Predictive Approaches for Acute Dialysis Requirement and Death in COVID-19. Issue 8 (August 2021)

• Record Type: Journal Article
• Authors: Vaid, Akhil; Chan, Lili; Chaudhary, Kumardeep; Jaladanki, Suraj K.; Paranjpe, Ishan; Russak, Adam; Kia, Arash; Timsina, Prem; Levin, Matthew A.; He, John Cijiang; Böttinger, Erwin P.; Charney, Alexander W.; Fayad, Zahi A.; Coca, Steven G.; Glicksberg, Benjamin S.; Nadkarni, Girish N.
• Other Names: Charney Alex author non-byline.; Just Allan C. author non-byline.; Glicksberg Benjamin author non-byline.; Nadkarni Girish author non-byline.; Huckins Laura author non-byline.; O'Reilly Paul author non-byline.; Miotto Riccardo author non-byline.; Fayad Zahi author non-byline.; Russak Adam J. author non-byline.; Rahman Adeeb author non-byline.; Vaid Akhil author non-byline.; Le Dobbyn Amanda author non-byline.; Leader Andrew author non-byline.; Moscati Arden author non-byline.; Kapoor Arjun author non-byline.; Chang Christie author non-byline.; Bellaire Christopher author non-byline.; Carrion Daniel author non-byline.; Chaudhry Fayzan author non-byline.; Richter Felix author non-byline.; Soultanidis Georgios author non-byline.; Paranjpe Ishan author non-byline.; Nabeel Ismail author non-byline.; De Freitas Jessica author non-byline.; Xu Jiayi author non-byline.; Rush Johnathan author non-byline.; Johnson Kipp author non-byline.; Vemuri Krishna author non-byline.; Chaudhary Kumardeep author non-byline.; Lepow Lauren author non-byline.; Cotter Liam author non-byline.; Liharska Lora author non-byline.; Pereanez Marco author non-byline.; Bicak Mesude author non-byline.; DeFelice Nicholas author non-byline.; Naik Nidhi author non-byline.; Beckmann Noam author non-byline.; Nadukuru Rajiv author non-byline.; O'Hagan Ross author non-byline.; Zhao Shan author non-byline.; Somani Sulaiman author non-byline.; Van Vleck Tielman T. author non-byline.; Mutetwa Tinaye author non-byline.; Wanyan Tingyi author non-byline.; Fauveau Valentin author non-byline.; Yang Yang author non-byline.; Lavin Yonit author non-byline.; Lanksy Alona author non-byline.; Atreja Ashish author non-byline.; Del Valle Diane author non-byline.; Meyer Dara author non-byline.; Golden Eddye author non-byline.; Fasihuddin Farah author non-byline.; Hsun Wen Huei author non-byline.; Rogers Jason author non-byline.; Lilly Gutierrez Jennifer author non-byline.; Walker Laura author non-byline.; Singh Manbir author non-byline.; Danieletto Matteo author non-byline.; Nieves Melissa A. author non-byline.; Zweig Micol author non-byline.; Pyzik Renata author non-byline.; Fayad Rima author non-byline.; Glowe Patricia author non-byline.; Calorossi Sharlene author non-byline.; Kaur Sparshdeep author non-byline.; Ascolillo Steven author non-byline.; Roa Yovanna author non-byline.; Lala-Trindade Anuradha author non-byline.; Coca Steven G. author non-byline.; Percha Bethany author non-byline.; Sigel Keith author non-byline.; Polak Paz author non-byline.; Hirten Robert author non-byline.; Swartz Talia author non-byline.; Do Ron author non-byline.; Loos Ruth J. F. author non-byline.; Charney Dennis author non-byline.; Nestler Eric author non-byline.; Murphy Barbara author non-byline.; Reich David author non-byline.; Böttinger Erwin author non-byline.; Chatani Kumar author non-byline.; Martin Glenn author non-byline.; Nestler Eric author non-byline.; Kovatch Patricia author non-byline.; Finkelstein Joseph author non-byline.; Murphy Barbara author non-byline.; Buxbaum Joseph author non-byline.; Cho Judy author non-byline.; Kasarskis Andrew author non-byline.; Horowitz Carol author non-byline.; Cordon-Cardo Carlos author non-byline.; Sohn Monica author non-byline.; Martin Glenn author non-byline.; Garcia-Sastre Adolfo author non-byline.; Bagiella Emilia author non-byline.; Krammer Florian author non-byline.; Aberg Judith author non-byline.; Narula Jagat author non-byline.; Wright Robert author non-byline.; Lium Erik author non-byline.; Wright Rosalind author non-byline.; Gelijns Annetine author non-byline.; Fuster Valentin author non-byline.; Merad Miriam author non-byline.;
• Is Part Of: Clinical journal of the American Society of Nephrology. Volume 16:Issue 8(2021)
• Page Start: 1158
• Page End: 1168
• Publication Date: 2021-08
• Subjects: COVID-19 -- dialysis -- machine learning -- prediction -- AKI
• DOI: 10.2215/CJN.17311120
• Languages: English
• ISSNs: 1555-9041

Abstract

Visual Abstract: Abstract : Background and objectives: AKI treated with dialysis initiation is a common complication of coronavirus disease 2019 (COVID-19) among hospitalized patients. However, dialysis supplies and personnel are often limited. Design, setting, participants, & measurements: Using data from adult patients hospitalized with COVID-19 from five hospitals from the Mount Sinai Health System who were admitted between March 10 and December 26, 2020, we developed and validated several models (logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), random forest, and eXtreme GradientBoosting [XGBoost; with and without imputation]) for predicting treatment with dialysis or death at various time horizons (1, 3, 5, and 7 days) after hospital admission. Patients admitted to the Mount Sinai Hospital were used for internal validation, whereas the other hospitals formed part of the external validation cohort. Features included demographics, comorbidities, and laboratory and vital signs within 12 hours of hospital admission. Results: A total of 6093 patients (2442 in training and 3651 in external validation) were included in the final cohort. Of the different modeling approaches used, XGBoost without imputation had the highest area under the receiver operating characteristic (AUROC) curve on internal validation (range of 0.93–0.98) and area under the precision-recall curve (AUPRC; range of 0.78–0.82) for all time points. XGBoost without imputation also had the highest test parameters on external validation (AUROC range of 0.85–0.87, and AUPRC range of 0.27–0.54) across all time windows. XGBoost without imputation outperformed all models with higher precision and recall (mean difference in AUROC of 0.04; mean difference in AUPRC of 0.15). Features of creatinine, BUN, and red cell distribution width were major drivers of the model's prediction. Conclusions: An XGBoost model without imputation for prediction of a composite outcome of either death or dialysis in patients positive for COVID-19 had the best performance, as compared with standard and other machine learning models. Podcast: This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2021_07_09_CJN17311120.mp3