Redefining physiological whole-body alignment according to pelvic incidence: normative values and prediction models

Abstract

Background context

Spinopelvic alignment assessment needs to account for pelvic incidence (PI). Purpose

This study aimed at providing normative values for commonly used parameters in whole-body alignment analysis based on PI. Design

Multicentric prospective study. Patient sample

This study included healthy volunteers with full-body biplanar radiograph in free-standing position. Outcome measures

All radiographic data were collected from 3D reconstructions: Sagittal vertical axis (SVA), T1 pelvic angle (TPA), spino-sacral angle (SSA), sagittal odontoid-hip axis angle (ODHA), pelvic parameters, sacro-femoral angle (SFA), knee flexion angle (KFA), ankle flexion angle (AA), Pelvic shift (PSh), lumbar lordosis (LL), thoracic kyphosis (TK) and cervical lordosis (CL). Methods

Population was divided into five groups according to PI. Normative values were described for each group. Linear regressions including age and PI provided prediction formulas for PT, TPA, SSA and SFA. Results

642 subjects were included. Mean age was 37.7 ± 16.3 years (range: 18–90). Mean PI in the cohort was 49.3 ± 9.5°. LL, PT, SFA, SSA and TPA correlated with PI and age. ODHA, TK, CL and the other lower limb parameters were not associated with PI. All normative values across PI groups are provided for segmental, regional and global alignment parameters. Prediction formulas were: PT=-12.7 + 0.38*PI + 0.14*Age, TPA=-16.9 + 0.34*PI + 0.15*Age, SSA = 109.8 + 0.58*PI-0.19*Age, and SFA = 173 + 0.39*PI + 0.11*Age. Conclusions

SSA, PT, TPA and SFA must be assessed according to patient’s PI. This study provides normative values for each PI group, and predictive formulas taking age and PI into account. PI cannot be used to define thoracic and cervical curvatures.