1
|
Xu C, Li H, Yang J, Peng Y, Cai H, Zhou J, Gu W, Chen L. Interpretable prediction of 3-year all-cause mortality in patients with chronic heart failure based on machine learning. BMC Med Inform Decis Mak 2023; 23:267. [PMID: 37985996 PMCID: PMC10662001 DOI: 10.1186/s12911-023-02371-5] [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: 05/27/2023] [Accepted: 11/08/2023] [Indexed: 11/22/2023] Open
Abstract
BACKGROUND The goal of this study was to assess the effectiveness of machine learning models and create an interpretable machine learning model that adequately explained 3-year all-cause mortality in patients with chronic heart failure. METHODS The data in this paper were selected from patients with chronic heart failure who were hospitalized at the First Affiliated Hospital of Kunming Medical University, from 2017 to 2019 with cardiac function class III-IV. The dataset was explored using six different machine learning models, including logistic regression, naive Bayes, random forest classifier, extreme gradient boost, K-nearest neighbor, and decision tree. Finally, interpretable methods based on machine learning, such as SHAP value, permutation importance, and partial dependence plots, were used to estimate the 3-year all-cause mortality risk and produce individual interpretations of the model's conclusions. RESULT In this paper, random forest was identified as the optimal aools lgorithm for this dataset. We also incorporated relevant machine learning interpretable tand techniques to improve disease prognosis, including permutation importance, PDP plots and SHAP values for analysis. From this study, we can see that the number of hospitalizations, age, glomerular filtration rate, BNP, NYHA cardiac function classification, lymphocyte absolute value, serum albumin, hemoglobin, total cholesterol, pulmonary artery systolic pressure and so on were important for providing an optimal risk assessment and were important predictive factors of chronic heart failure. CONCLUSION The machine learning-based cardiovascular risk models could be used to accurately assess and stratify the 3-year risk of all-cause mortality among CHF patients. Machine learning in combination with permutation importance, PDP plots, and the SHAP value could offer a clear explanation of individual risk prediction and give doctors an intuitive knowledge of the functions of important model components.
Collapse
Affiliation(s)
- Chenggong Xu
- The First Affiliated Hospital of Kunming Medical University, Kunming, China
| | - Hongxia Li
- The First Affiliated Hospital of Kunming Medical University, Kunming, China
| | - Jianping Yang
- College of Big Data, Yunnan Agricultural University, Kunming, China
| | - Yunzhu Peng
- The First Affiliated Hospital of Kunming Medical University, Kunming, China
| | - Hongyan Cai
- The First Affiliated Hospital of Kunming Medical University, Kunming, China
| | - Jing Zhou
- The First Affiliated Hospital of Kunming Medical University, Kunming, China
| | - Wenyi Gu
- The First Affiliated Hospital of Kunming Medical University, Kunming, China
| | - Lixing Chen
- The First Affiliated Hospital of Kunming Medical University, Kunming, China.
| |
Collapse
|
2
|
Nohno Y, Fujiu K, Kozu R, Maruyama S, Hasegawa T, Tsuchiya H, Yanagisawa T, Tachibana T, Kimura H, Yazaki Y. Conduction System Pacing Upgrade in Chronic Heart Failure with Severe Left Ventricular Dysfunction and Chronic Atrial Fibrillation. Int Heart J 2023; 64:768-774. [PMID: 37460323 DOI: 10.1536/ihj.22-614] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 08/01/2023]
Abstract
Cardiac resynchronization therapy (CRT) is a standard treatment for patients with severe congestive heart failure. However, one-third of patients receiving CRT are non-responders. Conduction system pacing (CSP), including His-bundle pacing (HBP) and left bundle branch area pacing (LBBAP), has become an alternative to CRT therapy. Furthermore, CRT therapy with CSP has shown to be more effective than CRT alone. When an implantable cardiac defibrillator or CRT-defibrillator is implanted with CSP, the problem of which port the HBP lead and LBBAP lead should be connected to arises. We report 2 cases of upgrading to CRT with CSP by utilizing the atrial ports for HBP and LBBAP leads. The procedure is a simple, reasonable, and effective therapy for end-stage heart failure.
Collapse
Affiliation(s)
- Yasumasa Nohno
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| | - Katsuhito Fujiu
- Department of Cardiovascular Medicine, The University of Tokyo
| | - Ryosuke Kozu
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| | - Shusaku Maruyama
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| | - Tomoya Hasegawa
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| | - Hiromi Tsuchiya
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| | | | | | - Hikaru Kimura
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| | - Yoshikazu Yazaki
- Department of Cardiology, Saku Central Hospital Advanced Care Center
| |
Collapse
|
3
|
Celikyurt U, Acar B, Agacdiken A, Vural A. Electrocardiographic Predictors of Complete Heart Block During Right Ventricular Lead Implantation in Patients Who Underwent Cardiac Resynchronization Therapy. Am J Cardiol 2023; 201:62-67. [PMID: 37352666 DOI: 10.1016/j.amjcard.2023.05.072] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/17/2023] [Revised: 05/08/2023] [Accepted: 05/29/2023] [Indexed: 06/25/2023]
Abstract
Cardiac resynchronization therapy (CRT) device procedures have their own complications in addition to the complications associated with standard pacemaker implantations. This study aimed to analyze the predictors of the right bundle branch injury resulting in complete heart block (CHB) during right ventricular (RV) lead implantation in patients who underwent CRT with defibrillator. We conducted an observational study of consecutive 790 patients who underwent CRT with defibrillator device implantation at our institution from 2010 to 2022. Relevant clinical information and complete data regarding the echocardiographic data, implantation procedure, and clinical follow-up were collected into a computerized database. A total of 29 patients (3.7%) had CHB during RV lead implantation. In multivariate analysis, left axis deviation (odds ratio [OR] 2.408, 95% confidence interval [CI] 1.025 to 5.658, p = 0.044), QRS width (OR 1.022, 95% CI 1.001 to 1.043, p = 0.035) and QRS alternans (OR 4.214, 95% CI 1.788 to 9.930, p = 0.001) were found independently related to right bundle branch injury resulting in CHB development during RV lead implantation. In conclusion, left axis deviation, QRS width, and QRS alternans were associated with a higher rate of CHB, and these findings provide insight into optimal and safe CRT device implantation strategies based on preprocedural characteristics.
Collapse
Affiliation(s)
- Umut Celikyurt
- Arrhythmia, Electrophysiology, Pacemaker Research and Management Center, Kocaeli University Medical Faculty, Kocaeli, Turkey
| | - Burak Acar
- Arrhythmia, Electrophysiology, Pacemaker Research and Management Center, Kocaeli University Medical Faculty, Kocaeli, Turkey.
| | - Aysen Agacdiken
- Arrhythmia, Electrophysiology, Pacemaker Research and Management Center, Kocaeli University Medical Faculty, Kocaeli, Turkey
| | - Ahmet Vural
- Arrhythmia, Electrophysiology, Pacemaker Research and Management Center, Kocaeli University Medical Faculty, Kocaeli, Turkey
| |
Collapse
|
4
|
Vichare R, Saleem F, Mansour H, Bojkovic K, Cheng F, Biswal M, Panguluri SK. Impact of age and sex on hyperoxia-induced cardiovascular pathophysiology. Mech Ageing Dev 2022; 208:111727. [PMID: 36075315 DOI: 10.1016/j.mad.2022.111727] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2022] [Revised: 08/17/2022] [Accepted: 08/23/2022] [Indexed: 12/30/2022]
Abstract
Hyperoxia is characterized by pronounced inflammatory responses, pulmonary cell apoptosis, and adverse cardiac remodeling due to an excess supply of oxygen. Hyperoxic episodes are frequent in mechanically ventilated patients and are associated with in-hospital mortality. This study extends the analysis of prior published research by our group as it investigates the influence of age in male and female rodents exposed to hyperoxic conditions. Age is an independent cardiovascular risk factor, often compounded by variables like obesity, diabetes, and a decline in sex hormones and their receptors. This study simulates clinical hyperoxia by subjecting rodents to > 90 % of oxygen for 72 h and compares the changes in cardiac structural and functional parameters with those exposed to normal air. While in both sexes conduction abnormalities with ageing were discernible, aged females owing to their inherent higher baseline QTc, were at a higher risk of developing arrhythmias as compared to age-matched males. Quantitative real-time RT-PCR and western blot analysis reflected altered expression of cardiac potassium channels, resulting in conduction abnormalities in aged female rodents. Unaffected by age and sex, hyperoxia-treated mice had altered body composition, as evidenced by a considerable reduction in body weight. Interestingly, compensatory hypertrophy observed as a protective mechanism in young males was absent in aged males, whereas protection of hearts from hyperoxia-induced cardiac hypertrophy was absent in aged female mice, both of which may be at least in part due to a reduction in sex steroid receptors and the systemic steroid levels. Finally, statistical analysis revealed that hyperoxia had the greatest impact on most of the cardiac parameters, followed by age and then sex. This data established an imperative finding that can change the provision of care for aged individuals admitted to ICU by elucidating the impact of intrinsic aging on hyperoxia-induced cardiac remodeling.
Collapse
Affiliation(s)
- Riddhi Vichare
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA
| | - Faizan Saleem
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA
| | - Hussein Mansour
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA
| | - Katarina Bojkovic
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA
| | - Feng Cheng
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA
| | - Manas Biswal
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA; College of Medicine Internal Medicine, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA
| | - Siva Kumar Panguluri
- Department of Pharmaceutical Sciences, Taneja College of Pharmacy, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA; Cell Biology, Microbiology and Molecular Biology, College of Arts and Sciences, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA.
| |
Collapse
|
5
|
Perry AS, Li S. Association of change in QRS duration with chronic heart failure outcomes. Am Heart J 2022; 254:77-80. [PMID: 36002049 DOI: 10.1016/j.ahj.2022.08.005] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/17/2022] [Revised: 07/26/2022] [Accepted: 08/17/2022] [Indexed: 06/15/2023]
Abstract
QRS duration is an established risk factor among patients with heart failure. How change in QRS duration relates to heart failure outcomes has had limited study. In this post-hoc analysis of the Beta-Blocker Evaluation of Survival Trial, we demonstrated that QRS duration change from baseline to 3 months is independently associated with long-term survival and left ventricle ejection fraction.
Collapse
Affiliation(s)
- Andrew S Perry
- Division of Cardiology, University of Washington School of Medicine, Seattle, WA
| | - Song Li
- Division of Cardiology, University of Washington School of Medicine, Seattle, WA.
| |
Collapse
|
6
|
de Oliveira Figueiredo MJ, Diez GR, Nava-Townsend S, Alfie A, Hardy C, Di Toro D, Márquez MF. Highlights and comments on EHRA/HRS/APHRS/LAHRS expert consensus on risk assessment in cardiac arrhythmias: use the right tool for the right outcome. J Interv Card Electrophysiol 2021; 62:461-467. [PMID: 33876382 PMCID: PMC8055211 DOI: 10.1007/s10840-021-00996-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/22/2021] [Accepted: 04/07/2021] [Indexed: 11/29/2022]
Abstract
Patients with heart disease, or at high risk of developing a cardiac condition, usually undergo risk assessment by primary care physicians, internal medicine doctors, or cardiologists. There are several methods that can be used for this risk assessment, and their applicability differs with respect to availability, complexity, and usefulness in different geographic populations. This document focuses on some of the many relevant clinical topics recently presented in the "Expert Consensus on Risk Assessment in Cardiac Arrhythmias: Use the Right Tool for the Right Outcome," which include statements based on the best available evidence. In this review, we want to highlight and make some pertinent comments on some of the most relevant points of this Consensus.
Collapse
Affiliation(s)
| | - Gerardo Rodriguez Diez
- Department of Electrophysiology and Hemodynamic, Arrhytmias Unity, CMN 20 de Noviembre, ISSSTE, Mexico City, Mexico
| | - Santiago Nava-Townsend
- Department of Electrocardiology, National Institute of Cardiology "Ignacio Chavez", Mexico City, Mexico
| | - Alberto Alfie
- Division of Electrophysiology, Instituto Cardiovascular AdventistaClinica Bazterrica, Buenos Aires, Argentina
| | - Carina Hardy
- Arrhythmia Unit, Heart Institute, University of São Paulo Medical School, Instituto do Coração -InCor, São Paulo, Brazil
| | - Dario Di Toro
- Department of Cardiology, Division of Electrophysiology, Argerich Hospital and CEMIC, Buenos Aires, Argentina
| | - Manlio F Márquez
- Department of Electrocardiology, National Institute of Cardiology "Ignacio Chavez", Mexico City, Mexico
| |
Collapse
|
7
|
Marquis-Gravel G, Raymond-Paquin A, McIntyre WF. The Prognostic Value of Intraventricular Conduction Disturbances: A Matter of Time? Can J Cardiol 2020; 36:1196-1198. [PMID: 32553813 DOI: 10.1016/j.cjca.2020.01.014] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/08/2020] [Accepted: 01/08/2020] [Indexed: 11/29/2022] Open
Affiliation(s)
- Guillaume Marquis-Gravel
- Montreal Heart Institute, Department of Medicine, Université de Montréal, Montréal, Québec, Canada; Duke Clinical Research Institute, Durham, North Carolina, USA.
| | - Alexandre Raymond-Paquin
- Montreal Heart Institute, Department of Medicine, Université de Montréal, Montréal, Québec, Canada
| | - William F McIntyre
- Population Health Research Institute, McMaster University, Hamilton, Ontario, Canada
| |
Collapse
|