Patients
The local ethics committee (University Hospital Aachen, EK 291/13) approved the analysis and waived the requirement of informed consent. Our analyses was conducted with adherence to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline [20].
From January 2010 to June 2015 we performed 290 liver transplantations in our tertiary care university hospital. Records of patients who received a liver transplantation during the observation period from 2010 to 2014 were reviewed for the intraoperative treatment of aprotinin. Using organ quality, donor age and CIT, we identified a propensity score matched control group with the nearest neighbor method (SPSS 24.0, MatchIt package for R, IBM Corporation, Armonk, USA). Matching variables were selected a priori, as these were the criteria used to apply aprotinin at our center at that time.
Donor data
The covering letter from Eurotransplant provided the donor data: age, sex, body mass index, cold ischemic time (CIT) as well as sodium concentration, alanine transaminase (ALT), aspartate transaminase (AST) and bilirubin (Bili).
Donor organ assessment
The transplanting surgeon assessed the liver graft macroscopically by inspection and palpations as described before [21], assessing liver texture, yellowness, absence of scratch marks and round edges[22] . The organs were classified to either good, acceptable or poor quality following the Eurotransplant criteria. Macro- and microvesicular fat content were determined by histology and described as affected hepatocytes in percentage [7] including microsteatosis (MIS; the cytoplasm of the hepatocyte contains multiple tiny lipid vesicles without nuclear dislocation) or macrovesicular steatosis (the cytoplasm of the hepatocyte contains a univacuole lipid vesicle with nuclear displacement) [23].
Liver transplantation management
Liver transplantation was performed using an extracorporeal venovenous/portalvenous bypass. Bypass, surgical and anesthesiologic management as well as peri- and postoperative immune suppression regimen have been described earlier in detail [7, 21]. The patients received a maximum of one liter of balanced electrolyte solution. Adjacent volume replacement was hold up by transfusion of fresh frozen plasma (FFP) in order to anticipate coagulation disorders. Transfusion triggers for red blood cell units (RBC) were dependent on the patient’s comorbidities and transfusions were conducted at the discretion of the providing anesthesiologist. Our standard operation procedure (SOP) scheduled a thrombelastometry (TEM, Rotem®) after induction of anesthesia as well as 15-30 min and 45-60 min after reperfusion for early correction of coagulation disorders [24].
Aprotinin application
After allocation of the organ, the treating anesthesiologist and transplant surgeon decided jointly considering i) the visual assessment of the liver, ii) to the cold ischemic time (CIT) and iii) the donor age, whether the patient should receive aprotinin in order to attenuate PRS and early allograft dysfunction. Aprotinin infusion was started immediately after the surgical incision with a testing dose of 1ml (equivalent to 10 000 IE) to rule out any allergic reaction. After that, aprotinin was infused at a rate of 2x106 IE/h, a rate of 4x106 IE/h during the an-hepatic phase and reduced to 2x106IE/h until the end of surgery.
Recipient data
Recipient data were abstracted from the patient’s medical chart: recipient age, diagnosis leading to transplantation and the laboratory model of end stage liver disease score (labMeld: 10x (3.8*ln(bilirubin[mg/dl])+11,2*loge(INR)+9,6*ln(creatinine level [mg/dl])+6,4*(etiology: 0 if cholestatic or alcoholic, 1 otherwise) [3] were recorded at the evaluation procedure before patients were enlisted for transplantation. Clinical chemistry data (creatinine, AST, ALT, Bili, GGT, GLDH) were extracted from the electronical chart after admission closest to the beginning of the surgery (preoperatively), at ICU admission immediately after surgery (postoperatively) and on day 1, 3, 7 and 14. The number of intraoperatively transfused units of red blood cell units (RBC; units), fresh frozen plasma (FFP), platelets, fibrinogen and 4 factor prothrombin complex concentrate (4F-PCC) were extracted from the paper-based anesthesia protocol. Bilirubin, INR, AST/ALT, acute rejection (clinical diagnosis), surgical revisions, re-transplantation, sepsis, need for renal replacement therapy (RRT), intensive care unit (ICU) length of stay (LOS). Early allograft dysfunction is defined as bilirubin 10mg/dl on postoperative day (POD)7 and/or INR 1,6 on POD 7 and/or AST or ALT >2000 IU/L within in the first 7 days, were abstracted from the patient’s chart after the transplantation.
Postreperfusion syndrome [7]
Postreperfusion syndrome was defined as occurrence of one of the following criteria: (1) decrease in mean arterial pressure (MAP) of at least 30% at time of reperfusion, (2) administration of an intravenous bolus of norepinephrine >2 μg kg body weight (BW)-1, (3) increase of continuous norepinephrine (NE) infusion of ≥0.1 μg kg BW− 1 within 5 to 30 min after reperfusion, or (4) initiation of continuous vasopressin infusion after reperfusion. According to our department’s SOP, PRS was treated as follows: (i) 0,5 mg atropine before reperfusion if heart rate < 80, (ii) NE boli and NE infusion to maintain MAP, (iii) epinephrine boli and infusion in case of significant bradycardia with hypotension and decrease of SVO2 during reperfusion, (iv) infusion of vasopressin if high doses of NA are necessary or NA therapy ineffective.
Statistics
For a detailed description of the propensity score matching please refer to the the patients section above. Differences between groups were analyzed using the t-test for continuous, Chi-Square test for categorical and variance analysis for repeated measurements for continuous variables over time. Kaplan-Meyers curves were generated to display the effect of trasylol on patient and graft survival. A logistic regression analysis was used to analyse the effect of trasylol, MELD, organ quality, donor BMI and donor AST on EAD. Parameter were included in multivariate (stepwise forward) approach if univariate analysis was significant (SPSS 24.0). Results were displayed as mean and standard deviation or absolute and relative number of cases. Figures were created using Prism 6.0 (GraphPad Software, San Diego, USA). A two-sided p-value ≤0.05 was considered statistically significant.