Hepatorenal Syndrome: a Proposal for Kidney After Liver Transplantation (KALT)

Transcription

1 LIVER TRANSPLANTATION 13: , 2007 ORIGINAL ARTICLE Hepatorenal Syndrome: a Proposal for Kidney After Liver Transplantation (KALT) Richard Ruiz, Yousri M. Barri, Linda W. Jennings, Srinath Chinnakotla, Robert M. Goldstein, Marlon F. Levy, Greg J. McKenna, Henry B. Randall, Edmund Q. Sanchez, and Goran B. Klintmalm Baylor Regional Transplant Institute, Dallas, TX. Hepatorenal syndrome (HRS) is a well-recognized complication of end-stage liver disease. Once thought to be a reversible condition with liver transplantation (LT) alone, HRS may directly contribute to the requirement for long-term dialysis posttransplant. As a result, discussion has now focused on whether or when a kidney allograft should be considered for these patients. Using the International Ascites Club guidelines with a pretransplant serum creatinine (SCr) 2.0 mg/dl to define HRS, 130 patients undergoing LT over a 10-yr period were identified, for an overall incidence of 9%. Patient survival rates at 1, 3, and 5 yr were 74%, and 68%, and 62%, respectively. Survival was significantly worse when compared to non-hrs patients undergoing LT over the same study period (P ). For patients presenting with type 2 HRS, 7 patients (6%) developed irreversible kidney failure posttransplant compared to 0.34% in the non-hrs population (P ). Five of these patients died within 1 yr with a median survival time of 139 days. Combined liver and kidney transplantation (CLKT) for patients with HRS is not recommended. However, an improvement in outcome can be accomplished by addressing those patients who require dialysis greater than 60 days posttransplant. We propose a role for kidney after liver transplantation (KALT) in select HRS patients. Liver Transpl 13: , AASLD. Received December 21, 2006; accepted January 31, Hepatorenal syndrome (HRS) is a complication of liver failure which carries a poor prognosis without transplantation. Type 1 HRS or liver failure accompanied by rapidly progressive renal failure carries a median patient survival of 2 to 4 weeks; patients with type 2 HRS or liver failure associated with a slower deterioration of renal function fare better with a median survival of approximately 6 months. 1,2 With liver transplantation (LT) alone, there is considerable improvement in survival although it still did not approach the survival of non-hrs patients in a study by Gonwa et al. 3 (60% vs. 68% at 5 yr, P 0.03). In an effort to improve outcome, some transplant centers have implemented combined liver and kidney transplantation (CLKT) for patients with HRS on hemodialysis (HD) for a predetermined length of time pretransplant. According to United Network for Organ Sharing data, HRS as an isolated indication for kidney transplant in CLKT has increased from 9% in the 5 yr prior to the Model for End-Stage Liver Disease (MELD) era to 16% since (Table 1, P 0.001). 4 The decision to transplant patients with type 1 HRS certainly should not rest on the degree of acute renal failure or need for HD pretransplant and thus, requires no further discussion. For those patients with type 2 presentation, however, there is no United Network for Organ Sharing policy with regard to the use of kidney allografts, mainly because the course of type 2 HRS is not well understood and the onset of end-stage renal disease (ESRD) in these patients is undefined. As a result, transplant centers are left to implement their own rule or strategy. With the current MELD system, in which renal dysfunction prioritizes patients awaiting transplant, we are faced with having to develop a global plan concerning kidney allocation for patients presenting with HRS. In a previous study at our institution, no preoperative risk factors were found that could identify HRS patients Abbreviations: HRS, hepatorenal syndrome; LT, liver transplantation; CLKT, combined liver and kidney transplantation; HD, hemodialysis; MELD, Model for End-Stage Liver Disease; ESRD, end-stage renal disease; SCr, serum creatinine; CNI, calcineurin inhibitor; KALT, kidney after liver transplantation. Address reprint requests to Richard Ruiz, MD, Baylor Regional Transplant Institute, 4th Floor Roberts Building, 3500 Gaston Avenue, Dallas, TX Telephone: (214) ; FAX: (214) ; richarru@baylorhealth.edu DOI /lt Published online in Wiley InterScience ( American Association for the Study of Liver Diseases.

2 HEPATORENAL SYNDROME 839 TABLE 1. Isolated Indications for Kidney Transplant in CLKT Before and After the MELD Era 4 January 1, 1997-December 31, 2001 N 601 (%) February 2, 2002-August 31, 2006 N 1350 (%) P value Polycystic kidney disease 45 (7.5) 74 (5.5) Diabetic nephropathy 66 (11) 196 (14.5) Hypertensive nephropathy 30 (5) 87 (6.4) Oxalate nephropathy 31 (5.2) 24 (1.8) Calcineurin inhibitor toxicity 42 (7) 61 (4.5) Graft failure/retransplant 61 (10.2) 86 (6.4) Acute tubular necrosis 11 (1.8) 37 (2.7) Glomerulonephritides (all types) 58 (9.7) 113 (8.4) Hepatorenal syndrome 54 (9) 213 (15.8) Unknown etiology 31 (5.2) 73 (5.4) Other 172 (28.6) 386 (28.6) requiring subsequent kidney transplant. 5 Using modified criteria from the International Ascites Club to define HRS, we reexamined this question by reviewing our data on this population of patients over the ensuing decade. PATIENTS AND METHODS From January 1995 to December 2004, 1,405 adult cadaveric liver-only transplants were performed at Baylor University Medical Center. All patients had been evaluated preoperatively by a transplant team nephrologist. The diagnosis of HRS was based on a pretransplant serum creatinine (SCr) 2.0 mg/dl or a SCr 2.0 mg/dl if the patient required HD. The following criteria as outlined by the International Ascites Club were also met: 1) presence of acute or chronic liver disease; 2) absence of preexisting documented renal disease; 3) urinary sodium 10 mmol/l; 4) normal-sized kidneys on radiographic imaging; 5) absence of sepsis; and 6) lack of ongoing fluid losses at the time of transplant despite appropriate resuscitation. Type 1 HRS was defined as a rapid decline in renal function associated with fulminant or subfulminant hepatitis. Type 2 HRS was defined as progressive renal dysfunction associated with cirrhosis. Patients with prior transplants were excluded to eliminate calcineurin inhibitor (CNI) toxicity as a possible etiology of renal insufficiency. Data regarding age, gender, etiology of liver disease, MELD score, preoperative and postoperative laboratory evaluation, preoperative and postoperative dialysis requirements, need for mechanical ventilation, donor demographics and laboratory values, operative statistics and fluid use, length of intensive care unit and hospital stay, subsequent liver or kidney transplantation, and patient and graft survival were obtained from our prospectively maintained database and chart review. The standard immunosuppressive protocol (CNI, mycophenolate mofetil, and steroids) varied minimally over the study period. In general, for patients with renal failure posttransplant (demonstrated by oliguria/anuria), the CNI was held for 72 hours. If the patient produced urine in the face of an increasing SCr, the CNI dose was reduced to a low maintenance level (cyclosporine ng/ml or tacrolimus 3-5 ng/ml). If there was no improvement in renal function within 72 hours, antibody therapy with OKT3, or rarely thymoglobulin, was initiated and the CNI withheld for 7 to 14 days or until the patient s renal function improved. Patient survival was defined as time from transplantation to death or last follow-up. HD days were defined as total inclusive days during which HD (conventional or continuous) was required, pretransplant or posttransplant. The MELD equation used to calculate pretransplant disease severity was as follows: (0.957 log e [creatinine md/dl] log e [bilirubin mg/dl] and 1.12 log e [international normalized ratio of prothrombin time] 0.643) 10. The maximum SCr considered within the MELD equation was 4.0 mg/dl. Categorical data were analyzed using the 2-tailed Fisher s exact test for 2 by 2 tables and the likelihood ratio chi-squared test for larger tables. Continuous variables were compared using the 2-sample Wilcoxon test. Survival curves were estimated using the Kaplan Meier method and compared using the log-rank test. A P value less than 0.05 was considered statistically significant. All analyses were performed using SAS, version (SAS Institute, Cary, NC). RESULTS Using our criteria, 130 patients with HRS (82 male and 48 female; mean age 49 yr) were identified, for an overall incidence of 9% over the study period. A total of 13 patients developed type 1 HRS and 117 patients evolved with type 2 HRS. The etiology of liver failure for each type is shown in Table 2. Seven patients required a liver-only retransplantation for primary nonfunction (1 patient), early hepatic artery thrombosis (2 patients), late onset hepatic artery thrombosis with hepatic abscesses (3 patients), and recurrent hepatitis C (1 patient). Three patients have subsequently had a kidney transplant (2 cadaveric, 1 live donor) for CNI-related nephrotoxicity, glomerulosclerosis, and diabetic nephropathy, and 1 patient has required a CLKT for late onset hepatic artery thrombosis and ESRD secondary to multifactorial causes (CNI toxicity, sepsis, and prolonged acute tubular necrosis).

3 840 RUIZ ET AL. TABLE 2. Primary Etiology of Liver Failure in 130 HRS Cases HRS Type 1 N HRS Type 2 N Fulminant hepatic failure 8 Hepatitis C 42 Subfulminant hepatic failure 5 Alcoholic cirrhosis 23 Cryptogenic cirrhosis 16 Primary sclerosing cholangitis 6 Autoimmune hepatitis 5 Hepatitis C/hepatocellular cancer 5 Primary biliary cirrhosis 4-1-antitrypsin syndrome 3 Budd Chiari syndrome 3 Hepatitis B 4 Alcohol/hepatocellular cancer 2 Nonalcoholic steatohepatitis 2 Caroli s disease 1 Secondary biliary cirrhosis 1 Total Figure 1. Comparison of patient survival in HRS and non- HRS patients undergoing liver-only transplantation. (Patients undergoing retransplantation were excluded from both groups.) At the completion of the analysis, 58 patients (45%) had died. Causes of death included sepsis or organ failure (21 patients), cardiac complications (7 patients), malignancy and graft failure (6 patients each), central nervous system and gastrointestinal complications (4 patients each), and unknown reasons in 10 patients. With a median follow-up of 47 months, the 1-, 3-, and 5-yr patient survival rates were 74%, 68%, and 62%, respectively, for the entire HRS group. As shown in Figure 1, there is a statistically significant difference in patient survival when compared to non-hrs patients over the same time period (P ). Survival rates at 1 and 5 yr for the 13 patients with type 1 HRS were 77% and 69%, respectively. For patients with type 2 HRS, the 1-, 3-, and 5-yr patient survival rates were 74%, 68%, and 61%, respectively. We analyzed donor and recipient risk factors for these patients and found none that correlated with poor 1-yr survival including the need for pretransplant dialysis (Table 3). We then compared those patients on HD, the median duration of which was 5 days (range, 1-36 days), to those not requiring HD. Although patient acuity was more severe in the HD group, as shown by significant differences in MELD score and preoperative life support, there was no difference in survival despite a longer postoperative course (Table 4). We also found no significant difference in survival when comparing patients on HD less than 14 days to those on HD greater than 14 days prior to transplant (P 0.52, data not shown). However, duration of posttransplant HD greater than 30 days did correlate with poor outcome (P , Fig. 2). A total of 11 patients required HD posttransplant for greater than 30 days (Table 5). Of the 3 patients requiring HD from 30 to 60 days, 1 patient recovered renal function (#2) only to die from sepsis at 2 months. Of the 8 patients on HD greater than 60-days, 7 became dialysis dependent with 1 patient (#4) recovering renal function after 103 days of dialysis. Patient #5 developed late onset acute renal failure. Four patients in this subgroup expired from sepsis within 1 yr of transplant. Among the preoperative and donor variables analyzed for these 7 patients, none had value in predicting posttransplant renal failure (data not shown). Over the same time period, the incidence of developing irreversible kidney failure in non-hrs patients was 0.34%, a significant difference when compared to the seven patients in the HRS group (P ). DISCUSSION As outlined in the major criteria from the International Ascites Club, HRS is defined as patients having a minimum baseline SCr of 1.5 mg/dl, with type 1 HRS characterized as a doubling of SCr to greater than 2.5 mg/dl in less than 2 weeks, and type 2 HRS classified as a more gradual decline in renal function. 6 We present some deficiencies with using this definition in transplantation. For example, patients presenting with fulminant hepatitis may not have time to double their

4 HEPATORENAL SYNDROME 841 TABLE 3. Effect of Various Patient and Donor Characteristics on 1-yr Survival for Patients With Type 2 HRS Undergoing LT Survival 1 yr (N 86) Survival 1 yr (N 31) P value Recipient factors Male 60 (70%) 17 (55%) Age, median, yr Pre-LT ICU status, patients 40 (47%) 20 (65%) Pre-LT mechanical ventilation, patients 10 (12%) 8 (26%) Pre-LT hemodialysis, patients 39 (45%) 19 (61%) Pre-LT calculated MELD Pre-LT albumin, median, gm/dl Donor factors Male 47 (55%) 17 (55%) Donor age, median, yr Peak donor Na, median, mmol/l Peak donor Cr, median, mg/dl Operative factors Piggyback placement 6 (8%) 4 (15%) Liver cold ischemia time, median, hours Liver warm ischemia, median, minutes Abbreviations: ICU, intensive care unit. TABLE 4. Demographic and Outcome Comparisons of Type 2 HRS Patients On and Off HD Prior to LT HRS HD (N 58) HRS HD (N 59) P value Recipient factors Male Age, median (yr) Pre-LT ICU status, patients 51 (88%) 9 (15%) Pre-LT mechanical ventilation, patients 17 (29%) 1 (2%) Calculated MELD pre-lt, median Operative factors PRBC (units) Systemic bypass, patients 30 (65%) 43 (80%) Liver cold ischemia time, median (hour) Liver warm ischemia, median (minute) Surgery time, median (hour) Outcomes Patients on HD after LT 55 (95%) 11 (19%) Duration of HD after LT, median (days) Length of stay, median (days) Patient survival, 1 yr 67% 80% GFR at 1-yr post-lt Abbreviations: ICU, intensive care unit; PRBC, packed red blood cells; GFR, glomerular filtration rate. SCr to greater than 2.5 mg/dl prior to death or transplantation. Additionally, using a baseline SCr of 1.5 mg/dl overrepresents patients with true intrinsic hepatorenal disease. By utilizing a minimum pretransplant SCr of 2.0 mg/dl, we were able to capture more patients with a type 1 presentation, exclude patients with probable reversible kidney dysfunction, if any, and eliminate patients based on a deviant or unexpected SCr value. For all HRS patients, we found a similar 5-yr patient survival as reported by Gonwa et al. 3 in an earlier institutional review from 1984 to 1992 (62% vs. 60%, respectively). Likewise, there was a statistically significant difference in patient survival when compared to non-hrs patients over this study period. For patients with type 2 HRS, no risk factors could be identified that would predict poor short-term patient survival, including the presence and duration of pretransplant HD. Based on our data, we cannot exclude or deny any patient with HRS a liver transplant alone.

5 842 RUIZ ET AL. Figure 2. Survival of type 2 HRS patients by time on HD posttransplantation. The overall course of HRS is unpredictable and likely influenced by the presence of other perioperative conditions or events. Recovery of renal function in the postoperative period may be delayed by hypovolemia, use of vasopressors, infection, subsequent reoperations, and nephrotoxic agents including the use of CNIs. Interestingly, renal function at 1-yr after LT was better preserved in the HD group (Table 3, P 0.012), which may argue in favor of early dialysis support pretransplant. In a previous study of HRS recipients at the University of California, Los Angeles, eventual return of native kidney function for patients on preoperative HD up to 1 month was demonstrated, with 7% of patients developing irreversible kidney failure (unpublished results from UCLA study). 7 In our current review, all but 1 patient required HD for less than 30 days with 7 patients (6%) requiring long-term HD. Thus, the authors of this study can also conclusively assert that there is no role for CLKT for patients with HRS on short-term dialysis pretransplant. We point out, however, that the incidence of ESRD progressing from acute renal failure in the non-hrs population has been reportedly less than 1%, 8-10 including 0.34% in our review. From this perspective, HRS patients are, at minimum, 6 times more likely to develop ESRD than their non-hrs counterparts. The discussion should thus shift to those HRS patients who become dialysis dependent after LT. In our analysis of 7 patients, all but 1 was initially discharged from the hospital. Five recipients subsequently died within 1 yr of transplant and 1 died secondary to complications of chronic renal failure. KALT should be strongly considered in this subset of patients. This approach, which has been previously recommended by Davis et al., 11 would allow time for patient stabilization and irreversible kidney function to declare itself. Instead of entertaining a CLKT for patients with HRS, a KALT in select patients would exemplify an improvement in organ utilization. For those institutions who argue that a kidney should be allocated to patients with HRS, our proposal would salvage 94 kidneys per 100 cases. For those concerned about line jumping or bypassing a potential cadaveric kidney transplant recipient, we point out that based on our data, a maximum of 7 kidneys over a 10-yr period would have been justly reallocated. It has been shown that post-lt patients who develop ESRD and receive a subsequent kidney transplant have significantly improved survival rates. 12 The relationship between transplant interval and outcome has been minimally studied however. In a review by Demirci et al patient had a kidney transplant 8 days after LT and died from sepsis on postoperative day 28. Another patient had a KALT 9 months afterward and was alive at 4-yr follow-up. Timing is the key component here. With the intent to eliminate the mortality of posttransplant renal failure, we propose a minimum 60-day waiting period before consideration for a priority KALT. While arbitrary, a 60-day period was chosen because our analysis shows that chances of renal recovery were low (1/8, or 13%) if a patient was on dialysis for greater than 60 days. Additionally, for the 5 patients who died within 1 yr of transplant, the median survival time was 139 days. A minimum 60-day period would have allowed ample time for a patient to be assessed, listed, and a KALT to be ultimately performed. A 30-day period was also entertained; however, 1 patient came off HD after 36 days and all 3 patients in this category remained acutely ill and never left the hospital. Thus, a 60-day period appears optimal from a utilitarian perspective. We understand that there are occasional patients who recover renal function after a prolonged period of time such as our single patient who came off HD more than 100 days after transplant. Clearly, the decision has to be individualized to make sure that there is no particular etiology that is likely to resolve. However, these occurrences are uncommon and based on our results, waiting a minimum of 100 days would not provide a sufficient window of opportunity to perform a KALT. In order to promptly list a patient for a kidney allograft, qualifying criteria must be met and medical clearance obtained. Unfortunately, published reports attempting to define the onset of ESRD in HRS patients are not available. Our study is the first to try to address this critical issue. An additional topic that merits some enlightening discussion is whether the hepatic immunoprotective effect applies to subsequently placed renal allografts. Immunoprotection by different donor KALTs, as defined by incidence of chronic rejection and allograft half-life, was not demonstrated in a United Network for Organ Sharing analysis of 352 cases. 14 This may be due to the effect of organs from different donors or the result of extended timing between transplants. Performing accelerated KALTs may answer the question regarding the effect of timing on immunoprotection. Our study is limited by its retrospective design and small sample size of patients developing irreversible kidney failure. Another perceived drawback includes using a minimum SCr 2.0 mg/dl to define HRS. We

6 HEPATORENAL SYNDROME 843 TABLE 5. Outcome of 11 HRS Patients on HD Greater Than 30 Days Posttransplantation Start of HD Patient Patient number Pre-LT HD (days) after LT (day) Days of HD posttransplantation status (A/D) Patient days Cause of death 1 Y (1) 0 31 D 31 Hemorrhage 2 Y (3) 0 36 D 64 Sepsis 3 Y (15) 0 48 D 48 Status epilepticus 4 Y (11) A 1028 n/a 5 N D 135 Sepsis 6 Y (5) D 127 Sepsis 7 Y (9) D 139 Sepsis 8 N D 214 MI 9 Y (15) D 308 Sepsis 10 Y (1) D 971 CRF 11 N A 1121 n/a Abbreviations: A/D, alive/dead; n/a, not applicable; MI, myocardial infarction; CRF, chronic renal failure. likely eliminated potential recipients with severe muscle wasting pretransplantation and smaller-sized patients. However, we argue using a higher SCr is a more specific test for inclusion of patients with significant renal dysfunction. Additionally, we were able to include 18 patients with a SCr 2.0 mg/dl but on HD at the time of transplant. While more reliable tests such as creatinine clearance, MAG-3 (Tc-99m-mercaptoacetyltriglycine) or Glofil (I 125 -iothalamate) scans are available to assess renal function, SCr levels are easily obtainable, reproducible, and less costly. CONCLUSION There is no role for CLKT for patients with HRS on short-term HD pretransplantation, as demonstrated by return of native renal function and an acceptable glomerular filtration rate at 1-yr posttransplant. Persistent renal dysfunction posttransplant contributes to 1-yr mortality. Thus, we recommend a minimum 60- day waiting period for consideration of subsequent KALT, whether deceased or live donor. The decision whether to prioritize these patients merits further discussion. REFERENCES 1. Alessandria C, Ozdogan O, Guevara M, Restuccia T, Jimenez W, Arroyo V, et al. MELD score and clinical type predict prognosis in hepatorenal syndrome: relevance to liver transplantation. Hepatology 2005;41: Gines A, Escorsell A, Gines P, Salo J, Jimenez W, Inglada L, et al. Incidence, predictive factors, and prognosis of the hepatorenal syndrome in cirrhosis with ascites. Gastroenterology 1993;105: Gonwa TA, Klintmalm GB, Levy M, Jennings LS, Goldstein RM, Husberg BS. Impact of pretransplant renal function on survival after liver transplantation. Transplantation 1995;59: Based on OPTN data as of November 24, Jeyarajah DR, Gonwa TA, McBride M, Testa G, Abbasoglu O, Husberg BS, et al. Hepatorenal syndrome: combined liver kidney transplants versus isolated liver transplant. Transplantation 1997;64: Arroyo V, Gines P, Gerbes AL, Dudley FJ, Gentilini P, Laffi G, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. International Ascites Club. Hepatology 1996;23: Ruiz R, Kunitake H, Wilkinson AH, Danovitch GM, Farmer DG, Ghobrial RM, et al. Long-term analysis of combined liver and kidney transplantation at a single center. Arch Surg 2006;141: Junge G, Schewior LV, Kohler S, Neuhaus R, Langrehr JM, Tullius S, et al. Acute renal failure after liver transplantation: Incidence, etiology, therapy, and outcome. Transplant Proc 2006;38: Chuang FR, Lin CC, Wang PH, Cheng YF, Hsu KT, Chen YS, et al. Acute renal failure after cadaveric related liver transplantation. Transplant Proc 2004;36: Gonwa TA, Morris CA, Goldstein RM, Husberg BS, Klintmalm GB. Long-term survival and renal function following liver transplantation in patients with and without hepatorenal syndrome experience in 300 patients. Transplantation 1991;51: Davis CL, Gonwa TA, Wilkinson AH. Identification of patients best suited for combined liver-kidney transplantation: Part II. Liver Transpl 2002;8: Paramesh AS, Roayaie S, Doan Y, Schwartz ME, Emre S, Fishbein T, et al. Post-liver transplant acute renal failure: factors predicting development of end-stage renal disease. Clin Transplant 2004;18: Demirci G, Becker T, Nyibata M, Lueck R, Bektas H, Lehner F, et al. Results of combined and sequential liverkidney transplantation. Liver Transpl 2003;9: [Erratum in: Liver Transpl 2004;10:329.] 14. Simpson N, Cho YW, Iwaki Y, Selby RR, Fong T. Comparison of renal allograft outcome in combined liver kidney transplant versus later kidney transplant in liver transplant recipients: analysis of UNOS database. Presented at the 2005 American Transplant Congress, sponsored by the American Society of Transplant Surgeons and the American Society of Transplantation, Seattle, WA, May 21-25, 2005.