Pediatr Transplantation 2014: 18: 497–502

© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Pediatric Transplantation DOI: 10.1111/petr.12277

Hepatic venous outflow obstruction in pediatric liver transplantation: Technical considerations in prevention, diagnosis, and management Sommovilla J , Doyle MM, Vachharajani N, Saad N, Nadler M, Turmelle YP, Weymann A, Chapman WC, Lowell JA. Hepatic venous outflow obstruction in pediatric liver transplantation: Technical considerations in prevention, diagnosis, and management.

J. Sommovilla1, M. M. Doyle1, N. Vachharajani1, N. Saad2, M. Nadler3, Y. P. Turmelle3, A. Weymann3, W. C. Chapman1 and J. A. Lowell 1 1

Abstract: HVOO creates significant diagnostic and management dilemmas in pediatric liver transplant recipients, particularly with TVGs (split or reduced-size grafts). Numerous technical variations for the hepatic vein to IVC anastomosis have been described to minimize the incidence of this complication, but no consensus for an optimal anastomotic technique exists. One hundred and thirty-four liver transplants (70 TVGs) were performed in 124 patients between 1994 and 2011. These were divided into two cohorts. Group 1 (95 transplants, 41 TVGs) utilized a continuous running anastomosis. Group 2 (39 transplants, 29 TVGs) implemented a triangulated (threestitch) anastomosis. All were reviewed for demographics, diagnostics, interventions, and outcome. The overall HVOO incidence was seven of 134 transplants (5.2%) and six of 70 transplants utilizing TVGs (8.6%). Group 1 incidence was five of 41 (12.2%) compared with one of 29 (3.4%; p = 0.20, OR 3.89) in Group 2. Liver Doppler was employed in all patients, and only three suggested HVOO. All patients with HVOO underwent venogram, at a median of 81 days post-transplant. All underwent percutaneous venoplasty and required 1–6 treatments, all resulting in HVOO resolution. Incidence of HVOO has improved since adopting the triangulated anastomosis, although not to a level of statistical significance. US is not adequately sensitive to exclude HVOO. Venogram is recommended in patients with prolonged ascites, and venoplasty has been highly successful in HVOO treatment.

Liver transplantation is an established curative treatment for children suffering from fulminant hepatic failure, end-stage liver disease, liver malignancies, and certain inborn errors of metabolism. The number of small children suffering from these conditions far exceeds the number of organs available for whole organ transplant. As transplant outcomes have improved, technical

Abbreviations: HVOO, hepatic venous outflow obstruction; IVC, inferior vena cava; L Lat, left lateral section; MRA, magnetic resonance angiogram; TVG, technical variant grafts; US, ultrasound.

Department of Surgery, Washington University in St Louis, St Louis Children’s Hospital, St Louis, MO, USA, 2Department of Radiology, Washington University in St Louis, St Louis Children’s Hospital, St Louis, MO, USA, 3Department of Pediatrics, Washington University in St Louis, St Louis Children’s Hospital, St Louis, MO, USA

Key words: hepatic vein – stenosis – venogram – piggyback Josh Sommovilla, 660 S. Euclid Ave, Campus Box 8901, St Louis, MO 63110, USA Tel.: +1 215 760 7330 Fax: +1 314 747 1288 E-mail: [email protected] The findings in this manuscript were delivered as a minioral presentation entitled “When the outflow slows: technical and diagnostic considerations for hepatic venous anastomosis in pediatric liver presentation,” at the AHPBA conference in Miami Beach, FL on 2/24/2013. Accepted for publication 25 March 2014

advances employing the use of split and reducedsized grafts have expanded the donor pool available to these patients (1). Several case series have demonstrated excellent patient survival with use of such grafts (2). In the earliest series of liver transplants, obstruction of the hepatic vein to IVC anastomosis was uncommon, as a suprahepatic cavo-caval anastomosis was typically employed. With the use of piggyback technique, however, the incidence of HVOO has increased. It is of particular concern in transplantation of split or reducedsize grafts, which are predisposed to this complication. This predisposition is likely due to a 497

Sommovilla et al.

combination of factors including vessel size, graft position, and organ growth. Published case series have reported incidences ranging between 0% and 27.7% of HVOO in split or reduced-size liver transplants (3, 4). Several techniques for anastomosing the IVC to the hepatic veins have been described, all with the aim of minimizing the likelihood of outflow obstruction. One such technique, initially described in 1993, involves creating a triangularshaped anastomosis along the anterior surface of the IVC (5). The goal of any technique is to provide a widely patent anastomosis and to minimize the possibility of acting as an axis for torsion. In addition to the anastomosis itself, further measures to prevent twisting or kinking include tacking the falciform ligament to the abdominal wall or using the right colon as “space filler” in the right upper quadrant, although this is not uniformly carried out. HVOO can also create significant diagnostic and management dilemmas in the pediatric liver transplant population. Patients with HVOO most commonly present with recurrent ascites and/or right-sided pleural effusion, often in the setting of otherwise normal graft function. Thus, the diagnosis may be difficult, as the presenting sign may only be a marked delay in resolution of ascites following transplant. Less commonly, patients have been reported to present with altered liver function or even fulminant graft failure (6). In patients biopsied for elevated liver enzymes, case series have also reported the centrilobular necrosis found in HVOO mistaken for acute cellular rejection (7). US is commonly employed as an initial diagnostic study, as it can also assess other critical hepatic vasculature. For HVOO, however, venogram with measurement of hepatic vein-vena caval pressure gradient is considered the gold standard. This allows for therapeutic intervention with angioplasty or even stent placement (8). Because of the difficult anatomic positions of this anastomosis, re-operation is rarely employed in the treatment of hepatic vein outflow obstruction. The most common therapy employed has been percutaneous venoplasty at the time of diagnosis. Stent use has been reported, but has been avoided at our institution due to concern over stent migration with patient and graft growth (9). In cases of vein stenosis and severe organ disfunction, need for retransplantation has been reported (6). The purpose of this study is to review our center’s experience with the incidence, diagnosis, and management of outflow obstruction, along with a change in technique to a “triangulated” caval anastomosis in split grafts. 498

Patients and methods Between 10/28/1994 and 11/8/2011, 134 liver transplants were performed at Saint Louis Children’s Hospital on 124 patients. Of these, 70 of the transplanted grafts were either split or reduced-sized grafts (TVGs). Prior to 2001, the hepatic vein to IVC anastomosis for all transplants was performed utilizing a standard running, single-suture anastomosis. In 2001, a triangulated, three-stitch technique was first utilized on a six-month-old patient receiving an L Lat graft for biliary atresia. Briefly, a wide orifice is created on the recipient anterior vena cava by dividing the bridges of tissues between the recipient hepatic veins. Three stay sutures are then placed at the cavotomy to form the corners of a triangle. The donor hepatic vein is enlarged with a longitudinal incision on the posterior surface of the vein. Each stay suture is used during the anastomosis to close one side of the triangle and anchored to the stay suture in an adjacent corner. The construction of this anastomosis is shown in Fig. 1. Beginning in 2006, this became the standard technique employed for the caval anastomosis on TVGs. In all cases, absorbable monofilament suture has been used for this anastomosis. In the years spanning this study, only minimal technical changes were made to other aspects of the transplant operation in very small recipients. Measures taken to minimize

Fig. 1. Construction of triangulated anastomosis: individual stay suture placed on the IVC at each corner of the triangle (top left inset); closure by sewing each stay suture to an adjacent corner (top right inset); completed triangulated anastomosis (bottom).

Venous outflow in pediatric liver transplant the risk of graft rotation and kinking included tacking of the falciform ligament to the abdominal wall; cutting the donor hepatic veins as short as possible; and utilization of the right colon to serve as “space filler” in the right upper quadrant. Changes over the years have included the increasing use of aortic conduits for arterial inflow and the use of silastic patches for temporary abdominal wall fascial closure. Regarding post-operative care, no changes to use of anticoagulant and antiplatelet therapy were made during the course of this series. Aspirin was used for antiplatelet therapy, and no anticoagulation was routinely used unless otherwise indicated. Antirejection medication regimens have also changed only slightly, with introduction of the use of an antimetabolite and steroids for just three months following transplant, and long-term maintenance with use of a single-drug regimen (tacrolimus) in absence of rejection or other medication-related complications. Workup for HVOO was initiated based on clinical suspicion, and diagnostic studies were performed, which included US, CT angiogram, MR angiography, venogram with pressure gradient measurement, or combinations thereof. Charts of all 134 transplants performed at St Louis Children’s Hospital were reviewed to analyze the incidence, diagnostic workup, treatment, and outcome of HVOO. This study focuses on the subgroup of TVG transplants, of which there were 70. The series was divided into two cohorts, based on the technique implemented for donor hepatic vein to recipient IVC anastomosis. The first group utilized a continuous running single-suture anastomotic technique, and the second group implemented a triangulated anastomosis to the recipient vena cava. Information regarding patient demographics, disease etiology, operative technique, and post-operative course (including mortality, morbidity, graft function, diagnostic imaging, subsequent procedures) was obtained through this chart review.

Results

In all, 70 liver transplants were performed using TVGs. Of these, 41 utilized a continuous running anastomosis (Group 1), and 29 utilized a triangulated technique (Group 2). The indications for transplantation in the two groups are shown in Table 1. The total incidence of HVOO in TVGs was 6 of 70 (8.6%). The incidence of hepatic arterial thrombosis in this group was four of 70 (5.7%). In Group 1, the incidence of HVOO was five of 41 (12.2%), compared with one of 29 (3.4%; p value = 0.20) in Group 2. Additionally, one recipient of a whole graft also developed HVOO, and the incidence of this complication was seven of 134 (5.2%) in all of the transplants performed over the time period covered in this series (TVGs and whole grafts). As Table 2 shows, Group 2 trended toward a slightly younger patient population, but this difference was not significant. There was no difference between groups regarding the method of graft splitting or type of graft used. Information pertaining to the patients diagnosed with HVOO is shown in Table 3.

Table 1. Indications for transplantation among TVG recipients

Biliary atresia Ornithine transcarbamylase deficiency Alagille syndrome Cryptogenic fulminant failure Hepatoblastoma Progressive familial intrahepatic cholestasis Alpha-1 antitrypsin deficiency Other inborn errors metabolism Autoimmune Choledochal cyst/Caroli disease Cystic fibrosis Refsum disease Arteriovenous malformations Initial graft failure Other malignancy

Running anastomosis (n = 41)

Triangulated anastomosis (n = 29)

16 1

12 3

– 2

1 3

3 2

3 1

4



2

1

1 –

1 1

3 1 1

– – –

3 2

3 –

Table 2. Recipient age and split technique

n Mean age, months Median age, months % split in situ (%) Number with HVOO (%)

Group 1 (running)

Group 2 (triangulated)

41 38.8 14.1 21.9 5 (12.2)

29 25.7* 10.1 20.7 1 (3.4)**

*p = 0.10; **p = 0.20.

Regarding presenting signs, all patients with HVOO had persistent or recurrent ascites. Two patients also presented with recurrent pleural effusions. Only two patients had liver function enzymes elevated beyond their baseline at time of diagnosis. No patients presented with outflow obstruction acutely; the earliest case was diagnosed at 41 days post-transplant, and HVOO was diagnosed as late as 254 days post-transplant. Liver doppler US examination was employed in all patients as an initial diagnostic study. In four of seven patients, US revealed no abnormality in hepatic venous flow, whereas three suggested some degree of outflow congestion. Patients in whom there was concern for HVOO, regardless of US result, underwent confirmatory venogram with measurement of pressure gradients. In addition to those with confirmed outflow obstruction, four patients with refractory ascites not explained by non-invasive studies underwent 499

500

Biliary atresia Primary non-function or previous transplant (for biliary atresia) 8 25.3

59

16.8

7.5 9.8

Triangulated Running

L Lat Whole

254 71

Ascites Ascites, pleural effusion

1 2 (stent)

HVOO resolved, patient death from primary disease HVOO resolved HVOO resolved; rejection 2 Ascites 41 L Lat Running

3 172 L Lat

Histiocytosis; sclerosing cholangitis Refsum disease 43

13.2

Running

62 203 L Lat L Lat Biliary atresia Hepatoblastoma 5.5 50

5.7 16.5

Running Running

Ascites

+

HVOO resolved Retransplanted, graft failure secondary to biliary strictures HVOO resolved; Chronic rejection 6 1 +

HVOO resolved 5

Effusion, ascites, extremity edema Ascites Ascites 83 L Lat Running Biliary atresia 10.1 11

Indication for transplantation Wt (kg) Age at Txp (month)

Table 3. Summary of patients diagnosed with HVOO

Anastomotic technique

Graft type

Interval to Dx (days)

Presenting signs

US result

Number of veno plasties

Outcome

Sommovilla et al.

venography for clinical suspicion of HVOO. These patients were found to have normal venous outflow. One patient with splenomegaly and abnormal liver function underwent MRA that revealed no HVOO. All patients with HVOO underwent venoplasty at the time of confirmatory venogram. Patients required between one and six venoplasties, and all patients eventually experienced resolution of HVOO. Pre- and post-venogram images are shown for one such patient in Fig. 2. One patient had a stent placed, which was performed during her second venoplasty. This was actually performed in the lone patient with HVOO after transplantation of a whole graft. Her outflow obstruction resolved after stent placement, and the stents were confirmed in proper position by venogram two and a half yr after placement. This patient has also experienced multiple episodes of acute rejection that have led to redevelopment of compensated cirrhosis, but has experienced no clinical or radiographic evidence of HVOO recurrence since stent placement. None of the patients in this series required re-operation for revision of the hepatic vein-caval anastomosis. One patient with HVOO, a child with Refsum disease, died of neurologic decline two yr following her transplant and had normal liver function at the time of death. One patient required retransplantation for biliary strictures refractory to stenting and drainage, causing recurrent cholangitis and eventually cholestatic graft failure. Two patients redeveloped cirrhosis secondary to chronic rejection. The remaining patients remain healthy with good graft function. There was no difference between groups in patient or graft survival. Overall patient survival

Fig. 2. Radiographic images of stenotic hepatic vein-IVC anastomosis before (left) and after (right) percutaneous venoplasty in an L Lat graft.

Venous outflow in pediatric liver transplant

at one and five yr was 89.5 and 78.7% for Group 1 and 89.7 and 86.1 for Group 2, respectively (p = 0.50). One- and five-yr graft survival was 86.8% and 76.1% for Group 1 and 82.8 and 75.0% for Group 2 (p = 0.98). Patient and graft survival are shown in Fig. 3 and Table 4.

Table 4. Overall patient and graft survival in Group 1 and Group 2 Time since transplant Patient survival 30 days 1 yr 3 yr 5 yr Graft survival 30 days 1 yr 3 yr 5 yr

Discussion

While HVOO is not the most feared anastomotic complication in the early post-transplant period, adequate hepatic venous outflow is essential for long-term graft function. Outflow obstruction is of much greater concern in split or reduced-size transplants, in which the outflow tract is threatened by vessel size and the possibility of torquing with graft growth and positional changes. Techniques commonly described for stabilization of graft position and orientation include the following: suturing of the falciform ligament anteriorly to the diaphragm, minimization of hepatic vein length, and utilization of the right colon as a space filler. There is no consensus, however, regarding the optimal technique for constructing the anastomosis itself. Reported techniques include the following: a continuous running (single stitch) closure across a transverse cavotomy; creation of a triangular caval orifice closed using three sutures (5); a wide longitudinal cavotomy anastomosed with a running stitch (10); and cutting donor hepatic vein flush with liver parenchyma (11). The present study focused on a comparison of the first two of these techniques. Our series shows an incidence of HVOO at 3.45% since switching to a triangulated anastomotic technique. Despite this low incidence, our series does not reveal a statistically significant decrease compared with our previous running technique. Other recent series have reported a similarly low incidence of this complication. Tannuri et al. describe an incidence decrease from 27.7 to 5.7% when switching to a triangulated technique (10). There is recent evidence to suggest other novel techniques can similarly decrease the incidence of HVOO. In the aforementioned study, Tannuri’s group further switched to a creation of a large

Group 1 Group 2

60 40 20 0

4

8

12

Years since transplant

16

Graft survival (%)

Patient survival (%)

89.5 89.5 81.4 78.7

89.7 89.7 86.1 86.1

89.5 86.8 78.8 76.1

86.2 82.8 79.2 75.0

100

80

p=.4970

Fig. 3. Overall patient and graft survival in Group 1 and Group 2.

Group 2

longitudinal incision on the recipient IVC after closure of the hepatic vein orifice, and using this to fashion a widely patent anastomosis. They reported a zero incidence of HVOO in 47 patients transplanted with this technique. Similarly, Heffron et al. report a zero incidence of HVOO in 102 transplants utilizing a donor hepatic vein cut very short – flush with the liver parenchyma – to minimize the likelihood of anastomotic kinking (11). Considering this, along with the fact that HVOO potentially develops as a result of several contributing factors, we cannot state with certainty that a triangulated anastomosis alone is responsible for a decreased likelihood of HVOO. We have achieved very good results with this anastomosis (only one episode of outflow obstruction), and expect the decrease in HVOO to become more pronounced with its continued use. As other centers have reported excellent results with other novel techniques, there are likely several options for minimizing HVOO. We found US to be insufficiently sensitive as a diagnostic tool for HVOO. Because of other clinical concerns and the possibility of concurrent hepatic arterial and portal venous abnormalities, however, US remains a reasonable first-choice screening study, primarily to exclude other causes of ascites. In a patient with refractory ascites and a normal US, however, further evaluation for HVOO should be initiated. While MR angiography has proven useful in diagnosis, it

100

0

Group 1

Group 1 Group 2

80 60 40 20 0

p=.9830 0

4

8

12

16

Years since transplant

501

Sommovilla et al.

provides no therapeutic opportunity despite often requiring sedation in this patient population. Venogram should be considered the gold standard, which also allows therapeutic intervention at the same procedure. In our series, venoplasty was successful in resolving HVOO in all patients. This resolution only rarely occurred with a single treatment, and most patients required multiple venoplasty treatments. Although stent placement can be considered in children with adult-sized hepatic veins, it should be avoided in very small children due to the possibility of stent migration with patient and graft growth (12). There have, however, been case series describing effective and safe stent use for this complication (13). In our series, no recipients of TVGs had stents placed. The lone patient stented, who had received a whole liver graft, has experienced no stent migration or recurrence of HVOO three yr after stent placement. This study is limited by its retrospective nature and its relatively small sample size. While this study focuses on a comparison of anastomotic techniques, other factors such as graft size, growth and orientation likely contribute to the occurrence of HVOO. Additionally, while several techniques have been described to avoid HVOO, only two have been utilized and compared at our institution. We have demonstrated a very low (3.4%) incidence of HVOO after switching to a triangulated technique, but other institutions have shown different techniques to be highly successful. Although we found no statistical significance between our two techniques, we strongly advocate continued use of a wide triangulated (three stitch) anastomosis. Conflict of interest

The authors have no financial or other disclosures to report.

502

References 1. BROELSCH CE, EMOND JC, WHITINGTON PF, et al. Application of reduced-size liver transplants as split grafts, auxiliary orthotopic grafts, and living related segmental transplants. Ann Surg 1990: 212: 368–375; discussion 375–377. 2. SHEHATA MR, YAGI S, OKAMURA Y, et al. Pediatric liver transplantation using reduced and hyper-reduced left lateral segment grafts: A 10-year single-center experience. Am J Transplant 2012: 12: 3406–3413. 3. KRISHNA KUMAR G, SHARIF K, MAYER D, et al. Hepatic venous outflow obstruction in paediatric liver transplantation. Pediatr Surg Int 2010: 26: 423–425. 4. BUELL JF, FUNAKI B, CRONIN DC, et al. Long-term venous complications after full-size and segmental pediatric liver transplantation. Ann Surg 2002: 236: 658–666. 5. EMOND JC, HEFFRON TG, WHITINGTON PF, et al. Reconstruction of the hepatic vein in reduced size hepatic transplantation. Surg Gynecol Obstet 1993: 176: 11–17. 6. SAKAMOTO S, NAKAZAWA A, SHIGETA T, et al. Devastating outflow obstruction after pediatric split liver transplantation. Pediatr Transplant 2012: 17: E25–E28. 7. GIBELLI NE, TANNURI AC, ANDRADE WC, et al. Centrilobular necrosis as a manifestation of venous outflow block in pediatric malnourished liver transplant recipients – case reports. Pediatr Transplant 2012: 16: E383–E387. 8. KO GY, SUNG KB, YOON HK, et al. Endovascular treatment of hepatic venous outflow obstruction after living-donor liver transplantation. J Vasc Interv Radiol 2002: 13: 591–599. 9. GOELITZ BW, DARCY M. Longitudinal stent fracture and migration of a stent fragment complicating treatment of hepatic vein stenosis after orthotopic liver transplantation. Semin Intervent Radiol 2007: 24: 333–336. 10. TANNURI U, SANTOS MM, TANNURI AC, et al. Which is the best technique for hepatic venous reconstruction in pediatric livingdonor liver transplantation? Experience from a single center. J Pediatr Surg 2011: 46: 1379–1384. 11. HEFFRON TG, PILLEN T, SMALLWOOD G, et al. Incidence, impact, and treatment of portal and hepatic venous complications following pediatric liver transplantation: A single-center 12 year experience. Pediatr Transplant 2010: 14: 722–729. 12. DARCY MD. Management of venous outflow complications after liver transplantation. Tech Vasc Interv Radiol 2007: 10: 240–245. 13. KO GY, SUNG KB, YOON HK, et al. Early posttransplant hepatic venous outflow obstruction: Long-term efficacy of primary stent placement. Liver Transpl 2008: 14: 1505–1511.

Hepatic venous outflow obstruction in pediatric liver transplantation: technical considerations in prevention, diagnosis, and management.

HVOO creates significant diagnostic and management dilemmas in pediatric liver transplant recipients, particularly with TVGs (split or reduced-size gr...
215KB Sizes 0 Downloads 3 Views