply not available in most countries and therefore not an option for children with congenital heart disease and severely elevated PVR. The 5-year survival for children after lung or heart-lung transplant is reported to be 40% [18]. The intermediate survival of patients in our series irrespective of PVR and defect complexity was greater than 60%. The double patch flap valve closure would, therefore, appear to be a promising alternative for children in certain countries who do not have access to these two sophisticated treatment plans. Previous reports have shown that intermediate results after closure of a large VSD with pulmonary hypertension appear to be dependent upon the Heath-Edwards classification at the time of closure [19]. Other studies,
Ann Thorac Surg 2005;79:21– 8
however, have produced conflicting results regarding the outcome of closure of a large VSD when PVR is elevated [20]. Our VSD method of closure confirms reports that early and intermediate survival of patients with an elevated PVR may not be dependent upon the preoperative value of the PVR. Reports in patients with primary pulmonary hypertension have shown improved hemodynamics and patient exercise capability after long-term treatment with oral calcium-channel blockers [21], endothelin inhibitors [22], or continuous intravenous prostacyclin [23], suggesting that vascular remodeling may take place, especially in children. Barst [24], however, has cautioned against assuming that these beneficial effects observed in primary pulmonary hypertension will necessarily be seen in children with congenital heart disease with pulmonary hypertension and elevated PVR. Recent reports of providing substrates for nitric oxide [25] or using endothelin blockers [26] and phosphodiesterase inhibitors [27] to diminish postoperative pulmonary hypertension and elevated PVR are encouraging. However, Stocker and coworkers [28] have suggested caution in using some of these agents. We have shown that with the placement of a unidirectional flap valve VSD patch children with pulmonary hypertension and elevated PVR can undergo operation with reasonable morbidity and mortality in the absence of sophisticated pharmacologic or mechanical intervention. Intermediate survival is to be expected, but longterm survival requires continued observation of these children. Furthermore, the use of this technique in highrisk pulmonary hypertensive high-flow normal PVR infants may reduce the need for ECMO or nitric oxide in these patients who are at risk for pulmonary hypertensive crises postoperatively.
Study Limitations Although we have follow-up on 93% of the patients, not all have undergone repeat echocardiographic evaluation. We, therefore, cannot determine if other children may have developed progressive pulmonary hypertension. Indeed, some of the children studied by echo-Doppler have developed suprasystemic pulmonary artery pressure with shunting right-to-left. Furthermore, it would be interesting to know if individual lesions had an increased risk of developing pulmonary hypertension after operation when compared with the simple VSD. The complex group is composed of a heterogeneous group of lesions, however, and no one subgroup is large enough for valid statistical comparison with the VSD group. Additionally, although short- and intermediate-term survivals are encouraging, it is the long-term survival that is important. Therefore, long-term follow-up is necessary to determine if the novel approach is successful.
A portion of the funding for this study was provided by the Paul Nemir, Jr, MD, Professorship and Endowment Fund in International Child Health, the University of Tennessee Health Sciences Center, and the International Children’s Heart Foundation.
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
References 1. Goldman AP, Delius RE, Deanfield JE, de Leval MR, Sigston PE, Macrae DJ. Nitric oxide might reduce the need for extracorporeal support in children with critical postoperative pulmonary hypertension. Ann Thorac Surg 1996;62: 750 –5. 2. Goldman AP, Delius RE, Deanfield JE, Macrae DJ. Nitric oxide is superior to prostacyclin for pulmonary hypertension after cardiac operations. Ann Thorac Surg 1995;60:300 –5; discussion 306. 3. Blackstone EH, Kirklin JW, Bradley EL, DuShane JW, Appelbaum A. Optimal age and results in repair of large ventricular septal defects. J Thorac Cardiovasc Surg 1976;72: 661–79. 4. Bando K, Turrentine MW, Sun K, et al. Surgical management of complete atrioventricular septal defects. A twentyyear experience. J Thorac Cardiovasc Surg 1995;110:1543–54. 5. Novick WM, Ivancan V, Colak Z, et al. Early extubation: the initial Croatian experience. Cardiovasc Eng 2002;7:60. 6. Schulze Neick I, Li J, Penny DJ, Redington AN. Pulmonary vascular resistance after cardiopulmonary bypass in infants: effect on postoperative recovery. J Thorac Cardiovasc Surg 2001;121:1033–9. 7. Novick WM, Gurbuz AT, Watson DC, et al. Double patch closure of ventricular septal defect with increased pulmonary vascular resistance. Ann Thorac Surg 1998;66:1533– 8. 8. Kannan BR, Sivasankaran S, Tharakan JA, et al. Long-term outcome of patients operated for large ventricular septal defects with increased pulmonary vascular resistance. Indian Heart J 2003;55:161– 6. 9. Harrison AM, Cox AC, Davis S, Piedmonte M, DrummondWebb JJ, Mee RBB. Failed extubation after cardiac surgery in young children: incidence, etiology, risk factors. Pediatr Crit Care Med 2002;3:148 –52. 10. Brown KL, Ridout DA, Goldman AP, Hoskote A, Penny DJ. Risk factors for long intensive care unit stay after cardiopulmonary bypass in children. Crit Care Med 2003;31:28 –33. 11. Milliken J, Tait GA, Ford Jones EL, Mindorff CM, Gold R, Mullins G. Nosocomial infections in a pediatric intensive care unit. Crit Care Med 1988;16:233–7. 12. Shekerdemian LS, Penny DJ, Novick W. Early extubation after surgical repair of tetralogy of Fallot. Cardiol Young 2000;10:636 –7. 13. Ikawa S, Shimazaki Y, Nakano S, Kobayashi J, Matsuda H, Kawashima Y. Pulmonary vascular resistance during exercise late after repair of large ventricular septal defects. Relation to age at the time of repair. J Thorac Cardiovasc Surg 1995;109:1218 –24. 14. Kidd L, Driscoll DJ, Gersony WM, et al. Second natural history study of congenital heart defects. Results of treat-
15. 16.
17. 18. 19. 20.
21. 22.
23.
24. 25.
26.
27. 28.
27
ment of patients with ventricular septal defects. Circulation 1993;87:I38 –51. de Leval M. Ventricular septal defects. In: Stark J, de Leval M, eds. Surgery for congenital heart defects. Philadelphia: WB Saunders, 1994:355–72. WHO Expert Committee. Health manpower requirements for the achievement of health for all by the year 2000 through primary health care. Report of a WHO Expert Committee. WHO Tech Report 1985;717:7–92. Wessel DL. Current and future strategies in the treatment of childhood pulmonary hypertension. Progr Pediatr Cardiol 2001;12:289 –318. Spray TL, Bridges ND. Lung transplantation for pediatric pulmonary hypertension. Progr Pediatr Cardiol 2001;12:319 –25. Fried R, Falkovsky G, Newburger J, et al. Pulmonary arterial changes in patients with ventricular septal defects and severe pulmonary hypertension. Pediatr Cardiol 1986;7:147–54. Hallidie Smith KA, Hollman A, Cleland WP, Bentall HH, Goodwin JF. Effects of surgical closure of ventricular septal defects upon pulmonary vascular disease. Br Heart J 1969; 31:246 – 60. Rich S, Kaufmann E, Levy PS. The effect of high doses of calcium-channel blockers on survival in primary pulmonary hypertension. N Engl J Med 1992;327:76 – 81. Hoeper MM, Taha N, Bekjarova A, Gatzke R, Spiekerkoetter E. Bosentan treatment in patients with primary pulmonary hypertension receiving nonparenteral prostanoids. Eur Respir J 2003;22:330 – 4. Barst RJ, Rubin LJ, Long WA, et al. A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional therapy for primary pulmonary hypertension. The Primary Pulmonary Hypertension Study Group. N Engl J Med 1996;334:296 –302. Barst RJ. Diagnosis and assessment of pulmonary hypertension in infants and children. Progr Pediatr Cardiol 2001;12: 279 – 88. Schulze Neick I, Penny DJ, Rigby ML, et al. L-arginine and substance P reverse the pulmonary endothelial dysfunction caused by congenital heart surgery. Circulation 1999;100: 749 –55. Schulze Neick I, Li J, Reader JA, Shekerdemian L, Redington AN, Penny DJ. The endothelin antagonist BQ123 reduces pulmonary vascular resistance after surgical intervention for congenital heart disease. J Thorac Cardiovasc Surg 2002;124: 435– 41. Schulze Neick I, Hartenstein P, Li J, et al. Intravenous sildenafil is a potent pulmonary vasodilator in children with congenital heart disease. Circulation 2003;108(Suppl 1):167–73. Stocker C, Penny DJ, Brizard CP, Cochrane AD, Soto R, Shekerdemian LS. Intravenous sildenafil and inhaled nitric oxide: a randomised trial in infants after cardiac surgery. Intens Care Med 2003;29:1996 –2003.
DISCUSSION DR GERHARD ZIEMER (Tuebingen, Germany): Doctor Sandoval, you have to be congratulated for this series. It is outstanding not only by the assembly of rare pathologies, but also in its geographical distribution of the patients and operating room sites. Once employing special measures or additional procedures, the question has to be raised whether these measures are really necessary. You advocate this double patch. First, I agree that these patients need to have a fenestrated patch at least early postoperatively. Most fatal outcomes are in those patients who really required this hole. But where is the advantage of this flap? At least in my experience, having a hole in this set of patients, you never have to go back 2 or 3 or 5 years later to only close this hole because
you have a hemodynamically significant left-to-right shunt. And in my own practice, I only closed these holes when I had to go back anyway because there were, for example, conduit obstructions or anything else.So what is really the rationale for especially going to Third World countries to employ a more complex technique than actually necessary in dealing with these patients? DR SANDOVAL: Thank you. I think this is not a very complex technique. It is fairly easy. You can spend usually no more than 10 minutes trying to construct the patch. And in some of these countries, we don’t have all these sophisticated and very expensive medications to treat these patients in the postoperative period. So when these patients develop pulmonary hyperten-
CARDIOVASCULAR
Ann Thorac Surg 2005;79:21– 8
CARDIOVASCULAR
28
NOVICK ET AL DOUBLE PATCH VSD CLOSURE
sion, the only thing that happens is that they shunt from right to left as systemic desaturation, but it will allow us enough blood to maintain the systemic cardiac output. So it’s just a simple thing that will maybe give us opportunity not to use all these medications. But I agree with you, maybe one of these patients could end up with a small shunt left to right, and that is why we use the double flap, to prevent left-to-right shunts later in those children who respond by a decrease in PVR. Thank you for the comment. DR ANTONIO CORNO (Lausanne, Switzerland): I have one comment, one question. The comment: I’m very grateful that you presented this patient population, because in Lausanne we are facing the same problem with patients coming from abroad with either VSD or complex diseases and very severe pulmonary hypertension because of late referral. We adopted a completely different approach. I know that we have a lot of distinguished surgeons talking about the fact that pulmonary artery banding is like going back to the time of the dinosaurs, but this is what we do. We prefer having a pulmonary artery banding with our adjustable device (FloWatch; EndoArt, Lausanne, Switzerland), and we wait; after a period of 2 or 3 months with normal pulmonary artery pressure, then we can go and close the VSD or do a repair. And I can tell you that the postoperative course is completely different than in a patient with VSD closure, even with perforated patch, done in presence of pulmonary hypertension. The question is, then, if this is a series of consecutive patients or if you have patients undergoing pulmonary artery banding instead of this approach in this period? DR SANDOVAL: Thank you, Dr Corno. I think your comment is very important. And we also are very concerned about that. We think that for the majority of these patients, we can give them the opportunity for pulmonary vascular remodeling if we just decrease the PA pressure. And we are allowed with this patch for the pressure not to be so high during maybe the first weeks or months, but we don’t know the correct answer. But I agree with you, with your technique these patients can have a remodeling of the pulmonary vasculature, and it could be another way to treat these difficult patients. None of the patients in this series was banded in preparation for the double patch; however, some patients had previously received a banding at other institutions and still had pulmonary hypertension and elevated pulmonary vascular resistance. DR KLAUS MEYER-DELIUS (Caracas, Venezuela): I would like to congratulate Dr Sandoval and his colleagues for this very important papaer, especially for our countries, where we do not have nitric oxide, any possibility of ventricular assist, or ECMO. In many cases we don’t even have a device to close a fenestrated VSD patch. I would like to know if you used any kind of pharmacologic intervention upon the pulmonary vascular resistance in those patients who had late right-to-left shunting through the flap valve VSD closure?
Ann Thorac Surg 2005;79:21– 8
DR SANDOVAL: Thank you, Dr Meyer, for your comments. We are willing to have some prospective study in these patients, who are doing now very well. We have used sildenafil. We have used some prostacyclin. But we would really like to be able to use Bosentan on these patients. Because we know that Bosentan works for the primary hypertensive patients, but we don’t know anything about the Bosentan in this large series of patients with secondary increased pulmonary vascular resistance. DR HUMBERTO RODRIGUEZ (Monterrey, Mexico): You chose different pathologies in which you used this patch. In which one do you consider that it is more difficult to use it, in all that you have shown to us? DR SANDOVAL: As everybody knows, a regular VSD patch is easy to make flat, and there is no special detail in construction. But sometimes, as with an AV canal, you have to make a hole in the lower part of the patch, so when the patch opens it won’t interfere with the chordae. Maybe with a double outlet right ventricle, when we have to make kind of a tunnel, it is hard to fix the patch so that we can assure complete closure of this hole. DR RANDAS BATISTA (Curitiba, Brazil): I have been able to reverse completely lung lesions “extracting oxygen from the PA.” The problem is quantity of oxygen in the PA. Once you extract it, the lung lesions reverse completely. The way I do it? I band the PA. This increases right-left shunt, the saturation in the aorta will fall, and so does the venous blood returning to the RA (maximally desaturated). The band will also lower PAP, too. I think shock is the most potent vasodilator in the body, mainly in the PA. For the lungs, shock means low oxygen and pressure in the PA. Remember, the PA is the only artery in the body that is not supposed to carry oxygen! Within 2 years, with repeated lung biopsies, you’ll see a complete regression of the lung lesions that before were “irreversible.” Then, you can close the VSD and take off the band. I have done it with 23 patients. The oldest is a 46-year-old man, and he is completely cured, he is jogging, he is doing very well with normal lung biopsies. It is written that what causes pulmonary hypertension is the PA hyperflow. But that is not true. What causes hyperresistant PA hypertension is the quantity of oxygen in the PA. If the PA would have had hyperflow with low oxygen, it would not have developed “hyperresistant pulmonary hypertension.” DR DUKE E. CAMERON (Baltimore, MD): You attach your secondary patch on the superior rim of the fenestration. Why that? Is there a reason why you attach it superiorly as opposed to inferiorly? DR SANDOVAL: In the beginning of the series, one of the flap valve patches was upside down. It was up in the top, and it caused a left ventricular outflow tract obstruction. So then we changed the patch, and we put it toward the apex of the ventricle.
