HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Uemura, H. Articles by Watanabe, K. Search for Related Content PubMed PubMed Citation Articles by Uemura, H. Articles by Watanabe, K.

Ann Thorac Surg 1999;68:1731-1735
© 1999 The Society of Thoracic Surgeons

---

Original Articles

Redirection of hepatic venous drainage after total cavopulmonary shunt in left isomerism

^a Department of Cardiovascular Surgery and Pediatrics, National Cardiovascular Center, Osaka, Japan

Address reprint requests to Dr Uemura, Department of Cardiovascular Surgery, National Cardiovascular Center, 5-7-1 Fujishirodai, Suita, 565-8565 Osaka, Japan

	Abstract

Top Abstract Introduction Patients and methods Results Comment References

 
Background. Conversion from total cavopulmonary shunt (TCPS) to the Fontan circulation can improve cyanosis in patients with potential risks of development of pulmonary arteriovenous fistula (PAVF).

Methods. Inclusion of the hepatic veins in the pulmonary circulation was employed using an intra-atrial tube graft in 5 patients with left isomerism previously undergoing TCPS. Prior to the conversion, abnormal communication was identified between the azygos vein and either the hepatic or the portal vein in all. PAVF was seen in 3.

Results. All patients survived the procedure. Postoperative catheterization showed 13 ± 2 mm Hg of superior caval venous pressure, and 2.3 ± 0.4 L/min/m² of cardiac index. Pulmonary arteriovenous fistula progressed markedly in the right lung even after the conversion in 2 patients, in whom the hepatic veins had been exclusively diverted to the left lung. Arterial oxygen saturation became below 65%, with exercise capacity reduced, in these 2 patients. The other patients remain asymptomatic.

Conclusions. Total cavopulmonary shunt can be efficiently converted to the Fontan circulation by appropriately redirecting hepatic venous drainage to perfuse both lungs in a balanced fashion.

	Introduction

Top Abstract Introduction Patients and methods Results Comment References

 
The so-called total cavopulmonary shunt (TCPS) by means of bidirectional cavopulmonary anastomosis [1] is a surgical option of choice in patients with interruption of the inferior caval vein (IVC), particularly when total bypass of the right heart is initially considered unsuitable [2–4]. We chose this surgical intervention as a definitive procedure until 1989. Residual cyanosis, however, can frequently progress after TCPS because of, in part, development of pulmonary arteriovenous fistula (PAVF) [5]. Development of collateral channels from the systemic veins to the hepatic venous system could be another cause of progressive cyanosis [6]. To avoid development, or to promote regression, of these hazardous communications [7, 8], we have decided to proceed to further operation for redirection of hepatic venous drainage to the pulmonary arteries (PA). Such conversion was herein described in our 5 patients.

	Patients and methods

Top Abstract Introduction Patients and methods Results Comment References

 
Total cavopulmonary shunt (TCPS) was employed in 11 patients with isomeric left appendages and interrupted IVC at our institution. One patient died, because of gastric bleeding, 1 month after the procedure. Another 3 patients have died in the longer term, 33, 43, and 115 months after the procedure, because of arrhythmia in 1 and intractably developed PAVF in 2. In another 5 patients, the hepatic veins were connected to PA. Such conversion from partial to total bypass of the right heart has not been considered feasible in the remaining 2 patients, because of severely impaired ventricular performance in 1, and presence of significantly abundant systemic-to-pulmonary collaterals [9] in the other.

The 5 patients undergoing the conversion possessed a dominant morphologically right ventricle (RV) with a hypoplastic left ventricle (LV). Atrioventricular (AV) connection was absent for LV in 2 patients, markedly deviated towards RV because of straddling of the morphologically mitral valve in 1, and double inlet to RV guarded by a common AV valve in 2. Ventriculoarterial connections were of double outlet RV in all patients, with pulmonary stenosis in 4. Prior to TCPS, the pulmonary trunk had been banded in 1 and a systemic-to-pulmonary shunt had been constructed in another 3. TCPS was carried out at 3 to 12 years-of-age, with a mean of 6.9 ± 3.7 years. At the procedure, forward flow from the ventricles to PA [10, 11] was maintained via the pulmonary trunk in 3 patients and via the previously constructed shunt in 1, to avoid unfavorable regression of PA size [12]. In 2 of these 4 patients, either pulmonary valvotomy or reconstruction of the RV outflow tract was additionally carried out 26 and 77 months after TCPS. Catheterization prior to establishment of the Fontan circulation showed 8–17 (11 ± 4) mm Hg of mean PA pressure, 0.56–1.03 (0.69 ± 0.19) of PA area index [9, 10, 12], 268–503 (339 ± 94) of Nakata index [13], 0.56–1.07 (0.83 ± 0.20) of estimated Qp/Qs value, and 0.50–0.61 (0.56 ± 0.05) and 7–13 (9 ± 3) mm Hg of ejection fraction and end diastolic pressure of the dominant RV, respectively.

Age at redirection of hepatic venous drainage to PA was 8–19 (12.2 ± 4.2) years old. The interval between initial TCPS and the conversion was 35–80 (64 ± 19) months. A Gore-Tex (W.L. Gore and Assoc, Flagstaff, AZ) tube graft was intra-atrially used [9, 14] for redirection of the hepatic vein, interposing between the hepatic venous orifices and PA. The size of the prosthesis used was 16 mm in 2, 18 mm in 1, and 20 mm in 2. The cardiac veins were not redirected to PA to avoid influence of elevated venous pressure on myocardial perfusion [15]. The prosthetic tube was fenestrated, placing a 4-mm-diameter circular hole, in 2 patients in whom pulmonary resistance was estimated as greater than 3 units · m² or mean PA pressure was higher than 16 mm Hg preoperatively. In one of these, moderate subaortic stenosis was concomitantly treated by myectomy. The other patient underwent plasty to the solitary AV valve. The additional channel for forward flow previously provided was closed. Orientation of diversion of the hepatic venous blood was shown (Fig 1) in relation to the bidirectional cavopulmonary anastomosis previously constructed as well as the connection of the azygos vein draining IVC.

View larger version (31K): [in this window] [in a new window]   Fig 1. Patterns of systemic venous connections and orientation of the hepatic venous channel in relation to the cavopulmonary anastomosis.

	Results

Top Abstract Introduction Patients and methods Results Comment References

Changes in arterial oxygen saturation were shown in Figure 2. Desaturation before establishment of the Fontan circulation was partly because of presence of abnormal communications between the systemic veins and either the hepatic or the portal vein (Fig 3), found in all the 5 patients (Table 1). The channel between the azygos and the hepatic veins remained patent after the conversion in patient 1. Similar collaterals developed in the other 3, however, could not be unequivocally detected after the conversion. The communication between the azygos and the portal veins, seen before establishment of the Fontan circulation in 3, could not be clearly identified after the conversion, either. In patient 2, however, abnormal communication was found between the hepatic vein and the atrial chamber, producing right-to-left shunt and arterial desaturation (Fig 4). Another obvious cause of desaturation was formation of PAVF seen in 4 patients (Table 1). PAVF was unequivocally detected, by contrast echocardiography as well as PA angiography, subsequent to TCPS and prior to establishment of the Fontan circulation in 3, and, in one of these, the fistulous communication could be successfully occluded by transcatheter embolization using coils. PAVF developed markedly even after the conversion in 2 patients. In these 2, the hepatic venous effluent directed exclusively to the left lung (Fig 5A), and the deleterious PAVF progressed within the contra-lateral right lung. Probably, this is because of orientation of the cavopulmonary anastomoses or the presence of stenosis at the central PA (Fig 1). In the other 3 patients, the hepatic venous effluent proved to be delivered to the right and the left lungs in a balanced fashion (Fig 5B), and PAVF has not significantly progressed thus far.

View larger version (29K): [in this window] [in a new window]   Fig 2. Changes in arterial oxygen saturation. Forward flow from the ventricle to the pulmonary arteries was added subsequent to total cavopulmonary shunt (*) by reconstruction of the right ventricular outflow tract in patient 3 and by valvotomy in patient 5. (TCPS = total cavopulmonary shunt.)

View larger version (153K): [in this window] [in a new window]   Fig 3. Abnormal venous-to-venous communications. (A) A dilated collateral between the azygos and the hepatic veins seen in patient 3. (B) A tortuous channel between the azygos and the portal vein found in patient 2. (C) In patient 1, communication was present between the azygos, the hepatic, and the portal veins.

View larger version (96K): [in this window] [in a new window]   Fig 4. Residual right-to-left shunt via the collaterals between the hepatic vein and the atrial chamber seen in patient 2 after conversion to the Fontan circulation.

View larger version (128K): [in this window] [in a new window]   Fig 5. Patterns in pulmonary perfusion. (A) Angiography carried out 19 months after the conversion in patient 1. Hepatic venous drainage had been diverted exclusively to the left lung. (B) In patient 5, hepatic venous drainage had been diverted to the right and the left lungs in a balanced fashion. The right superior caval vein drained mainly to the right lung, and the left superior caval vein exclusively to the left lung. In this patient, pulmonary arteriovenous fistula had been successfully embolized. No further development of such fistula was detected.

	Comment

Top Abstract Introduction Patients and methods Results Comment References

 
Pulmonary arteriovenous fistula has been well documented in patients with interrupted IVC, and several factors have been speculated in terms of causes of the deleterious communication. Recent findings [16, 17] may support the opinion that presence of isomeric left appendages, or else polysplenia, and exclusion of the hepatic venous blood from the pulmonary circulation are two of the major risk factors [5]. Inclusion of the hepatic venous effluent to the pulmonary perfusion, indeed, promoted regression of PAVF in some patients, this circulatory modification being provided by achieving total right heart bypass procedure [7, 8]. Moore and associates [18], in contrast, reported development of PAVF after the Fontan procedure. In their first patient, the hepatic veins were diverted exclusively to the left PA, and PAVF occurred in the right lung. In their second patient, the hepatic veins were left unrerouted and remained as an intracardiac right-to-left shunt. In this particular patient, therefore, hepatic venous drainage was excluded from the pulmonary circulation just like TCPS, and PAVF occurred in both lungs.

In our present study, development of PAVF, after establishment of the Fontan circulation, was intimately associated with biased perfusion of hepatic venous drainage to the left lung. In this respect, the angiographic figures seen in the articles by Knight and Mee [7], as well as by Shah and associates [8], beautifully demonstrated bilateral and balanced pulmonary perfusion of hepatic venous drainage. It is therefore likely essential to appropriately design the hepatic venous channel in relation to the anastomosis between SVC and PA. Proper arrangements can be readily achieved in patients having bilaterally SVCs unless stenosis is present at the central PA. In patients with a solitary SVC, in contrast, balanced perfusion of hepatic venous blood to the bilateral lungs is not an easy matter to achieve. This is because the surgeon, when connecting the hepatic venous channel to PA, usually constructs the anastomosis away from the cavopulmonary anastomosis for SVC, to minimize possible hemodynamic disadvantage associated with unfavorable turbulence of blood flow. To drain the hepatic veins bilaterally to the lungs, a Y-shaped branching conduit might have been needed in patients with a solitary SVC.

As for the options for redirection of hepatic venous drainage, not only intra-atrial grafting but also construction of an extracardiac channel would be of attractive use. In patients with visceral heterotaxy, venoatrial connections are commonly abnormal, and use of the extracardiac Fontan procedure can be less complicated in such settings [19]. Presence of the independent hepatic veins are, nonetheless, not rare in patients with isomeric left appendages [20]. In patients with dual hepatic veins on both sides of the vertebrae, an extracardiac channel might be difficult to construct without obstruction or distortion. Taking orientation of all the connections of the pulmonary, the caval, and the hepatic veins to the atriums into account, the optimal option should be chosen for redirection of hepatic venous drainage in patients with visceral heterotaxy.

Progressive cyanosis after TCPS can be caused not only by development of PAVF but also by formation of abnormal communications between the azygos vein and the portal or hepatic vein. Surprisingly, such communications were identified in all 5 patients on precise examinations by angiography and scintigraphy. Similar venous-to-venous collaterals may develop also in those without interruption of IVC and undergoing the bidirectional Glenn procedure [6], particularly when the azygos vein is not ligated. Since the azygos vein cannot be occluded in patients with interrupted IVC, venous collaterals via either the hepatic or the portal vein can develop, more or less and sooner or later, in the majority of patients in this particular circumstance.

In addition to progressive cyanosis, regression of PA size, which implies underdevelopment of PA vasculature, can become remarkable after formation of the collaterals from the systemic to the hepatic venous system, because the amount of pulmonary blood flow may be reduced by the shunt. This is obviously an unfavorable factor when aiming towards successful establishment of the Fontan circulation. To avoid excessive regression of PA size after partial bypass of the right heart, maintenance of additional forward flow from the ventricles to PA can be of optional use. For that purpose, amount of flow through such additional channel would be sufficient when cross-sectional area of the channel indexed by body surface area is greater than 25 x 10^-6 [12]. Our preference is use of forward flow through the pulmonary trunk, if available, because size of the channel can be readily adjusted by placement of banding. Additional forward flow can also allow part of the hepatic venous effluent to join the PA perfusate. In 4 patients with forward flow in this study, PA size remained suitable for the Fontan procedure. Although minor or localized PAVF progressed in 3 of these, diffuse and intractably severe PAVF was not seen. The surgeon, in contrast, should mind the disadvantageous aspects of the additional forward flow. We would severely restrict such channel in patients with significant atrioventricular valvar regurgitation and end diastolic volume of the dominant systemic ventricle greater than 240% of the anticipated normal value, to off-load the ventricle as much as possible. Another deleterious aspect of the option may be potentially higher PA pressure, which could accelerate development of venous-to-venous collaterals.

It is fortunate that the venous-to-venous collaterals have clinically posed no significant problems subsequent to conversion to the Fontan circulation thus far. The channel between the azygos and the hepatic veins, although proven clearly patent in one of our patients, would hopefully have no fundamental disadvantages with the Fontan circulation. It has yet to be determined, however, whether or not the communication between the systemic and the portal veins, once developed, has no impact on the bodily physiology. Basically, pressure of the portal vein must be higher than that of the hepatic vein. Converting to the Fontan circulation, hepatic venous pressure becomes as high as other systemic venous pressures. The abnormally developed channel between the portal and the systemic veins, therefore, can remain unclosed. Through such a shunt, portal venous blood can appear in PA without perfusing the hepatic parenchyma. This is the exact analogue of the Eck shunt procedure, which used to be employed in patients with portal hypertension, and functionally similar to the circulation in patients with liver cirrhosis. Accordingly, those patients undergoing TCPS and possessing abnormal venous communications between the portal venous system and the systemic veins must be carefully followed up, even after the conversion, in terms of possible progress of hepatic dysfunction and PAVF.

It remains controversial, and is difficult to determine, whether conversion from TCPS to the Fontan circulation is advantageous or not. In terms of progressive cyanosis and potential risk of PAVF, such surgical procedure is to be efficient if the hepatic venous channel is well designed to perfuse both the right and the left lungs. To avoid presence of different systemic and hepatic venous pressures within a body, and to avoid exclusion of hepatic venous blood from the pulmonary circulation at any stage of life, our current preference is establishment of the Fontan circulation without previous TCPS unless the surgical approach is deemed to be contraindicated.

	References

Top Abstract Introduction Patients and methods Results Comment References

 

1. Kawashima Y., Kitamura S., Matsuda H., Shimazaki Y., Nakano S., Hirose H. Total cavopulmonary shunt operation in complex cardiac anomalies. J Thorac Cardiovasc Surg 1984;87:74-81.[Abstract]
2. Hopkins R.A., Armstrong B.E., Serwer G.A., Peterson R.J., Oldham H.N., Jr Physiological rationale for a bidirectional cavopulmonary shunt. J Thorac Cardiovasc Surg 1985;90:391-398.[Abstract]
3. Mazzera E., Corno A., Picardo S., et al. Bidirectional cavopulmonary shunts. Ann Thorac Surg 1989;47:415-420.[Abstract]
4. Bridges N.D., Jonas R.A., Mayer J.E., Flanagan M.F., Keane J.F., Castañeda A.R. Bidirectional cavopulmonary anastomosis as interim palliation for high-risk Fontan candidates. Circulation 1990;82(suppl IV):IV170-IV176.
5. Kawashima Y. Cavopulmonary shunt and pulmonary arteriovenous malformations. Ann Thorac Surg 1997;63:930-932.[Free Full Text]
6. Gatzoulis M.A., Shinebourne E.A., Redington A.N., Rigby M.L., Ho S.Y., Shore D.F. Increasing cyanosis early after cavopulmonary connection caused by abnormal systemic venous channels. Br Heart J 1995;73:182-186.[Abstract/Free Full Text]
7. Knight W.B., Mee R.B.B. A cure for pulmonary arteriovenous fistulas?. Ann Thorac Surg 1995;59:999-1001.[Abstract/Free Full Text]
8. Shah M.J., Rychik J., Fogel M.A., Murphy J.D., Jacobs M.L. Pulmonary arteriovenous malformations after superior cavopulmonary connection. Ann Thorac Surg 1997;63:960-963.[Abstract/Free Full Text]
9. Ichikawa H., Yagihara T., Kishimoto H., et al. Extent of aortopulmonary collateral blood flow as a risk factor for Fontan operations. Ann Thorac Surg 1995;59:433-437.[Abstract/Free Full Text]
10. Kobayashi J., Matsuda H., Nakano S., et al. Hemodynamic effects of bidirectional cavopulmonary shunt with pulsatile pulmonary flow. Circulation 1991;84(suppl III):III219-III225.
11. Albanese S.B., Carotti A., Di Donato R.M., et al. Bidirectional cavopulmonary anastomosis in patients under two years of age. J Thorac Cardiovasc Surg 1992;104:904-909.[Abstract]
12. Uemura H., Yagihara T., Kawashima Y., Okada K., Kamiya T., Anderson R.H. Use of the bidirectional Glenn procedure in the presence of forward flow from the ventricles to the pulmonary arteries. Circulation 1995;92(suppl II):II228-II232.
13. Nakata S., Imai Y., Takanashi Y., et al. A new method for the quantitative standardization of cross-sectional areas of the pulmonary arteries in congenital heart disease with decreased pulmonary blood flow. J Thorac Cardiovasc Surg 1984;88:610-619.[Abstract]
14. Uemura H., Yagihara T., Kawashima Y., et al. What factors affect ventricular performance after a Fontan-type operation?. J Thorac Cardiovasc Surg 1995;110:405-415.[Abstract/Free Full Text]
15. Uemura H., Ho S.Y., Anderson R.H., et al. The surgical anatomy of coronary venous return in hearts with isomeric atrial appendages. J Thorac Cardiovasc Surg 1995;110:436-444.[Abstract/Free Full Text]
16. Srivastava D., Preminger T., Lock J.E., 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]
17. Laberge J.M., Braudt M.L., Lebecque P., et al. Reversal of cirrhosis-related pulmonary shunting in two children by orthotopic liver transplantation. Transplantation 1992;53:1135-1138.[Medline]
18. Moore J.W., Kirby W.C., Madden W.A., Gaither N.S. Development of pulmonary arteriovenous malformations after modified Fontan operations. J Thorac Cardiovasc Surg 1989;98:1045-1050.[Abstract]
19. Uemura H. The Fontan type procedure in patients with visceral heterotaxy. Cardiol Young 1998;8:419-422.[Medline]
20. Uemura H., Ho S.Y., Devine W.A., Kilpatrick L.L., Anderson R.H. Atrial appendages and venoatrial connections in hearts from patients with visceral heterotaxy. Ann Thorac Surg 1995;60:561-569.[Abstract/Free Full Text]

This article has been cited by other articles:

				Y. Nakamura, T. Yagihara, K. Kagisaki, I. Hagino, and J. Kobayashi Pulmonary arteriovenous malformations after a Fontan operation in the left isomerism and absent inferior vena cava Eur. J. Cardiothorac. Surg., July 1, 2009; 36(1): 69 - 76. [Abstract] [Full Text] [PDF]

				Y. Koteda, K. Suda, S. Kishimoto, and M. Iemura Portal-systemic encephalopathy after Fontan-type operation in patient with polysplenia syndrome Eur. J. Cardiothorac. Surg., June 1, 2009; 35(6): 1083 - 1085. [Abstract] [Full Text] [PDF]

				S.-J. Kim, E.-J. Bae, J.-Y. Lee, H.-G. Lim, C. Lee, and C.-H. Lee Inclusion of hepatic venous drainage in patients with pulmonary arteriovenous fistulas. Ann. Thorac. Surg., February 1, 2009; 87(2): 548 - 553. [Abstract] [Full Text] [PDF]

				J. W. Brown, M. Ruzmetov, P. Vijay, M. D. Rodefeld, and M. W. Turrentine Pulmonary Arteriovenous Malformations in Children After the Kawashima Operation Ann. Thorac. Surg., November 1, 2005; 80(5): 1592 - 1596. [Abstract] [Full Text] [PDF]

				D. B. McElhinney, J. Kreutzer, P. Lang, J. E. Mayer Jr, P. J. del Nido, and J. E. Lock Incorporation of the Hepatic Veins Into the Cavopulmonary Circulation in Patients With Heterotaxy and Pulmonary Arteriovenous Malformations After a Kawashima Procedure Ann. Thorac. Surg., November 1, 2005; 80(5): 1597 - 1603. [Abstract] [Full Text] [PDF]

				H. Ichikawa, N. Fukushima, M. Ono, T. Kita, T. Matsushita, Y. Miyamoto, and H. Matsuda Resolution of pulmonary arteriovenous fistula by redirection of hepatic venous blood Ann. Thorac. Surg., May 1, 2004; 77(5): 1825 - 1827. [Abstract] [Full Text] [PDF]

				B. W. Duncan and S. Desai Pulmonary arteriovenous malformations after cavopulmonary anastomosis Ann. Thorac. Surg., November 1, 2003; 76(5): 1759 - 1766. [Abstract] [Full Text] [PDF]

				J. Steinberg, G. M. Alfieris, B. Brandt III, F. Smith, C. J. Byrum, G. W. Fink, and J. Halter New approach to the surgical management of pulmonary arteriovenous malformations after cavopulmonary anastomosis Ann. Thorac. Surg., May 1, 2003; 75(5): 1640 - 1642. [Abstract] [Full Text] [PDF]

				Y. Kaneko, A. Murakami, T. Miyamoto, and S. Takamoto Hepatic vein-to-azygos vein connection in a patient with interrupted inferior vena cava Eur. J. Cardiothorac. Surg., March 1, 2002; 21(3): 582 - 584. [Abstract] [Full Text] [PDF]

				I. E. Konstantinov, F. J. Puga, and V. V. Alexi-Meskishvili Thrombosis of intracardiac or extracardiac conduits after modified Fontan operation in patients with azygous continuation of the inferior vena cava Ann. Thorac. Surg., November 1, 2001; 72(5): 1641 - 1644. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Uemura, H. Articles by Watanabe, K. Search for Related Content PubMed PubMed Citation Articles by Uemura, H. Articles by Watanabe, K.