An Artificial Neural Network Stratifies the Risks of Reintervention and Mortality after Endovascular Aneurysm Repair; a Retrospective Observational study

Background Lifelong surveillance after endovascular repair (EVAR) of abdominal aortic aneurysms (AAA) is considered mandatory to detect potentially life-threatening endograft complications. A minority of patients require reintervention but cannot be predictively identified by existing methods. This study aimed to improve the prediction of endograft complications and mortality, through the application of machine-learning techniques. Methods Patients undergoing EVAR at 2 centres were studied from 2004-2010. Pre-operative aneurysm morphology was quantified and endograft complications were recorded up to 5 years following surgery. An artificial neural networks (ANN) approach was used to predict whether patients would be at low- or high-risk of endograft complications (aortic/limb) or mortality. Centre 1 data were used for training and centre 2 data for validation. ANN performance was assessed by Kaplan-Meier analysis to compare the incidence of aortic complications, limb complications, and mortality; in patients predicted to be low-risk, versus those predicted to be high-risk. Results 761 patients aged 75 +/- 7 years underwent EVAR. Mean follow-up was 36+/- 20 months. An ANN was created from morphological features including angulation/length/areas/diameters/volume/tortuosity of the aneurysm neck/sac/iliac segments. ANN models predicted endograft complications and mortality with excellent discrimination between a low-risk and high-risk group. In external validation, the 5-year rates of freedom from aortic complications, limb complications and mortality were 95.9% vs 67.9%; 99.3% vs 92.0%; and 87.9% vs 79.3% respectively (p<0.001) Conclusion This study presents ANN models that stratify the 5-year risk of endograft complications or mortality using routinely available pre-operative data.

[1]  Andrew Baker,et al.  Clinical Outcome Prediction in Aneurysmal Subarachnoid Hemorrhage Using Bayesian Neural Networks with Fuzzy Logic Inferences , 2013, Comput. Math. Methods Medicine.

[2]  R W Raven,et al.  The british association of surgical oncology , 1973, Annals of the Royal College of Surgeons of England.

[3]  Alan Karthikesalingam,et al.  Predictive Value of Peak Systolic Velocity for the Development of Graft Limb Complications After Endovascular Aneurysm Repair , 2012, Journal of endovascular therapy : an official journal of the International Society of Endovascular Specialists.

[4]  P. Lin,et al.  Aneurysm rupture after EVAR: can the ultimate failure be predicted? , 2009, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[5]  W. Baxt,et al.  A neural computational aid to the diagnosis of acute myocardial infarction. , 2002, Annals of emergency medicine.

[6]  T. Carrell,et al.  Long‐term surveillance with computed tomography after endovascular aneurysm repair may not be justified , 2009, The British journal of surgery.

[7]  A. Karthikesalingam,et al.  Secondary interventions following endovascular aneurysm repair (EVAR) and the enduring value of graft surveillance. , 2010, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[8]  F. Dick,et al.  The influence of thrombus, calcification, angulation, and tortuosity of attachment sites on the time to the first graft-related complication after endovascular aneurysm repair. , 2011, Journal of vascular surgery.

[9]  M M Thompson,et al.  Heterogeneity in surveillance after endovascular aneurysm repair in the UK. , 2011, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[10]  S. Haulon,et al.  Management of abdominal aortic aneurysms clinical practice guidelines of the European society for vascular surgery. , 2011, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[11]  R. L. Kennedy,et al.  Artificial neural network models for prediction of acute coronary syndromes using clinical data from the time of presentation. , 2005, Annals of emergency medicine.

[12]  R. Edwards,et al.  Surveillance of patients post-endovascular abdominal aortic aneurysm repair (EVAR). A web-based survey of practice in the UK. , 2013, Clinical radiology.

[13]  G. Moneta.  Predictors of Abdominal Aortic Aneurysm Sac Enlargement After Endovascular Repair , 2011 .

[14]  Matthew J. Bown,et al.  Endovascular Aortic Aneurysm Repair in Patients with Hostile Neck Anatomy , 2013, Journal of endovascular therapy : an official journal of the International Society of Endovascular Specialists.

[15]  Vincenzo Lagani,et al.  Structure-based variable selection for survival data , 2010, Bioinform..

[16]  S. Walsh,et al.  Radiation exposure during endovascular aneurysm repair , 2008, The British journal of surgery.

[17]  A. Karthikesalingam,et al.  Systematic Review and Meta-analysis of Duplex Ultrasonography, Contrast- enhanced Ultrasonography or Computed Tomography for Surveillance After Endovascular Aneurysm Repair , 2013 .

[18]  J. van der Meulen,et al.  Identifying co‐morbidity in surgical patients using administrative data with the Royal College of Surgeons Charlson Score , 2010, The British journal of surgery.

[19]  P. Harris,et al.  Diameter of abdominal aortic aneurysm and outcome of endovascular aneurysm repair: does size matter? A report from EUROSTAR. , 2004, Journal of vascular surgery.

[20]  M M Thompson,et al.  Risk of reintervention after endovascular aortic aneurysm repair , 2010, The British journal of surgery.

[21]  A. Karthikesalingam,et al.  Predicting aortic complications after endovascular aneurysm repair , 2013, British Journal of Surgery.

[22]  Juanmei Liu,et al.  Comparison of Prediction Model for Cardiovascular Autonomic Dysfunction Using Artificial Neural Network and Logistic Regression Analysis , 2013, PloS one.

[23]  P. Stather,et al.  Outcomes of endovascular aneurysm repair in patients with hostile neck anatomy. , 2012, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[24]  Mohammad H. Eslami,et al.  Predictors of Abdominal Aortic Aneurysm Sac Enlargement After Endovascular Repair , 2011, Circulation.

[25]  M. Sculpher,et al.  The UK EndoVascular Aneurysm Repair (EVAR) trials: randomised trials of EVAR versus standard therapy. , 2012, Health technology assessment.

[26]  S. Vallabhaneni,et al.  Surveillance after EVAR based on duplex ultrasound and abdominal radiography. , 2011, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[27]  L. Lawler,et al.  Use of colour duplex ultrasound as a first line surveillance tool following EVAR is associated with a reduction in cost without compromising accuracy. , 2012, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[28]  D. Mant,et al.  Systematic review and validation of prediction rules for identifying children with serious infections in emergency departments and urgent-access primary care. , 2012, Health technology assessment.

[29]  S. Walsh,et al.  Renal Consequences of Endovascular Abdominal Aortic Aneurysm Repair , 2008, Journal of endovascular therapy : an official journal of the International Society of Endovascular Specialists.

[30]  Mark J Sculpher,et al.  Endovascular versus Open Repair of Abdominal Aortic Aneurysm , 2011 .

[31]  Ayman Salem,et al.  St George's Vascular Institute Protocol: An Accurate and Reproducible Methodology to Enable Comprehensive Characterization of Infrarenal Abdominal Aortic Aneurysm Morphology in Clinical and Research Applications , 2012, Journal of endovascular therapy : an official journal of the International Society of Endovascular Specialists.

[32]  Paolo Emilio Puddu,et al.  Artificial neural networks versus proportional hazards Cox models to predict 45-year all-cause mortality in the Italian Rural Areas of the Seven Countries Study , 2012, BMC Medical Research Methodology.

[33]  J.,et al.  The New England Journal of Medicine , 2012 .

[34]  L. Spelt,et al.  Artificial neural networks--a method for prediction of survival following liver resection for colorectal cancer metastases. , 2013, European journal of surgical oncology : the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology.

[35]  F. Setacci,et al.  Surveillance imaging modality does not affect detection rate of asymptomatic secondary interventions following EVAR. , 2012, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[36]  C. Kalbaugh,et al.  Lost to follow-up: a potential under-appreciated limitation of endovascular aneurysm repair. , 2007, Journal of vascular surgery.

[37]  P. Geisbüsch,et al.  Current clinical practice in postoperative endovascular aneurysm repair imaging surveillance. , 2012, Journal of vascular and interventional radiology : JVIR.

[38]  D. Chambers,et al.  Endovascular stents for abdominal aortic aneurysms: a systematic review and economic model. , 2009, Health technology assessment.

[39]  Chen-chiang Lin,et al.  Comparison of artificial neural network and logistic regression models for predicting mortality in elderly patients with hip fracture. , 2010, Injury.

[40]  B. Sonesson,et al.  Is there a benefit of frequent CT follow-up after EVAR? , 2009, European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery.

[41]  S. Thomas,et al.  Cost‐effectiveness of endovascular abdominal aortic aneurysm repair , 2005, The British journal of surgery.

[42]  Mattias Ohlsson,et al.  Risk factor identification and mortality prediction in cardiac surgery using artificial neural networks. , 2006, The Journal of thoracic and cardiovascular surgery.

[43]  Diederick E Grobbee,et al.  Long-term outcome of open or endovascular repair of abdominal aortic aneurysm. , 2010, The New England journal of medicine.

[44]  Panagiotis Kougias,et al.  Long-term comparison of endovascular and open repair of abdominal aortic aneurysm. , 2012, The New England journal of medicine.

[45]  Alberto Vidal-Diez,et al.  Predicting Aortic Complications After Endovascular Aneurysm Repair , 2014 .

[46]  J. H. Kim,et al.  A case-matched validation study of anatomic severity grade score in predicting reinterventions after endovascular aortic aneurysm repair. , 2012, Journal of vascular surgery.

[47]  ScienceDirect.  European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery. , 1995 .

[48]  J. Nilsson,et al.  Artificial neural networks predict survival from pancreatic cancer after radical surgery. , 2013, American journal of surgery.

[49]  J. Nilsson,et al.  Artificial neural networks in pancreatic disease , 2008, The British journal of surgery.

[50]  Roland L. Dunbrack,et al.  The Role of Balanced Training and Testing Data Sets for Binary Classifiers in Bioinformatics , 2013, PloS one.

[51]  G. Moneta,et al.  Compliance with long-term surveillance recommendations following endovascular aneurysm repair or type B aortic dissection. , 2012, Journal of vascular surgery.