The Reliability of the Projection Area Per Length Squared for Measuring Lumbar Lordosis on Lateral Radiographs: A Comparison with Cobb Method

Abstract

Objective: The assessment of the degree of lumbar lordosis in patients with spinal disorders is essential to determine disease progression and the effectiveness of treatment. The aim of this study was to examine the reliability of the projection area per length squared (PAL) for measuring lumbar lordosis on lateral radiographs and to compare it with the Cobb method. Methods: Two independent investigators measured lumbar lordosis twice on 100 lateral radiographs using PAL and Cobb methods. Intra- and interobserver agreements of each radiological method were evaluated using intraclass correlation coefficients (ICC) and Bland-Altman plots. Correlations between the PAL estimations and Cobb angle measurements were tested using the Spearman rank correlation coefficient. Results: Intra- and interobserver agreements for PAL and Cobb methods were excellent with all ICC values>0.976. The Bland-Altman plots indicated strong intra-observer and interobserver concordance in the measurement of the lumbar lordosis using the PAL method. A strong correlation was determined between the PAL and Cobb angle values in the first and second measurements (r=0.825; p<0.001 and r=0.815; p<0.001, respectively). Conclusion: The PAL technique is easy to apply on digital images and provides quantitative information independent of the vertebral surface pathologies of the end vertebrae. It could be used as an alternative and potent diagnostic criterion for evaluating lumbar lordosis.