Methods
This study was conducted under Mayo Clinic IRB 19-005899. It is a
single-center, retrospective review of patients who underwent orthotopic
liver transplantation between 01/01/2015 to 05/31/2018 at Mayo Clinic,
Florida. We included patients who were part of the “fast track”
protocol at our institution as they would receive neuromuscular blockade
antagonism and be extubated in the operating room at the conclusion of
the case. “Fast track” liver transplantation is defined as extubating
in the operating room, bypassing the ICU, and going directly from the
post-anesthesia care unit to the surgical ward.6 We
collected the patient’s baseline characteristics, intraoperative
variables, and postoperative variables using our institution’s medical
record system (Epic, Verona, WI). Baseline characteristics included age,
sex, weight, body mass index (BMI), physiologic model for end-stage
liver disease (MELD) score on the day of surgery, and liver graft types,
including donation after circulatory death (DCD) or donation after brain
death (DBD). Both preoperative and immediate postoperative laboratory
variables were collected, which included international normalized ratio
(INR), aPTT, and creatinine. We also collected blood products transfused
in the postoperative day (POD) 0 to day 1, POD 0-1 respiratory
complication, POD 0 PONV status, POD 0-1 need for re-operation for
bleeding, POD 0-7 organ rejection, and the length of stay. PONV status
is defined as if the patient needed rescue antiemetics. POD 0-7 organ
rejection was confirmed by biopsy in the first-week post-transplant.
Induction of anesthesia was at the anesthesiologist’s discretion and
performed using 1 to 2 ug/kg fentanyl, 0.6 mg/kg rocuronium or 1 mg/kg
succinylcholine, and 2 to 2.5 mg/kg propofol. Intraoperatively patient
received rocuronium based on subjective evaluation with a peripheral
nerve stimulator (PNS); maintenance of anesthetic with sevoflurane. All
patients were kept normothermic with forced-air warming devices, warming
mattresses, and a fluid warmer. Intraoperative coagulation labs included
platelet count (Plts), PT/INR, aPTT, fibrinogen, and thromboelastography
(TEG) were collected post-induction, 10 minutes before and after the
anhepatic phase, 5 minutes before and after the reperfusion phase.
Additional labs and TEG were collected at the discretion of the
attending anesthesiologist. Patients received various blood products
based on the lab results with a goal of the following: Hgb 9 to 10
mg/dl, INR 1.5 to 2.0, fibrinogen above 180, Plts around 100, 000.
The anesthesiologist and the transplant surgeon made the decision to
“fast-track” based on their clinical judgment rather than a defined
clinical protocol. At the time of emergence, all patients received
either sugammadex (2-4 mg/kg) or neostigmine (0.03-0.07 mg/kg) based on
values obtained from the PNS. All patients were extubated after the NMBA
antagonist was administered, and the anesthesia team felt neuromuscular
function had been restored based on values from a PNS.
The primary outcomes were POD 0-1 bleeding events (re-operation for
bleeding, packed red blood cell (PRBC) transfusion), and POD 0 values of
aPTT and INR. Secondary outcomes were POD 0-1 respiratory complication,
POD 0 PONV, POD 0-7 organ rejection, and length of stay.
Continuous variables were summarized as mean (standard deviation) or
median (range), while categorical variables were reported as frequency
(percentage). Continuous baseline and preoperative variables were
compared between the standard and sugammadex groups using the Wilcoxon
rank-sum test, and categorical variables were compared using the
Chi-squared test. All tests were two-sided, with an alpha level set at
0.05 for statistical significance. The difference and corresponding 95%
confidence intervals (CI) in the continuous postoperative outcome
variables between the two groups were estimated using linear regression
models, and odds ratios and 95% CI in dichotomous outcomes were
estimated using logistic regression models.