Methodology for Estimation of Annual Risk of Rupture for Abdominal Aortic Aneurysm

• Published in: Computer methods and programs in biomedicine Vol. 200; p. 105916
• Main Authors: Polzer, Stanislav, Kracík, Jan, Novotný, Tomáš, Kubíček, Luboš, Staffa, Robert, Raghavan, Madhavan L.
• Format: Journal Article
• Language: English
• Published: Ireland Elsevier B.V 01.03.2021
• Subjects: Abdominal Aortic Aneurysm; Annual risk of rupture; Aortic Aneurysm, Abdominal - diagnostic imaging; Aortic Rupture - diagnostic imaging; Bayes Theorem; Bayesian statistics; Humans; Models, Cardiovascular; Patient-Specific Modeling; Risk Assessment; Risk Factors; Rupture Risk Assessment; Stress, Mechanical; Tomography, X-Ray Computed; Abdominal Aortic Aneurysm; Rupture Risk Assessment; Annual risk of rupture; Bayesian statistics
• ISSN: 0169-2607; 1872-7565
• DOI: 10.1016/j.cmpb.2020.105916

•Methodology for computing annual risk of rupture of aortic aneurysm is presented.•Combination of finite element analysis and Bayesian statistics is applied.•It respects patient-specific blood pressure variations for the first time.•It respects the fact intact aneurysm was intact also for certain time before scanning. Background and Objective: Estimating patient specific annual risk of rupture of abdominal aortic aneurysm (AAA) is currently based only on population. More accurate knowledge based on patient specific data would allow surgical treatment of only those AAAs with significant risk of rupture. This would be beneficial for both patients and health care system. Methods: A methodology for estimating annual risk of rupture (EARR) of abdominal aortic aneurysms (AAA) that utilizes Bayesian statistics, mechanics and patient-specific blood pressure monitoring data is proposed. EARR estimation takes into consideration, peak wall stress in AAA computed by patient-specific finite element modeling, the probability distributions of wall thickness, wall strength, systolic blood pressure and the period of time that the patient is known to have already survived with the intact AAA. Initial testing of proposed approach was performed on fifteen patients with intact AAA (mean maximal diameter 51mm±8mm). They were equipped with a pressure holter and their blood pressure was recorded over 24 hours. Then, we calculated EARR values for four possible scenarios – without considering any days of survival prior identification of AAA at computed tomography scans (EARR_0), considering past survival of 30 (EARR_30), 90 (EARR_90) and 180 days (EARR_180). Finally, effect of patient-specific blood pressure variability was analyzed. Results: Consideration of past survival does indeed significantly improve predictions of future risk: EARR_30 (1.04%± 0.87%), EARR_90 (0.67%± 0.56%) and EARR_180 (0.47%± 0.39%) which are unrealistically high otherwise (EARR_0 5.02%± 5.24%). Finally, EARR values were observed to vary by an order as a consequence of blood pressure variability and by factor of two as a consequence of neglected growth. Conclusions: Methodology for computing annual risk of rupture of AAA was developed for the first time. Sensitivity analyses showed respecting patient specific blood pressure is important factor and should be included in the AAA rupture risk assessment. Obtained EARR values were generally low and in good agreement with confirmed survival time of investigated patients so proposed method should be further clinically validated.