Portomesenteric venous thrombosis associated with rectal venous malformations

Purpose

We report thrombosis of portal and mesenteric veins in patients with a pattern of rectal venous malformations (VMs) and ectatic major mesenteric veins.

Methods

Eight patients having rectal VMs with either ectatic mesenteric veins and/or evidence of portomesenteric venous thrombosis (PVT), evaluated from 1995-2009, were reviewed.

Results

Portomesenteric venous thrombosis was evident in 5 patients at presentation. Three had patent ectatic mesenteric veins, 2 with demonstrated reversal of flow, and 2 of whom went on to thrombosis during observation. Six patients developed portal hypertension. Five remain on long-term anticoagulation. After recognizing this pattern, one patient underwent preemptive proximal ligation of the inferior mesenteric vein (IMV) to enhance antegrade portal vein flow and prevent propagation or embolization of venous thrombus from the IMV to the portal vein.

Conclusion

Rectal VMs should be evaluated for associated ectatic mesenteric veins. The ectatic vein siphons flow from the portal vein down to the rectal VM, leading to stagnation of blood in the portal vein and resultant thrombosis. Primary thrombosis in the stagnant rectal VM and/or mesenteric vein can also predispose to embolization up into the portal vein. This pattern of rectal VM and ectatic mesenteric vein should be considered a risk factor for devastating PVT.     

Posttransplantation portal thrombosis secondary to splenorenal shunt persistence

Purpose

The aim of this study was to analyze our experience with portal vein thrombosis after liver transplantation with a persistent splenorenal shunt.

Materials and methods

The study population included 780 liver transplantations from 1990 to 2009. We analyzed the existence of portal vein thrombosis in the immediate posttransplant period, selecting cases with a persistent splenorenal shunt requiring surgery.

Results

The incidence of posttransplant portal vein thrombosis was 1.41% (n = 11), of which 3 (27%) had a splenorenal shunt as a possible cause (0.38% of the total). Two cases required liver retransplantation due to portal vein thrombosis, and the third a thrombectomy. In all cases the shunt was also closed. During the early postoperative follow-up of these 3 patients, 2 needed repeat surgeries because of a new portal vein thrombosis (thrombectomy) in one and a bilioperitoneum in the other. After a median follow-up of 11 months, the patients showed a good evolution with no primary graft dysfunction.

Discussion

The portal steal phenomenon secondary to persistence of a splenorenal shunt promotes the occurrence of portal vein thrombosis. Although it is a rare cause of graft dysfunction, it must be treated early, because it can lead to a small-for-size syndrome.     

Portal-to-right portal vein bypass for extrahepatic portal vein obstruction

Objective

Rex shunt (mesenteric-to-left portal vein bypass) is considered a more physiologically rational treatment for EHPVO than other portosystemic systemic shunts in children. However, about 13.6% of children with EHPVO do not have usable left portal veins and up to 28.1%. Rex operations in children are not successful. Hence, a Rex shunt in these children was impossible. This study reports a novel approach by portal-to-right portal vein bypass for treatment of children with failed Rex shunts.

Previous portal hypertension surgery negatively affects results of mesenteric to left portal vein bypass

Purpose

The mesenteric to left portal vein bypass (MLPVB) has been successfully used to treat extrahepatic portal vein obstruction (EHPVO) in children. We examined the effect of failed prior surgical or radiological procedures intended to treat complications of portal hypertension on the success rate of subsequent MLPVB surgery.

Methods

Sixty-two patients younger than 18 years with EHPVO underwent MLPVB between 1997 and 2006. Children were divided into 3 groups: those with no prior surgery related to portal hypertension, those with prior portosystemic shunts, and those with either splenectomy or mesenteric vascular embolization procedures. The effect of prior procedures on the patency rate of the MLPVB was then examined.

Results

Of 62 children, 11 (17.7%) had significant procedures to treat symptoms of portal hypertension: 6 had at least 1 portosystemic shunt attempt, 3 had isolated splenectomy, and 2 had embolization of the splenic artery or coronary and peripancreatic varices. Patients with previous portal hypertension surgery were significantly older and larger than those with no surgery. Patients with no prior interventions had a significantly higher MLPVB patency rate (88.2%, 45/51) than those with no prior interventions (63.6%, 7/11). Prior splenectomy alone was not found to adversely affect MLPVB. Patients with prior embolization procedures or unsuccessful shunts had significantly poorer successful outcomes (0% and 66.7%) than those with no prior interventions (88.2%; P < .005).

Conclusions

The results demonstrate that prior portosystemic shunts or mesenteric embolizations have a deleterious effect on outcome after MLPVB and should be avoided whenever possible. This study suggests that patients with symptomatic EHPVO should undergo MLPVB as a primary intervention rather than as a rescue procedure to optimize MLPVB patency.     

Extrahepatic portal vein morphology in children with extrahepatic portal hypertension assessed by 3-dimensional computed tomographic portography: a new etiology of extrahepatic portal hypertension

Background/Purpose

The cause of extrahepatic portal hypertension in children has not been clarified. Our aim was to determine the morphological features of the extrahepatic portal vein in children with extrahepatic portal hypertension by 3-dimensional computed tomographic portography and to clarify the etiology of this disorder.

Materials and Methods

Six patients ranging in age from 10 to 18 years (median age, 12.8 years) who had portal hypertension presented with hematemesis. They underwent intravenous computed tomographic portography using a helical computed tomography scanner and 3-dimensional image reconstruction.

Results

The extrahepatic portal vein was visualized in all patients by 3-dimensional computed tomographic portography. None of the patients showed extrahepatic portal vein obstruction or cavernous transformation. All patients had a tortuous η-shaped extrahepatic portal vein, and a line could be drawn through the flexures of the portal vein to the hepatic hilum.

Conclusion

In children, extrahepatic portal hypertension is not caused by extrahepatic portal vein obstruction and may be of embryological origin.     

Preoperative imaging of left portal vein at the Rex recess for Rex shunt formation using wedged hepatic vein carbon dioxide portography

Background

In children with extrahepatic portal vein obstruction (EHPVO), formation of a mesentericoportal bypass (Rex shunt) restores hepatopetal flow, relieves portal hypertension, and reduces variceal bleeding and hypersplenism. The Rex shunt is created by inserting a vein graft between the superior mesenteric vein and the umbilical segment (Rex) of the left portal vein within the Rex recess of the liver. The preoperative evaluation of a patient with EHPVO includes an accurate assessment of the venous inflow and outflow. The inflow portal vein is readily assessed by ultrasound and magnetic resonance imaging. The outflow intrahepatic portal vein is harder to assess. We report our experience of patients evaluated with wedged hepatic vein carbon dioxide portography (WHVCP).

Method

All children referred for venography from October 2001 to October 2007 were prospectively identified, and clinical and radiologic data were reviewed retrospectively. The imaging findings were correlated to findings at surgery.

Results

Eleven children (range, 3-14 years, median, 6 years) were referred for preoperative wedged hepatic venography. The left portal vein at the Rex recess was clearly identified in 9 patients (82%). In the other 2 patients (18%), the Rex segment was not identified despite opacification of left and right intrahepatic portal veins; this was taken to indicate an occluded segment. Wedged venography was performed with carbon dioxide in 10 patients (91%). Carbon dioxide was contraindicated in the final patient because of the presence of a ventricular septal defect.

Conclusion

Our series demonstrates the use of WHVCP as a diagnostic tool in preoperative assessment of the Rex segment of left portal vein in children with extrahepatic portal vein obstruction.     

Extrahepatic portal hypertension treated by anastomosing inferior mesenteric vein to left portal vein at Rex recessus

For the treatment of recurrent bleeding despite sclerotherapy or clinically significant hypersplenism, portosystemic shunt procedures should be performed in cases of extrahepatic portal hypertension caused by extrahepatic portal vein thrombosis. A novel alternative to portosystemic shunt procedures in extrahepatic portal hypertension is mesenterico-left portal bypass. Portal vein thrombosis is bypassed by an autologous vein graft (usually left internal jugular vein) interposed between superior mesenteric vein and left portal vein. In the presence of an enlarged right gastroepiploic vein, the distal end of this vein can be anastomosed to left portal vein without disturbing its proximal end. Herein, the authors report a case of extrahepatic portal hypertension treated by anastomosing enlarged inferior mesenteric vein to left portal vein to bypass portal vein thrombosis.     

Rex shunt for acute portal vein thrombosis after pediatric liver transplantation in children with biliary atresia

Background/Purpose

Posttransplantation portal vein thrombosis (PVT) can have severe health consequences, and portal hypertension and other consequences of the long-term privation of portal inflow to the graft may be hazardous, especially in young children. The Rex shunt has been used successfully to treat PVT patients since 1998. In 2007, we started to perform this surgery in patients with idiopathic PVT and late posttransplantation PVT. Herein we have reported our experience with this technique in acute posttransplantation PVT.

Methods

Three patients of ages 12, 15, and 18 months underwent cadaveric (n = 1) or living donor (n = 2) orthotopic liver transplantation (OLT). All patients had biliary atresia with portal vein hypoplasia; they developed acute PVT on the first postoperative day. They underwent a mesenteric-portal surgical shunt (Rex shunt) using a left internal jugular vein autograft (n = 2) or cadaveric iliac vein graft (n = 1) on the first postoperative day.

Results

The 8-month follow-up has confirmed shunt patency by postoperative Doppler ultrasound. There have been no biliary complications to date.

Conclusions

The mesenteric-portal shunt (Rex shunt) using an autograft of the left internal jugular or a cadaveric vein graft should be considered for children with acute PVT after OLT. These children usually have small portal veins; reanastomosis is often unsuccessful. In addition, this technique has the advantage to avoid manipulation of the hepatic hilum and biliary anastomosis. Although this study was based on a limited experience, we concluded that this technique is feasible, with great benefits to and low risks for these patients.     

Long-term follow-up of children with extrahepatic portal vein obstruction: impact of an endoscopic sclerotherapy program on bleeding episodes, hepatic function, hypersplenism, and mortality

Background

Endoscopic sclerotherapy (ES) has been the standard treatment for children with idiopathic extrahepatic portal vein obstruction (EHPVO). Portosystemic shunts are indicated when variceal bleeding cannot be controlled by ES. Recently, mesenteric left portal vein bypass was indicated as a surgical intervention and preventative measure for hepatic dysfunction in children with long-term EHPVO. Nevertheless, there is a lack of published data confirming the extent of hepatic dysfunction, hypersplenism, and physical development in children with long-term follow-up.

Method

We retrospectively verified the long-term outcomes in 82 children with EHPVO treated with ES protocol, focusing on mortality, control of bleeding, hypersplenism, and consequent hepatic dysfunction.

Results

Of the children, 56% were free from bleeding after the initiation of ES. The most frequent cause of rebleeding was gastric varices (30%). Four patients had recurrent bleeding from esophageal varices (4.6%). Four patients underwent surgery as a consequence of uncontrolled gastric varices. There were no deaths. Most patients showed good physical development. We observed a mild but statistically significant drop in factor V motion, as well as leukocyte and platelet count.

Conclusion

Endoscopic sclerotherapy is an efficient treatment for children with EHPVO. The incidence of rebleeding is low, and there was no mortality. Children develop mild liver dysfunction and hypersplenism with long-term follow-up. Only a few patients manifest symptoms of hypersplenism, portal biliopathy, or liver dysfunction before adolescence.     

Modified technique of meso-Rex shunt in case of insufficient length of the jugular vein graft

Meso-Rex shunt (MRS) can relieve portal hypertension and restore a physiological portal flow in patients with portal vein thrombosis. We describe a technical variant where the autologous internal jugular vein (IJV) was too short to bridge the superior mesenteric vein (SMV) and the Rex recessus.

Patient

A 15-year-old boy with portal cavernoma had several episodes of gastrointestinal bleeding despite repeated sclerotherapy. Preoperative assessment, including retrograde transjugular portography, showed persistent esophageal and gastric varices, severe hypertensive gastropathy, obstructed portal vein, patent SMV and splenomesenteric confluence, patent intrahepatic portal branches, and normal transhepatic pressure gradient. An MRS was planned. The left IJV was retrieved from its infracranial part to its confluence with subclavian vein. After performing the Rex recessus to IJV graft anastomosis, the IJV graft proved to be too short for classical end-to-side anastomosis onto the SMV. After clamp testing showing good tolerance of the small bowel, the proximal jejunal branches of the SMV were tied, the proximal SMV was mobilized and transsected 4 cm below the pancreas, and an end-to-end anastomosis between SMV and IJV was performed. Portal pressure decreased from 23 to 13 mm Hg, and intraoperative Ultra Sound Doppler (US Doppler) showed good flows in the shunt. Postoperative course was uneventful, and 1 year after surgery, the child is clinically well, off medication, with a patent shunt, and no portal hypertension.

Conclusion

This modified MRS technique may be useful when the autologous IJV graft is too short, avoiding the need for prosthetic conduits and prolonged postoperative anticoagulation.     

Extrahepatic portal vein thrombosis after umbilical catheterization: is it a good choice for Rex shunt?

Background

Extrahepatic portal vein thrombosis (EHPVT) is an important cause of portal hypertension in children. Rex shunt has been used successfully to treat these patients.

Methods

We report our experience in 19 infants and children (5 months to 14 years) with HPVT eligible for a mesenteric-portal surgical shunt with left internal jugular vein autograft. Eight children had idiopathic EHPVT, nine had post-umbilical catheterization EHPVT, one had portal vein agenesis, and one had posttransplant EHPVT.

Results

It was possible to perform the Rex shunt in all patients except for 8 of 9 cases in the post-umbilical catheterization EHPVT group. A Warren procedure was performed in 4 of those patients and a proximal splenorenal shunt in 1. Current follow-up ranges from 3 to 26 months. Shunt thrombosis occurred in one patient with portal vein agenesis and associated cardiac anomaly. Portal hypertension has significantly improved after surgery. None of our patients have experienced new bleeding episodes until now.

Conclusions

The Rex shunt should be considered in the treatment of children with idiopathic EHPVT experiencing repeated gastrointestinal bleeding episodes refractory to endoscopic treatment. Nevertheless, the role of this operation for children with post-umbilical catheterization EHPVT is yet to be clearly evaluated.     

成人肝外型门静脉高压症诊治分析

目的: 探讨成人肝外型门静脉高压症(EHPH)的临床特点和治疗策略. 方法: 回顾分析对6例EHPH病人的诊治过程. 结果: 5例病人以反复上消化道出血及脾亢为主要临床表现,经彩色多普勒超声及门静脉造影检查,3例有典型门静脉海绵样变(CTPV)表现,1例为门静脉血栓形成,1例为卵巢癌转移压闭脾静脉.断流术3例,肠腔分流术1例,内镜下曲张静脉硬化及结扎治疗2例,平均随访8.4个月,死亡2例. 结论: 成人EHPH病因复杂,诊断主要依靠彩超及门静脉造影,治疗方案根据病人具体情况而定.     

Effectiveness of Rex shunt in the treatment of portal hypertension

Background

Children with portal venous thrombosis often have severe symptoms secondary to portal hypertension including recurrent upper gastrointestinal (UGI) bleeds and hypersplenism. We report results of the use of the mesenterico-left portal bypass (Rex shunt) in 5 consecutive patients.

Methods

A retrospective chart review of all patients with portal venous thrombosis who underwent Rex shunt procedure was performed. Children were evaluated preoperatively with magnetic resonance angiogram, Doppler ultrasound, and mesenteric angiogram. Postoperative ultrasounds were performed in follow-up.

Results

All patients had evidence of portal hypertension and hypersplenism. The average age of the patients was 13.2 ± 4.9 (7-19) years. The patient had an average of 2.6 ± 1.7 UGI bleeds requiring banding and 3.4 ± 4.2 U of blood transfused before undergoing the shunt. The mean operative time was 383 ± 46 minutes, and length of stay was 10.4 ± 7.1 days. In follow-up of 18.8 ± 5.2 months (11-24 months), all but 1 patient had a patent shunt. One narrowed shunt was successfully dilated by percutaneous angioplasty. Thrombocytopenia improved significantly in patients with functioning shunts but did not correlate with a significant decrease in splenic size.

Conclusions

The Rex shunt reestablishes normal hepatopedal portal flow, and in patients with functioning shunts, no recurrent UGI bleeds or transfusions were required or evidence of encephalopathy were noted.     

Extrahepatic portal vein thrombosis is associated with an increased incidence of cholelithiasis

Background/Purpose

The effect of portal flow deprivation to the liver on bile composition and the biliary system remains undefined in children. This report catalogues the authors’ experience with biliary tract problems in children with extrahepatic portal vein thrombosis (EHPVT).

Methods

Twenty-nine children with symptomatic idiopathic EHPVT were evaluated for the Rex shunt procedure (mesenterico-left portal bypass) over a 4-year period. The authors retrospectively reviewed all operative reports and pre- and postoperative abdominal ultrasound findings with regard to associated congenital anomalies and abnormal biliary tract findings.

Results

Seven of the 29 patients with EHPVT (24%) had associated nonbiliary congenital abnormalities. Twenty-four of 29 (83%) patients had detectable biliary tract pathology by ultrasound examination. Biliary symptoms developed in 3 of the 9 (33%) patients with either stones or sludge (10.3% of all patients). Two patients were treated by cholecystectomy. There was no statistical correlation between biliary tract pathology and the age of presentation, symptoms of portal hypertension, gender, or underlying medical condition.

Conclusions

The authors have noted a high incidence of biliary tract pathology in patients with EHPVT compared with the normal population and a 10% incidence of symptomatic biliary pathology in this series.     

Endoscopic ligation of esophageal varices for prophylaxis of first bleeding in children and adolescents with portal hypertension: preliminary results of a prospective study

Background/purpose

Endoscopic variceal ligation (EVL) is effective in controlling rebleeding from esophageal varices in children, but there is no data on the use of EVL to prevent initial bleeding. The objective of this study was to prospectively evaluate the efficacy of EVL in preventing the first hemorrhage from esophageal varices in children.

Methods

Thirty-seven children with portal hypertension (22 liver cirrhosis, 15 portal vein thrombosis), aged 4 to 17 years (M = 9.5 ± 4.4 years) were included in the study. The criteria for inclusion were (1) no previous variceal bleeding; (2) the presence of esophageal varices classified grade II or more, and (3) their enlargement by at least I grade after 6 months of observation without endoscopic treatment or appearance of endoscopic signs of high bleeding risk. A Multi-Band Ligator was used, and 2 to 6 bands were fixed under general anesthesia during one procedure depending on the number and size of varices. Follow-up examinations were performed every 3 months, repeating the procedure if necessary. In total, 75 procedures of EVL were performed, from one to 5 in each patient

Results

Four patients underwent liver transplantation before eradication of varices. Two others were excluded from the observation because of lack of compliance to the protocol. Of the remaining 31 patients, eradication of varices was achieved in 28 children (90.3%) after 2.0 EVL sessions performed at 3-month intervals. The average time of follow-up after cessation of treatment is 16 months. No bleeding from varices occurred in any child during or after treatment. There were no differences in results between children with liver cirrhosis and portal vein thrombosis. Development of hypertensive gastropathy was observed in 2 children with one episode of bleeding. Recurrence of varices without bleeding occurred in 3 children after 12, 13, and 28 months from eradication.

Conclusions

The study results confirmed that endoscopic variceal ligation is a safe and highly effective procedure in children with portal hypertension, regardless of its etiology. Eradication of esophageal varices was followed by 16 months free of bleeding. Prolonged observation is mandatory to conclude if preventive EVL influences the natural history of disease and diminishes the risk of first bleeding onset.     

Liver nodules after portal systemic shunt surgery for extrahepatic portal vein obstruction in children

Background

Liver nodules have been reported after portal systemic shunt surgery (PSSS) in animal experiments or in humans with liver cirrhosis. The aim of our study was to assess the incidence of liver nodules after surgery for extrahepatic portal vein obstruction (EHPVO) in children without associated liver disease.

Methods

We retrospectively reviewed the charts of 45 children who had surgery from 1979 to 2005 for EHPVO in our institution, consisting of 38 PSSS and 7 portal reperfusion procedures (PRPs). We assessed the presence of liver nodules on ultrasonography.

Results

Of 45 patients, 7 (15%) had liver nodules during a median of 80 months of follow-up. All the nodules occurred after PSSS. Five nodules were subjected to biopsy; we found 2 liver cell adenomas and 3 focal nodular hyperplasias.

Conclusions

In this study, liver nodules occurred in 18% of cases after PSSS for EHPVO in children and not after PRP. As many children have undergone PSSS throughout the world, the presence of liver nodules should be considered during the follow-up of those patients.     

Mesoportal bypass for extrahepatic portal vein obstruction in children: close to a cure for most!

Aim

Extrahepatic portal vein obstruction (EPVO) is a common cause of portal hypertension in children and can lead to life-threatening bleeding, thrombocytopenia, and coagulation disorders. Mesoportal bypass (MPB) restores normal physiologic portal flow to the liver and corrects portal hypertension. There is, however, little long-term outcome data after MPB.The aim of our study was to analyze the long-term outcome after MPB in children.

Methods

Retrospective single-center review of all MPB with more than 5-year follow-up was performed in children between 1998 and 2003.

Results

Thirty children underwent MPB, which was successful in 29. Long-term follow-up is available for 24. Median age at the time of bypass was 8.5 years (range, 0.4-14.2 years). Material used for bypass was as follows: left internal jugular vein (n = 20), recanalized umbilical vein (n = 2), gastric vein, and a large colic vein (n = 2). Median time since MPB is 8 years (range, 5.3-8.8 years).One MPB using recanalized umbilical vein thrombosed at 4 months but was successfully refashioned using Gortex. One MPB with left internal jugular vein was thrombosed at 1 year after cardiac surgery. Overall, 23 of 24 children have a patent bypass and resolution of portal hypertension. All showed an decrease in spleen size. Recurrent blood flow in the cavernoma with secondary increase in spleen size occurred in 5 children (median time, 3.4 years after MPB). Further evaluation of these 5 revealed no recurrence of portal hypertension.

Conclusion

The MPB provides long-term correction of portal hypertension owing to EPVO. Where technically feasible, MPB is the preferred surgical procedure for treatment of EPVO.     

Outcomes of liver transplantation in candidates with portal vein thrombosis

Objective

The objective of the present study was to analyze the incidence of portal vein thrombosis (PVT), comparing morbidity and mortality rates among those affected with and those free of this complication. In the PVT group, we also analyzed mortality related to partial (PPVT) and total (TPVT) thrombosis.

Methods

We undertook a retrospective study of orthotopic liver transplantations from deceased donors in 617 recipients from January 1991 until October 2008. Recipients were classified according to whether they had PVT. In all cases, we considered age, sex, Model for End-stage Liver Disease score, Child-Pugh score, indication for transplantation, type of thrombosis, surgical technique blood product transfusion, and survival rate.

Results

There were 48 patients with PVT (7.78%) among 670 transplantations in 617 recipients in our institution. Concerning the type of thrombosis, 28 (58.3%) were partial and 20 (41.7%) total with complete occlusion of the portal vein lumen.

Conclusion

PVT in liver transplant candidates is a rare event (7.8%) that entails greater difficulty in the procedure, expressed as a longer operative time, greater consumption of blood products, and complex surgical techniques. The prognosis for these patients depends on the type of thrombosis: patients with TPVT showed a higher mortality, whereas those with PPVT had survival rates comparable to those of candidates with a permeable portal vein.     

Growth impairment in children with extrahepatic portal vein obstruction is improved by mesenterico-left portal vein bypass

Background

Extrahepatic portal vein obstruction (EHPVO) has been associated with growth impairment in children. We hypothesized that growth parameters improve after reversal of portal hypertension and restoration of mesenteric venous blood flow to the liver by the mesenterico-left portal vein bypass (MLPVB).

Methods

A retrospective review of 45 children with idiopathic EHPVO who underwent MLPVB between 1997 and 2007 and had follow-up data for analysis was carried out. Growth was assessed using SD scores (z scores) for height, weight, and body mass index (BMI) at the time of operation and at early (5-12 months) and late (13-24 months) follow-up.

Results

The mean height and weight of children with EHPVO was significantly lower than the general population before surgery. Mean BMI was also lower, although statistically insignificant. All parameters increased significantly after MLPVB as follows: height from −0.42 before surgery to −0.12 (P = .027) at 5 to 12 months and −0.14 (P = .026) at 13 to 24 months; weight from −0.49 before surgery to 0.03 (P < .001) at 5 to 12 months and 0.35 (P < .001) at 13 to 24 months; and BMI from −0.22 before surgery to 0.17 (P = .001) at 5 to 12 months and 0.48 (P < .001) at 13 to 24 months.

Conclusion

Restoration of portal blood flow to the liver by MLPVB improves growth in children with EHPVO.     

Balloon-occluded retrograde transvenous obliteration performed in a pediatric patient with isolated gastric fundal varices

Background

For giant gastric varices in association with portal hypertension, endoscopic treatment often is difficult. Although balloon-occluded retrograde transvenous obliteration (B-RTO) has been performed successfully in adult cirrhotic patients, there has been no report in pediatric patients.

Methods

A 10-year-old girl with biliary atresia (BA) who had been free of jaundice after hepatic portoenterostomy was detected to have isolated gastric fundal varices by routine endoscopy. They gradually enlarged up to 4 cm in diameter, showing a tense appearance, so prophylactic treatment was conducted. Magnetic resonance angiography showed the blood flow of the varices mainly drained by a large gastrorenal shunt. A balloon catheter was introduced into the gastrorenal shunt via the femoral vein and was inflated to occlude the outflow of the varices. Five percent ethanolamine oleate was injected into the varices, and the outflow occlusion was kept for more than 30 minutes. Extensive thrombosis was achieved by an additional embolotherapy after 17 months. Throughout the course, the patient has been doing well without bleeding or worsening of the liver function tests.

Conclusions

B-RTO for isolated gastric fundal varices has been performed safely in a pediatric patient and seems effective in reducing the variceal size and tension.