RETROSPECTIVE ANALYSIS ASSESSING ECONOMIC IMPACT OF APROTININ REINSTATEMENT IN 4 FRENCH CARDIAC SURGERY CENTRES

Abstract

<h3>Objective</h3> Until 2008, aprotinin (APR) was used to reduce perioperative bleeding and transfusion needs in cardiac surgery but was suspended following the publication of the BART study,1 leaving only tranexamic acid (TXA) as available antifibrinolytic. In 2012, the European Medicine Agency revisited all APR data and restored APR to be given to adult patients at high risk of major blood loss undergoing isolated coronary artery bypass graft (CABG) surgery and requested the marketing authorization holder to gather more information on the profile of APR use through a registry (NAPaR). The economic impact of APR reintroduction on costs was compared to TXA. <h3>Design and Method</h3> This multicenter before-after study comparing APR with TXA was carried out in 4 French university centers. APR use followed ARCOTHOVA (French association of cardiothoracic and vascular anaesthetists) guidance in 2 indication classes: class 1, on-label (iCABG); class 2, off-label, patient with at least 3 risk factors (Trust score), 2 and/or exposed to high risk surgeries. APR data were retrieved from the NAPaR registry. TXA data were retrieved from each center's database, and were matched to APR data based on indication classes. Budget impact was evaluated using both direct costs associated with antifibrinolytics and transfusion products (within first 48h) and indirect costs such as surgery duration and ICU stay. <h3>Results</h3> Between December 2018 and November 2020, 236 patients were treated with APR and were matched to 223 matched patients treated with TXA between May 2016 and October 2019. The 459 patients were distributed as: 17% in class 1, 83% in class 2; 25 APR patients and 23 TXA patients died before ICU discharge and were excluded from the cost analysis. Mean costs per patient tended to be lower in APR group versus TXA group (21K€ vs 24K€), which resulted in an estimated gross saving > 660 K€ for the whole cohort in favour APR. The mean cost per patient differed with the indication class, 22K€ on-label vs 47K€ off-label. Differential costs between APR vs TXA were in the opposite direction for on-label (+4.7K€) and off-label (-5.3K€) use (Figure). Higher costs for TXA patients were related to longer ICU stay, mainly due to those patients who stayed more than 16 days in the ICU (stays exceeding the 90th percentile, p=0.002 versus APR patients). <h3>Conclusions</h3> Costs were driven by the ICU stay. The substantial savings observed with APR treatment were because fewer patients in the APR group stayed long at the ICU compared to TXA group. These unforeseen results raise the question of the role of the unspecific anti-protease action of APR, resulting in anti-inflammatory effect,3 in the observed improvement, and needs further investigation.