General Session: Value and Outcomes in Spine Surgery - Hall F
Presented by: C. Bortz
P. Passias(1), S. Horn(1), G. Poorman(1), A. Daniels(2), K. Hamilton(3), H. Kim(1), D. Sciubba(4), J. Smith(5), B. Neuman(6), N. Stekas(1), C. Bortz(1), F. Segreto(1), L. Steinmetz(1), B. Diebo(7), C. Shaffrey(5), V. Lafage(8), R. Lafage(8), T. Protopsaltis(1), C. Ames(9), R. Hart((1)0), G. Mundis((1)(1)), R. Eastlack((1)(1)), International Spine Study Group
(1) New York University Langone Orthopedic Hospital, Division of Spinal Surgery, New York, NY, United States
(2) Brown University Alpert Medical School, Department of Orthopaedic Surgery, Providence, RI, United States
(3) University of Pittsburgh, Department of Neurologic Surgery, Pittsburgh, PA, United States
(4) Johns Hopkins University, Department of Neurosurgery, Baltimore, MD, United States
(5) University of Virginia, Department of Neurosurgery, Charlottesville, VA, United States
(6) Johns Hopkins University, Department of Orthopaedics, Baltimore, MD, United States
(7) University Hospital of Brooklyn, Department of Orthopaedic Surgery, New York, NY, United States
(8) Hospital for Special Surgery, Department of Orthopaedic Surgery, New York, NY, United States
(9) University of California San Francisco, Department of Neurosurgery, San Francisco, CA, United States
(10) Swedish Neuroscience Institute, Department of Orthopaedic Surgery, Seattle, WA, United States
(11) San Diego Center for Spinal Disorders, La Jolla, CA, United States
Introduction: Adult cervical deformities have variable structural causes. Recent findings show that the primary driver (PD) of malalignment, most frequently the apex of the curve, may be important in characterizing cervical deformity (CD) and to include in fusion to achieve post-op alignment and quality of life goals. This study aims to define how PD's improve understanding of the disease's mechanisms and the impact of driver region on re-alignment and outcomes. We hypothesize that there are differences in radiographic correction success and outcomes at 1-year postoperatively based on the location of the primary cervical deformity driver.
Methods: Retrospective review of a prospective cervical deformity database. Inclusion: radiographically defined cervical deformity, >18 years, and 1-year follow-up. PD apex was classified based on spinal region: cervical, cervicothoracic junction (CTJ), thoracic, or spino-pelvic by spine surgeons using a modified Delphi approach. Primary analysis evaluated PD groups meeting alignment goals (by Ames modifiers cSVA, TS-CL, CBVA, and mJOA) and HRQL goals (evaluated by EQ5D, NDI, and mJOA) using t-tests. Secondary analysis grouped interventions according to fusion constructs including the PD apex based on lowest instrumented vertebra: cervical: LIV≤C7, CTJ: LIV≤T3, thoracic: LIV≤T12.
Results: 73 patients (mean age 62.2 years, 60% female) were evaluated with the following PDs of their cervical deformity: cervical 50.6%, CTJ 27.3%, thoracic 13.0%, and spino-pelvic 2.6%. They were corrected with average 7.6 levels fused, 21.9% including three-column osteotomy. Primary analysis comparing different PD groups showed varying changes in deformity. Cervical drivers (n=39) showed the greatest 1Y post-op cervical and global alignment changes (improvement in T1S, CL, cSVA, C1 Slope). Thoracic drivers were more likely to have persistent severe TS-CL modifier grade at 1Y (0=20.0%, +=0.0%, ++=80.0%, p=0.031). Secondary analysis compared fusion constructs between the 68% of patients whose fusion included PD apex to those who did not. Patients whose intervention included the PD apex met quality of life MCID at similar rates: EQ5D (Included: 12% vs. not:9%, p=0.373), mJOA (Included: 8% vs. not: 11%, p=0.404, and NDI (Included: 23% vs. not:32%, p=0.556). However, cervical deformity modifiers were relieved from 'severe' more frequently in patients whose construct included the PD apex: cSVA (Included: 64% vs. not: 40.4%, p=0.006) and Horizontal gaze (Included: 44% vs. not: 21.2%, p=0.006). Thoracic and Spino-pelvic PD apex patients did not significantly improve in any HRQLs when PD apex was not treated.
Conclusions: Structural drivers of cervical deformities have an important effect on treatment rendered and 1 year outcomes following correction. Importantly, patients with thoracic or spino-pelvic drivers not included in the construct result in residual deformity and inferior HRQLs.