1
|
Benovic S, Ajlani AH, Leinert C, Fotteler M, Wolf D, Steger F, Kestler H, Dallmeier D, Denkinger M, Eschweiler GW, Thomas C, Kocar TD. Introducing a machine learning algorithm for delirium prediction-the Supporting SURgery with GEriatric Co-Management and AI project (SURGE-Ahead). Age Ageing 2024; 53:afae101. [PMID: 38776213 PMCID: PMC11110913 DOI: 10.1093/ageing/afae101] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/19/2023] [Indexed: 05/24/2024] Open
Abstract
INTRODUCTION Post-operative delirium (POD) is a common complication in older patients, with an incidence of 14-56%. To implement preventative procedures, it is necessary to identify patients at risk for POD. In the present study, we aimed to develop a machine learning (ML) model for POD prediction in older patients, in close cooperation with the PAWEL (patient safety, cost-effectiveness and quality of life in elective surgery) project. METHODS The model was trained on the PAWEL study's dataset of 878 patients (no intervention, age ≥ 70, 209 with POD). Presence of POD was determined by the Confusion Assessment Method and a chart review. We selected 15 features based on domain knowledge, ethical considerations and a recursive feature elimination. A logistic regression and a linear support vector machine (SVM) were trained, and evaluated using receiver operator characteristics (ROC). RESULTS The selected features were American Society of Anesthesiologists score, multimorbidity, cut-to-suture time, estimated glomerular filtration rate, polypharmacy, use of cardio-pulmonary bypass, the Montreal cognitive assessment subscores 'memory', 'orientation' and 'verbal fluency', pre-existing dementia, clinical frailty scale, age, recent falls, post-operative isolation and pre-operative benzodiazepines. The linear SVM performed best, with an ROC area under the curve of 0.82 [95% CI 0.78-0.85] in the training set, 0.81 [95% CI 0.71-0.88] in the test set and 0.76 [95% CI 0.71-0.79] in a cross-centre validation. CONCLUSION We present a clinically useful and explainable ML model for POD prediction. The model will be deployed in the Supporting SURgery with GEriatric Co-Management and AI project.
Collapse
Affiliation(s)
- Samuel Benovic
- Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany
- Agaplesion Bethesda Clinic Ulm, Ulm, Germany
| | - Anna H Ajlani
- Institute of the History, Philosophy and Ethics of Medicine, Ulm University, Ulm, Germany
- Department of Sociology with a Focus on Innovation and Digitalization, Institute of Sociology, Johannes Kepler University Linz, Linz, Austria
| | - Christoph Leinert
- Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany
- Agaplesion Bethesda Clinic Ulm, Ulm, Germany
| | - Marina Fotteler
- Agaplesion Bethesda Clinic Ulm, Ulm, Germany
- DigiHealth Institute, Neu-Ulm University of Applied Sciences, Neu-Ulm, Germany
| | - Dennis Wolf
- Institute of Medical Systems Biology, Ulm University, Ulm, Germany
| | - Florian Steger
- Institute of the History, Philosophy and Ethics of Medicine, Ulm University, Ulm, Germany
| | - Hans Kestler
- Institute of Medical Systems Biology, Ulm University, Ulm, Germany
| | - Dhayana Dallmeier
- Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany
- Department of Epidemiology, Boston University School of Public Health, Boston, USA
| | - Michael Denkinger
- Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany
- Agaplesion Bethesda Clinic Ulm, Ulm, Germany
| | - Gerhard W Eschweiler
- Geriatric Center, University Hospital Tübingen, Tubingen, Germany
- Department of Psychiatry and Psychotherapy, Tübingen University Hospital, Tübingen, Germany
| | - Christine Thomas
- Department of Psychiatry and Psychotherapy, Tübingen University Hospital, Tübingen, Germany
- Department of Geriatric Psychiatry and Psychotherapy, Klinikum Stuttgart, Stuttgart, Germany
| | - Thomas D Kocar
- Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany
- Agaplesion Bethesda Clinic Ulm, Ulm, Germany
| |
Collapse
|
2
|
Snigurska UA, Liu Y, Ser SE, Macieira TGR, Ansell M, Lindberg D, Prosperi M, Bjarnadottir RI, Lucero RJ. Risk of bias in prognostic models of hospital-induced delirium for medical-surgical units: A systematic review. PLoS One 2023; 18:e0285527. [PMID: 37590196 PMCID: PMC10434879 DOI: 10.1371/journal.pone.0285527] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/12/2022] [Accepted: 04/25/2023] [Indexed: 08/19/2023] Open
Abstract
PURPOSE The purpose of this systematic review was to assess risk of bias in existing prognostic models of hospital-induced delirium for medical-surgical units. METHODS APA PsycInfo, CINAHL, MEDLINE, and Web of Science Core Collection were searched on July 8, 2022, to identify original studies which developed and validated prognostic models of hospital-induced delirium for adult patients who were hospitalized in medical-surgical units. The Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies was used for data extraction. The Prediction Model Risk of Bias Assessment Tool was used to assess risk of bias. Risk of bias was assessed across four domains: participants, predictors, outcome, and analysis. RESULTS Thirteen studies were included in the qualitative synthesis, including ten model development and validation studies and three model validation only studies. The methods in all of the studies were rated to be at high overall risk of bias. The methods of statistical analysis were the greatest source of bias. External validity of models in the included studies was tested at low levels of transportability. CONCLUSIONS Our findings highlight the ongoing scientific challenge of developing a valid prognostic model of hospital-induced delirium for medical-surgical units to tailor preventive interventions to patients who are at high risk of this iatrogenic condition. With limited knowledge about generalizable prognosis of hospital-induced delirium in medical-surgical units, existing prognostic models should be used with caution when creating clinical practice policies. Future research protocols must include robust study designs which take into account the perspectives of clinicians to identify and validate risk factors of hospital-induced delirium for accurate and generalizable prognosis in medical-surgical units.
Collapse
Affiliation(s)
- Urszula A. Snigurska
- Department of Family, Community, and Health Systems Science, College of Nursing, University of Florida, Gainesville, FL, United States of America
| | - Yiyang Liu
- Department of Epidemiology, College of Public Health and Health Professions and College of Medicine, University of Florida, Gainesville, FL, United States of America
| | - Sarah E. Ser
- Department of Epidemiology, College of Public Health and Health Professions and College of Medicine, University of Florida, Gainesville, FL, United States of America
| | - Tamara G. R. Macieira
- Department of Family, Community, and Health Systems Science, College of Nursing, University of Florida, Gainesville, FL, United States of America
| | - Margaret Ansell
- Health Science Center Libraries, George A. Smathers Libraries, University of Florida, Gainesville, FL, United States of America
| | - David Lindberg
- Department of Statistics, College of Liberal Arts and Sciences, University of Florida, Gainesville, FL, United States of America
| | - Mattia Prosperi
- Department of Epidemiology, College of Public Health and Health Professions and College of Medicine, University of Florida, Gainesville, FL, United States of America
| | - Ragnhildur I. Bjarnadottir
- Department of Family, Community, and Health Systems Science, College of Nursing, University of Florida, Gainesville, FL, United States of America
| | - Robert J. Lucero
- Department of Family, Community, and Health Systems Science, College of Nursing, University of Florida, Gainesville, FL, United States of America
- School of Nursing, University of California Los Angeles, Los Angeles, CA, United States of America
| |
Collapse
|
3
|
Dodsworth BT, Reeve K, Falco L, Hueting T, Sadeghirad B, Mbuagbaw L, Goettel N, Schmutz Gelsomino N. Development and validation of an international preoperative risk assessment model for postoperative delirium. Age Ageing 2023; 52:7192246. [PMID: 37290122 DOI: 10.1093/ageing/afad086] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2022] [Indexed: 06/10/2023] Open
Abstract
BACKGROUND Postoperative delirium (POD) is a frequent complication in older adults, characterised by disturbances in attention, awareness and cognition, and associated with prolonged hospitalisation, poor functional recovery, cognitive decline, long-term dementia and increased mortality. Early identification of patients at risk of POD can considerably aid prevention. METHODS We have developed a preoperative POD risk prediction algorithm using data from eight studies identified during a systematic review and providing individual-level data. Ten-fold cross-validation was used for predictor selection and internal validation of the final penalised logistic regression model. The external validation used data from university hospitals in Switzerland and Germany. RESULTS Development included 2,250 surgical (excluding cardiac and intracranial) patients 60 years of age or older, 444 of whom developed POD. The final model included age, body mass index, American Society of Anaesthesiologists (ASA) score, history of delirium, cognitive impairment, medications, optional C-reactive protein (CRP), surgical risk and whether the operation is a laparotomy/thoracotomy. At internal validation, the algorithm had an AUC of 0.80 (95% CI: 0.77-0.82) with CRP and 0.79 (95% CI: 0.77-0.82) without CRP. The external validation consisted of 359 patients, 87 of whom developed POD. The external validation yielded an AUC of 0.74 (95% CI: 0.68-0.80). CONCLUSIONS The algorithm is named PIPRA (Pre-Interventional Preventive Risk Assessment), has European conformity (ce) certification, is available at http://pipra.ch/ and is accepted for clinical use. It can be used to optimise patient care and prioritise interventions for vulnerable patients and presents an effective way to implement POD prevention strategies in clinical practice.
Collapse
Affiliation(s)
| | - Kelly Reeve
- Institute of Data Analysis and Process Design, Zurich University of Applied Sciences, Winterthur 8400, Switzerland
| | - Lisa Falco
- Zühlke Engineering AG, Zürcherstrasse 39J, Schlieren 8952, Switzerland
| | - Tom Hueting
- Evidencio, Irenesingel 19, Haaksbergen 7481 GJ, Netherlands
| | - Behnam Sadeghirad
- Department of Health Research Methods, Evidence, and Impact, McMaster University, Hamilton ON L8S 4L8, Canada
- Department of Anesthesia, McMaster University, Hamilton ON L8S 4L8, Canada
| | - Lawrence Mbuagbaw
- Department of Health Research Methods, Evidence, and Impact, McMaster University, Hamilton ON L8S 4L8, Canada
- Department of Anesthesia, McMaster University, Hamilton ON L8S 4L8, Canada
- Department of Pediatrics, McMaster University, Hamilton, ON L8S 4L8, Canada
- Biostatistics Unit, Father Sean O'Sullivan Research Centre, St Joseph's Healthcare, Hamilton, ON L8S 4L8, Canada
- Centre for Development of Best Practices in Health (CDBPH), Yaoundé Central Hospital, Yaoundé 12117, Cameroon
- Division of Epidemiology and Biostatistics, Department of Global Health, Stellenbosch University, Cape Town 7600, South Africa
| | - Nicolai Goettel
- Department of Anesthesiology, University of Florida College of Medicine, Gainesville FL 32610, USA
- Department of Clinical Research, University of Basel, Basel 4031, Switzerland
| | - Nayeli Schmutz Gelsomino
- PIPRA AG, Zurich 8005, Switzerland
- Department of Anaesthesia, University Hospital Basel, Spitalstrasse 21, Basel 4031, Switzerland
| |
Collapse
|
4
|
Wueest AS, Berres M, Bettex DA, Steiner LA, Monsch AU, Goettel N. Independent External Validation of a Preoperative Prediction Model for Delirium After Cardiac Surgery: A Prospective Observational Cohort Study. J Cardiothorac Vasc Anesth 2023; 37:415-422. [PMID: 36567220 DOI: 10.1053/j.jvca.2022.11.038] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/19/2022] [Revised: 11/09/2022] [Accepted: 11/30/2022] [Indexed: 12/12/2022]
Abstract
OBJECTIVE This investigation provided independent external validation of an existing preoperative risk prediction model. DESIGN A prospective observational cohort study of patients undergoing cardiac surgery covering the period between April 16, 2018 and January 18, 2022. SETTING Two academic hospitals in Switzerland. PARTICIPANTS Adult patients (≥60 years of age) who underwent elective cardiac surgery, including coronary artery bypass graft, mitral or aortic valve replacement or repair, and combined procedures. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS The primary outcome measure was the incidence of postoperative delirium (POD) in the intensive or intermediate care unit, diagnosed using the Intensive Care Delirium Screening Checklist. The prediction model contained 4 preoperative risk factors to which the following points were assigned: Mini-Mental State Examination (MMSE) score ≤23 received 2 points; MMSE 24-27, Geriatric Depression Scale (GDS) >4, prior stroke and/or transient ischemic attack (TIA), and abnormal serum albumin (≤3.5 or ≥4.5 g/dL) received 1 point each. The missing data were handled using multiple imputation. In total, 348 patients were included in the study. Sixty patients (17.4%) developed POD. For point levels in the prediction model of 0, 1, 2, and ≥3, the cumulative incidence of POD was 12.6%, 22.8%, 25.8%, and 35%, respectively. The validation resulted in a pooled area under the receiver operating characteristics curve of 0.60 (median CI, 0.525-0.679). CONCLUSIONS The evaluated predictive model for delirium after cardiac surgery in this patient cohort showed only poor discriminative capacity but fair calibration.
Collapse
Affiliation(s)
- Alexandra S Wueest
- Memory Clinic, University Department of Geriatric Medicine FELIX PLATTER, Basel, Switzerland; Clinic for Anaesthesia, Intermediate Care, Prehospital Emergency Medicine and Pain Therapy, University Hospital Basel, Basel, Switzerland
| | - Manfred Berres
- Department of Mathematics and Technology, University of Applied Sciences Koblenz, Germany
| | - Dominique A Bettex
- Division of Cardiovascular Anaesthesia, Institute of Anaesthesia, University Hospital Zurich, Zurich, Switzerland
| | - Luzius A Steiner
- Clinic for Anaesthesia, Intermediate Care, Prehospital Emergency Medicine and Pain Therapy, University Hospital Basel, Basel, Switzerland; Department of Clinical Research University of Basel, Basel, Switzerland
| | - Andreas U Monsch
- Memory Clinic, University Department of Geriatric Medicine FELIX PLATTER, Basel, Switzerland
| | - Nicolai Goettel
- Department of Clinical Research University of Basel, Basel, Switzerland; Department of Anaesthesiology, University of Florida College of Medicine, Gainesville, FL, USA.
| |
Collapse
|