Intra- and inter-observer reliability of spinal inflection point localization: a cross-sectional radiographic study.

Area of Science:

• Orthopedics
• Radiology
• Biomechanical analysis

Background:

• Spinal inflection point localization is crucial for sagittal alignment analysis.
• Accurate localization impacts biomechanical interpretation and surgical planning.
• Reliability of this measurement by non-specialists is not well-established.

Purpose of the Study:

• To evaluate the intra- and inter-observer reliability of spinal inflection point identification.
• To assess the consistency of this measurement on lateral radiographs in asymptomatic adults.
• To determine the influence of evaluator experience on measurement reliability.

Main Methods:

• Cross-sectional study of 10 asymptomatic adults.
• Lateral and anteroposterior panoramic radiographs analyzed.
• Three orthopedic residents with <2 years experience localized inflection points using Surgimap® software.
• Measurements repeated across three sessions with 3-week intervals.
• Intra- and inter-observer agreement assessed using ICC and Cohen's kappa.

Main Results:

• Most common cervicothoracic inflection point: C7-T1; thoracolumbar: T12-L1.
• Low inter-observer agreement for cervicothoracic (kappa 0.19-0.41) and thoracolumbar (kappa -0.03-0.22) inflection points.
• Low intra-observer reproducibility (ICC 0.29-0.56 for both regions).
• Wide confidence intervals indicated low precision, potentially due to small sample size and evaluator inexperience.

Conclusions:

• Localization of spinal inflection points by non-specialist evaluators demonstrated low reliability.
• Significant variability in both intra- and inter-observer agreement was observed.
• Need for standardized protocols and enhanced training for accurate spinal inflection point identification.