Patterns and predictors of mechanical complications following 3-column osteotomies for correction of adult thoracolumbar spinal deformity with 2 years of follow up

Wednesday, February 21, 2024

Presenting Author(s)

Winward Choy, MD

Resident
University of California, San Francisco
San Francisco, CA, US

Co-Author(s)

Christopher P. Ames, MD

Neurosurgeon
University of California, San Francisco
San Francisco, California, United States
TC

Tony Catalan, B.S.

Researcher
University of California, San Francisco, United States
PC

Ping-Yeh Chiu, MD

Professor
Chang Gung Memorial Hospital, United States
AC

Aaron J. Clark, MD, PhD

Associate Professor
UCSF
San Francisco, California, United States
VD

Vedat Deviren, MD

Professor
University of California, San Francisco
San Francisco, California, United States
Darryl Lau, MD

Neurosurgeon, Director of Spinal Neurosurgery Research
NYU Langone
New York, NY, US
Austin Lui, M.S. (he/him/his)

Medical Student
Touro University California
Union City, California, United States
David J. Mazur-Hart, MD (he/him/his)

Clinical Fellow
University of California, San Francisco
San Francisco, CA, US

Disclosure(s):

Winward Choy, MD: No financial relationships to disclose

David J. Mazur-Hart, MD: No financial relationships to disclose

Introduction: Correction of fixed thoracolumbar adult spinal deformities (ASD) may require three column osteotomies to restore adequate spinal alignment. However, mechanical complications (MC) remain significant following ASD surgery. We report the outcomes of patients with thoracolumbar deformities undergoing 3-column osteotomies.

Methods: Consecutive patients who underwent 3-column osteotomies from 2006 to 2021 performed by the senior author were identified. Those with a minimum of 2 years of clinical and radiographic follow up were included. Main outcome comprised MC, defined by rod fracture, proximal junctional failure (PJF)/distal junctional failure (DJF), or other implant related complications. Univariate and multivariate analysis of clinical, radiographic and operative features was used to identify predictive factors.

Results: 253 patients follow up were included. Mean age was 63 years, 36.8% of patients were male, 71.9% of patients had a previous fusion. Mean number of levels fused was 12.5. The most common site of the 3-column osteotomy was at L3 in 30.8% and L4 in 33.2% of patients. 13.8% of patients were smokers. Mean follow up was 38 months. Rates of MC were 59.3% and overall reoperation rate was 31.7%. The most common types of MC were rod fracture (32.5%), pseudoarthrosis (21.4%) and proximal junctional failure (10.7%). Reoperation was required in 52% of all MC. Rates of rod fracture and PJF have significantly decreased in the last 5 years of the series. Only half of MC occurred within 2 years of index operation, and rate of mechanical complications increases with length of follow up. On multivariate analysis, ligamentoplasty and use of multi-rod constructs were correlated with decreased rates of MC.

Conclusion : Mechanical complications following 3-column osteotomy are significant, and require long term follow up for detection. However, advancements in surgical techniques have significantly decreased rates of MC.