INTRODUÇÃO: O transplante hepático é o único tratamento efetivo para uma variedade de doenças hepáticas irreversíveis. No entanto, o número limitado de doadores pediátricos leva ao uso de enxertos hepáticos de doadores adultos, com necessidade de anastomoses vasculares mais complexas. Essas anastomoses tornam-se complicadas pela diferença no calibre dos vasos entre o doador e o receptor, resultando em alterações do fluxo sanguíneo, estenose da anastomose venosa ou arterial e trombose. Os efeitos para regeneração hepática decorrentes da privação do fluxo sanguíneo pela veia porta ou pela artéria hepática não estão completamente elucidados. Experimentalmente, quando um lobo do fígado não recebe o fluxo venoso portal, é observada atrofia deste segmento e hipertrofia do restante do órgão perfundido. Embora existam vários modelos experimentais para estudo da regeneração hepática, poucos são focados em animais em crescimento. Além disso, os efeitos regenerativos de drogas como o tacrolimus e a insulina precisam ser pesquisados, com o objetivo de encontrar um tratamento ideal para a insuficiência hepática ou um método de estimular a regeneração do fígado após ressecções ou transplantes parciais. O objetivo do presente estudo é descrever modelos de regeneração hepática em ratos em crescimento com: 1) ausência de fluxo hepático arterial e 2) redução do fluxo portal. Adicionalmente, o estudo avalia o efeito pró-regenerativo do tacrolimus e da insulina nesses modelos descritos. MÉTODOS: cento e vinte ratos (entre 50 e 100g de peso) foram divididos em 6 grupos, de acordo com o tipo de intervenção cirúrgica: Grupo 1, incisão abdominal sem intervenção hepática; Grupo 2, hepatectomia a 70%; Grupo 3, hepatectomia a 70% + estenose de veia porta; Grupo 4, hepatectomia a 70% + ligadura da artéria hepática; Grupo 5, hepatectomia a 70% + estenose de veia porta + insulina; Grupo 6, hepatectomia a 70% + estenose de veia porta + tacrolimus. Os animais dos grupos 1 ao 4 foram subdivididos em 5 subgrupos de acordo com o momento da morte: 1, 2, 3, 5 e 10 dias após a intervenção cirúrgica. Os animais dos grupos 5 e 6 foram subdividos em 2 subgrupos de acordo com o momento da morte: 2 e 10 dias após a intervenção cirúrgica. Os lobos hepáticos remanescentes foram submetidos à análise histomorfométrica, imuno-histoquímica e molecular. RESULTADOS: Verificou-se que no grupo com hepatectomia a 70% houve recuperação do peso do fígado no terceiro dia com aumento da atividade mitótica, enquanto que no grupo com estenose portal não se observou esse fenômeno (p < 0,001). A insulina e o tacrolimus promoveram aumento do peso do fígado e do índice mitótico. A atividade mitótica foi considerada aumentada nos animais dos grupos hepatectomia, hepatectomia + ligadura da artéria, insulina e tacrolimus; e esse parâmetro estava reduzido no grupo submetido à hepatectomia + estenose portal (p < 0,001). A expressão de interleucina 6 estava presente em todos os animais, sendo significativamente maior nos grupos hepatectomia, hepatectomia + ligadura da artéria e significativamente menor no grupo hepatectomia + estenose portal. Entretanto, a administração de tacrolimus ou insulina recuperou os níveis teciduais de interleucina 6 no grupo com estenose portal. CONCLUSÕES: No presente estudo foi padronizado um modelo simples e facilmente reprodutível para estudar a regeneração hepática em ratos em crescimento com redução do fluxo arterial ou venoso para o fígado. Foi demonstrado que a administração de insulina ou tacrolimus é capaz de reverter os efeitos deletérios da estenose portal na regeneração hepática. A obstrução do fluxo arterial não afetou a capacidade regenerativa hepática / BACKGROUND/PURPOSE: Liver transplantation is an effective treatment for a variety of irreversible liver diseases. However, the limited number of pediatric donor livers leads to the use of adult livers, which usually require more complex vascular anastomoses. These anastomoses are complicated by differences in vessel caliber between donors and recipients, resulting in vascular flow anomalies, stenosis of the venous or arterial anastomosis and thrombosis . The effects of portal vein or hepatic arterial flow privation in hepatic regeneration have not been completely elucidated. Experimentally, when a liver lobe is deprived of portal vein flow, atrophy is observed with hypertrophy of the other perfused parts of the organ, and interleukin-6 (IL-6) is required for normal liver regeneration. Although several experimental models are currently used to study the liver regeneration mechanisms, few studies have focused on the growing animal. In addition, the regenerative effects of drugs (e.g., tacrolimus and insulin) have been experimentally studied, aiming to find an ideal treatment for hepatic failure or a method of stimulating liver regeneration after extensive resection or partial transplants. The aim of the present investigation was to describe the new models of liver regeneration in growing rats with: 1) absence of arterial blood hepatic inflow and 2) reduced portal flow. Additionally, it was studied whether tacrolimus or insulin could have any pro-regenerative effect under such conditions. METHODS/MATERIALS: one hundred and twenty rats (50-100 g body weight) were divided into 6 groups based on the intervention type: Group 1 (sham), abdominal incision without intervention; Group 2, 70% hepatectomy; Group 3, 70% hepatectomy + portal vein stenosis; Group 4, 70% hepatectomy + ligation of the hepatic artery; Group 5, 70% hepatectomy + portal vein stenosis + insulin; and Group 6, 70% hepatectomy + portal vein stenosis + tacrolimus. Animals in groups 1 to 4 were subdivided into 5 groups according to the moment of death: 1, 2, 3, 5 and 10 days after surgical intervention. The animals in groups 5 and 6 were subdivided into 2 other groups according to the moment of death: 2 and 10 days after surgical intervention. The remnant liver lobes were harvested for morphological, histological histomorphometric, immunohistochemical and molecular analyses. RESULTS: it was verified that the hepatectomy group regained liver weight on the third day and had increased mitotic activity, and the portal vein stenosis prevented these phenomena, as well as the increased mitotic index (P < 0.001). In addition, insulin and tacrolimus promoted a significant increase of liver weight. Mitotic activity was considerably increased in the hepatectomy, hepatectomy + arterial ligature, insulin and tacrolimus groups and this parameter was reduced by portal vein stenosis. The expression of the interleukin-6 (IL-6) gene was present in all the animal groups. Tissue levels of IL- 6 were significantly increased by hepatectomy and hepatectomy + hepatic artery ligature; portal vein stenosis prevented this change. However, the administration of tacrolimus or insulin could recuperate the tissue levels of IL-6. CONCLUSION: In the present study a simple and highly reproducible model was standardized to study liver regeneration with portal vein or hepatic artery blood inflow reduction in growing rats. It was demonstrated that insulin or tacrolimus administration may partially reverse the harmful effects of portal vein stenosis. The obstruction of the arterial flow did not affect liver regeneration

INTRODUÇÃO: O transplante hepático é o único tratamento efetivo para uma variedade de doenças hepáticas irreversíveis. No entanto, o número limitado de doadores pediátricos leva ao uso de enxertos hepáticos de doadores adultos, com necessidade de anastomoses vasculares mais complexas. Essas anastomoses tornam-se complicadas pela diferença no calibre dos vasos entre o doador e o receptor, resultando em alterações do fluxo sanguíneo, estenose da anastomose venosa ou arterial e trombose. Os efeitos para regeneração hepática decorrentes da privação do fluxo sanguíneo pela veia porta ou pela artéria hepática não estão completamente elucidados. Experimentalmente, quando um lobo do fígado não recebe o fluxo venoso portal, é observada atrofia deste segmento e hipertrofia do restante do órgão perfundido. Embora existam vários modelos experimentais para estudo da regeneração hepática, poucos são focados em animais em crescimento. Além disso, os efeitos regenerativos de drogas como o tacrolimus e a insulina precisam ser pesquisados, com o objetivo de encontrar um tratamento ideal para a insuficiência hepática ou um método de estimular a regeneração do fígado após ressecções ou transplantes parciais. O objetivo do presente estudo é descrever modelos de regeneração hepática em ratos em crescimento com: 1) ausência de fluxo hepático arterial e 2) redução do fluxo portal. Adicionalmente, o estudo avalia o efeito pró-regenerativo do tacrolimus e da insulina nesses modelos descritos. MÉTODOS: cento e vinte ratos (entre 50 e 100g de peso) foram divididos em 6 grupos, de acordo com o tipo de intervenção cirúrgica: Grupo 1, incisão abdominal sem intervenção hepática; Grupo 2, hepatectomia a 70%; Grupo 3, hepatectomia a 70% + estenose de veia porta; Grupo 4, hepatectomia a 70% + ligadura da artéria hepática; Grupo 5, hepatectomia a 70% + estenose de veia porta + insulina; Grupo 6, hepatectomia a 70% + estenose de veia porta + tacrolimus. Os animais dos grupos 1 ao 4 foram subdivididos em 5 subgrupos de acordo com o momento da morte: 1, 2, 3, 5 e 10 dias após a intervenção cirúrgica. Os animais dos grupos 5 e 6 foram subdividos em 2 subgrupos de acordo com o momento da morte: 2 e 10 dias após a intervenção cirúrgica. Os lobos hepáticos remanescentes foram submetidos à análise histomorfométrica, imuno-histoquímica e molecular. RESULTADOS: Verificou-se que no grupo com hepatectomia a 70% houve recuperação do peso do fígado no terceiro dia com aumento da atividade mitótica, enquanto que no grupo com estenose portal não se observou esse fenômeno (p < 0,001). A insulina e o tacrolimus promoveram aumento do peso do fígado e do índice mitótico. A atividade mitótica foi considerada aumentada nos animais dos grupos hepatectomia, hepatectomia + ligadura da artéria, insulina e tacrolimus; e esse parâmetro estava reduzido no grupo submetido à hepatectomia + estenose portal (p < 0,001). A expressão de interleucina 6 estava presente em todos os animais, sendo significativamente maior nos grupos hepatectomia, hepatectomia + ligadura da artéria e significativamente menor no grupo hepatectomia + estenose portal. Entretanto, a administração de tacrolimus ou insulina recuperou os níveis teciduais de interleucina 6 no grupo com estenose portal. CONCLUSÕES: No presente estudo foi padronizado um modelo simples e facilmente reprodutível para estudar a regeneração hepática em ratos em crescimento com redução do fluxo arterial ou venoso para o fígado. Foi demonstrado que a administração de insulina ou tacrolimus é capaz de reverter os efeitos deletérios da estenose portal na regeneração hepática. A obstrução do fluxo arterial não afetou a capacidade regenerativa hepática / BACKGROUND/PURPOSE: Liver transplantation is an effective treatment for a variety of irreversible liver diseases. However, the limited number of pediatric donor livers leads to the use of adult livers, which usually require more complex vascular anastomoses. These anastomoses are complicated by differences in vessel caliber between donors and recipients, resulting in vascular flow anomalies, stenosis of the venous or arterial anastomosis and thrombosis . The effects of portal vein or hepatic arterial flow privation in hepatic regeneration have not been completely elucidated. Experimentally, when a liver lobe is deprived of portal vein flow, atrophy is observed with hypertrophy of the other perfused parts of the organ, and interleukin-6 (IL-6) is required for normal liver regeneration. Although several experimental models are currently used to study the liver regeneration mechanisms, few studies have focused on the growing animal. In addition, the regenerative effects of drugs (e.g., tacrolimus and insulin) have been experimentally studied, aiming to find an ideal treatment for hepatic failure or a method of stimulating liver regeneration after extensive resection or partial transplants. The aim of the present investigation was to describe the new models of liver regeneration in growing rats with: 1) absence of arterial blood hepatic inflow and 2) reduced portal flow. Additionally, it was studied whether tacrolimus or insulin could have any pro-regenerative effect under such conditions. METHODS/MATERIALS: one hundred and twenty rats (50-100 g body weight) were divided into 6 groups based on the intervention type: Group 1 (sham), abdominal incision without intervention; Group 2, 70% hepatectomy; Group 3, 70% hepatectomy + portal vein stenosis; Group 4, 70% hepatectomy + ligation of the hepatic artery; Group 5, 70% hepatectomy + portal vein stenosis + insulin; and Group 6, 70% hepatectomy + portal vein stenosis + tacrolimus. Animals in groups 1 to 4 were subdivided into 5 groups according to the moment of death: 1, 2, 3, 5 and 10 days after surgical intervention. The animals in groups 5 and 6 were subdivided into 2 other groups according to the moment of death: 2 and 10 days after surgical intervention. The remnant liver lobes were harvested for morphological, histological histomorphometric, immunohistochemical and molecular analyses. RESULTS: it was verified that the hepatectomy group regained liver weight on the third day and had increased mitotic activity, and the portal vein stenosis prevented these phenomena, as well as the increased mitotic index (P < 0.001). In addition, insulin and tacrolimus promoted a significant increase of liver weight. Mitotic activity was considerably increased in the hepatectomy, hepatectomy + arterial ligature, insulin and tacrolimus groups and this parameter was reduced by portal vein stenosis. The expression of the interleukin-6 (IL-6) gene was present in all the animal groups. Tissue levels of IL- 6 were significantly increased by hepatectomy and hepatectomy + hepatic artery ligature; portal vein stenosis prevented this change. However, the administration of tacrolimus or insulin could recuperate the tissue levels of IL-6. CONCLUSION: In the present study a simple and highly reproducible model was standardized to study liver regeneration with portal vein or hepatic artery blood inflow reduction in growing rats. It was demonstrated that insulin or tacrolimus administration may partially reverse the harmful effects of portal vein stenosis. The obstruction of the arterial flow did not affect liver regeneration

Upper gastrointestinal bleeding episodes from variceal structures are severe complications in patients with portal hypertension. Endoscopic sclerotherapy and variceal ligation are the treatment options preferred for upper variceal bleeding owing to extrahepatic portal hypertension due to portal vein thrombosis (PVT). Recurrent duodenal variceal bleeding in non-cirrhotic patients with diffuse porto-splenic vein thrombosis and subsequent portal. cavernous transformation represent a clinical challenge if classic shunt surgery is not possible or suitable. In this study, we represent a case of recurrent bleeding of duodenal varices in a non-cirrhotic patient with cavernous transformation of the portal vein that was successfully treated with a collateral caval shunt operation.

Upper gastrointestinal bleeding episodes from variceal structures are severe complications in patients with portal hypertension. Endoscopic sclerotherapy and variceal ligation are the treatment options preferred for upper variceal bleeding owing to extrahepatic portal hypertension due to portal vein thrombosis (PVT). Recurrent duodenal variceal bleeding in non-cirrhotic patients with diffuse porto-splenic vein thrombosis and subsequent portal. cavernous transformation represent a clinical challenge if classic shunt surgery is not possible or suitable. In this study, we represent a case of recurrent bleeding of duodenal varices in a non-cirrhotic patient with cavernous transformation of the portal vein that was successfully treated with a collateral caval shunt operation.

Hintergrund: Ziel der Studie war der Vergleich von Patienten mit duktalem Pankreaskarzinom (PDAC) im progressionsfreien und Gesamtüberleben sowie im Rezidivmuster in Abhängigkeit einer Resektion der Vena portae oder der Vena mesenterica superior (PV/SMV). Methoden: Es wurde eine retrospektive Analyse durchgeführt. Hierbei wurden Patienten betrachtet, die zwischen 2005 und 2015 eine pyloruserhaltende partielle Pankreatoduodenektomie (PPPD), eine klassische Pankreatoduodenektomie (kPD) oder eine totale Pankreatektomie (TP) erhielten. Diese wurden in drei Gruppen eingeteilt. Die P+I+- Gruppe bestand aus Patienten mit Venenresektion (P+), bei denen eine pathohistologische Infiltration der PV oder SMV vorlag (I+). Fand sich bei durchgeführter Venenresektion keine portalvenöse Infiltration (I-), wurden die Patienten der P+I--Gruppe zugeordnet. Als Kontrollgruppe galten Patienten ohne Venenresektion (P-I-), welche zu denen der P+I+- Gruppe gematcht wurden. Die statistischen Analysen wurden mit dem R Softwarepaket durchgeführt. Das Signifikanzlevel wurde für alle Berechnungen auf α = 0,05 festgelegt. Ergebnisse: Insgesamt wurden 179 Patienten eingeschlossen. 113 erhielten eine portalvenöse Resektion. Davon hatten 36 (31,9%) eine pathohistologische Lumeninfiltration (P+I+), bei 77 (68,1%) lag dagegen keine Infiltration vor (P+I-). 66 Patienten ohne Venenresektion wurden zu den Patienten der P+I+-Gruppe gematcht (P-I-). Zwischen den drei Gruppen waren die meisten pathohistologischen Parameter vergleichbar. 17 Patienten (9,5%) wurden neoadjuvant therapiert, davon erhielten 16 eine Venenresektion (P+). Für das Gesamtüberleben konnten signifikante Unterschiede nachgewiesen werden (11,9 Monate [P+I+] vs. 16,1 Monate [P+I-] vs. 20,1 Monate [P-I-]; p=0,01). In der univariaten Überlebensanalyse konnte für den erhöhten präoperativen CA19-9 Wert, den Resektionsstatus (R), den Lymphknotenstatus (N), das Lymphknotenverhältnis (LNR), die mikroskopische Veneninvasion (V) sowie die pathohistologisch gesicherte Infiltration der PV/SMV ein negativer Einfluss nachgewiesen werden. In der multivariaten Analyse blieb die wahre Infiltration der PV/SMV als einziger signifikanter negativer Einflussfaktor auf das Gesamtüberleben erhalten (p=0,014). Die Inzidenz an Fernmetastasen war in der P+I+- Gruppe signifikant erhöht (75% [P+I+] vs. 45,8% [P+I-] vs. 54,7% [P-I-], p=0,01). Für ein Lokalrezidiv fanden sich dagegen keine Häufigkeitsunterschiede zwischen den Gruppen (p=0,96). Das mediane progressionsfreie Überleben war für Patienten der P+I+-Gruppe signifikant verkürzt (7,4 Monate [P+I+] vs. 10,9 Monate [P+I-] vs. 11,6 Monate [P-I-]; p=0,02). Die Lumeninfiltration der PV/SMV, die mikroskopische Veneninvasion (V), der präoperative CA19-9 Wert sowie der Differenzierungsgrad (G) waren negative Einflussfaktoren auf das progressionsfreie Überleben. In der multivariaten Analyse blieben die pathohistologisch gesicherte Infiltration sowie das Grading als negative unabhängige Einflussfaktoren nachweisbar. In 25% der Fälle manifestierte sich das Rezidiv initial in der Leber. Schlussfolgerung: Die pathohistologisch gesicherte Infiltration der PV/SMV ist ein unabhängiger Risikofaktor für das progressionsfreie und das Gesamtüberleben. Die Inzidenz an Fernmetastasen ist für die Patienten der P+I+-Gruppe erhöht. Eine potentiell kurative venöse Resektion kann den Einfluss der aggressiven Tumorbiologie und des fortgeschrittenen Krankheitsbildes nicht vollständig kompensieren. / Background. The present study aims to evaluate the longterm outcome and metastatatic pattern of patients who underwent an operation for pancreatic ductal adenocarcinoma (PDAC) with portal or superior mesenteric vein (PV/SMV) resection. Methods. Patients who underwent a pylorus preserving pancreaticoduodenectomy (PPPD), Whipple procedure (kPD) or total pancreatoduodenectomy (TP) between 2005 and 2015 were retrospectively analyzed. The patients were categorized in three subgroups. Those whom received a vein resection with pathohistological tumor invasion of the PV/SMV (P+I+) those at whom underwent vein resection but without pathohistological tumor invasion (P+I-) and lastly a third group (P-I-) matched to the P+I+ included patients without vein resection. Statistical analysis was performed using the R software package. The significance level for all calculations was set at α = 0.05. Results. The study cohort included 179 patients, 113 of whom underwent simultaneous PV/SMV resection. 36 patients (31,9%) had pathohistological tumor infiltration (P+I+), 77 (68,1%) did not (P+I-). 66 patients without vein resection (P-I-) were balanced by the P+I+ group. Most of pathohistological tumor characteristics were comparable between groups. 17 patients (9.5%) received neoadjuvant therapy, 16 of them were in vein resection group (P+). The study revealed differences in overall median survival (11.9 months [P+I+] vs. 16.1 months [P+I-] vs. 20.1 months [P-I-]; p=0.01). Univariate survival analysis shown negative consequences for CA19-9, resection margin (R), status of nodal metastasis (N), lymph node ratio (LNR), microvascular vein invasion (V) and true invasion of the PV/SMV. Multivariate survival analysis identified true invasion of the PV/SMV as the only significant, negative prognostic factor (p= 0.01). Whereas the incidence of local tumor recurrence was comparable (p=0.96), the proportion of patients with distant metastasis showed significant differences (75% [P+I+] vs. 45.8% [P+I-] vs 54.7% [P-I-]; p=0.01). The median time to progression were significantly shorter if the PV/SMV was infiltrated (7,4 months [P+I+] vs. 10,9 months [P+I-] vs. 11,6 months [P-I-]; p=0.02). Univariate progression analysis revealed significances for true invasion of the PV/SMV, microvascular vein invasion (V), CA19-9 and histologic classification (G). Multivariate progression analysis detected pathohistological invasion of the PV/SMV and histologic classification (G) as independent factors. Initial liver metastasis occurred in 25% of the patients. Conclusions. Pathohistological invasion of the PV/SMV is an independent risk factor for overall and progression free survival. Patients of P+I+-group had a higher incidence of distant metastasis, local progression is comparable. Even radical and complete resection cannot completely compensate for aggressive tumor biology and advanced disease. Modifiziert nach Mierke et al., 2016

Possibilities of Portal Vein Reconstruction during Surgical Treatment of Pancreatic Cancer - Experiment on a Large Animal Introduction: Pancreatic cancer is a fatal malignancy that is known as one of the leading causes of cancer mortality worldwide. The only potentially curative treatment is radical surgical resection. Because of the lack of early symptoms, the diagnosis is usually made in advanced stages of the disease. In the majority of patients, the tumor is already locally advanced or it has distant metastases at the time of diagnosis. Pancreatic cancer tends to infiltrate the portal vein (PV) or the superior mesenteric vein (SMV). Nowadays, resection of infiltrated parts of PV/SMV is recommended in specialized centers. There are several established techniques of PV/SMV reconstruction. The use of allogeneic venous grafts seems to be a method with minimal risk of adverse effects but there is only limited experience with these grafts. The optimal anatomical origin of allogeneic venous grafts for PV/SMV reconstruction remains unknown. Aims: The aim of this experiment was to compare two types of allogeneic venous grafts used for PV reconstruction in a large animal model of pancreatico- duodenectomy. These grafts were harvested from the systemic venous system (inferior caval vein grafts - IVC grafts) and...

INTRODUÇÃO: Nos pacientes pediátricos, em razão das limitações relacionadas ao tamanho dos receptores, são utilizados segmentos do fígado provenientes de \"cadáver\" ou de doador vivo (\"fígado reduzido\"). As complicações decorrentes das anastomoses cirúrgicas podem ser de natureza vascular e biliar, sendo que a maior causa de perda do enxerto deve-se à trombose ou estenose da artéria hepática, veia porta e veias hepáticas. A ultra-sonografia convencional e com Doppler colorido (US-DC) pode ser utilizada para avaliação dessas complicações. Todavia, apesar da alta sensibilidade e especificidade deste método, há poucas descrições a respeito de parâmetros que podem ser utilizados para o estudo das alterações vasculares. OBJETIVO: Estabelecer parâmetros de ultra-sonografia convencional e com Doppler colorido (US-DC) para o diagnóstico de estenose da VP e VH no transplante hepático reduzido em crianças. MÉTODO: Estudo retrospectivo de 134 US-DC realizado em 48 crianças submetidas a transplante hepático segmentar no Instituto da Criança do Hospital das Clinicas da Faculdade de Medicina da Universidade de São Paulo (ICr-HC-FMUSP), entre janeiro de 2002 e julho de 2005. No estudo da VP, os seguintes parâmetros foram analisados: calibre da VP na anastomose; velocidade máxima na anastomose (VP1); velocidade máxima no segmento pré-anastomose (VP2) e relação entre as velocidades máximas na anastomose/préanastomose (VP1/VP2). No estudo da VH, foram analisados: velocidade máxima na anastomose (VH1); velocidade na VH (VH2) e relação entre as velocidades máximas na anastomose/VH (VH1/VH2). Pacientes com estenose confirmada pela angiografia foram incluídos no grupo estenose e pacientes com angiografia ou cirurgia normal ou com boa evolução clínica foram incluídos grupo controle. RESULTADOS: Quatorze US-DC tiveram estenose da VP confirmadas pela portografia. O calibre da VP na anastomose foi menor no grupo estenose do que no grupo controle (média 2,5mm e 6,3mm respectivamente); PV1 e PV1/PV2 foram maiores no grupo estenose do que no grupo controle (média PV1 = 172cm/s, PV1/PV2 = 5,1 / PV1 = 83cm/s, PV1/PV2 = 1,8 respectivamente). O calibre da VP < 3,3mm apresentou a melhor correlação com a portografia (sensibilidade = 93% e especificidade = 88%), seguidos de VP1 > 128cm/s (sensibilidade = 86% e especificidade = 84%) e VP1/VP2 > 2,4 (sensibilidade = 79% e especificidade = 86%). Doze US-DC tiveram estenose da VH confirmada pela angiografia. A VH1 e HV1/VH2 foram maiores no grupo estenose do que no grupo controle (média VH1 = 202.2cm/s, VH1/VH2 = 6,0 / VH1 = 136,8cm/s, VH1/VH2 = 3,0 respectivamente). A relação VH1/VH2 > 4 apresentou a melhor correlação com a angiografia (sensibilidade = 83% e especificidade = 84%) seguido de VH1 > 128cm/s (sensibilidade = 83% e especificidade = 52%). CONCLUSÃO: Calibre da VP < 3,3mm na anastomose e relação das velocidades VH1/VH2 > 4 são altamente sensíveis e específicos no diagnóstico das estenoses da anastomose da VP e VH respectivamente, no póstransplante hepático em crianças / INTRODUCTION: Cadáveric split liver and living donor liver transplantation is mostly performed in children because of a shortage of suitable hepatic allograft in this population. Real time and Doppler ultrasound (CD-US) is the initial imaging modality for detection and follow-up of early and delayed vascular and non-vascular complications after liver transplant, but there are only few studies concerning its value in the diagnosis of venous complications. OBJECTIVE: Determine real time and Doppler ultrasound (CD-US) diagnostic parameters of portal (PV) and hepatic vein (HV) stenoses after segmental liver transplantation in children and assess their sensitivity and specificity. METHOD: Retrospective study of 134 CD-US performed in 48 patients submitted to segmental liver transplantation at Child Institute of University of Sao Paulo Medical School from January 2002 through July 2005. PV parameters: diameter at the anastomosis, velocities at the anastomosis (PV1) and at the pre-anastomosis segments (PV2) and the ratio PV1/PV2. HV parameters: velocities at the HV anastomosis to the inferior vena cava (HV1), at HV (HV2) and the ratio HV1/HV2. Stenosis group: patients with stenosis confirmed by angiography. Control group: patients presenting normal angiography or uneventhfull surgery or good clinical outcome. RESULTS: Fourteen CD-US had PV stenosis confirmed by portography. Diameter of PV at the anastomosis: narrower in the stenosis group (2.5mm) than in the control group (6.3mm) (p<.5). Mean PV1 and PV1/PV2: higher in the stenosis group than in the control group (PV1 = 172 cm/s, PV1/PV2 = 5.1/PV1 = 83cm/s, PV1/PV2 = 1.8). Statistical analysis determined as predictive of PV stenosis: PV diameter < 3.3mm (sensitivity of 93% and specificity of 88.4%); PV1 > 128cm/s (sensitivity of 86% and specificity of 84%) and PV1/PV2 > 2.4 (sensitivity of 79% and specificity of 86%). Twelve CD-US had HV stenosis confirmed by angiography. Mean HV1 and HV1/HV2: higher in the study group (HV1 = 202.2cm/s, HV1/HV2 = 6.0 / HV1 = 136.8cm/s, HV1/HV2 = 3.0) (p<.05). Statistical analysis determined as predictive of HV stenosis: PV1 > 128cm/s (sensitivity of 83% and specificity of 52%) and HV1/HV2 > 4 (sensitivity of 83% and specificity of 84%). CONCLUSION: PV diameter < 3.3mm at the anastomosis and HV1/HV2 > 4.0 are highly predictive for detection of PV and HV stenosis respectively, after pediatric segmental liver transplantation

La transplantation d’hépatocytes est un procédé séduisant pour remplacer les cellules déficientes dans un foie anatomiquement normal. Dans les maladies métaboliques héréditaires hépatiques (MMHH), la thérapie cellulaire présente un potentiel espoir thérapeutique. Le remplacement d'un pourcentage restreint (5-10%) d’hépatocytes déficients par des hépatocytes normaux pourrait rétablir durablement la fonction métabolique. Les résultats des essais cliniques de transplantation d'hépatocytes génétiquement modifiés ou non sont moins concluants, montrent une prise de greffe insuffisante et, dans la plupart des études, un effet thérapeutique transitoire. L’efficacité limitée de la transplantation d’hépatocytes isolés dans le traitement des MMHH semble en partie liée au faible pourcentage de la masse hépatocytaire reconstituée par les hépatocytes définitivement greffés et fonctionnels. De nombreux modèles animaux ont été développés pour étudier les facteurs pouvant augmenter le nombre et le pourcentage d’hépatocytes transplantés et greffés. Cependant, la majorité de ces modèles ne sont pas transposables en clinique car ils présentent des risques importants ou mal évalués pour les patients. Les principaux objectifs de ce travail ont été d’étudier des moyens peu invasifs pour induire une régénération hépatique et une prise de greffe hépatocytaire significatives dans le but de développer une nouvelle approche de transplantation d’hépatocytes génétiquement modifiés ex vivo pour le traitement de l’hypercholestérolémie familiale. L’effet d’une embolisation portale partielle (EPP) réversible sur la prolifération hépatocytaire et la régénération hépatique a été évalué chez le macaque. A la différence de l’EPP par un produit non résorbable, il ne s’agit pas d’une approche potentiellement délétère à long terme. Une obstruction veineuse plus complète a été provoquée en utilisant le Curaspon®, une gélatine biodégradable, en forme de poudre. Nous avons démontré pour la première fois dans la littérature l’efficacité d’une EPP réversible à induire une importante prolifération hépatocytaire et régénération hépatique. Nos données suggèrent qu’une occlusion portale initiale et temporaire est suffisante pour déclencher les mécanismes responsables d’une régénération hépatique dans le foie non-embolisé. L’utilisation du Curaspon® en poudre peut être considérée comme la forme la plus évoluée d’EPP : très distale, résorbable, qui dure suffisamment pour induire l’hypertrophie hépatique. Cette technique pourrait être indiquée dans des situations cliniques nécessitant une régénération hépatique de courte durée (ex. le traitement des cancers du foie en plusieurs étapes) ou dans des cas qui ne nécessitent pas une résection hépatique, comme la transplantation d’hépatocytes pour le traitement de MMHH. Ces résultats nous ont permis d’évaluer cette approche dans notre protocole préclinique de thérapie génique pour le traitement de l’hypercholestérolémie familiale chez le primate, par autotransplantation d’hépatocytes génétiquement modifiés ex vivo par un vecteur lentiviral. Nous avons démontré que l’EPP réversible induit une régénération hépatique du foie non-embolisé et améliore notablement les résultats de la transplantation d’hépatocytes isolés génétiquement modifiés exprimant la GFP. Seize semaines après la transplantation, les hépatocytes transduits et greffés exprimaient le transgène contrôlé par le promoteur apo-AII humain. Notre protocole a montré pour la première fois chez un gros animal que l’EPP par un produit résorbable entraine avec des conditions de sécurité optimales une repopulation hépatique importante par des hépatocytes transduits par un vecteur lentiviral, et ceci même à distance de la transplantation hépatocytaire. Les résultats encourageants de ces travaux nous ont ouvert la voie pour avancer sur notre projet préclinique et envisager la réalisation d’une étude clinique de phase I/II pour le traitement de l’hypercholestérolémie familiale. / Hepatocyte transplantation is an attractive process for replacing deficient cells in an anatomically normal liver. In metabolic liver diseases, cell therapy could be an interesting alternative to orthotopic liver transplantation. The replacement of a small percentage (5-10%) of deficient hepatocytes by normal hepatocytes could restore the metabolic defect at a long term. Data from clinical studies of hepatocyte autotransplantation or allotransplantation, genetically modified or not, provided poor results, insufficient cell engraftment in the liver parenchyma and, in the majority of cases, a transient therapeutic effect. The limited efficacy of hepatocyte transplantation in metabolic liver diseases is mainly due to the poor percentage of engrafted and finally functional hepatocytes. Numerous animal models have been developed in order to study the factors that could increase the number and the percentage of transplanted and engrafted hepatocytes. However, the majority of these models cannot be used in patients since they present important risks for them. The aim of this work was to evaluate less invasive procedures for inducing liver regeneration and significant hepatocyte engraftment in order to develop a new approach of transplantation of ex vivo genetically modified hepatocytes for the treatment of familial hypercholesterolemia. The effect of reversible portal vein embolization (PVE) on liver regeneration and hepatocyte proleferation was evaluated in monkeys. In contrast to PVE by a permanent embolizing agent, reversible PVE has not a long term deleterious effect on embolized liver. A more complete venous occlusion was obtained by using the powdered form of an absorbable gelatin sponge (Curaspon®). We showed for the first time in the literature the safe and successful use of reversible PVE for inducing significant hepatocyte proliferation and liver regeneration. Our data support that an initial occlusion of the portal branch, even if not permanent, is sufficient to start the mechanisms of liver regeneration in the contralateral lobe. Embolization with Curaspon® powder could be considered to be the ultimate form of embolization: very distal, reversible and lasting sufficiently in order to induce substantial liver hypertrophy. Our findings suggest that this method could reliably be used for clinical purposes, particularly in situations in which short-term regeneration is required (i.e. multi-step management of hepatic malignancies) or in cases where resection of the liver is not finally necessary, such as in hepatocyte transplantation for the treatment of metabolic liver diseases. These promising results on reversible PVE allowed us to evaluate this approach in our preclinical study of gene therapy for the treatment of familial hypercholesterolemia in macaques. Our protocol consisted of an autotransplantation of ex vivo genetically modified hepatocytes by a lentiviral vector. We showed that reversible PVE induces liver regeneration of the non-embolized liver segments and improves considerably hepatocyte transplantation of genetically modified cells expressing Green Fluorescent Protein (GFP). Sixteen weeks after transplantation, transduced engrafted hepatocytes expressed the transgene, which was under control of the human apo-AII promoter. Our protocol showed for the first time in a big animal that PVE by an absorbable agent leads safely to an important and long-term repopulation of the liver by lentivirally transduced hepatocytes. The extremely encouraging results of this work opened our way advancing in our preclinical study and preparing a phase I/II clinical trial for the treatment of familial hypercholesterolemia based on our protocol of autotransplantation of ex vivo genetically modified hepatocytes by a lentiviral vector.

Objetivo: A trombose da veia porta é uma causa importante de hiper-tensão porta em crianças e adolescentes, porém, em uma proporção importante dos casos, não apresenta fator etiológico definido. O objetivo desse estudo é determinar a freqüência de deficiência das proteínas inibidoras da coagulação – proteínas C, S e antitrombina − e das mutações fator V Leiden, G20210A no gene da protrombina e C677T da metileno-tetraidrofolato redutase em crianças e adolescentes com trom-bose da veia porta, definir o padrão hereditário de uma eventual deficiência das pro-teínas inibidoras da coagulação nesses pacientes e avaliar a freqüência da deficiên-cia dessas proteínas em crianças e adolescentes com cirrose. Casuística e Métodos: Foi realizado um estudo prospectivo com 14 crianças e adolescentes com trombose da veia porta, seus pais (n = 25) e dois gru-pos controles pareados por idade, constituídos por um grupo controle sem hepato-patia (n = 28) e um com cirrose (n = 24). A trombose da veia porta foi diagnosticada por ultra-sonografia abdominal com Doppler e/ou fase venosa do angiograma celíaco seletivo. A dosagem da atividade das proteínas C, S e antitrombina foi determinada em todos os indivíduos e a pesquisa das mutações fator V Leiden, G20210A da pro-trombina e C677T da metileno-tetraidrofolato redutase, nas crianças e adolescentes com trombose da veia porta, nos pais, quando identificada a mutação na criança, e nos controles sem hepatopatia. Resultados: Foram avaliados 14 pacientes caucasóides, com uma média e desvio padrão de idade de 8 anos e 8 meses ± 4 anos e 5 meses e do diagnóstico de 3 anos e 8 meses ± 3 anos e seis meses. Metade dos pacientes pertenciam ao gênero masculino. O motivo da investigação da trombose da veia porta foi hemorra-gia digestiva alta em 9/14 (64,3%) e achado de esplenomegalia ao exame físico em 5/14 (35,7%). Anomalias congênitas extra-hepáticas foram identificadas em 3/14 (21,4%) e fatores de risco adquiridos em 5/14 (35,7%) dos pacientes. Nenhum pa-ciente tinha história familiar de consangüinidade ou trombose venosa. A deficiência das proteínas C, S e antitrombina foi constatada em 6/14 (42,9%) (p < 0,05 vs con-troles sem hepatopatia), 3/14 (21,4%) (p > 0,05) e 1/14 (7,1%) (p > 0,05) pacientes com trombose da veia porta, respectivamente. A deficiência dessas proteínas não foi identificada em nenhum dos pais ou controles sem hepatopatia. A mutação G20210A no gene da protrombina foi identificada em um paciente com trombose da veia porta e em um controle sem hepatopatia (p = 0,999), mas em nenhum desses foi identificado a mutação fator V Leiden. A mutação C677T da metileno-tetraidrofo-lato redutase foi observada na forma homozigota, em 3/14 (21,4%) dos pacientes com trombose da veia porta e em 5/28 (17,9%) controles sem hepatopatia (p = 0,356). A freqüência da deficiência das proteínas C, S e antitrombina nos pacientes com cir-rose foi de 14/24 (58,3%), 7/24 (29,2%) e 11/24 (45,8%), respectivamente (p < 0,05 vs controles sem hepatopatia), sendo mais freqüente nos pacientes do subgrupo Child-Pugh B ou C, que foi de 11/12 (91,7%), 5/12 (41,7%) e 9/12 (75%), respectivamente (p < 0,05 vs controles sem hepatopatia). Conclusões: A deficiência de proteína C foi freqüente nas crianças e adolescentes com trombose da veia porta e não parece ser de origem genética. A deficiência de proteína S, antitrombina e as presenças das mutações G20210A da protrombina e C677T da metileno-tetraidrofolato redutase foram observadas mas não apresentaram diferença estatística significativa em relação ao grupo controle sem hepatopatia. O fator V Leiden não foi identificado. Os resultados deste estudo sugerem que a deficiência da proteína C pode ocorre como conseqüência da hiper-tensão porta. Os distúrbios pró-trombóticos hereditários não parecem apresentar um papel importante em relação à trombose nas crianças e adolescentes estudadas. / Objective: Portal vein thrombosis is a major cause of portal hypertension in children and adolescents; yet, its etiology is not clearly defined in a considerable number of cases. The present study aims at determining the prevalence of blood coagulation disorders – protein C, protein S and antithrombin – and factor V Leiden, G20210A prothrombin, and C677T methylenetetrahydrofolate reductase mutations in children and adolescents with portal vein thrombosis, as well as assessing the hereditary character of these disorders in these patients, and also evaluating the prevalence of blood coagulation disorders in children and adolescents with cirrhosis. Study design: A prospective study was carried out, including children and adolescents with portal vein thrombosis (n = 14), their parents (n = 25), two age-matched control groups, one without liver disease (n = 28), and another with cirrhosis (n = 24). Portal vein thrombosis was diagnosed through abdominal Doppler ultrasonography and/or venous phase of selective coeliac angiograms. The activity of protein C, protein S and antithrombin was evaluated for all individuals; the presence of factor V Leiden, G20210A prothrombin, and C677T methylenetetrahydrofolate reductase gene mutations was investigated in children and adolescents with portal vein thrombosis, in parents when their respective children presented any of these mutations, and in the control group without liver disease. Results: 14 Caucasian patients were assessed. The mean and standard deviation for age were 8 years and 8 months ± 4 years and 5 months while the mean and standard deviation for diagnosis were 3 years and 8 months ± 3 years and six months. Half of the patients were males. Initial clinical manifestations upon diagnosis were digestive hemorrhage in 9/14 (64.3%) and splenomegaly on physical examination in 5/14 individuals (35.7%). Patients presented extrahepatic anomalies in 3/14 (21.4%) and acquired risk factors in 5/14 (35.7%) of the cases. None of the patients had a family history of consanguinity or venous thrombosis. The frequency of protein C, protein S and antithrombin deficiency was observed in 6/14 (42.9%) (p < 0.05 vs. controls without liver disease), 3/14 (21.4%) (p > 0.05) and 1/14 (7.1%) (p > 0.05) of patients, respectively. None of the portal vein thombosis patients or controls presented protein C, S or antithrombin deficiency. One portal vein patient and one control (p = 0.999) presented G20210A prothrombin mutation. None of these patients presented the factor V Leiden. The homozygous form of C677T methylenetetrahydrofolate reductase mutation was observed in 3/14 patients with portal vein thrombosis (21.4%) and in 5/28 controls (17.9%) (p = 0.356). The frequency of coagulation inhibitor deficiency was high in cirrhotic patients (14/24 (58.3%) PC, 7/24 (29.2%) PS and 11/24 (45.8%) AT; p < 0.05 vs. controls), especially in Child-Pugh B and C patients (11/12 (91.7%) PC, 5/12 (41.7%) PS and 9/12 (75%) AT; p < 0.05 vs. controls). Conclusions: Protein C deficiency was frequent in children and adolescents with portal vein thrombosis and does not seem to be an inherited condition. Protein S and antithrombin deficiency, and G20210A prothrombin and C677T methylenetetrahydrofolate reductase mutations were observed but did not present statistically significant differences when compared to the controls without liver disease. Factor V Leiden was not observed. The results suggest the protein C deficiency may originates from portal hypertension. The hereditary prothrombotic disorders do not seem to play a vital role in thrombosis in children and adolescents with portal vein thrombosis.

Objetivo: A trombose da veia porta é uma causa importante de hiper-tensão porta em crianças e adolescentes, porém, em uma proporção importante dos casos, não apresenta fator etiológico definido. O objetivo desse estudo é determinar a freqüência de deficiência das proteínas inibidoras da coagulação – proteínas C, S e antitrombina − e das mutações fator V Leiden, G20210A no gene da protrombina e C677T da metileno-tetraidrofolato redutase em crianças e adolescentes com trom-bose da veia porta, definir o padrão hereditário de uma eventual deficiência das pro-teínas inibidoras da coagulação nesses pacientes e avaliar a freqüência da deficiên-cia dessas proteínas em crianças e adolescentes com cirrose. Casuística e Métodos: Foi realizado um estudo prospectivo com 14 crianças e adolescentes com trombose da veia porta, seus pais (n = 25) e dois gru-pos controles pareados por idade, constituídos por um grupo controle sem hepato-patia (n = 28) e um com cirrose (n = 24). A trombose da veia porta foi diagnosticada por ultra-sonografia abdominal com Doppler e/ou fase venosa do angiograma celíaco seletivo. A dosagem da atividade das proteínas C, S e antitrombina foi determinada em todos os indivíduos e a pesquisa das mutações fator V Leiden, G20210A da pro-trombina e C677T da metileno-tetraidrofolato redutase, nas crianças e adolescentes com trombose da veia porta, nos pais, quando identificada a mutação na criança, e nos controles sem hepatopatia. Resultados: Foram avaliados 14 pacientes caucasóides, com uma média e desvio padrão de idade de 8 anos e 8 meses ± 4 anos e 5 meses e do diagnóstico de 3 anos e 8 meses ± 3 anos e seis meses. Metade dos pacientes pertenciam ao gênero masculino. O motivo da investigação da trombose da veia porta foi hemorra-gia digestiva alta em 9/14 (64,3%) e achado de esplenomegalia ao exame físico em 5/14 (35,7%). Anomalias congênitas extra-hepáticas foram identificadas em 3/14 (21,4%) e fatores de risco adquiridos em 5/14 (35,7%) dos pacientes. Nenhum pa-ciente tinha história familiar de consangüinidade ou trombose venosa. A deficiência das proteínas C, S e antitrombina foi constatada em 6/14 (42,9%) (p < 0,05 vs con-troles sem hepatopatia), 3/14 (21,4%) (p > 0,05) e 1/14 (7,1%) (p > 0,05) pacientes com trombose da veia porta, respectivamente. A deficiência dessas proteínas não foi identificada em nenhum dos pais ou controles sem hepatopatia. A mutação G20210A no gene da protrombina foi identificada em um paciente com trombose da veia porta e em um controle sem hepatopatia (p = 0,999), mas em nenhum desses foi identificado a mutação fator V Leiden. A mutação C677T da metileno-tetraidrofo-lato redutase foi observada na forma homozigota, em 3/14 (21,4%) dos pacientes com trombose da veia porta e em 5/28 (17,9%) controles sem hepatopatia (p = 0,356). A freqüência da deficiência das proteínas C, S e antitrombina nos pacientes com cir-rose foi de 14/24 (58,3%), 7/24 (29,2%) e 11/24 (45,8%), respectivamente (p < 0,05 vs controles sem hepatopatia), sendo mais freqüente nos pacientes do subgrupo Child-Pugh B ou C, que foi de 11/12 (91,7%), 5/12 (41,7%) e 9/12 (75%), respectivamente (p < 0,05 vs controles sem hepatopatia). Conclusões: A deficiência de proteína C foi freqüente nas crianças e adolescentes com trombose da veia porta e não parece ser de origem genética. A deficiência de proteína S, antitrombina e as presenças das mutações G20210A da protrombina e C677T da metileno-tetraidrofolato redutase foram observadas mas não apresentaram diferença estatística significativa em relação ao grupo controle sem hepatopatia. O fator V Leiden não foi identificado. Os resultados deste estudo sugerem que a deficiência da proteína C pode ocorre como conseqüência da hiper-tensão porta. Os distúrbios pró-trombóticos hereditários não parecem apresentar um papel importante em relação à trombose nas crianças e adolescentes estudadas. / Objective: Portal vein thrombosis is a major cause of portal hypertension in children and adolescents; yet, its etiology is not clearly defined in a considerable number of cases. The present study aims at determining the prevalence of blood coagulation disorders – protein C, protein S and antithrombin – and factor V Leiden, G20210A prothrombin, and C677T methylenetetrahydrofolate reductase mutations in children and adolescents with portal vein thrombosis, as well as assessing the hereditary character of these disorders in these patients, and also evaluating the prevalence of blood coagulation disorders in children and adolescents with cirrhosis. Study design: A prospective study was carried out, including children and adolescents with portal vein thrombosis (n = 14), their parents (n = 25), two age-matched control groups, one without liver disease (n = 28), and another with cirrhosis (n = 24). Portal vein thrombosis was diagnosed through abdominal Doppler ultrasonography and/or venous phase of selective coeliac angiograms. The activity of protein C, protein S and antithrombin was evaluated for all individuals; the presence of factor V Leiden, G20210A prothrombin, and C677T methylenetetrahydrofolate reductase gene mutations was investigated in children and adolescents with portal vein thrombosis, in parents when their respective children presented any of these mutations, and in the control group without liver disease. Results: 14 Caucasian patients were assessed. The mean and standard deviation for age were 8 years and 8 months ± 4 years and 5 months while the mean and standard deviation for diagnosis were 3 years and 8 months ± 3 years and six months. Half of the patients were males. Initial clinical manifestations upon diagnosis were digestive hemorrhage in 9/14 (64.3%) and splenomegaly on physical examination in 5/14 individuals (35.7%). Patients presented extrahepatic anomalies in 3/14 (21.4%) and acquired risk factors in 5/14 (35.7%) of the cases. None of the patients had a family history of consanguinity or venous thrombosis. The frequency of protein C, protein S and antithrombin deficiency was observed in 6/14 (42.9%) (p < 0.05 vs. controls without liver disease), 3/14 (21.4%) (p > 0.05) and 1/14 (7.1%) (p > 0.05) of patients, respectively. None of the portal vein thombosis patients or controls presented protein C, S or antithrombin deficiency. One portal vein patient and one control (p = 0.999) presented G20210A prothrombin mutation. None of these patients presented the factor V Leiden. The homozygous form of C677T methylenetetrahydrofolate reductase mutation was observed in 3/14 patients with portal vein thrombosis (21.4%) and in 5/28 controls (17.9%) (p = 0.356). The frequency of coagulation inhibitor deficiency was high in cirrhotic patients (14/24 (58.3%) PC, 7/24 (29.2%) PS and 11/24 (45.8%) AT; p < 0.05 vs. controls), especially in Child-Pugh B and C patients (11/12 (91.7%) PC, 5/12 (41.7%) PS and 9/12 (75%) AT; p < 0.05 vs. controls). Conclusions: Protein C deficiency was frequent in children and adolescents with portal vein thrombosis and does not seem to be an inherited condition. Protein S and antithrombin deficiency, and G20210A prothrombin and C677T methylenetetrahydrofolate reductase mutations were observed but did not present statistically significant differences when compared to the controls without liver disease. Factor V Leiden was not observed. The results suggest the protein C deficiency may originates from portal hypertension. The hereditary prothrombotic disorders do not seem to play a vital role in thrombosis in children and adolescents with portal vein thrombosis.