Machine Learning Prediction Models for Mitral Valve Repairability and Mitral Regurgitation Recurrence in Patients Undergoing Surgical Mitral Valve Repair.
heart
machine learning
mitral valve prolapse
mitral valve repair
prediction model
primary mitral regurgitation
Journal
Bioengineering (Basel, Switzerland)
ISSN: 2306-5354
Titre abrégé: Bioengineering (Basel)
Pays: Switzerland
ID NLM: 101676056
Informations de publication
Date de publication:
25 Aug 2021
25 Aug 2021
Historique:
received:
18
06
2021
revised:
20
08
2021
accepted:
23
08
2021
entrez:
25
9
2021
pubmed:
26
9
2021
medline:
26
9
2021
Statut:
epublish
Résumé
Mitral valve regurgitation (MR) is the most common valvular heart disease and current variables associated with MR recurrence are still controversial. We aim to develop a machine learning-based prognostic model to predict causes of mitral valve (MV) repair failure and MR recurrence. 1000 patients who underwent MV repair at our institution between 2008 and 2018 were enrolled. Patients were followed longitudinally for up to three years. Clinical and echocardiographic data were included in the analysis. Endpoints were MV repair surgical failure with consequent MV replacement or moderate/severe MR (>2+) recurrence at one-month and moderate/severe MR recurrence after three years. 817 patients (DS1) had an echocardiographic examination at one-month while 295 (DS2) also had one at three years. Data were randomly divided into training (DS1: n = 654; DS2: n = 206) and validation (DS1: n = 164; DS2 n = 89) cohorts. For intra-operative or early MV repair failure assessment, the best area under the curve (AUC) was 0.75 and the complexity of mitral valve prolapse was the main predictor. In predicting moderate/severe recurrent MR at three years, the best AUC was 0.92 and residual MR at six months was the most important predictor. Machine learning algorithms may improve prognosis after MV repair procedure, thus improving indications for correct candidate selection for MV surgical repair.
Sections du résumé
BACKGROUND
BACKGROUND
Mitral valve regurgitation (MR) is the most common valvular heart disease and current variables associated with MR recurrence are still controversial. We aim to develop a machine learning-based prognostic model to predict causes of mitral valve (MV) repair failure and MR recurrence.
METHODS
METHODS
1000 patients who underwent MV repair at our institution between 2008 and 2018 were enrolled. Patients were followed longitudinally for up to three years. Clinical and echocardiographic data were included in the analysis. Endpoints were MV repair surgical failure with consequent MV replacement or moderate/severe MR (>2+) recurrence at one-month and moderate/severe MR recurrence after three years.
RESULTS
RESULTS
817 patients (DS1) had an echocardiographic examination at one-month while 295 (DS2) also had one at three years. Data were randomly divided into training (DS1: n = 654; DS2: n = 206) and validation (DS1: n = 164; DS2 n = 89) cohorts. For intra-operative or early MV repair failure assessment, the best area under the curve (AUC) was 0.75 and the complexity of mitral valve prolapse was the main predictor. In predicting moderate/severe recurrent MR at three years, the best AUC was 0.92 and residual MR at six months was the most important predictor.
CONCLUSIONS
CONCLUSIONS
Machine learning algorithms may improve prognosis after MV repair procedure, thus improving indications for correct candidate selection for MV surgical repair.
Identifiants
pubmed: 34562939
pii: bioengineering8090117
doi: 10.3390/bioengineering8090117
pmc: PMC8469985
pii:
doi:
Types de publication
Journal Article
Langues
eng
Références
J Thorac Cardiovasc Surg. 2018 Jul;156(1):122-129.e16
pubmed: 29709354
Am J Cardiol. 2010 Aug 1;106(3):395-401
pubmed: 20643253
Yearb Med Inform. 2019 Aug;28(1):16-26
pubmed: 31419814
Circulation. 2003 Apr 1;107(12):1609-13
pubmed: 12668494
J Am Coll Cardiol. 2017 May 30;69(21):2657-2664
pubmed: 28545640
Circ Res. 2017 Oct 13;121(9):1092-1101
pubmed: 28794054
Am Heart J. 2006 Jun;151(6):1325-33
pubmed: 16781250
J Cardiovasc Dev Dis. 2020 Oct 20;7(4):
pubmed: 33092178
J Gen Intern Med. 2018 Jun;33(6):921-928
pubmed: 29383551
Eur J Cardiothorac Surg. 2018 Jun 1;53(6):1244-1250
pubmed: 29309559
Glob Cardiol Sci Pract. 2017 Mar 31;2017(1):e201703
pubmed: 31139637
Interact Cardiovasc Thorac Surg. 2013 Jul;17(1):120-5
pubmed: 23587525
Lancet. 2006 Sep 16;368(9540):1005-11
pubmed: 16980116
Eur Heart J. 2019 Jul 14;40(27):2194-2202
pubmed: 31121021
J Thorac Dis. 2014 Mar;6 Suppl 1:S39-51
pubmed: 24672698
Eur J Cardiothorac Surg. 2012 Mar;41(3):518-24
pubmed: 22223695
Ann Thorac Surg. 2016 Nov;102(5):1459-1465
pubmed: 27720370
Eur Heart J. 2017 Sep 21;38(36):2739-2791
pubmed: 28886619
J Am Coll Cardiol. 2018 Jun 12;71(23):2668-2679
pubmed: 29880128
J Am Coll Cardiol. 2016 Feb 9;67(5):488-98
pubmed: 26846946
J Clin Med. 2019 Apr 17;8(4):
pubmed: 30999593
Lancet. 2009 Apr 18;373(9672):1382-94
pubmed: 19356795
J Thorac Cardiovasc Surg. 2008 Feb;135(2):274-82
pubmed: 18242250
J Am Coll Cardiol. 2006 Dec 19;48(12):2524-30
pubmed: 17174193
J Cardiovasc Dev Dis. 2021 Apr 16;8(4):
pubmed: 33923465
PLoS One. 2017 Apr 4;12(4):e0174944
pubmed: 28376093
J Thorac Cardiovasc Surg. 2014 Oct;148(4):1400-6
pubmed: 24589201
J Thorac Cardiovasc Surg. 2008 Apr;135(4):885-93, 893.e1-2
pubmed: 18374775
J Am Soc Echocardiogr. 2017 Apr;30(4):303-371
pubmed: 28314623
J Am Soc Echocardiogr. 1994 Jan-Feb;7(1):20-6
pubmed: 8155330
N Engl J Med. 1999 Jul 1;341(1):1-7
pubmed: 10387935
J Heart Valve Dis. 2004 Nov;13(6):914-20
pubmed: 15597581
Circulation. 2017 Jun 20;135(25):e1159-e1195
pubmed: 28298458
J Thorac Cardiovasc Surg. 1998 Nov;116(5):734-43
pubmed: 9806380
JAMA Netw Open. 2019 Oct 2;2(10):e1915997
pubmed: 31651973
Eur J Cardiothorac Surg. 2007 Aug;32(2):301-7
pubmed: 17561410
Circulation. 1991 Jul;84(1):23-34
pubmed: 2060099
Heart Lung Circ. 2019 Dec;28(12):1852-1865
pubmed: 30377076
J Cardiothorac Surg. 2019 Nov 27;14(1):205
pubmed: 31775821
J Thorac Cardiovasc Surg. 2007 Apr;133(4):995-1003
pubmed: 17382640