{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:06:53Z","timestamp":1782864413920,"version":"3.54.5"},"reference-count":30,"publisher":"BMJ","content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["Reg Anesth Pain Med"],"accepted":{"date-parts":[[2026,4,5]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Up to 20% of total knee arthroplasty (TKA) patients experience significant postoperative pain that delays recovery and increases risk of chronic pain. Early identification of high-risk patients may allow for timely targeted interventions. Prior studies on postoperative pain trajectories have been limited by small cohorts and restricted methodology. This exploratory study characterized postoperative pain archetypes using machine learning and developed a perioperative predictive model to identify patients at higher risk for postoperative pain.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This single-center retrospective study analyzed 17,200 primary unilateral TKAs (2021\u20132024), randomly divided into 80% training (n=13,760) and 20% testing (n=3,440) sets. Pain scores (Numeric Rating Scale (NRS) 0\u201310) collected 0\u201372\u2009hours postoperatively were modeled using long short-term memory with contrastive learning and K-means clustering. Least Absolute Shrinkage and Selection Operator regression was applied to 107 preoperative and intraoperative variables for feature selection, and selected variables were used to train an eXtreme Gradient Boosting (XGBoost) model. Model performance was assessed using accuracy and area under the receiver operating characteristic curve (ROC-AUC).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Two distinct pain archetypes were identified on the training set: a \u201clow pain\u201d (n=7,082) and \u201chigh pain\u201d cluster (n=6,678). The high pain cluster had higher mean postoperative NRS scores, greater cumulative pain burden, higher opioid consumption, and more chronic pain consultations. XGBoost predicted high-pain cluster membership with 64% accuracy and ROC-AUC of 0.68, with key predictors included younger age, ambulatory surgery, higher mode of tolerable NRS score, genicular block, and higher Patient-Reported Outcomes Measurement Information System-10 pain scores. Clustering using NRS data from only the first 12 or 24 postoperative hours showed moderate concordance with 72-hour results (61.3% and 62.4%, respectively).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Machine learning identified distinct postoperative pain trajectories after TKA, and early pain data predicted later pain patterns. Incorporating early pain profiles into perioperative care may support proactive, individualized pain management.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/rapm-2026-107626","type":"journal-article","created":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T05:20:16Z","timestamp":1776316816000},"page":"rapm-2026-107626","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":1,"title":["Classification and stratification of patient pain archetypes following total knee arthroplasty: a machine learning approach"],"prefix":"10.1136","author":[{"given":"Justin","family":"Chew","sequence":"first","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"},{"name":"Department of Anesthesiology, Perioperative Care, and Pain Management, NYU Langone Health, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Junying","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renee","family":"Ren","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"},{"name":"Pain Prevention Research Center, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Maalouf","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9093-0030","authenticated-orcid":false,"given":"Stavros G","family":"Memtsoudis","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7058-5102","authenticated-orcid":false,"given":"Jashvant","family":"Poeran","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-9839-4553","authenticated-orcid":false,"given":"Alexandra","family":"Sideris","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"},{"name":"Pain Prevention Research Center, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1029-2786","authenticated-orcid":false,"given":"Jiabin","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Anesthesiology, Critical Care and Pain Management, Hospital for Special Surgery, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,4,15]]},"reference":[{"key":"2026041522201289000_rapm-2026-107626v1.1","doi-asserted-by":"crossref","first-page":"2781","DOI":"10.1016\/j.arth.2025.05.124","article-title":"Trends in Medicare Arthroplasty Procedure Volume: Projecting From 2025 to 2040","volume":"40","author":"Jones","year":"2025","journal-title":"J Arthroplasty"},{"key":"2026041522201289000_rapm-2026-107626v1.2","doi-asserted-by":"publisher","DOI":"10.1097\/AAP.0b013e318251b688"},{"key":"2026041522201289000_rapm-2026-107626v1.3","doi-asserted-by":"publisher","DOI":"10.1111\/papr.12311"},{"key":"2026041522201289000_rapm-2026-107626v1.4","doi-asserted-by":"crossref","DOI":"10.1136\/rapm-2018-100036","article-title":"Acute postoperative pain is an independent predictor of chronic postsurgical pain following total knee arthroplasty at 6 months: a prospective cohort study","volume":"44","author":"Buvanendran","year":"2019","journal-title":"Reg Anesth Pain Med"},{"key":"2026041522201289000_rapm-2026-107626v1.5","doi-asserted-by":"publisher","DOI":"10.2147\/JPR.S144066"},{"key":"2026041522201289000_rapm-2026-107626v1.6","doi-asserted-by":"crossref","first-page":"2466","DOI":"10.1016\/j.arth.2021.02.044","article-title":"Early Postoperative Pain After Total Knee Arthroplasty Is Associated With Subsequent Poorer Functional Outcomes and Lower Satisfaction","volume":"36","author":"Lo","year":"2021","journal-title":"J Arthroplasty"},{"key":"2026041522201289000_rapm-2026-107626v1.7","doi-asserted-by":"crossref","first-page":"S169","DOI":"10.1016\/j.arth.2023.04.005","article-title":"Age-Related Differences in Pain, Function, and Quality of Life Following Primary Total Knee Arthroplasty: Results From a FORCE-TJR (Function and Outcomes Research for Comparative Effectiveness in Total Joint Replacement) Cohort","volume":"38","author":"Ayers","year":"2023","journal-title":"J Arthroplasty"},{"key":"2026041522201289000_rapm-2026-107626v1.8","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1093\/pm\/pnac154","article-title":"Predictors of Persistent Post-Surgical Pain Following Total Knee Arthroplasty: A Systematic Review and Meta-Analysis of Observational Studies","volume":"24","author":"Ashoorion","year":"2023","journal-title":"Pain Med"},{"key":"2026041522201289000_rapm-2026-107626v1.9","doi-asserted-by":"publisher","DOI":"10.1007\/s11999-013-2803-3"},{"key":"2026041522201289000_rapm-2026-107626v1.10","doi-asserted-by":"crossref","DOI":"10.1186\/s12891-022-04997-4","article-title":"Does obesity affect patient-reported outcomes following total knee arthroplasty?","volume":"23","author":"Baghbani-Naghadehi","year":"2022","journal-title":"BMC Musculoskelet Disord"},{"key":"2026041522201289000_rapm-2026-107626v1.11","doi-asserted-by":"crossref","first-page":"1812","DOI":"10.2106\/JBJS.17.00022","article-title":"Effect of Obesity on Pain and Functional Recovery Following Total Knee Arthroplasty","volume":"99","author":"Collins","year":"2017","journal-title":"J Bone Joint Surg Am"},{"key":"2026041522201289000_rapm-2026-107626v1.12","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1186\/s13018-024-04778-w","article-title":"Risk factors of chronic postoperative pain after total knee arthroplasty: a systematic review","volume":"19","author":"Li","year":"2024","journal-title":"J Orthop Surg Res"},{"key":"2026041522201289000_rapm-2026-107626v1.13","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0283446","article-title":"Factors correlated with pain after total knee arthroplasty: A systematic review and meta-analysis","volume":"18","author":"Olsen","year":"2023","journal-title":"PLoS One"},{"key":"2026041522201289000_rapm-2026-107626v1.14","doi-asserted-by":"crossref","first-page":"421","DOI":"10.1097\/ALN.0000000000003681","article-title":"Temporal Postoperative Pain Signatures G. Patient and Procedural Determinants of Postoperative Pain Trajectories","volume":"134","author":"Vasilopoulos","year":"2021","journal-title":"Anesthesiology"},{"key":"2026041522201289000_rapm-2026-107626v1.15","doi-asserted-by":"crossref","DOI":"10.1097\/AJP.0000000000001270","article-title":"Factors Associated With Postoperative Pain Trajectory, Nausea, and Vomiting After Total Knee Arthroplasty: A Retrospective Study","volume":"41","author":"Chen","year":"2025","journal-title":"Clin J Pain"},{"key":"2026041522201289000_rapm-2026-107626v1.16","doi-asserted-by":"crossref","first-page":"437","DOI":"10.1016\/j.bja.2023.11.024","article-title":"Preoperative pain screening and optimisation by a perioperative pain service to support complex surgical patients: no patient left behind","volume":"132","author":"Rim","year":"2024","journal-title":"Br J Anaesth"},{"key":"2026041522201289000_rapm-2026-107626v1.17","article-title":"Predicting acute postsurgical pain in the postanesthesia care unit: risk tool development and internal validation","volume":"10","author":"Papadomanolakis-Pakis","year":"2025","journal-title":"Pain Rep"},{"key":"2026041522201289000_rapm-2026-107626v1.18","doi-asserted-by":"publisher","DOI":"10.1097\/j.pain.0000000000001118"},{"key":"2026041522201289000_rapm-2026-107626v1.19","doi-asserted-by":"publisher","DOI":"10.1136\/rapm-2024-105603"},{"key":"2026041522201289000_rapm-2026-107626v1.20","doi-asserted-by":"crossref","first-page":"1386","DOI":"10.1111\/pme.12713","article-title":"Teaching a Machine to Feel Postoperative Pain: Combining High-Dimensional Clinical Data with Machine Learning Algorithms to Forecast Acute Postoperative Pain","volume":"16","author":"Tighe","year":"2015","journal-title":"Pain Med"},{"key":"2026041522201289000_rapm-2026-107626v1.21","doi-asserted-by":"crossref","first-page":"526","DOI":"10.1213\/ANE.0000000000003794","article-title":"Addition of Infiltration Between the Popliteal Artery and the Capsule of the Posterior Knee and Adductor Canal Block to Periarticular Injection Enhances Postoperative Pain Control in Total Knee Arthroplasty: A Randomized Controlled Trial","volume":"129","author":"Kim","year":"2019","journal-title":"Anesth Analg"},{"key":"2026041522201289000_rapm-2026-107626v1.22","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"2026041522201289000_rapm-2026-107626v1.23","unstructured":"Chen T , Kornblith S , Norouzi M , et al . A simple framework for contrastive learning of visual representations. Proceedings of the International Conference on Machine Learning (ICML).1597\u2013607."},{"key":"2026041522201289000_rapm-2026-107626v1.24","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From Local Explanations to Global Understanding with Explainable AI for Trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nat Mach Intell"},{"key":"2026041522201289000_rapm-2026-107626v1.25","unstructured":"S.M. Lundberg SIL . A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems 30. 2017: 4766\u201375."},{"key":"2026041522201289000_rapm-2026-107626v1.26","doi-asserted-by":"crossref","first-page":"1966","DOI":"10.1016\/j.arth.2021.02.005","article-title":"Factors Associated With the Risk of Developing Moderate to Severe Acute Postoperative Pain After Primary Total Knee Arthroplasty: Results From the PAIN OUT Registry","volume":"36","author":"Garc\u00eda-L\u00f3pez","year":"2021","journal-title":"J Arthroplasty"},{"key":"2026041522201289000_rapm-2026-107626v1.27","doi-asserted-by":"crossref","first-page":"5912","DOI":"10.3390\/jcm12185912","article-title":"Age-Related Variations in Postoperative Pain Intensity across 10 Surgical Procedures: A Retrospective Study of Five Hospitals in South Korea","volume":"12","author":"Kim","year":"2023","journal-title":"JCM"},{"key":"2026041522201289000_rapm-2026-107626v1.28","doi-asserted-by":"crossref","first-page":"1322","DOI":"10.1111\/aas.14301","article-title":"Psychological predictors of acute postoperative pain after total knee and hip arthroplasty: A systematic review","volume":"67","author":"Springborg","year":"2023","journal-title":"Acta Anaesthesiol Scand"},{"key":"2026041522201289000_rapm-2026-107626v1.29","doi-asserted-by":"crossref","DOI":"10.1186\/s12871-025-03034-w","article-title":"Development and validation of machine learning models for predicting post-cesarean pain and individualized pain management strategies: a multicenter study","volume":"25","author":"Lv","year":"2025","journal-title":"BMC Anesthesiol"},{"key":"2026041522201289000_rapm-2026-107626v1.30","doi-asserted-by":"crossref","DOI":"10.1177\/17448069251376199","article-title":"Application of XGBoost in the prediction of acute postoperative pain after major noncardiac surgery in older patients","volume":"21","author":"Sun","year":"2025","journal-title":"Mol Pain"}],"container-title":["Regional Anesthesia &amp; Pain Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/rapm-2026-107626","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T05:20:19Z","timestamp":1776316819000},"score":1,"resource":{"primary":{"URL":"https:\/\/rapm.bmj.com\/lookup\/doi\/10.1136\/rapm-2026-107626"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,15]]},"references-count":30,"alternative-id":["10.1136\/rapm-2026-107626"],"URL":"https:\/\/doi.org\/10.1136\/rapm-2026-107626","relation":{},"ISSN":["1098-7339","1532-8651"],"issn-type":[{"value":"1098-7339","type":"print"},{"value":"1532-8651","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,15]]}}}