Technical Risk Factors for Hepatic Artery Thrombosis After Orthotopic Liver Transplantation: The Hungarian Experience B. Nemes, G. Gaman, F. Gelley, A. Doros, G. Zadori, D. Gorog, I. Fehervari, and L. Kobori ABSTRACT Hepatic artery thrombosis (HAT) signiﬁcantly affects graft loss and mortality after orthotopic liver transplantation (OLT). The aim of this study was to analyze the risk factors of HAT in our program, with special regard to the personal-technical factor. We retrospectively analyzed the data of 500 adult liver transplant recipients between 1995 and 2011. Operations were performed by a certain group of surgeons, with standardized technique. The incidence rate of HAT decreased since 1995 from 12% to 7.8%. In accordance with the literature, HAT associated with acute rejection, polytransfusion, and the duration of the hepatectomy, arterial variations/reconstructions, tiny arteries, and furthermore, the timing of the anastomosis in Hungary. However we did not ﬁnd an association with other parameters, like cytomegalovirus infection, and hepatocellular carcinoma as indication. We created a “difﬁculty index” that consists of the technical parameters. The difﬁculty index together with surgical experience (number of OLTs performed) had an outstanding association with HAT. In conclusion, the incidence and risk factors for HAT are similar to the results published by others. However, personal factors, such as experience, timing, given anatomy, and tiredness, might also play a signiﬁcant role in the occurrence of HAT.
HE INCIDENCE OF hepatic artery thrombosis (HAT) after orthotopic liver transplantation (OLT) is 5% to 15% in the literature.1e7 HAT occurs mostly in the immediate postoperative period. The known risk factors for HAT are tiny arteries, anatomical variations in both donors and recipients, a need for arterial reconstruction at the bench table, prolonged reperfusion time, mismatch in the cytomegalovirus (CMV) and ABO-Rh serostate of the donor and recipient, acute rejection, polytransfusion, pediatric recipient, Leiden mutation, and liver cancer.1e7 However, HAT is mainly a technical complication that needs urgent reintervention. In early onset cases, embolectomy, reanastomosis, or a jumping graft (aorta conduit) will resolve the problem, but in majority of cases, retransplantation (reOLT) remains the only option. Despite all efforts, a small percent of the patients will die. Therefore, better knowledge of risk factors is essential. Surgical experience reduces risk while the anatomical disorders, like aberrant arterial anatomy requiring complex reconstruction or tiny vessels, increase the risk of HAT. In this study we also focused on the personal factor, like timing of the anastomosis (overnight operation) or after a difﬁcult, long hepatectomy. These four factors were recognized as difﬁculty index for arterial anastomosis.
MATERIALS AND METHODS Data of 500 adult liver transplant recipients from 1995 to 2011 were retrospectively analyzed. Patients with HAT (HAT group) were compared with those having no arterial complication (NOR group). The following factors were analyzed: donor and recipient demographics data, operative parameters (duration of hepatectomy, cold ischemic time, warm ischemic time, transfusion need), technical parameters (variations and caliber of the arteries), personal parameters (experience of the surgeon, timing of the operation), and peri- and postoperative data, and complications, as well as reinterventions. A “difﬁculty index” was deﬁned. It consisted of tiny arteries (either donor/recipient side 3.5 mm in diameter), arterial variation required, prolonged hepatectomy (more than 180 minutes), and overnight timing of the arterial anastomosis (between 1:00 and 5:00 AM). One score was given after each positive item. The arterial anastomosis was called difﬁcult if at least 2 scores were given. Statistical analysis was performed using SPSS 15.0 software. P < .05 was considered to be signiﬁcant.
From the Department of Transplantation and Surgery, Budapest, Hungary. Address reprint requests to Balazs Nemes, Department of Transplantation and Surgery, Semmelweis University, Baross u. 23e25, H-1082 Budapest, Hungary. E-mail: [email protected]
ª 2013 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/13/$esee front matter http://dx.doi.org/10.1016/j.transproceed.2013.10.007
Transplantation Proceedings, 45, 3691e3694 (2013)
NEMES, GAMAN, GELLEY ET AL
Fig 1. Receiver operating characteristic analysis of hepatic artery thrombosis (HAT) in relation to individual factors: (A) arterial variations (area under the curve [AUC] ¼ 0.49, P ¼ .87); (B) tiny arteries (AUC ¼ 0.56; P ¼ .35); (C) prolonged/difﬁcult hepatectomy (AUC ¼ 0.57, P ¼ .53); (D) overnight arterial anastomosis (AUC ¼ 0.53, P ¼ .61).
Data of HAT-group compared to normal are indicated on Tables 1 and 2. HAT occurred in 7.8% of the cases altogether: 12.1% from 1995 to 2000, then 8.3% from 2011 to 2006, and 5.4% since 2007. The 1-, 3-, 5-, and 10-year
cumulative patient survivals were 80%, 75%, 70%, and 61% for NOR group and 60%, 50%, 43%, and 14% for HAT group respectively (P ¼ .008). Cumulative graft survivals were 78%, 72%, 70%, and 52% for normal and 47%, 35%, 29%, and 0% for HAT group (P ¼ .001). There were no
TECHNICAL RISK FACTORS FOR HAT
3693 Table 2. Operation and Technical Data for HAT-Group and Normal-Group HAT group
Operation n ¼ 38 n ¼ 462 Hepatectomy (min) 210 101 183 62 NS Arterial anastomosis (min) 19 11 15 12 NS WIT (min) 58 17 55 16 NS Operation total (min) 478 146 450 107 NS CIT (min) 470 108 490 137 NS Aorta-conduit used 21% 11.6% .08 Transfusion (units) OR 9.6 6.5 8 6.9 .09 Transfusion (units) ICU 15 18 8 11 .08 Colloid (mL) 1500 1200 900 1000 .006 Diuresis in OR (mL) 1200 900 900 650 .05 Technical n ¼ 28 n ¼ 214 Hepatectomy >180 min 57% 48% NS Arterial anatomy variation 18% 15% .1 Tiny artery (caliber 3.0 mm) 50% 34% .1 Timing: 1.00e5.00 AM 30% 26% NS Difﬁculty index 1.37 1.1 0.76 0.6 .001 Score 0 21% 35% Score 1 321% 53% Score 2 36% 11% Score 3 11% 1 (0.5%) .0009
Fig 2. Receiver operating characteristic analysis for the hepatic artery thrombosis (HAT) and the difﬁculty index (positive predictive value ¼ 0.393; negative predictive value ¼ 0.926; area under the curve ¼ 0.68; P ¼ .002, cutoff 2).
differences in the donor parameters except for the “zero” biopsy: in the HAT group moderate (10%) fatty changes was observed in more than 40% of liver grafts, compared to 25% in the normal group (not signiﬁcant). There were no Table 1. Donor and Recipient Data for HAT-Group and NormalGroup HAT group n ¼ 38
Donor Age 40 12 BMI 22.8 3 ASAT (U/L) 52 73 Female donor 19 (50%) Marginal donora 10% Graft fatty changes >10%a 38% Recipient Age 38 14 Gender male (288)/female (222) 50% BMI 25 1.1 Child-P score 9.2 2.2 MELD score 14.7 6 Indication Parenchymal 70% Hepatitis C 50% Alcohol/toxic 5% Acute fulminant 12%
Normal group n ¼ 462
37 13 23.4 3 46 40 222 (48%) 17% 26%
NS NS NS NS NS .1
44 13 50% 25 1.3 8.9 1.9 14.2 6
.02 NS NS NS NS
70% 37% 17% 4%
NS .1 .04 .08
ASAT, aspartate aminotransferase; BMI, body mass index; MELD, Model for End-stage Liver Disease; HAT, hepatic artery thrombosis; NS, not signiﬁcant. a Zero biopsies taken in 252 cases (26 HAT and 252 normal group).
WIT, warm ischemic time; CIT, cold ischemic time; OR, operation room; ICU, intensive care unit; NS, not signiﬁcant. a Data were available.
differences in the recipient age, body mass index, Model for End-stage Liver Disease, and Child score, as well as liver related complications prior to OLT. Indication for the OLT also made no difference, nor did hepatocellular carcinoma. Surprisingly there were no association with CMV serostate and ABO minor incompatibility and HAT. Total blood transfusion requirement was larger in the HAT group compared to normal: 9.5 (6.5) U vs 8 (6.9) U intraoperatively (not signiﬁcant), and 15 (18) U vs 8 (11) U postoperatively (not signiﬁcant). Regarding baseline immunosuppression, HAT occurred more than among cyclosporine treated patients compared to patients on FK regimen (12.3% vs 6.1%; P ¼ .022). Acute rejection was more frequent among HAT group compared to normal (48% vs 31%; P ¼ .34). Postoperative bleeding, biliary complications, and sepsis were more frequent in the HAT Table 3. Data of Surgeons in Order of Exposure (No. of OLTs Performed) Operator
A C B D E F J G H
9.6% 5.9% 3.1% 8% 15% 15% 5.6% 1 (20%) 1 (33%)
1.03 0.79 0.68 1.15 1.72 0.64 0.96 NA NA
0.7 0.6 0.6 0.7 0.9 0.7 0.7
No. of OLTs
136 101 96 87 33 26 18 5 3
HAT, hepatic artery thrombosis; OLT, orthotopic liver transplantation; NA, not available.
group compared to normal (12% vs 6%; P ¼ .02; 42% vs 25%; P ¼ .018 and 31% vs 10%; P ¼ .001). Especially early biliary complications were more frequent in the HAT group (35% vs 15%; P ¼ .001). Postoperative reintervention was performed in 90% of HAT cases, compared to 50% in normal group (P ¼ .001). Of these interventions, 65% were surgical, the rest were radiologic. Technical and personal data did not prove to be independent risk factors one by one, this is presented on receiver operating characteristic (ROC) analysis (Fig 1). Tiny arteries had a positive predictive value (PPD) of 0.289; negative predictive value (NPD) of 0.825 (P ¼ .17); and area under the curve (AUC) of 0.579; presence of arterial variation had PPD of 0.148; NPD of 0.869 (P ¼ .8) and AUC of 0.586. Prolonged/difﬁcult hepatectomy had PPD of 0.14; NPD of 0.891 (P ¼ .47) ns; AUC of 0.536; overnight arterial anastomosis had PPD of 0.138; NPD of 0.89; (P ¼ .5) and AUC of 0.529. However the difﬁculty index as a composite of the 4 factors had signiﬁcant correlation with the HAT. It had a PPD of 0.393 NPD of 0.926; sensitivity of 0.46, and speciﬁcity of 0.87 with AUC of 0.679 (P ¼ .002). The cutoff was a score 2 (Fig. 2). With a difﬁculty index score 2, the PPD, NPD, and P value are .75; .92, and .0001 respectively. Data clearly showed up that the number of OLTs performed by an individual surgeon had a clear negative correlation to HAT, if this number exceeded at least 35 OLTs (Table 3).
NEMES, GAMAN, GELLEY ET AL
living related liver transplantation (LRLT), to reconstruct the arteries. The anastomoses were performed under an operating microscope (up to 24 magniﬁcation). One of 10 cases experienced HAT.10 Our study conﬁrms that the incidence of HAT can be reduced by practice. It is mainly inﬂuenced by the experience of the surgeon and the operative situation and technique. The more experienced the surgeon, the less the HAT rate is, and in contrast, the more difﬁcult the situation the higher is the HAT incidence. Difﬁculty in liver transplantation is a composite of anatomical (caliber, variations) and personal factors (overnight operation following a long, complicated hepatectomy). In conclusion, our result clearly proved that none of these factors will inﬂuence the outcome on its own; however, the presence of more than 2 factors might warn the implant team to take special care on the situation. If the difﬁculty index was administered separately to the surgeons, it turned out that despite great experience, a difﬁcult OLT can increase the risk of HAT even in experts’ hands. It also proved that under a limited exposure (below 35 OLTs/person) the incidence of HAT is independent of the difﬁculty index. The difﬁculty index might be used for the assessment of risk for HAT prior to the operation; therefore, it can also be used to adapt anticoagulant therapy.
HAT remains an important cause of graft loss after liver transplantation.1e7 HAT can be caused by technical, hemodynamic, or immunologic factors. Donor age and prothrombotic gene mutation might also be factors for HAT.3,4 Bench reconstruction of anatomical variants of the hepatic artery is considered to be a major risk related to HAT, despite of some outstanding experience.6 The incidence and risk factors of HAT in Hungary are similar to other reported results.1e7 Also the consequences and management of HAT is similar to the literature.8 Our results improved since 1995,9 a consequence of several factors, such as the introduction of intraoperative Doppler according to international standards,1 the more active use of aortal conduits in doubtful cases, the reduction in blood transfusion, and more precise suture technique. Our approach to hepatic artery reconstruction, similar to others reports,3 systematically preferred the Carrel patch/celiac trunk, and/or end-to-end anastomosis between short cut hepatic artery ends to prevent kinking, while the reconstructions were made of donor accessory right hepatic artery on the donor gastroduodenal or lineal artery. Anastomosis was preferably made by running 6-0, but in cases of tiny arteries, knotted and 7e0 suture threads. Mangal et al published a similar attempt to understand the practical difﬁculties encountered while performing hepatic artery anastomosis by microsurgical technique in living donor liver transplantation. They involved plastic surgeon, in
1. Warner P, Fusai G, Glantzounis GK, et al. Risk factors associated with early hepatic artery thrombosis after orthotopic liver transplantationdunivariable and multivariable analysis. Transpl Int. 2011;24(4):401e408. 2. Duffy JP, Hong JC, Farmer DG, et al. Vascular complications of orthotopic liver transplantation: experience in more than 4,200 patients. J Am Coll Surg. 2009;208:896e903. 3. Cescon M, Zanello M, Grazi GL, et al. Impact of very advanced donor age on hepatic artery thrombosis after liver transplantation. Transplantation. 2011;92(4):439e445. 4. Pereboom IT, Adelmeijer J, van der Steege G, et al. Prothrombotic gene polymorphisms: possible contributors to hepatic artery thrombosis after orthotopic liver transplantation. Transplantation. 2011;92(5):587e593. 5. Pareja E, Cortes M, Navarro R, et al. Vascular complications after orthotopic liver transplantation: hepatic artery thrombosis. Transplant Proc. 2010;42(8):2970e2972. 6. Kornasiewicz O, Dudek K, Lewandowski Z, et al. Low incidence of hepatic artery thrombosis after hepatic artery reconstruction during liver transplantation. Transplant Proc. 2009;41(8): 3138e3140. 7. Nemes B, Zadori G, Hartmann E, et al. [Biliary complications following orthotopic liver transplantation. The Hungarian experience]. Orv Hetil. 2008;149(21):963e973. 8. Wojcicki M, Milkiewicz P, Silva M. Biliary tract complications after liver transplantation: a review. Dig Surg. 2008;25(4):245e257. 9. Nemes B, Polak W, Ther G, et al. Analysis of differences in outcome of two European liver transplant centers. Transpl Int. 2006;19(5):372e380. 10. Mangal M, Gambhir S, Gupta A, Shah A. Role of plastic surgeons in hepatic artery anastomosis in living donor liver transplantation: our experience of 10 cases. J Reconstr Microsurg. 2012;28(6):359e362.