Machine Learning Predicts Unplanned Care Escalations for Post-Anesthesia Care Unit Patients during the Perioperative Period: A Single-Center Retrospective Study.
Artificial Intelligence
Patient Safety
Precision Medicine
Predictive Analytics
Risk Stratification
Journal
Journal of medical systems
ISSN: 1573-689X
Titre abrégé: J Med Syst
Pays: United States
ID NLM: 7806056
Informations de publication
Date de publication:
23 Jul 2024
23 Jul 2024
Historique:
received:
30
11
2023
accepted:
06
07
2024
medline:
23
7
2024
pubmed:
23
7
2024
entrez:
23
7
2024
Statut:
epublish
Résumé
Despite low mortality for elective procedures in the United States and developed countries, some patients have unexpected care escalations (UCE) following post-anesthesia care unit (PACU) discharge. Studies indicate patient risk factors for UCE, but determining which factors are most important is unclear. Machine learning (ML) can predict clinical events. We hypothesized that ML could predict patient UCE after PACU discharge in surgical patients and identify specific risk factors. We conducted a single center, retrospective analysis of all patients undergoing non-cardiac surgery (elective and emergent). We collected data from pre-operative visits, intra-operative records, PACU admissions, and the rate of UCE. We trained a ML model with this data and tested the model on an independent data set to determine its efficacy. Finally, we evaluated the individual patient and clinical factors most likely to predict UCE risk. Our study revealed that ML could predict UCE risk which was approximately 5% in both the training and testing groups. We were able to identify patient risk factors such as patient vital signs, emergent procedure, ASA Status, and non-surgical anesthesia time as significant variable. We plotted Shapley values for significant variables for each patient to help determine which of these variables had the greatest effect on UCE risk. Of note, the UCE risk factors identified frequently by ML were in alignment with anesthesiologist clinical practice and the current literature. We used ML to analyze data from a single-center, retrospective cohort of non-cardiac surgical patients, some of whom had an UCE. ML assigned risk prediction for patients to have UCE and determined perioperative factors associated with increased risk. We advocate to use ML to augment anesthesiologist clinical decision-making, help decide proper disposition from the PACU, and ensure the safest possible care of our patients.
Sections du résumé
BACKGROUND
BACKGROUND
Despite low mortality for elective procedures in the United States and developed countries, some patients have unexpected care escalations (UCE) following post-anesthesia care unit (PACU) discharge. Studies indicate patient risk factors for UCE, but determining which factors are most important is unclear. Machine learning (ML) can predict clinical events. We hypothesized that ML could predict patient UCE after PACU discharge in surgical patients and identify specific risk factors.
METHODS
METHODS
We conducted a single center, retrospective analysis of all patients undergoing non-cardiac surgery (elective and emergent). We collected data from pre-operative visits, intra-operative records, PACU admissions, and the rate of UCE. We trained a ML model with this data and tested the model on an independent data set to determine its efficacy. Finally, we evaluated the individual patient and clinical factors most likely to predict UCE risk.
RESULTS
RESULTS
Our study revealed that ML could predict UCE risk which was approximately 5% in both the training and testing groups. We were able to identify patient risk factors such as patient vital signs, emergent procedure, ASA Status, and non-surgical anesthesia time as significant variable. We plotted Shapley values for significant variables for each patient to help determine which of these variables had the greatest effect on UCE risk. Of note, the UCE risk factors identified frequently by ML were in alignment with anesthesiologist clinical practice and the current literature.
CONCLUSIONS
CONCLUSIONS
We used ML to analyze data from a single-center, retrospective cohort of non-cardiac surgical patients, some of whom had an UCE. ML assigned risk prediction for patients to have UCE and determined perioperative factors associated with increased risk. We advocate to use ML to augment anesthesiologist clinical decision-making, help decide proper disposition from the PACU, and ensure the safest possible care of our patients.
Identifiants
pubmed: 39042285
doi: 10.1007/s10916-024-02085-9
pii: 10.1007/s10916-024-02085-9
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
69Subventions
Organisme : NIH HHS
ID : GM127584
Pays : United States
Informations de copyright
© 2024. The Author(s).
Références
D. Bainbridge, J. Martin, M. Arango, D. Cheng, and G. Evidence-based Peri-operative Clinical Outcomes Research, “Perioperative and anaesthetic-related mortality in developed and developing countries: a systematic review and meta-analysis,” Lancet, vol. 380, no. 9847, pp. 1075-81, Sep 22 2012, doi: https://doi.org/10.1016/S0140-6736(12)60990-8 .
D. A. Watters et al, “Perioperative mortality rate (POMR): a global indicator of access to safe surgery and anaesthesia,” World J Surg, vol. 39, no. 4, pp. 856 – 64, Apr 2015, doi: https://doi.org/10.1007/s00268-014-2638-4 .
N. Katori, K. Yamakawa, K. Yagi, Y. Kimura, M. Doi, and S. Uezono, “Characteristics and outcomes of unplanned intensive care unit admission after general anesthesia,” BMC Anesthesiol, vol. 22, no. 1, p. 191, Jun 20 2022, doi: https://doi.org/10.1186/s12871-022-01729-y .
doi: 10.1186/s12871-022-01729-y
pubmed: 35725372
pmcid: 9208222
M. S. Melton et al, “Unplanned hospital admission after ambulatory surgery: a retrospective, single cohort study,” Can J Anaesth, vol. 68, no. 1, pp. 30–41, Jan 2021, doi: 10.1007/s12630-020-01822-1. Admission non planifiee a l’hopital apres une chirurgie ambulatoire: une etude retrospective de cohorte unique.
T. J. Loftus et al, “Overtriage, Undertriage, and Value of Care after Major Surgery: An Automated, Explainable Deep Learning-Enabled Classification System,” J Am Coll Surg, vol. 236, no. 2, pp. 279–291, Feb 1 2023, doi: https://doi.org/10.1097/XCS.0000000000000471 .
doi: 10.1097/XCS.0000000000000471
pubmed: 36648256
H. Ohbe, H. Matsui, R. Kumazawa, and H. Yasunaga, “Intensive care unit versus high dependency care unit admission after emergency surgery: a nationwide in-patient registry study,” Br J Anaesth, vol. 129, no. 4, pp. 527–535, Oct 2022, doi: https://doi.org/10.1016/j.bja.2022.06.030 .
doi: 10.1016/j.bja.2022.06.030
pubmed: 35961814
G. Costa et al, “Gastro-intestinal emergency surgery: Evaluation of morbidity and mortality. Protocol of a prospective, multicenter study in Italy for evaluating the burden of abdominal emergency surgery in different age groups. (The GESEMM study),” Front Surg, vol. 9, p. 927044, 2022, doi: https://doi.org/10.3389/fsurg.2022.927044 .
doi: 10.3389/fsurg.2022.927044
pubmed: 36189400
pmcid: 9524583
I. Garutti et al, “Spontaneous recovery of neuromuscular blockade is an independent risk factor for postoperative pulmonary complications after abdominal surgery: A secondary analysis,” Eur J Anaesthesiol, vol. 37, no. 3, pp. 203–211, Mar 2020, doi: https://doi.org/10.1097/EJA.0000000000001128 .
doi: 10.1097/EJA.0000000000001128
pubmed: 32028288
D. Chandler et al, “Perioperative strategies for the reduction of postoperative pulmonary complications,” Best Pract Res Clin Anaesthesiol, vol. 34, no. 2, pp. 153–166, Jun 2020, doi: https://doi.org/10.1016/j.bpa.2020.04.011 .
doi: 10.1016/j.bpa.2020.04.011
pubmed: 32711826
H. Mufti et al, “The association between preoperative anemia, blood transfusion need, and postoperative complications in adult cardiac surgery, a single center contemporary experience,” J Cardiothorac Surg, vol. 18, no. 1, p. 10, Jan 7 2023, doi: https://doi.org/10.1186/s13019-023-02132-5 .
doi: 10.1186/s13019-023-02132-5
pubmed: 36611177
pmcid: 9824911
A. Hosny, C. Parmar, J. Quackenbush, L. H. Schwartz, and H. Aerts, “Artificial intelligence in radiology,” Nat Rev Cancer, vol. 18, no. 8, pp. 500–510, Aug 2018, doi: https://doi.org/10.1038/s41568-018-0016-5 .
J. Seah, T. Boeken, M. Sapoval, and G. S. Goh, “Prime Time for Artificial Intelligence in Interventional Radiology,” Cardiovascular and interventional radiology, vol. 45, no. 3, pp. 283–289, Mar 2022, doi: https://doi.org/10.1007/s00270-021-03044-4 .
doi: 10.1007/s00270-021-03044-4
pubmed: 35031822
pmcid: 8921296
V. Baxi, R. Edwards, M. Montalto, and S. Saha, “Digital pathology and artificial intelligence in translational medicine and clinical practice,” Mod Pathol, vol. 35, no. 1, pp. 23–32, Jan 2022, doi: https://doi.org/10.1038/s41379-021-00919-2 .
doi: 10.1038/s41379-021-00919-2
pubmed: 34611303
K. Bera, K. A. Schalper, D. L. Rimm, V. Velcheti, and A. Madabhushi, “Artificial intelligence in digital pathology - new tools for diagnosis and precision oncology,” Nat Rev Clin Oncol, vol. 16, no. 11, pp. 703–715, Nov 2019, doi: https://doi.org/10.1038/s41571-019-0252-y .
doi: 10.1038/s41571-019-0252-y
pubmed: 31399699
pmcid: 6880861
S. Mainali and S. Park, “Artificial Intelligence and Big Data Science in Neurocritical Care,” Crit Care Clin, vol. 39, no. 1, pp. 235–242, Jan 2023, doi: https://doi.org/10.1016/j.ccc.2022.07.008 .
doi: 10.1016/j.ccc.2022.07.008
pubmed: 36333034
R. Thirunavukarasu, G. P. D. C, G. R, M. Gopikrishnan, and V. Palanisamy, “Towards computational solutions for precision medicine based big data healthcare system using deep learning models: A review,” Comput Biol Med, vol. 149, p. 106020, Oct 2022, doi: https://doi.org/10.1016/j.compbiomed.2022.106020 .
doi: 10.1016/j.compbiomed.2022.106020
pubmed: 36088715
K. Palla et al, “Intraoperative prediction of postanaesthesia care unit hypotension,” Br J Anaesth, vol. 128, no. 4, pp. 623–635, Apr 2022, doi: https://doi.org/10.1016/j.bja.2021.10.052 .
doi: 10.1016/j.bja.2021.10.052
pubmed: 34924175
A. Abujaber, A. Fadlalla, D. Gammoh, H. Al-Thani, and A. El-Menyar, “Machine Learning Model to Predict Ventilator Associated Pneumonia in patients with Traumatic Brain Injury: The C.5 Decision Tree Approach,” Brain Inj, vol. 35, no. 9, pp. 1095–1102, Jul 29 2021, doi: https://doi.org/10.1080/02699052.2021.1959060 .
P. Sinha, A. Spicer, K. L. Delucchi, D. F. McAuley, C. S. Calfee, and M. M. Churpek, “Comparison of machine learning clustering algorithms for detecting heterogeneity of treatment effect in acute respiratory distress syndrome: A secondary analysis of three randomised controlled trials,” EBioMedicine, vol. 74, p. 103697, Dec 2021, doi: https://doi.org/10.1016/j.ebiom.2021.103697 .
N. H. Badner, R. L. Knill, J. E. Brown, T. V. Novick, and A. W. Gelb, “Myocardial infarction after noncardiac surgery,” Anesthesiology, vol. 88, no. 3, pp. 572-8, Mar 1998, doi: https://doi.org/10.1097/00000542-199803000-00005 .
G. Landesberg et al, “Importance of long-duration postoperative ST-segment depression in cardiac morbidity after vascular surgery,” Lancet, vol. 341, no. 8847, pp. 715-9, Mar 20 1993, doi: https://doi.org/10.1016/0140-6736(93)90486-z .
I. Vascular Events In Noncardiac Surgery Patients Cohort Evaluation Study et al, “Association between postoperative troponin levels and 30-day mortality among patients undergoing noncardiac surgery,” JAMA, vol. 307, no. 21, pp. 2295 – 304, Jun 6 2012, doi: https://doi.org/10.1001/jama.2012.5502 .
X. Dastile, Celik, T, Potsane, M, “Statistical and machine learning models in credit scoring: A systematic literature survey,” Applied Soft Computing, vol. 91, p. 106263, 2020, doi: https://doi.org/10.1016/j.asoc.2020.106263 .
doi: 10.1016/j.asoc.2020.106263
L. Jehi et al, “Development and validation of a model for individualized prediction of hospitalization risk in 4,536 patients with COVID-19,” PLoS One, vol. 15, no. 8, p. e0237419, 2020, doi: https://doi.org/10.1371/journal.pone.0237419 .
doi: 10.1371/journal.pone.0237419
pubmed: 32780765
pmcid: 7418996
L. Jehi et al, “Individualizing Risk Prediction for Positive Coronavirus Disease 2019 Testing: Results From 11,672 Patients,” Chest, vol. 158, no. 4, pp. 1364–1375, Oct 2020, doi: https://doi.org/10.1016/j.chest.2020.05.580 .
T. K. K. Mamidi, T. K. Tran-Nguyen, R. L. Melvin, and E. A. Worthey, “Development of An Individualized Risk Prediction Model for COVID-19 Using Electronic Health Record Data,” Front Big Data, vol. 4, p. 675882, 2021, doi: https://doi.org/10.3389/fdata.2021.675882 .
doi: 10.3389/fdata.2021.675882
pubmed: 34151259
pmcid: 8211871
G. Navas-Palencia, “Optimal binning: mathematical programming formulation,” doi: https://doi.org/10.48550/arXiv.2001.08025 .
G. Navas-Palencia, “Optimal Counterfactual Explanations for Scorecard modelling,” doi: https://doi.org/10.48550/arXiv.2104.08619 .
L. P. Zdravevski E, Kulakov A, “Weight of evidence as a tool for attribute transformation in the preprocessing stage of supervised learning algorithms,” The 2011 International Joint Conference on Neural Networks, pp. 181–188, 2011, doi: https://doi.org/10.1109/IJCNN.2011.6033219 .
H. Zou, Hastie, T, “Regularization and variable selection via the elastic net.,” Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 67, pp. 301–320, 2005, doi: https://doi.org/10.1111/j.1467-9868.2005.00503.x .
doi: 10.1111/j.1467-9868.2005.00503.x
M. J. Alves and J. Climaco, “A review of interactive methods for multiobjective integer and mixed-integer programming,” European Journal of Operational Research, vol. 180, no. 1, pp. 99–115, 2007.
doi: 10.1016/j.ejor.2006.02.033
The Shapley Value: Essays in Honor of Lloyd S. Shapley. Cambridge: Cambridge University Press, 1998.
R. Aumann, Game Theory. Palgrave Macmillan UK, 1989.
S. Lundberg, Lee, SI, “A unified approach to interpreting model predictions,” Advances in neural information processing systems, 2017, doi: https://doi.org/10.48550/arXiv.1705.07874 .
doi: 10.48550/arXiv.1705.07874
S. Y. Chen, Z. Feng, and X. Yi, “A general introduction to adjustment for multiple comparisons,” J Thorac Dis, vol. 9, no. 6, pp. 1725–1729, Jun 2017, doi: https://doi.org/10.21037/jtd.2017.05.34 .
doi: 10.21037/jtd.2017.05.34
pubmed: 28740688
pmcid: 5506159
T. J. Pollard, A. E. W. Johnson, J. D. Raffa, and R. G. Mark, “tableone: An open source Python package for producing summary statistics for research papers,” JAMIA Open, vol. 1, no. 1, pp. 26–31, Jul 2018, doi: https://doi.org/10.1093/jamiaopen/ooy012 .
doi: 10.1093/jamiaopen/ooy012
pubmed: 31984317
pmcid: 6951995
J. V. Carter, J. Pan, S. N. Rai, and S. Galandiuk, “ROC-ing along: Evaluation and interpretation of receiver operating characteristic curves,” Surgery, vol. 159, no. 6, pp. 1638–1645, Jun 2016, doi: https://doi.org/10.1016/j.surg.2015.12.029 .
G. Aad et al, “Observation of associated near-side and away-side long-range correlations in sqrt[s(NN)] = 5.02 TeV proton-lead collisions with the ATLAS detector,” Physical review letters, vol. 110, no. 18, p. 182302, May 3 2013. [Online]. Available: http://www.ncbi.nlm.nih.gov/pubmed/23683193 .
M. H. Huang et al, “Validation of a Deep Learning-based Automatic Detection Algorithm for Measurement of Endotracheal Tube-to-Carina Distance on Chest Radiographs,” Anesthesiology, vol. 137, no. 6, pp. 704–715, Dec 1 2022, doi: https://doi.org/10.1097/ALN.0000000000004378 .
J. A. Aldrete and D. Kroulik, “A postanesthetic recovery score,” Anesth Analg, vol. 49, no. 6, pp. 924 – 34, Nov-Dec 1970. [Online]. Available: https://www.ncbi.nlm.nih.gov/pubmed/5534693 .
D. Yamaguchi et al, “Usefulness of discharge standards in outpatients undergoing sedative endoscopy: a propensity score-matched study of the modified post-anesthetic discharge scoring system and the modified Aldrete score,” BMC Gastroenterol, vol. 22, no. 1, p. 445, Nov 4 2022, doi: https://doi.org/10.1186/s12876-022-02549-7 .
doi: 10.1186/s12876-022-02549-7
pubmed: 36333660
pmcid: 9635164
A. Kia et al, “MEWS++: Enhancing the Prediction of Clinical Deterioration in Admitted Patients through a Machine Learning Model,” J Clin Med, vol. 9, no. 2, Jan 27 2020, doi: https://doi.org/10.3390/jcm9020343 .
M. J. Rothman, S. I. Rothman, and J. t. Beals, “Development and validation of a continuous measure of patient condition using the Electronic Medical Record,” J Biomed Inform, vol. 46, no. 5, pp. 837 – 48, Oct 2013, doi: https://doi.org/10.1016/j.jbi.2013.06.011 .
S. Jahandideh, G. Ozavci, B. W. Sahle, A. Z. Kouzani, F. Magrabi, and T. Bucknall, “Evaluation of machine learning-based models for prediction of clinical deterioration: A systematic literature review,” Int J Med Inform, vol. 175, p. 105084, Jul 2023, doi: https://doi.org/10.1016/j.ijmedinf.2023.105084 .
doi: 10.1016/j.ijmedinf.2023.105084
pubmed: 37156168
Y. M. Hydoub et al, “Risk Prediction Models for Hospital Mortality in General Medical Patients: A Systematic Review,” Am J Med Open, vol. 10, Dec 2023, doi: https://doi.org/10.1016/j.ajmo.2023.100044 .
M. E. Smith et al, “Early warning system scores for clinical deterioration in hospitalized patients: a systematic review,” Annals of the American Thoracic Society, vol. 11, no. 9, pp. 1454–65, Nov 2014, doi: https://doi.org/10.1513/AnnalsATS.201403-102OC .
doi: 10.1513/AnnalsATS.201403-102OC
pubmed: 25296111
F. Pesapane et al, “Myths and facts about artificial intelligence: why machine- and deep-learning will not replace interventional radiologists,” Med Oncol, vol. 37, no. 5, p. 40, Apr 3 2020, doi: https://doi.org/10.1007/s12032-020-01368-8 .
A. Di Ieva, “AI-augmented multidisciplinary teams: hype or hope?,” Lancet, vol. 394, no. 10211, p. 1801, Nov 16 2019, doi: https://doi.org/10.1016/S0140-6736(19)32626-1 .
O. Asan, A. E. Bayrak, and A. Choudhury, “Artificial Intelligence and Human Trust in Healthcare: Focus on Clinicians,” J Med Internet Res, vol. 22, no. 6, p. e15154, Jun 19 2020, doi: https://doi.org/10.2196/15154 .