Thrombotic Microangiopathy After LivingDonor ... - Wiley Online Library

American Journal of Transplantation 2012; 12: 728–736 Wiley Periodicals Inc.

C

Copyright 2011 The American Society of Transplantation and the American Society of Transplant Surgeons doi: 10.1111/j.1600-6143.2011.03841.x

Thrombotic Microangiopathy After Living-Donor Liver Transplantation J. Shindoha , Y. Sugawaraa, *, N. Akamatsub , J. Kanekoa , S. Tamuraa , N. Yamashikic , T. Aokia , Y. Sakamotoa , K. Hasegawaa and N. Kokudoa a Artificial Organ and Transplantation Division, Department of Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan b Department of Hepato-Biliary-Pancreatic Surgery, Saitama Medical Center, Saitama Medical University, Saitama, Japan c Organ Transplantation Service, University of Tokyo, Tokyo, Japan *Corresponding author: Yasuhiko Sugawara, [email protected]

Thrombotic microangiopathy (TMA) is an infrequent but severe life-threatening disorder in solid organ transplant recipients. Few studies of TMA in living donor liver transplant (LDLT) recipients, however, have been reported. We investigated the clinical characteristics and prognostic factors of TMA after LDLT. Among 393 adult LDLT recipients, 30 patients (7.6%) were identified to have TMA. The 1-, 3- and 5-year survival rates of these patients were lower (60.6%, 52.5% and 47.7%, respectively) than those of patients without TMA (93.0%, 89.0% and 87.3%, respectively). Multivariate analysis confirmed that reduced administration of fresh frozen plasma and sensitization against HLA are closely related with TMA (odds ratio [OR]: 2.6 and 16.1, respectively). However, a review of the cases revealed that individual responses to treatment varied considerably and the main etiologies were difficult to determine. A comparison of the clinical factors suggested that late onset (>30 days), poor response to treatment and delayed diagnosis and/or treatment are associated with a poor outcome. Because the prevention of TMA in LDLT patients is difficult, early diagnosis and initiation of intensive therapies may be crucial to improve the prognosis. Key words: Hemolytic uremic syndrome, liver transplantation, living donor, thrombotic microangiopathy Abbreviations: CP, cleaving protease; CNI, calcineurin inhibitor; FFP, fresh frozen plasma; HCV, hepatitis C virus; LDH, lactate dehydrogenase; LDLT, living donor liver transplantation; OR, odds ratio; POD, postoperative day; TMA, thrombotic microangiopathy; vWF, von Willebrand factor.

728

Received 19 July 2011, revised 08 September 2011 and accepted for publication 26 September 2011

Introduction Thrombotic microangiopathy (TMA) is an infrequent but severe life-threatening disorder characterized by systemic or intrarenal aggregation of platelets, thrombocytopenia and mechanical injury to erythrocytes. Without effective treatment, the mortality of TMA is as high as 90% (1). On the basis of the previous reports, TMA develops after solid organ transplantation with an incidence of 0.5– 15% (2–5). Although the actual etiology of this disorder in posttransplant patients is difficult to determine, a history of transplantation itself and the administration of several drugs, including calcineurin inhibitors (CNIs), are considered to be risk factors for posttransplant TMA (1). To date, there are only sporadic reports of TMA in liver transplant recipients, whereas this disorder is well recognized in the field of renal transplantation (6). A recent histopathologic study of renal biopsy specimens in posttransplant patients, however, indicated that TMA might be more prevalent in liver transplant recipients than in renal transplant recipients when including subclinical cases (7). Once TMA occurs in liver transplant recipients, the clinical course severely deteriorates and the mortality rate is greater than 40%, even after intensive treatments (8). Therefore, knowledge of posttransplant TMA has become more important in the perioperative management of liver transplant recipients. We previously reported our experience of 10 patients complicated with TMA after living donor liver transplantation (LDLT) in terms of the clinical features and outcomes (5). The underlying risk factors, posttransplant triggering events and optimal treatments for posttransplant TMA in LDLT recipients, however, remain unclear. In this study, we reviewed our LDLT recipients and analyzed the risk factors and clinical outcomes of posttransplant TMA.

TMA After LDLT

Methods

frozen plasma (FFP) used within the first week after transplantation was also investigated to reveal the correlation with TMA.

Patients A total of 470 patients (77 children, 393 adults) underwent LDLT at our institution between January 1996 and April 2011. Clinical records of the 393 adult recipients (209 men and 184 women ranging in age from 18 to 67 years [median age: 52 years]) were enrolled in the present analysis. Our surgical technique and basic perioperative procedures have been described in detail elsewhere (9–15). Living donors were selected after considering their age, blood type, graft size, liver function and confirming their desire to volunteer (16).

Immunosuppression protocol Continuous intravenous infusion of tacrolimus was started immediately after transplantation at a dose of 2.5 lg/kg/h. After confirming that the trough level exceeded 20 ng/mL, the dose was adjusted to maintain a target level of 15–20 ng/mL during the first week. Then, tacrolimus was administered orally to maintain a trough level of 14–16 ng/mL on postoperative days (POD) 8–14, 12–14 ng/mL on POD 15–29, 10–12 ng/mL on POD 30–89, 8–10 ng/mL on POD 90–179 and 5–8 ng/mL on POD 180 onward. If adverse events, such as convulsion or encephalopathy, were observed, tacrolimus was converted to cyclosporin. Also, if needed, 2000–3000 mg of mycofenolate mofetil or anti-CD25 antibody (Basiliximab) was added with a dose reduction or discontinuation of CNIs. Silorimus was not used in our institution.

Diagnosis and treatment of TMA According to the reported criteria (1,5,17), diagnosis of TMA was made on the basis of all or some of the following: thrombocytopenia (platelet count 5.0 × 104 /mm3 within 24 h), microangiopathic hemolytic anemia (hemoglobin level of ≤8 g/dL with sharply elevated serum lactate dehydrogenase [LDH] levels [typically >500 U/L] and the presence of fragmented erythrocytes in the blood smear) and progressive renal failure (serum creatinine level ≥1.2 mg/dL) during treatment. The differential diagnosis of disseminated intravascular coagulation was made on the basis of the lack of laboratory findings suggestive of the consumption of clotting factors with an international normalized ratio of prothrombin time less than 1.5.

Second, individual clinical data of TMA were studied in detail. Trends of laboratory data for platelet count, total bilirubin level, creatinine level, LDH level, hemoglobin level, ferritin level and number of fragmented red cells counted per high power field (×400) were assessed in every case. To determine the prognostic factors, clinical data were compared between cases with early onset (≤30 days) and late onset (>30 days) and also between the survivors and deceased cases. Factors associated with TMA were inferred from the individual response to treatment. Additionally, graft regeneration rate was also assessed and compared with that in cases without TMA following the method reported previously (19).

Results The clinical records of the 393 adult-to-adult LDLT revealed 30 patients (7.6%) complicated with TMA after transplantation. Among the 30 patients, 1 patient experienced two episodes of TMA. Consequently, 31 TMA cases were identified in the present series. Of the 31 cases, 10 were deceased cases (32.3%) that were refractory to intensive care. Figure 1 represents the survival curves of LDLT recipients complicated with (n = 30) and without (n = 363) TMA. The 1-, 3- and 5-year survival rates of patients with TMA were significantly lower (60.6%, 52.5% and 47.7%, respectively) than those of patients without TMA (93.0%, 89.0% and 87.3%, respectively; p < 0.0001). Median time interval from diagnosis of TMA to death was 42 days (range: 11–91 days). Main causes of death were multiple organ failure (MOF) (n = 6) and sepsis (n = 4). Background characteristics of TMA Comparison of the patient characteristics and perioperative factors between cases with and without TMA is shown in

Once TMA was suspected, treatments, including conversion of calcineurin inhibitor (CNI) and/or plasma infusion or plasma exchange, were initiated immediately according to the therapeutic protocols at our institution (8,18).

Data analysis All analyses in this study were performed in accordance with the ethical guidelines for clinical studies at the University of Tokyo Hospital. Clinical data were recorded using an Excel 2010 (Microsoft) spreadsheet and analyzed using the statistical software JMP 9 (SAS Institute Japan, Tokyo, Japan). For the statistical analysis, a Wilcoxon’s rank-sum test was used for continuous data and chi-square test or Fischer’s exact test was used for categorical data as appropriate. A p-value of less than 0.05 was considered statistically significant. First, the background characteristics and clinical data were compared between cases with and without TMA. The analyzed factors included the recipient’s demographics, the donor’s demographics, graft types and volumes, surgical parameters and postoperative graft status. Survival curves of the patients were generated by the Kaplan–Meier method and compared by the log-rank test. The risk factors of TMA were then determined by multivariate analysis using factors with p < 0.20 in the univariate analysis as independent factors. Because TMA is a coagulation disorder and plasma infusion or exchange is a main firstline therapy, the total amount of fresh

American Journal of Transplantation 2012; 12: 728–736

Figure 1: Posttransplant survival curves of LDLT cases with and without TMA.

729

Shindoh et al. Table 1: Comparison of clinical factors between the cases with and without TMA

n Recipient factors Age1 Sex (male/female) MELD score2 HBV positive HCV positive Autoimmune diseases (PBC, PSC, AIH) Anti-HLA antibody Donor factors Age1 Blood type (identical/compatible/incompatible) Graft type (R/non R) Graft volume (ml)2 Graft volume vs. recipient’s SLV (%)2 Surgical factors OP time (min)2 Anhepatic phase (min)2 Warm ischemic time (min)2 Cold ischemic time (min)2 Blood loss (g)2 Transfusion (mL)2 Splenectomy Postoperative factors PV flow (cm/s)

FFP usage during POD from 0 to 7 Acute rejection

Positive Negative

Yes No After LDLT2 POD32 POD72 POD 0–3 mean2 POD 0–7 mean2 Yes No

TMA (−)

TMA (+)

363

30

p-Value

52 (18–67) 198/165 13.2 (10.0–18.4) 65 (17.9%) 104 (28.6%) 100 (27.6%) 2 (0.6%) 345 (99.4%)

53 (28–65) 11/19 14.2 (10.9–22.8) 3 (10.0%) 16 (53.3%) 8 (26.7%) 4 (13.3%) 26 (86.7%)

0.28 0.06 0.12 0.45 0.004 1 0.0004

36 (17–66) 275/88/0 200/163 525 (453–618) 46 (40–52)

38 (20–63) 24/6/0 15/15 526 (467–654) 47 (42–55)

0.92 0.96 0.59 0.59 0.25

867 (780–978) 165 (125–205) 60 (50–75) 109 (76–140) 5030 (3188–7952) 6100 (4360–8980) 147 (40.5%) 216 (59.5%)

868 (779–1020) 184 (168–221) 56 (35–79) 102 (79–127) 5545 (3225–9098) 6380 (3720–11500) 21 (70.0%) 9 (30.0%)

0.94 0.03 0.41 0.43 0.22 0.87 0.002

70.6 (49.6–98.5) 59.3 (41.0–85.2) 47.3 (32.2–63.5) 68.4 (49.8–93.3) 64.9 (48.3–86.6) 238 (64.6%) 103 (28.4%) 260 (71.6%)

69.6 (48.0–89.8) 51.1 (34.9–82.9) 47.3 (32.2–54.8) 60.1 (48.1–81.2) 57.1 (43.1–73.1) 12 (40.0%) 9 (30.0%) 21 (70.0%)

0.80 0.30 0.25 0.21 0.16 0.006 0.83

BMI = body mass index; MELD = model for endstage liver disease; HBV = hepatitis B virus; HCV = hepatitis C virus; PBC = primary biliary cirrhosis; PSC = primary sclerosing cholangitis; AIH = autoimmune hepatitis; R = right liver graft; SLV = standard liver volume; PV = portal vein; FFP = fresh frozen plasma. 1 Median (range). 2 Median (interquartile range).

Table 1. Median age of the patients with TMA was 53 (range: 28–65) and female sex was predominant (male 36.7% vs. female 63.3%). Fifty-three percent of the cases were positive for hepatitis C virus (HCV) antibodies. Median onset of TMA was POD 18 (range: POD 1–708). Median time to diagnosis from the onset of TMA was 2 days (range: 0–9 days). When TMA was clinically suspected, adequate treatments, including CNI conversions and/or plasma exchange, were initiated immediately.

culation in the kidney and mechanical hemolysis. Although hemoglobin levels did not change dramatically in comparison with the other laboratory data, a gradual decrease was observed accompanying abnormal ferritin levels and fragmented red cells in a blood smear. Generally, these laboratory data recovered when the disorder was clinically resolved. The serum ferritin levels and fragmented red cell count, however, fluctuated and remained at relatively high levels even after clinical resolution of the disorder.

The trends of the laboratory data in the present TMA series are shown in Figure 2. Typically, the primary changes began with a sudden decrease in the platelet count (within 2 days) and a gradual increase in the total bilirubin level with no clinical evidence of hepatic failure or obstructive jaundice (peaked at day 4). Then, increases in serum creatinine level (peaked at day 5) and LDH level (peaked at day 6) were observed, reflecting impairment of the microcir-

Risk factors for TMA Univariate analysis revealed significant differences in the existence of preformed anti-HLA antibody, the length of the anhepatic phase during transplantation, history of splenectomy and usage of FFP within the first week after transplantation (Table 1). To search for independent factors of posttransplant TMA, a multivariate analysis was performed by selecting the factors with p < 0.20 in the univariate

730


TMA After LDLT

Figure 2: Trends of laboratory data in TMA patients. Each graph represents median data of the 31 TMA cases. LDH = lactate dehydrogenase; FRC = fragmented red cells.

analysis as variables, which revealed that lack of infusion of FFP after transplantation (odds ratio [OR]: 2.6) and sensitization against HLA (OR: 16.1) might be significantly correlated with posttransplant TMA (Table 2). In the present series, preformed anti-HLA antibody was observed in six patients (1.5%). Among these, four patients were complicated with TMA and all of them died in the first or repeated episode of TMA. Posttransplant trough levels of tacrolimus were maintained almost within the target level and no apparent overdose was confirmed around the event of TMA in the present series. There was no significant difference in the trough

Table 2:

level between cases with and without preformed anti-HLA antibodies (Figure 3). Response to the treatments Figure 4 summarizes individual responses to the treatments after diagnosis of TMA in the present series. Of the 31 cases, 27 cases (87.1%) at least partially responded to the treatment based on the recovery of laboratory data or clinical manifestations. Close relation between the TMA and CNIs was strongly suspected in at least eight cases on the basis of positive response to a discontinuation or dose down of the CNIs (indicated with gray boxes in Figure 4). In contrast, seven cases responded to and were relieved by

Risk factors for TMA in post-LDLT recipients

Anti-HLA antibody (+) Disuse of FFP within 1wk after LDLT Splenectomy Female Mean PV flow (POD 0–7) 180 min HCV positive MELD score >14 Blood loss > 5000g

OR

[95% CI]

p-Value

16.1 2.6 2.3 2.1 1.8 1.5 1.5 1.4 1.0

[2.7–133] [1.1–6.6] [0.7–7.4] [0.8–5.3] [0.7–4.2] [0.6–3.7] [0.5–4.9] [0.6–1.7] [0.4–2.7]

0.003 0.03 0.15 0.11 0.20 0.36 0.46 0.43 0.99

OR = odds ratio; 95% CI = 95% confidence interval; FFP = fresh frozen plasma; PV = portal vein.


731

Shindoh et al.

Dose down or discontinuation of CNIs was selected in two cases after conversion from tacrolimus to cyclosporin and three cases during administration of cyclosporin. Of these, cyclosporin was discontinued and replaced with mycofenolate mofetil in two severe cases because of progression of TMA even after dose down of CNIs. Although one of them showed a temporary response to the treatment, both patients died of MOF or sepsis. The remaining three cases were successfully treated with dose down of cyclosporine. Mycophenolate mofetil was added in every case and anti-CD25 antibodies were used in two of three cases with temporal discontinuation of immunosuppressive agents.

Figure 3: Posttransplant tacrolimus trough levels in 30 TMA cases. Continuous line represents median trough levels of tacrolimus in the cases without preformed anti-HLA antibodies (n = 26) and dashed line indicates those in cases sensitized against HLA before transplantation (n = 4). Intergroup comparison revealed no significant difference at any postoperative period. POD, postoperative days.

plasma infusion or plasma exchange only, suggesting that deficiency or disorders in coagulation cascades contribute to TMA (indicated with black boxes in Figure 4). Of the 14 cases that underwent both CNI conversion and plasma therapies, 12 cases (85.7%) responded to the treatment well. However, the factors associated with TMA were difficult to specify in these cases.

Graft regeneration in TMA The graft regeneration rate was compared between the present cases (n = 30) and the cases without TMA (n = 199) reported in a previous study (19). As shown in Figure 5, the regeneration rates did not differ significantly between groups at 1 or 3 months after LDLT.

Comparison of clinical features and treatments of TMA according to clinical outcome To determine the poor prognostic factors of TMA, clinical features and treatment of TMA were reviewed and compared between the surviving cases (n = 21) and deceased cases (n = 10; Table 3). Baseline characteristics showed that in the deceased cases the median age of the patient was younger and preformed ant-HLA antibody was more frequent.

Figure 4: Therapeutic menu and individual responses of 31 TMA cases. a Tacrolimus was discontinued and micofenolate mofetil was introduced, b Treated only with continuous hemodiafiltration, c Cyclosporin was discontinued and micofenolate mofetil was introduced. FK = tacrolimus; CyA = cyclosporine.

732


TMA After LDLT

(Figure 4), however, making it difficult to specify the actual pathophysiology of TMA in the present series. Although the pathogenesis of TMA after solid organ transplantation is still poorly understood, a number of risk factors have been reported especially in the field of renal transplantation. They include cytomegalovirus infection, parvovirus B 19 infection, antiphospholipid antibodies, anticardiolipin antibodies in HCV-positive patients or malignancy and most importantly, cyclosporine and tacrolimus for immunosuppression (20). The development of TMA associated with CNIs is well documented in transplantation. In renal transplant patients treated with cyclosporine, the incidence of TMA is 4–15% (6). De novo TMA has also been documented in approximately 1% of patients receiving tacrolimus (2,21). The disease triggering effects of CNIs have been related to vasoconstriction, endothelial toxicity and prothrombotic and antifibrinolytic actions (22).

Figure 5: Graft regeneration after LDLT in cases with and without TMA. Changes in graft regeneration rates at 1 and 3 months after transplantation are shown for cases with (n = 27) and without (n = 199) TMA. Comparison between the cases with and without TMA revealed no significant difference in graft regeneration rates at each observation time (p > 0.05). NS = statistically not significant.

Regarding clinical features, late onset (>30 POD) seemed to be associated with poor outcome. Reflecting a poor response to treatment, minimum platelet count was lower, maximum serum LDH, bilirubin and creatinine levels were higher and time to recovery of these laboratory data were significantly longer in the deceased cases. Conversion or discontinuation of CNIs was performed in 14 (67%) cases among survivors and 8 (80%) cases among patients who died, respectively. Plasma infusion or exchange therapies were required in 16 (76%) and 8 (80%) cases, respectively. The time from onset to diagnosis and the lag time from diagnosis to the initiation of plasma exchange or CNI conversion also tended to be longer in the deceased cases.

Discussion This retrospective study revealed that TMA occurred in 7.6% of the adult LDLT recipients and their clinical courses were severely deteriorated, with a mortality rate of 32.2% despite intensive care. Analysis of the patient’s background data indicated that reduced infusion of FFP and sensitization against HLA may be potent risk factors of posttransplant TMA (Table 2). The severity of TMA and response to treatments varied considerably among individuals American Journal of Transplantation 2012; 12: 728–736

In the present series, trough levels of tacrolimus were maintained almost within the target level (Figure 3) and no apparent continuous overdose was confirmed around the event of TMA. A review of the individual responses to treatment, however, revealed a close relation with CNIs in, at least, 8 of 30 TMA cases. Therefore, toxicity of CNIs might be a major causes of TMA in LDLT patients (Figure 4). In addition to this, deficiency of some coagulation factors is suspected to be another potent factor associated with TMA. The deficiency of vWF-cleaving protease (CP) was recently reported to be an important contributor to the occurrence of familial or idiopathic TTP (23,24) and a similar pathophysiology has been suspected in posttransplant TMA. Nagazawa et al. (25) hypothesized that the severity of the vWF-CP deficit is involved in the occurrence of TMA in LDLT recipients because vWF-CP is exclusively synthesized in the liver (23,24) and significant reduction of vWF-CP and an increase of vWF in the patient with postLDLT TMA were reported in a recent study (26). Although plasma levels of vWF-CP and vWF were not measured in the present study, a close connection between the reduction of vWF-CP and TMA might be plausible in LDLT, because partial liver graft that meets both the “minimal” requirements of the recipients and the “minimal” risk of the donor is usually selected in LDLT and accordingly, it is easy to speculate that plasma vWF-CP level could be severely decreased in patients that received a partial liver graft. Actually, posttransplant TMA seems less frequent in deceased-donor liver transplantation (DDLT; Refs. 1,25,27– 35) compared with that in LDLT. In our institution, we experienced 14 DDLT cases during the same period and posttransplant TMA was diagnosed in only one patient, whose clinical features were far milder than those observed in LDLT cases and rapidly relieved by continuous infusion 733

Shindoh et al. Table 3:

Comparison of clinical factors based on the clinical outcomes of TMA

n Backgrounds Age Sex Male Female Background disease HBV HCV Autoimmune diseases (PBC, PSC, AIH) Graft type R1 nonR1 Splenectomy Yes No Anti-HLA antibody Positive Negative Acute rejection Yes No CNIs Tacrolimus Cyclosporine Clinical features Onset of TMA (POD) Early onset TMA (≤30 POD) Late onset TMA (>30 POD) Symptoms Low grade fever Psychiatric symptoms Laboratory data Minimum platelet count (104 /mm3 ) Time to nadir (days) Maximum LDH (U/L) Time to peak (days) Maximum bilirubin (mg/dL) Maximum creatinine (mg/dL) Fragmented red cells in blood smear Treatments CNI conversion or discontinuation PI/PE Interval from onset to diagnosis (days)2 Interval from diagnosis to PI/PE (days) Interval from diagnosis to Conversion (days)

Overall

Survival case

Deceased case

31

21

10

p-Value

53 (28–63)

54 (37–65)

44 (28–63)

0.04

11 (35%) 20 (65%)

7 (33%) 14 (67%)

4 (40%) 6 (60%)

0.97

3 (10%) 16 (5%) 9 (3%)

3 (14%) 10 (5%) 7 (33%)

0 6 (60%) 2 (20%)

0.53 0.79 0.42

16 (52%) 15 (48%)

9 (43%) 12 (57%)

7 (70%) 3 (30%)

0.25

21 (68%) 10 (32%)

15 (71%) 6 (29%)

6 (60%) 4 (40%)

0.82

5 (16%) 26 (84%)

1 (5%) 20 (95%)

4 (40%) 6 (60%)

0.03

10 (32%) 21 (68%)

7 (33%) 14 (67%)

3 (30%) 7 (70%)

0.70

25 (81%) 6 (19%)

18 (86%) 3 (14%)

7 (70%) 3 (30%)

0.36

19 (4–77) 17 (55%) 14 (45%)

15 (4–88) 13 (62%) 8 (38%)

46 (7–120) 4 (40%) 6 (60%)

0.48 0.44

10 (32%) 2 (6%)

7 (33%) 1 (5%)

3 (30%) 1 (10%)

1 1

2.0 (0.7–3.1) 5 (1–16) 1142 (500–25201) 6 (1–19) 6.3 (1.3–71.5) 1.48 (0.87–3.65) 31 (100%) 22 (71%) 24 (77%) 2 (1–5) 0 (0–1) 2 (0–3)

2.1 (1.0–6.0) 4 (1–8) 1032 (500–5465) 6 (1–11) 5.6 (1.3–22.7) 1.38 (0.87–2.86) 21 (100%) 14 (67%) 16 (76%) 2 (0–8) 0 (0–2) 1 (0–5)

1.6 (0.7–3.1) 11 (4–16) 2634 (834–25201) 7.5 (1–19) 19.5 (6.3–71.5) 1.65 (1.32–3.65) 10 (100%) 8 (80%) 8 (80%) 3 (0–9) 1 (0–6) 2 (1–3)

0.04 0.002 0.04 0.09 0.001 0.32 1 0.68 1 0.62 0.07 0.20

HBV = hepatitis B virus; HCV = hepatitis C virus; PBC primary biliary cirrhosis; PSC = primary sclerosing cholangitis; AIH = autoimmune hepatitis; CNI = calcineurin inhibitors; PI/PE = plasma infusion or exchange. Figures represent median (range): 1 Right liver graft versus others. 2 Time from the beginning of changes in laboratory data to meet the diagnostic criteria.

of FFP. In the present series, graft injury or malfunction because of small-for-size graft syndrome was less plausible for induction of TMA, because posttransplant portal flow (Table 1) and graft regeneration rates (Figure 5) were not significantly different between the cases with and with-

734

out TMA. Considering the higher incidence of TMA and the effectiveness of plasma infusion/exchange therapy for posttransplant TMA in LDLT recipients (Figure 4), however, the deficiency of some coagulation factors might also play a central role in the induction of TMA.


TMA After LDLT

Regarding the prognostic factors of TMA, late onset was reported to be a risk factor for poor outcome among patients who underwent plasma exchange at our institution (8). The present results also confirmed that the onset of TMA tended to be late in the deceased cases. In the present series, mortality rate in cases with early onset (≤30 days) and late onset (>30 days) TMA were 42.0% and 29.4% (p = 0.70), respectively. The reason for the higher mortality rate in cases with late onset is unclear, but the delay in diagnosis and in the initiation of treatment seemed to contribute to the comparatively poor prognosis between survivors and nonsurvivors (Table 3). Additionally, sensitization against HLA seemed to be a potent risk factor for both the induction of TMA and the poor outcome. Although a high prevalence of posttransplant TMA in ABO-incompatible LDLT recipients has been reported (4), the association between preformed antiHLA antibodies and TMA has not yet been demonstrated. We experienced six patients who were already sensitized against HLA before transplantation because of a history of multiple transfusions. Preoperative screening with lymphocytotoxic cross match tests revealed negative results in every case. In four of six patients, however, TMA developed postoperatively and all of them died during the first (n = 3) or repeated (n = 1) episode of TMA. Although the actual mechanism of induction of TMA in a patient possessing anti-HLA antibodies is unremarkable, sensitization against HLA is associated with immune-mediated platelet destruction (36) and all forms of humoral allograft rejection: hyperacute, acute and chronic (37–42). Therefore, patients with preformed anti-HLA antibody should be carefully managed after LDLT, even when the lymphocytotoxic cross match test is negative. Currently, there is no standard protocol for the treatment of posttransplant TMA. Considering that delayed treatment in deceased cases and the response rate of TMA to treatment (plasma infusion/exchange and CNI conversion) are estimated to be as high as 87%, however, early diagnosis of TMA and early treatment are likely critical in the management of posttransplant patients. In cases refractory to the standard treatments, a dose decrease or discontinuation of CNIs with other immunosuppressive agents, including mycofenolate mofetil or anti-CD 25 antibody (43), should be used as an alternative strategy to avoid CNIs. In conclusion, the pathophysiology of TMA seems to vary considerably and remains inconclusive. The development of TMA might result from the presence of invasive stimuli in association with some risk factors, including a deficiency of factors associated with coagulation cascades and/or immunosuppressive treatment with CNIs. Therefore, replenishment of coagulation factors with plasma infusion/exchange and conversion/discontinuation of CNIs might be a reasonable approach in the treatment of TMA. Because prevention of TMA is difficult, early diagnosis and American Journal of Transplantation 2012; 12: 728–736

early initiation of intensive therapies are crucial to improve the prognosis.

Disclosure The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References 1. Moake JL. Thrombotic microangiopathies. N Engl J Med 2002; 347: 589–600. 2. Humar A, Jessurun J, Sharp HL, Gruessner RW. Hemolytic uremic syndrome in small-bowel transplant recipients; the first two case reports. Transpl Int 1999; 12: 387–390. 3. Trimarchi HM, Truong LD, Brennan S, Gonzalez JM, Suki WN. FK506-associated thrombotic microangiopathy; report of two cases and review of the literature. Transplantation 1999; 67: 539– 544. 4. Miyata R, Shimazu M, Tanabe M, et al. Clinical characteristics of thrombotic microangiopathy following ABO incompatible living donor liver transplantation. Liver Transpl 2007; 13: 1455–1462. 5. Tamura S, Sugawara Y, Matsui Y, et al. Thrombotic microangiopathy in living-donor liver transplantation. Liver Transpl 2005; 80: 169–175. 6. Noris M, Remuzzi G. Thrombotic microangiopathy after kidney transplantation. Am J Transplant 2010; 10: 1517–1523. 7. Schwarz A, Haller H, Schmitt R, et al. Biopsy-diagnosed renal disease in patients after transplantation of other organs and tissues. Am J Transpl 2010; 10: 2017–2025. 8. Nishi H, Hanafusa N, Kondo Y, et al. Clinical outcome of thrombotic microangiopathy after living-donor liver transplantation treated with plasma exchange therapy. Clin J Am Soc Nephrol 2006; 1: 811–819. 9. Sugawara Y, Makuuchi M. Technical advances in living-related liver transplantation. J Hepatobiliary Pancreat Surg 1999; 6: 245–253. 10. Sugawara Y, Makuuchi M. Right lateral sector graft as a feasible option for partial liver transplantation. Liver Transpl 2004; 10: 1156– 1157. 11. Sugawara Y, Makuuchi M. Advances in adult living donor liver transplantation: A review based on reports from the 10th anniversary of the adult-to-adult living donor liver transplantation meeting in Tokyo. Liver Transpl 2004; 10: 715–720. 12. Sugawara Y, Makuuchi M, Akamatsu N, et al. Refinement of venous reconstruction using cryopreserved veins in right liver grafts. Liver Transpl 2004; 10:541–547. 13. Kokudo N, Sugawara Y, Imamura H, Sano K, Makuuchi M. Sling suspension of the liver in donor operation: A gradual taperepositioning technique. Transplantation 2003; 76: 803–807. 14. Kokudo N, Sugawara Y, Kaneko J, Imamura H, Sano K, Makuuchi M. Reconstruction of isolated caudate portal vein in left liver graft. Liver Transpl 2004; 10: 1163–1165. 15. Sugawara Y, Makuuchi M, Kaneko J, Ohkubo T, Imamura H, Kawarasaki H. Correlation between optimal tacrolimus doses and the graft weight in living donor liver transplantation. Clin Transplant 2002; 16: 102–106. 16. Kokudo N, Sugawara Y, Imamura H, Sano K, Makuuchi M. Tailoring the type of donor hepatectomy for adult living donor liver transplantation. Am J Transplantation 2005; 5: 1694–703.

735

Shindoh et al. 17. Zeigler ZR, Shadduck RK, Nemunaitis J, Andrews DF, Rosenfeld CS. Bone marrow transplant-associated thrombotic microangiopathy: A case series. Bone Marrow Transplant 1995; 15: 247–253. 18. Tamura S, Sugawara Y, Kishi Y, et al. Conversion to cyclosporine provides valuable rescue therapy for living related adult liver transplantation intolerant of tacrolimus: A single center experience at the University of Tokyo. Transplant Proc 2004; 36: 3242–3344. 19. Akamatsu N, Sugawara Y, Tamura S, Imamura H, Kokudo N, Makuuchi M. Regeneration and function of hemiliver graft: Right versus left. Surgery 2006; 139: 765–772. 20. Ponticelli C, Banfi G. Thrombotic microangiopathy after kidney transplantation. Transplant Int 2006; 19: 789–794. 21. Burke GW, Ciancio G, Cirocco R, et al. Microangiopathy in kidney and simultaneous pancreas/kidney recipients treated with tacrolimus: Evidence of endothelin and cytokine involvement. Transplantation 1999; 68: 1336–1342. 22. Noris M, Remuzzi G. Thrombotic microangiopathy after kidney transplantation. Am J Transplant 2010; 10: 1517–1523. 23. Tsai HM, Lian EC. Antibodies to von Willebrand factor-Cleaving protease in acute thrombotic thrombocytopenic purpura. N Engl J Med 1998; 339: 1585–1594. 24. Furlan M, Robles R, Galbusera M, et al. von Willebrand factorcleaving protease in thrombotic thrombocytopenic purpura and the hemolyticuremic syndrome. N Engl J Med 1998; 339: 1578– 1584. 25. Nakazawa Y, Hashikura Y, et al. Von Willebrand factor– Cleaving protease activity in thrombotic microangiopathy after living donor liver transplantation: A case report. Liver Transpl 2003; 9: 1328– 1333. 26. Kobayashi T, Wada H, Usui M, et al. Decreased ADAMTS13 levels in patients after living donor liver transplantation. Thromb Res 2009; 124: 541–545. 27. Furlan M, Lammle B. Haemolytic uraemic syndrome and thrombotic thrombocytopenic purpura: New insights into underlying biochemical mechanisms. Nephrol Dial Transplant 2000; 15: 1112– 1114. 28. Allford SL, Machin SJ. Current understanding of the pathophysiology of thrombotic thrombocytopenic purpura. J Clin Pathol 2000; 53: 497–501. 29. Bonser RS, Adu D, Franklin I, McMaster P. Cyclosporin-induced haemolytic uraemic syndrome in liver allograft recipient. Lancet 1984; 2: 1337.

736

30. Dzik WH, Georgi BA, Khettry U, Jenkins RL. Cyclosporineassociated thrombotic thrombocytopenic purpura following liver transplantation: Successful treatment with plasma exchange. Transplantation 1987; 44: 570–572. 31. McCauley J, Bronsther O, Fung J, Todo S, Starzl TE. Treatment of cyclosporin-induced haemolytic uraemic syndrome with FK506. Lancet 1989; 2(8678–8679): 1516. 32. Holman MJ, Gonwa TA, Cooper B, et al. FK506-associated thrombotic thrombocytopenic purpura. Transplantation 1993; 55: 205– 206. 33. Hochstetler LA, Flanigan MJ, Lager DJ. Transplant-associated thrombotic microangiopathy: The role of IgG administration as initial therapy. Am J Kidney Dis 1994; 23: 444–450. 34. Mach-Pascual S, Samii K, Beris P. Microangiopathic hemolytic anemia complicating FK506 (tacrolimus) therapy. AmJ Hematol 1996; 52: 310–312. 35. Singh N, Gayowski T, Marino IR. Hemolytic uremic syndrome in solid-organ transplant recipients. Transpl Int 1996; 9: 68–75. 36. Hod E, Schwartz J. Platelet transfusion refractoriness. Br J Haematol 2008; 142: 348–360. 37. Feucht HE, Schneeberger H, Hillebrand G, et al. Capillary deposition of C4d complement fragment and early renal graft loss. Kidney Int 1993; 43: 1333–1338. 38. McKenna RM, Takemoto SK, Terasaki PI. Anti-HLA antibodies after solid organ transplantation. Transplantation 2000; 69:319– 326. 39. Terasaki PI. Humoral theory of transplantation. Am J Transplant 2003; 3: 665–673. 40. Takemoto SK, Zeevi A, Feng S, et al. National conference to assess antibody-mediated rejection in solid organ transplantation. Am J Transplant 2004; 4: 1033–1041. 41. Ionescu DN, Girnita AL, Zeevi A, et al. C4d deposition in lung allografts is associated with circulating anti-HLA alloantibody. Transpl Immunol 2005; 15: 63–68. 42. Girnita AL, Webber SA, Zeevi A. Anti-HLA alloantibodies in pediatric solid organ transplantation. Pediatr Transplant 2006; 10: 146–153. 43. Nashan B, Moore R, Amlot P, Schmidt AG, Abeywickrama K, Soulillou JP. Randomised trial of basiliximab versus placebo for control of acute cellular rejection in renal allograft recipients. CHIB 201 International Study Group. Lancet 1997; 350: 1193–1198.
