Poster Presentation 50th International Society for the Study of the Lumbar Spine Annual Meeting 2024

Introduction: The T1 pelvic angle (T1PA)—angle centered at the bicoxofemoral axis with vertices at the sacral endplate and T1 centroid—has been proposed as a more reliable measure of global sagittal alignment than traditional measures due to its independence from postural compensation. Likewise, other vertebral pelvic angles (VPA) may serve as robust, alternative spinopelvic parameters, especially for adult spinal deformity (ASD). Here, we assess intra- and interobserver agreement and minimum detectable change (MDC) of VPAs, specifically T1, T4, T9, L1, and L4PA, in ASD patients.

Methods: Patients over 18 years old with preoperative standing PA and lateral films without prior fusion surgery and at least one of the following ASD criteria were selected: coronal Cobb angle ≥ 20°, sagittal vertical axis ≥ 5cm, pelvic tilt (PT) ≥ 25°, or thoracic kyphosis (TK) ≥ 60°. Three examiners—one orthopedic surgeon (ground truth, GT) and two medical students—blindly measured T1, T4, T9, L1, and L4PA twice within a four-week window. The intraclass correlation coefficients (ICC) were calculated to assess reliability. ICC ≥ 0.90 was considered excellent. Additionally, we determined the MDC for each VPA, which denotes the smallest detectable change in a true value with 95% confidence while accounting for measurement error.

Results: Thirty-nine consecutive ASD patients were included. Mean age was 65.5±9.1 years and 76.9% were female. Mean spinopelvic parameters (°) were as follows: PT 24.6±9.6, pelvic incidence (PI) 54.2±14.1, lumbar lordosis (LL) 33.2±18.9, PI-LL mismatch 33.2±18.9, TK 25.5±14.3, and coronal max Cobb angle 40.7±15.5. For all VPAs, mean absolute error to GT for each medical student was 2.3-2.5° and 1.3-1.6°, respectively, resulting in an overall mean absolute error to GT of less than 2°. The intraobserver ICC values ranged from 0.90-0.99 for each VPA. For T1, T4, T9, L1, and L4PA, the ICC values for overall interobserver agreement were 0.97, 0.97, 0.96, 0.95, and 0.92, and the MDCs were 5.3°, 5.1°, 4.8°, 4.9°, and 4.1°, respectively (Table).

Discussion: All VPAs demonstrated excellent intra- and interobserver reliability with the MDC calculated to be approximately 5°, indicating at least comparable reliability to traditional spinopelvic parameters using Cobb angle techniques.