Perioperative lethality after endovascular and open repair of ruptured abdominal aortic aneurysms: An analysis of administrative data of the AOK health insurance fund

Objective: In this paper we will report the perioperative outcome after endovascular (EVAR) and open (OAR) repair of ruptured abdominal aortic aneurysms (rAAA) in Germany based on data of the AOK health insurance fund.Methods: Anonymised data of all patients with rAAA (n = 3,227) who were treated from 01/01/2010 to 12/31/2016 were analysed, using SPSS 27 (IBM Deutschland GmbH, Ehningen, Germany).Results: 41.9% (1,353/3,227) of the patients were treated with EVAR and 58.1% (1,874/3,227) with OAR. Patients >= 80 years made up 38.4% for EVAR and 32.9% for OAR (p = 0.002). The proportion of patients undergoing surgery within 24 hours after admission was significantly higher for OAR (87.8%) than for EVAR (73.0%) (p = 0.000). The perioperative lethality rate for OAR was 42.4%, and thus almost twice as high as for EVAR with 21.3% (p = 0.000). Women had higher perioperative lethality rates for both EVAR (perioperative lethality 24.6%) and OAR (perioperative lethality 51.7%) compared to men with 20.6% (EVAR) and 40.2% (OAR), respectively. With EVAR, 35.8% of the patients showed a complication-free post-operative course, with OAR it was 17.7% (p = 0.000). Blood transfusions (whole blood, red cell concen-trates, and autotransfusions) were administered in 57.6% of the patients with EVAR, but in 92.3% with OAR (p = 0.000). The highest perioperative lethality was found in EVAR and OAR patients who received both surgery within 24 hours after admission and blood transfusions (perioperative lethality EVAR 36.0%, OAR 46.0%; p = 0.000). In contrast, patients who did not require blood transfusions and were trea-ted later than 24 hours after admission had the lowest perioperative lethality with 3.2% for EVAR vs. 5.4% for OAR (p = 0.623).Conclusion: The data confirm the observation that the perioperative mortality of rAAA patients is lower with EVAR than with OAR. However, strict attention must be paid to the time of the intervention. The low perioperative lethality of patients who were treated later than 24 hours after hospital admission and who did not require blood transfusions indicates that cases of symptomatic AAA without rupture have also been recorded in this administrative database under the diagnosis rAAA. One point of criticism is that the decision not to adjust for the patient groups with EVAR and with OAR in order to be able to better analyse the properties of routine data includes a considerable risk of bias in the statements of this work due to confounding variables.