Effects of Unrestricted Kinematically Aligned Total Knee Arthroplasty with a Modified Soft-Tissue Respecting Technique on the Deformity of Limb Alignment in Japanese Patients
| Autor: | Masahiro Ishikawa, Masaaki Ishikawa, Hideaki Nagashima, Shinya Ishizuka, Kazuhiko Michishita, Yoshinori Soda, Takafumi Hiranaka |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2023 |
| Předmět: | |
| Zdroj: | Medicina, Vol 59, Iss 11, p 1969 (2023) |
| Druh dokumentu: | article |
| ISSN: | 1648-9144 1010-660X |
| DOI: | 10.3390/medicina59111969 |
| Popis: | Background and Objectives: Unrestricted kinematic alignment total knee arthroplasty (KA-TKA) with a soft-tissue respecting technique (STRT) is a soft-tissue-dependent tibial resection entailing the restoration of the original soft-tissue tension using ligamentotaxis after resurfacing the femur, based on the concept of restoring the native or pre-osteoarthritis alignment in each patient. However, there is no consensus on the indications of unrestricted KA-TKA with the STRT. We modified the STRT, followed by an investigation of the effects of surgery on the postoperative hip–knee–ankle angle (HKAA). Materials and Methods: We retrospectively analyzed the clinical background data, including the preoperative and postoperative HKAA, of 87 patients who underwent unrestricted KA-TKA with the modified STRT. Univariate and multivariate analyses were performed to determine the factors affecting the postoperative HKAA. A receiver operating characteristic (ROC) curve was plotted to investigate the change in the cut-off values of preoperative HKAA with respect to the safe zone of the postoperative HKAA. We generated two regression models, the linear regression model and generalized additive model (GAM) using machine learning, to predict the postoperative HKAA. Results: Univariate and multivariate analyses revealed the preoperative HKAA as the factor most relevant to the postoperative HKAA. ROC analysis revealed that the preoperative HKAA exhibited a high predictive utility, with a cut-off value of −10°, when the safe range of postoperative HKAA was set at ±5°. The GAM was the superior machine learning model, indicating a non-linear association between the preoperative and postoperative HKAA. Patients with preoperative HKAAs ranging from −18° to 4° were more likely to fall within the ±5° safe range of the postoperative HKAA. Conclusions: The preoperative HKAA influences the postoperative HKAA in unrestricted KA-TKA with the modified STRT. Machine learning using the GAM may contribute to the selection of patients eligible for the surgical approach. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|