Re-transplantation for Hepatic Artery Thrombosis: A National Perspective

  • Shu Kwun Lui
  • Catherine R. Garcia
  • Xiaonan Mei
  • Roberto Gedaly
Original Scientific Report
  • 48 Downloads

Abstract

Background

Hepatic artery thrombosis (HAT) is a major complication after liver transplantation that commonly requires re-transplantation.

Methods

We queried the UNOS dataset for all patients transplanted between 1995 and 2015 for HAT.

Results

We identified 623 patients who underwent re-transplantation for HAT with a mean age of 51.25 + 10.4 years. The mean BMI was 26.72 kg/m2, and mean MELD score was 19.62 + 9.09. There was a higher proportion of male patients, with higher prevalence of pre-transplant portal vein thrombosis (7.4 vs. 5.4%, p = 0.04), lower incidence of hepatitis C virus infection (29.5 vs. 35.8%, p = 0.002), and shorter waiting time (61 vs. 111 days, p = 0.001) in the HAT group compared to those re-transplanted for other indications. The perioperative 90-day mortality was lower in patients re-transplanted for HAT (16 vs. 20%, p = 0.02). Patients undergoing re-transplantation for HAT had 13% decreased graft survival and 13% increased long-term survival. After case–control matched analysis, graft survival and patient survival were significantly better in the HAT group. Late re-transplantation (>30 days) for HAT was linked to decreased graft and patient survival when compared to those undergoing early re-transplantation (within 30 days).

Conclusions

Improved outcomes were seen in patients undergoing re-transplantation for HAT compared to patients who underwent re-transplantation for other indications. Those re-transplanted late after HAT (>30 days) were associated with worse outcomes when compared to early re-transplantation.

Abbreviations

HAT

Hepatic artery thrombosis

LT

Liver transplantation

BMI

Body mass index

Introduction

Hepatic artery thrombosis (HAT) is a serious post-liver transplant (LT) complication associated with a 50% rate of re-transplantation. Studies indicate that HAT leads to increase risk of morbidity, graft loss, and mortality [2, 3]. HAT is classified into early (within 30 days post-LT) and late HAT (after 30 days post-LT) [4, 5, 6, 7]. The incidence of HAT varies from 2 to 9% with higher incidence in the pediatric population, particularly early childhood [8]. HAT is typically most frequently identified in the first week following LT; however, it may appear even years after the procedure [2, 9].

The mechanism of HAT is most likely multifactorial, involving both surgical and medical risk factors [7]. Some of the risk factors linked to increased rates of HAT include kinking of the hepatic artery, arterial anastomotic stricture, prolonged surgical time, and abnormal arterial anatomy of either the donor or recipient arteries requiring complex reconstructions [7]. Clotting abnormalities, tobacco use, immunological and infectious complications, and low donor/recipient age ratio have also been identified as critical determinants for the development of HAT [7].

Prompt identification of HAT may allow for urgent revascularization options, including surgical thrombectomy, and arterial reconstruction or thrombolysis [8]. Regardless of the treatment, re-transplantation is frequently needed, especially in patients with significant ischemic cholangiopathy [2]. Yet, the available data regarding re-transplantation for HAT are limited. The aim of this study is to analyze outcomes of patients re-transplanted for HAT compared to those re-transplanted for other indications.

Materials and methods

The UNOS database was queried to find all patients who underwent liver re-transplantation between 1995 and 2015 in the USA. Due to the lack of a specific diagnosis code for HAT, we used the “Primary Diagnosis/Text of Other, Specify at Listing” file for the following keywords “hepatic artery thrombosis” and “HAT” to carefully identify patients re-transplanted for this indication. Out of the 5902 liver re-transplantation procedures performed during the study period, 623 were performed for HAT. Recipient information including age, gender, serum creatinine, total bilirubin, INR, MELD, BMI, blood type, presence of pre-transplant portal vein thrombosis, length of hospital stay, waiting time, acute rejection at 6 months, early graft failure (within 7 days), and 30- and 90-day mortality was collected. Donor characteristics including living versus deceased donor, age, gender, and cold ischemia time were also collected. Pediatric patients were excluded from this analysis. We calculated the Donor Risk Index (DRI) on the first LT of patients who underwent re-transplantation for HAT and compared with all patients undergoing isolated LT (excluding living/partial graft transplants). We also compared our HAT cohort (first transplant) to all other patients undergoing re-transplantation for other indications by DRI. DRI is a validated tool commonly used to assess risk of graft loss after LT based on donor factors [10].

Statistical analysis

Univariate analyses were performed using Chi-squared or Fisher’s exact test as appropriate for categorical variables. Nonparametric continuous variables were reported with means (range) and compared using the Mann–Whitney U test. Survival was calculated from the time of transplantation using the Kaplan–Meier method with the log rank test to analyze differences in survival. Logistic regression analysis was performed to find independent predictors of early graft loss and perioperative mortality. A Cox regression analysis was performed to identify predictors of graft and patients survival after re-transplantation for HAT.

Factors found to be significant in the univariate analysis were included in the multivariate analysis (MVA). Missing data of continuous variables were assigned using SPSS automatic imputation method. The results of the Cox models were presented as adjusted hazard ratios (aHRs), and logistic regression results were presented as odds ratios. Each was accompanied by 95% CI. A case-match analysis 1:1, adjusting for age, gender, HCV infection, HCC diagnosis, and living vs deceased donors, was performed. In addition, we performed a subgroup analysis for those re-transplanted early or late, using 30 days as cutoff, to compare early graft loss, perioperative mortality, as well as graft and patient survival. All tests were two-sided with significance defined as p < 0.05. Analyses were conducted using SPSS software version 22.0.6 (SPSS Inc., Chicago, IL).

Results

Patient characteristics and demographics

During the study period, 623 patients (10.6%) were re-transplanted for HAT, while 5279 were re-transplanted for other causes. Eighty-one percent of the cases were performed after year 2000. The mean age in patients re-transplanted for HAT was 51.25 + 10.4 years, with a mean BMI of 26.72 kg/m2. MELD score at the time of transplantation for this group was 19.62 + 9.09 (lower than the MELD score of patients re-transplanted for other indications) (p < 0.001). Patients who underwent re-transplantation for HAT had a median waiting time of 60.77 + 165.67 days (Table 1).
Table 1

Demographics and transplant-specific characteristics for patients undergoing re-transplantation for HAT and non-HAT

 

Non-HAT (n = 5279)

HAT (n = 623)

p value

Recipient age (years)

49.01 ± 11.67

51.25 ± 10.40

<0.001*

Recipient gender (male)

3367 (63.8%)

423 (67.9%)

0.043*

Serum creatinine

1.87 ± 1.28

1.44 ± 0.94

<0.001*

Total bilirubin

16.39 ± 14.51

7.83 ± 9.80

<0.001*

INR

2.00 ± 1.72

1.56 ± 1.06

<0.001*

MELD

27.19 ± 9.92

19.62 ± 9.09

<0.001*

BMI

26.39 ± 5.65

26.72 ± 5.56

0.070

Blood type

A

1986 (37.6%)

257 (41.3%)

0.418

A1

7 (0.1%)

1 (0.2%)

A2

1 (0.0%)

0 (0.0%)

A2B

3 (0.1%)

0 (0.0%)

AB

277 (5.2%)

37 (5.9%)

B

701 (13.3%)

83 (13.3%)

O

2304 (43.6%)

245 (39.3%)

Pre-transplant portal vein thrombosis

287 (5.4%)

46 (7.4%)

0.046*

HCV

1891 (35.8%)

184 (29.5%)

0.002*

Length of hospitalization

26.46 ± 35.58

23.34 ± 35.93

0.015*

Waiting time

111.27 ± 298.96

60.77 ± 165.67

0.001*

Acute cellular rejection within 6 month

508 (9.6%)

52 (8.3%)

0.304

Early graft failure

390 (7.4%)

35 (5.6%)

0.106

Perioperative mortality (30 days)

721 (13.7%)

69 (11.1%)

0.073

Perioperative mortality (90 days)

1052 (19.9%)

100 (16.1%)

0.021*

Donor type (deceased)

5237 (99.2%)

622 (99.8%)

0.078

Donor age (years)

36.92 ± 16.29

36.14 ± 15.94

0.305

Donor gender (male)

3105 (58.8%)

370 (59.4%)

0.784

Cold ischemia time (hours)

7.50 ± 3.80

7.42 ± 3.35

0.846

Mean ± SD for continuous variables and N (%) for categorical variables

*Significant

We compared patients undergoing re-transplantation for HAT with patients re-transplanted for other indications. HAT recipients were significantly older (51 vs. 49 year, p < 0.001), with a higher preponderance of males (67.9 vs. 63.8%, p = 0.04). The incidence of hepatitis C viral infection (29.5 vs. 35.8%, p = 0.002) and the MELD score (19.6 vs. 27.2, p < 0.001) were significantly lower in the HAT cohort, while pre-transplant portal vein thrombosis was significantly higher (7.4 vs. 5.4%, p = 0.04). In the HAT group, length of hospital stay (23 vs. 26 days, p = 0.02), waiting time (61 vs. 111 days, p = 0.001), 90-day mortality (16 vs. 20%, p = 0.02) were significantly lower compared to the non-HAT group. BMI, blood type, cellular rejection, early graft failure, 30-day mortality, donor age, donor gender, or cold ischemia time were not significantly different between the two groups.

The most common indications for re-transplantation in non-HAT recipients were recurrent HCV (21.4%), graft failure unspecified (9.4%), alcoholic liver disease (4.8%), recurrent PSC (3.8%), recurrent autoimmune hepatitis (3%), and chronic rejection (3%).

Graft survival for the HAT group was 94.4, 88.9, 77.2, 61.2, and 46.9% at 7, 30 days, 1, 3, and 5 years, respectively. Patient survival was 95.0, 89.4, 78.0, 62.0, and 47.0% at 7, 30 days, 1, 3, and 5 years, respectively. Acute cellular rejection at 6 months was 8.3% for those re-transplanted for HAT.

The UNOS regions with the higher frequency of these patients were region two (13.2%) (Delaware, District of Columbia, Maryland, New Jersey, Pennsylvania, West Virginia), three (19.7%) (Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, and Puerto Rico), and nine (10.8%), (Colorado, Iowa, Kansas, Missouri, Nebraska, Wyoming).

We found that patients undergoing transplantation in the HAT group had slightly higher median DRI compared to those undergoing LT alone (1.88 vs. 1.71, respectively). The DRI on the first transplant of HAT vs patients re- transplanted for other indications was 1.88 and 1.84, respectively.

Multivariate analysis of prognostic factors associated with early graft loss (<7 days), perioperative mortality, graft survival, and patient survival after re-transplantation

A multivariate logistic regression analysis was performed to find factors associated with early graft failure in patients who underwent liver re-transplantation in the USA. Recipient age, male gender, MELD score, presence of pre-transplant portal vein thrombosis were associated with increased risk of early graft loss after liver re-transplantation (Table 2).
Table 2

Cox proportional hazards model for graft loss for patients undergoing re-transplantation in the USA

 

p value

HR

95% CI

HAT (Y vs. N)

0.039

0.879

0.778–0.994

Recipient age (10 years)

<0.001

1.113

1.077–1.150

MELD

<0.001

1.008

1.004–1.011

Recipient BMI

0.002

1.010

1.004–1.016

Donor age (10 years)

<0.001

1.128

1.104–1.152

Cold ischemic time (hours)

0.044

1.009

1.001–1.018

*The analysis only included significant variables in Table 1. Only significant variables were shown in this table

A risk-adjusted analysis was performed to find factors associated with perioperative mortality (30 day) after re-transplantation. Factors associated with greater risk of perioperative death were recipient age, male gender, MELD score, and presence of pre-transplant portal vein thrombosis (Table 3).
Table 3

Cox proportional hazards model for patient death for patients undergoing re-transplantation in the USA

 

p value

HR

95% CI

HAT (Y vs. N)

0.031

0.875

0.775–0.998

Recipient age (10 years)

<0.001

1.130

1.094–1.168

MELD

<0.001

1.007

1.004–1.011

Recipient BMI

0.004

1.009

1.003–1.016

Donor age (10 years)

<0.001

1.124

1.101–1.148

*The analysis only included significant variables in Table 1. Only significant variables were shown in this table

We performed a multivariate analysis to find predictors of overall graft and patient survival in patients undergoing re-transplantation in the USA to assess the impact of HAT. Interestingly, HAT was associated with better graft and patient survival than those transplanted for other indications. In fact, patients undergoing re-transplantation for HAT had 13% decreased risk of graft loss and increased patient survival.

Case-match control-adjusted analysis for graft and patient survival comparing patients re-transplanted for HAT and non-HAT

Case-match control analysis was performed adjusting for factors that could influence graft and patient survival. These factors included age, gender, HCV infection, HCC diagnosis, and living vs deceased donors. We found that patients undergoing re-transplantation for HAT had significantly better graft survival (p = 0.015, Fig. 1) and patient survival (p = 0.018, Fig. 2) than the non-HAT cohort.
Fig. 1

Kaplan–Meier curves for graft survival in the case-match analysis (1:1) comparing patients with and without HAT from the UNOS database (p value = 0.015)

Fig. 2

Kaplan–Meier curves for patient survival in the case-match analysis (1:1) comparing patients with and without HAT from the UNOS database (p value = 0.018)

Subgroup analysis of patients undergoing re-transplantation for early and late HAT (30 days or more)

We reviewed patient demographics and clinical characteristics between late and early re-transplantation and found no statistically significant difference between groups on most factors. Importantly, BMI, PVT, and cold ischemia time were significantly higher in the early HAT group, while waiting time was higher in the late HAT group.

We found that patients who underwent late re-transplantation for HAT had increased risk of early graft loss. Seven-day graft loss for early re-transplantation 2.36% versus 7.9%, for late re-transplantation (p = 0.005). Perioperative mortality was numerically higher in the late re-transplantation group (13%) compared to those undergoing early re-transplantation (8.6%, ps = ns). Graft survival was worse on the late re-transplantation group compared to the early group (p = 0.036). Our model found no statistically differences in patient survival (Fig. 3).
Fig. 3

Subgroup analysis of patients re-transplanted with HAT defined as late and early (p value = 0.036)

Discussion

HAT is a major complication after LT that is often the cause of re-transplantation in a significant number of patients [2]. It has been reported a higher incidence of this complication in pediatric patients ranging from 5 to 20% [11, 12]. Graft salvage in patients with early HAT may be possible with surgical or interventional revascularization [13, 14]. The success rate of surgical revascularization has been reported to be from 10 to 50% [2, 15, 16]. During the initial 30-day postoperative period after transplantation, the allograft may be more sensitive to ischemia and the development of significant biliary complications. The management of patients with late HAT depends on many factors, including liver function, collateralization with preservation of hepatic function, and the development of life-threatening complications such as hepatic abscesses and ischemic cholangiopathy with recurrent cholangitis and sepsis.

In this study, patients re-transplanted due to HAT have better outcomes than those re-transplanted for other indications. We observed higher rates of perioperative 30- and 90-day mortality in the non-HAT group. Patients undergoing re-transplantation for non-HAT indications had a 13% increased risk of graft loss and decreased patient survival compared to those transplanted for HAT. We found using data from the first LT in the HAT cohort that the DRI was similar to those re-transplanted for other indications. The DRI was slightly higher in the HAT group compared to all other isolated LT (data from first transplant). Previously published data report similar risk of graft loss to patients with these DRIs (1.88 and 1.71). Our case-match control analysis demonstrated higher graft and patient survival in recipients re-transplanted for HAT compared to the non-HAT cohort. These results are in contrast to what was found by Maggi et al. [17] in a multicenter study in Italy in which there was no difference in patient and graft survival in those re-transplanted for HAT or other indications. Interestingly, patients in the HAT group had higher incidence of PVT.

Our subgroup analysis of patients re-transplanted for early and late HAT demonstrated that graft survival and patient survival were more favorable in those re-transplanted when early HAT occurred. We hypothesize that patients re-transplanted in the late phase had a higher incidence of biliary-related complications, including hepatic abscesses, ischemic cholangiopathy, cholangitis, multidrug resistant infections, and chronic drain placements and exchanges.

This study has several limitations that should be disclosed. Variables such as onset of HAT, type of revascularization, decision-making to decide revascularization versus re-transplantation, the presence of biliary complications, number of episodes of cholangitis and sepsis, and type of treatment for the biliary complications are not included in the UNOS dataset. The strength of our study comes from the significant number of patients re-transplanted for HAT included in a single database but also from its clinically robust and uniform definitions of patient characteristics and events.

In summary, patients with failing revascularization with severe ischemia or those with late biliary complication secondary to ischemic cholangiopathy should be considered for re-transplantation. Our overall and risk-adjusted analysis demonstrated better overall outcomes in patients undergoing re-transplantation for HAT compared to those re-transplanted for other indications. Further studies are needed to identify modifiable factors associated with poor outcomes in patients undergoing re-transplantation for HAT.

Notes

Authors’ contribution

SKL, CRG, XM, and RG designed research/study. SKL, CRG, XM, and RG performed research/study. XM collected data. SKL, XM, and RG analyzed data. SKL, CRG, XM, and RG wrote the paper.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

References

  1. 1.
    Garcia C, Mei X, Berger J et al (2018) Re-transplantation for hepatic artery thrombosis: analysis of the UNOS dataset. Am J Transplant, abstract # p 119, presented at the State of the Art Winter Symposium, 2018, 8, p 75Google Scholar
  2. 2.
    Bekker J, Ploem S, de Jong KP (2009) Early hepatic artery thrombosis after liver transplantation: a systematic review of the incidence, outcome and risk factors. Am J Transplant 9:746–757CrossRefPubMedGoogle Scholar
  3. 3.
    Wu L, Zhang J, Guo Z et al (2011) Hepatic artery thrombosis after orthotopic liver transplant: a review of the same institute 5 years later. Exp Clin Transplant 9:191–196PubMedGoogle Scholar
  4. 4.
    Oh CK, Sawyer RG, Pelletier SJ, Pruett TL, Sanfey HA (2004) Independent predictors for primary non-function after liver transplantation. Yonsei Med J 45:1155–1161CrossRefPubMedGoogle Scholar
  5. 5.
    Silva MA, Jambulingam PS, Gunson BK et al (2006) Hepatic artery thrombosis following orthotopic liver transplantation: a 10-year experience from a single centre in the United Kingdom. Liver Transplant 12:146–151CrossRefGoogle Scholar
  6. 6.
    Kok T, Slooff MJ, Thijn CJ et al (1998) Routine Doppler ultrasound for the detection of clinically unsuspected vascular complications in the early postoperative phase after orthotopic liver transplantation. Transpl Int 11:272–276CrossRefPubMedGoogle Scholar
  7. 7.
    Pastacaldi S, Teixeira R, Montalto P, Rolles K, Burroughs AK (2001) Hepatic artery thrombosis after orthotopic liver transplantation: a review of nonsurgical causes. Liver Transplant 7:75–81CrossRefGoogle Scholar
  8. 8.
    Jurim O, Csete M, Gelabert HA et al (1995) Reduced-size grafts—the solution for hepatic artery thrombosis after pediatric liver transplantation? J Pediatr Surg 30:53–55CrossRefPubMedGoogle Scholar
  9. 9.
    Sidhu PS, Ryan SM, Karani JB (2009) Collateral transformation of the hepatic artery after liver transplantation: significance of the tardus-parvus waveform. AJR Am J Roentgenol 192:W264 (author reply W5) CrossRefPubMedGoogle Scholar
  10. 10.
    Feng S, Goodrich NP, Bragg-Gresham JL et al (2006) Characteristics associated with liver graft failure: the concept of a donor risk index. Am J Transplant 6:783–790CrossRefPubMedGoogle Scholar
  11. 11.
    Sanchez SE, Javid PJ, Lao OB et al (2012) Hepatic artery thrombosis and liver malignancy in pediatric liver transplantation. J Pediatr Surg 47:1255–1260CrossRefPubMedGoogle Scholar
  12. 12.
    Rela M, Muiesan P, Bhatnagar V et al (1996) Hepatic artery thrombosis after liver transplantation in children under 5 years of age. Transplantation 61:1355–1357CrossRefPubMedGoogle Scholar
  13. 13.
    Singhal A, Mukherjee I, Stokes K, Wright HI, Sebastian A, Kohli V (2010) Continuous intraarterial thrombolysis for early hepatic artery thrombosis following liver transplantation: case report. Vasc Endovasc Surg 44:134–138CrossRefGoogle Scholar
  14. 14.
    Levy EB, Zhang H, Lindisch D, Wood BJ, Cleary K (2007) Electromagnetic tracking-guided percutaneous intrahepatic portosystemic shunt creation in a swine model. J Vasc Interv Radiol JVIR 18:303–307CrossRefPubMedGoogle Scholar
  15. 15.
    Duffy JP, Hong JC, Farmer DG et al (2009) Vascular complications of orthotopic liver transplantation: experience in more than 4200 patients. J Am Coll Surg 208:896–903 (discussion 5) CrossRefPubMedGoogle Scholar
  16. 16.
    Scarinci A, Sainz-Barriga M, Berrevoet F et al (2010) Early arterial revascularization after hepatic artery thrombosis may avoid graft loss and improve outcomes in adult liver transplantation. Transplant Proc 42:4403–4408CrossRefPubMedGoogle Scholar
  17. 17.
    Maggi U, Andorno E, Rossi G et al (2012) Liver retransplantation in adults: the largest multicenter Italian study. PLoS ONE 7:e46643CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Société Internationale de Chirurgie 2018

Authors and Affiliations

  • Shu Kwun Lui
    • 1
  • Catherine R. Garcia
    • 1
  • Xiaonan Mei
    • 1
  • Roberto Gedaly
    • 1
  1. 1.Department of Surgery - Transplant Division, College of MedicineUniversity of KentuckyLexingtonUSA

Personalised recommendations