Continuous versus intermittent infusion of human antithrombin III concentrate in the immediate postoperative period after liver transplantation

Abstract Antithrombin‐III (AT‐III) concentrates have been used in the immediate postoperative period after liver transplantation to prevent critical thrombosis. We aimed to investigate a more appropriate method for AT‐III concentrate administration to maintain plasma AT‐III activity level within the target range. In this randomized controlled trial, 130 adult patients undergoing living‐donor liver transplantation were randomized to either the intermittent group or continuous group. In the intermittent group, 500 international units (IU) of AT‐III concentrate were administered after liver transplantation and repeated every 6 h for 72 h. In the continuous group, 3000 IU of AT‐III were continuously infused for 71 h after a loading dose of 2000 IU over 1 h. Plasma AT‐III activity level was measured at 12, 24, 48, 72, and 84 h from the first AT‐III administration. The primary outcome was the target (80%–120%) attainment rate at 72 h. Target attainment rates at other timepoints and associated complications were collected as secondary outcomes. A total of 107 patients were included in the analysis. The target attainment rates at 72 h post‐dose were 30% and 62% in the intermittent group and continuous group, respectively (p = 0.003). Compared to the intermittent group, patients in the continuous group reached the target level more rapidly (12 vs. 24 h, median time, p < 0.001) and were more likely to remain in the target range until 84 h. For maintaining the target plasma AT‐III activity level after living‐donor liver transplantation, continuous infusion of AT‐III seemed to be more appropriate compared to the conventional intermittent infusion regimen.

trate were administered after liver transplantation and repeated every 6 h for 72 h.
In the continuous group, 3000 IU of AT-III were continuously infused for 71 h after a loading dose of 2000 IU over 1 h. Plasma AT-III activity level was measured at 12, 24, 48, 72, and 84 h from the first AT-III administration. The primary outcome was the target (80%-120%) attainment rate at 72 h. Target attainment rates at other timepoints and associated complications were collected as secondary outcomes. A total of 107 patients were included in the analysis. The target attainment rates at 72 h post-dose were 30% and 62% in the intermittent group and continuous group, respectively (p = 0.003). Compared to the intermittent group, patients in the continuous group reached the target level more rapidly (12 vs. 24 h, median time, p < 0.001) and were more likely to remain in the target range until 84 h. For maintaining the target plasma AT-III activity level after livingdonor liver transplantation, continuous infusion of AT-III seemed to be more appropriate compared to the conventional intermittent infusion regimen.

INTRODUCTION
Liver transplantation is a life-saving treatment option for patients with end-stage liver disease. The most recent 5year survival rate after liver transplantation in the United States reached 81.2% with advances in perioperative care including immunosuppressive agents, antibiotics, tissue preservation, and anticoagulation. 1 During the immediate postoperative period, antithrombin-III (AT-III) has been advocated as an adjunct therapy to prevent critical thrombosis and improve surgical outcomes. [2][3][4] AT-III is a glycoprotein produced in the liver that inhibits thrombin and other factors in the coagulation cascade such as IXa, Xa, XIa, and XIIa, and plasmin. 5,6 The impact of surgery on the impaired coagulation function of a liver transplant recipient may cause further dysfunction of the coagulation system. After liver transplantation, plasma AT-III activity level may decrease to 80% of the preoperative baseline level after liver transplantation and gradually recover with the resumed function of the liver graft over a few days. [7][8][9] In the immediate postoperative phase, there may be dysfunction of the anticoagulation mechanism with reduced plasma AT-III activity level, increasing the possibility of thrombosis of anastomosed major vessels such as the hepatic artery and portal vein. The incidences of post-transplantation hepatic artery and portal vein thrombosis have been reported to be around 4%-15% and 2%-7%, respectively, which might be fatal to the prognosis. 10,11 When using AT-III, titration of plasma AT-III activity level is essential as an excessive dose of AT-III concentrate may promote postoperative bleeding and an insufficient dose may not be effective. Consequently, we performed a preceding study that retrospectively analyzed the pharmacokinetics of AT-III in liver transplantation recipients and showed simulations of various dosing scenarios according to the developed model. According to the simulations, the plasma AT-III activity level was expected to be maintained more stably within the target range (80%-120%) with a smaller total dose by the continuous infusion method with a loading dose than by the conventional intermittent infusion method. 12 Therefore, a randomized controlled trial (RCT) was performed to compare the two AT-III administration methods in liver transplantation recipients. We hypothesized that the continuous infusion method of AT-III

WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?
Antithrombin III (AT-III) concentrate has been used to prevent critical thrombosis immediately after liver transplantation. Simulation of various dosing strategies of AT-III concentrate in liver transplantation recipients suggested that continuous infusion would be superior to intermittent infusion in attaining the target level.

WHAT QUESTION DID THIS STUDY ADDRESS?
This randomized controlled trial aimed to investigate a more appropriate method for AT-III concentrate administration to maintain plasma AT-III activity level within the target range.

WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?
Using continuous infusion, plasma AT-III activity level more rapidly reached the target level and was more stably maintained within the target range compared to intermittent infusion. However, the incidence of post-transplantation hemorrhagic or thrombotic complications was not significantly different between the two groups.

HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?
This study showed that continuous infusion of AT-III concentrate is a more efficient and safer administration strategy compared to conventional intermittent infusion in the immediate postoperative period after liver transplantation. Although it was difficult to demonstrate the clinical usefulness of AT-III concentrate due to the extremely low incidence of thrombosis, this study drew on the clear difference between the two groups based on pharmacokinetic analysis.
concentrate would be more appropriate in maintaining the plasma AT-III activity level within the target range compared to the intermittent infusion method.

Patients
The study was approved by the Institutional Review Board of Seoul National University Hospital (IRB number: H-1911-100-1081) and registered at Clini calTr ials.gov (NCT 04219579) before recruitment. The study was designed and executed adhering to the guidelines for Good Clinical Practice and the Consolidated Standards of Reporting Trials.
Adult patients (>18 years old) scheduled to undergo elective living-donor liver transplantation between December 2019 and June 2022 were assessed for eligibility and written informed consent was obtained. Patients who refused to participate in the study or decided to participate in other clinical trials were excluded. Additionally, patients with high risk of significant postoperative bleeding, canceled operation, or protocol violation were omitted. In case of bleeding or thrombosis requiring intervention/re-operation, patients were excluded from the trial and the complications were recorded as secondary outcomes.
Patients were randomly allocated into an intermittent group or a continuous group according to a computergenerated randomization table in a 1:1 ratio. An assistant who was not involved in the study performed the randomization, concealed the allocation in opaque envelopes, and passed it on to the assigned nurse in the intensive care unit (ICU). The assigned nurse administered AT-III concentrate to the patients following the predetermined protocol according to the group allocation. The investigators were not involved in the AT-III concentrate administration during the study period for concealment of the group assignment. The patients were also unaware of the group allocation as the drug was administered behind a screen shield over the patients' heads.

AT-III concentrate administration
AT-III concentrate was administered to all participants immediately after admission to the ICU after liver transplantation. In the intermittent group, 500 international units (IU) of AT-III concentrate were injected over an hour every 6 h for 72 h after admission to the ICU. In the continuous group, a loading dose of 2000 IU was administered over an hour, followed by continuous infusion of 3000 IU of AT-III over 71 h. In compliance with the package insert of AT-III concentrate, the syringe for continuous infusion was replaced every 12 h to maintain the stability of reconstituted AT-III. In total, 6000 IU and 5000 IU of AT-III concentrate for the intermittent group and continuous group were administered, respectively ( Figure 1).
The plasma AT-III activity level at the end of the liver transplantation was recorded as the baseline value (AT-III base ). Plasma AT-III activity level was analyzed at 12, 24, 48, 72, and 84 h (before the dose in the intermittent group) and the measured values were labeled as AT-III 12h , AT-III 24h , AT-III 48h , AT-III 72h , and AT-III 84h (Figure 1), respectively. If the plasma AT-III activity level at a specific timepoint was higher than 120%, AT-III concentrate administration was withheld until the next measured plasma F I G U R E 1 Schematic diagram of the study protocol. In the intermittent group, 500 IU of antithrombin-III (AT-III) concentrate were injected over an hour every 6 h for 72 h. In the continuous group, a loading dose of 2000 IU was administered over an hour, followed by continuous infusion of 3000 IU of AT-III over 71 h. Plasma AT-III activity level was analyzed at baseline, 12, 24, 48, 72, and 84 h, and the measured values were labeled as AT-III base , AT-III 12h , AT-III 24h , AT-III 48h , AT-III 72h , and AT-III 84h , respectively. AT-III, antithrombin III; IU, international unit.
AT-III activity level was 120% or lower, at which point AT-III concentrate administration was resumed.

Determination of plasma AT-III activity level
The plasma AT-III activity levels in plasma were determined by chromogenic assay using ACL-TOP 750 CTS (Instrumentation Laboratory) and factor Xa was used as the enzymatic source. When factor Xa is added to a plasma sample it forms a complex with functionally active AT-III. Chromogenic substrate (HemosIL Liquid Antithrombin; Instrumentation Laboratory) was added to the mixture and the active AT-III level was determined by the color change generated by residual factor Xa.

Outcome measurement
The primary end point was the proportion of patients whose plasma AT-III activity level at 72 h was in the target range of 80%-120%. The proportions of patients who had a plasma AT-III activity level within the target range at 12, 24, 48, and 84 h were collected as the secondary end points. In addition, the proportion of plasma AT-III activity level values within the target range among all the collected samples, the time required for plasma AT-III activity level to reach the target range, and the incidence of postoperative bleeding requiring intervention or thrombosis events were also collected.

Safety evaluation
Safety was evaluated based on the emergence rate of bleeding or thrombosis events requiring intervention and clinical laboratory tests. Prothrombin time, platelet count, hemoglobin level, and serum albumin were evaluated at baseline and 12, 24, 48, 72, 84 h after admission to the ICU.

Statistical analysis
Based on the pharmacokinetic model of the preceding study, the target attainment rates at 72 h after liver transplantation were 43.19% and 60.21% for the intermittent group and continuous group, respectively. 12 The sample size to show superiority of the primary end point in the continuous group compared to the intermittent group was 57 patients in each treatment group with 70% power at a significance level of 0.05. Assuming a 15% dropout rate, a total of 132 subjects were required to be recruited. SAS software version 9.4 (SAS Institute Inc.) was used for statistical analyses. Patients' demographics, baseline characteristics, results of primary and secondary end points, and safety data were summarized and presented by descriptive statistics. Two-sample t-test was used to compare the demographics and baseline characteristics between the treatment groups. A mixed-effect model was developed to compare plasma AT-III activity levels between the treatment groups. Chi-square test was used to compare the proportion data in the primary and the secondary end points between the treatment groups. The time to reach the target plasma AT-III activity level range was compared between the treatment groups using Kruskal-Wallis rank sum test. A p-value of less than 0.05 was considered statistically significant.

Patients characteristics
A total of 157 liver transplantation recipients were screened for eligibility. After excluding 27 patients, 130 patients were enrolled and randomly assigned to either the intermittent group (n = 64) or the continuous group (n = 66) ( Figure 2). AT-III concentrate was administered in accordance with the allocated group in 58 patients in each group and 9 patients were additionally excluded from the analysis due to complications or protocol violations. The remaining 107 patients completed the study as planned and were included in the final analysis. Baseline patient characteristics were similar between the intermittent group and continuous group, except for sex and body weight (Table 1). Total AT-III doses during 72 h were significantly lower in the continuous group compared to the intermittent group (Table 1).

Plasma AT-III activity levels after ICU admission
In the continuous group, the plasma AT-III activity level reached the target range (80%-120%) more rapidly and more stably remained within the target range 84 h postoperatively compared to the intermittent group (Figure 3a). The median time to reach target concentration was 12 h and 24 h in the continuous group and intermittent group, respectively (p < 0.001; Figure 4). The target attainment rate at 72-h postoperatively was significantly higher in the continuous group compared to the intermittent group (62% vs. 30%, p = 0.003). The target attainment rates at 12 h (72% vs. 24%, p < 0.001) and 48 h (77% vs. 52%, p = 0.019) postoperatively were also significantly higher in the continuous group (Figure 3b). The proportion of patients who did not reach the target range at 12 h postoperatively were significantly higher in the intermittent group compared to the continuous group (76% vs. 11%, p < 0.001). Also, the proportion of patients who surpassed the target range at 48 h (33% vs. 2%, p < 0.001) and 72 h (63% vs. 26%, p < 0.001) postoperatively were significantly higher in the intermittent group compared to the continuous group. The proportion of the plasma AT-III activity levels within target range was significantly higher in the continuous group (53% vs. 31%, p < 0.001).

Safety
The number of patients who developed postoperative bleeding requiring intervention during the study period was similar, with two patients in the intermittent group and four patients in the continuous group (3.1% vs. 6.1%, p = 0.425). Critical thrombosis requiring intervention was observed in two patients (3.1%), both in the intermittent group (p = 0.496). One patient developed hepatic artery, portal vein, and hepatic vein thromboses on postoperative Day 1 and underwent surgical thrombectomy. The other patient underwent balloon angioplasty on postoperative Day 3 due to portal vein stenosis and thrombosis. The overall recovery was similar between the two groups as shown by the similar improving trends in prothrombin time, hemoglobin levels, platelet counts, and serum albumin ( Figure 5).

DISCUSSION
Compared to conventional intermittent infusion, continuous infusion of AT-III was more likely to maintain plasma AT-III activity within the target range after living-donor liver transplantation. The AT-III level reached the target range more rapidly and remained within the range longer with continuous infusion.
Several studies have investigated the administration method of AT-III concentrates in various clinical situations. A recent study compared continuous and intermittent administration of AT-III concentrate in pediatric patients using extracorporeal membrane oxygenation. The plasma AT-III activity level remained in the target range longer when using the continuous infusion method, which was adjusted based on the daily AT-III level according to the predetermined protocol compared to the intermittent method which was based on the physician's discretion. 13 Another study that investigated AT-III concentrate regimens in severe sepsis patients showed comparable tolerability and effectiveness between the two regimens. 14 However, the plasma AT-III activity level depending on the administration strategy has not been compared in liver transplant patients in the immediate postoperative period. Our study is the first report that suggests the superiority of continuous administration over intermittent administration when using AT-III concentrate in liver transplant patients in the immediate postoperative period.
Studies regarding the use of AT-III concentrates in liver transplantation are generally scarce. A pilot study of 25 patients showed reduced fibrin degradation product  D-dimer levels and platelet transfusion requirement with AT-III concentrate supplementation in the early postoperative period. 4 Furthermore, a recent retrospective study of 181 children who underwent liver transplantation suggested an association between low plasma AT-III activity level early after liver transplantation and postoperative thrombosis, advocating for postoperative AT-III supplementation. 2 In an RCT in patients with liver disease, AT-III concentrate showed promise for preventing portal vein thrombosis in patients with low plasma AT-III activity level. 15 However, no studies have investigated the dosing strategies in liver transplantation recipients despite the significant perioperative changes in plasma AT-III activity level. In this respect, the results of this RCT may suggest grounds for AT-III concentrate dosing in the immediate postoperative period after liver transplantation. The total dose of AT-III concentrate administered was on average 4764 IU for the continuous group and 5824 IU for the intermittent group. Clinicians tend to side with overdose due to the fear of critical thrombosis and its fatal consequences since there is no clear guidance on AT-III concentrate dosing. As post-transplantation hemorrhage requiring re-operation and the number of transfused red blood cells are known as significant risk factors for mortality, use of optimal AT-III concentrate dosage based on our results may be helpful in improving clinical outcomes. 16 The plasma AT-III activity level was more likely to be within the target range in the continuous group compared to the intermittent group at 72 h (AT-III 72h ), as well as at other timepoints. The plasma AT-III activity level at 12 h (AT-III 12h ) was within the target range in more than 70% of the patients in the continuous group, which was in contrast to 24.1% in the intermittent group. Based on these results, the plasma AT-III activity level seems to reach the target range more quickly and stays within the target range more often in the continuous group, which may be particularly beneficial in patients with the highest risk of thrombogenesis.
There were no significant differences in posttransplantation complications between the two groups. Therefore, it remains unclear whether the delicate control of the plasma AT-III activity level results in the improved clinical outcomes. Two patients in the intermittent group developed thrombosis of the hepatic artery and portal vein requiring intervention. The incidence was significantly lower compared to previous reports, suggesting the efficacy of AT-III concentrate in preventing thrombosis. 10,11 However, the relatively small sample size of our study and the extremely low incidence of thrombosis makes it difficult to draw a reliable conclusion. This study has several limitations that should be considered. First, the study only included living-donor liver transplantations performed at a single medical center. Therefore, it may be difficult to expect similar results in different circumstances. However, considering the significant number of liver transplantations (more than 100 cases per year) performed by several surgeons in this medical center, comparable results can be anticipated in centers with surgical proficiency. Second, significant clinical advantages of AT-III concentrate after liver transplantation were not identified. Due to the low incidence of critical thrombosis during the study period, the clinical utility of AT-III concentrate remains unclear. However, as the focus of this study was on pharmacokinetic analysis, plasma AT-III activity levels showed a clear difference between the two groups.
In conclusion, continuous infusion of AT-III concentrate can be more efficient and safer in maintaining the plasma AT-III activity level within the target range compared to intermittent infusion in the immediate postoperative period after liver transplantation.