Article
Novel classifications in kneemorphotypes in 8,739 knees prior to TKA using artificial intelligence
Search Medline for
Authors
Published: | October 23, 2023 |
---|
Outline
Text
Objectives: Recent classifications, as the CPAK (MacDessie et al.) and phenotype classification (Hirschmann et al.), describe the knee morphology based on the bony anatomy of the articulating surfaces. Data on gender specific distribution though is not sufficient in large populations.
The aim of this study was to analyse the preoperative alignment and knee morphology with regards to the aforementioned classifications in long leg radiographs prior to TKA surgery in the biggest dataset so far using Artificial Intelligence.
Methods: In this single centre study, we analysed 8,739 preoperative LLRs (4,143 left, 4,596 right) of 7,456 patients of all total knee arthroplasty surgeries between 2009 and 2021 from out institutional database. The measurements were performed with the AI software LAMA (ImageBiopsy Lab, Vienna) and included the hip-knee-ankle angle, mechanical lateral distal femur angle, mechanical medial proximal tibia angle, mechanical axis deviation, and joint line convergence angle.
Results and conclusion: Varus alignment was more common in men (68.5%, 50.8%) while neutral (19.7%, 23.4%) and valgus (11.8%, 25.8%) alignment was more common in women.
Figure 1 [Fig. 1] shows the distribution of CPAK types.
76.3% of all knees showed an apex distal joint line (CPAK-I[28.1%], -II[27.3%], -III[20.9%]). An apex proximal joint line was only seen in 1.3% of all cases (CPAK-VII[0.3%], -VIII[0.3%], -IX[0.7%]). Men were more common in CPAK-I(38,8%) and CPAK II(27,3%) and women were spread more equally between CPAK-I(22.7%), -II(27.3%) and -III(25.7%) (p<0.001).
The most common combination of femur and tibia types was NEU¬mLDFA0°,NEU¬mMPTA0° (f: 17.3%; m: 17.5%),but men showed more often femoral valgus.Patients with a higher BMI showed a significantly lower age at surgery (R²=0.09, p<0.001). There were significant differences between men and women for all radiographic parameters (p<0.001).
Differences in knee morphology with regards to gender, age and BMI highlight the broad variability in osteoarthritic knees that may influence future surgical planning.