72-Hour Hemodynamic and Urine Output Trajectories Predict Renal Recovery in Continuous Kidney Replacement Therapy: A Machine Learning Approach

72-Hour Hemodynamic and Urine Output Trajectories Predict Renal Recovery in Continuous Kidney Replacement Therapy: A Machine Learning Approach | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article 72-Hour Hemodynamic and Urine Output Trajectories Predict Renal Recovery in Continuous Kidney Replacement Therapy: A Machine Learning Approach Daseul Huh, Juyeon Park, In Mee Han, Youn Kyung Kee, Hee Jung Jeon, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9073059/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Background: Predicting renal recovery in patients receiving continuous kidney replacement therapy (CKRT) remains challenging. Conventional single-time-point hemodynamic assessments at CKRT initiation have shown limited prognostic value. We investigated whether 72-hour hemodynamic and urine output (UO) trajectories, analyzed through machine learning, could improve prediction of 28-day renal recovery. Methods: This single-center retrospective cohort study included 331 CKRT episodes (318 patients). Renal recovery was defined as successful CKRT discontinuation with no kidney replacement therapy for at least 72 hours within 28 days of initiation. Phase 1 employed Cox proportional hazards regression to evaluate day-0 variables. Phase 2 developed machine learning models (logistic regression, XGBoost, bidirectional LSTM) using 105 engineered features from 72-hour MAP, SBP, DBP, heart rate (HR), and UO time-series data. Model performance was assessed using 5-fold stratified cross-validation with bootstrap 95% confidence intervals. SHAP analysis was used for model interpretability. Results: Of 331 episodes, 71 (21.5%) achieved renal recovery. In Phase 1, day-0 UO was non-significant in both univariable (HR=1.010 per 100 mL/day; p=0.146) and multivariable Cox regression (HR=1.003 per 100 mL/day; p=0.703), even when forced into a model with clinically important covariates. The multivariable model (C-index=0.768) identified CCI (HR=0.749; p<0.001) and SOFA (HR=0.805; p<0.001) as the strongest independent predictors. In Phase 2, incorporating 72-hour trajectories significantly improved prediction: XGBoost Combined (day-0 + 72h features) achieved an AUC of 0.848 (95% CI, 0.765–0.890), compared with 0.762 for baseline-only XGBoost (AUC difference=+0.086, DeLong p=0.035). SHAP analysis revealed late-phase (36–72h) UO variability (UO_late_std) as the dominant predictor (mean |SHAP|=1.311), exceeding the second-ranked feature by four-fold. The 72-hour UO trajectory features occupied two of the top three positions in feature importance. Conclusions: Day-0 hemodynamic variables fail to predict renal recovery in CKRT patients when assessed as single-time-point measurements. However, 72-hour trajectories—particularly late-phase UO variability—are powerful predictors when captured through machine learning. These findings support integrating temporal hemodynamic patterns into clinical decision-making for CKRT patients. acute kidney injury continuous kidney replacement therapy renal recovery machine learning XGBoost time-series analysis urine output trajectory SHAP prediction model Full Text Additional Declarations No competing interests reported. Supplementary Files Additionalfile1.docx FigureS1AUCboxplot.tiff FigureS2Calibration.tiff FigureS3ROCwithRF.tiff Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 29 Apr, 2026 Reviewers agreed at journal 18 Apr, 2026 Reviewers invited by journal 19 Mar, 2026 Editor assigned by journal 11 Mar, 2026 Submission checks completed at journal 11 Mar, 2026 First submitted to journal 09 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9073059","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":609170602,"identity":"34ae1aee-e2ef-4834-a95a-a9d362421af5","order_by":0,"name":"Daseul Huh","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Daseul","middleName":"","lastName":"Huh","suffix":""},{"id":609170604,"identity":"eda09c53-45d0-4f22-8662-d73a39c7f5db","order_by":1,"name":"Juyeon Park","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Juyeon","middleName":"","lastName":"Park","suffix":""},{"id":609170605,"identity":"97e61ca6-a884-4d18-9a98-cf96d7a3da0d","order_by":2,"name":"In Mee Han","email":"","orcid":"","institution":"Yonsei Hanmaeum Internal Medicine Clinic","correspondingAuthor":false,"prefix":"","firstName":"In","middleName":"Mee","lastName":"Han","suffix":""},{"id":609170606,"identity":"eb0db039-8508-4a77-b754-5e845b6120fb","order_by":3,"name":"Youn Kyung Kee","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Youn","middleName":"Kyung","lastName":"Kee","suffix":""},{"id":609170607,"identity":"f0cb124b-5c8c-46ef-8837-06133a08f1fe","order_by":4,"name":"Hee Jung Jeon","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hee","middleName":"Jung","lastName":"Jeon","suffix":""},{"id":609170608,"identity":"51f32f09-1bb3-4583-bc11-6de07eb7fa72","order_by":5,"name":"Jieun Oh","email":"","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jieun","middleName":"","lastName":"Oh","suffix":""},{"id":609170609,"identity":"190db9cd-6891-4f47-9b1d-06b7f8beefd8","order_by":6,"name":"Jung Tak Park","email":"","orcid":"","institution":"Yonsei University College of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jung","middleName":"Tak","lastName":"Park","suffix":""},{"id":609170610,"identity":"a66c7984-511f-4d22-a4ce-cfe226f178da","order_by":7,"name":"Dong Ho Shin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIiWNgGAWjYBACNjBZ8Z9Hnr0ByDCwIFbLGWYZw54DIC0SRFrF2MZsw3AjAcQkQgsf/+GHH362sfEwznx+dcOPAgkG/vbuBPwOk0gzluw5x8PDLp1TdrMH6DCJM2c3ENDCYCDNUCbBwzg7J+0GD1CLgUQuAS38xz//ZmAz4GG4eSbt5h+itDDkmEkztCXwMNxgP3abOFskcsose84c4DHsyWG7LWMgwUPQL/L9xzff+FFxwF6e/fizm2/+2Mjxt/fi14IEeAzAJLHKQYD9ASmqR8EoGAWjYAQBANdjQiEbTvN8AAAAAElFTkSuQmCC","orcid":"","institution":"Hallym University College of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Dong","middleName":"Ho","lastName":"Shin","suffix":""}],"badges":[],"createdAt":"2026-03-09 12:40:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9073059/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9073059/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105568787,"identity":"3507a554-d398-4c71-b365-0931335948c2","added_by":"auto","created_at":"2026-03-27 13:10:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":882000,"visible":true,"origin":"","legend":"","description":"","filename":"CKRTRenalRecoveryManuscriptv7CCformat.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9073059/v1_covered_4917a140-73fb-4696-9651-7ee148552b16.pdf"},{"id":105175716,"identity":"11e623a2-525c-413d-a015-91d5c4cc2c56","added_by":"auto","created_at":"2026-03-23 05:51:42","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":41559,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-9073059/v1/d45e8340603f606acc7e3ae4.docx"},{"id":105563955,"identity":"da42bb6b-562e-4b7b-9b6f-c98970569e34","added_by":"auto","created_at":"2026-03-27 12:48:17","extension":"tiff","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":312872,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1AUCboxplot.tiff","url":"https://assets-eu.researchsquare.com/files/rs-9073059/v1/60ead37c66fc692852200ff3.tiff"},{"id":105175713,"identity":"fd1b9741-e436-4579-b555-635f9b7cf376","added_by":"auto","created_at":"2026-03-23 05:51:42","extension":"tiff","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":411532,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS2Calibration.tiff","url":"https://assets-eu.researchsquare.com/files/rs-9073059/v1/197bda9bd65a52043d3963c0.tiff"},{"id":105175715,"identity":"295cb700-dd90-424c-a247-8a90c05bd5c4","added_by":"auto","created_at":"2026-03-23 05:51:42","extension":"tiff","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":422752,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3ROCwithRF.tiff","url":"https://assets-eu.researchsquare.com/files/rs-9073059/v1/d1c28b5968ca823ae9c28918.tiff"}],"financialInterests":"No competing interests reported.","formattedTitle":"72-Hour Hemodynamic and Urine Output Trajectories Predict Renal Recovery in Continuous Kidney Replacement Therapy: A Machine Learning Approach","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"critical-care","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cric","sideBox":"Learn more about [Critical Care](http://ccforum.biomedcentral.com/)","snPcode":"13054","submissionUrl":"https://submission.nature.com/new-submission/13054/3","title":"Critical Care","twitterHandle":"@Crit_Care","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"acute kidney injury, continuous kidney replacement therapy, renal recovery, machine learning, XGBoost, time-series analysis, urine output trajectory, SHAP, prediction model","lastPublishedDoi":"10.21203/rs.3.rs-9073059/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9073059/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003ePredicting renal recovery in patients receiving continuous kidney replacement therapy (CKRT) remains challenging. Conventional single-time-point hemodynamic assessments at CKRT initiation have shown limited prognostic value. We investigated whether 72-hour hemodynamic and urine output (UO) trajectories, analyzed through machine learning, could improve prediction of 28-day renal recovery.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eThis single-center retrospective cohort study included 331 CKRT episodes (318 patients). Renal recovery was defined as successful CKRT discontinuation with no kidney replacement therapy for at least 72 hours within 28 days of initiation. Phase 1 employed Cox proportional hazards regression to evaluate day-0 variables. Phase 2 developed machine learning models (logistic regression, XGBoost, bidirectional LSTM) using 105 engineered features from 72-hour MAP, SBP, DBP, heart rate (HR), and UO time-series data. Model performance was assessed using 5-fold stratified cross-validation with bootstrap 95% confidence intervals. SHAP analysis was used for model interpretability.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eOf 331 episodes, 71 (21.5%) achieved renal recovery. In Phase 1, day-0 UO was non-significant in both univariable (HR=1.010 per 100 mL/day; p=0.146) and multivariable Cox regression (HR=1.003 per 100 mL/day; p=0.703), even when forced into a model with clinically important covariates. The multivariable model (C-index=0.768) identified CCI (HR=0.749; p\u0026lt;0.001) and SOFA (HR=0.805; p\u0026lt;0.001) as the strongest independent predictors. In Phase 2, incorporating 72-hour trajectories significantly improved prediction: XGBoost Combined (day-0 + 72h features) achieved an AUC of 0.848 (95% CI, 0.765–0.890), compared with 0.762 for baseline-only XGBoost (AUC difference=+0.086, DeLong p=0.035). SHAP analysis revealed late-phase (36–72h) UO variability (UO_late_std) as the dominant predictor (mean |SHAP|=1.311), exceeding the second-ranked feature by four-fold. The 72-hour UO trajectory features occupied two of the top three positions in feature importance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eDay-0 hemodynamic variables fail to predict renal recovery in CKRT patients when assessed as single-time-point measurements. However, 72-hour trajectories—particularly late-phase UO variability—are powerful predictors when captured through machine learning. These findings support integrating temporal hemodynamic patterns into clinical decision-making for CKRT patients.\u003c/p\u003e","manuscriptTitle":"72-Hour Hemodynamic and Urine Output Trajectories Predict Renal Recovery in Continuous Kidney Replacement Therapy: A Machine Learning Approach","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-23 05:51:37","doi":"10.21203/rs.3.rs-9073059/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-29T13:41:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"135521374132011022022074133224898058758","date":"2026-04-18T23:28:42+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-19T07:07:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-11T04:21:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-11T04:20:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Critical Care","date":"2026-03-09T12:35:18+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"critical-care","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cric","sideBox":"Learn more about [Critical Care](http://ccforum.biomedcentral.com/)","snPcode":"13054","submissionUrl":"https://submission.nature.com/new-submission/13054/3","title":"Critical Care","twitterHandle":"@Crit_Care","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"3ee75139-df15-400a-9272-8cd1135bc7b8","owner":[],"postedDate":"March 23rd, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-04-29T13:41:46+00:00","index":33,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-23T05:51:38+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-23 05:51:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9073059","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9073059","identity":"rs-9073059","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}