Novel Preoperative Risk Stratification Using Digital Phenotyping Applying a Scalable Machine-Learning Approach.
Journal
Anesthesia and analgesia
ISSN: 1526-7598
Titre abrégé: Anesth Analg
Pays: United States
ID NLM: 1310650
Informations de publication
Date de publication:
05 Dec 2023
05 Dec 2023
Historique:
pubmed:
6
12
2023
medline:
6
12
2023
entrez:
5
12
2023
Statut:
aheadofprint
Résumé
Classification of perioperative risk is important for patient care, resource allocation, and guiding shared decision-making. Using discriminative features from the electronic health record (EHR), machine-learning algorithms can create digital phenotypes among heterogenous populations, representing distinct patient subpopulations grouped by shared characteristics, from which we can personalize care, anticipate clinical care trajectories, and explore therapies. We hypothesized that digital phenotypes in preoperative settings are associated with postoperative adverse events including in-hospital and 30-day mortality, 30-day surgical redo, intensive care unit (ICU) admission, and hospital length of stay (LOS). We identified all laminectomies, colectomies, and thoracic surgeries performed over a 9-year period from a large hospital system. Seventy-seven readily extractable preoperative features were first selected from clinical consensus, including demographics, medical history, and lab results. Three surgery-specific datasets were built and split into derivation and validation cohorts using chronological occurrence. Consensus k -means clustering was performed independently on each derivation cohort, from which phenotypes' characteristics were explored. Cluster assignments were used to train a random forest model to assign patient phenotypes in validation cohorts. We reconducted descriptive analyses on validation cohorts to confirm the similarity of patient characteristics with derivation cohorts, and quantified the association of each phenotype with postoperative adverse events by using the area under receiver operating characteristic curve (AUROC). We compared our approach to American Society of Anesthesiologists (ASA) alone and investigated a combination of our phenotypes with the ASA score. A total of 7251 patients met inclusion criteria, of which 2770 were held out in a validation dataset based on chronological occurrence. Using segmentation metrics and clinical consensus, 3 distinct phenotypes were created for each surgery. The main features used for segmentation included urgency of the procedure, preoperative LOS, age, and comorbidities. The most relevant characteristics varied for each of the 3 surgeries. Low-risk phenotype alpha was the most common (2039 of 2770, 74%), while high-risk phenotype gamma was the rarest (302 of 2770, 11%). Adverse outcomes progressively increased from phenotypes alpha to gamma, including 30-day mortality (0.3%, 2.1%, and 6.0%, respectively), in-hospital mortality (0.2%, 2.3%, and 7.3%), and prolonged hospital LOS (3.4%, 22.1%, and 25.8%). When combined with the ASA score, digital phenotypes achieved higher AUROC than the ASA score alone (hospital mortality: 0.91 vs 0.84; prolonged hospitalization: 0.80 vs 0.71). For 3 frequently performed surgeries, we identified 3 digital phenotypes. The typical profiles of each phenotype were described and could be used to anticipate adverse postoperative events.
Sections du résumé
BACKGROUND
BACKGROUND
Classification of perioperative risk is important for patient care, resource allocation, and guiding shared decision-making. Using discriminative features from the electronic health record (EHR), machine-learning algorithms can create digital phenotypes among heterogenous populations, representing distinct patient subpopulations grouped by shared characteristics, from which we can personalize care, anticipate clinical care trajectories, and explore therapies. We hypothesized that digital phenotypes in preoperative settings are associated with postoperative adverse events including in-hospital and 30-day mortality, 30-day surgical redo, intensive care unit (ICU) admission, and hospital length of stay (LOS).
METHODS
METHODS
We identified all laminectomies, colectomies, and thoracic surgeries performed over a 9-year period from a large hospital system. Seventy-seven readily extractable preoperative features were first selected from clinical consensus, including demographics, medical history, and lab results. Three surgery-specific datasets were built and split into derivation and validation cohorts using chronological occurrence. Consensus k -means clustering was performed independently on each derivation cohort, from which phenotypes' characteristics were explored. Cluster assignments were used to train a random forest model to assign patient phenotypes in validation cohorts. We reconducted descriptive analyses on validation cohorts to confirm the similarity of patient characteristics with derivation cohorts, and quantified the association of each phenotype with postoperative adverse events by using the area under receiver operating characteristic curve (AUROC). We compared our approach to American Society of Anesthesiologists (ASA) alone and investigated a combination of our phenotypes with the ASA score.
RESULTS
RESULTS
A total of 7251 patients met inclusion criteria, of which 2770 were held out in a validation dataset based on chronological occurrence. Using segmentation metrics and clinical consensus, 3 distinct phenotypes were created for each surgery. The main features used for segmentation included urgency of the procedure, preoperative LOS, age, and comorbidities. The most relevant characteristics varied for each of the 3 surgeries. Low-risk phenotype alpha was the most common (2039 of 2770, 74%), while high-risk phenotype gamma was the rarest (302 of 2770, 11%). Adverse outcomes progressively increased from phenotypes alpha to gamma, including 30-day mortality (0.3%, 2.1%, and 6.0%, respectively), in-hospital mortality (0.2%, 2.3%, and 7.3%), and prolonged hospital LOS (3.4%, 22.1%, and 25.8%). When combined with the ASA score, digital phenotypes achieved higher AUROC than the ASA score alone (hospital mortality: 0.91 vs 0.84; prolonged hospitalization: 0.80 vs 0.71).
CONCLUSIONS
CONCLUSIONS
For 3 frequently performed surgeries, we identified 3 digital phenotypes. The typical profiles of each phenotype were described and could be used to anticipate adverse postoperative events.
Identifiants
pubmed: 38051671
doi: 10.1213/ANE.0000000000006753
pii: 00000539-990000000-00672
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Informations de copyright
Copyright © 2023 International Anesthesia Research Society.
Déclaration de conflit d'intérêts
Conflicts of Interest: See Disclosures at the end of the article.
Références
Nepogodiev D, Martin J, Biccard B, Makupe A, Bhangu A; National Institute for Health Research Global Health Research Unit on Global Surgery. Global burden of postoperative death. Lancet. 2019;393:401.
Weiser TG, Regenbogen SE, Thompson KD, et al. An estimation of the global volume of surgery: a modelling strategy based on available data. Lancet. 2008;372:139–144.
Pearse RM, Harrison DA, James P, et al. Identification and characterisation of the high-risk surgical population in the United Kingdom. Crit Care. 2006;10:R81.
Le Manach Y, Collins G, Rodseth R, et al. Preoperative Score to Predict Postoperative Mortality (POSPOM): derivation and validation. Anesthesiology. 2016;124:570–579.
Stavem K, Hoel H, Skjaker SA, Haagensen R. Charlson comorbidity index derived from chart review or administrative data: agreement and prediction of mortality in intensive care patients. Clin Epidemiol. 2017;9:311–320.
Hofer IS, Lee C, Gabel E, Baldi P, Cannesson M. Development and validation of a deep neural network model to predict postoperative mortality, acute kidney injury, and reintubation using a single feature set. NPJ Digit Med. 2020;3:58.
Hackett NJ, De Oliveira GS, Jain UK, Kim JY. ASA class is a reliable independent predictor of medical complications and mortality following surgery. Int J Surg. 2015;18:184–190.
Horvath B, Kloesel B, Todd MM, Cole DJ, Prielipp RC. The evolution, current value, and future of the American Society of Anesthesiologists physical status classification system. Anesthesiology. 2021;135:904–919.
Kwa CXW, Cui J, Lim DYZ, Sim YE, Ke Y, Abdullah HR. Discordant American Society of Anesthesiologists physical status classification between anesthesiologists and surgeons and its correlation with adverse patient outcomes. Sci Rep. 2022;12:7110.
Tollinche LE, Yang G, Tan KS, Borchardt R. Interrater variability in ASA physical status assignment: an analysis in the pediatric cancer setting. J Anesth. 2018;32:211–218.
Hill BL, Brown R, Gabel E, et al. An automated machine learning-based model predicts postoperative mortality using readily-extractable preoperative electronic health record data. Br J Anaesth. 2019;123:877–886.
Lee CK, Hofer I, Gabel E, Baldi P, Cannesson M. Development and validation of a deep neural network model for prediction of postoperative in-hospital mortality. Anesthesiology. 2018;129:649–662.
Lee CK, Samad M, Hofer I, Cannesson M, Baldi P. Development and validation of an interpretable neural network for prediction of postoperative in-hospital mortality. NPJ Digit Med. 2021;4:8.
Misic VV, Gabel E, Hofer I, Rajaram K, Mahajan A. Machine learning prediction of postoperative emergency department hospital readmission. Anesthesiology. 2020;132:968–980.
Tseng PY, Chen YT, Wang CH, et al. Prediction of the development of acute kidney injury following cardiac surgery by machine learning. Crit Care. 2020;24:478.
Gray GM, Ahumada LM, Rehman MA, et al. A machine-learning approach for decision support and risk stratification of pediatric perioperative patients based on the APRICOT dataset. Paediatr Anaesth. 2023;33:710–719.
Wongtangman K, Aasman B, Garg S, et al. Development and validation of a machine learning ASA-score to identify candidates for comprehensive preoperative screening and risk stratification. J Clin Anesth. 2023;87:111103.
Zhang L, Fabbri D, Lasko TA, Ehrenfeld JM, Wanderer JP. A system for automated determination of perioperative patient acuity. J Med Syst. 2018;42:123.
Data Science Collaborative G. Differences in clinical deterioration among three sub-phenotypes of COVID-19 patients at the time of first positive test: results from a clustering analysis. Intensive Care Med. 2021;47:113–115.
Jain SH, Powers BW, Hawkins JB, Brownstein JS. The digital phenotype. Nat Biotechnol. 2015;33:462–463.
Oellrich A, Collier N, Groza T, et al. The digital revolution in phenotyping. Brief Bioinform. 2016;17:819–830.
Ferreira JC, Patino CM. Types of outcomes in clinical research. J Bras Pneumol. 2017;43:5.
Seymour CW, Kennedy JN, Wang S, et al. Derivation, validation, and potential treatment implications of novel clinical phenotypes for sepsis. JAMA. 2019;321:2003–2017.
Moura I, Teles A, Viana D, Marques J, Coutinho L, Silva F. Digital phenotyping of mental health using multimodal sensing of multiple situations of interest: a systematic literature review. J Biomed Inform. 2023;138:104278.
Jayakumar P, Lin E, Galea V, et al. Digital phenotyping and patient-generated health data for outcome measurement in surgical care: a scoping review. J Pers Med. 2020;10:282.
Grant RW, McCloskey J, Hatfield M, et al. Use of latent class analysis and k-means clustering to identify complex patient profiles. JAMA Netw Open. 2020;3:e2029068.
Ranti D, Warburton AJ, Hanss K, Katz D, Poeran J, Moucha C. K-Means clustering to elucidate vulnerable subpopulations among Medicare patients undergoing total joint arthroplasty. J Arthroplasty. 2020;35:3488–3497.
Luo W, Phung D, Tran T, et al. Guidelines for developing and reporting machine learning predictive models in biomedical research: a multidisciplinary view. J Med Internet Res. 2016;18:e323.
Hofer IS, Gabel E, Pfeffer M, Mahbouba M, Mahajan A. A systematic approach to creation of a perioperative data warehouse. Anesth Analg. 2016;122:1880–1884.
Kamdar NV, Huverserian A, Jalilian L, et al. Development, implementation, and evaluation of a telemedicine preoperative evaluation initiative at a major academic medical center. Anesth Analg. 2020;131:1647–1656.
Delrieu L, Hamy AS, Coussy F, et al. Digital phenotyping in young breast cancer patients treated with neoadjuvant chemotherapy (the NeoFit Trial): protocol for a national, multicenter single-arm trial. BMC Cancer. 2022;22:493.
Cote DJ, Barnett I, Onnela JP, Smith TR. Digital phenotyping in patients with spine disease: a novel approach to quantifying mobility and quality of life. World Neurosurg. 2019;126:e241–e249.
Lamarche Y, Elmi-Sarabi M, Ding L, Abel JG, Sirounis D, Denault AY. A score to estimate 30-day mortality after intensive care admission after cardiac surgery. J Thorac Cardiovasc Surg. 2017;153:1118–1125.e4.
Van Buuren S. Multiple imputation of discrete and continuous data by fully conditional specification. Stat Methods Med Res. 2007;16:219–242.
Wilkerson MD, Hayes DN. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics. 2010;26:1572–1573.
MacQueen JB. Some methods for classification and analysis of multivariate observations. Proc Berkeley Symp Math Statist Prob. 1967;1:281–297.
Alyousef AA, Nihtyanova S, Denton C, Bosoni P, Bellazzi R, Tucker A. Nearest consensus clustering classification to identify subclasses and predict disease. J Healthc Inform Res. 2018;2:402–422.
Jalali A, Lonsdale H, Do N, et al. Deep learning for improved risk prediction in surgical outcomes. Sci Rep. 2020;10:9289.
Barnett I, Torous J, Staples P, Sandoval L, Keshavan M, Onnela JP. Relapse prediction in schizophrenia through digital phenotyping: a pilot study. Neuropsychopharmacology. 2018;43:1660–1666.
Benoit J, Onyeaka H, Keshavan M, Torous J. Systematic review of digital phenotyping and machine learning in psychosis spectrum illnesses. Harv Rev Psychiatry. 2020;28:296–304.
Kleiman EM, Turner BJ, Fedor S, et al. Digital phenotyping of suicidal thoughts. Depress Anxiety. 2018;35:601–608.
Brittany N, Burton CC, Jennifer L, Maxime C. Personalized surgical transfusion risk prediction: comment. Anesthesiology. 2022;138:116–117.
Lou SS, Liu H, Lu C, Wildes TS, Hall BL, Kannampallil T. Personalized surgical transfusion risk prediction using machine learning to guide preoperative type and screen orders. Anesthesiology. 2022;137:55–66.
Lou SS, Wildes TS, Hall BL, Avidan MS, Kannampallil T. Personalized surgical transfusion risk prediction: reply. Anesthesiology. 2022;138:118–119.