Does an Additional Source of Pulmonary Blood Flow Alter Outcome After a Bidirectional Cavopulmonary Shunt?
Background The bidirectional cavopulmonary shunt has become an important intermediate step in the treatment of pediatric patients with single ventricle physiology who are ultimately destined for palliative surgery. We wanted to know whether there would be risks or benefits if an additional source of pulmonary blood flow was left after a bidirectional cavopulmonary shunt.
Methods and Results We retrospectively reviewed the medical and surgical records of all patients who underwent a bidirectional cavopulmonary shunt at the Children’s Hospital of Wisconsin between January 1991 and December 1993. A total of 43 patients were identified. Anatomic diagnoses included double inlet left ventricle (14 patients), tricuspid atresia (8 patients), pulmonary atresia with intact septum (6 patients), single right ventricle (5 patients), hypoplastic left heart (3 patients), unbalanced atrioventricular septal defect (3 patients), and other complex lesions (4 patients). We then divided the patients into two groups for purposes of analysis. Group 1 had only the cavopulmonary shunt as a source of pulmonary flow (22 patients); group 2 had an additional source of pulmonary flow (21 patients). Patient age at the time of cavopulmonary shunt ranged from 6 months to 12 years, with group 1 patients being younger (31 versus 45 months, P=.05). Group 2 patients had higher postoperative central venous pressures (17.8 versus 14.1 mm Hg, P<.001) and oxygen saturations (86% versus 81%, P<.001) than did group 1 patients. There was no statistical difference between groups in the number of chest tube days or hospital days. There was 1 early death in group 1 related to severe ventricular dysfunction and 1 late death in group 2 related to sepsis. Five patients in group 2 were readmitted to the hospital for drainage of a large chylothorax compared with none in group 1 (P<.02).
Conclusions We conclude that patients with an additional source of pulmonary blood flow after bidirectional cavopulmonary shunt have higher postoperative central venous pressures, have higher oxygen saturations, and are at risk for the late development of a chylothorax.
The bidirectional cavopulmonary shunt has become an important intermediate step in the treatment of infants and children with single ventricle physiology ultimately destined for Fontan palliation.1 2 3 4 5 6 Although the surgical technique has been previously described, few reports have focused on the risks and/or benefits of leaving an additional source of pulmonary blood flow at the time of cavopulmonary shunt. Those who advocate leaving an additional source of pulmonary blood flow postulate that oxygen saturations should be higher and that therefore single ventricle function may be better preserved.7 Also, the pulsatility associated with leaving an additional source of pulmonary blood flow may alter the long-term development of pulmonary arteriovenous fistulae8 and may promote pulmonary artery growth. However, the additional source of pulmonary blood flow also may potentially elevate pulmonary artery pressures and alter long-term single ventricle function by maintaining an excessive volume load. In this study, we reviewed our most recent 3-year experience with the bidirectional cavopulmonary shunt at the Children’s Hospital of Wisconsin, comparing the early outcome of patients left with an additional source of pulmonary blood flow with the early outcome of those having only the bidirectional cavopulmonary shunt as a source of pulmonary blood flow.
We retrospectively reviewed the medical and surgical records of all patients who underwent a bidirectional cavopulmonary shunt for treatment of complex single ventricle at the Children’s Hospital of Wisconsin between January 1991 and December 1993. The patients were then divided into two groups for purposes of analysis. Group 1 patients had only the bidirectional cavopulmonary shunt as a source of pulmonary blood flow; group 2 patients had an additional source of pulmonary blood flow such as persistent antegrade main pulmonary artery flow or a systemic-to–pulmonary artery shunt. Groups 1 and 2 were contemporaneous; however, only 7 of the 22 patients in group 1 had the cavopulmonary shunt before 1993 compared with 19 of the 21 patients in group 2. The decision of whether to leave an additional source of pulmonary blood flow at the time of cavopulmonary shunt was a surgical one, and there were no absolute selection criteria. Group 1 represents a later surgical experience, during which it was felt that the reduction in volume work of the single ventricle was of utmost importance.
Preoperative and postoperative data were analyzed and compared between groups. Data are presented as mean±1 SD unless otherwise indicated. The significance of differences between groups was estimated by the Mann-Whitney U test and the Fisher exact test. A value of P<.05 was considered significant.
Preoperative patient characteristics are summarized in Table 1⇓ (group 1) and Table 2⇓ (group 2). The patient age at the time of bidirectional cavopulmonary shunt in group 1 was 31±29 months (range, 7 months to 10 years) compared with 45±31 months (range, 6 months to 12 years) in group 2. This was of borderline statistical significance (P=.05). There was no significant difference in preoperative oxygen saturation (80% in both groups) or hemoglobin concentration (17.4±1.9 versus 16.6±1.6 g%) between groups. Anatomic diagnoses in group 1 included double inlet left ventricle (6 patients), pulmonary atresia with intact septum (4 patients), hypoplastic left heart syndrome (3 patients), single right ventricle (3 patients), tricuspid atresia (3 patients), unbalanced atrioventricular septal defect (2 patients), and mitral atresia with double outlet right ventricle (1 patient). Anatomic diagnoses in group 2 included double inlet left ventricle (8 patients), tricuspid atresia (5 patients), pulmonary atresia with intact septum (2 patients), single right ventricle (2 patients), unbalanced atrioventricular septal defect (1 patient), and other complex lesions (3 patients—2 with a hypoplastic right ventricle and tricuspid valve straddle and 1 with superoinferior ventricles, ventricular septal defect, and transposition of the great arteries). All but 2 patients in each group had one or more surgical procedures before the bidirectional cavopulmonary shunt. Fourteen patients in group 1 had previous placement of a systemic-to–pulmonary artery shunt compared with 12 patients in group 2. A Norwood procedure was performed in the 3 patients with hypoplastic left heart syndrome and in 1 patient with an unbalanced atrioventricular septal defect, all in group 1. Other surgical procedures included pulmonary artery banding, coarctation repair, enlargement of the bulboventricular foramen, and atrial septectomy.
Surgery was performed via a right thoracotomy in 8 patients, all in group 2. This was the initial approach of choice in patients with adequate antegrade pulmonary blood flow, since cardiopulmonary bypass could usually be avoided. Six of these patients had surgery without bypass, using only systemic heparinization and a bypassing tubular shunt placed in the high superior vena cava and right atrium. In the other 2 patients, cardiopulmonary bypass was used because of inadequate oxygenation with clamping of the right pulmonary artery. Surgery was performed via a median sternotomy with cardiopulmonary bypass in the remaining 35 patients. This is currently the approach of choice, as a hemi-Fontan connection is now performed at the time of bidirectional cavopulmonary shunt, with anastomosis of the lower caval segment to the caudad surface of the right pulmonary artery.
The majority of patients in both groups had end-to-side anastomoses of the right superior vena cava to the right pulmonary artery. However, 5 patients in group 1 and 1 patient in group 2 had bilateral superior venae cavae–to–pulmonary artery anastomoses because of a large left superior vena cava and the absence of a bridging vein between the cavae. Other surgical procedures performed at the time of the cavopulmonary shunt included pulmonary artery reconstruction (9 patients in group 1 and 3 patients in group 2), resection of subaortic stenosis or enlargement of the bulboventricular foramen (4 patients in group 2), and repair of total anomalous pulmonary venous connection (1 patient in group 1).
During this study, there were no intraoperative deaths. There was one early death in group 1 (Table 1⇑, patient 15), a 10-year-old patient with asplenia, single right ventricle, and total anomalous pulmonary venous return. She developed severe ventricular dysfunction after the surgical procedure and could not be stabilized despite maximal inotropic support.
One patient in each group required early reoperation secondary to markedly elevated central venous pressures and clinical superior vena cava syndrome. The patient in group 1 (Table 1⇑, patient 19) was found to have significant right pulmonary artery stenosis and did well after surgical angioplasty. The patient in group 2 (Table 2⇑, patient 7) required plication of the main pulmonary artery secondary to excessive antegrade flow and also did well after reoperation. Postoperative characteristics in these two patients were analyzed after reoperation.
Postoperative Patient Characteristics
No other significant hospital complications occurred in the 42 patients who survived to hospital discharge. There was no significant difference between group 1 and group 2 in the mean hospital stay [7.7±2 days (range, 5 to 13 days) versus 9.2±6.7 days (range, 5 to 31 days)] or the mean number of days with pleural fluid drainage [2.9±1.7 days (range, 1 to 6 days) versus 3.0±1.7 days (range, 1 to 8 days)]. However, the mean central venous pressure during the first 24 hours after surgery in group 1 was 14.1±2.1 mm Hg, significantly lower than the mean of 17.8±3.2 mm Hg in group 2 (P<.001, Fig 1⇓). Also, oxygen saturations at the time of discharge were significantly higher in group 2 (86±3.8%) compared with group 1 (81±2.7%) (P<.001, Fig 2⇓).
Five patients in group 2 were readmitted to the hospital 2 to 4 weeks after discharge for drainage of a large chylothorax (2 left-sided and 3 right-sided). This complication was not observed in any group 1 patient and was statistically different between groups (P<.02). Patient age ranged from 11 to 67 months (mean, 41.2±24.4 months). All of these patients required placement of central venous lines for prolonged total parenteral nutrition, and their hospital stays ranged from 6 to 55 days. There was one late death in group 2 (Table 2⇑, patient 2) secondary to sepsis while the patient was hospitalized for treatment of persistent chylous drainage.
The bidirectional cavopulmonary shunt has become an important staging procedure in the management of patients with complex single ventricle physiology. This retrospective review compared early outcome after bidirectional cavopulmonary shunt in patients with and without an additional source of pulmonary blood flow. Patients with an additional source of pulmonary blood flow had significantly elevated central venous pressures and increased oxygen saturations compared with those patients with the bidirectional cavopulmonary shunt as the sole source of pulmonary blood flow. In addition, patients with an additional source of pulmonary blood flow were at significantly higher risk for the late development of chylothoraxes. Operative mortality, hospital stay, and early chest tube drainage were not statistically different between the two groups. Patient age at operation was different between the two groups. This was of borderline statistical significance, however, and is not reflected in the subset of patients who developed chylothoraxes (three of the patients were less than 4 years of age and the oldest was 6.5 years of age at the time of their surgeries).
Few reports have addressed the issue of whether or not to leave an additional source of pulmonary blood flow at the time of bidirectional cavopulmonary shunt. The potential benefits of a bidirectional cavopulmonary shunt include improvement in effective pulmonary blood flow and a reduction in volume work of the single ventricle. Difficulties inherent in maintaining an additional source of pulmonary blood flow include adequate restriction to flow and pressure transmitted to the pulmonary vascular bed so that these benefits are not lost. If the additional source of pulmonary blood flow is not carefully controlled, continued pulmonary recirculation and ventricular dysfunction secondary to a chronic volume overload may result. Several studies have shown that having a chronically volume-overloaded single ventricle increases morbidity and mortality after the Fontan operation.9 10 In addition, it appears that the chronically dilated single ventricle is less likely to have a significant decrease in volume after a bidirectional cavopulmonary shunt.11
There are several theoretical advantages to maintaining an additional source of pulmonary blood flow after the bidirectional cavopulmonary shunt. As this study shows, the arterial oxygen saturation is increased and may improve long-term ventricular function by avoiding excessive hypoxemia.12 The maintenance of pulsatile pulmonary blood flow may help prevent the late complications that had been seen previously with the classic Glenn shunt, including abnormalities in regional pulmonary perfusion, the development of collaterals from the superior vena cava to the inferior vena cava, and the development of pulmonary arteriovenous malformations.13 14 15 Finally, the absence of pulsatile pulmonary blood flow may hinder pulmonary artery growth16 and may therefore influence survival after the Fontan operation.17
Elevated pulmonary artery pressures can complicate outcome after a bidirectional cavopulmonary shunt. In this study, patients with an additional source of pulmonary blood flow had significantly elevated central venous pressures. Although early chest tube drainage was not significantly different between groups, the development of chylothoraxes in the patients with an additional source of pulmonary blood flow may be related to the higher central venous pressures.18 The late development of chylothoraxes could be explained by improved oral (and fat) intake after hospital discharge. This led to significant morbidity and contributed to mortality in our series.
This study was limited by the lack of long-term follow-up. Further long-term studies are certainly needed to determine whether an additional source of pulmonary blood flow after a bidirectional cavopulmonary shunt has more deleterious effects on single ventricle function related to the chronic volume overload or more beneficial effects on function related to improved oxygen saturation. Also, since pulmonary arteriovenous malformations were a late complication of the standard Glenn shunt, long-term follow-up is needed to see whether the incidence of malformations decreases when the bidirectional cavopulmonary shunt is associated with an additional pulsatile source of pulmonary blood flow. Finally, long-term follow-up is needed to determine whether pulmonary artery growth is affected by an additional source of pulmonary blood flow after the bidirectional cavopulmonary shunt.
- Copyright © 1995 by American Heart Association
Castaneda AR. From Glenn to Fontan: a continuing evolution. Circulation. 1992;86(suppl II):II-80-II-84.
Bridges ND, Jonas RA, Mayer JE, Flanagan MF, Keane JF, Castaneda AR. Bidirectional cavopulmonary anastomosis as interim palliation for high-risk Fontan candidates. Circulation. 1990;82(suppl IV):IV-170-IV-176.
Kobayashi J, Matsuda H, Nakano S, Shimazaki Y, Ikawa S, Mitsuno M, Takahashi Y, Kawashima Y, Arisawa J, Matsushita T. Hemodynamic effects of bidirectional cavopulmonary shunt with pulsatile pulmonary flow. Circulation. 1991;84(suppl III):III-219-III-225.
Trusler GA, Williams WG, Cohen AJ, Rabinovitch M, Moes F, Smallhorn JF, Coles JG, Lightfoot NE, Freedom RM. The cavopulmonary shunt: evolution of a concept. Circulation. 1990;82(suppl IV):IV-131-IV-138.
Sluysmans T, Saunnders SP, van der Velde M, Matitiau A, Parness IA, Spevak PJ, Mayer JE, Colan SD. Natural history and patterns of recovery of contractile function in single left ventricle after Fontan operation. Circulation. 1992;86:1753-1761.
Graham TP, Franklin RCG, Wyse RKH, Gooch V, Deanfield JE. Left ventricular wall stress and contractile function in childhood: normal values and comparison of Fontan repair versus palliation only in patients with tricuspid atresia. Circulation. 1986;74(suppl I):I-61-I-69.
Cloutier A, Ash JM, Smallhorn JF, Williams WG, Trusler GA, Rowe RD, Rabinovitch M. Abnormal distribution of pulmonary blood flow after the Glenn shunt or Fontan procedure: risk of development of arteriovenous fistulae. Circulation. 1985;72:471-479.
McFaul RC, Tajik AJ, Mair DD, Danielson GK, Seward JB. Development of pulmonary arteriovenous shunt after superior vena cava-right pulmonary artery (Glenn) anastomosis: report of four cases. Circulation. 1977;55:212-216.
Szabo G, Magyar Z. Effect of increased systemic venous pressure on lymph pressure and flow. Am J Physiol. 1967;212:1469-1474.