|
|
||||||||
J Thorac Cardiovasc Surg 1995;110:916-923
© 1995 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
Rondebosch, Republic of South Africa
From the Department of Pediatric Surgery, Red Cross Children's Hospital, Rondebosch, Republic of South Africa.
Received for publication April 20, 1994. Accepted for publication Feb. 6, 1995. Address for reprints: H. Rode, MMed(Surg), FRCS(Ed), FCS(SA), Department of Pediatric Surgery, Red Cross Children's Hospital, Rondebosch 7700, Republic of South Africa.
Abstract
Membranous obstruction of the inferior vena cava at the level of the diaphragm is a rare cause of Budd-Chiari syndrome in children. Medical therapy usually fails. Surgical intervention aims at reestablishing patency of the inferior vena cava and hepatic venous outflow to the right atrium. We report on the management of this condition in 19 children of whom 7 were treated surgically. Indications for operation were persistent ascites, deteriorating liver function, and hepatic and inferior vena caval obstruction without significant collateral circulation. Three pathologic types were identified by ultrasonography and cavography and were confirmed at operation. These were type I (4 cases), with a thin membrane occluding the inferior vena cava at the level of the diaphragm; type II (12 cases), with segmental fibrotic obstruction of the inferior vena cava with variable involvement of hepatic veins; and type III (3 cases), with complete absence or nonvisualization of the inferior vena cava. All procedures were done with an extended midline sternotomy incision, cardiopulmonary bypass, core cooling to 16° to 20°C, and periods of circulatory arrest. Type I lesions necessitated membranectomy; type II lesions necessitated transcaval resection of the occluded confluence of the inferior vena cava and the hepatic vein with repair of the defect with an autogenous pericardial patch. One type II lesion, in addition, called for use of a 14 cm polytetrafluoroethylene tube graft to restore inferior vena caval flow. After the operation, marked clinical improvement was observed with an immediate reduction in liver and spleen size and resolution of ascites. Repeat cavography 10 to 30 days after the operation revealed complete patency in four cases and residual stenosis, which required transiliac balloon angioplasty to normalize the inferior vena cava/right atrial pressure gradient, in 3 cases. Thus eventual relief of hepatic venous outflow obstruction and inferior vena caval flow was restored in all cases. We advocate transcardiac membranectomy and pericardial patch grafting for symptomatic and deteriorating membranous obstruction of the inferior vena cava in children. (J THORACCARDIOVASCSURG1995;110:916-23)
Budd-Chiari syndrome is a loosely defined term that may be applied to any clinical entity in which any one of a variety of pathologic processes, but excluding congestive heart failure, causes hepatic venous outflow obstruction.
1 An uncommon cause for the Budd-Chiari syndrome in children is a membranous obstruction of the inferior vena cava (IVC) with variable involvement of the hepatic venous outflow tracts. The clinical picture is determined by the rate of onset, development of collateral circulation, and liver disease. If hepatic venous outflow obstruction persists, liver cirrhosis and the subsequent development of hepatocellular carcinoma may ensue.
2-4 In adults the mortality rates are 30% and 60% at 1 and 2 years, respectively.
5
Although the lesion of membranous obstruction of the IVC is thought to be congenital in origin,
6-8 it has rarely been described in children. Hoffman, Stockland, and Heyden,
9 in a 1987 report of nine cases in children, could find reference to only eight other cases in the literature.
3,4,8,10-12 A large series by Gentil-Kocher and colleagues
13 of 22 cases included patients with sole hepatic venous obstruction and patients with membranous obstruction of the IVC. Further reports of two other children who underwent surgical repair are described.
14-16
Spontaneous resolution is rare and medical therapy results in only limited and temporary amelioration of symptoms.
13 The aim of surgery is to reestablish hepatic venous outflow to the caval system and right atrium, but the best method of attaining this goal remains a topic of controversy.
17
We report the cases of 19 children, of whom seven were treated surgically with what we believe to be the most appropriate form of management on the basis of the clinical presentation, severity of symptoms, and findings at operation.
PATIENTS AND METHODS
Patients.
Nineteen children with membranous obstruction of the IVC were seen by us during a 9-year period (1984 to 1993). All except one child came from the same geographic area in Namibia.
18 There was a male preponderence of 14 children. The children's ages ranged from 1 to 11 years and 16 children were younger than 5 years old.
The clinical picture was dominated by hepatomegaly (19/19), splenomegaly (8/19), ascites (6/19), and peripheral edema (2/19). Hematologic findings were unremarkable but transaminase values were slightly elevated in all. Hepatitis serologic testing gave negative results. An enlarged azygos vein was observed on the chest radiograph of seven children. Ultrasonography of the hepatic veins and IVC accurately identified the site and extent of obstruction in all of the most recent patients. Limited cavography was done in the first six cases, and cavography was extended to include simultaneous injection of contrast medium through catheters positioned in the right atrium and upper IVC in the remainder. Results obtained with ultrasonography and cavography defined three types of membranous obstruction of the IVC. Type I was defined as a thin membrane and was present in four of the 19 patients; type II, a segmental short obstruction with involvement of the hepatic veins, was present in 12 patients; and type III was defined as absent IVC or nonvisualization of the IVC and was present in three patients.
Seven of this group of 19 children aged 4.3 ± 3.5 years were treated surgically. Indications for operation were progressive symptoms, deteriorating liver function, and hepatic and IVC outflow obstruction without significant collateral circulation.
Table I depicts the pathologic abnormalities as seen on ultrasonography and cavography and subsequently confirmed surgically.
|
Surgical procedure
In all patients a midline sternotomy incision was made and extended to approximately midway between the xiphisternum and umbilicus. In cases in which the obliterative process was more extensive the incision was extended inferiorly and obliquely to the right as a Kocher incision. Large venous collateral and engorged lymphatic vessels were usually present in the falciform ligament.
All procedures were done with the use of cardiopulmonary bypass and core cooling to 16° to 20°C with periods of circulatory arrest (Fig. 1). In patients with membranous obstruction of the IVC, venous drainage to the cardiopulmonary bypass machine was established by a single venous cannula placed directly into the right atrium and the patient was moderately cooled. In those with more complex obstruction an additional venous cannula was placed in the infrahepatic IVC and the patient cooled to 18°C. Exposure was facilitated by duodenal kocherization. The exposed IVC was invariably tense and thick walled. A purse-string suture was inserted and a cannula advanced cephalad until the site of obstruction was encountered. Palpation and visualization of this cannula identified the obstruction and the same cannula provided venous drainage from the lower half of the body.
|
The often dense fibrous membranous areas were excised, exposing the hepatic veins. With more extensive involvement of the IVC the occluded liver-caval junction was excised (cored out) toward the palpable retrohepatic IVC cannula, exposing as many hepatic veins as possible in the process. Usually this extended approximately 2 cm deep and 3 cm in diameter into the IVChepatic vein confluence. The resultant defect was roofed by autogenous pericardium sutured to the capsule of the liver, the fibrous edges of the IVC, and the right atrium. Biopsy specimens were taken from the membranes and liver through the opened IVC. In one patient with more extensive obliteration of the retrohepatic IVC, in addition to coring out the liver-caval junction, a 14 cm polytetrafluoroethylene (PTFE) tube graft*
was sutured to the retrohepatic IVC and anastomosed to the right atrium. This graft was positioned posterior to the liver.
RESULTS
All children survived the surgical procedure and were discharged from the hospital. The postoperative course was uncomplicated in all but one child in whom a right pleural effusion developed. Long-term follow-up was not possible in all cases because the patients resided in the rural areas of northern Namibia.
The types and extent of hepatic vein and IVC obstruction are schematically demonstrated in Fig. 2 and summaries of the operative procedures and results obtained are given in
Table I.
|
Cavography done 10 to 30 days after operation showed complete patency in two children (types I and II); residual IVC stenosis in two (both type II), but with complete resolution of the clinical component of the Budd-Chiari syndrome; and mild residual stenosis in three children (types I and II) that necessitated transiliac balloon angioplasty to normalize the IVCright atrium pressure gradients.
DISCUSSION
Although we have defined the pathologic process it is difficult to determine the exact cause of the Budd-Chiari syndrome in our patients. Because all but one of the patients in our series came from the Ovambo tribe in Namibia, and because of the sporadic occurrence of this disease in India and Japan, a genetic or environmental factor is likely. There was no familial tendency to suggest a genetic disposition although membranous obstruction of the IVC has been described in two sisters.
13
However, the constant position of the defect, the histologic similarities between the membrane and the wall of the IVC, the thin membranous character in some instances, and frequent involvement of the left hepatic vein suggest a developmental defect possibly related to extension of the normal obliterative process of the ductus venosus soon after birth.
7 Hirooka and Kimura
7 postulated an alteration in the development of the hepatic sinusoidal labyrinths that give rise to the hepatic portion of the IVC, which may either represent an arrest of the primitive hepatic extension of the subcardinal vein that forms that segment of the IVC
19 or persistence of the eustachian valves.
20
Simson
6,21 favors an acquired origin because of its usual manifestation during the fourth decade of life, the absence of associated congenital abnormalities, and the absence of cases reported in the neonatal period. Also, Terabayashi and colleagues
22 , published strong evidence that a vena caval thrombosis can organize and in the chronic phase produce a lesion identical to that in membranous obstruction of the IVC.
The natural history has not been defined in children.
9,13 Onset may be acute or insidious and the obstructive nature of the lesion masked by development of collateral circulation and the maintenance of relatively normal liver function. Hoffman, Stockland, and Van der Heyden
9 and Cameron and associates
23 have shown that the histologic evidence of progressive liver damage, which can develop during a short period, may improve rapidly with diminished congestion and hepatocellular regeneration after surgical decompression. Thus early diagnosis and possible surgical repair are essential to prevent permanent liver damage. Long-term outcome is less predictable and Koja and associates
24 reported on two adult patients in whom hepatomas developed after surgical relief. Thrombolytic therapy may have a place in the acute phase.
The most useful classification is the one proposed by Simson.
2 He described three types and although this is a working classification there is no evidence that the three types are related or that they represent a progression of the same disease. The classification, however, allows for a rational approach to surgery.
The common type of obstruction in our series and also in those from Japan involved obstruction of the retrohepatic or intrapericardial IVC, commonly from a web although in some this may be more extensive.
Accurate delineation of the extent of IVC and hepatic vein obstruction is crucial in determining the surgical approach. Ultrasonography is most useful in identifying the obstruction and in evaluating the patency of the main outflow tract. By placing catheters in the right atrium and IVC in close proximity to either side of the obstruction for simultaneous injection of contrast, accurate visualization can be obtained of the type and length of obstruction, patency of the hepatic veins, and extent of collateral circulation. Furthermore, venous pressure gradients between the right atrium and the IVC can be recorded and the degree of intrahepatic venous communicating channels can be evaluated.
The object of surgical repair is primarily to relieve the hepatic venous outflow tract.
25 Relief of obstruction to the IVC is of secondary importance because none of our patients had dependent edema or impairment of renal function.
25 The literature is scattered with reports of surgical procedures that unfortunately, despite much ingenuity, have been unsuccessful in the long term. Previous operative efforts only decompressed the portal system by using portacaval,
26 mesocaval, and mesoatrial shunts.
23,27 An excellent review of surgical procedures is provided by Stringer and associates.
28 These procedures may have moderate success in those instances in which there is obstruction of hepatic venous outflow without IVC obstruction. Reports from Japan advocate finger fracture of isolated membranous webs.
7 Recently percutaneous transluminal techniques with the use of balloon and atherectomy catheters, lasers, intravascular saws, and expandable metallic stents have been used with safety and success in patients with membranous obstruction of the IVC, but they have yet to be evaluated in the long term.
29,30 An absolute requirement of all these procedures is patency of at least one of the hepatic vein ostia for liver decompression.
5
Our surgical management has evolved with experience and it offers a rational approach to the surgical management of the Budd-Chiari syndrome caused by membranous obstruction of the IVC.
31-33 The technique of transcaval resection of the liver was initially described by Senning
34 and with modifications has been used in our patients. Murphy, Gregoric, and Cooley
35 resected a web and removed thrombus with the use of profound hypothermia and circulatory arrest and repair of the atriocaval junction with autologous pericardial patch. Similarly Kilman and associates
20 obtained symptomatic relief by transatrial resection of an obstructed eustachian valve. Our modifications include cannulation of the infrahepatic portion of the IVC rather than the iliac vein and the use of pericardium to roof the resultant defect rather than direct anastomosis of the capsule of the liver to the right atrium. The direct cannulation of the IVC distal to the obstruction allowed exact evaluation of the extent of the required surgical excision. Nojek and colleagues
36 used a venous cannula placed in the iliac vein and an additional solid catheter positioned so that it protruded into the occluded area, which was excised superiorly. Similarly Kimura and associates
37 in their finger fracture technique identified the membrane by palpation of an IVC catheter. Frimpong-Boateng and colleagues
12 were unsuccessful in their first attempt to correct a defect similar to the ones we studied but at repeat operation passed a balloon-tipped catheter into the IVC, which greatly facilitated the procedure. Our technique serves both purposes and offers greater potential for cure.
The hepatic circulation on the left side may be adequately decompressed by drainage into large diaphragmatic collateral vessels and it may not be necessary to extend the resection on this side. In one of our cases the hepatic veins could not be adequately opened and a PTFE graft was used in addition to bypass the obstructed IVC and to decompress the caudate lobe. The same approach has been used by Lygidakis, Bhalerao, and Udwadia,
38 who assumed that extensive intrahepatic venous collateral vessels would be sufficient to decompress the liver.
Although the early improvement in symptoms and signs suggests that the operation advocated by Senning
34 and also by us relieves hepatic venous obstruction, the long-term results are uncertain. Liver transplantation has been necessary after such a procedure.
39
Victor and associates,
40 via a right posterolateral thoracotomy with division of the hemidiaphragm, bypassed the IVC obstruction with a PTFE graft or homograft aorta but this procedure did not effectively relieve the hepatic venous obstruction, although it does offer an alternative approach to that used by Koja and associates
24 to open occluded hepatic veins and IVC. The reported patency of up to 5 years is encouraging because we were concerned about our use of a PTFE graft in one of our patients; this is an option for those patients in whom there is hepatic venous outflow obstruction.
14,40
Of concern has been the symptomatic restenosis of the IVC in three children. Evidence for this was present within the first month after primary membranectomy. All three responded to transiliac vein balloon angioplasty with restoration of normal pressure gradients in the short term. Unfortunately, no long-term follow-up was possible.
In some patients signs of fluid overload develop after relief of the IVC obstruction.
11,36,41 It is postulated that this is due to reabsorption of interstitial fluid.
36 Other patients have had shock and "asthma-like"dyspnea.
7 Whether this is due to the mechanism mentioned or pulmonary embolism is uncertain.
42 Prompt and vigorous diuresis should be induced if signs of hypervolemia develop. This syndrome was not observed in our patients.
In summary, we advocate percutaneous techniques if a thin membranous obstruction of the IVC without thrombus is demonstrated and the hepatic veins are patent. Although results are not always perfect in that gradients can increase with time and reocclusion can develop, the lesser morbidity with these procedures may be preferable.
43,44 If there is obstruction of hepatic venous drainage, resection of liver at the site of the obstruction and roofing of the resultant defect by pericardium with use of the technique described are recommended. If hepatic venous obstruction persists after these measures, balloon angioplasty or a shunt procedure may be required. Liver transplantation should be considered after shunting in cases with deteriorating clinical status and established cirrhosis.
45,46
Footnotes
*Gore-Tex brand PTFE tube graft. Gore-Tex is a registered trademark of W. L. Gore & Associates, Inc., Newark, Del. ![]()
References
This article has been cited by other articles:
![]() |
J. W. Gaynor, P. M. Weinberg, and T. L. Spray Congenital Heart Surgery Nomenclature and Database Project: systemic venous anomalies Ann. Thorac. Surg., April 1, 2000; 69(4): S70 - S76. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |