- Email: [email protected]
Risk factors for death following liver retransplantation
Risk Factors for Death Following Liver Retransplantation I. Bilbao, J. Figueras, L. Grande, M. Cle`ries, E. Jaurrieta, J. Visa, and C. Margarit ABSTRACT Aim. Our goal was to retrospectively analyze graft loss and mortality risk factors using multi-centre data on liver retransplantation. Material and methods. Between 1991–1995, 640 patients underwent 718 liver transplants in Barcelona. Mean age of the 74 patients receiving a second transplant was 47.6 years (range 19 – 65). Causes of retransplantation were immunologic in 26 patients (35.1%), technical in 23 (31.1%), primary dysfunction in 12 (16.2%), recurrent original disease in 7 (9.5%), and other causes in 6 (8.1%). Mean time between first and second transplant was less than 7 days in 20 patients (27%), between 8 and 30 days in 4 (5.4%) and more than 30 days in 50 patients (67.6%). Recipient, donor, and operative variables were analyzed using univariate (Kaplan-Meier curves) and multivariate techniques (Cox regression) to identify risk factors. Results. Retransplant patient survival at 1 and 5 years was 60.8% and 49.5%, respectively, compared to 75.6% and 64.8% in a series of 640 first transplant patients. Mortality risk factors identified by multivariate analysis were bilirubin ⬎12 mg/dL (RR 2.3; P ⫽ .010), recipient age (RR increase 0.04 for each additional year; P ⫽ .02), cause for retransplant (immunologic RR 4.01, technical RR 2.7 and other causes RR 6.9; compared to primary dysfunction RR 1; P ⫽ .020). Urea ⬎54 mg/dL (0.02) and multiple transfusions ⬎15 units red blood cells (0.001) were only significant in the univariate analysis. Conclusions. In our experience, retransplantation for primary dysfunction is the setting that has the best prognosis. Of the other causes, retransplantation should be performed before the total bilirubin reaches ⬎12 mg/dL or before the appearance of variables indicative of severe renal insufficiency.
C
URRENTLY, the incidence of retransplantation varies between 5% and 20% of all liver transplants. As such, it is one of the most important causes of transplantation and utilization of donor organs.1–5 Retransplantation outcomes are inferior to primary ones; Long-term survival is 10% to 30% lower than the first transplant.1,3,4,6 –11 Also, the incidence of morbidity is higher, especially sepsis2 and neurologic complications. As a consequence, hospital stays are longer, and costs are nearly doubled.2,7,9 To offer one patient more than one transplant opportunity while others are dying on the waiting list for their first graft has been the subject of considerable controversy.6 Hence it is important to analyse the risk factors that lead to death after retransplantation so as to preclude erroneous clinical judgment. Retransplantation may be indicated early or late in the course of transplant follow-up; the relative frequency of each has changed over recent years.5 The advent of more
effective immunosuppressants, advances in technical skills, and increasing experience are some reasons for the decreasing rate of acute rejection. In the past, technical complications were the most important causes of early retransplant, while chronic rejection was the most important cause for late retransplant. However, the increasing number of patients dying while on the waiting list together with donor scarcity has necessitated the acceptance of suboptimal donors, which, in turn, contributes to maintaining the
From the Hospital Vall d’Hebro´n; Hospital de Bellvitge; Hospital Clinic i Provincial; and Servicio Catala´n de la Salud; Barcelona, Spain. Address reprint requests to Carlos Margarit, MD and Itxarone Bilbao, MD, Unidad de Trasplante Hepa´tico, Hospital Vall d’Hebro´n, Paseo Vall d’Hebro´n, 119-129 08035 - Barcelona, Spain. E-mail: [email protected] or [email protected]
© 2003 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/03/$–see front matter doi:10.1016/S0041-1345(03)00585-2
Transplantation Proceedings, 35, 1871–1873 (2003)
1871
1872
incidence of primary graft dysfunction as the leading indication for early retransplantation. Similarly, the use of more potent immunosuppressants has decreased chronic rejection but has increased recurrent hepatitis C in the graft as the most important cause for late retransplant. MATERIALS AND METHODS Between January 1991 and December 1995, 640 patients underwent 718 orthotopic liver transplantation at the three university hospitals authorised to perform liver transplants in Barcelona (Catalunya). Over this period, 74 patients (11.5%) received a second graft and four patients received a third. We collated the recipient, donor, and operative variables and utilized univariate analysis (Kaplan-Meier method) and multivariate analysis (Cox regression) to identify factors associated with mortality. Recipient variables included: age, gender, UNOS status (UNOS 1, 2a, 2b versus UNOS 3), clinical indication for first transplant (postnecrotic cirrhosis, cholostatic cirrhosis, acute hepatic insufficiency, and other causes), indication for retransplantation (primary graft dysfunction, immunologic causes, technical problems, original disease recurrence, and other causes), previous history of alcohol abuse, nutritional status, cross-match, ABO compatibility, serologic status related to hepatitis B and C, time between first and second transplant (⬍7 days, between 7 and 30 days, and ⬎30 days), and biochemical parameters (including total bilirubin cut-off 12 mg/dL, AST cut-off 220 IU/L, ALT cut-off 175 IU/L, urea cut-off 54 mg/dL, serum creatinine cut-off 1.2 mg/dL, platelets cut-off 116 ⫻ 10E9/L, albumin cut-off 3.3 g/dL). Donor variables were age, gender, cause of death (craneoencephalic trauma, cerebrovascular stroke, and others), graft steatosis (presence or absence), days in ICU prior to harvest, ALT (cut-off 40 IU/L), and sodium prior to harvest (cut-off 145 mmol/L). Operative variables were type of anhepatic technique performed (piggyback versus cross-clamping with or without bypass), portal thrombosis, iliac graft for arterial anastomosis, ischemia time (cut-off 10 hours), multiple transfusions of red blood cells (cut-off 10 units, cut-off 15 units). All the patients were followed for 5 years. Status such as dead or alive and cause of death were among the recorded outcomes. To establish cut-off points to discriminate poor outcomes, continuous variables were assessed as means or medians or quartiles, and qualitative variables were dichotomised
RESULTS
The mean age was 47.6 years, the median was 50 years, and the range was 19 to 65 years. Five patients (6.8%) were over 60 years or age. There were 47 men (63.5%) and 27 women (36.5%). Causes of retransplantation were primary graft dysfunction in 12 (16.2%); immunological causes in 26 (35.1%); acute rejection in 2 and chronic rejection in 24; technical problems in 23 (31.1%); vascular complications in 16; biliary complications in 7; disease recurrence (hepatitis C reinfection of the graft) in 7 (9.5%); and other causes in 6 (8.1%). All patients received ABO-compatible grafts except two cases, both of whom died at 18 and 6 months subsequent to the retransplant. Two patients (2.7%) had positive cross-matches. Half of the patients (52.7%) were UNOS status I, IIa, or IIb at the time of retransplant. Indications for first transplant were postnecrotic cirrhosis in 49 patients (66.2%), cholestatic cirrhosis in 10 (13.5%),
BILBAO, FIGUERAS, GRANDE ET AL
acute hepatic insufficiency in 9 (12.2%), and other causes in 6 patients (8.1%). The time interval between the first and second transplant was less than 7 days in 20 (27%), between 8 and 30 days in 4 (5.4%), and over 30 days in 50 patients (67.6%). Six (8.1%) patients were HBV positive, and 26 (35.1%) were HCV positive. Thirteen patients (17.6%) had hepatocellular carcinoma in addition to cirrhosis. Twentyfour patients (32.4%) had a history of alcohol abuse. At the time of retransplant the mean bilirubin concentration was 16.5% mg/dL, and the median was 12.1 mg/dL (range 0.5– 85); mean ALT was 1.012 UI/L, median 176 (range 23–11,460); mean AST was 1021, median 219 UI/L (range 23–7190); mean serum creatinine was 1.6 mg/dL, median 1.1 mg/dL (range 0.3–5); mean urea was 78 mg/dL, median 54 mg/dL (range 6 –319); mean platelet count was 125 ⫻ 109/L, median 116 ⫻ 109/L (range 19 – 438); mean Quick rate was 73%, median 80% (range 15–107); and mean albumin was 3.3 g/dL, median 3.3 g/dL (range 2– 4.9). Donor mean age was 37.5 years and the median was 36 (range 6 – 67). Causes of death were craneoencephalic trauma in 35 (47.3%), cerebrovascular stroke in 28 (37.8%), and other causes in 9 (12.2%). Donor mean ALT was 57 UI/L, median 37 UI/L (range 5–321); mean sodium was 145.6 mmol/L, median 144.5 mmol/L (range 113–170). Mean ischemia time was 7.6 hours, median 6.9 hours (range 2–17). Anhepatic technique employed was inferior vena cava preservation in 44 (59.5%) and cross-clamping with or without bypass in 30 patients (40.5%). Mean intraoperative transfusion of red blood cells was 15 units, median 10 units (range 0 – 86) Actuarial survival at 1 and 5 years for retransplanted patients was 60.8% and 49.5%, respectively compared to 75.6% and 64.8% for the 640 patients undergoing a first transplantation over the same study period. Thirty-four patients died during follow-up. Causes of death were primary dysfunction in 2 (5.9%), technical problems in 3 (8.8%), original disease recurrence in 5 (14.7%), infectionsepsis in 10 (29.4%), and other causes in 14 patients (41.2%). Prognostic factors for mortality in univariate analysis (Kaplan-Meier) were total bilirubin ⬎12 mg/dL (P ⫽ .020), urea ⬎54 mg/dL (P ⫽ .020), retransplantation cause other than primary dysfunction (P ⫽ .010) and multiple transfusions ⬎15 units (P ⫽ .001). In the multivariate analysis (Cox regression) the risk factors independently associated with mortality were total bilirubin ⬎12 mg/dL RR: 2.3; P ⫽ .019), age above 50 years (each additional year increased RR 0.04, P ⫽ .020), and certain causes of retransplantation (immunologic causes RR: 4.01, technical problems RR: 2.7, and other causes RR: 6.9, compared to primary graft dysfunction RR: 1; P ⫽ .020) Table 1. DISCUSSION
Many reports suggest that outcomes in liver retransplantation depend on the interval between the first transplant and the retransplant. The best results are obtained when re-
RISK OF DEATH AFTER LIVER TRANSPLANT
1873
Table 1. Univariate and Multivariate Analysis Variables Recipient Variables
Univariate Analysis; P
Age ⬎50 years Gender Indication for first transplant UNOS status Cause of retransplantation Primary dysfunction Immunological causes Technical problems Other causes Days between first and second transplant (7 d/30 d) Alcohol abuse Hepatitis C virus positive Hepatocellular carcinoma Median bilirubin (cut-off ⬎12 mg/dL) Median ALT (cut-off 175 IU/L) Median AST (cut-off 220 IU/L) Median platelets (cut-off 116 ⫻ 109/L) Median urea (cut-off 54 mg/dL; 40 mg/dL) Median serum creatinine (cutoff 1.2 mg/dL) Median albumin (cut-off 3.3 g/dL) Donor Variables ABO compatibility Cause of death ALT (cut-off 40 IU/L) Median sodium (cut-off 144 mmol/L) Surgery Variables Median ischaemic time (cut-off 10 hours) Anhepatic technique Transfusion red blood cells (median cut-off 10 units) Transfusion red blood cells (mean cut-off 15 units)
.50 .34 .15 .97 .012
Multivariate Analysis
P ⫽ .02 RR 1 RR 4.01 RR 2.7 RR 6.9
.24/.07 .69 .49 .47 .02
RR 2.3; P ⫽ .019
.58 .11 .65 .02/.006 .13 .43
.76 .28 .13 .59
.26 .21 .02 .001
transplantation is indicated before the third postoperative day, while the period between the third and thirtieth days are the worst. Beyond the thirtieth day, the results begin to improve again. The main causes during the first 3 days posttransplant are primary dysfunction or technical complications. If liver retransplantation is indicated before the first week posttransplant, the repeat operation is easier and the general condition of the patient has not been progressively deteriorating. Several studies (2,7,10,12,13) report that retransplantation should be performed before deterioration of more than three organs (renal failure, ventilatory requirements, cerebrovascular stroke, liver failure with severe cholestasis, progressive encephelopathy, coagulopathy, and so on). However, problems arise when sub-acute graft dysfunc-
tion occurs requiring intensive medical treatment or surgical revascularisations in cases of hepatic artery thrombosis or portal thrombosis. The patient is deteriorating progressively, and when retransplantation is finally indicated, the patient may have lost priority for organ allocation, thereby undergoing the procedure in poor general condition. Another situation is that of late retransplantation performed months, or even years, after the first transplant due to chronic rejection, late-onset technical complications, or original disease recurrence. In such cases an elective procedure may be planned when the patient presents in optimal condition.2 In such cases, it should be noted that surgery may be difficult due to previous adhesions and fibrosis. Facciuto et al 14 reported multiple transfusions, age over 50 years, and renal insufficiency (serum creatinine ⬎2 mg/dL) as risk factors for late retransplantation beyond 6 months. Based on our experience, patients with primary dysfunction needing a retransplant have the best prognosis, probably because the indication for retransplantation occurs before deterioration of the patient’s other organs. Other causes of retransplantation (technical problems, refractory acute rejection, and so forth), which are usually more delayed in the course of transplantation, must be indicated before deterioration of other organs due to renal insufficiency, coagulopathy requiring multiple transfusions, and liver failure with total bilirubin ⬎12 mg/dL. The indication of original disease recurrence, especially with hepatitis virus C, is more controversial awaiting resolution when better treatments become available. We think that retransplantation in such cases should be performed only for investigative purposes under controlled experimental conditions. REFERENCES 1. Registro Espan ˜ol de Transplante Hepa´tico (R.E.T.H.): Primera Memoria de Resultados 1984-1997 2. Markmann JF, Markowitz JS, Yersiz H, et al: Ann Surg 226:408, 1997 3. De Carlis L, Slim AO, Glacomoni A, et al: Transplant Proc 33:1411, 2001 4. Meneu-Diaz JC, Moreno Gonzalez E, Vicente E, et al: Transplant Proc 34:301, 2002 5. Kashyap R, Jain A, Reyes J, et al: Transplant Proc 33:1486, 2001 6. Powelson JA, Cosimi AB, Lewis WD, et al: Transplantation 55:802, 1993 7. D’Alessandro AM, Ploeg RJ, Knechtle SJ, et al: Transplantation 55:1083, 1993 8. Doyle HR, Morelli F, McMichael J, et al: Transplantation 61:1499, 1996 9. Wong T, Devlin J, Rolando N, et al: Transplantation 64:878, 1997 10. Rosen HR, Madden JP, Martin P: Hepatology 29:365, 1999 11. Biggins SW, Beldecos A, Rabkin JM, et al: Liver Transplant 8:313, 2002 12. Markmann JF, Gornbein J, Markowitz JS, et al: Transplantation 67:422, 1999 13. Bilbao I, Hidalgo E, La´zaro JL, et al: Transplant Proc 29:368, 1997 14. Facciuto M, Heidt D, Guarrera J, et al: Liver Transpl 6:174, 2000
C
URRENTLY, the incidence of retransplantation varies between 5% and 20% of all liver transplants. As such, it is one of the most important causes of transplantation and utilization of donor organs.1–5 Retransplantation outcomes are inferior to primary ones; Long-term survival is 10% to 30% lower than the first transplant.1,3,4,6 –11 Also, the incidence of morbidity is higher, especially sepsis2 and neurologic complications. As a consequence, hospital stays are longer, and costs are nearly doubled.2,7,9 To offer one patient more than one transplant opportunity while others are dying on the waiting list for their first graft has been the subject of considerable controversy.6 Hence it is important to analyse the risk factors that lead to death after retransplantation so as to preclude erroneous clinical judgment. Retransplantation may be indicated early or late in the course of transplant follow-up; the relative frequency of each has changed over recent years.5 The advent of more
effective immunosuppressants, advances in technical skills, and increasing experience are some reasons for the decreasing rate of acute rejection. In the past, technical complications were the most important causes of early retransplant, while chronic rejection was the most important cause for late retransplant. However, the increasing number of patients dying while on the waiting list together with donor scarcity has necessitated the acceptance of suboptimal donors, which, in turn, contributes to maintaining the
From the Hospital Vall d’Hebro´n; Hospital de Bellvitge; Hospital Clinic i Provincial; and Servicio Catala´n de la Salud; Barcelona, Spain. Address reprint requests to Carlos Margarit, MD and Itxarone Bilbao, MD, Unidad de Trasplante Hepa´tico, Hospital Vall d’Hebro´n, Paseo Vall d’Hebro´n, 119-129 08035 - Barcelona, Spain. E-mail: [email protected] or [email protected]
© 2003 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/03/$–see front matter doi:10.1016/S0041-1345(03)00585-2
Transplantation Proceedings, 35, 1871–1873 (2003)
1871
1872
incidence of primary graft dysfunction as the leading indication for early retransplantation. Similarly, the use of more potent immunosuppressants has decreased chronic rejection but has increased recurrent hepatitis C in the graft as the most important cause for late retransplant. MATERIALS AND METHODS Between January 1991 and December 1995, 640 patients underwent 718 orthotopic liver transplantation at the three university hospitals authorised to perform liver transplants in Barcelona (Catalunya). Over this period, 74 patients (11.5%) received a second graft and four patients received a third. We collated the recipient, donor, and operative variables and utilized univariate analysis (Kaplan-Meier method) and multivariate analysis (Cox regression) to identify factors associated with mortality. Recipient variables included: age, gender, UNOS status (UNOS 1, 2a, 2b versus UNOS 3), clinical indication for first transplant (postnecrotic cirrhosis, cholostatic cirrhosis, acute hepatic insufficiency, and other causes), indication for retransplantation (primary graft dysfunction, immunologic causes, technical problems, original disease recurrence, and other causes), previous history of alcohol abuse, nutritional status, cross-match, ABO compatibility, serologic status related to hepatitis B and C, time between first and second transplant (⬍7 days, between 7 and 30 days, and ⬎30 days), and biochemical parameters (including total bilirubin cut-off 12 mg/dL, AST cut-off 220 IU/L, ALT cut-off 175 IU/L, urea cut-off 54 mg/dL, serum creatinine cut-off 1.2 mg/dL, platelets cut-off 116 ⫻ 10E9/L, albumin cut-off 3.3 g/dL). Donor variables were age, gender, cause of death (craneoencephalic trauma, cerebrovascular stroke, and others), graft steatosis (presence or absence), days in ICU prior to harvest, ALT (cut-off 40 IU/L), and sodium prior to harvest (cut-off 145 mmol/L). Operative variables were type of anhepatic technique performed (piggyback versus cross-clamping with or without bypass), portal thrombosis, iliac graft for arterial anastomosis, ischemia time (cut-off 10 hours), multiple transfusions of red blood cells (cut-off 10 units, cut-off 15 units). All the patients were followed for 5 years. Status such as dead or alive and cause of death were among the recorded outcomes. To establish cut-off points to discriminate poor outcomes, continuous variables were assessed as means or medians or quartiles, and qualitative variables were dichotomised
RESULTS
The mean age was 47.6 years, the median was 50 years, and the range was 19 to 65 years. Five patients (6.8%) were over 60 years or age. There were 47 men (63.5%) and 27 women (36.5%). Causes of retransplantation were primary graft dysfunction in 12 (16.2%); immunological causes in 26 (35.1%); acute rejection in 2 and chronic rejection in 24; technical problems in 23 (31.1%); vascular complications in 16; biliary complications in 7; disease recurrence (hepatitis C reinfection of the graft) in 7 (9.5%); and other causes in 6 (8.1%). All patients received ABO-compatible grafts except two cases, both of whom died at 18 and 6 months subsequent to the retransplant. Two patients (2.7%) had positive cross-matches. Half of the patients (52.7%) were UNOS status I, IIa, or IIb at the time of retransplant. Indications for first transplant were postnecrotic cirrhosis in 49 patients (66.2%), cholestatic cirrhosis in 10 (13.5%),
BILBAO, FIGUERAS, GRANDE ET AL
acute hepatic insufficiency in 9 (12.2%), and other causes in 6 patients (8.1%). The time interval between the first and second transplant was less than 7 days in 20 (27%), between 8 and 30 days in 4 (5.4%), and over 30 days in 50 patients (67.6%). Six (8.1%) patients were HBV positive, and 26 (35.1%) were HCV positive. Thirteen patients (17.6%) had hepatocellular carcinoma in addition to cirrhosis. Twentyfour patients (32.4%) had a history of alcohol abuse. At the time of retransplant the mean bilirubin concentration was 16.5% mg/dL, and the median was 12.1 mg/dL (range 0.5– 85); mean ALT was 1.012 UI/L, median 176 (range 23–11,460); mean AST was 1021, median 219 UI/L (range 23–7190); mean serum creatinine was 1.6 mg/dL, median 1.1 mg/dL (range 0.3–5); mean urea was 78 mg/dL, median 54 mg/dL (range 6 –319); mean platelet count was 125 ⫻ 109/L, median 116 ⫻ 109/L (range 19 – 438); mean Quick rate was 73%, median 80% (range 15–107); and mean albumin was 3.3 g/dL, median 3.3 g/dL (range 2– 4.9). Donor mean age was 37.5 years and the median was 36 (range 6 – 67). Causes of death were craneoencephalic trauma in 35 (47.3%), cerebrovascular stroke in 28 (37.8%), and other causes in 9 (12.2%). Donor mean ALT was 57 UI/L, median 37 UI/L (range 5–321); mean sodium was 145.6 mmol/L, median 144.5 mmol/L (range 113–170). Mean ischemia time was 7.6 hours, median 6.9 hours (range 2–17). Anhepatic technique employed was inferior vena cava preservation in 44 (59.5%) and cross-clamping with or without bypass in 30 patients (40.5%). Mean intraoperative transfusion of red blood cells was 15 units, median 10 units (range 0 – 86) Actuarial survival at 1 and 5 years for retransplanted patients was 60.8% and 49.5%, respectively compared to 75.6% and 64.8% for the 640 patients undergoing a first transplantation over the same study period. Thirty-four patients died during follow-up. Causes of death were primary dysfunction in 2 (5.9%), technical problems in 3 (8.8%), original disease recurrence in 5 (14.7%), infectionsepsis in 10 (29.4%), and other causes in 14 patients (41.2%). Prognostic factors for mortality in univariate analysis (Kaplan-Meier) were total bilirubin ⬎12 mg/dL (P ⫽ .020), urea ⬎54 mg/dL (P ⫽ .020), retransplantation cause other than primary dysfunction (P ⫽ .010) and multiple transfusions ⬎15 units (P ⫽ .001). In the multivariate analysis (Cox regression) the risk factors independently associated with mortality were total bilirubin ⬎12 mg/dL RR: 2.3; P ⫽ .019), age above 50 years (each additional year increased RR 0.04, P ⫽ .020), and certain causes of retransplantation (immunologic causes RR: 4.01, technical problems RR: 2.7, and other causes RR: 6.9, compared to primary graft dysfunction RR: 1; P ⫽ .020) Table 1. DISCUSSION
Many reports suggest that outcomes in liver retransplantation depend on the interval between the first transplant and the retransplant. The best results are obtained when re-
RISK OF DEATH AFTER LIVER TRANSPLANT
1873
Table 1. Univariate and Multivariate Analysis Variables Recipient Variables
Univariate Analysis; P
Age ⬎50 years Gender Indication for first transplant UNOS status Cause of retransplantation Primary dysfunction Immunological causes Technical problems Other causes Days between first and second transplant (7 d/30 d) Alcohol abuse Hepatitis C virus positive Hepatocellular carcinoma Median bilirubin (cut-off ⬎12 mg/dL) Median ALT (cut-off 175 IU/L) Median AST (cut-off 220 IU/L) Median platelets (cut-off 116 ⫻ 109/L) Median urea (cut-off 54 mg/dL; 40 mg/dL) Median serum creatinine (cutoff 1.2 mg/dL) Median albumin (cut-off 3.3 g/dL) Donor Variables ABO compatibility Cause of death ALT (cut-off 40 IU/L) Median sodium (cut-off 144 mmol/L) Surgery Variables Median ischaemic time (cut-off 10 hours) Anhepatic technique Transfusion red blood cells (median cut-off 10 units) Transfusion red blood cells (mean cut-off 15 units)
.50 .34 .15 .97 .012
Multivariate Analysis
P ⫽ .02 RR 1 RR 4.01 RR 2.7 RR 6.9
.24/.07 .69 .49 .47 .02
RR 2.3; P ⫽ .019
.58 .11 .65 .02/.006 .13 .43
.76 .28 .13 .59
.26 .21 .02 .001
transplantation is indicated before the third postoperative day, while the period between the third and thirtieth days are the worst. Beyond the thirtieth day, the results begin to improve again. The main causes during the first 3 days posttransplant are primary dysfunction or technical complications. If liver retransplantation is indicated before the first week posttransplant, the repeat operation is easier and the general condition of the patient has not been progressively deteriorating. Several studies (2,7,10,12,13) report that retransplantation should be performed before deterioration of more than three organs (renal failure, ventilatory requirements, cerebrovascular stroke, liver failure with severe cholestasis, progressive encephelopathy, coagulopathy, and so on). However, problems arise when sub-acute graft dysfunc-
tion occurs requiring intensive medical treatment or surgical revascularisations in cases of hepatic artery thrombosis or portal thrombosis. The patient is deteriorating progressively, and when retransplantation is finally indicated, the patient may have lost priority for organ allocation, thereby undergoing the procedure in poor general condition. Another situation is that of late retransplantation performed months, or even years, after the first transplant due to chronic rejection, late-onset technical complications, or original disease recurrence. In such cases an elective procedure may be planned when the patient presents in optimal condition.2 In such cases, it should be noted that surgery may be difficult due to previous adhesions and fibrosis. Facciuto et al 14 reported multiple transfusions, age over 50 years, and renal insufficiency (serum creatinine ⬎2 mg/dL) as risk factors for late retransplantation beyond 6 months. Based on our experience, patients with primary dysfunction needing a retransplant have the best prognosis, probably because the indication for retransplantation occurs before deterioration of the patient’s other organs. Other causes of retransplantation (technical problems, refractory acute rejection, and so forth), which are usually more delayed in the course of transplantation, must be indicated before deterioration of other organs due to renal insufficiency, coagulopathy requiring multiple transfusions, and liver failure with total bilirubin ⬎12 mg/dL. The indication of original disease recurrence, especially with hepatitis virus C, is more controversial awaiting resolution when better treatments become available. We think that retransplantation in such cases should be performed only for investigative purposes under controlled experimental conditions. REFERENCES 1. Registro Espan ˜ol de Transplante Hepa´tico (R.E.T.H.): Primera Memoria de Resultados 1984-1997 2. Markmann JF, Markowitz JS, Yersiz H, et al: Ann Surg 226:408, 1997 3. De Carlis L, Slim AO, Glacomoni A, et al: Transplant Proc 33:1411, 2001 4. Meneu-Diaz JC, Moreno Gonzalez E, Vicente E, et al: Transplant Proc 34:301, 2002 5. Kashyap R, Jain A, Reyes J, et al: Transplant Proc 33:1486, 2001 6. Powelson JA, Cosimi AB, Lewis WD, et al: Transplantation 55:802, 1993 7. D’Alessandro AM, Ploeg RJ, Knechtle SJ, et al: Transplantation 55:1083, 1993 8. Doyle HR, Morelli F, McMichael J, et al: Transplantation 61:1499, 1996 9. Wong T, Devlin J, Rolando N, et al: Transplantation 64:878, 1997 10. Rosen HR, Madden JP, Martin P: Hepatology 29:365, 1999 11. Biggins SW, Beldecos A, Rabkin JM, et al: Liver Transplant 8:313, 2002 12. Markmann JF, Gornbein J, Markowitz JS, et al: Transplantation 67:422, 1999 13. Bilbao I, Hidalgo E, La´zaro JL, et al: Transplant Proc 29:368, 1997 14. Facciuto M, Heidt D, Guarrera J, et al: Liver Transpl 6:174, 2000