CAN THE SAGITTAL PELVIC TILT BE PREDICTED FROM ANTEROPOSTERIOR PELVIC RADIOGRAPHS?

The relevance of the hip-spine interaction in understanding the biomechanical behavior of the hip has led to surgeons assessing spinopelvic characteristics prior to total hip arthroplasty (THA), to identify patients at risk. However, to identify patients with abnormal spinopelvic characteristics, sagittal X-rays of the pelvis and spine are suggested/required, which are associated with increased radiation/cost. The sagittal pelvic tilt is an important parameter to assess as it governs sagittal balance for weight transfer and high values (>19°) have been associated with increased dislocation risk. The aim of this study was to assess whether pre-operative anteroposterior pelvic radiographs (the gold standard of pre-THA) can be used to identify patients with abnormal tilt. In doing so, the secondary aim was to describe features that would allow surgeons to estimate sagittal tilt from AP pelvic radiographs. This is a prospective, IRB-approved, consecutive, case series from a tertiary, academic, center. 300 patients (mean age: 64.8±11.4years; 53.5% females; mean BMI: 28.9±6.0kg/m 2 ), with osteoarthritis, listed for THA underwent detailed pre-operative radiographic evaluations of their hip- and spine anatomy/characteristics. Patients underwent standing and supine AP pelvic radiographs to measure several parameters i. distance between several anatomical landmarks (sacrococcygeal joint, femoral heads, sacro-iliac joint, transischial line, trans-ASIS line, height and width of obturator foramen), which allowed for calculation of various ratios and ii. the sacro-femoral pubic angle (SFP) (Figure 1). SFP has been suggested as an evaluator of sagittal pelvic tilt (SFP-derived-PT = 74 – SFP). On a standing, lateral, spinopelvic radiograph, pelvic tilt (PT – measure of sagittal, weight-transfer, balance) and pelvic incidence (PI – reflects pelvic morphology) were measured. A PT >19° was considered high. The mean standing pelvic tilt in this population was 15.8°±7.6°, 34% had a high pelvic tilt. Moderate correlation was found between pelvic tilt and the vertical distance between symphysis and a line connecting distal end of sacroiliac joint (Fig.1 C) ( rho = −0.410; p<0.001), pubic symphysis to sacroiliac index (PS-SI) (Fig.1 C/D) ( rho = −0.426; p<0.001), ratio between height of foramen obturator and distance between tear drops (Fig.1 E/G) ( rho = 0.455; p<0.001), sacro-femoral-pelvic angle (Fig.1 SFP) ( rho = −0.421;; p<0.001) and vertical distance from symphysis to transischial line (Fig.1 TSITA) ( rho = 0.414; p<0.001). Multivariate analysis with a model predicting 60% of the variation found that E/G, TSITA and sex were the most important predictors. Correlation between PT and SFP-derived PT was moderate ( rho = 0.421; p<0.001). The difference between PT and SFP-derived PT, correlated with PI ( rho = −0.4; p<0.001). An SFP value less than 60° was indicative of risk for high pelvic tilt (sensitivity 88%; specificity 59%). TSITA (vertical distance from symphysis to transischial line) and ratio between height of foramen and distance between tear drops are variables that can help to assess pelvic tilt on standing AP pelvic radiographs. An SFP value <60° can be used as a relevant screening tool for increased pelvic tilt (PT >19°). However, none of the variables on an AP pelvic radiograph strongly correlated with pelvic tilt, showing the need for additional spinopelvic radiographs in at least those with low SFP and patients at-risk of adverse spinopelvic characteristics (age, degenerative lumbar spine or arthrodesis). For any figures or tables, please contact the authors directly.