Identification of high-risk preoperative blood indicators and baseline characteristics for multiple postoperative complications in rheumatoid arthritis patients undergoing total knee arthroplasty: a multi-machine learning feature contribution analysis_Chinese Journal of Reparative and Reconstructive Surgery

Authors：

ZHU Kejia ¹ , HUANG Zhiyang ¹ , WANG Biao ² , LI Hang ¹ , WU Yuangang ¹ ,  SHEN Bin ¹ ,  NIE Yong ¹

1. Department of Orthopedics, Orthopedic Research Institute, West China Hospital, Sichuan University, Chengdu Sichuan, 610041, P. R. China;
2. West China Biomedical Big Data Center, West China Hospital, Sichuan University, Chengdu Sichuan, 610041, P. R. China;

Corresponding author：

SHEN Bin, Email: shenbin_1971@163.com; NIE Yong, Email: nieyong1983@wchscu.cn

Keywords：

Rheumatoid arthritis; total knee arthroplasty; postoperative complications; preoperative risk factors; machine learning; blood biomarkers

DOI：

10.7507/1002-1892.202508008

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Objective To explore, identify, and develop novel blood-based indicators using machine learning algorithms for accurate preoperative assessment and effective prediction of postoperative complication risks in patients with rheumatoid arthritis (RA) undergoing total knee arthroplasty (TKA). Methods A retrospective cohort study was conducted including RA patients who underwent unilateral TKA between January 2019 and December 2024. Inpatient and 30-day postoperative outpatient follow-up data were collected. Six machine learning algorithms, including decision tree (DT), random forest (RF), logistic regression (LR), support vector machine (SVM), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM), were used to construct predictive models. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), F1-score, accuracy, precision, and recall. SHAP values were employed to interpret and rank the importance of individual variables. Results According to the inclusion criteria, a total of 1 548 patients were enrolled. Ultimately, 18 preoperative indicators were identified as effective predictive features, and 8 postoperative complications were defined as prediction labels for inclusion in the study. Within 30 days after surgery, 453 patients (29.2%) developed one or more complications. Considering overall accuracy, precision, recall, and F1-score, the random forest model [AUC=0.930, 95%CI(0.910, 0.950)] and the extreme gradient boosting model [AUC=0.909, 95%CI(0.880, 0.938)] demonstrated the best predictive performance. SHAP analysis revealed that anti-cyclic citrullinated peptide antibody, C-reactive protein, rheumatoid factor, interleukin-6, body mass index, age, and smoking status made significant contributions to the overall prediction of postoperative complications. Conclusion Machine learning-based models enable accurate prediction of postoperative complication risks among RA patients undergoing TKA. Inflammatory and immune-related blood biomarkers, such as anti- cyclic citrullinated peptide antibody, C-reactive protein, and rheumatoid factor, interleukin-6, play key predictive roles, highlighting their potential value in perioperative risk stratification and individualized management.

1.	Zhang AR, Cheng QH, Yang YZ, et al. Meta-analysis of outcomes after total knee arthroplasty in patients with rheumatoid arthritis and osteoarthritis. Asian J Surg, 2024, 47(1): 43-54.
2.	Jauregui JJ, Kapadia BH, Dixit A, et al. Thirty-day complications in rheumatoid patients following total knee arthroplasty. Clin Rheumatol, 2016, 35(3): 595-600.
3.	Curtis GL, Jawad M, Samuel LT, et al. Incidence, causes, and timing of 30-day readmission following total knee arthroplasty. J Arthroplasty, 2019, 34(11): 2632-2636.
4.	Goodman SM, Springer BD, Chen AF, et al. 2022 American College of Rheumatology/American Association of Hip and Knee Surgeons Guideline for the Perioperative Management of Antirheumatic Medication in Patients With Rheumatic Diseases Undergoing Elective Total Hip or Total Knee Arthroplasty. Arthritis Rheumatol, 2022, 74(9): 1464-1473.
5.	Fraenkel L, Bathon JM, England BR, et al. 2021 American College of Rheumatology Guideline for the Treatment of Rheumatoid Arthritis. Arthritis Rheumatol, 2021, 73(7): 1108-1123.
6.	Goodman SM, Springer B, Guyatt G, et al. 2017 American College of Rheumatology/American Association of Hip and Knee Surgeons Guideline for the Perioperative Management of Antirheumatic Medication in Patients With Rheumatic Diseases Undergoing Elective Total Hip or Total Knee Arthroplasty. Arthritis Care Res (Hoboken), 2017, 69(8): 1111-1124.
7.	Sun Z, Hesler BD, Makarova N, et al. The association between rheumatoid arthritis and adverse postoperative outcomes: a retrospective analysis. Anesth Analg, 2016, 122(6): 1887-1893.
8.	Baek JH, Lee SC, Kim JW, et al. Inferior outcomes of primary total knee arthroplasty in patients with rheumatoid arthritis compared to patients with osteoarthritis. Knee Surg Sports Traumatol Arthrosc, 2022, 30(8): 2786-2792.
9.	Centers for Medicare & Medicaid Services (CMS), HHS. Medicare program; revisions to payment policies under the Physician Fee Schedule, Clinical Laboratory Fee Schedule, access to identifiable data for the Center for Medicare and Medicaid Innovation Models & other revisions to Part B for CY 2015. Final rule with comment period. Fed Regist, 2014, 79(219): 67547-8010.
10.	Li J, Li Y, Huang Y, et al. Predictors and complications of blood transfusion in rheumatoid arthritis patients undergoing total joint arthroplasty. Clin Rheumatol, 2023, 42(1): 67-73.
11.	Castano-Betancourt MC, Fruschein Annichino R, de Azevedo E Souza Munhoz M, et al. Identification of high-risk groups for complication after arthroplasty: predictive value of patient's related risk factors. J Orthop Surg Res, 2018, 13(1): 328. doi: 10.1186/s13018-018-1036-2.
12.	Buus AAØ, Udsen FW, Laugesen B, et al. Patient-reported outcomes for function and pain in total knee arthroplasty patients. Nurs Res, 2022, 71(5): E39-E47.
13.	Harris AH, Kuo AC, Bowe T, et al. Prediction models for 30-day mortality and complications after total knee and hip arthroplasties for veteran health administration patients with osteoarthritis. J Arthroplasty, 2018, 33(5): 1539-1545.
14.	Shah AA, Devana SK, Lee C, et al. Development of a novel, potentially universal machine learning algorithm for prediction of complications after total hip arthroplasty. J Arthroplasty, 2021, 36(5): 1655-1662.
15.	Harris AHS, Kuo AC, Weng Y, et al. Can machine learning methods produce accurate and easy-to-use prediction models of 30-day complications and mortality after knee or hip arthroplasty? Clin Orthop Relat Res, 2019, 477(2): 452-460.
16.	Mohammed H, Huang Y, Memtsoudis S, et al. Utilization of machine learning methods for predicting surgical outcomes after total knee arthroplasty. PLoS One, 2022, 17(3): e0263897. doi: 10.1371/journal.pone.0263897.
17.	Blackburn AZ, Katakam A, Roberts T, et al. Visceral fat as a risk factor for periprosthetic joint infection after total hip and knee arthroplasty. J Arthroplasty, 2023, 38(9): 1839-1845.
18.	Kolin DA, Lyman S, Della Valle AG, et al. Predicting postoperative anemia and blood transfusion following total knee arthroplasty. J Arthroplasty, 2023, 38(7): 1262-1266.
19.	Konzett V, Aletaha D. Management strategies in rheumatoid arthritis. Nat Rev Rheumatol, 2024, 20(12): 760-769.
20.	Sutton JC, Antoniou J, Epure LM, et al. Hospital discharge within 2 days following total hip or knee arthroplasty does not increase major-complication and readmission rates. J Bone Joint Surg (Am), 2016, 98(17): 1419-1428.
21.	McGrory BJ, Weber KL, Jevsevar DS, et al. Surgical Management of Osteoarthritis of the Knee: Evidence-based Guideline. J Am Acad Orthop Surg, 2016, 24(8): e87-e93.
22.	Lundberg S, Lee SI. A unified approach to interpreting model predictions [EB/OL]. (2017-12-04)[2025-10-28]. https://dl.acm.org/doi/10.5555/3295222.3295230.
23.	Jo C, Ko S, Shin WC, et al. Transfusion after total knee arthroplasty can be predicted using the machine learning algorithm. Knee Surg Sports Traumatol Arthrosc, 2020, 28(6): 1757-1764.
24.	Hinterwimmer F, Lazic I, Langer S, et al. Prediction of complications and surgery duration in primary TKA with high accuracy using machine learning with arthroplasty-specific data. Knee Surg Sports Traumatol Arthrosc, 2023, 31(4): 1323-1333.
25.	Jung JW, Hwang S, Ko S, et al. A machine-learning model to predict postoperative delirium following knee arthroplasty using electronic health records. BMC Psychiatry, 2022, 22(1): 436. doi: 10.1186/s12888-022-04067-y.
26.	Shohat N, Ludwick L, Sherman MB, et al. Using machine learning to predict venous thromboembolism and major bleeding events following total joint arthroplasty. Sci Rep, 2023, 13(1): 2197. doi: 10.1038/s41598-022-26032-1.
27.	Tran P, Knecht S, Tamine L, et al. Risk prediction of kalaemia disturbance and acute kidney injury after total knee arthroplasty: use of a machine learning algorithm. Orthop Traumatol Surg Res, 2025, 111(1): 103958. doi: 10.1016/j.otsr.2024.103958.
28.	van de Stadt LA, Witte BI, Bos WH, et al. A prediction rule for the development of arthritis in seropositive arthralgia patients. Ann Rheum Dis, 2013, 72(12): 1920-1926.
29.	Aletaha D, Alasti F, Smolen JS. Rheumatoid factor determines structural progression of rheumatoid arthritis dependent and independent of disease activity. Ann Rheum Dis, 2013, 72(6): 875-880.
30.	Jämsen E, Huhtala H, Puolakka T, et al. Risk factors for infection after knee arthroplasty. A register-based analysis of 43, 149 cases. J Bone Joint Surg (Am), 2009, 91(1): 38-47.
31.	Westerlind H, Kastbom A, Rönnelid J, et al. The association between autoantibodies and risk for venous thromboembolic events among patients with rheumatoid arthritis. Rheumatology (Oxford), 2023, 62(6): 2106-2112.
32.	Mackey RH, Kuller LH, Deane KD, et al. Rheumatoid arthritis, anti-cyclic citrullinated peptide positivity, and cardiovascular disease risk in the women's health initiative. Arthritis Rheumatol, 2015, 67(9): 2311-2322.
33.	Zhang J, Chen L, Delzell E, et al. Republished: The association between inflammatory markers, serum lipids and the risk of cardiovascular events in patients with rheumatoid arthritis. Postgrad Med J, 2014, 90(1070): 722-729.
34.	Navarro-Millán I, Yang S, DuVall S L, et al. Association of hyperlipidaemia, inflammation and serological status and coronary heart disease among patients with rheumatoid arthritis: data from the National Veterans Health Administration. Ann Rheum Dis, 2016, 75(2): 341-347.
35.	Myasoedova E, Crowson CS, Kremers HM, et al. Lipid paradox in rheumatoid arthritis: the impact of serum lipid measures and systemic inflammation on the risk of cardiovascular disease. Ann Rheum Dis, 2011, 70(3): 482-487.
36.	Su X, Zhu B, Qin L, et al. Joint fluid interleukin-6 combined with the neutral polymorphonuclear leukocyte ratio (PMN%) as a diagnostic index for chronic periprosthesis infection after arthroplasty. J Orthop Traumatol, 2023, 24(1): 34. doi: 10.1186/s10195-023-00712-8.
37.	Magzal F, Sela S, Szuchman-Sapir A, et al. In-vivo oxidized albumin- a pro-inflammatory agent in hypoalbuminemia. PLoS One, 2017, 12(5): e0177799. doi: 10.1371/journal.pone.0177799.
38.	Mori C, Ribsam V. Best Practice Guideline Total Knee Replacement (Arthroplasty). 2018. https://docslib.org/doc/4697832/best-practice-guideline-total-knee-replacement-arthroplasty#google_vignette.
39.	Villa-Blanco J I, Calvo-Alén J. Elderly onset rheumatoid arthritis: differential diagnosis and choice of first-line and subsequent therapy. Drugs Aging, 2009, 26(9): 739-750.

1. Zhang AR, Cheng QH, Yang YZ, et al. Meta-analysis of outcomes after total knee arthroplasty in patients with rheumatoid arthritis and osteoarthritis. Asian J Surg, 2024, 47(1): 43-54.
2. Jauregui JJ, Kapadia BH, Dixit A, et al. Thirty-day complications in rheumatoid patients following total knee arthroplasty. Clin Rheumatol, 2016, 35(3): 595-600.
3. Curtis GL, Jawad M, Samuel LT, et al. Incidence, causes, and timing of 30-day readmission following total knee arthroplasty. J Arthroplasty, 2019, 34(11): 2632-2636.
4. Goodman SM, Springer BD, Chen AF, et al. 2022 American College of Rheumatology/American Association of Hip and Knee Surgeons Guideline for the Perioperative Management of Antirheumatic Medication in Patients With Rheumatic Diseases Undergoing Elective Total Hip or Total Knee Arthroplasty. Arthritis Rheumatol, 2022, 74(9): 1464-1473.
5. Fraenkel L, Bathon JM, England BR, et al. 2021 American College of Rheumatology Guideline for the Treatment of Rheumatoid Arthritis. Arthritis Rheumatol, 2021, 73(7): 1108-1123.
6. Goodman SM, Springer B, Guyatt G, et al. 2017 American College of Rheumatology/American Association of Hip and Knee Surgeons Guideline for the Perioperative Management of Antirheumatic Medication in Patients With Rheumatic Diseases Undergoing Elective Total Hip or Total Knee Arthroplasty. Arthritis Care Res (Hoboken), 2017, 69(8): 1111-1124.
7. Sun Z, Hesler BD, Makarova N, et al. The association between rheumatoid arthritis and adverse postoperative outcomes: a retrospective analysis. Anesth Analg, 2016, 122(6): 1887-1893.
8. Baek JH, Lee SC, Kim JW, et al. Inferior outcomes of primary total knee arthroplasty in patients with rheumatoid arthritis compared to patients with osteoarthritis. Knee Surg Sports Traumatol Arthrosc, 2022, 30(8): 2786-2792.
9. Centers for Medicare & Medicaid Services (CMS), HHS. Medicare program; revisions to payment policies under the Physician Fee Schedule, Clinical Laboratory Fee Schedule, access to identifiable data for the Center for Medicare and Medicaid Innovation Models & other revisions to Part B for CY 2015. Final rule with comment period. Fed Regist, 2014, 79(219): 67547-8010.
10. Li J, Li Y, Huang Y, et al. Predictors and complications of blood transfusion in rheumatoid arthritis patients undergoing total joint arthroplasty. Clin Rheumatol, 2023, 42(1): 67-73.
11. Castano-Betancourt MC, Fruschein Annichino R, de Azevedo E Souza Munhoz M, et al. Identification of high-risk groups for complication after arthroplasty: predictive value of patient's related risk factors. J Orthop Surg Res, 2018, 13(1): 328. doi: 10.1186/s13018-018-1036-2.
12. Buus AAØ, Udsen FW, Laugesen B, et al. Patient-reported outcomes for function and pain in total knee arthroplasty patients. Nurs Res, 2022, 71(5): E39-E47.
13. Harris AH, Kuo AC, Bowe T, et al. Prediction models for 30-day mortality and complications after total knee and hip arthroplasties for veteran health administration patients with osteoarthritis. J Arthroplasty, 2018, 33(5): 1539-1545.
14. Shah AA, Devana SK, Lee C, et al. Development of a novel, potentially universal machine learning algorithm for prediction of complications after total hip arthroplasty. J Arthroplasty, 2021, 36(5): 1655-1662.
15. Harris AHS, Kuo AC, Weng Y, et al. Can machine learning methods produce accurate and easy-to-use prediction models of 30-day complications and mortality after knee or hip arthroplasty? Clin Orthop Relat Res, 2019, 477(2): 452-460.
16. Mohammed H, Huang Y, Memtsoudis S, et al. Utilization of machine learning methods for predicting surgical outcomes after total knee arthroplasty. PLoS One, 2022, 17(3): e0263897. doi: 10.1371/journal.pone.0263897.
17. Blackburn AZ, Katakam A, Roberts T, et al. Visceral fat as a risk factor for periprosthetic joint infection after total hip and knee arthroplasty. J Arthroplasty, 2023, 38(9): 1839-1845.
18. Kolin DA, Lyman S, Della Valle AG, et al. Predicting postoperative anemia and blood transfusion following total knee arthroplasty. J Arthroplasty, 2023, 38(7): 1262-1266.
19. Konzett V, Aletaha D. Management strategies in rheumatoid arthritis. Nat Rev Rheumatol, 2024, 20(12): 760-769.
20. Sutton JC, Antoniou J, Epure LM, et al. Hospital discharge within 2 days following total hip or knee arthroplasty does not increase major-complication and readmission rates. J Bone Joint Surg (Am), 2016, 98(17): 1419-1428.
21. McGrory BJ, Weber KL, Jevsevar DS, et al. Surgical Management of Osteoarthritis of the Knee: Evidence-based Guideline. J Am Acad Orthop Surg, 2016, 24(8): e87-e93.
22. Lundberg S, Lee SI. A unified approach to interpreting model predictions [EB/OL]. (2017-12-04)[2025-10-28]. https://dl.acm.org/doi/10.5555/3295222.3295230.
23. Jo C, Ko S, Shin WC, et al. Transfusion after total knee arthroplasty can be predicted using the machine learning algorithm. Knee Surg Sports Traumatol Arthrosc, 2020, 28(6): 1757-1764.
24. Hinterwimmer F, Lazic I, Langer S, et al. Prediction of complications and surgery duration in primary TKA with high accuracy using machine learning with arthroplasty-specific data. Knee Surg Sports Traumatol Arthrosc, 2023, 31(4): 1323-1333.
25. Jung JW, Hwang S, Ko S, et al. A machine-learning model to predict postoperative delirium following knee arthroplasty using electronic health records. BMC Psychiatry, 2022, 22(1): 436. doi: 10.1186/s12888-022-04067-y.
26. Shohat N, Ludwick L, Sherman MB, et al. Using machine learning to predict venous thromboembolism and major bleeding events following total joint arthroplasty. Sci Rep, 2023, 13(1): 2197. doi: 10.1038/s41598-022-26032-1.
27. Tran P, Knecht S, Tamine L, et al. Risk prediction of kalaemia disturbance and acute kidney injury after total knee arthroplasty: use of a machine learning algorithm. Orthop Traumatol Surg Res, 2025, 111(1): 103958. doi: 10.1016/j.otsr.2024.103958.
28. van de Stadt LA, Witte BI, Bos WH, et al. A prediction rule for the development of arthritis in seropositive arthralgia patients. Ann Rheum Dis, 2013, 72(12): 1920-1926.
29. Aletaha D, Alasti F, Smolen JS. Rheumatoid factor determines structural progression of rheumatoid arthritis dependent and independent of disease activity. Ann Rheum Dis, 2013, 72(6): 875-880.
30. Jämsen E, Huhtala H, Puolakka T, et al. Risk factors for infection after knee arthroplasty. A register-based analysis of 43, 149 cases. J Bone Joint Surg (Am), 2009, 91(1): 38-47.
31. Westerlind H, Kastbom A, Rönnelid J, et al. The association between autoantibodies and risk for venous thromboembolic events among patients with rheumatoid arthritis. Rheumatology (Oxford), 2023, 62(6): 2106-2112.
32. Mackey RH, Kuller LH, Deane KD, et al. Rheumatoid arthritis, anti-cyclic citrullinated peptide positivity, and cardiovascular disease risk in the women's health initiative. Arthritis Rheumatol, 2015, 67(9): 2311-2322.
33. Zhang J, Chen L, Delzell E, et al. Republished: The association between inflammatory markers, serum lipids and the risk of cardiovascular events in patients with rheumatoid arthritis. Postgrad Med J, 2014, 90(1070): 722-729.
34. Navarro-Millán I, Yang S, DuVall S L, et al. Association of hyperlipidaemia, inflammation and serological status and coronary heart disease among patients with rheumatoid arthritis: data from the National Veterans Health Administration. Ann Rheum Dis, 2016, 75(2): 341-347.
35. Myasoedova E, Crowson CS, Kremers HM, et al. Lipid paradox in rheumatoid arthritis: the impact of serum lipid measures and systemic inflammation on the risk of cardiovascular disease. Ann Rheum Dis, 2011, 70(3): 482-487.
36. Su X, Zhu B, Qin L, et al. Joint fluid interleukin-6 combined with the neutral polymorphonuclear leukocyte ratio (PMN%) as a diagnostic index for chronic periprosthesis infection after arthroplasty. J Orthop Traumatol, 2023, 24(1): 34. doi: 10.1186/s10195-023-00712-8.
37. Magzal F, Sela S, Szuchman-Sapir A, et al. In-vivo oxidized albumin- a pro-inflammatory agent in hypoalbuminemia. PLoS One, 2017, 12(5): e0177799. doi: 10.1371/journal.pone.0177799.
38. Mori C, Ribsam V. Best Practice Guideline Total Knee Replacement (Arthroplasty). 2018. https://docslib.org/doc/4697832/best-practice-guideline-total-knee-replacement-arthroplasty#google_vignette.
39. Villa-Blanco J I, Calvo-Alén J. Elderly onset rheumatoid arthritis: differential diagnosis and choice of first-line and subsequent therapy. Drugs Aging, 2009, 26(9): 739-750.

Chinese Journal of Reparative and Reconstructive Surgery

Latest ArticlesIdentification of high-risk preoperative blood indicators and baseline characteristics for multiple postoperative complications in rheumatoid arthritis patients undergoing total knee arthroplasty: a multi-machine learning feature contribution analysis

Abstract Full text Figures/Tables Video References Cited by

Format

Content