- Email: [email protected]
Nodular regenerative hyperplasia associated with aplasic anemia: Case report
574
Journal of Gastrointestinal Surgery
Abstracts
1 case of metastases from breast cancer and 1 case of metastase from kidney cancer (together with caval thrombectomy), 3 left lobectomies, 9 bisegmentectomies, and 8 “wedge resections.” There was neither perioperative mortality nor morbidity (incidence of wound infection: 3.5%; incidence of biliary leakage: 0%). Intraoperative blood losses ranged from 200 to 800 ml, while mean postoperative hospital stay was 8.5 days. One patient died 3 months after adrenalectomy and bisegmentectomy for recurrent adrenal cancer, 1 patient died of peritoneal carcinosis 1 year after “wedge resections” for metastases from uterine cancer, 1 patient died of recurrent breast cancer 3.5 years after hepatic resection, 1 patient died 11 months after central hepatectomy for metastatic renal cancer, 1 patient died with lung metastases 2 months after “wedge resections” and adrenalectomy for metastatic renal cancer, 1 patient with ampullary cancer died of peritoneal carcinosis 6 months after surgery, and 1 patient had bilateral hepatic metastases 24 months after extended right hepatectomy for metastases from breast cancer treated 12 years before. Although we have initially observed an incidence of cancer-related deaths after resective surgery of noncolorectal hepatic metastases higher than reported by other authors, it is not yet possible to draw statistically significant conclusions from our limited experience, due to the short follow-up period.
174 OUTCOMES OF HEPATIC RESECTION FOR COLORECTAL METASTASIS FOLLOWING NEOADJUVANT CHEMOTHERAPY Prosanto Chaudhury, MD, Giuseppina Campisi, MS, Tammy Znajda, MD, John Martinie, MD, Caroline Rochon, MD, Peter Horton, MD, Peter Metrakos, MD, McGill University Health Centre, Montreal, PQ, Canada The use of systemic chemotherapy prior to resection of colorectal cancer (CRC) metastasis to the liver is becoming more and more prevalent. Despite this, many surgeons remain concerned about potentially increased morbidity and mortality associated with neoadjuvant therapy. The purpose of this study was to review our perioperative complications in patients receiving neoadjuvant systemic chemotherapy within 6 months of liver resection. Chemotherapy consisted of 5-fluouracil (5-FU), leucovorin (LCV), with or without irinotecan (CPT-11) or oxaliplatin. A retrospective chart review of consecutive patients undergoing liver resection for CRC metastasis from January 2002 to July 2004 was performed. 120 patients were identified. Perioperative parameters, patient demographics, tumor characteristics, and details of chemotherapy were measured in addition to morbidity and mortality. The analysis of the first 40 patients is presented here; 40 patients were identified and their charts reviewed. Sixteen patients (40%) received no chemotherapy, 12 (30%) received 5-FU/LCV, and 12 (30%) received 5-FU/LCV/CPT-11 or oxaliplatin. Patients receiving 5-FU/LCV/CPT-11 or oxaliplatin tended to have larger and more numerous tumors. Overall complication rates between groups were not significantly different. Hospital length of stay was similar for both groups. Intraoperative blood loss was also similar for both groups. There was a trend towards an increased rate of wound infection in the group receiving neoadjuvant chemotherapy (16% vs. 21%) but this did not reach statistical significance. There was no perioperative mortality in either group. From this analysis, we conclude that an aggressive use of neoadjuvant chemotherapy (5-FU/ LCV ⫾ CPT-11 or oxaliplatin) prior to liver resection does not result in increased morbidity or mortality. This strategy offers several potential advantages. Neoadjuvant chemotherapy may downstage tumors making unresectable lesions resectable. It may also allow more appropriate selection of patients for surgery: those who progress to extrahepatic disease during a short course of chemotherapy were unlikely to benefit from liver resection in the first place. The analysis of the remaining 80 patients will be presented at the meeting.
175 RADIOFREQUENCY: NEW TECHNOLOGY FOR THE TREATMENT OF LIVER TUMORS Mariano Moro, Rodrigo Sa´nchez Claria´, Martı´n Palavecino, Christian Bertona, Juan Pekolj, Eduardo de Santiban˜es, Hospital Italiano, Buenos Aires, Argentina Only 30-40% of patients with malignant hepatic tumors (primary or secondary) are candidates to a resection with curative intention. This stimulates the development of new technologies to perform a local ablation to modify the evolution. We sought to evaluate the results of the thermoablation with radiofrequency (RF) in the treatment of malignant liver tumors, even primary and secondary. The design was retrospective and observational. Between January 1999 and December 2003, 54 patients were treated with RF. Population was divided according to the original tumor: primary hepatic tumor (PHT), colorectal metastasis (CRM) and nonprimary noncolorectal metastasis (NPNCM). RF was applied alone or associated with another ablative procedure. RF was performed by laparotomy, laparoscopy or using percutaneous puncture. Mean age was 64 (range: 45-80). The tumors were CRM in 32 patients (60%), PHT in 14 (26%), and NPNCM in 8 (14%). RF was percutaneous in 8 (14%), laparoscopic in 4 (9%) and applied by laparotomy in 42 patients (77%). In 31 patients (57%) RF was the only therapy and in 23 (43%) it was associated with another resective or local ablative procedure. Mean hospital stay was 4 days (range: 3-10). Complication rate was 11%. There was no mortality. No reoperations were needed. Mean survival in PHT was 16.5 months, and the global overall survival rate was 47% at 4 years. In CRM mean survival was 16.3 months and overall survival rate was 27%. In NPNCM mean survival was 16.8 months and overall survival rate was 53%. Thermoablation with RF should be considered in patients with no chances of resection. This procedure can be associated with other resective procedures. Laparotomy is the safest way to perform it. The RF has a low rate of complications and it is safe in selected patients.
176 NODULAR REGENERATIVE HYPERPLASIA ASSOCIATED WITH APLASIC ANEMIA: CASE REPORT S. Gonzalez-Huezo, R. Mondragon, L. Villela, C. Carrillo, Centro Medico Issemym, Metepec, Edo Mexico, Mexico Nodular regenerative hyperplasia (NRH) is a rare condition associated with immune, neoplastic, and vascular disorders, as well as with drugs and toxins. Pathogenesis is related to abnormal hepatic blood flow. Portal hypertension is the main complication. We report the case of a woman with NRH and aplastic anemia, association not previously described. A 31-year-old Hispanic female presented with pancitopenia and cholestasis. PMH was relevant for epilepsy treated with phenytoin for 5 years, which was discontinued 7 years before presentation. At admission she was pale and slightly jaundiced with diffuse petechiae. Lab report was as follows: CBC: leukocytes 2700/µl (with 500 total PMN cells); platelets 35,000/µl; Hb 6.7 g/dl; reticulocytes 0%. LFT: ALT 61 U/L, AST 70 U/L, Tot. Bil. 3.3 g/dl, Dir. Bil. 1.0 g/dl, Alk. Phos. 203 U/L, Alb 3.7 g/dl, globulins 3.9 g/dl, INR 1.4. Bone marrow biopsy was diagnostic of aplastic anemia. Abdominal ultrasound revealed multiple liver nodules, without hepato-splenomegaly. No abnormalities were noted in hepatic vessels. Serology for HBV and HCV was negative. Antimitochondrial (AMAs) and antinuclear antibodies were positive. Coagulation tests, antiphospholipid antibodies, C and S proteins, antithrombin III and factor V Leiden were normal. Liver biopsy showed hepatocyte nodules with central atrophy and sinusoidal dilatation in the absence of significant fibrosis. No arteritis, thrombosis or bile duct damage was found. Anti-thymocyte antiglobulin and cyclosporine were started with partial response at 5 weeks. Pathogenesis of NRH remains unclear. The most accepted theory was developed
Vol. 9, No. 4 2005
by Wanless 24 years ago, and reinforced in 1997. It states that alterations in hepatic blood flow (portal obliterative venopathy) originate a specific response, initially consisting of apoptosis secondary to acute ischemia, with ensuing parenchymal atrophy and surrounding areas of liver regeneration and chronic ischemia. The common pathologic mechanism in all the associated diseases in NRH involves liver flow abnormalities secondary to microthromboses, vascular congestion, or vasculitis. In the current case we were unable to document evidence of liver flow abnormalities. Noteworthy there are previous descriptions of the association of NRH with primary biliary cirrhosis, and this patient was AMA positive, but without bile duct damage in the biopsy. Uncommonly, the only laboratory abnormality in NRH can be cholestasis. This case is relevant because we cannot sustain Wanless’ theory and because it is the first description of the association of NRH with aplastic anemia.
177 A CASE OF PSEUDOLYMPHOMA OF THE LIVER Tsuyoshi Hatsuno, Nagoya Medical Center, Nagoya, Japan Pseudolymphoma (reactive lymphoid hyperplasia) of the liver is an extremely rare entity, with a few cases reported so far. We encountered a 50-year-old Japanese female with pseudolymphoma of the liver. A tumor-like lesion was discovered incidentally as a hypoechoic mass with a hyperechoic rim during clinical follow-up of endometriosis. Radiological studies, such as contrast-enhanced computed tomography (CT) and angiography demonstrated a hypervascular lesion. The hepatic lesion was resected because hepatocellular carcinoma was suspected after a needle biopsy. Grossly, the lesion was well defined and measured 1.0 × 1.5 cm. Microscopically, the lesion consisted of hyperplastic lymphoid follicles with distinctive germinal centers and interfollicular areas consistng of mature lymphocytes and plasma cells. An immunohistological study revealed that the lymphoid cells of the lesion were polyclonal in immunophenotypes. These histological and immunohistochemical findings strongly suggested a pseudolymphoma of the liver. The following features have characterized the images in past cases, as well as ours: hypoechoic mass, occasionally with a rim, in ultrasonography and hypervascularity, shown by angiography and enhanced CT.
178 TOO MANY COOKS SPOIL THE BROTH: THE DILEMMA OF SELECTING PATIENTS WITH HEPATOCELLULAR CARCINOMA FOR LIVER TRANSPLANTATION Claudia Hriesik, MD, PhD, Burkhardt Ringe, MD, Drexel University College of Medicine, Philadelphia, PA Criteria for liver transplantation in patients with hepatocellular carcinoma (HCC) differ amongst staging systems and amongst transplant centers. Originally used staging systems, e.g., TNM/UICC, are replaced by selection criteria described by Mazzaferro and the Milan criteria. The latter, however, are too tight and might breed the dilemma to falsely withhold an organ. Examining different selection criteria, it becomes obvious that a potential risk is that selected criteria may be used as one pleases in order to achieve improved results or in order to expand patient selection. This in return poses the risk of wasting an organ. Furthermore, established staging criteria (e.g., Milan criteria, Barcelona criteria) are conflicting as they use postoperative data, e.g., histological results, in order to select therapy. Thus, a comparison of results and outcome amongst transplant centers following liver transplantation and an outlook on prognosis is difficult if not impossible. The aim of this study was to analyze currently used
Abstracts
575
selection criteria for liver transplantation and to determine their reliability related to patient selection, therapy selection, and prognosis. We have reviewed the medical literature since 1954 to identify criteria published for the staging of HCC. To evaluate and compare the reliability of these criteria we focused on two questions: (1) Can the criteria be used in the decision making process for liver transplantation before and for treatment, and (2) How accurately can these criteria predict prognosis? As a tool we used the certainty factor (C factor) defining the level of diagnostic judgment and reliability: C1 clinical, C2 radiological, C3 biopsy, C4 resection, C5 autopsy. As an additional criterion liver function parameters were selected. Twentyone staging systems were used for this comparative analysis. 1) Five staging systems can be used for therapy selection (C1-C3). 2) Three systems provided prognosis assessment only (C4-C5). 3) Thirteen staging systems allowed to both assist in decision making for therapy based on clinical criteria as well as the assessment of a prognosis based on pathological criteria (C1-C5). 4) Only five staging systems assessed tumors factors as well as liver function (C1-C5, LF). The ideal staging system should 1) facilitate the selection of transplant candidates based on tumor factors and clinical parameters and 2) predict long term prognosis as indicated by pathological tumor extent. It would be desirable to use a uniform staging system as suggested by the AHPBA consensus statement of 2003.
179 RADIOFREQUENCY ABLATION IN THE PORCINE MODEL: COMPARISON OF MULTIPLE COOL TIP PROBES AND A SWITCHING CONTROLLER WITH A STANDARD THREE-PROBE CLUSTER David A. Iannitti, MD, Dionisios Vrochides, MD, PhD, Jason McKee, MD, Damian E. Dupuy, MD, Brown Medical School, Providence, RI Evaluation of the Switching Controller with multiple, combined cool tip probes for an output-based radiofrequency device. Radiofrequency (RF) ablation was performed in three randomly selected segments of the liver in 12 pigs. Half of the ablations were performed with a Pringle maneuver and half without. Ablations were performed with three 3.0 cm active tip, cooled RF probes spaced 2.0 cm apart and a 200 watt RF generator. Ablations using the standard cluster of probes were performed for 12 minutes in the non-Pringle subgroup and 6 minutes in the Pringle subgroup. Ablations using the Switching Controller with multiple, combined cool tip probes were performed for 16 minutes in the non-Pringle subgroup and 8 minutes in the Pringle subgroup. The Switching Controller delivered current that alternated among the probes, based on impedance (to 30 Ohms above baseline) and time (30 seconds maximum). The ablation dimensions, volume, temperature, current delivery parameters, and postablation histology were evaluated. A total of 34 successful ablations were performed. The minimum ablation diameter (Dmin) in the standard cluster group was 40.3 ⫾ 4.3 mm (42.5 ⫾ 4.1 mm for the Pringle group and 38.1 ⫾ 3.6 mm for the non-Pringle group). The maximum ablation surface area (SAmax) in this group was 16.0 ⫾ 3.3 cm2 (17.6 ⫾ 2.2 sqcm for the Pringle group and 14.2 ⫾ 3.5 cm2 for the nonPringle group). The ablation volume in the cluster group was 36.9 ⫾ 9.5 ml (40.1 ⫾ 8.1 ml for the Pringle group and 33.7 ⫾ 10.4 ml for the non-Pringle group). In the Switching Controller Multiprobe group, the ablation Dmin was 51.4 ⫾ 7.4 mm (54.5 ⫾ 8.5 mm for the Pringle group and 48.2 ⫾ 4.8 mm for the non-Pringle group). The ablation SAmax in this group was 22.4 ⫾ 4.7 cm2 (23.9 ⫾ 4.9 cm2 for the Pringle group and 20.8 ⫾ 4.1 cm2 for the non-Pringle group). The ablation volume in the multiprobe group was 66.1 ⫾ 17.9 ml (68 ⫾ 16.8 ml for the Pringle group and 64.3 ⫾ 19.6 ml for the nonPringle group). Technology to increase the size of ablated areas and shorten the procedural time continues to evolve. The use of the
Journal of Gastrointestinal Surgery
Abstracts
1 case of metastases from breast cancer and 1 case of metastase from kidney cancer (together with caval thrombectomy), 3 left lobectomies, 9 bisegmentectomies, and 8 “wedge resections.” There was neither perioperative mortality nor morbidity (incidence of wound infection: 3.5%; incidence of biliary leakage: 0%). Intraoperative blood losses ranged from 200 to 800 ml, while mean postoperative hospital stay was 8.5 days. One patient died 3 months after adrenalectomy and bisegmentectomy for recurrent adrenal cancer, 1 patient died of peritoneal carcinosis 1 year after “wedge resections” for metastases from uterine cancer, 1 patient died of recurrent breast cancer 3.5 years after hepatic resection, 1 patient died 11 months after central hepatectomy for metastatic renal cancer, 1 patient died with lung metastases 2 months after “wedge resections” and adrenalectomy for metastatic renal cancer, 1 patient with ampullary cancer died of peritoneal carcinosis 6 months after surgery, and 1 patient had bilateral hepatic metastases 24 months after extended right hepatectomy for metastases from breast cancer treated 12 years before. Although we have initially observed an incidence of cancer-related deaths after resective surgery of noncolorectal hepatic metastases higher than reported by other authors, it is not yet possible to draw statistically significant conclusions from our limited experience, due to the short follow-up period.
174 OUTCOMES OF HEPATIC RESECTION FOR COLORECTAL METASTASIS FOLLOWING NEOADJUVANT CHEMOTHERAPY Prosanto Chaudhury, MD, Giuseppina Campisi, MS, Tammy Znajda, MD, John Martinie, MD, Caroline Rochon, MD, Peter Horton, MD, Peter Metrakos, MD, McGill University Health Centre, Montreal, PQ, Canada The use of systemic chemotherapy prior to resection of colorectal cancer (CRC) metastasis to the liver is becoming more and more prevalent. Despite this, many surgeons remain concerned about potentially increased morbidity and mortality associated with neoadjuvant therapy. The purpose of this study was to review our perioperative complications in patients receiving neoadjuvant systemic chemotherapy within 6 months of liver resection. Chemotherapy consisted of 5-fluouracil (5-FU), leucovorin (LCV), with or without irinotecan (CPT-11) or oxaliplatin. A retrospective chart review of consecutive patients undergoing liver resection for CRC metastasis from January 2002 to July 2004 was performed. 120 patients were identified. Perioperative parameters, patient demographics, tumor characteristics, and details of chemotherapy were measured in addition to morbidity and mortality. The analysis of the first 40 patients is presented here; 40 patients were identified and their charts reviewed. Sixteen patients (40%) received no chemotherapy, 12 (30%) received 5-FU/LCV, and 12 (30%) received 5-FU/LCV/CPT-11 or oxaliplatin. Patients receiving 5-FU/LCV/CPT-11 or oxaliplatin tended to have larger and more numerous tumors. Overall complication rates between groups were not significantly different. Hospital length of stay was similar for both groups. Intraoperative blood loss was also similar for both groups. There was a trend towards an increased rate of wound infection in the group receiving neoadjuvant chemotherapy (16% vs. 21%) but this did not reach statistical significance. There was no perioperative mortality in either group. From this analysis, we conclude that an aggressive use of neoadjuvant chemotherapy (5-FU/ LCV ⫾ CPT-11 or oxaliplatin) prior to liver resection does not result in increased morbidity or mortality. This strategy offers several potential advantages. Neoadjuvant chemotherapy may downstage tumors making unresectable lesions resectable. It may also allow more appropriate selection of patients for surgery: those who progress to extrahepatic disease during a short course of chemotherapy were unlikely to benefit from liver resection in the first place. The analysis of the remaining 80 patients will be presented at the meeting.
175 RADIOFREQUENCY: NEW TECHNOLOGY FOR THE TREATMENT OF LIVER TUMORS Mariano Moro, Rodrigo Sa´nchez Claria´, Martı´n Palavecino, Christian Bertona, Juan Pekolj, Eduardo de Santiban˜es, Hospital Italiano, Buenos Aires, Argentina Only 30-40% of patients with malignant hepatic tumors (primary or secondary) are candidates to a resection with curative intention. This stimulates the development of new technologies to perform a local ablation to modify the evolution. We sought to evaluate the results of the thermoablation with radiofrequency (RF) in the treatment of malignant liver tumors, even primary and secondary. The design was retrospective and observational. Between January 1999 and December 2003, 54 patients were treated with RF. Population was divided according to the original tumor: primary hepatic tumor (PHT), colorectal metastasis (CRM) and nonprimary noncolorectal metastasis (NPNCM). RF was applied alone or associated with another ablative procedure. RF was performed by laparotomy, laparoscopy or using percutaneous puncture. Mean age was 64 (range: 45-80). The tumors were CRM in 32 patients (60%), PHT in 14 (26%), and NPNCM in 8 (14%). RF was percutaneous in 8 (14%), laparoscopic in 4 (9%) and applied by laparotomy in 42 patients (77%). In 31 patients (57%) RF was the only therapy and in 23 (43%) it was associated with another resective or local ablative procedure. Mean hospital stay was 4 days (range: 3-10). Complication rate was 11%. There was no mortality. No reoperations were needed. Mean survival in PHT was 16.5 months, and the global overall survival rate was 47% at 4 years. In CRM mean survival was 16.3 months and overall survival rate was 27%. In NPNCM mean survival was 16.8 months and overall survival rate was 53%. Thermoablation with RF should be considered in patients with no chances of resection. This procedure can be associated with other resective procedures. Laparotomy is the safest way to perform it. The RF has a low rate of complications and it is safe in selected patients.
176 NODULAR REGENERATIVE HYPERPLASIA ASSOCIATED WITH APLASIC ANEMIA: CASE REPORT S. Gonzalez-Huezo, R. Mondragon, L. Villela, C. Carrillo, Centro Medico Issemym, Metepec, Edo Mexico, Mexico Nodular regenerative hyperplasia (NRH) is a rare condition associated with immune, neoplastic, and vascular disorders, as well as with drugs and toxins. Pathogenesis is related to abnormal hepatic blood flow. Portal hypertension is the main complication. We report the case of a woman with NRH and aplastic anemia, association not previously described. A 31-year-old Hispanic female presented with pancitopenia and cholestasis. PMH was relevant for epilepsy treated with phenytoin for 5 years, which was discontinued 7 years before presentation. At admission she was pale and slightly jaundiced with diffuse petechiae. Lab report was as follows: CBC: leukocytes 2700/µl (with 500 total PMN cells); platelets 35,000/µl; Hb 6.7 g/dl; reticulocytes 0%. LFT: ALT 61 U/L, AST 70 U/L, Tot. Bil. 3.3 g/dl, Dir. Bil. 1.0 g/dl, Alk. Phos. 203 U/L, Alb 3.7 g/dl, globulins 3.9 g/dl, INR 1.4. Bone marrow biopsy was diagnostic of aplastic anemia. Abdominal ultrasound revealed multiple liver nodules, without hepato-splenomegaly. No abnormalities were noted in hepatic vessels. Serology for HBV and HCV was negative. Antimitochondrial (AMAs) and antinuclear antibodies were positive. Coagulation tests, antiphospholipid antibodies, C and S proteins, antithrombin III and factor V Leiden were normal. Liver biopsy showed hepatocyte nodules with central atrophy and sinusoidal dilatation in the absence of significant fibrosis. No arteritis, thrombosis or bile duct damage was found. Anti-thymocyte antiglobulin and cyclosporine were started with partial response at 5 weeks. Pathogenesis of NRH remains unclear. The most accepted theory was developed
Vol. 9, No. 4 2005
by Wanless 24 years ago, and reinforced in 1997. It states that alterations in hepatic blood flow (portal obliterative venopathy) originate a specific response, initially consisting of apoptosis secondary to acute ischemia, with ensuing parenchymal atrophy and surrounding areas of liver regeneration and chronic ischemia. The common pathologic mechanism in all the associated diseases in NRH involves liver flow abnormalities secondary to microthromboses, vascular congestion, or vasculitis. In the current case we were unable to document evidence of liver flow abnormalities. Noteworthy there are previous descriptions of the association of NRH with primary biliary cirrhosis, and this patient was AMA positive, but without bile duct damage in the biopsy. Uncommonly, the only laboratory abnormality in NRH can be cholestasis. This case is relevant because we cannot sustain Wanless’ theory and because it is the first description of the association of NRH with aplastic anemia.
177 A CASE OF PSEUDOLYMPHOMA OF THE LIVER Tsuyoshi Hatsuno, Nagoya Medical Center, Nagoya, Japan Pseudolymphoma (reactive lymphoid hyperplasia) of the liver is an extremely rare entity, with a few cases reported so far. We encountered a 50-year-old Japanese female with pseudolymphoma of the liver. A tumor-like lesion was discovered incidentally as a hypoechoic mass with a hyperechoic rim during clinical follow-up of endometriosis. Radiological studies, such as contrast-enhanced computed tomography (CT) and angiography demonstrated a hypervascular lesion. The hepatic lesion was resected because hepatocellular carcinoma was suspected after a needle biopsy. Grossly, the lesion was well defined and measured 1.0 × 1.5 cm. Microscopically, the lesion consisted of hyperplastic lymphoid follicles with distinctive germinal centers and interfollicular areas consistng of mature lymphocytes and plasma cells. An immunohistological study revealed that the lymphoid cells of the lesion were polyclonal in immunophenotypes. These histological and immunohistochemical findings strongly suggested a pseudolymphoma of the liver. The following features have characterized the images in past cases, as well as ours: hypoechoic mass, occasionally with a rim, in ultrasonography and hypervascularity, shown by angiography and enhanced CT.
178 TOO MANY COOKS SPOIL THE BROTH: THE DILEMMA OF SELECTING PATIENTS WITH HEPATOCELLULAR CARCINOMA FOR LIVER TRANSPLANTATION Claudia Hriesik, MD, PhD, Burkhardt Ringe, MD, Drexel University College of Medicine, Philadelphia, PA Criteria for liver transplantation in patients with hepatocellular carcinoma (HCC) differ amongst staging systems and amongst transplant centers. Originally used staging systems, e.g., TNM/UICC, are replaced by selection criteria described by Mazzaferro and the Milan criteria. The latter, however, are too tight and might breed the dilemma to falsely withhold an organ. Examining different selection criteria, it becomes obvious that a potential risk is that selected criteria may be used as one pleases in order to achieve improved results or in order to expand patient selection. This in return poses the risk of wasting an organ. Furthermore, established staging criteria (e.g., Milan criteria, Barcelona criteria) are conflicting as they use postoperative data, e.g., histological results, in order to select therapy. Thus, a comparison of results and outcome amongst transplant centers following liver transplantation and an outlook on prognosis is difficult if not impossible. The aim of this study was to analyze currently used
Abstracts
575
selection criteria for liver transplantation and to determine their reliability related to patient selection, therapy selection, and prognosis. We have reviewed the medical literature since 1954 to identify criteria published for the staging of HCC. To evaluate and compare the reliability of these criteria we focused on two questions: (1) Can the criteria be used in the decision making process for liver transplantation before and for treatment, and (2) How accurately can these criteria predict prognosis? As a tool we used the certainty factor (C factor) defining the level of diagnostic judgment and reliability: C1 clinical, C2 radiological, C3 biopsy, C4 resection, C5 autopsy. As an additional criterion liver function parameters were selected. Twentyone staging systems were used for this comparative analysis. 1) Five staging systems can be used for therapy selection (C1-C3). 2) Three systems provided prognosis assessment only (C4-C5). 3) Thirteen staging systems allowed to both assist in decision making for therapy based on clinical criteria as well as the assessment of a prognosis based on pathological criteria (C1-C5). 4) Only five staging systems assessed tumors factors as well as liver function (C1-C5, LF). The ideal staging system should 1) facilitate the selection of transplant candidates based on tumor factors and clinical parameters and 2) predict long term prognosis as indicated by pathological tumor extent. It would be desirable to use a uniform staging system as suggested by the AHPBA consensus statement of 2003.
179 RADIOFREQUENCY ABLATION IN THE PORCINE MODEL: COMPARISON OF MULTIPLE COOL TIP PROBES AND A SWITCHING CONTROLLER WITH A STANDARD THREE-PROBE CLUSTER David A. Iannitti, MD, Dionisios Vrochides, MD, PhD, Jason McKee, MD, Damian E. Dupuy, MD, Brown Medical School, Providence, RI Evaluation of the Switching Controller with multiple, combined cool tip probes for an output-based radiofrequency device. Radiofrequency (RF) ablation was performed in three randomly selected segments of the liver in 12 pigs. Half of the ablations were performed with a Pringle maneuver and half without. Ablations were performed with three 3.0 cm active tip, cooled RF probes spaced 2.0 cm apart and a 200 watt RF generator. Ablations using the standard cluster of probes were performed for 12 minutes in the non-Pringle subgroup and 6 minutes in the Pringle subgroup. Ablations using the Switching Controller with multiple, combined cool tip probes were performed for 16 minutes in the non-Pringle subgroup and 8 minutes in the Pringle subgroup. The Switching Controller delivered current that alternated among the probes, based on impedance (to 30 Ohms above baseline) and time (30 seconds maximum). The ablation dimensions, volume, temperature, current delivery parameters, and postablation histology were evaluated. A total of 34 successful ablations were performed. The minimum ablation diameter (Dmin) in the standard cluster group was 40.3 ⫾ 4.3 mm (42.5 ⫾ 4.1 mm for the Pringle group and 38.1 ⫾ 3.6 mm for the non-Pringle group). The maximum ablation surface area (SAmax) in this group was 16.0 ⫾ 3.3 cm2 (17.6 ⫾ 2.2 sqcm for the Pringle group and 14.2 ⫾ 3.5 cm2 for the nonPringle group). The ablation volume in the cluster group was 36.9 ⫾ 9.5 ml (40.1 ⫾ 8.1 ml for the Pringle group and 33.7 ⫾ 10.4 ml for the non-Pringle group). In the Switching Controller Multiprobe group, the ablation Dmin was 51.4 ⫾ 7.4 mm (54.5 ⫾ 8.5 mm for the Pringle group and 48.2 ⫾ 4.8 mm for the non-Pringle group). The ablation SAmax in this group was 22.4 ⫾ 4.7 cm2 (23.9 ⫾ 4.9 cm2 for the Pringle group and 20.8 ⫾ 4.1 cm2 for the non-Pringle group). The ablation volume in the multiprobe group was 66.1 ⫾ 17.9 ml (68 ⫾ 16.8 ml for the Pringle group and 64.3 ⫾ 19.6 ml for the nonPringle group). Technology to increase the size of ablated areas and shorten the procedural time continues to evolve. The use of the