Real-time prediction of intradialytic hypotension using machine learning and cloud computing infrastructure.
end-stage kidney disease
intradialytic hypotension
machine learning
real-time prediction
Journal
Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association
ISSN: 1460-2385
Titre abrégé: Nephrol Dial Transplant
Pays: England
ID NLM: 8706402
Informations de publication
Date de publication:
30 Jun 2023
30 Jun 2023
Historique:
received:
15
12
2022
medline:
3
7
2023
pubmed:
14
4
2023
entrez:
13
4
2023
Statut:
ppublish
Résumé
In maintenance hemodialysis patients, intradialytic hypotension (IDH) is a frequent complication that has been associated with poor clinical outcomes. Prediction of IDH may facilitate timely interventions and eventually reduce IDH rates. We developed a machine learning model to predict IDH in in-center hemodialysis patients 15-75 min in advance. IDH was defined as systolic blood pressure (SBP) <90 mmHg. Demographic, clinical, treatment-related and laboratory data were retrieved from electronic health records and merged with intradialytic machine data that were sent in real-time to the cloud. For model development, dialysis sessions were randomly split into training (80%) and testing (20%) sets. The area under the receiver operating characteristic curve (AUROC) was used as a measure of the model's predictive performance. We utilized data from 693 patients who contributed 42 656 hemodialysis sessions and 355 693 intradialytic SBP measurements. IDH occurred in 16.2% of hemodialysis treatments. Our model predicted IDH 15-75 min in advance with an AUROC of 0.89. Top IDH predictors were the most recent intradialytic SBP and IDH rate, as well as mean nadir SBP of the previous 10 dialysis sessions. Real-time prediction of IDH during an ongoing hemodialysis session is feasible and has a clinically actionable predictive performance. If and to what degree this predictive information facilitates the timely deployment of preventive interventions and translates into lower IDH rates and improved patient outcomes warrants prospective studies.
Sections du résumé
BACKGROUND
BACKGROUND
In maintenance hemodialysis patients, intradialytic hypotension (IDH) is a frequent complication that has been associated with poor clinical outcomes. Prediction of IDH may facilitate timely interventions and eventually reduce IDH rates.
METHODS
METHODS
We developed a machine learning model to predict IDH in in-center hemodialysis patients 15-75 min in advance. IDH was defined as systolic blood pressure (SBP) <90 mmHg. Demographic, clinical, treatment-related and laboratory data were retrieved from electronic health records and merged with intradialytic machine data that were sent in real-time to the cloud. For model development, dialysis sessions were randomly split into training (80%) and testing (20%) sets. The area under the receiver operating characteristic curve (AUROC) was used as a measure of the model's predictive performance.
RESULTS
RESULTS
We utilized data from 693 patients who contributed 42 656 hemodialysis sessions and 355 693 intradialytic SBP measurements. IDH occurred in 16.2% of hemodialysis treatments. Our model predicted IDH 15-75 min in advance with an AUROC of 0.89. Top IDH predictors were the most recent intradialytic SBP and IDH rate, as well as mean nadir SBP of the previous 10 dialysis sessions.
CONCLUSIONS
CONCLUSIONS
Real-time prediction of IDH during an ongoing hemodialysis session is feasible and has a clinically actionable predictive performance. If and to what degree this predictive information facilitates the timely deployment of preventive interventions and translates into lower IDH rates and improved patient outcomes warrants prospective studies.
Identifiants
pubmed: 37055366
pii: 7117966
doi: 10.1093/ndt/gfad070
pmc: PMC10310501
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1761-1769Informations de copyright
© The Author(s) 2023. Published by Oxford University Press on behalf of the ERA.
Références
Nat Mach Intell. 2020 Jan;2(1):56-67
pubmed: 32607472
Curr Opin Nephrol Hypertens. 2016 Nov;25(6):545-550
pubmed: 27606498
Am J Nephrol. 2019;49(6):497-506
pubmed: 31129661
Clin Kidney J. 2020 Jul 08;13(6):981-993
pubmed: 33391741
Clin J Am Soc Nephrol. 2014 Dec 5;9(12):2124-32
pubmed: 25376764
J Biomed Inform. 2019 Oct;98:103271
pubmed: 31454648
Nephrol Dial Transplant. 2015 Dec;30(12):2075-9
pubmed: 26333543
Sensors (Basel). 2018 Aug 27;18(9):
pubmed: 30150592
Kidney Int. 2021 Jun;99(6):1408-1417
pubmed: 33607178
Am J Nephrol. 2013;38(5):388-96
pubmed: 24192428
Kidney Dis (Basel). 2019 Feb;5(1):28-33
pubmed: 30815462
J Am Soc Nephrol. 2015 Mar;26(3):724-34
pubmed: 25270068
Hemodial Int. 2014 Apr;18(2):415-22
pubmed: 24467830
Kidney Int. 2002 Sep;62(3):1046-53
pubmed: 12164889
Semin Dial. 2016 Jul;29(4):295-6
pubmed: 27082064
PLoS One. 2017 Oct 19;12(10):e0185846
pubmed: 29049308
BMC Nephrol. 2021 Aug 9;22(1):274
pubmed: 34372809
Clin J Am Soc Nephrol. 2021 Mar 8;16(3):396-406
pubmed: 33574056
Nephrol Dial Transplant. 2018 Jan 1;33(1):149-159
pubmed: 28444336
J Formos Med Assoc. 2018 Oct;117(10):888-893
pubmed: 29941330
Nephrol Dial Transplant. 2011 Aug;26(8):2656-63
pubmed: 21212166
Kidney Int. 2013 Oct;84(4):795-802
pubmed: 23783241