Analysis of preoperative imaging data to predict the selection of unicompartmental and total knee prostheses.

Aims Finding an imaging method that can inform preoperative planning and streamline prosthetic selection to improve surgical decision-making between UKA and TKA. Methods This study enrolled 68 patients. 34 patients in experimental group underwent TKA for ACL issues for whom UKA was initially planned preoperatively. A cohort of 34 patients was randomly chosen for control group, consisting of patients scheduled for UKA preoperatively and ultimately undergoing TKA during surgery. Preoperative measurement parameters included the hip-knee-ankle angle(HKAA),H1(the height of the medial tibial intercondylar eminence), H2(the height of the lateral tibial intercondylar eminence) on AP radiographs, R (the ratio of the distance the medial femoral condyle moves posteriorly on the tibial plateau to the anterior-posterior distance from the medial tibial plateau) on a fully extended lateral view of the knee, and the medial joint space width (D) at the valgus stress site. MRI was used to measure the area S and circumference C of the ACL at the widest section of the sagittal plane, and the differences were compared between the two groups. A prediction model was established using univariate and multivariate linear regression analyses. Results There are significant differences in variables H1, H2, and HKAA between the two groups, with no differences in variables D and R. Variable S had a strong positive correlation with H1 and a weak positive correlation with H2，no correlation with HKAA, D, and R. Variable C had a strong positive correlation with H1 but did not correlate with H2, A, D, and R. The correlation between C and H1 is C=38.156+3.521×H1.The predictive model for S with H1 and H2 was S=-40.538+14.416×H1+11.296×H2. Conclusion Patients with lightweight and high intercondylar eminence on knee AP radiographs may be better candidates for UKA than those with lower intercondylar eminence and heavy weight.