Performance of Multi-Layer Feedforward Neural Networks to Predict Liver Transplantation Outcome

A novel multisolutional clustering and quantization (MCO) algorithm has been developed that provides a flexible way to preprocess data. It was tested whether it would impact the neural network’s performance favorably and whether the employment of the proposed algorithm would enable neural networks to handle missing data. This was assessed by comparing the performance of neural networks using a well-documented data set to predict outcome following liver transplantation. This new approach to data preprocessing leads to a statistically significant improvement in network performance when compared to simple linear scaling. The obtained results also showed that coding missing data as zeroes in combination with the MCO algorithm, leads to a significant improvement in neural network performance on a data set containing missing values in 59.4% of cases when compared to replacement of missing values with either series means or medians.

Liver transplantation using donation after cardiac death donors: long-term follow-up from a single center

There is a lack of universally accepted clinical parameters to guide the utilization of donation after cardiac death (DCD) donor livers and it is unclear as to which patients would benefit most from these organs. We reviewed our experience in 141 patients who underwent liver transplantation using DCD allografts from 1993 to 2007. Patient outcomes were analyzed in comparison to a matched cohort of 282 patients who received livers from donation after brain death (DBD) donors. Patient survival was similar, but 1-, 5- and 10-year graft survival was significantly lower in DCD (69%, 56%, 44%) versus DBD (82%, 73%, 63%) subjects (p < 0.0001). Primary nonfunction and biliary complications were more common in DCD patients, accounting for 67% of early graft failures. A donor warm ischemia time >20 min, cold ischemia time >8 h and donor age >60 were associated with poorer DCD outcomes. There was a lack of survival benefit in DCD livers utilized in patients with model for end-stage liver disease (MELD) < or =30 or those not on organ-perfusion support, as graft survival was significantly lower compared to DBD patients. However, DCD and DBD subjects transplanted with MELD >30 or on organ-perfusion support had similar graft survival, suggesting a potentially greater benefit of DCD livers in critically ill patients.

Causes of death in patients with unresectable hepatocellular carcinoma

Most patients with hepatocellular carcinoma (HCC) also have cirrhosis, an independent cause of death. We considered an alternative definition of tumor-related death in patients with HCC and attempted to validate our definition. Two hundred thirty-seven HCC patients were diagnosed, followed, and died over a 12-year period and were evaluated every 2 months, including their last 6 months of life. We defined death by cancer if there was, in the last 6 months of life, a CT scan increase of >25% in the sum of tumor index lesions' cross-sectional areas or new onset of, or increase in, either vascular invasion or metastatic disease (Group 1). Patients with stable cancer were considered to have died from any other cause (Group 2). We found that 135 (57%) patients died from cancer progression (Group 1), whereas 102 (43%) patients did not (Group 2). There was a statistically significant difference between Group 1 and Group 2 patients in percentage with bilobar disease (P = 0.03), more than one tumor (P = 0.01), an increase in AFP (P = 0.04), vascular invasion (P = 0.001), and the presence of metastases (P = 0.01). We conclude that 57% of patients with unresectable HCC died as a direct result of cancer progression, but 43% did not. The latter died from complications of their cirrhosis, including sepsis, GI bleeds, and renal failure.

Prognostic models in hepatocellular carcinoma (HCC) and statistical methodologies behind them

Hepatocellular carcinoma (HCC) is estimated to be responsible for 250,000 deaths worldwide yearly. Aggressive surgical resection or liver transplantation still remain the only viable curative options for patients suffering the disease despite the multitude of emerging therapies for HCC. However, even with the most aggressive surgical intervention, survival varies widely within each particular stage of HCC. In order to improve utilization of available therapeutic modalities, a number of outcome prognostic models have been developed. This manuscript reviews the prognostic models most commonly utilized in clinical practice and the statistical methodologies on which these models are based. A multitude of statistical and mathematical techniques can be used for prognostic model development. The most common methodologies used for HCC prognostic model development can be generally divided into four groups: survival, artificial neural networks, analysis of variance, and cluster analysis. Survival methodologies (such as Cox proportional hazard model) are commonly employed for estimation of relative significance of risk factors for patient survival or cancer recurrence. Artificial neural networks (such as back-propagation network) can be supreme approximation tools for any continuous or binary function, and as such can be employed for prognostication of HCC recurrence (death). Analysis of variance and cluster analysis are the most common statistical tools of recently evolved microarrays technology, which, in turn, is one of the most promising tools available to the cancer researcher.

Chronic allograft nephropathy score before sirolimus rescue predicts allograft function in renal transplant patients

Chronic allograft nephropathy (CAN) is a major indication for initiation of sirolimus (SRL) in renal transplantation (TX) to prevent deterioration of renal function. We evaluated whether the CAN score at time of sirolimus rescue (SRL-R) predicts renal allograft function. CAN score is the sum of the following 4 categories: glomerulopathy (cg, 0-3), interstitial fibrosis (ci, 0-3), tubular atrophy (ct, 0-3), and vasculopathy (cv, 0-3). This is a retrospective cohort study of renal transplant recipients from July 2001 to March 2004. Immunosuppression consisted of preconditioning with rabbit anti-thymocyte globulin or alemtuzumab and maintenance with tacrolimus (TAC) monotherapy with spaced weaning, if applicable, SRL-R was achieved by conversion from TAC, or by addition to reduced doses of TAC. Ninety patients received SRL. Thirty-three of these patients met the inclusion criteria of the following: (1) receipt of SRL for >6 months, and (2) follow-up of > or =6 months. There were 16 patients in the low-CAN (0-4) group and 17 patients in the high-CAN (>4) group. Cockcroft-Gault (C-G) glomerular filtration rate (GFR) was calculated at SRL-R and at 1, 3, 6, and 12 months. The DeltaGFR was significantly better in the low-CAN group at 1, 3, and 6 months. A trend toward an improved DeltaGFR was present at 12 months in the low-CAN group (P = .16). CAN scoring at the time of SRL-R predicts recovery of renal allograft function (as measured using DeltaGFR), and should be used in preference to biochemical markers (Cr and C-G GFR), which may not be reliable predictors.

Survival of liver transplant patients coinfected with HIV and HCV is adversely impacted by recurrent hepatitis C

Although liver transplantation (LTx) in HIV-positive patients receiving highly active antiretroviral therapy (HAART) has been successful, some have reported poorer outcomes in patients coinfected with hepatitis C virus (HCV). Here we discuss the impact of recurrent HCV on 27 HIV-positive patients who underwent LTx. HIV infection was well controlled post-transplantation. Survival in HIV-positive/HCV-positive patients was shorter compared to a cohort of HIV-negative/HCV-positive patients matched in age, model for end-stage liver disease (MELD) score, and time of transplant, with cumulative 1-, 3- and 5-year patient survival of 66.7%, 55.6% and 33.3% versus 75.7%, 71.6% and 71.6%, respectively, although not significantly (p = 0.07), and there was a higher likelihood of developing cirrhosis or dying from an HCV-related complication in coinfected subjects (RR = 2.6, 95% CI, 1.06-6.35; p = 0.03). Risk factors for poor survival included African-American race (p = 0.02), MELD score > 20 (p = 0.05), HAART intolerance postLTx (p = 0.01), and postLTx HCV RNA > 30000000 IU/mL (p = 0.00). Recurrent HCV in 18 patients was associated with eight deaths, including three from fibrosing cholestatic hepatitis. Among surviving coinfected recipients, five are alive at least 3 years after LTx, and of 15 patients treated with interferon-alpha/ribavirin, six (40%) are HCV RNA negative, including four with sustained virological response. Hepatitis C is a major cause of graft loss and patient mortality in coinfected patients undergoing LTx.

Living donor renal transplantation using alemtuzumab induction and tacrolimus monotherapy

Alemtuzumab was used as an induction agent in 205 renal transplant recipients undergoing 207 living donor renal transplants. All donor kidneys were recovered laparoscopically. Postoperatively, patients were treated with tacrolimus monotherapy, and immunosuppression was weaned when possible. Forty-seven recipients of living donor renal transplants prior to the induction era who received conventional triple drug immunosuppression without antibody induction served as historic controls. The mean follow-up was 493 days in the alemtuzumab group and 2101 days in the historic control group. Actuarial 1-year patient and graft survival were 98.6% and 98.1% in the alemtuzumab group, compared to 93.6% and 91.5% in the control group, respectively. The incidence of acute cellular rejection (ACR) at 1 year was 6.8% in the alemtuzumab group and 17.0% (p < 0.05) in the historic control group. Most (81.3%) episodes of ACR in the alemtuzumab group were Banff 1 (a or b) and were sensitive to steroid pulses for the treatment of rejection. There was no cytomegalovirus disease or infection. The incidence of delayed graft function was 0%, and the incidence of posttransplant insulin-dependent diabetes mellitus was 0.5%. This study represents the largest series to date of live donor renal transplant recipients undergoing alemtuzumab induction, and confirms the short-term safety and efficacy of this approach.

Real-time monitoring of acute liver-allograft rejection using the Banff schema

BACKGROUND

The Banff schema is the internationally accepted standard for grading acute liver-allograft rejection, but it has not been prospectively tested.

METHODS

Complete Banff grading was prospectively applied to 2,038 liver-allograft biopsies from 901 adult tacrolimus-treated primary hepatic allograft recipients between August 1995 and September 2001. Histopathologic data was melded with demographic, clinical, and laboratory data into a database on an ongoing basis using locally developed software.

RESULTS

Acute rejection developed in 575 of 901 (64%) patients and the worst grade was mild in 422 of 575 (73%). At least one episode of moderate or severe acute rejection developed in 153 of 901 (17%) patients and most episodes, irrespective of severity, occurred within the first year after transplantation. Patients with moderate or severe acute rejection showed higher alanine aminotransferase (P =0.007) and aspartate aminotransferase ( P=0.07) levels and were more likely to develop perivenular fibrosis on follow-up biopsies (P =0.001) and graft failure from acute or chronic rejection ( P=0.004) than those with mild rejection. Regardless of severity, 80% of patients with acute rejection did not develop significant fibrosis in follow-up biopsies, and graft failure from acute or chronic rejection occurred in only 11 of 901 (1%) allografts.

CONCLUSIONS

Most acute-rejection episodes are mild and do not lead to clinically significant architectural sequelae. When tested prospectively under real-life and -time conditions, the Banff schema can be used to identify those few patients who are potentially at risk for more significant problems. Creation, capture, and integration of non-free text, or "digital," pathology data can be used to prospectively conduct outcomes-based research in transplantation.

De novo malignancies after liver transplantation: a major cause of late death

1. Recurrent and de novo malignancies are the second leading causes of late death in liver transplant recipients, following age-related cardiovascular complications. 2. The increased incidence of de novo malignancies in liver transplant recipients compared with the general population reflects their demographic makeup, known preexistent risk factors for cancer, greater rate of chronic viral infection, and actions of exogenous immunosuppression. 3. The greatest incidence of de novo malignancies is seen in cancers associated with chronic viral infections, such as Epstein-Barr virus-associated posttransplant lymphoproliferative disease, and skin cancers, including squamous cell carcinoma and Kaposi's sarcoma. 4. Although a greater incidence of such malignancies as oropharyngeal malignancy and colorectal cancer was noted, there did not appear to be an increased risk for liver transplant recipients matched for age, sex, and length of follow-up using modified life-table technique and Surveillance Epidemiology End Result data with a similar at-risk group. However, they may present with more advanced stages of disease. 5. An increased incidence of de novo cancers in chronically immunocompromised liver transplant recipients demands careful long-term screening protocols to help facilitate diagnosis at an earlier stage of disease.

Clinical intestinal transplantation: a decade of experience at a single center

OBJECTIVE

To assess the long-term efficacy of intestinal transplantation under tacrolimus-based immunosuppression and the therapeutic benefit of newly developed adjunct immunosuppressants and management strategies.

SUMMARY BACKGROUND DATA

With the advent of tacrolimus in 1990, transplantation of the intestine began to emerge as therapy for intestinal failure. However, a high risk of rejection, with the consequent need for acute and chronic high-dose immunosuppression, has inhibited its widespread application.

METHODS

During an 11-year period, divided into two segments by a 1-year moratorium in 1994, 155 patients received 165 intestinal allografts under immunosuppression based on tacrolimus and prednisone: 65 intestine alone, 75 liver and intestine, and 25 multivisceral. For the transplantations since the moratorium (n = 99), an adjunct immunosuppressant (cyclophosphamide or daclizumab) was used for 74 transplantations, adjunct donor bone marrow was given in 39, and the intestine of 11 allografts was irradiated with a single dose of 750 cGy.

RESULTS

The actuarial survival rate for the total population was 75% at 1 year, 54% at 5 years, and 42% at 10 years. Recipients of liver plus intestine had the best long-term prognosis and the lowest risk of graft loss from rejection (P =.001). Since 1994, survival rates have improved. Techniques for early detection of Epstein-Barr and cytomegaloviral infections, bone marrow augmentation, the adjunct use of the interleukin-2 antagonist daclizumab, and most recently allograft irradiation may have contributed to the better results.

CONCLUSION

The survival rates after intestinal transplantation have cumulatively improved during the past decade. With the management strategies currently under evaluation, intestinal transplant procedures have the potential to become the standard of care for patients with end-stage intestinal failure.

An analysis of early renal transplant protocol biopsies--the high incidence of subclinical tubulitis

To investigate the possibility that we have been underestimating the true incidence of acute rejection, we began to perform protocol biopsies after kidney transplantation. This analysis looks at the one-week biopsies. Between March 1 and October 1, 1999, 100 adult patients undergoing cadaveric kidney or kidney/pancreas transplantation, or living donor kidney transplantation, underwent 277 biopsies. We focused on the subset of biopsies in patients without delayed graft function (DGF) and with stable or improving renal function, who underwent a biopsy 8.2+/-2.6 d (range 3-18 d) after transplantation (n = 28). Six (21%) patients with no DGF and with stable or improving renal function had borderline histopathology, and 7 (25%) had acute tubulitis on the one-week biopsy. Of the 277 kidney biopsies, there was one (0.4%) serious hemorrhagic complication, in a patient receiving low molecular weight heparin; she ultimately recovered and has normal renal function. Her biopsy showed Banff 1B tubulitis. In patients with stable or improving renal allograft function early after transplantation, subclinical tubulitis may be present in a substantial number of patients. This suggests that the true incidence of rejection may be higher than is clinically appreciated.

Liver transplantation for hepatocellular carcinoma: a proposal of a prognostic scoring system

BACKGROUND

The current staging system of hepatocellular carcinoma established by the International Union Against Cancer and the American Joint Committee on Cancer does not necessarily predict the outcomes after hepatic resection or transplantation.

STUDY DESIGN

Various clinical and pathologic risk factors for tumor recurrence were examined on 344 consecutive patients who received hepatic transplantation in the presence of nonfibrolamellar hepatocellular carcinoma to establish a reliable risk scoring system.

RESULTS

Multivariate analysis identified three factors as independently significant poor prognosticators: 1) bilobarly distributed tumors, 2) size of the greatest tumor (2 to 5 cm and > 5 cm), and 3) vascular invasion (microscopic and macroscopic). Prognostic risk score (PRS) of each patient was calculated from the relative risks of multivariate analysis. The patients were grouped into five grades of tumor recurrence risk: grade 1: PRS = 0 to < 7.5; grade 2: PRS = 7.5 to < or = 11.0; grade 3: PRS > 11.0 to 15.0; grade 4: PRS > or = 15.0; and grade 5: positive node, metastasis, or margin. The proposed PRS system correlated extremely well with tumor-free survival after liver transplantation (100%, 61%, 40%, 5%, and 0%, from grades 1 to 5, respectively, at 5 years), but current pTNM staging did not.

CONCLUSIONS

1) Patients with grades 1 and 2 are effectively treated with liver transplantation, 2) patients with grades 4 and 5 are poor candidates for liver transplantation, and 3) patients with grade 1 do not benefit from adjuvant chemotherapy.

Liver transplantation and chemotherapy for hepatoblastoma and hepatocellular cancer in childhood and adolescence

OBJECTIVE

To describe our experience with total hepatectomy and liver transplantation as treatment for primary hepatoblastoma (HBL) and hepatocellular carcinoma (HCC) in children.

STUDY DESIGN

A retrospective analysis of the perioperative course of 31 children with unresectable primary HBL (n = 12) and HCC (n = 19) who underwent transplantation between May 1989 and December 1998. Systemic (n = 18) and intraarterial (n = 7) neoadjuvant chemotherapy were administered; follow-up ranged from 1 to 185 months.

RESULTS

For HBL, 1-year, 3-year, and 5-year posttransplantation survival rates were 92%, 92%, and 83%, respectively. Intravenous invasion, positive hilar lymph nodes, and contiguous spread did not have a significant adverse effect on outcome; distant metastasis was responsible for 2 deaths. Intraarterial chemotherapy was effective in all patients treated. For HCC, the overall 1-year, 3-year, and 5-year disease-free survival rates were 79%, 68%, and 63%, respectively. Vascular invasion, distant metastases, lymph node involvement, tumor size, and gender were significant risk factors for recurrence. Intraarterial chemotherapy was effective in 1 of 3 patients. Six patients died of recurrent HCC, and 3 deaths were unrelated to recurrent tumor.

CONCLUSION

Liver transplantation for unresectable HBL and HCC can be curative. Risk factors for recurrence were significant only for HCC, with more advanced stages amenable to cure in the HBL group.

Is the pathologic TNM staging system for patients with hepatoma predictive of outcome?

BACKGROUND

The pathologic TNM (pTNM) staging system was designed to aid in determining the prognosis of cancer patients and in planning and evaluating their treatment. The current pTNM classification system was not found to be predictive for patients undergoing orthotopic liver transplantation (OLTx) in the presence of hepatocellular carcinoma (HCC). Therefore, the authors examined the current system to determine whether improvements would allow the development of a more predictive system.

METHODS

Three hundred seven patients with HCC underwent OLTx between 1981 and 1997. Risk factors for recurrence were identified using the Kaplan-Meier method with the log rank test. The Cox proportional hazards model was used to identify factors independently predictive of recurrence which were then used to create a new staging system.

RESULTS

There was neither a direct correlation between the current pTNM system and tumor free survival nor homogeneity in outcomes for patients within certain current pTNM categories. Depth of vascular invasion, lobar distribution, lymph node status, and largest tumor size were found to be independent predictors of tumor free survival; tumor number was not found to be significant in multivariate analysis. A new staging system is proposed, which takes into account the results of the multivariate analysis in which tumor free survival correlates directly with stage.

CONCLUSIONS

The proposed staging system is superior to the current pTNM staging system in predicting tumor free survival following OLTx with HCC. Further studies will determine the appropriateness of this system for staging HCC after subtotal hepatic resection.

Is the pathologic TNM staging system for patients with hepatoma predictive of outcome?

BACKGROUND

The pathologic TNM (pTNM) staging system was designed to aid in determining the prognosis of cancer patients and in planning and evaluating their treatment. The current pTNM classification system was not found to be predictive for patients undergoing orthotopic liver transplantation (OLTx) in the presence of hepatocellular carcinoma (HCC). Therefore, the authors examined the current system to determine whether improvements would allow the development of a more predictive system.

METHODS

Three hundred seven patients with HCC underwent OLTx between 1981 and 1997. Risk factors for recurrence were identified using the Kaplan-Meier method with the log rank test. The Cox proportional hazards model was used to identify factors independently predictive of recurrence which were then used to create a new staging system.

RESULTS

There was neither a direct correlation between the current pTNM system and tumor free survival nor homogeneity in outcomes for patients within certain current pTNM categories. Depth of vascular invasion, lobar distribution, lymph node status, and largest tumor size were found to be independent predictors of tumor free survival; tumor number was not found to be significant in multivariate analysis. A new staging system is proposed, which takes into account the results of the multivariate analysis in which tumor free survival correlates directly with stage.

CONCLUSIONS

The proposed staging system is superior to the current pTNM staging system in predicting tumor free survival following OLTx with HCC. Further studies will determine the appropriateness of this system for staging HCC after subtotal hepatic resection.

Is the pathologic TNM staging system for patients with hepatoma predictive of outcome?

BACKGROUND

The pathologic TNM (pTNM) staging system was designed to aid in determining the prognosis of cancer patients and in planning and evaluating their treatment. The current pTNM classification system was not found to be predictive for patients undergoing orthotopic liver transplantation (OLTx) in the presence of hepatocellular carcinoma (HCC). Therefore, the authors examined the current system to determine whether improvements would allow the development of a more predictive system.

METHODS

Three hundred seven patients with HCC underwent OLTx between 1981 and 1997. Risk factors for recurrence were identified using the Kaplan-Meier method with the log rank test. The Cox proportional hazards model was used to identify factors independently predictive of recurrence which were then used to create a new staging system.

RESULTS

There was neither a direct correlation between the current pTNM system and tumor free survival nor homogeneity in outcomes for patients within certain current pTNM categories. Depth of vascular invasion, lobar distribution, lymph node status, and largest tumor size were found to be independent predictors of tumor free survival; tumor number was not found to be significant in multivariate analysis. A new staging system is proposed, which takes into account the results of the multivariate analysis in which tumor free survival correlates directly with stage.

CONCLUSIONS

The proposed staging system is superior to the current pTNM staging system in predicting tumor free survival following OLTx with HCC. Further studies will determine the appropriateness of this system for staging HCC after subtotal hepatic resection.

Should hepatomas be treated with hepatic resection or transplantation?

BACKGROUND

The aim of this collaborative study was to compare the long term results of hepatic resection (Hx) with those of orthotopic liver transplantation (OLTx) in large numbers of cirrhotic patients with hepatocellular carcinoma (HCC) and to delineate the roles of these two surgical treatments.

METHODS

The databases of the National Cancer Center Hospital in Japan and the University of Pittsburgh Medical Center in the U. S. were exchanged and 294 cirrhotic patients who underwent curative Hx and 270 cirrhotic patients who underwent curative OLTx were selected for comparison.

RESULTS

The mortality rate within 30 days and that within 150 days after Hx were significantly lower than those after OLTx (P = 0.001 and P = 0.00007, respectively). Overall survival was similar between the Hx group and the OLTx group (P = 0.40). When compared in the HCC patients without macroscopic vascular invasion and lymph node metastases, the overall survival rate after OLTx was significantly higher than that after Hx (P = 0.006). However, this difference was not significant between the patients with Child-Pugh Grade A tumors in the Hx group and all patients (majority with Child-Pugh Grade C tumors) in the OLTx group (P = 0.25). Tumor free survival after OLTx was significantly higher than that after Hx (P < 0.0001), particularly in HCCs measuring </=5 cm, unilobarly distributed tumors, and HCCs with either no or only microscopic vascular invasion. In HCCs measuring > 5 cm and those with macroscopic vascular invasion, the tumor free survival rate was similar between the Hx group and the OLTx group.

CONCLUSIONS

In the face of organ shortage, HCC developing in a well compensated cirrhotic liver initially may be treated with Hx. However, the authors believe OLTx should be applied selectively to those patients with tumor recurrence and/or progressive hepatic failure.

Should hepatomas be treated with hepatic resection or transplantation?

BACKGROUND

The aim of this collaborative study was to compare the long term results of hepatic resection (Hx) with those of orthotopic liver transplantation (OLTx) in large numbers of cirrhotic patients with hepatocellular carcinoma (HCC) and to delineate the roles of these two surgical treatments.

METHODS

The databases of the National Cancer Center Hospital in Japan and the University of Pittsburgh Medical Center in the U. S. were exchanged and 294 cirrhotic patients who underwent curative Hx and 270 cirrhotic patients who underwent curative OLTx were selected for comparison.

RESULTS

The mortality rate within 30 days and that within 150 days after Hx were significantly lower than those after OLTx (P = 0.001 and P = 0.00007, respectively). Overall survival was similar between the Hx group and the OLTx group (P = 0.40). When compared in the HCC patients without macroscopic vascular invasion and lymph node metastases, the overall survival rate after OLTx was significantly higher than that after Hx (P = 0.006). However, this difference was not significant between the patients with Child-Pugh Grade A tumors in the Hx group and all patients (majority with Child-Pugh Grade C tumors) in the OLTx group (P = 0.25). Tumor free survival after OLTx was significantly higher than that after Hx (P < 0.0001), particularly in HCCs measuring </=5 cm, unilobarly distributed tumors, and HCCs with either no or only microscopic vascular invasion. In HCCs measuring > 5 cm and those with macroscopic vascular invasion, the tumor free survival rate was similar between the Hx group and the OLTx group.

CONCLUSIONS

In the face of organ shortage, HCC developing in a well compensated cirrhotic liver initially may be treated with Hx. However, the authors believe OLTx should be applied selectively to those patients with tumor recurrence and/or progressive hepatic failure.

Should hepatomas be treated with hepatic resection or transplantation?

BACKGROUND

The aim of this collaborative study was to compare the long term results of hepatic resection (Hx) with those of orthotopic liver transplantation (OLTx) in large numbers of cirrhotic patients with hepatocellular carcinoma (HCC) and to delineate the roles of these two surgical treatments.

METHODS

The databases of the National Cancer Center Hospital in Japan and the University of Pittsburgh Medical Center in the U. S. were exchanged and 294 cirrhotic patients who underwent curative Hx and 270 cirrhotic patients who underwent curative OLTx were selected for comparison.

RESULTS

The mortality rate within 30 days and that within 150 days after Hx were significantly lower than those after OLTx (P = 0.001 and P = 0.00007, respectively). Overall survival was similar between the Hx group and the OLTx group (P = 0.40). When compared in the HCC patients without macroscopic vascular invasion and lymph node metastases, the overall survival rate after OLTx was significantly higher than that after Hx (P = 0.006). However, this difference was not significant between the patients with Child-Pugh Grade A tumors in the Hx group and all patients (majority with Child-Pugh Grade C tumors) in the OLTx group (P = 0.25). Tumor free survival after OLTx was significantly higher than that after Hx (P < 0.0001), particularly in HCCs measuring </=5 cm, unilobarly distributed tumors, and HCCs with either no or only microscopic vascular invasion. In HCCs measuring > 5 cm and those with macroscopic vascular invasion, the tumor free survival rate was similar between the Hx group and the OLTx group.

CONCLUSIONS

In the face of organ shortage, HCC developing in a well compensated cirrhotic liver initially may be treated with Hx. However, the authors believe OLTx should be applied selectively to those patients with tumor recurrence and/or progressive hepatic failure.

An analysis of pretransplantation variables associated with long-term allograft outcome in pediatric liver transplant recipients receiving primary tacrolimus (FK506) therapy

BACKGROUND

The present study analyzes pretransplantation variables associated with long-term liver allograft survival in 278 children who underwent transplantation under primary tacrolimus (FK506) therapy at a single center between October 1989 and October 1996.

METHODS

The influence of 17 pretransplantation variables on long-term liver allograft outcome was analyzed. Donor variables included age, weight, gender, and cold ischemia time. Recipient variables included age, weight, gender, original liver disease, pretransplantation waiting time, previous abdominal surgery, United Network of Organ Sharing (UNOS) status, ABO blood group, bilirubin level, prothrombin time, ammonia level, creatinine level, and reduced-size/split liver grafts.

RESULTS

Overall actuarial graft survival was 79.9% at 1 year, 79.1% at 2 years, and 78.3% at 3, 4, and 5 years. Retransplantation rate was 10.8%. Pretransplantation variables with a significant adverse effect on graft survival by univariate analysis were donor age < or = 1 year (P<0.004), donor weight < or = 10 kg (P<0.003), UNOS status I and II (P<0.007), ABO type O, B, and AB (P<0.03), and reduced-size/split liver grafts (P<0.02). Pretransplantation variables significant by multivariate analysis and therefore independent predictors of inferior graft outcome were donor weight '10 kg (relative risk [RR] 2.91, confidence interval [CI] 1.53-5.51); reduced-size/split liver grafts (RR 2.53, CI 1.30-5.64); and UNOS status I (RR 2.22, CI 1.11-4.43).

CONCLUSIONS

Pediatric liver transplant recipients receiving primary tacrolimus therapy have long-term graft survival rates approaching 80%. UNOS status, donor weight, and the use of reduced-size/split liver grafts are the most important factors affecting survival.

Hyperbaric oxygen therapy for hepatic artery thrombosis after liver transplantation in children

Early hepatic artery thrombosis (HAT) after pediatric orthotopic liver transplantation (OLT) can cause significant morbidity and mortality, leading to liver failure or septic complications requiring urgent retransplantation. Experimental evidence that hyperbaric oxygen (HBO) may ameliorate hepatic ischemic-reperfusion injury led to this study of HBO in pediatric liver transplant recipients who developed HAT. Children undergoing OLT under primary tacrolimus immunosuppression and University of Wisconsin organ preservation between August 1, 1989, and December 31, 1998, who developed HAT were the basis for this study. Patients who developed HAT between March 1, 1994, and December 31, 1998, were treated with HBO therapy until signs of ischemia resolved (absence of fever, normalizing liver injury test results) or for 2 weeks. The pediatric OLTs performed from August 1, 1989, to February 28, 1994, who developed HAT served as a control group. Primary outcome measures were survival, retransplantation rate, time to retransplantation, incidence of hepatic gangrene, and days to collateral formation. Three hundred seventy-five consecutive pediatric patients underwent 416 OLTs between August 1, 1989, and December 31, 1998. Thirty-one patients (7.5%) developed HAT at a mean time of 8.2 days (range, 1 to 52 days) post-OLT. In 17 patients, HBO treatment was begun within 24 hours of HAT or immediately after the revascularization attempt and performed twice daily for 90 minutes at 2.4 atmospheres pressure. Fourteen patients were treated without HBO. None of the HBO-treated patients developed hepatic gangrene. Eight HBO patients (47%) were bridged to retransplantation at a mean time of 157 days (range, 3 to 952 days) after initial OLT and all survived. Mean time to retransplant in the control group was 12.7 days (range, 1 to 64 days). HBO was well tolerated without significant complications. Although there was no significant difference in survival or retransplantation rates, HBO significantly delayed retransplantation, potentially by hastening the development of hepatic artery collaterals.

Hepatic resection for metastatic colorectal adenocarcinoma: a proposal of a prognostic scoring system

BACKGROUND

Hepatic resection for metastatic colorectal cancer provides excellent longterm results in a substantial proportion of patients. Although various prognostic risk factors have been identified, there has been no dependable staging or prognostic scoring system for metastatic hepatic tumors.

STUDY DESIGN

Various clinical and pathologic risk factors were examined in 305 consecutive patients who underwent primary hepatic resections for metastatic colorectal cancer. Survival rates were estimated by the Cox proportional hazards model using the equation: S(t) = [So(t)]exp(R-Ro), where So(t) is the survival rate of patients with none of the identified risk factors and Ro = 0.

RESULTS

Preliminary multivariate analysis revealed that independently significant negative prognosticators were: (1) positive surgical margins, (2) extrahepatic tumor involvement including the lymph node(s), (3) tumor number of three or more, (4) bilobar tumors, and (5) time from treatment of the primary tumor to hepatic recurrence of 30 months or less. Because the survival rates of the 62 patients with positive margins or extrahepatic tumor were uniformly very poor, multivariate analysis was repeated in the remaining 243 patients who did not have these lethal risk factors. The reanalysis revealed that independently significant poor prognosticators were: (1) tumor number of three or more, (2) tumor size greater than 8 cm, (3) time to hepatic recurrence of 30 months or less, and (4) bilobar tumors. Risk scores (R) for tumor recurrence of the culled cohort (n = 243) were calculated by summation of coefficients from the multivariate analysis and were divided into five groups: grade 1, no risk factors (R = 0); grade 2, one risk factor (R = 0.3 to 0.7); grade 3, two risk factors (R = 0.7 to 1.1); grade 4, three risk factors (R= 1.2 to 1.6); and grade 5, four risk factors (R > 1.6). Grade 6 consisted of the 62 culled patients with positive margins or extrahepatic tumor. Kaplan-Meier and Cox proportional hazards estimated 5-year survival rates of grade 1 to 6 patients were 48.3% and 48.3%, 36.6% and 33.7%, 19.9% and 17.9%, 11.9% and 6.4%, 0% and 1.1%, and 0% and 0%, respectively (p < 0.0001).

CONCLUSIONS

The proposed risk-score grading predicted the survival differences extremely well. Estimated survival as determined by the Cox proportional hazards model was similar to that determined by the Kaplan-Meier method. Verification and further improvements of the proposed system are awaited by other centers or international collaborative studies.

Simultaneous administration of adjuvant donor bone marrow in pancreas transplant recipients

OBJECTIVE

The effect of donor bone marrow was evaluated for its potentially favorable effect in the authors' simultaneous pancreas/kidney transplant program.

METHODS

From July 1994 to January 1999, 177 pancreas transplants were performed, 151 of which were simultaneous pancreas/kidney transplants. All patients received tacrolimus, mycophenolate mofetil, and steroids for immunosuppression (azathioprine was used in the first year of the program). Fifty-three simultaneous pancreas/kidney transplant recipients received perioperative unmodified donor bone marrow, 3 to 6 x 10(8) cells/kg.

RESULTS

Overall actuarial survival rates at 1 and 3 years were 98% and 95% (patient), 95% and 87% (kidney), and 86% and 80% (pancreas), respectively. In the adjuvant bone marrow group, 1- and 3-year survival rates were 96% and 91 % (patient), 95% and 87% (kidney), and 83% and 83% (pancreas), respectively. For 98 recipients who did not receive bone marrow, survival rates at 1 and 3 years were 100% and 98% (patient), 96% and 86% (kidney), and 87% and 79% (pancreas), respectively. No pancreas allografts were lost after 3 months in bone marrow recipients, and seven in the non-bone marrow recipients were lost to rejection at 0.7, 6.7, 8.8, 14.6, 24.1, 24.3, and 25.5 months. Twenty-two percent of bone marrow patients were steroid-free at 1 year, 45% at 2 years, and 67% at 3 years. Nineteen percent of the non-bone marrow recipients were steroid-free at 1 year, 38% at 2 years, and 45% (p = 0.02) at 3 years. The mean acute cellular rejection rate was 0.94+/-1.1 in the bone marrow group and 1.57+/-1.3 (p = 0.003) in the non-bone marrow group (includes borderline rejection and multiple rejections). The level of donor cell chimerism in the peripheral blood of bone marrow patients was at least two logs higher than in controls.

CONCLUSION

In this series, which represents the largest experience with adjuvant bone marrow infusion in pancreas recipients, there was a higher steroid withdrawal rate (p = 0.02), fewer rejection episodes, and no pancreas graft loss after 3 months in bone marrow recipients compared with contemporaneous controls. All pancreas allografts lost to chronic rejection (n = 6) were in the non-bone marrow group. Donor bone marrow administered around the time of surgery may have a protective effect in pancreas transplantation.

Hepatobiliary malignancies. Primary hepatic malignant neoplasms

Primary hepatic tumors are epithelial, mesenchymal, or mixed in origin. Of these, epithelial tumors are the most common and include hepatocellular carcinoma, cholangiocarcinoma, mixed hepatocholangiocarcinoma, hepatoblastoma, and a variety of more rare tumors. Hepatocellular carcinoma, also know as hepatoma or malignant hepatoma, is the most common, followed by cholangiocarcinoma. This article discusses these two malignancies.

Strategies to expand the donor pool for pancreas transplantation

BACKGROUND

Our organ procurement organization has been forced to liberalize the donor criteria in order to expand the donor pool for pancreas transplantation. In this report, we describe our experience using whole organ pancreatic grafts from "marginal" donors, which include grafts obtained from donors over 45 years of age and from donors who were identified to be hemodynamically unstable at the time of organ retrieval.

METHODS

A prospective study was performed between July 1994 and March 1998, during which time 137 pancreas transplants were performed at our center using organs procured by our own surgeons (organs sent by other teams were excluded). The rapid en bloc technique was used exclusively. The use of pancreatic grafts from marginal donors was analyzed for short-term and overall graft survival, and for delayed graft function and complications.

RESULTS

Overall pancreas graft survival for our series was 83%, with a mean follow-up of 23 months. There were 22 pancreas grafts from donors over 45 years of age, 13 of whom were greater than 50 years of age. The actual graft survival rate of the over-45 donor group was 86%. Fifty-one grafts were removed from hemodynamically unstable donors on high-dose vasopressors. The actual graft survival in this group was 86%. There was no significant difference found in graft survival between recipients of pancreatic grafts from marginal and nonmarginal donors. Delayed graft function was exhibited by more recipients of grafts from donors on high-dose vasopressors (P<0.05), but this had no effect on long-term graft survival and endocrine function. Recipients of marginal donor grafts did not have higher rates of complication compared to recipients of nonmarginal grafts.

CONCLUSIONS

Based on our results, we currently employ a graft selection strategy not limited by donor age or hemodynamic stability. Our selection of pancreas organs for transplantation is based on careful inspection of the pancreas and determination of the adequacy of the ex vivo flush. Our results suggest that the current pancreas donor pool may be expanded substantially.

Factors impacting the survival of children with intestinal failure referred for intestinal transplantation

PURPOSE

The aim of this study was to analyze factors impacting on the survival of pediatric patients with intestinal failure referred for intestinal transplantation (ITx).

METHODS

Two hundred fifty-seven children (mean age, 3.4+/-0.26 years) with intestinal failure were evaluated for ITx between 1990 and 1998. All patients were dependent on total parenteral nutrition (TPN) for a mean of 31+/-2.7 months. The mean follow-up time from the date of evaluation was 9.2+/-0.9 months.

RESULTS

Eighty-two (32%) children underwent ITx with a mean waiting time of 10.1+/-1.3 months (simultaneous liver-intestinal allograft in 68% instances). Of the 175 patients who did not undergo transplantation, 120 died, 23 were lost to follow-up, and 32 are alive. Younger patients (< or =1 year) had poorer survival rates than patients older than 1 year (P<.0001). The patients with the worse prognosis were those with necrotizing enterocolitis, and those with the best prognosis were those with Hirschsprung's disease. Patients with "surgical" causes had poorer survival rates than those with "nonsurgical" causes (P<.04). Patients with bridging fibrosis or established cirrhosis had an earlier mortality than patients with portal fibrosis (P<.003). The worst survival rate was found for patients with bilirubin levels of greater than 3 mg/dL (P<.0001), plateletcounts less than 100.000/mL (P<.0001), prothrombin time greater than 15 seconds (P = .03) or partial thromboplastin time greater than 40 seconds (P<.04). Children who at the time of evaluation needed only an isolated intestinal allograft had a better prognosis than those who required a combined liver-intestine allograft (P<.00001). With multivariate analysis independent prognosis risk factors of poor outcome were hyperbilirubinemia and severity of histopathologic damage.

CONCLUSIONS

Early referral for ITx should occur before the development of liver dysfunction, taking into consideration the aforementioned risk factors that would facilitate the development and ominous evolution to liver failure.

Hepatic hemangioendothelioma: clinical experience and management strategy

PURPOSE

This study sought to define management strategies based on clinical experience in treating infantile hepatic hemangioendothelioma.

METHODS

A retrospective analysis of patients with hemangioendothelioma presenting to a tertiary liver transplantation center between 1989 and 1997 was performed.

RESULTS

Thirteen patients (median age, 14 days) with hemangioendothelioma were identified. Congestive heart failure (P<.03) and abdominal mass (P<.081) were predictive of 5-month mortality rates. Ultrasonography and computerized axial tomography were the diagnostic modalities most commonly used. Treatment strategies consisted of medical management (steroids and alpha-interferon) and interventional modalities (hepatic artery ligation or embolization, resectional surgery, or orthotopic liver transplantation). Patients who underwent resectional surgery, with or without orthotopic liver transplantation, had a lower 5-month mortality rate (P<.02) and a greater 2-year survival rate (P<.003) than did those who underwent hepatic artery ligation or embolization. Early morbidity and mortality tended to be a consequence of the primary lesion, whereas late morbidity and mortality were reflective of the treatment modality used.

CONCLUSIONS

In cases of failed medical management, resectional therapy should be used when possible. If partial hepatectomy is not technically achievable, hepatic artery embolization should be used either as definitive therapy or as a temporizing measure until orthotopic liver transplantation is possible.

Undetected hepatocellular carcinoma: clinical features and outcome after liver transplantation

The aim of the study was to define the clinical characteristics and outcome of patients found to have an undetected hepatocellular carcinoma (HCC) at liver transplantation. Patients who underwent liver transplantation and were found to have a hepatoma with a prior workup showing normal alpha-fetoprotein levels and no corresponding lesion on radiological evaluation were defined as having an undetected HCC. Detailed information was collected, and the last abdominal computed tomographic (CT) scan before transplantation was performed was retrospectively reviewed. Thirty-nine patients had a tumor that met the criteria for an undetected hepatoma. The most common causes for pretransplantation liver disease were hepatitis C virus (HCV) (49%) and alcohol use (28%). Tumor size was 2 cm or less in 85% of the patients, vascular invasion was detected in 31% of the patients, and tumor, node, metastasis (TNM) classification was stage I or II in 77% of the patients. Review of the last CT scan before transplantation showed that the lesion was evident in retrospect in only 15% of the patients. Thirty-two patients (82%) remained alive at the time of the study with a mean follow-up of 30 months. Metastatic HCC was detected in 1 patient 7 months after transplantation. There were no other tumor recurrences. Survival analysis showed no significant differences when tumor size, stage, presence of vascular invasion, or causes of pretransplantation liver disease were compared. Undetected HCCs represent a significant percentage of total hepatomas in patients undergoing liver transplantation. Most patients have small, early-stage tumors, but tumors greater than 2 cm or of advanced stage are also frequently found in this population. Overall and tumor-free survival appear to be favorable.

Treatment of hilar cholangiocarcinoma (Klatskin tumors) with hepatic resection or transplantation

BACKGROUND

Because of the rarity of hilar cholangiocarcinoma, its prognostic risk factors have not been sufficiently analyzed. This retrospective study was undertaken to evaluate various pathologic risk factors which influenced survival after curative hepatic resection or transplantation.

METHODS

Between 1981 and 1996, 72 patients (43 males and 29 females) with hilar cholangiocarcinoma underwent hepatic resection (34 patients) or transplantation (38 patients) with curative intent. Medical records and pathologic specimens were reviewed to examine the various prognostic risk factors. Survival was calculated by the method of Kaplan-Meier using the log rank test with adjustment for the type of operation. Survival statistics were calculated first for each kind of treatment separately, and then combined for the calculation of the final significance value.

RESULTS

Survival rates for 1, 3, and 5 years after hepatic resection were 74%, 34%, and 9%, respectively, and those after transplantation were 60%, 32%, and 25%, respectively. Univariate analysis revealed that T-3, positive lymph nodes, positive surgical margins, and pTNM stage III and IV were statistically significant poor prognostic factors. Multivariate analysis revealed that pTNM stage 0, I, and II, negative lymph node, and negative surgical margins were statistically significant good prognostic factors. For the patients in pTNM stage 0-II with negative surgical margins, 1-, 3-, and 5-year survivals were 80%, 73%, and 73%, respectively. For patients in pTNM stage IV-A with negative lymph nodes and surgical margins, 1-, 3-, and 5-year survivals were 66%, 37%, and 37%, respectively.

CONCLUSIONS

Satisfactory longterm survivals can be obtained by curative surgery for hilar cholangiocarcinoma either with hepatic resection or liver transplantation. Redefining pTNM stage III and IV-A is proposed to better define prognosis.

Liver transplantation in the treatment of hepatocellular carcinoma

We aimed to determine the most appropriate candidates for liver transplantation based on their survival outcomes. Two hundred and fourteen patients who were transplanted in the presence of hepatocellular carcinoma (HCC) were analyzed. Patient groups were selected as "good risk" candidates for transplantation by our previously developed artificial network model or by the classic pTNM pathological classification system. The survival of the model-selected candidate groups was then compared to the survival of the candidates chosen as "good risk" by the pTNM classification (i.e. , pTNM stages I + II and pTNM stages I + II + III). Suitability for transplantation was judged by long-term survival rates (i.e., 1-10 years post-transplant). By using the neural network prediction model and the subsequent subgroup case analysis, it was possible to generate those combinations of risk factors which predetermined patient survival through HCC recurrence. By applying the developed neural network model to the transplant candidate pool for patients with HCC, it was possible to select the maximum number of suitable candidates for transplantation while minimizing donor organ loss to recurrent HCC.

Hepatic resection and transplantation for peripheral cholangiocarcinoma

BACKGROUND

Recent publications have questioned the role of orthotopic liver transplantation (OLT) in treating advanced or unresectable peripheral cholangiocarcinoma (Ch-Ca).

STUDY DESIGN

We reviewed our experience with Ch-Ca to determine survival rates, recurrence patterns, and risk factors in 54 patients who underwent either hepatic resection or OLT between 1981 and 1994. Liver transplantation was performed in patients with unresectable tumors (n = 12) and in those with advanced cirrhosis (n = 8). There were 33 women (61%) and 21 men (39%), with a mean age of 54.3 years. The median followup period was 6.8 years. Prognostic risk factors were analyzed by univariate and multivariate analyses.

RESULTS

Mortality within 30 days was 7.4%. Overall patient and tumor-free survival rates were 64% and 57% at 1 year, 34% and 34% at 3 years, and 26% and 27% at 5 years after operation. Thirty-two patients (59.3%) experienced tumor recurrence. Univariate analysis revealed that multiple tumors, bilobar tumor distribution, regional lymph node involvement, presence of metastasis, positive surgical margins, and advanced pTNM stages were significant negative predictors of both tumor-free and patient survival. Multivariate analysis revealed that positive margins, multiple tumors, and lymph node involvement were independently associated with poor prognosis. When patients with these three negative predictors were excluded, the patient survivals at 1, 3, and 5 years were 74%, 64%, and 62%, respectively.

CONCLUSIONS

Both hepatic resection and OLT are effective therapies for Ch-Ca when the tumor can be removed with adequate margins, the lesion is singular, and lymph nodes are not involved.

Hepatic resection and transplantation for peripheral cholangiocarcinoma

BACKGROUND

Recent publications have questioned the role of orthotopic liver transplantation (OLT) in treating advanced or unresectable peripheral cholangiocarcinoma (Ch-Ca).

STUDY DESIGN

We reviewed our experience with Ch-Ca to determine survival rates, recurrence patterns, and risk factors in 54 patients who underwent either hepatic resection or OLT between 1981 and 1994. Liver transplantation was performed in patients with unresectable tumors (n = 12) and in those with advanced cirrhosis (n = 8). There were 33 women (61%) and 21 men (39%), with a mean age of 54.3 years. The median followup period was 6.8 years. Prognostic risk factors were analyzed by univariate and multivariate analyses.

RESULTS

Mortality within 30 days was 7.4%. Overall patient and tumor-free survival rates were 64% and 57% at 1 year, 34% and 34% at 3 years, and 26% and 27% at 5 years after operation. Thirty-two patients (59.3%) experienced tumor recurrence. Univariate analysis revealed that multiple tumors, bilobar tumor distribution, regional lymph node involvement, presence of metastasis, positive surgical margins, and advanced pTNM stages were significant negative predictors of both tumor-free and patient survival. Multivariate analysis revealed that positive margins, multiple tumors, and lymph node involvement were independently associated with poor prognosis. When patients with these three negative predictors were excluded, the patient survivals at 1, 3, and 5 years were 74%, 64%, and 62%, respectively.

CONCLUSIONS

Both hepatic resection and OLT are effective therapies for Ch-Ca when the tumor can be removed with adequate margins, the lesion is singular, and lymph nodes are not involved.

Treatment of fibrolamellar hepatoma with subtotal hepatectomy or transplantation

Fibrolamellar hepatoma (FL-HCC) is an uncommon variant of hepatocellular carcinoma (HCC), distinguished by histopathological features suggesting greater differentiation than conventional HCC. However, the optimal treatment and the prognosis of FL-HCC have been controversial. Follow-up studies are available from 1 year to 27 years, after 41 patients with FL-HCC were treated with partial hepatectomy (PHx) (28 patients) or liver transplantation (13 patients). In this retrospective study, the effect on outcome was determined for the pTNM stage and other prognostic factors routinely recorded at the time of surgery. Cumulative survival at 1, 3, 5, and 10 years was 97.6%, 72.3%, 66.2%, and 47.4%. Tumor-free survival at these times was 80.3%, 49.4%, 33%, and 29.3%. The TNM stage was significantly associated with tumor-free survival. Patients with positive nodes had a shorter tumor-free survival than those with negative nodes (P < .015). Patient survival was most adversely affected by the presence of vascular invasion (P < .05). FL-HCC is an indolently growing tumor of the liver, which usually was diagnosed in our patients at a stage too advanced for effective surgical treatment of most conventional HCC. Nevertheless, long-term survival frequently was achieved with aggressive surgical treatment. When a subtotal hepatectomy could not be performed, total hepatectomy (THx) with liver transplantation was a valuable option.

The prediction of risk of recurrence and time to recurrence of hepatocellular carcinoma after orthotopic liver transplantation: a pilot study

Orthotopic liver transplantation (OLTx) in the presence of hepatocellular carcinoma (HCC) has been complicated by high recurrence rates. The ability to determine the risk and timing of HCC recurrence on an individual basis would greatly aid in the candidate selection process resulting in a more efficient use of donated organs and allow the individualization and better evaluation of adjuvant chemotherapy. The 214 patients who underwent OLTx in the presence of HCC were analyzed. From the 178 patients who survived more than 150 days, 71 (40%) have suffered HCC recurrence. Based on five risk factors, that is, gender, tumor number, lobar tumor distribution, tumor size, grade of vascular invasion, artificial neural network models predicting the likelihood of HCC recurrence within 1, 2, and 3 consecutive years after transplantation were developed. Based on model predictions, those combinations of risk factors that should/should not lead to recurrence were generated, allowing stratification of patients into the following three groups: 1) patients who should not suffer HCC recurrence and who should not need adjuvant therapy, 2) patients who will suffer recurrence and for whom postoperative chemotherapy significantly prolonged survival (but did not prevent recurrence), and 3) patients who may or may not suffer HCC recurrence and whose recurrence may be prevented by adjuvant chemotherapy. The outcome of OLTx for patients with HCC can be prognosticated based on a number of clinical variables. If verified through multicenter trials, these models could be made available to transplantation programs performing OLTx in the presence of HCC.

Liver transplantation for acute liver failure: outcome analysis

Acute liver failure (ALF) remains a major cause of morbidity and mortality. Before the availability of liver transplantation only 20% of patients with ALF survived. The clinical dilemma relates to the prognostication of these patients, as early liver transplantation has been associated with better outcomes. The eligibility for liver transplantation must therefore be quickly established. The patient's age, aetiology of disease, interval between the onset of jaundice and encephalopathy, blood pH, prothrombin time, serum bilirubin and serum creatinine levels has been identified as useful prognostic markers. The degree of hepatocyte necrosis on liver biopsy and estimated hepatic volume by computed tomography may also be valuable predictors of survival; however, further studies are needed.

Multisolutional clustering and quantization algorithm (MCQ)

We have developed a novel clustering and quantization algorithm that allows the user to create multiple one-to-one correspondences between the actual data and its transformed (clustered and quantized) values, based on the user's hypothesis regarding the nature of the classification task. The types of problems for which the algorithm can be beneficial are discussed. We report experiments employing simulated and real data that suggest the proposed algorithm may be useful in neural network analysis of various phenomena in medicine and biology.