CT-Based Software-Generated Measurements Permit More Objective Assessments of Arithmetic Hip-Knee-Ankle Axis and...

Area of Science:

• Orthopedic surgery
• Radiology
• Medical imaging analysis

Background:

• Robotic-assisted total knee arthroplasty (RATKA) adoption increases CT scan availability.
• Mako SmartRobotics software uses CT scans for pre-operative planning.
• Key measurements include arithmetic hip-knee-ankle axis (aHKA), joint line obliquity (JLO), and Coronal Plane Alignment of the Knee (CPAK) phenotype.

Purpose of the Study:

• To compare CT-based software measurements with manual long limb radiograph (LLR) measurements.
• To assess the correlation between CT-derived and LLR-derived knee alignment parameters.
• To determine if the Coronal Plane Alignment of the Knee (CPAK) distribution differs between the two imaging modalities.

Main Methods:

• Retrospective radiological study of 500 knees from 367 patients undergoing RATKA.
• Comparison of Stryker Mako CT-based software measurements against manual LLR measurements.
• Statistical analysis of differences and correlations for LDFA, MPTA, aHKA, and JLO.

Main Results:

• Statistically significant differences found in average LDFA, MPTA, aHKA, and JLO measurements between CT and LLR.
• CT measurements showed higher LDFA, lower MPTA, more varus aHKA, and increased apex distal JLO compared to LLRs.
• Moderately good correlations observed for LDFA (r=0.409), MPTA (r=0.683), aHKA (r=0.595), and JLO (r=0.456).
• CPAK distribution differed significantly, with LLRs underestimating varus and JLO, leading to more CPAK Types I and IV with CT measurements.

Conclusions:

• Despite moderately good correlation, significant differences exist between CT-based and LLR-based knee alignment measurements.
• CT-based measurements reveal a different CPAK distribution, potentially impacting surgical planning.
• LLRs tend to underestimate constitutional varus and JLO compared to CT scans in RATKA patients.