Machine Learning Algorithm Predicts Cardiac Resynchronization Therapy Outcomes: Lessons From the COMPANION Trial

Cardiac resynchronization therapy (CRT) reduces morbidity and mortality in heart failure patients with reduced left ventricular function and intraventricular conduction delay. However, individual outcomes vary significantly. This study sought to use a machine learning algorithm to develop a model to...

Full description

Saved in:

Bibliographic Details
Published in:	Circulation. Arrhythmia and electrophysiology Vol. 11; no. 1; p. e005499
Main Authors:	Kalscheur, Matthew M, Kipp, Ryan T, Tattersall, Matthew C, Mei, Chaoqun, Buhr, Kevin A, DeMets, David L, Field, Michael E, Eckhardt, Lee L, Page, C David
Format:	Journal Article
Language:	English
Published:	United States 01.01.2018
Subjects:	Aged Algorithms Cardiac Resynchronization Therapy - methods Decision Making Deep Learning Female Heart Conduction System - physiopathology Heart Failure - physiopathology Heart Failure - therapy Humans Machine Learning Male Middle Aged Predictive Value of Tests Stroke Volume - physiology Ventricular Function, Left - physiology algorithms heart failure hospitalization machine learning cardiac resynchronization therapy
ISSN:	1941-3084, 1941-3084
Online Access:	Get more information
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Cardiac resynchronization therapy (CRT) reduces morbidity and mortality in heart failure patients with reduced left ventricular function and intraventricular conduction delay. However, individual outcomes vary significantly. This study sought to use a machine learning algorithm to develop a model to predict outcomes after CRT. Models were developed with machine learning algorithms to predict all-cause mortality or heart failure hospitalization at 12 months post-CRT in the COMPANION trial (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure). The best performing model was developed with the random forest algorithm. The ability of this model to predict all-cause mortality or heart failure hospitalization and all-cause mortality alone was compared with discrimination obtained using a combination of bundle branch block morphology and QRS duration. In the 595 patients with CRT-defibrillator in the COMPANION trial, 105 deaths occurred (median follow-up, 15.7 months). The survival difference across subgroups differentiated by bundle branch block morphology and QRS duration did not reach significance ( =0.08). The random forest model produced quartiles of patients with an 8-fold difference in survival between those with the highest and lowest predicted probability for events (hazard ratio, 7.96; <0.0001). The model also discriminated the risk of the composite end point of all-cause mortality or heart failure hospitalization better than subgroups based on bundle branch block morphology and QRS duration. In the COMPANION trial, a machine learning algorithm produced a model that predicted clinical outcomes after CRT. Applied before device implant, this model may better differentiate outcomes over current clinical discriminators and improve shared decision-making with patients.
AbstractList	Cardiac resynchronization therapy (CRT) reduces morbidity and mortality in heart failure patients with reduced left ventricular function and intraventricular conduction delay. However, individual outcomes vary significantly. This study sought to use a machine learning algorithm to develop a model to predict outcomes after CRT.BACKGROUNDCardiac resynchronization therapy (CRT) reduces morbidity and mortality in heart failure patients with reduced left ventricular function and intraventricular conduction delay. However, individual outcomes vary significantly. This study sought to use a machine learning algorithm to develop a model to predict outcomes after CRT.Models were developed with machine learning algorithms to predict all-cause mortality or heart failure hospitalization at 12 months post-CRT in the COMPANION trial (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure). The best performing model was developed with the random forest algorithm. The ability of this model to predict all-cause mortality or heart failure hospitalization and all-cause mortality alone was compared with discrimination obtained using a combination of bundle branch block morphology and QRS duration. In the 595 patients with CRT-defibrillator in the COMPANION trial, 105 deaths occurred (median follow-up, 15.7 months). The survival difference across subgroups differentiated by bundle branch block morphology and QRS duration did not reach significance (P=0.08). The random forest model produced quartiles of patients with an 8-fold difference in survival between those with the highest and lowest predicted probability for events (hazard ratio, 7.96; P<0.0001). The model also discriminated the risk of the composite end point of all-cause mortality or heart failure hospitalization better than subgroups based on bundle branch block morphology and QRS duration.METHODS AND RESULTSModels were developed with machine learning algorithms to predict all-cause mortality or heart failure hospitalization at 12 months post-CRT in the COMPANION trial (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure). The best performing model was developed with the random forest algorithm. The ability of this model to predict all-cause mortality or heart failure hospitalization and all-cause mortality alone was compared with discrimination obtained using a combination of bundle branch block morphology and QRS duration. In the 595 patients with CRT-defibrillator in the COMPANION trial, 105 deaths occurred (median follow-up, 15.7 months). The survival difference across subgroups differentiated by bundle branch block morphology and QRS duration did not reach significance (P=0.08). The random forest model produced quartiles of patients with an 8-fold difference in survival between those with the highest and lowest predicted probability for events (hazard ratio, 7.96; P<0.0001). The model also discriminated the risk of the composite end point of all-cause mortality or heart failure hospitalization better than subgroups based on bundle branch block morphology and QRS duration.In the COMPANION trial, a machine learning algorithm produced a model that predicted clinical outcomes after CRT. Applied before device implant, this model may better differentiate outcomes over current clinical discriminators and improve shared decision-making with patients.CONCLUSIONSIn the COMPANION trial, a machine learning algorithm produced a model that predicted clinical outcomes after CRT. Applied before device implant, this model may better differentiate outcomes over current clinical discriminators and improve shared decision-making with patients. Cardiac resynchronization therapy (CRT) reduces morbidity and mortality in heart failure patients with reduced left ventricular function and intraventricular conduction delay. However, individual outcomes vary significantly. This study sought to use a machine learning algorithm to develop a model to predict outcomes after CRT. Models were developed with machine learning algorithms to predict all-cause mortality or heart failure hospitalization at 12 months post-CRT in the COMPANION trial (Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure). The best performing model was developed with the random forest algorithm. The ability of this model to predict all-cause mortality or heart failure hospitalization and all-cause mortality alone was compared with discrimination obtained using a combination of bundle branch block morphology and QRS duration. In the 595 patients with CRT-defibrillator in the COMPANION trial, 105 deaths occurred (median follow-up, 15.7 months). The survival difference across subgroups differentiated by bundle branch block morphology and QRS duration did not reach significance ( =0.08). The random forest model produced quartiles of patients with an 8-fold difference in survival between those with the highest and lowest predicted probability for events (hazard ratio, 7.96; <0.0001). The model also discriminated the risk of the composite end point of all-cause mortality or heart failure hospitalization better than subgroups based on bundle branch block morphology and QRS duration. In the COMPANION trial, a machine learning algorithm produced a model that predicted clinical outcomes after CRT. Applied before device implant, this model may better differentiate outcomes over current clinical discriminators and improve shared decision-making with patients.
Author	Kalscheur, Matthew M Kipp, Ryan T Page, C David Mei, Chaoqun DeMets, David L Eckhardt, Lee L Buhr, Kevin A Field, Michael E Tattersall, Matthew C
Author_xml	– sequence: 1 givenname: Matthew M surname: Kalscheur fullname: Kalscheur, Matthew M email: mmkalsch@medicine.wisc.edu organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison. mmkalsch@medicine.wisc.edu – sequence: 2 givenname: Ryan T surname: Kipp fullname: Kipp, Ryan T organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 3 givenname: Matthew C surname: Tattersall fullname: Tattersall, Matthew C organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 4 givenname: Chaoqun surname: Mei fullname: Mei, Chaoqun organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 5 givenname: Kevin A surname: Buhr fullname: Buhr, Kevin A organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 6 givenname: David L surname: DeMets fullname: DeMets, David L organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 7 givenname: Michael E surname: Field fullname: Field, Michael E organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 8 givenname: Lee L surname: Eckhardt fullname: Eckhardt, Lee L organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison – sequence: 9 givenname: C David surname: Page fullname: Page, C David organization: From the Division of Cardiovascular Medicine, Department of Medicine, School of Medicine and Public Health (M.M.K., R.T.K., M.C.T., M.E.F., L.L.E.), Department of Biostatistics and Medical Informatics (C.M., K.A.B., D.L.D., C.D.P.), University of Wisconsin Institute for Clinical and Translational Research (C.M.), and Department of Computer Sciences (C.D.P.), University of Wisconsin-Madison
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/29326129$$D View this record in MEDLINE/PubMed
BookMark	eNpNkDtPwzAYRS1URB_wA1iQR5YU23nZbFXUQqW2qaoyR67ztTFK7GInQ_n1RKJITPcM957hjtHAWAMIPVIypTShL9lyl823PadTQuJIiBs0oiKiQUh4NPjHQzT2_pOQhHKa3KEhEyFLKBMjdF5LVWkDeAXSGW1OeFafrNNt1eCtg1Kr1uNMulJLhXfgL0ZVzhr9LVttDd5X4OT5gvOuVbYB_9p7vLfG44WzDW4rwFm-3s42y3yD907L-h7dHmXt4eGaE_SxmO-z92CVvy2z2SpQEaFxwFIm4zIkoMIQOE8gUYcYeMxpKlLVg1IHTtghjTgnYQzimHJWlmGc8LTflWyCnn-9Z2e_OvBt0WivoK6lAdv5ggou4oRFvWSCnq7V7tBAWZydbqS7FH8vsR_vy2tb
CitedBy_id	crossref_primary_10_1016_j_healun_2019_07_002 crossref_primary_10_1093_eurheartj_ehz565 crossref_primary_10_1161_CIRCEP_119_008210 crossref_primary_10_1016_j_intimp_2023_109879 crossref_primary_10_1038_s41598_024_51240_2 crossref_primary_10_1016_j_cvdhj_2022_09_001 crossref_primary_10_1016_j_jcmg_2019_03_009 crossref_primary_10_1161_CIRCRESAHA_120_317872 crossref_primary_10_3389_fphar_2021_759782 crossref_primary_10_1016_j_jchf_2022_06_011 crossref_primary_10_1186_s12911_022_02015_0 crossref_primary_10_1007_s00259_023_06259_4 crossref_primary_10_1038_s41598_021_03914_4 crossref_primary_10_1007_s00059_025_05298_x crossref_primary_10_1093_eurheartj_ehac617 crossref_primary_10_1097_CRD_0000000000000715 crossref_primary_10_1016_j_jacep_2021_08_003 crossref_primary_10_3390_jcm10235710 crossref_primary_10_1161_CIRCEP_117_006104 crossref_primary_10_3389_fcvm_2021_611055 crossref_primary_10_1136_bmjopen_2018_027688 crossref_primary_10_1007_s10741_023_10357_8 crossref_primary_10_1007_s12170_021_00684_6 crossref_primary_10_1007_s10278_025_01666_5 crossref_primary_10_1016_j_jaad_2019_10_060 crossref_primary_10_1016_j_jacep_2021_06_009 crossref_primary_10_1161_CIRCEP_119_007316 crossref_primary_10_1161_CIRCEP_119_007952 crossref_primary_10_15420_aer_2021_14 crossref_primary_10_1111_fcp_12747 crossref_primary_10_3389_fsurg_2022_935656 crossref_primary_10_1016_j_ebiom_2024_105447 crossref_primary_10_1007_s11936_023_01004_4 crossref_primary_10_1007_s12170_020_00649_1 crossref_primary_10_1093_eurheartj_ehz902 crossref_primary_10_1007_s00399_022_00837_z crossref_primary_10_3389_fdgth_2023_1201392 crossref_primary_10_3389_fcvm_2021_765693 crossref_primary_10_3390_diagnostics12030689 crossref_primary_10_3389_fphys_2023_1162520 crossref_primary_10_4274_atfm_galenos_2022_36449 crossref_primary_10_1002_ejhf_1333 crossref_primary_10_1016_j_ahj_2020_07_009 crossref_primary_10_1016_j_mayocp_2020_01_038 crossref_primary_10_1111_jce_15171 crossref_primary_10_1016_j_cjca_2021_07_016 crossref_primary_10_1016_j_nexres_2025_100492 crossref_primary_10_5662_wjm_v15_i4_106854 crossref_primary_10_1016_j_bbe_2021_05_002 crossref_primary_10_1093_eurheartj_ehab544 crossref_primary_10_1016_j_echo_2020_12_025 crossref_primary_10_1186_s13063_021_05489_x crossref_primary_10_1016_j_ihj_2022_07_004 crossref_primary_10_1097_MLR_0000000000001140 crossref_primary_10_1007_s40471_020_00259_w crossref_primary_10_3389_fcvm_2022_945726 crossref_primary_10_1016_j_rec_2019_05_014 crossref_primary_10_3389_fcvm_2021_736491 crossref_primary_10_1186_s12872_020_01502_4 crossref_primary_10_1016_j_jjcc_2021_11_017 crossref_primary_10_1007_s40119_022_00273_7 crossref_primary_10_3389_fphys_2021_753282 crossref_primary_10_1371_journal_pone_0222397 crossref_primary_10_1109_ACCESS_2023_3286346
ContentType	Journal Article
Copyright	2018 American Heart Association, Inc.
Copyright_xml	– notice: 2018 American Heart Association, Inc.
DBID	CGR CUY CVF ECM EIF NPM 7X8
DOI	10.1161/CIRCEP.117.005499
DatabaseName	Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic
DatabaseTitle	MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic MEDLINE
Database_xml	– sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	no_fulltext_linktorsrc
Discipline	Medicine
EISSN	1941-3084
ExternalDocumentID	29326129
Genre	Multicenter Study Randomized Controlled Trial Journal Article Research Support, N.I.H., Extramural
GrantInformation_xml	– fundername: NCATS NIH HHS grantid: UL1 TR000427
GroupedDBID	--- .XZ .Z2 18M 53G 5VS 6J9 AAAAV AAHPQ AAIQE AAJCS AARTV AASCR ABASU ABBUW ABDIG ABVCZ ABXVJ ABXYN ABZAD ABZZY ACDDN ACEWG ACGFO ACILI ACWDW ACWRI ACXJB ACXNZ ADBBV ADGGA ADHPY ADNKB AEBDS AEETU AFBFQ AFDTB AFEXH AFNMH AFUWQ AGINI AHQNM AHQVU AHRYX AHVBC AINUH AJCLO AJIOK AJNWD AJNYG AJZMW AKCTQ ALKUP ALMA_UNASSIGNED_HOLDINGS ALMTX AMJPA AMKUR AMNEI AOHHW AOQMC BAWUL BQLVK C45 CGR CS3 CUY CVF DIK DIWNM DUNZO E.X E3Z EBS ECM EEVPB EIF EJD EX3 F5P FCALG FL- GNXGY GQDEL H13 HLJTE IKREB IN~ IPNFZ KD2 KQ8 KQB L-C NPM ODMTH ODZKP OHYEH OK1 OPUJH OUVQU OVD OVDNE OXXIT RAH RIG RLZ S4S TEORI TR2 TSPGW V2I W2D W3M W8F WOW ZZMQN 7X8 ABPXF ADKSD
ID	FETCH-LOGICAL-c4015-272a5d30ec33e886e6cb5e8581797ce85ccb802b7488035e9f782dd356875d3d2
IEDL.DBID	7X8
ISICitedReferencesCount	68
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000422637000004&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1941-3084
IngestDate	Mon Sep 29 04:42:23 EDT 2025 Thu Apr 03 06:59:38 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	1
Keywords	algorithms heart failure hospitalization machine learning cardiac resynchronization therapy
Language	English
License	2018 American Heart Association, Inc.
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c4015-272a5d30ec33e886e6cb5e8581797ce85ccb802b7488035e9f782dd356875d3d2
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 ObjectType-Undefined-3
PMID	29326129
PQID	1989562480
PQPubID	23479
ParticipantIDs	proquest_miscellaneous_1989562480 pubmed_primary_29326129
PublicationCentury	2000
PublicationDate	2018-01-00 20180101
PublicationDateYYYYMMDD	2018-01-01
PublicationDate_xml	– month: 01 year: 2018 text: 2018-01-00
PublicationDecade	2010
PublicationPlace	United States
PublicationPlace_xml	– name: United States
PublicationTitle	Circulation. Arrhythmia and electrophysiology
PublicationTitleAlternate	Circ Arrhythm Electrophysiol
PublicationYear	2018
References	29326131 - Circ Arrhythm Electrophysiol. 2018 Jan;11(1):e006104
References_xml	– reference: 29326131 - Circ Arrhythm Electrophysiol. 2018 Jan;11(1):e006104
SSID	ssj0061816
Score	2.4667897
Snippet	Cardiac resynchronization therapy (CRT) reduces morbidity and mortality in heart failure patients with reduced left ventricular function and intraventricular...
SourceID	proquest pubmed
SourceType	Aggregation Database Index Database
StartPage	e005499
SubjectTerms	Aged Algorithms Cardiac Resynchronization Therapy - methods Decision Making Deep Learning Female Heart Conduction System - physiopathology Heart Failure - physiopathology Heart Failure - therapy Humans Machine Learning Male Middle Aged Predictive Value of Tests Stroke Volume - physiology Ventricular Function, Left - physiology
Title	Machine Learning Algorithm Predicts Cardiac Resynchronization Therapy Outcomes: Lessons From the COMPANION Trial
URI	https://www.ncbi.nlm.nih.gov/pubmed/29326129 https://www.proquest.com/docview/1989562480
Volume	11
WOSCitedRecordID	wos000422637000004&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings	1
isFullTextHit
isPrint
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LS8NAEF7Uinjx_agvVvAaTbN5bLxICS0KJg2lSm8h2Z1UwSa1SQX_vbNJqjcRvOS0GZbd2Zlv3oRcpZZ0WBq7mjAMRzMNPdY4aikNOmAnOnPtuOpb8PzoBAEfj92wcbgVTVrlUiZWglrmQvnIb1RuD-pqk-t3s3dNTY1S0dVmhMYqaTGEMiqlyxl_RxFs1F5VdZFrKh8LN5uoJoKcG-9h6PVCFbK8VqjF_QVhVpqmv_3fPe6QrQZj0m7NFLtkBbI9suE3UfR9MvOrDEqgTXPVCe2-TZBO-TKl4VwtKwvqVawj6BCKz0xULXTrik06qhsR0MGixG1BcYt08JKzgvbn-ZQioqTewA-7AYppOlIMfkCe-r2Rd681kxc0gfaWKlIzYksyHQRjwLkNtkgs4BZeo-uoQadCJFw3Ekc9f2aBmyLQkJJZNpo_kknjkKxleQbHhOopqmEFizqQovEo3RSQAqLAhMdMQqdNLpdnGSFnq3BFnEG-KKKf02yTo_pColndgiMyFOxEqHLyh79PySaiHF77Tc5IK8V3DedkXXyUr8X8omIZ_Aah_wWgXsrQ
linkProvider	ProQuest
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Machine+Learning+Algorithm+Predicts+Cardiac+Resynchronization+Therapy+Outcomes%3A+Lessons+From+the+COMPANION+Trial&rft.jtitle=Circulation.+Arrhythmia+and+electrophysiology&rft.au=Kalscheur%2C+Matthew+M&rft.au=Kipp%2C+Ryan+T&rft.au=Tattersall%2C+Matthew+C&rft.au=Mei%2C+Chaoqun&rft.date=2018-01-01&rft.eissn=1941-3084&rft.volume=11&rft.issue=1&rft.spage=e005499&rft_id=info:doi/10.1161%2FCIRCEP.117.005499&rft_id=info%3Apmid%2F29326129&rft_id=info%3Apmid%2F29326129&rft.externalDocID=29326129
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1941-3084&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1941-3084&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1941-3084&client=summon