The High Incidence of Occult Carcinoma in Total Hepatectomy Specimens of Patients Treated for Unresectable Colorectal Liver Metastases With Liver Transplant.

OBJECTIVE: To describe the rate of occult carcinoma deposits in total hepatectomy specimens from patients treated with liver transplant (LT) for colorectal liver metastases (CRLM). BACKGROUND: Previous studies have shown that patients with CRLM treated with systemic therapy demonstrate a high rate of complete radiographic response or may have disappearing liver metastases. However, this does not necessarily translate into a complete pathologic response, and residual invasive cancer may be found in up to 80% of the disappearing tumors after resection. METHODS: Retrospective review of 14 patients who underwent LT for CRLM, at 2 centers. Radiographic and pathologic correlation of the number of tumors and their viability before and after LT was performed. RESULTS: The median (interquartile range) number of tumors at diagnosis was 11 (4-23). The median number of chemotherapy cycles was 24 (16-37). Hepatic artery infusion was used in 5 patients (35.7%); 6 (42.9%) underwent surgical resection, and 5 (35.7%) received locoregional therapy. The indication for LT was unresectability in 8 patients (57.1%) and liver failure secondary to oncologic treatment in the remaining 6 (42.9%). Before LT, 7 patients (50%) demonstrated fluorodeoxyglucose-avid tumors and 7 (50%) had a complete radiographic response. Histopathologically, 11 patients (78.6%) had a viable tumor. Nine (64.2%) of the 14 patients were found to have undiagnosed metastases on explant pathology, with at least 22 unaccounted viable tumors before LT. Furthermore, 4 (57.1%) of the 7 patients who demonstrated complete radiographic response harbored viable carcinoma on explant pathology. CONCLUSIONS: A complete radiographic response does not reliably predict a complete pathologic response. In patients with unresectable CRLM, total hepatectomy and LT represent a promising treatment options to prevent indolent disease progression from disappearing CRLM.

Early postoperative ERAS compliance predicts decreased length of stay and complications following liver resection.

BACKGROUND: Enhanced recovery after surgery (ERAS) components for liver resection lack standardization and compliance. We evaluated our ERAS protocol and describe the association of postoperative ERAS compliance with length of stay (LOS) and complications. METHODS: We retrospectively reviewed patients undergoing liver resection at our institution from 2016 to 2020. Pre- and post-ERAS outcomes and compliance at 72 h were compared with LOS and complications. LOS beyond 72 h was defined as LOS72. RESULTS: 210 patients were included. Post-ERAS patients had significantly shorter LOS (5.1 vs. 7.3 days, p = 0.0014) with no difference in 30-day mortality, morbidity, or readmissions. ERAS components associated with shorter LOS72 were regular diet (HR 1.73), fluid discontinuation (HR 1.63), drain removal (HR 1.94), multimodal and oral analgesia (HR 1.51), and ambulation >100 ft (HR 2.23). LOS72 was 1-day for ≥9 ERAS component compliance, 4-days for 6-8 components, and 6-days for <6 components. 30-day complication rates for patients with ≥9 components by postoperative day 3 (POD3) were significantly lower than those with 6-8 (12 vs 32%). CONCLUSION: ERAS decreases LOS after liver resection. Nutritional advancement, drain discontinuation, multimodal and oral analgesia, and ambulation >100 ft by POD3 are associated with decreased LOS72. Achieving ≥6 components by POD3 predicts decreased LOS72 and complications.

The meaning of confounding adjustment in the presence of multiple versions of treatment: an application to organ transplantation.

Causal inference for treatments with many versions requires a careful specification of the versions of treatment. Specifically, the existence of multiple relevant versions of treatment has implications for the selection of confounders. To illustrate this, we estimate the effect of organ transplantation using grafts from donors who died due to anoxic drug overdose, on recipient graft survival in the US. We describe how explicitly outlining the target trial (i.e. the hypothetical randomized trial which would answer the causal question of interest) to be emulated by an observational study analysis helps conceptualize treatment versions, guides selection of appropriate adjustment variables, and helps clarify the settings in which causal effects of compound treatments will be of value to decision-makers.

Survey of Living Organ Donors' Experience and Directions for Process Improvement.

INTRODUCTION: Understanding living organ donors' experience with donation and challenges faced during the process is necessary to guide the development of effective strategies to maximize donor benefit and increase the number of living donors. METHODS: An anonymous self-administered survey, specifically designed for this population based on key informant interviews, was mailed to 426 individuals who donated a kidney or liver at our institution. Quantitative and qualitative methods including open and axial coding were used to analyze donor responses. FINDINGS: Of the 141 survey respondents, 94% would encourage others to become donors; however, nearly half (44%) thought the donation process could be improved and offered numerous suggestions. Five major themes arose: (1) desire for greater convenience in testing and scheduling; (2) involvement of previous donors throughout the process; (3) education and promotion of donation through social media; (4) unanticipated difficulties, specifically pain; and (5) financial concerns. DISCUSSION: Donor feedback has been translated into performance improvements at our hospital, many of which are applicable to other institutions. Population-specific survey development helps to identify vital patient concerns and provides valuable feedback to enhance the delivery of care.

Coronavirus disease 2019-related hepatic transplant pseudoaneurysm.

Coronavirus disease 2019-related transplant hepatic pseudoaneurysms have not been reported but can be life-threatening. They can be either solitary or multiple and can grow rapidly within weeks. They should be classified as mycotic and treated on an emergent basis. Both stenting of the vessel and coil embolization can potentially be viable treatment options of coronavirus disease 2019-related pseudoaneurysms.

HCV core and NS3 proteins manipulate human blood-derived dendritic cell development and promote Th 17 differentiation.

Hepatitis C virus (HCV) chronic infection is characterized by low-level or undetectable cellular immune response against HCV antigens. HCV proteins affect various intracellular events and modulate immune responses, although the mechanisms that mediate these effects are not fully understood. In this study, we examined the effect of HCV proteins on the differentiation of human peripheral blood monocytes to dendritic cells (DCs). The HCV core (HCVc) and non-structural 3 (NS3) proteins inhibited the expression of CD1a, CD1b and DC-SIGN during monocyte differentiation to DCs, while increasing some markers characteristic of macrophages (CD14 and HLA-DR) and also PD-L1 expression. Meanwhile, HCVc and NS3 could induce differentiating monocytes to secrete IL-10. However, anti-IL-10 mAb could not reverse HCVc and NS3 inhibition of monocyte differentiation into DCs. The HCVc and NS3 proteins increased IL-6 secretion both in immature and in fully differentiated DCs and also promoted CD4+ T-cell IL-17 production. Since T(h) 17 cells are active in many examples of immunopathology, these effects may contribute to HCV autoimmune responses in chronically infected patients.

Diabetes mellitus impacts risk of macrovascular invasion in patients undergoing transplantation for hepatocellular carcinoma.

BACKGROUND: Diabetes mellitus (DM) is identified as a negative prognostic indicator in hepatocellular carcinoma (HCC), though the basis for this is unknown. METHODS: This is a retrospective analysis of a prospectively collected database of 191 HCC patients treated at the University of Rochester Medical Center (URMC) with orthotopic liver transplantation between 1998-2008. Clinical characteristics were compared between patients with and without DM prior to liver transplantation and logistic regression analyses were conducted to assess the effect of DM on clinical outcomes including vascular invasion. RESULTS: Eighty-four of 191 (44%) transplanted patients had DM at time of transplantation. An association of DM with invasive disease was found among transplanted HCC patients where histologically confirmed macrovascular invasion was found in 20.2% (17/84) of diabetics compared to 9.3% of non-diabetics (10/107) (p=0.032). This difference also remained significant when adjusting for tumor size, number of nodules, age, obesity and etiologic risk factors in multivariate logistic regression analysis (OR=3.2, p=0.025). CONCLUSIONS: DM is associated with macrovascular invasion among a cohort of transplanted HCC patients.

Recipient and Donor Outcomes After Living-Donor Liver Transplant for Unresectable Colorectal Liver Metastases.

Importance: Colorectal cancer is a leading cause of cancer-related death, and nearly 70% of patients with this cancer have unresectable colorectal cancer liver metastases (CRLMs). Compared with chemotherapy, liver transplant has been reported to improve survival in patients with CRLMs, but in North America, liver allograft shortages make the use of deceased-donor allografts for this indication problematic. Objective: To examine survival outcomes of living-donor liver transplant (LDLT) for unresectable, liver-confined CRLMs. Design, Setting, and Participants: This prospective cohort study included patients at 3 North American liver transplant centers with established LDLT programs, 2 in the US and 1 in Canada. Patients with liver-confined, unresectable CRLMs who had demonstrated sustained disease control on oncologic therapy met the inclusion criteria for LDLT. Patients included in this study underwent an LDLT between July 2017 and October 2020 and were followed up until May 1, 2021. Exposures: Living-donor liver transplant. Main Outcomes and Measures: Perioperative morbidity and mortality of treated patients and donors, assessed by univariate statistics, and 1.5-year Kaplan-Meier estimates of recurrence-free and overall survival for transplant recipients. Results: Of 91 evaluated patients, 10 (11%) underwent LDLT (6 [60%] male; median age, 45 years [range, 35-58 years]). Among the 10 living donors, 7 (70%) were male, and the median age was 40.5 years (range, 27-50 years). Kaplan-Meier estimates for recurrence-free and overall survival at 1.5 years after LDLT were 62% and 100%, respectively. Perioperative morbidity for both donors and recipients was consistent with established standards (Clavien-Dindo complications among recipients: 3 [10%] had none, 3 [30%] had grade II, and 4 [40%] had grade III; donors: 5 [50%] had none, 4 [40%] had grade I, and 1 had grade III). Conclusions and Relevance: This study's findings of recurrence-free and overall survival rates suggest that select patients with unresectable, liver-confined CRLMs may benefit from total hepatectomy and LDLT.

Hepatitis C virus core protein subverts the antiviral activities of human Kupffer cells.

BACKGROUND & AIMS: Kupffer cells (KC) are important innate immune cells of the liver, functioning as scavenging sinusoidal phagocytes and transducers of pattern recognition signals, including those of toll-like receptors (TLRs). The hepatitis C virus core protein (HCVc) engages TLR2 on peripheral blood monocytes and induces production of multiple inflammatory cytokines. We examined the effects of HCVc on human primary KC functions. METHODS: KC were isolated from living donor allografts and stimulated with HCVc and/or ligands for TLRs. KC were examined for production of cytokines, expression of programmed death-ligand 1 (PD-L1), secretion of type 1 interferons (IFNs), and expression of the apoptosis-inducing protein tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL). RESULTS: HCVc acts as a ligand for TLR2 on human KC, inducing them to secrete interleukin (IL)-1beta, TNF-alpha, and IL-10 and up-regulate cell surface PD-L1. HCVc blocked TLR3-mediated secretion of IFN-alpha, IFN-beta, and cell surface expression of the cytotoxic molecule TRAIL. Inhibition of phosphoinositide 3 kinase with LY294002 blocked the up-regulation of PD-L1 by TLR ligands and the TLR3-specific induction of TRAIL and type 1 IFNs. CONCLUSIONS: KC are intravascular macrophages that are continuously exposed to, and tolerant of, bacterial TLR ligands, which are delivered via the portal circulation. By mimicking a bacterial TLR2 ligand and effectively blocking the TLR3-mediated, double-stranded RNA-induced antiviral response, HCVc might appear to exploit this unique aspect of immunity in the liver.

Synergy between TLR3 and IL-18 promotes IFN-γ dependent TRAIL expression in human liver NK cells.

Natural killer (NK) cells are a component of innate immunity against viral infections through their rapid cytotoxic activity and cytokine production. However, intra-hepatic NK cells' ability to respond to virus is still mostly unknown. Our results show that the synthetic dsRNA polyinosinic-polycytidylic acid (poly I:C), a mimic of a common product of viral infections, activates NK cells directly in the context of cytokines found in the liver, i.e.: poly I:C plus inflammatory cytokines (IL-18, IL-12, and IL-2) induced NK cell IFN-γ production and TRAIL expression, and anti-inflammatory cytokines (TGF-ß and IL-10) inhibit NK cell IFN-γ production. Neutralization of IFN-γ blocks poly I:C plus inflammatory cytokines-induced NK cell TRAIL expression, suggesting that IFN-γ is an autocrine differentiation factor for these cells. A better understanding of the intra-hepatic NK cell activation against viral infection may help in the design of therapies and vaccines for the control of viral hepatitis.

High mortality in orthotopic liver transplant recipients who require hemodialysis.

Acute renal failure is a significant risk factor for death in patients with liver failure. The goal of this study was to analyze the impact of peri-transplant dialysis on the long-term mortality of liver transplant recipients. We performed a single-center, retrospective cohort study of 743 adult liver transplants; patients who received first liver transplants were divided into four groups: those who received more than one dialysis treatment (hemodialysis [HD], continuous veno-venous hemodialysis [CVVH]) pre-orthotopic liver transplantation (OLT), post OLT, pre- and post OLT, and those not dialyzed. There was no statistically significant difference in the mean survival time for patients who were not dialyzed or dialyzed only pre-OLT. Mean survival times were markedly reduced in patients dialyzed post OLT or both pre- and post OLT compared with those never dialyzed. Mortality risk in a Cox proportional hazards model correlated with hemodialysis post OLT, intra-operative vasopressin or neosynephrine, donor age >50 yr, Cr >1.5 mg/dL at transplant, and need for subsequent retransplant. Risk of post-OLT dialysis was correlated with pre-OLT dialysis, intra-operative levophed, pre-OLT diabetes, African American race, pre-OLT Cr >1.5, and male gender. We conclude that renal failure requiring hemodialysis post liver transplant, irrespective of pre-transplant dialysis status, is a profound risk factor for death in liver transplant recipients.

Accuracy of vascular invasion reporting in hepatocellular carcinoma before and after implementation of subspecialty surgical pathology sign-out.

CONTEXT: Liver cancers (including hepatocellular carcinoma [HCC] and cholangiocarcinoma) are the fifth most common cause of cancer death. The most powerful independent histologic predictor of overall survival after transplantation for HCC is the presence of microscopic vascular invasion (VI). AIMS: Given that VI is known to have somewhat high interobserver variability in both HCC and other tumors, we hypothesized that pathologists with special interest and training in liver pathology would be more likely to identify and report VI in HCC than would general surgical pathologists. SETTINGS AND DESIGN: We searched our departmental surgical pathology archives for transplant hepatectomies performed for HCC. SUBJECTS AND METHODS: We identified 143 such cases with available sign-out reports and hematoxylin and eosin-stained slides. STATISTICAL ANALYSIS USED: Kappa results (level of agreement) were calculated. RESULTS: Before surgical pathology subspecialty sign-out (SSSO) implementation, 49 of 88 HCC cases were reported as negative for VI; on rereview, 20 of these had VI. After SSSO implementation, 39 of 55 cases were reported as negative for VI; on our review, 8 of these had VI. Kappa (agreement) between general SO and subspecialty rereview was 0.562 (95% confidence interval [CI] = 0.411-0.714) "weak agreement." Kappa (agreement) between SSSO and rereview by select liver pathologists was 0.693 (95% CI = 0.505-0.880) "moderate agreement." CONCLUSIONS: Our study is one of only a few so far that have suggested improved accuracy of certain parameters under SSSO.

Tobacco Use is a Modifiable Risk Factor for Post-Transplant Biliary Complications.

PURPOSE: Biliary complications following liver transplantation are a significant source of morbidity, potentially leading to graft failure necessitating retransplantation. We sought to evaluate smoking as an independent risk factor for post-transplant biliary complications. METHODS: The clinical course of all adult primary deceased donor liver transplants at our center from 1992 to 2012 was reviewed. Eligible patients were assigned to cohorts based on their lifetime tobacco exposure: never smokers indicating 0 pack-year exposure and all others were ever smokers. Biliary complications were defined as strictures, leaks, or bilomas requiring intervention. Complication rates were analyzed using univariate regression models correlated with donor and recipient characteristics. Associations found during univariate analysis were included in the final multivariate Cox model. RESULTS: Eight hundred sixty-five subjects were followed for a median of 65 months; 482 (55.7%) of patients had a positive smoking history at the time of transplant. In univariate analysis, positive tobacco smoking history (HR = 1.36; p = 0.037) and increased time from quit date to transplantation (HR = 0.998; p = 0.011) were positive and negative predictors of biliary complication, respectively. Lifetime tobacco exposure remained a significant predictor of biliary complication on multivariate analysis (HR = 1.408; p = 0.023). CONCLUSIONS: Smoking status is an independent predictor of post-transplant biliary complications, and the data presented reinforces the importance of early smoking cessation in the pre-transplantation period.

Comparison of outcomes between SBRT, yttrium-90 radioembolization, transarterial chemoembolization, and radiofrequency ablation as bridge to transplant for hepatocellular carcinoma.

PURPOSE: To evaluate and compare outcome of stereotactic body radiation therapy (SBRT), yttrium-90 radioembolization, radiofrequency ablation (RFA), or transarterial chemoembolization (TACE) as bridge to liver transplant (LT) in patients with hepatocellular carcinoma. METHODS AND MATERIALS: We retrospectively reviewed patients treated at our institution with SBRT, TACE, RFA, or yttrium-90 as bridge to LT between 2006 and 2013. We analyzed radiologic and pathologic response and rate of failure after bridge therapy. Toxicities were reported using Common Terminology Criteria for Adverse Events, 4.0. Kaplan-Meier method was used to calculate disease-free survival (DFS) and overall survival after LT. RESULTS: Sixty patients with a median age 57.5 years (range, 44-70) met inclusion criteria. Thirty-one patients (50.7%) had hepatitis C cirrhosis, 14 (23%) alcoholic cirrhosis, and 8 (13%) nonalcoholic steatohepatitis cirrhosis. Patients received a total of 79 bridge therapies: SBRT (n = 24), TACE (n = 37), RFA (n = 9), and Y90 (n = 9). Complete response (CR) was 25% for TACE, 8.6% for SBRT, 22% for RFA, and 33% for Y90. Grade 3 or 4 acute toxicity occurred following TACE (n = 4) and RFA (n = 2). Transplant occurred at a median of 7.4 months after bridge therapy. Pathological response among 57 patients was 100% necrosis (n = 23, 40%), >50% necrosis (n = 20, 35%), <50% necrosis (n = 9, 16%), and no necrosis (n = 5, 9%). Pathologic complete response was as follows: SBRT (28.5%), TACE (41%), RFA (60%), Y90 (75%), and multiple modalities (33%). At a median follow-up of 35 months, 7 patients had recurrence after LT. DFS was 85.8% and overall survival was 79% at 5 years. CONCLUSION: All bridge therapies demonstrated good pathological response and DFS after LT. SBRT and Y90 demonstrated significantly less grade ≥3 acute toxicity. Choice of optimal modality depends on tumor size, pretreatment bilirubin level, Child-Pugh status, and patient preference. Such a decision is best made at a multidisciplinary tumor board as is done at our institution.

Bleeding esophagogastric varices from extrahepatic portal hypertension: 40 years' experience with portal-systemic shunt.

BACKGROUND: This article discusses the largest and longest experience reported to date of the use of portal-systemic shunt (PSS) to treat recurrent bleeding from esophagogastric varices caused by extrahepatic portal hypertension associated with portal vein thrombosis (PVT). STUDY DESIGN: Two hundred consecutive children and adults with extrahepatic portal hypertension caused by PVT who were referred between 1958 and 1998 after recovering from at least two episodes of bleeding esophagogastric varices requiring blood transfusions were managed according to a well-defined and uniformly applied protocol. All but 14 of the 200 patients were eligible for and received 5 or more years of regular followup (93%); 166 were eligible for and received 10 or more years of regular followup (83%). RESULTS: The etiology of PVT was unknown in 65% of patients. Identifiable causes of PVT were neonatal omphalitis in 30 patients (15%), umbilical vein catheterization in 14 patients (7%), and peritonitis in 14 patients (7%). The mean number of bleeding episodes before PSS was 5.4 (range 2 to 18). Liver biopsies showed normal morphology in all patients. The site of PVT was the portal vein alone in 134 patients (76%), the portal vein and adjacent superior mesenteric vein in 10 patients (5%), and the portal and splenic veins in 56 patients (28%). Postoperative survival to leave the hospital was 100%. Actuarial 5-year, 10-year, and 15-year survival rates were 99%, 97%, and 95%, respectively. Five patients (2.5%), all with central end-to-side splenorenal shunts, developed thrombosis of the PSS, and these were the only patients who had recurrent variceal bleeding. During 10 or more years of followup, 97% of the eligible patients were shown to have a patent shunt and were free of bleeding. No patient developed portal-systemic encephalopathy, liver function tests remained normal, liver biopsies in 100 patients showed normal architecture, hypersplenism was corrected. CONCLUSION: PSS is the only consistently effective therapy for bleeding esophagogastric varices from PVT and extrahepatic portal hypertension, resulting in many years of survival, freedom from recurrent bleeding, normal liver function, and no encephalopathy.

Unshuntable extrahepatic portal hypertension revisited: 43 years' experience with radical esophagogastrectomy treatment of bleeding esophagogastric varices.

BACKGROUND: In 1994, the authors reported their experience with radical esophagogastrectomy for bleeding esophagogastric varices due to unshuntable extra-hepatic portal hypertension. Since then, the series has expanded from 22 to 44 patients. The aim of this study was to assess the validity of the previous observations and conclusions in the largest series with the longest follow-up. METHODS: From 1968 to 2005, 44 patients with unshuntable extra-hepatic portal hypertension were treated by total gastrectomy and resection of the distal two thirds of the esophagus. Before referral, the patients experienced 4 to 24 episodes of variceal bleeding requiring a mean 130 U of blood transfusion, 15 hospital admissions, and 6 previous unsuccessful operations. RESULTS: Transient postoperative complications occurred in 50% of patients. The survival rate is 100%, with no recurrence of variceal bleeding during 7 to 43 years of follow-up. Liver function and biopsy results have been normal. Quality of life has been excellent or good in 91%. Eighty-six percent have resumed employment or full-time housekeeping. CONCLUSIONS: In unshuntable extra-hepatic portal hypertension, radical esophagogastrectomy is the only consistently effective treatment of variceal hemorrhage. Prompt use of this lifesaving procedure is warranted.

Comparative analysis of hepatitis C recurrence and fibrosis progression between deceased-donor and living-donor liver transplantation: 8-year longitudinal follow-up.

BACKGROUND: Hepatitis C virus (HCV) recurrence is universal after liver transplantation (LT). Whether the progression of recurrent HCV is faster after live-donor LT (LDLT) compared with deceased-donor LT (DDLT) is debatable. METHODS AND RESULTS: We retrospectively examined 100 consecutive LTs (65 DDLTs and 35 LDLTs) performed between July 2000 and July 2003. A total of 147 liver biopsies were performed between 6 months post-LT and last follow-up. Mean donor age and model for end-stage liver disease (MELD) score were significantly lower in LDLT (P<0.01). On a mean follow-up of 86.6±6.8 months, overall patient and graft survivals were 61% (51% DDLT vs. 77.1% LDLT; P=0.026) and 56% (46.2% DDLT vs. 71.4% LDLT; P=0.042), respectively. Eight of 39 (20.5%) deaths (7 DDLT and 1 LDLT) and two of nine (22.2%) retransplants (one in each group) were related to recurrent HCV. Mean fibrosis scores for DDLT and LDLT were 1.9±1.7 and 1.6±1.4, respectively (P=0.01). When donor age less than 50 years and MELD score less than 25 were matched among 64 patients (32 DDLT and 32 LDLT), the overall patient and graft survivals were 73.4% (68.8% DDLT vs. 78.1% LDLT; P=0.439) and 71.9% (71.9% DDLT vs. 71.9% LDLT; P=0.978), respectively. CONCLUSIONS: Long-term survival rates were better, and fibrosis scores were lower for LDLT. The survivals between LDLT and DDLT were comparable for patients with MELD score less than 25 and donor age less than 50 years.

Utility of transjugular intrahepatic portosystemic shunts in liver-transplant recipients.

BACKGROUND: Transjugular intrahepatic portosystemic shunts (TIPS) have been used to control symptomatic portal hypertension in patients awaiting liver transplant. Although their role in pretransplantation patients is well established, their role in posttransplantation patients is unclear. STUDY DESIGN: Retrospective analyses were performed for 18 liver-transplant recipients who underwent TIPS for recurrent end-stage liver disease. Patients were evaluated in regard to gender, age, diagnoses, allograft type, indication for TIPS, portal pressures, laboratory results, Model for End-Stage Liver Disease (MELD) score, and outcomes. RESULTS: Median days from transplant to TIPS was 939 days (range, 122 to 3,415 days). Indications included variceal bleeding (n=2) and ascites (n=16). Ten patients (56%) responded to TIPS; TIPS prevented bleeding in both patients with varices, and it achieved symptomatic benefit in half of all patients with ascites. TIPS reduced median portal pressures from 22 mmHg (range, 17 to 50 mmHg) to 16 mmHg (range, 11 to 22 mmHg) and median portosystemic pressure gradients from 18 mmHg (range, 8 to 30 mmHg) to 8 mmHg (range, 2 to 12 mmHg). It increased median Model for End-Stage Liver Disease scores from 16 (range, 12 to 29) to 17 (range, 10 to 34) immediately and to 22 (range, 10 to 35) at 1 month. Six patients (33%) underwent retransplantation at a median of 58 days (range, 21 to 71 days) post-TIPS. Of the remaining 12 patients, 3 (25%) were alive and well at a median of 90 days (range, 78 to 1,169 days) post-TIPS; 9 (75%) died at a median of 99 days (range, 13 to 1,400 days) post-TIPS. Subgroup analysis failed to demonstrate significant differences between patients whose ascites responded to TIPS (n=8) and patients whose ascites did not (n=8). Responders were younger, had higher baseline portal pressures, greater reductions in portal-systemic pressure gradients, and better hepatic function. CONCLUSIONS: Though small, this was the largest series to date of TIPS in liver-transplant recipients. Overall, 56% of patients responded to TIPS. No single factor predicted response or nonresponse of ascites to TIPS. Without retransplantation, 75% of patients died. Careful selection is necessary when considering TIPS for patients with ascites.