1
|
Li B, Aljabri B, Verma R, Beaton D, Hussain MA, Lee DS, Wijeysundera DN, de Mestral C, Mamdani M, Al‐Omran M. Predicting Outcomes Following Lower Extremity Endovascular Revascularization Using Machine Learning. J Am Heart Assoc 2024; 13:e033194. [PMID: 38639373 PMCID: PMC11179886 DOI: 10.1161/jaha.123.033194] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/18/2023] [Accepted: 03/01/2024] [Indexed: 04/20/2024]
Abstract
BACKGROUND Lower extremity endovascular revascularization for peripheral artery disease carries nonnegligible perioperative risks; however, outcome prediction tools remain limited. Using machine learning, we developed automated algorithms that predict 30-day outcomes following lower extremity endovascular revascularization. METHODS AND RESULTS The National Surgical Quality Improvement Program targeted vascular database was used to identify patients who underwent lower extremity endovascular revascularization (angioplasty, stent, or atherectomy) for peripheral artery disease between 2011 and 2021. Input features included 38 preoperative demographic/clinical variables. The primary outcome was 30-day postprocedural major adverse limb event (composite of major reintervention, untreated loss of patency, or major amputation) or death. Data were split into training (70%) and test (30%) sets. Using 10-fold cross-validation, 6 machine learning models were trained using preoperative features. The primary model evaluation metric was area under the receiver operating characteristic curve. Overall, 21 886 patients were included, and 30-day major adverse limb event/death occurred in 1964 (9.0%) individuals. The best performing model for predicting 30-day major adverse limb event/death was extreme gradient boosting, achieving an area under the receiver operating characteristic curve of 0.93 (95% CI, 0.92-0.94). In comparison, logistic regression had an area under the receiver operating characteristic curve of 0.72 (95% CI, 0.70-0.74). The calibration plot showed good agreement between predicted and observed event probabilities with a Brier score of 0.09. The top 3 predictive features in our algorithm were (1) chronic limb-threatening ischemia, (2) tibial intervention, and (3) congestive heart failure. CONCLUSIONS Our machine learning models accurately predict 30-day outcomes following lower extremity endovascular revascularization using preoperative data with good discrimination and calibration. Prospective validation is warranted to assess for generalizability and external validity.
Collapse
Affiliation(s)
- Ben Li
- Department of SurgeryUniversity of TorontoCanada
- Division of Vascular SurgerySt. Michael’s Hospital, Unity Health Toronto, University of TorontoTorontoCanada
- Institute of Medical Science, University of TorontoTorontoCanada
- Temerty Centre for Artificial Intelligence Research and Education in Medicine (T‐CAIREM)University of TorontoTorontoCanada
| | - Badr Aljabri
- Department of SurgeryKing Saud UniversityRiyadhSaudi Arabia
| | - Raj Verma
- School of Medicine, Royal College of Surgeons in IrelandUniversity of Medicine and Health SciencesDublinIreland
| | - Derek Beaton
- Data Science & Advanced Analytics, Unity Health TorontoUniversity of TorontoTorontoCanada
| | - Mohamad A. Hussain
- Division of Vascular and Endovascular Surgery and the Center for Surgery and Public Health, Brigham and Women’s HospitalHarvard Medical SchoolBostonMAUSA
| | - Douglas S. Lee
- Division of Cardiology, Peter Munk Cardiac CentreUniversity Health NetworkTorontoCanada
- Institute of Health Policy, Management and Evaluation, University of TorontoTorontoCanada
- ICES, University of TorontoTorontoCanada
| | - Duminda N. Wijeysundera
- Institute of Health Policy, Management and Evaluation, University of TorontoTorontoCanada
- ICES, University of TorontoTorontoCanada
- Department of AnesthesiaSt. Michael’s Hospital, Unity Health TorontoTorontoCanada
- Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Unity Health TorontoTorontoCanada
| | - Charles de Mestral
- Department of SurgeryUniversity of TorontoCanada
- Division of Vascular SurgerySt. Michael’s Hospital, Unity Health Toronto, University of TorontoTorontoCanada
- Institute of Health Policy, Management and Evaluation, University of TorontoTorontoCanada
- ICES, University of TorontoTorontoCanada
- Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Unity Health TorontoTorontoCanada
| | - Muhammad Mamdani
- Institute of Medical Science, University of TorontoTorontoCanada
- Temerty Centre for Artificial Intelligence Research and Education in Medicine (T‐CAIREM)University of TorontoTorontoCanada
- Data Science & Advanced Analytics, Unity Health TorontoUniversity of TorontoTorontoCanada
- Institute of Health Policy, Management and Evaluation, University of TorontoTorontoCanada
- ICES, University of TorontoTorontoCanada
- Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Unity Health TorontoTorontoCanada
- Leslie Dan Faculty of PharmacyUniversity of TorontoTorontoCanada
| | - Mohammed Al‐Omran
- Department of SurgeryUniversity of TorontoCanada
- Division of Vascular SurgerySt. Michael’s Hospital, Unity Health Toronto, University of TorontoTorontoCanada
- Institute of Medical Science, University of TorontoTorontoCanada
- Temerty Centre for Artificial Intelligence Research and Education in Medicine (T‐CAIREM)University of TorontoTorontoCanada
- Li Ka Shing Knowledge Institute, St. Michael’s Hospital, Unity Health TorontoTorontoCanada
- Department of SurgeryKing Faisal Specialist Hospital and Research CenterRiyadhSaudi Arabia
| |
Collapse
|
2
|
Hönemann JN, Jordan J. Will obesity break your heart - cardiac biomarkers in the Japan Morning Surge-Home Blood Pressure study. Hypertens Res 2024; 47:808-810. [PMID: 38200217 PMCID: PMC10912013 DOI: 10.1038/s41440-023-01560-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/02/2023] [Revised: 11/21/2023] [Accepted: 12/02/2023] [Indexed: 01/12/2024]
Affiliation(s)
- Jan-Niklas Hönemann
- Department of Internal Medicine III, Division of Cardiology, Pneumology, Angiology, and Intensive Care, University of Cologne, Cologne, Germany
- Institute of Aerospace Medicine, German Aerospace Center (DLR), Cologne, Germany
| | - Jens Jordan
- Institute of Aerospace Medicine, German Aerospace Center (DLR), Cologne, Germany.
- Medical Faculty, University of Cologne, Cologne, Germany.
| |
Collapse
|
3
|
Kazantsev AN, Korotkikh AV, Wang S, Nonye OG, Artyukhov SV, Mukhtorov OS, Roshkovskaya LV, Unguryan VM, Shmatov DV, Urokov DA, Choriev AA, Zabolotniy VN, Lebedev OV, Zarkua NE, Kravchuk VN, Koplik VO, Kudryavtsev ME, Bagdavadze GS, Chernyavin MP, Leader RY, Kazantseva EG, Belov YV. Hospital and long-term results of carotid endarterectomy in patients with different severity of coronary artery lesion according to syntax score. Curr Probl Cardiol 2024; 49:102244. [PMID: 38043882 DOI: 10.1016/j.cpcardiol.2023.102244] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/27/2023] [Accepted: 11/28/2023] [Indexed: 12/05/2023]
Abstract
AIM Analysis of in-hospital and long-term results of carotid endarterectomy (CEE) in patients with different severity of coronary atherosclerosis. MATERIAL AND METHODS This comparative, retrospective, open study for the period from January 2013 to April 2020 included 1719 patients operated on for occlusive-stenotic lesions of the internal carotid arteries (ICA). Classical and eversion CEA were used as revascularization strategies. The criteria for inclusion in the study were: 1. Presence of coronary angiography within six months before the present CEE; 2. A history of myocardial revascularization in patients with severe coronary lesions. Depending on the severity of coronary atherosclerosis, all patients were divided into 3 groups: Group 1-871 (50.7 %) patients - with the presence of hemodynamically significant stenosis of the coronary arteries (CA) with a history of myocardial revascularization; Group 2-496 (28.8 %) patients - with the presence of hemodynamically insignificant lesions of the coronary artery (up to 70 %, not inclusive, and the trunk of the left coronary artery, up to 50 %, not inclusive); Group 3-352 (20.5 %) patients - without signs of atherosclerotic lesions of the coronary artery. In group 1, the observation period was 56.8±23.2 months, in group 2-62.0±15.6 months, in group 3-58.1±20.4 months. RESULTS During the hospital observation period, there were no significant intergroup differences in the number of complications. All cardiovascular events were detected in isolated cases. The most common injury was damage to the cranial nerves, diagnosed in every fifth patient in the total sample. The combined endpoint (CET), including death + myocardial infarction (MI) + acute cerebrovascular accident/transient ischemic attack (stroke/TIA), was 0.75 % (n=13). In the long-term follow-up period, when comparing survival curves, group 3 revealed the largest number of ischemic strokes (p = 0.007), myocardial infarction (p = 0.03), and CCT (p = 0.005). There were no intergroup differences in the number of deaths (p=0.62). CONCLUSION The results of the study showed that there was no significant intergroup difference in the development of complications at the hospital postoperative stage. However, in the long-term follow-up period, a group of patients with isolated lesions of the ICA demonstrated a rapid increase in the number of MI, stroke/TIA, and a combined endpoint, which was apparently associated with low compliance and progression of atherosclerosis in previously unaffected arteries.
Collapse
Affiliation(s)
- Anton N Kazantsev
- Kostroma Regional Clinic, Kostroma, Russian Federation; Russian Scientific Center for Surgery Named After Academician B.V. Petrovsky, Moscow, Russian Federation; Kostroma Oncology Dispensary, Kostroma, Russian Federation.
| | - Alexander V Korotkikh
- Cardiac Surgery Clinic, Amur State Medical Academy, Blagoveshchensk, Russian Federation
| | - Showen Wang
- First Moscow State Medical University, THEM. Sechenov, Moscow, Russian Federation
| | | | - Sergey V Artyukhov
- State Budgetary Healthcare Institution "City Alexander Hospital", St. Petersburg, Russian Federation
| | | | - Lyudmila V Roshkovskaya
- State Budgetary Healthcare Institution "City Alexander Hospital", St. Petersburg, Russian Federation
| | | | - Dmitry V Shmatov
- St. Petersburg State University, St. Petersburg, Russian Federation
| | | | | | | | | | - Nona E Zarkua
- Northwestern State Medical University Named After Mechnikov, St. Petersburg, Russian Federation
| | - Vyacheslav N Kravchuk
- Northwestern State Medical University Named After Mechnikov, St. Petersburg, Russian Federation
| | - Victoria O Koplik
- Novgorod State University Named After Yaroslav the Wise, Veliky Novgorod, Russian Federation
| | - Mikhail E Kudryavtsev
- Novgorod State University Named After Yaroslav the Wise, Veliky Novgorod, Russian Federation
| | | | - Maxim P Chernyavin
- Clinical Hospital of the Administration of the President of the Russian Federation, Moscow, Russian Federation
| | - Roman Yu Leader
- Kemerovo State Medical University, Kemerovo, Russian Federation
| | | | - Yuri V Belov
- Russian Scientific Center for Surgery Named After Academician B.V. Petrovsky, Moscow, Russian Federation; First Moscow State Medical University, THEM. Sechenov, Moscow, Russian Federation
| |
Collapse
|