Ann Thorac Surg 2004;78:1089-1090
© 2004 The Society of Thoracic Surgeons
Case report
Partial hepatic vein diversion in pulmonary arteriovenous malformations in congenital heart disease
Enrico Aidala, MDa,*,
Enrico Chiappa, MDb,
Maria Teresa Cascarano, MDa,
Andrea Valori, MDa,
Pietro Angelo Abbruzzese, MDa
a Department of Pediatric Cardiac Surgery, Ospedale Infantile "Regina Margherita," Torino, Italy
b Department of Pediatric Cardiology, Ospedale Infantile "Regina Margherita," Torino, Italy
Accepted for publication June 25, 2003.
* Address reprint requests to Dr Aidala, Via Canova 46, 10126 Torino, Italy
enri.ema{at}tiscalinet.it
 |
Abstract
|
|---|
The late appearance of pulmonary arteriovenous malformations after operations for complex congenital cardiac conditions may be explained in many ways. The necessity of hepatic blood flow to the lungs is now generally accepted for carrying some labile substance that is secreted by the liver and that has not yet been identified. Rerouting of hepatic blood to the lungs in cardiac patients can be accomplished with different methods, most of which are predisposed to thrombosis because of the slow flow in the tunnel. My colleagues and I describe a case of partial diversion of hepatic blood into the pulmonary circulation and suggest a strategy to decrease the likelihood of conduit thrombosis and to treat pulmonary arteriovenous malformations.
|
Introduction
|
|---|
Late development of pulmonary arteriovenous malformations (PAVM) with consequent severe hypoxia is a well-known complication of the Glenn operation and its modifications. Specifically, it is frequent after the Kawashima procedure, which is a unilateral or bilateral bidirectional modified Glenn operation performed in patients with azygous continuation of the inferior vena cava. Rerouting of hepatic flow to the lungs reverses this phenomenon and increases the oxygen saturation of the patient [1]. However, a high incidence of thrombosis of the rerouting conduit has been reported [2]. My colleagues and I describe a successful case of partial diversion of hepatic blood into the pulmonary circulation and suggest a strategy to decrease the likelihood of conduit thrombosis and to treat PAVM.
A cyanotic male neonate had left isomerism, dextrocardia with d-loop, double superior vena cava, azygous continuation of an interrupted inferior vena cava, single atrium, common atrioventricular valve, and ventriculoarterial discordance with pulmonary atresia. He underwent a right modified Blalock-Taussig shunt and, subsequently, at the age of 12 months, a bilateral bidirectional Glenn operation. He did well for more than 1 year, with oxygen saturations approaching 90%, and then an increasing cyanosis appeared, reaching a saturation of 65% over the next 3 years. Development of PAVM was therefore hypothesized and confirmed by contrast transthoracic echocardiography, which showed the rapid (< 4 beats) appearance of microbubbles in the left atrium. At the age of 4 years, the child underwent an operation to reroute hepatic blood to the pulmonary circulation. With the patient under deep hypothermia (16°C) and circulatory arrest, we opened the left side of the single atrium and found the polysplenia arrangement of the hepatic veins, with a single large right hepatic vein draining into the right side of the atrium while the median and the left hepatic veins drained together, distant from the right one, into the left side of the atrium. A very large tube would have been needed to reroute all veins with an in situ technique. We therefore elected to leave the right hepatic vein draining into the atrium while the button of the atrial wall surrounding the left-sided veins was detached from the atrium and anastomosed to one end of a 12-mm polytetrafluoroethylene (PTFE) tube, which was connected on the other side to the left pulmonary artery. The atrium was closed, and extracorporeal circulation was resumed to rewarm the patient. Before the end of the operation, a narrow 20-mm-long strip of PTFE was placed around the right-sided hepatic vein at its entrance into the atrium, and it was anchored to the diaphragm to prevent dilation of the vein and excessive hepatic blood flow to the atrium. At the end of the procedure, the mean atrial pressure was 8 mm Hg, and the mean pulmonary artery pressure was 13 mm Hg. Perioperative transesophageal echocardiography showed phasic slow flow through the conduit and unobstructed flow through the vein draining into the atrium. After the operation, a chylothorax was treated medically, and the patient was started on oral anticoagulation; the international normalized ratio was maintained between 2.5 and 3. At discharge, the oxygen saturation was 70%. One year later, the oxygen saturation reached 85%. Echocardiography continued to show phasic respiratory variation of a slow flow through the conduit and a nonaccelerated flow in the right hepatic vein; the diameter of this vein remained unchanged.
|
Comment
|
|---|
Many factors have been advocated to explain the late appearance of PAVM after certain operations for complex congenital cardiac conditions in children [3]. It is now generally accepted that a labile substance, secreted by the liver and not yet identified, must enter the pulmonary circulation in an undefined amount to prevent opening of the anatomically present precapillary pulmonary shunts. This explains why some Glenn and the Kawashima operations, which exclude hepatic blood from the pulmonary arterial tree, may cause such a problem. Furthermore, PAVM is a recognized complication of severe liver disease (hepatopulmonary syndrome), which might prevent the synthesis of that labile substance. Accordingly, surgical diversion of hepatic blood to the lungs in cardiac patients [1] and liver transplantation in patients with cirrhosis [3] reverse this phenomenon.
The diagnosis of PAVM in inexplicably cyanotic patients is not always simple: recently, contrast (shaken saline solution) transthoracic echocardiography has proven to be a more sensitive method than pulmonary angiography, which may not clearly detect diffuse small fistulas [4]. Rerouting of hepatic blood to the lungs can be accomplished in cardiac patients with different methods, including intracardiac tunnels, hepatic vein–to–azygos vein connections, or extracardiac conduits. With the exception of a direct hepatic vein–to–azygos vein connection, which has been reported twice [5, 6] and which is not always technically feasible, all these techniques have the disadvantage of requiring large conduits that are predisposed to thrombosis because of the slow venous flow from the liver. In fact, we know from physiology that hepatic flow accounts for 20% to 25% of total venous return and that its velocity is relatively low and inversely related to the diameter of the conduit. The consequences of this fact are that we should use as small a conduit as possible and that we have to manage a higher thrombotic risk than in normal extracardiac Fontan operations.
A recent report [2] focused on thrombosis of pericardial and PTFE conduits used for hepatic flow diversion. Their 3 patients had been treated with large conduits and no anticoagulation: all conduits thrombosed, and 1 patient died.
In our patient, the left isomerism altered the hepatic venous drainage into the atrium so that a large right-sided vein draining the right lobe entered the atrium distant from the other 2. We used a 12-mm conduit to divert the blood of these 2 veins to the lungs. The relatively small diameter of the conduit should increase the velocity of the hepatic venous blood, thus reducing thrombotic risk while maintaining a flow adequate for the disappearance of PAVM.
We elected to leave the solitary right-sided vein draining into the atrium, but we stabilized its natural diameter (approximately 7 mm) with an external PTFE fixation. This band did not modify the size or geometry of the vein, thus avoiding the risks of stenosis, distortion, and, consequently, thrombosis of such a low-flow vessel.
This right-sided vein should maintain liver function if the conduit thromboses; moreover, the band will prevent dilation of the vein because of the preferential flow toward the lower-pressure atrial chamber. Such a "steal phenomenon" would preclude the regression of PAVM and would increase the risk of thrombosis of the hepatic conduit because of the reduced flow to the lungs. In an effort to further decrease the risk of thrombosis, well emphasized in a recent review [2], we kept the patient on a relatively high level of oral anticoagulation, even though a universally accepted protocol is not yet defined.
In conclusion, an adequate but not yet defined amount of hepatic blood seems to be needed in the pulmonary circulation to prevent the development of PAVM or to reverse this phenomenon once it has developed. A small conduit, together with effective anticoagulation, may be a good option to decrease the tendency of slow hepatic flow toward a thrombotic occlusion. Leaving part of the hepatic venous flow draining into the atrium may be advisable as a safety valve, provided that some type of stabilization of the vein's diameter is performed to avoid preferential flow caused by pressure gradients. Doubts remain about the use of a small conduit in a growing child and about the need for a higher anticoagulation level; 14 months after the operation, neither clinical nor echocardiographic problems were evident.
|
References
|
|---|
- Shah MJ, Rychic J, Fogel MA, Murphy JO, Jacobs ML. Pulmonary AV malformations after superior cavopulmonary connections: resolution after inclusion of hepatic veins in the pulmonary circulation. Ann Thorac Surg. 1997;63:960–963[Abstract/Free Full Text]
- Kostantinov IE, Puga FJ, Alexi-Meskishvili VV. Thrombosis of intracardiac or extracardiac conduits after modified Fontan operation in patients with azygous continuation of the inferior vena cava. Ann Thorac Surg. 2001;72:1641–1644[Abstract/Free Full Text]
- Srivastava D, Preminger T, Lock JE, et al. Hepatic venous blood and the development of pulmonary arteriovenous malformations in congenital heart disease. Circulation. 1995;92:1217–1222[Abstract/Free Full Text]
- Bernstein HS, Brook MM, Silverman NH, Bristow J. Development of pulmonary arteriovenous fistulae in children after cavopulmonary shunt. Circulation. 1995;92(Suppl 2):II309–314
- Baskett RJF, Ross DB, Warren AE, Sharratt GP, Murphy DA. Hepatic vein to the azygos vein anastomosis for pulmonary arteriovenous fistulae. Ann Thorac Surg. 1999;68:232–233[Abstract/Free Full Text]
- Kaneko Y, Murakami A, Miyamoto T, Takamoto S. Hepatic vein-to-azygous vein connection in a patient with interrupted inferior vena cava. Eur J Cardiothorac Surg. 2002;21:582–584[Abstract/Free Full Text]
Top
Abstract
Introduction
