Unsupervised clustering reveals phenotypes of AKI in ICU COVID-19 patients.
AKI
COVID-19
clustering
critical care
machine learning
Journal
Frontiers in medicine
ISSN: 2296-858X
Titre abrégé: Front Med (Lausanne)
Pays: Switzerland
ID NLM: 101648047
Informations de publication
Date de publication:
2022
2022
Historique:
received:
28
06
2022
accepted:
20
09
2022
entrez:
24
10
2022
pubmed:
25
10
2022
medline:
25
10
2022
Statut:
epublish
Résumé
Acute Kidney Injury (AKI) is a very frequent condition, occurring in about one in three patients admitted to an intensive care unit (ICU). AKI is a syndrome defined as a sudden decrease in glomerular filtration rate. However, this unified definition does not reflect the various mechanisms involved in AKI pathophysiology, each with its own characteristics and sensitivity to therapy. In this study, we aimed at developing an innovative machine learning based method able to subphenotype AKI according to its pattern of risk factors. We adopted a three-step pipeline of analyses. First, we looked for factors associated with AKI using a generalized additive model. Second, we calculated the importance of each identified AKI related factor in the estimated AKI risk to find the main risk factor for AKI, at the single patient level. Lastly, we clusterized AKI patients according to their profile of risk factors and compared the clinical characteristics and outcome of every cluster. We applied this method to a cohort of severe COVID-19 patients hospitalized in the ICU of the Geneva University Hospitals. Among the 248 patients analyzed, we found 7 factors associated with AKI development. Using the individual expression of these factors, we identified three groups of AKI patients, based on the use of Lopinavir/Ritonavir, baseline eGFR, use of dexamethasone and AKI severity. The three clusters expressed distinct characteristics in terms of AKI severity and recovery, metabolic patterns and hospital mortality. We propose here a new method to phenotype AKI patients according to their most important individual risk factors for AKI development. When applied to an ICU cohort of COVID-19 patients, we were able to differentiate three groups of patients. Each expressed specific AKI characteristics and outcomes, which probably reflect a distinct pathophysiology.
Sections du résumé
Background
UNASSIGNED
Acute Kidney Injury (AKI) is a very frequent condition, occurring in about one in three patients admitted to an intensive care unit (ICU). AKI is a syndrome defined as a sudden decrease in glomerular filtration rate. However, this unified definition does not reflect the various mechanisms involved in AKI pathophysiology, each with its own characteristics and sensitivity to therapy. In this study, we aimed at developing an innovative machine learning based method able to subphenotype AKI according to its pattern of risk factors.
Methods
UNASSIGNED
We adopted a three-step pipeline of analyses. First, we looked for factors associated with AKI using a generalized additive model. Second, we calculated the importance of each identified AKI related factor in the estimated AKI risk to find the main risk factor for AKI, at the single patient level. Lastly, we clusterized AKI patients according to their profile of risk factors and compared the clinical characteristics and outcome of every cluster. We applied this method to a cohort of severe COVID-19 patients hospitalized in the ICU of the Geneva University Hospitals.
Results
UNASSIGNED
Among the 248 patients analyzed, we found 7 factors associated with AKI development. Using the individual expression of these factors, we identified three groups of AKI patients, based on the use of Lopinavir/Ritonavir, baseline eGFR, use of dexamethasone and AKI severity. The three clusters expressed distinct characteristics in terms of AKI severity and recovery, metabolic patterns and hospital mortality.
Conclusion
UNASSIGNED
We propose here a new method to phenotype AKI patients according to their most important individual risk factors for AKI development. When applied to an ICU cohort of COVID-19 patients, we were able to differentiate three groups of patients. Each expressed specific AKI characteristics and outcomes, which probably reflect a distinct pathophysiology.
Identifiants
pubmed: 36275817
doi: 10.3389/fmed.2022.980160
pmc: PMC9579431
doi:
Types de publication
Journal Article
Langues
eng
Pagination
980160Commentaires et corrections
Type : ErratumIn
Informations de copyright
Copyright © 2022 Legouis, Criton, Assouline, Le Terrier, Sgardello, Pugin, Marchi and Sangla.
Déclaration de conflit d'intérêts
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Références
PLoS One. 2021 May 11;16(5):e0249760
pubmed: 33974624
Lancet. 2021 Apr 3;397(10281):1293-1300
pubmed: 33812488
J Clin Med. 2021 Feb 25;10(5):
pubmed: 33668833
BMC Nephrol. 2017 Feb 8;18(1):55
pubmed: 28178929
Kidney Int Rep. 2020 Oct;5(10):1787-1790
pubmed: 32838087
Best Pract Res Clin Anaesthesiol. 2004 Mar;18(1):189-203
pubmed: 14760882
N Engl J Med. 2018 Oct 11;379(15):1431-1442
pubmed: 30304656
Crit Care Med. 2000 Nov;28(11):3697-704
pubmed: 11098976
Front Med (Lausanne). 2021 Dec 02;8:779627
pubmed: 34926523
Crit Care. 2020 Apr 15;24(1):150
pubmed: 32295614
JAMA Netw Open. 2019 Nov 1;2(11):e1916021
pubmed: 31755952
Curr Opin Crit Care. 2020 Dec;26(6):519-524
pubmed: 33044239
J Anesth. 1988 Sep 1;2(2):119-23
pubmed: 15236068
Am J Respir Crit Care Med. 2019 Apr 1;199(7):863-872
pubmed: 30334632
Front Med (Lausanne). 2021 Dec 06;8:719472
pubmed: 34938742
Br J Anaesth. 2018 Nov;121(5):1025-1033
pubmed: 30336846
Crit Care. 2016 Nov 17;20(1):372
pubmed: 27852290
BMC Nephrol. 2020 Jun 11;21(1):223
pubmed: 32527305
Clin J Am Soc Nephrol. 2017 Jun 7;12(6):874-884
pubmed: 28495862
Tohoku J Exp Med. 2015;237(4):287-95
pubmed: 26607258
Crit Care. 2021 Jul 16;25(1):249
pubmed: 34271956
J Med Virol. 2021 Jul;93(7):4411-4419
pubmed: 33792956
Commun Biol. 2022 Jul 19;5(1):719
pubmed: 35853932
Intensive Care Med. 2015 Aug;41(8):1411-23
pubmed: 26162677
Front Big Data. 2021 Jul 08;4:693674
pubmed: 34308343
J Am Soc Nephrol. 2022 Apr;33(4):810-827
pubmed: 35273087
Proc Soc Exp Biol Med. 1979 Sep;161(4):574-8
pubmed: 384401
Front Genet. 2020 Dec 14;11:620143
pubmed: 33381156
Anesthesiology. 1983 Feb;58(2):136-41
pubmed: 6337527
Intensive Care Med. 2013 Mar;39(3):420-8
pubmed: 23291734
Clin J Am Soc Nephrol. 2020 Nov 6;15(11):1557-1565
pubmed: 33033164
Intensive Care Med. 2019 Jul;45(7):1025-1028
pubmed: 31062051
Circ Cardiovasc Qual Outcomes. 2021 Sep;14(9):e007071
pubmed: 34517728
N Engl J Med. 2020 Jul 16;383(3):240-251
pubmed: 32668114
Sci Rep. 2020 Jul 23;10(1):12316
pubmed: 32704054
N Engl J Med. 2016 Jul 14;375(2):122-33
pubmed: 27181456
Anesthesiology. 1974 Nov;41(5):452-61
pubmed: 4611279
Source Code Biol Med. 2008 Dec 16;3:17
pubmed: 19087314
Sci Rep. 2019 Jan 28;9(1):797
pubmed: 30692568
Ann Intensive Care. 2020 Oct 6;10(1):131
pubmed: 33025225
J Am Soc Nephrol. 2012 Jan;23(1):13-21
pubmed: 22021710
Kidney Int. 2020 Jul;98(1):209-218
pubmed: 32416116
Anesthesiology. 1988 Dec;69(6):862-9
pubmed: 2973759
Brain. 2017 Jul 1;140(7):1959-1976
pubmed: 28549077
PLoS One. 2019 Apr 4;14(4):e0214904
pubmed: 30947282
PLoS One. 2021 May 25;16(5):e0251048
pubmed: 34033655
Clin J Am Soc Nephrol. 2020 Sep 7;15(9):1358-1360
pubmed: 32238366
Nat Metab. 2020 Aug;2(8):732-743
pubmed: 32694833
Am J Respir Crit Care Med. 2017 Mar 15;195(6):784-791
pubmed: 27635668
Chest. 1995 Apr;107(4):1095-100
pubmed: 7705122