Reliability and interobserver variability in radiological patellar height ratios

Area of Science:

• Orthopedics
• Radiology
• Biomechanics

Background:

• Patellar height is crucial for diagnosing knee conditions.
• Existing radiographic ratios for patellar height assessment include Insall-Salvati (IS), modified Insall-Salvati (MIS), Blackburne-Peel (BP), Caton-Deschamps (CD), and Labelle-Laurin (LL).
• Variability in classification may arise from differing normative data and patient anatomy.

Purpose of the Study:

• To evaluate the reliability and interobserver variability of five common patellar height ratios.
• To compare the classification of patellar height (alta, baja, norma) using these different radiographic methods.
• To recommend the most reliable and discriminatory method for assessing patellar height.

Main Methods:

• Two examiners measured five patellar height ratios (IS, MIS, BP, CD, LL) on lateral knee radiographs of 22 knees.
• Radiographs were taken at specific knee flexion angles (20 degrees for IS, MIS, BP, CD; 90 degrees for LL).
• Interobserver variability was assessed using correlation coefficients and mean errors.

Main Results:

• All five ratios demonstrated good interobserver reliability with high correlation coefficients (0.81-0.92).
• However, classification of patellar height (alta, baja, norma) varied significantly across methods in 68% of patients.
• The Blackburne-Peel (BP) ratio showed intermediate results and the best discrimination between patellar height categories.

Conclusions:

• While interobserver reliability is high for common patellar height ratios, the classification of patellar height is highly dependent on the chosen index.
• The Blackburne-Peel (BP) method is recommended due to its lowest interobserver variability and best ability to discriminate between patellar height groups.
• Future recommendations suggest using ratios that relate the patellar articular surface to the joint line.