Artificial neural network score potential in in-hospital mortality after ruptured aortic aneurysm repair in a tertiary hospital
DOI:
https://doi.org/10.48750/acv.684Keywords:
Aortic Rupture, Mortality, Artificial InteligenceAbstract
INTRODUCTION: Isolated use of scoring systems is not recommended for selecting patients for repair or palliation of ruptured abdominal aortic aneurysm (rAAA). Artificial intelligence may lead to improved scoring systems in the future, but current evidence is scarce. The aim of this study is to assess the applicability and potential of an artificial neural network (ANN) score in a tertiary hospital.
METHODS: A retrospective review of patients who underwent repair for rAAA at a tertiary institution from January 2012 to December 2024 was conducted. One point was assigned for each ANN item: age >70>70 years, loss of consciousness, cardiac arrest, and shock. The primary outcome was in-hospital mortality.
RESULTS: A total of 150 patients were admitted for rAAA repair between 2012 and 2024. Three patients were excluded for lack of relevant information. The mean age was 73.7 (±9.5)(±9.5) years, and the majority were male (91.2%)(91.2%) n=134n=134 ). The prevalence of smoking history was 47.6%47.6% (n=70)(n=70) . Open repair was performed in 58.5%58.5% (n=86)(n=86) . According to the ANN score, 17.0%17.0% (n=25)(n=25) scored 0, 25.2%25.2% (n=37)(n=37) scored 1, 32.0%32.0% (n=47)(n=47) scored 2, 19.0%19.0% (n=28)(n=28) scored 3, and 6.8%6.8% (n=10)(n=10) scored 4. In the univariable analysis, patients with scores of 3 and 4 had a higher risk of in-hospital mortality (OR 11.8; 3.2-43.0, CI 95%95% p<0.001p<0.001 and OR 10.3; 1.7-60.9, CI 95%95% p=0.010p=0.010 , respectively), whereas patients with a score of 2 had higher mortality but did not reach statistical significance (OR 2.7; 0.9-7.7, CI 95%95% p=0.064p=0.064 ). The area under the curve for the ANN score was 0.72 (CI 95%95% 0.63-0.80).
CONCLUSION: The ANN score had moderate predictive power for in-hospital mortality in our population, especially among patients with higher scores. However, the study's uncentric and retrospective nature are limitations. Future models based on a larger number of patients may yield better predictive performance.
Downloads
References
Karthikesalingam A, Holt PJ, Vidal-Diez A, Ozdemir BA, Poloniecki JD, Hinchliffe RJ, et al. Mortality from ruptured abdominal aortic aneurysms: clinical lessons from a comparison of outcomes in England and the USA. Lancet. 2014;383:963-9.
Zhang S, Feng J, Li H, Zhang Y, Lu Q, Jing Z. Open surgery (OS) versus endovascular aneurysm repair (EVAR) for hemodynamically stable and unstable ruptured abdominal aortic aneurysm (rAAA). Heart Vessels. 2016;31:1291-302.
Wanhanen A, Van Herzele I, Bastos Goncalves F, Bellmunt Montoya S, Berard X, Boyle JR, et al. European Society for Vascular Surgery (ESVS) 2024 Clinical Practice Guidelines on the Management of Abdominal Aorto-lliac Artery Aneurysms. Eur J Vasc Endovasc Surg. 2024;67:192-331.
Robinson WP, Schanzer A, Li Y, Goodney PP, Nolan BW, Eslami MH, et al. Derivation and validation of a practical risk score for prediction of mortality after open repair of ruptured abdominal aortic aneurysms in a U.S. regional cohort and comparison to existing scoring systems. J Vasc Surg. 2013;57:354-61.
von Meijenfeldt GCI, van Beek SC, Bastos Goncalves F, Verhagen HJM, Zeebregts CJ, Vahl AC, et al. Development and External Validation of a Model Predicting Death After Surgery in Patients With a Ruptured Abdominal Aortic Aneurysm: The Dutch Aneurysm Score. Eur J Vasc Endovasc Surg. 2017;53:168-74.
Garland BT, Danaher PJ, Desikan S, Tran NT, Quiroga E, Singh N, et al. Preoperative risk score for the prediction of mortality after repair of ruptured abdominal aortic aneurysms. J Vasc Surg. 2018;68:991-7.
Hemingway JF, Caps M, Zettervall SL, Benyakorn T, Quiroga E, Tran N, et al. Modified Harborview Risk Score improves ease in predicting mortality after ruptured abdominal aortic aneurysm repair. J Vasc Surg. 2024;79:562-8.
Wise ES, Hocking KM, Brophy CM. Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network. J Vasc Surg. 2015;62:8-15.
Schad CA, Fallon BP, Monteagudo J, Okochi S, Cheung EW, Morrissey NJ, et al. Routine Use of Distal Arterial Perfusion in Pediatric Femoral Venoarterial Extracorporeal Membrane Oxygenation. Artif Organs. 2017;41:11-6.
Luebke T, Majd P, Mylonas SN, Brunkwall J. Artificial neural network for prediction of in-hospital mortality after open repair of ruptured abdominal aortic aneurysm. J Cardiovasc Surg (Torino). 2017;58:794-6.
Li SR, Mazroua MS, Reitz KM, Phillips AR, Tzeng E, Liang NL. External Validation of Eight Ruptured Abdominal Aortic Aneurysm Mortality Prediction Models Demonstrates Limited Predictive Accuracy. Eur J Vasc Endovasc Surg. 2025;69:824-34.
