Validation and clinical application of the ΔC2 pelvic angle - ΔC2 tilt = Δpelvic tilt equation for predicting pelvic tilt in spinal deformity surgery.

Abstract

Objective: Notably, studies have established a consistent link between global sagittal alignment and pelvic tilt (PT) using C2 tilt (C2T) and C2 pelvic angle (C2PA), described by the following equation: C2PA = PT + C2T. The present study aimed to validate the proposed relationship (predicted ΔPT = ΔC2PA - ΔC2T) based on the assumption that patients aim to maintain a neutral C2T. Additionally, this study sought to evaluate the accuracy of intraoperative C2PA measurements for predicting postoperative PT.

Methods: The medical records of patients > 21 years of age undergoing spinal fusion were retrospectively reviewed. Inclusion criteria were spinal instrumentation and fusion extending to the sacrum with upper instrumented vertebrae at or above L3. Patients without complete preoperative and 6-week postoperative anteroposterior and lateral scoliosis radiographs were excluded. Patients were stratified into short (fusion at or below T10) and long thoracolumbar fusions. Pre- and postoperative measurements included the C2-7 Cobb angle (CA), C6-T4 CA, T5-12 CA, T4 pelvic angle, L1 pelvic angle, pelvic incidence, and lumbar lordosis. A subset of patients with intraoperative radiographs visualizing a visible C2 vertebral body and bilateral femoral heads were analyzed for intraoperative prediction accuracy. ΔC2PA was defined as postoperative C2PA - preoperative C2PA, and predicted ΔC2T was defined as 0 - preoperative C2T. The mean absolute error (MAE) was calculated as the mean absolute difference between the predicted and actual postoperative PT values.

Results: In total, 298 patients (mean age 65.4 ± 11.4 years, 71.8% female) met the inclusion criteria; 126 (42.3%) underwent short thoracolumbar fusions, and 172 (57.7%) underwent long thoracolumbar fusions. Preoperatively aligned patients had a mean postoperative C2T of -2.43° ± 2.48°, and preoperatively malaligned patients had a mean postoperative C2T of 0.72° ± 5.32°. The equation demonstrated excellent accuracy in the full cohort, with an MAE of 3.56° and an R2 value of 0.77. Of the total cohort, 69 patients (23.2%) met criteria for intraoperative measurements. Intraoperatively, the equation retained clinical utility (MAE = 5.75°, R2 = 0.576) and maintained high accuracy across stratified analyses by fusion length (MAE in long fusion = 5.89°, R2 = 0.595; MAE in short fusion = 5.31°, R2 = 0.603).

Conclusions: This study validates the equation (predicted ΔPT = ΔC2PA - ΔC2T) as a reliable tool for predicting PT in spinal deformity surgery. The equation's dual functionality as a preoperative planning tool and intraoperative predictive guide underscores its clinical utility.