Lightning Podiums: Spinal Potpourri - 803B

Presented by: J. Elysee


J. Elysee(1), R. Lafage(1), S. Pesenti(2), J. Varghese(3), J. Moon(4), P. Zhou(4), J. Tishelman(4), B. Beaubrun(4), D. Vasquez-Montes(4), T. Protopsaltis(4), T. Errico(4), A. Buckland(4), F. Schwab(1), V. Lafage(1)

(1) Hospital for Special Surgery, New York, NY, United States
(2) Hopital d'Enfants de la Timone, Marseille, France
(3) SUNY Downstate Medical Center, New York, NY, United States
(4) New York University Langone Orthopedic Hospital, New York, NY, United States


Background: From a mechanical point of view, it has been demonstrated that the gravity line remained fixed in predefined position over the feet. As a consequence, sagittal plane deformity is associated with recruitment of lower limb compensatory mechanisms, with the PShift being the main regulator of gravity line position.

Purpose: To investigate the relationship between SVA and PShift, and demographic factors influencing this relationship Design: Single center retrospective review

Methods: Patients>18yo were drawn from a database of full body images. After evaluation of the spinopelvic and lower-limb radiographic parameters, correlation analysis was carried out between SVA and PShift as well as between T1SPi and PAngle (the angular version of PShift). Multivariate regression analysis was carried out based on spinal global alignment (SVA, T1SPi) and demographics to predict the global alignment of the lower limbs (Pshift or PAngle)

Results: A total of 853 pts (age=59yo, BMI=25kg/m2, 59% F) were included in the analysis. Mean sagittal spinopelvic parameters were PI-LL=10°±19, PT=21°±11, SVA=44mm±59 and T1SPi=-1.36˚±6.2. The lower limb alignment was: Hip extension=201±11, Knee flexion=6±8, PShift=40mm±43 and PAngle=2.6°±2.8. There was a strong correlation between SVA and PShift (r=0.73 p< 0.01) and between T1SPi and PAngle (r=0.60 p< 0.01). Larger BMI and older age were associated with a larger PShift and PAngle for the same SVA. Multivariate regression analysis (R2=0.56) demonstrated that SVA was the strongest predictor of PShift (Standardized coefficient 0.67), followed by BMI (0.13) and Age (0.10). Analysis of the beta coefficients demonstrated that for a given age and BMI, an increase of 5cm in SVA was associated with an increase of 2.5cm in PShift (49%). Similar results were established between PAngle and T1SPi (Standardize Beta: 0.53), Age (0.22) and BMI (0.16).

Conclusion: Lower limb compensation is directly associated with thoracolumbar spinal alignment and counteracts anterior malalignment by repositioning the gravity line over the feet. Age and BMI also impact lower limb compensation required to maintain a free-standing posture. Patient specific demographics and radiographic parameters influence lower limb compensation.

Summary of multivariate regression analysis for PS