Association of preoperative D-dimer-to-albumin ratio and prognostic nutritional index with aseptic wound complications and periprosthetic joint infection after primary total knee arthroplasty: a single-center retrospective cohort study
- 1Department of Orthopedic Surgery, The Affiliated Hospital of Qingdao University
- 2Department of Gynecology, The Women and Children’s Hospital of Qingdao
Abstract
Objective To investigate the associations of the preoperative D-dimer-to-albumin ratio (DAR) and prognostic nutritional index (PNI) with aseptic wound complications (AWC) and periprosthetic joint infection (PJI) after primary total knee arthroplasty (TKA), and to evaluate their within-sample discrimination for distinct postoperative adverse outcomes. Methods This single-center retrospective cohort study included patients who underwent primary unilateral TKA at the Affiliated Hospital of Qingdao University between April 2019 and January 2024. Patients with both AWC and PJI were classified in the PJI group only. Deep venous thrombosis was excluded preoperatively using lower-extremity color Doppler ultrasonography. DAR and PNI were calculated from laboratory tests obtained within 48 h before surgery. AWC was assessed within 30 days after surgery, and PJI was assessed within 2 years according to the 2018 International Consensus Meeting criteria. Multivariable logistic regression, receiver operating characteristic (ROC) analysis with 95% confidence intervals (CIs), DeLong tests, bootstrap internal validation, calibration assessment, and sensitivity analyses were performed. Results A total of 1,197 patients were included, including 45 isolated AWC events and 35 PJI events. After adjustment for age, sex, body mass index, diabetes mellitus, and Charlson Comorbidity Index, preoperative PNI was independently associated with isolated AWC (odds ratio [OR], 0.613; 95% CI, 0.543–0.693; P < 0.001), whereas preoperative DAR was not associated with isolated AWC (OR, 0.952; 95% CI, 0.852–1.064; P = 0.388). In contrast, DAR was independently associated with PJI (OR, 1.454; 95% CI, 1.328–1.592; P < 0.001), whereas PNI was not significantly associated with PJI (OR, 0.915; 95% CI, 0.825–1.015; P = 0.094). PNI showed an AUC of 0.873 (95% CI, 0.792–0.955) for isolated AWC, and DAR showed an AUC of 0.920 (95% CI, 0.865–0.975) for PJI. Bootstrap optimism-corrected AUCs were 0.862 for the adjusted PNI-AWC model and 0.900 for the adjusted DAR-PJI model. Conclusions In this exploratory single-center retrospective cohort, preoperative PNI was more strongly associated with isolated AWC, whereas preoperative DAR was more strongly associated with PJI. Although both markers showed good apparent discrimination and high negative predictive values, the low event numbers, limited positive predictive values, imperfect calibration, and absence of external validation require cautious interpretation. Larger multicenter prospective studies are needed before these cutoffs can be applied as clinical decision thresholds.
Related
- Preoperative nutritional status and its association with postoperative complications and survival in patients undergoing total hip arthroplasty: a retrospective cohort analysis
- Preoperative Staphylococcus aureus nasal screening and decolonization for the prevention of periprosthetic joint infection in elective total joint arthroplasty: A single-center retrospective comparative study.
- A smaller preoperative-to-prosthetic center of rotation shift is associated with better range of motion after bony increased-offset reverse shoulder arthroplasty: A retrospective cohort study of 97 patients by the FP-UCBM Shoulder Group
- Impact of intertrochanteric bone mineral density on the accuracy of preoperative templating in cementless total hip arthroplasty: a retrospective cohort study of 510 hips.
- Association between body mass index and early clinical efficacy and complications after total knee arthroplasty
BibTeX
@article{han2026association,
title = {{Association of preoperative D-dimer-to-albumin ratio and prognostic nutritional index with aseptic wound complications and periprosthetic joint infection after primary total knee arthroplasty: a single-center retrospective cohort study}},
author = {Zhi-Chao Han and Xiao-Kai Liu and Peng-Yuan Su and Yuan-He Wang and Shuo Yuan and Bin-Hai Yu and Zhen Liu and Hong-Yu Wang and Yan Wang and Shao-Qi Tian},
journal = {BMC Musculoskeletal Disorders},
year = {2026},
month = jul,
doi = {10.1186/s12891-026-10158-8},
url = {https://doi.org/10.1186/s12891-026-10158-8},
note = {CC BY},
}
RIS
TY - JOUR
TI - Association of preoperative D-dimer-to-albumin ratio and prognostic nutritional index with aseptic wound complications and periprosthetic joint infection after primary total knee arthroplasty: a single-center retrospective cohort study
AU - Zhi-Chao Han
AU - Xiao-Kai Liu
AU - Peng-Yuan Su
AU - Yuan-He Wang
AU - Shuo Yuan
AU - Bin-Hai Yu
AU - Zhen Liu
AU - Hong-Yu Wang
AU - Yan Wang
AU - Shao-Qi Tian
PY - 2026
DA - 2026/07/01
JO - BMC Musculoskeletal Disorders
DO - 10.1186/s12891-026-10158-8
UR - https://doi.org/10.1186/s12891-026-10158-8
AB - Objective To investigate the associations of the preoperative D-dimer-to-albumin ratio (DAR) and prognostic nutritional index (PNI) with aseptic wound complications (AWC) and periprosthetic joint infection (PJI) after primary total knee arthroplasty (TKA), and to evaluate their within-sample discrimination for distinct postoperative adverse outcomes. Methods This single-center retrospective cohort study included patients who underwent primary unilateral TKA at the Affiliated Hospital of Qingdao University between April 2019 and January 2024. Patients with both AWC and PJI were classified in the PJI group only. Deep venous thrombosis was excluded preoperatively using lower-extremity color Doppler ultrasonography. DAR and PNI were calculated from laboratory tests obtained within 48 h before surgery. AWC was assessed within 30 days after surgery, and PJI was assessed within 2 years according to the 2018 International Consensus Meeting criteria. Multivariable logistic regression, receiver operating characteristic (ROC) analysis with 95% confidence intervals (CIs), DeLong tests, bootstrap internal validation, calibration assessment, and sensitivity analyses were performed. Results A total of 1,197 patients were included, including 45 isolated AWC events and 35 PJI events. After adjustment for age, sex, body mass index, diabetes mellitus, and Charlson Comorbidity Index, preoperative PNI was independently associated with isolated AWC (odds ratio [OR], 0.613; 95% CI, 0.543–0.693; P < 0.001), whereas preoperative DAR was not associated with isolated AWC (OR, 0.952; 95% CI, 0.852–1.064; P = 0.388). In contrast, DAR was independently associated with PJI (OR, 1.454; 95% CI, 1.328–1.592; P < 0.001), whereas PNI was not significantly associated with PJI (OR, 0.915; 95% CI, 0.825–1.015; P = 0.094). PNI showed an AUC of 0.873 (95% CI, 0.792–0.955) for isolated AWC, and DAR showed an AUC of 0.920 (95% CI, 0.865–0.975) for PJI. Bootstrap optimism-corrected AUCs were 0.862 for the adjusted PNI-AWC model and 0.900 for the adjusted DAR-PJI model. Conclusions In this exploratory single-center retrospective cohort, preoperative PNI was more strongly associated with isolated AWC, whereas preoperative DAR was more strongly associated with PJI. Although both markers showed good apparent discrimination and high negative predictive values, the low event numbers, limited positive predictive values, imperfect calibration, and absence of external validation require cautious interpretation. Larger multicenter prospective studies are needed before these cutoffs can be applied as clinical decision thresholds.
SN - 1471-2474
ER -