- Email: [email protected]
Preservation injury patterns in liver transplantation associated with poor prognosis
Preservation Injury Patterns in Liver Transplantation Associated With Poor Prognosis Y.-M. Lee, C.B. O’Brien, N. Yamashiki, M. Behro, D. Weppler, A.G. Tzakis, and E.R. Schiff ABSTRACT Preservation injury (PI) is defined as hepatic dysfunction that occurs within 10 days of liver transplantation (OLT) but spontaneously resolves. However, we noted two new patterns: one characterized by histologic evidence of preservation injury that occurs at later than 10 days post-OLT (late PI), and a second, of persistent charge in liver biopsies ⬎ 10 days post-OLT (persistent PI). To characterize these new patterns, we performed a retrospective study of patients who underwent liver biopsies for hepatic dysfunction post-OLT from September 1993 to March 1998. The outcome of the 61 patients with preservation injury on liver biopsy after OLT was followed until the last clinic visit or death. Thirty patients had early PI, 16 patients had persistent preservation injury and 15 patients, late onset preservation injury. There were no significant differences in the age (P ⫽ .28), sex (P ⫽ .77), follow-up time (P ⫽ .78), cold ischemia (P ⫽ .3), or warm ischemia time (P ⫽ .16) between these groups. There was also no significant association between early preservation injury or persistent preservation injury with the development of acute or chronic rejection (P ⫽ .19). The overall survival rates at 1, 3, and 5 years was 52%, 45%, and 45%, respectively. There was no significant difference in survival between early, persistent, and late PI patterns (P ⫽ .59), although there was a trend toward better survival for patients with early preservation injury. The survival of OLT patients with persistent or late preservation injury is poor and should prompt consideration for retransplantation.
P
RESERVATION INJURY (PI) has been described as hepatic dysfunction occurring early in the posttransplant period, usually within the first 2 weeks.1 It occurs as a result of injury during the harvesting transportation and reperfusion processes. If mild, it resolves spontaneously over several days to weeks, but a severe insult may resolve slowly over several months with residual histological damage.1 PI is a diagnosis of exclusion, based on a characteristic histological pattern where acute cellular rejection, recurrent hepatitis B or C, hepatic artery thrombosis, or biliary leak are absent. Williams et al found that among 44 patients with 55 episodes of hepatic dysfunction within 48 hours post–liver transplant, 33 episodes were ascribed to PI.1 Subsequently, others reported PI in up to 54% of patients.2 Tillery et al described preservation injury as sinusoidal neutrophilia, microvesicular steatosis, hepatocellular cytoaggregation, and pyknosis in the early phases and centrilobular necrosis, hepatocyte swelling, and cholestasis developing in the later stages.3 The risk factors for preservation injury appear to be cold ischemia time, higher peak ALT and AST,3 donor age, high inotropic drug use, moderate to
severe macrovesicular steatosis, and prolonged stay in the intensive care unit.4 PI appears to result in increased subsequent rejection and biliary complications.5–7 Goldstein et al described a pattern of liver injury similar to PI occurring after the early posttransplant period, which was reversible except when superimposed with perivenular necrosis that resulted in retransplantation or death.8 We have also observed the appearance of PI that did not resolve spontaneously (which we defined as the “persistent PI pattern”) as well as the de novo appearance of PI at greater than 10 days posttransplant (which we defined as the “late PI pattern”). We sought to study both new patterns of PI, particularly with respect to their effect on the prognosis of both the patient and the transplanted organ. From the National University Hospital Singapore, Department of Medicine, Singapore, Singapore. Address reprint requests to Yin-Mei Lee, MD, National University Hospital Singapore, Department of Medicine, 5 Lower Kent Ridge Rd, Singapore 119074, Singapore. E-mail: mdeleeym.nus. edu.sg
0041-1345/03/$–see front matter doi:10.1016/j.transproceed.2003.10.084
© 2003 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
2964
Transplantation Proceedings, 35, 2964⫺2966 (2003)
PRESERVATION INJURY PATTERNS
2965 Table 1. Patient Demographics
No. of patients Age (range) Male/female Follow-up time (range)
Early PI
Persistent PI
Late PI
P value
30 48.8 (0.8 –75) 15/15 25.5 (0.33– 87.2)
16 51.3 (0.8 –75) 12/4 26.9 (0.33–75.7)
15 52.3 (0.8 –75) 12/4 26.3 (0.57– 69.1)
.28 .77 .78
Age is median years. Follow-up time is mean months.
MATERIALS AND METHODS Data were collected retrospectively from a computerized liver transplant database of all patients who underwent liver biopsies for hepatic dysfunction post–liver transplantation between September 1993 and March 1998. The liver biopsies were performed with either a Jamshidi or Trucut needle, rapidly embedded in paraffin and stained with hematoxylin and eosin. We excluded acute rejection, hepatic artery thrombosis, biliary leak, drug toxicity, sepsis, and viral infections (including CMV and hepatitis C). The diagnosis of PI was reconfirmed by a second pathologist to avoid interobserver bias. The database included the donor information of age, liver steatosis, intensive care stay under ventilatory support, and cold ischemia time. We examined the subsequent biopsies of these patients and recorded the final outcome of these subjects until the last clinic visit or death.
Statistics Student t test, chi-square test, Kaplan-Meier, and log-rank tests were used and a P value of ⬍ .05 was considered significant.
RESULTS
There were 61 patients in our study diagnosed with PI, a prevalence of 13%. Patient demographics are shown in Table 1. The diagnosis of late PI was made at an average of 112 days after transplant (range 11 to 1098). Only three patients had prior biopsies with two reported as mild acute rejection while the other showed acute pericholangitis. There was also no significant association between early or persistent PI with acute or chronic rejection (P ⫽ .19). We found no significant differences between the three groups for donor age (P ⫽ .1), donor sex mismatch (P ⫽ .23), presence of moderate (30% to 60%) or severe steatosis (⬎60%) in the donor liver (P ⫽ .06), prolonged ICU stay of more than 4 days (P ⫽ .38), intraoperative blood loss (P ⫽ .2), cold (P ⫽ .3) and warm (P ⫽ .16) ischemia times. Overall survival rates of the whole group of patients at 1, 3, and 5 years was 52%, 47%, and 47%, respectively, with no differences in survival by the log-rank test (P ⫽ .59). This observation is significantly reduced compared to the 1- and 3-year survival of 74.2% and 66.7%, respectively, for patients without PI at our center. Furthermore, there was a trend toward the worst survival in the group with persistent PI as their 1-, 3-, and 5-year survival rates were 41%, 28%, and 28%, respectively, as compared with the early PI group survivals of 57%, 53%, and 53% (see Fig 1). DISCUSSION
We have shown that almost 50% of PI occurs outside the classical definition. Since our study is a retrospective one,
many patients with late PI did not have liver biopsies during the early posttransplant period and may actually belong to the persistent PI group, who continue to demonstrate the changes of PI up to 7 years post–liver transplant. The exact pathologic mechanism causing this phenomenon is unknown. The three groups of PI may represent a spectrum of morphological changes secondary to an insult sustained during the process of harvesting, transportation, and reperfusion. Clearly, protocol biopsies beginning in the postreperfusion period need to be performed so that we can better define these two groups of late and persistent PI in subsequent studies. We did not find any differences between the three groups in terms of patient demographics or warm and cold ischemia times3 as well as donor age, donor sex mismatch, presence of moderate or severe steatosis in the donor liver, prolonged donor ICU stay of more than 4 days, and intraoperative blood loss, factors that have reportedly been associated with PI.4 We also did not find an increased risk of subsequent rejection associated with PI as reported by earlier studies,5,6 possibly because they diagnosed for PI by a rise in AST and hepatocyte swelling, which subsequent studies found to not be specific for PI.9,10 We have shown that the 1-, 3-, and 5-year survival rates of patients diagnosed with PI were worse than the overall post-OLT survival rates reported for those without PI during the same period. Furthermore, we observed a trend toward the worst survival among patients with the persistent PI pattern compared to those with early PI. Although this did not reach a level of significance (due to the small
Fig 1. Kaplan-Meier survival curves for early; persistent; and late preservation injury patterns. Comparison of survival of all three groups by log-rank test, P ⫽ .59.
2966
numbers of patients), nevertheless, it indicates that the presence of persistent PI (or the development of the late PI pattern in subsequent liver biopsies) is an ominous sign. In conclusion, we have described patterns of PI that are distinct from those that occur within the early transplant period, which is found in almost 50% of patients with PI. We have also shown that all forms of PI are associated with poor survival post-OLT, warranting consideration for early retransplantation. REFERENCES 1. Williams JW, Santiago V, Peters TG, et al: Cholestatic jaundice after hepatic transplantation. Am J Surg 151:65, 1986 2. Ng IO, Bourroughs AK, Rolles K, et al: Hepatocellular ballooning after liver transplantation, a light and electronmicroscopic study with clinicopathological correlation. Histopathology 18:323, 1991 3. Tillery W, Demetris J, Watkins D, et al: Pathological recognition of preservation injury in hepatic allografts within six months of follow up. Transplant Proc 21:1330, 1989
LEE, O’BRIEN, YAMASHIKI ET AL 4. Briceno J, Marchal T, Padillo J, et al: Influence of marginal donors on liver preservation injury. Transplantation 74:522, 2002 5. Howard TK, Klintmalm GB, Cofer JB, et al: The influence of preservation injury or rejection in the hepatic transplant recipient. Transplantation 49:103, 1990 6. Abraham S, Furth EE: Quantitative evaluation of histological features in “time zero” liver allograft biopsies as predictors of rejection or graft failure. receiver operating characteristic analysis application. Hum Pathol 27:1077, 1996 7. Busquests J, Figueras J, Serrano T, et al: Postperfusion biopsies are useful in predicting complications after liver transplantation. Liver Transpl 7:432, 2001 8. Goldstein NS, Hart J, Lewin KJ: Diffuse hepatocyte ballooning in liver biopsies from orthotopic liver transplant patients. Histopathology 8:331, 1991 9. Ray R, Lewin KJ, Colonna J, et al: The role of liver biopsy in evaluating acute allograft dysfunction following liver transplantation. a histological correlation of 34 liver transplants. Hum Pathol 19:835, 1988 10. Hubscher SG: Histological finding in liver allograft rejection—new insights into the pathogenesis of hepatocellular damage in liver allografts. Histopathology 18:377, 1991
P
RESERVATION INJURY (PI) has been described as hepatic dysfunction occurring early in the posttransplant period, usually within the first 2 weeks.1 It occurs as a result of injury during the harvesting transportation and reperfusion processes. If mild, it resolves spontaneously over several days to weeks, but a severe insult may resolve slowly over several months with residual histological damage.1 PI is a diagnosis of exclusion, based on a characteristic histological pattern where acute cellular rejection, recurrent hepatitis B or C, hepatic artery thrombosis, or biliary leak are absent. Williams et al found that among 44 patients with 55 episodes of hepatic dysfunction within 48 hours post–liver transplant, 33 episodes were ascribed to PI.1 Subsequently, others reported PI in up to 54% of patients.2 Tillery et al described preservation injury as sinusoidal neutrophilia, microvesicular steatosis, hepatocellular cytoaggregation, and pyknosis in the early phases and centrilobular necrosis, hepatocyte swelling, and cholestasis developing in the later stages.3 The risk factors for preservation injury appear to be cold ischemia time, higher peak ALT and AST,3 donor age, high inotropic drug use, moderate to
severe macrovesicular steatosis, and prolonged stay in the intensive care unit.4 PI appears to result in increased subsequent rejection and biliary complications.5–7 Goldstein et al described a pattern of liver injury similar to PI occurring after the early posttransplant period, which was reversible except when superimposed with perivenular necrosis that resulted in retransplantation or death.8 We have also observed the appearance of PI that did not resolve spontaneously (which we defined as the “persistent PI pattern”) as well as the de novo appearance of PI at greater than 10 days posttransplant (which we defined as the “late PI pattern”). We sought to study both new patterns of PI, particularly with respect to their effect on the prognosis of both the patient and the transplanted organ. From the National University Hospital Singapore, Department of Medicine, Singapore, Singapore. Address reprint requests to Yin-Mei Lee, MD, National University Hospital Singapore, Department of Medicine, 5 Lower Kent Ridge Rd, Singapore 119074, Singapore. E-mail: mdeleeym.nus. edu.sg
0041-1345/03/$–see front matter doi:10.1016/j.transproceed.2003.10.084
© 2003 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
2964
Transplantation Proceedings, 35, 2964⫺2966 (2003)
PRESERVATION INJURY PATTERNS
2965 Table 1. Patient Demographics
No. of patients Age (range) Male/female Follow-up time (range)
Early PI
Persistent PI
Late PI
P value
30 48.8 (0.8 –75) 15/15 25.5 (0.33– 87.2)
16 51.3 (0.8 –75) 12/4 26.9 (0.33–75.7)
15 52.3 (0.8 –75) 12/4 26.3 (0.57– 69.1)
.28 .77 .78
Age is median years. Follow-up time is mean months.
MATERIALS AND METHODS Data were collected retrospectively from a computerized liver transplant database of all patients who underwent liver biopsies for hepatic dysfunction post–liver transplantation between September 1993 and March 1998. The liver biopsies were performed with either a Jamshidi or Trucut needle, rapidly embedded in paraffin and stained with hematoxylin and eosin. We excluded acute rejection, hepatic artery thrombosis, biliary leak, drug toxicity, sepsis, and viral infections (including CMV and hepatitis C). The diagnosis of PI was reconfirmed by a second pathologist to avoid interobserver bias. The database included the donor information of age, liver steatosis, intensive care stay under ventilatory support, and cold ischemia time. We examined the subsequent biopsies of these patients and recorded the final outcome of these subjects until the last clinic visit or death.
Statistics Student t test, chi-square test, Kaplan-Meier, and log-rank tests were used and a P value of ⬍ .05 was considered significant.
RESULTS
There were 61 patients in our study diagnosed with PI, a prevalence of 13%. Patient demographics are shown in Table 1. The diagnosis of late PI was made at an average of 112 days after transplant (range 11 to 1098). Only three patients had prior biopsies with two reported as mild acute rejection while the other showed acute pericholangitis. There was also no significant association between early or persistent PI with acute or chronic rejection (P ⫽ .19). We found no significant differences between the three groups for donor age (P ⫽ .1), donor sex mismatch (P ⫽ .23), presence of moderate (30% to 60%) or severe steatosis (⬎60%) in the donor liver (P ⫽ .06), prolonged ICU stay of more than 4 days (P ⫽ .38), intraoperative blood loss (P ⫽ .2), cold (P ⫽ .3) and warm (P ⫽ .16) ischemia times. Overall survival rates of the whole group of patients at 1, 3, and 5 years was 52%, 47%, and 47%, respectively, with no differences in survival by the log-rank test (P ⫽ .59). This observation is significantly reduced compared to the 1- and 3-year survival of 74.2% and 66.7%, respectively, for patients without PI at our center. Furthermore, there was a trend toward the worst survival in the group with persistent PI as their 1-, 3-, and 5-year survival rates were 41%, 28%, and 28%, respectively, as compared with the early PI group survivals of 57%, 53%, and 53% (see Fig 1). DISCUSSION
We have shown that almost 50% of PI occurs outside the classical definition. Since our study is a retrospective one,
many patients with late PI did not have liver biopsies during the early posttransplant period and may actually belong to the persistent PI group, who continue to demonstrate the changes of PI up to 7 years post–liver transplant. The exact pathologic mechanism causing this phenomenon is unknown. The three groups of PI may represent a spectrum of morphological changes secondary to an insult sustained during the process of harvesting, transportation, and reperfusion. Clearly, protocol biopsies beginning in the postreperfusion period need to be performed so that we can better define these two groups of late and persistent PI in subsequent studies. We did not find any differences between the three groups in terms of patient demographics or warm and cold ischemia times3 as well as donor age, donor sex mismatch, presence of moderate or severe steatosis in the donor liver, prolonged donor ICU stay of more than 4 days, and intraoperative blood loss, factors that have reportedly been associated with PI.4 We also did not find an increased risk of subsequent rejection associated with PI as reported by earlier studies,5,6 possibly because they diagnosed for PI by a rise in AST and hepatocyte swelling, which subsequent studies found to not be specific for PI.9,10 We have shown that the 1-, 3-, and 5-year survival rates of patients diagnosed with PI were worse than the overall post-OLT survival rates reported for those without PI during the same period. Furthermore, we observed a trend toward the worst survival among patients with the persistent PI pattern compared to those with early PI. Although this did not reach a level of significance (due to the small
Fig 1. Kaplan-Meier survival curves for early; persistent; and late preservation injury patterns. Comparison of survival of all three groups by log-rank test, P ⫽ .59.
2966
numbers of patients), nevertheless, it indicates that the presence of persistent PI (or the development of the late PI pattern in subsequent liver biopsies) is an ominous sign. In conclusion, we have described patterns of PI that are distinct from those that occur within the early transplant period, which is found in almost 50% of patients with PI. We have also shown that all forms of PI are associated with poor survival post-OLT, warranting consideration for early retransplantation. REFERENCES 1. Williams JW, Santiago V, Peters TG, et al: Cholestatic jaundice after hepatic transplantation. Am J Surg 151:65, 1986 2. Ng IO, Bourroughs AK, Rolles K, et al: Hepatocellular ballooning after liver transplantation, a light and electronmicroscopic study with clinicopathological correlation. Histopathology 18:323, 1991 3. Tillery W, Demetris J, Watkins D, et al: Pathological recognition of preservation injury in hepatic allografts within six months of follow up. Transplant Proc 21:1330, 1989
LEE, O’BRIEN, YAMASHIKI ET AL 4. Briceno J, Marchal T, Padillo J, et al: Influence of marginal donors on liver preservation injury. Transplantation 74:522, 2002 5. Howard TK, Klintmalm GB, Cofer JB, et al: The influence of preservation injury or rejection in the hepatic transplant recipient. Transplantation 49:103, 1990 6. Abraham S, Furth EE: Quantitative evaluation of histological features in “time zero” liver allograft biopsies as predictors of rejection or graft failure. receiver operating characteristic analysis application. Hum Pathol 27:1077, 1996 7. Busquests J, Figueras J, Serrano T, et al: Postperfusion biopsies are useful in predicting complications after liver transplantation. Liver Transpl 7:432, 2001 8. Goldstein NS, Hart J, Lewin KJ: Diffuse hepatocyte ballooning in liver biopsies from orthotopic liver transplant patients. Histopathology 8:331, 1991 9. Ray R, Lewin KJ, Colonna J, et al: The role of liver biopsy in evaluating acute allograft dysfunction following liver transplantation. a histological correlation of 34 liver transplants. Hum Pathol 19:835, 1988 10. Hubscher SG: Histological finding in liver allograft rejection—new insights into the pathogenesis of hepatocellular damage in liver allografts. Histopathology 18:377, 1991