Predictors of Distal Adding-on in Thoracic Major Curves With AR Lumbar Modifiers.

Study Design: Retrospective review of prospectively collected data.
Objective: To determine whether the last substantially touched vertebra (LSTV) is a valid lowest instrumented vertebra (LIV) for both Lenke 1 and 2 curve patterns with AR lumbar modifiers, and to identify preoperative risk factors of distal adding-on.
Summary of Background Data: Previous studies have recommended selecting the LSTV as the LIV for Lenke 1AR curves (main thoracic curve with A lumbar modifier and L4 tilt to the right (thoracic overhang/King type IV curve).
Methods: One-hundred sixty patients with a Lenke 1 or 2 curve pattern and AR lumbar modifier who underwent posterior spinal fusion between 2008 and 2012 were reviewed. All patients had minimum 2-year follow up. Patients were identified with distal adding-on between first erect radiographs and 2-year follow up based on previously defined parameters. Factors predictive of the adding-on phenomenon were identified in a multivariate binary logistic regression model.
Results: Twenty-seven patients (17%) were identified as having distal adding-on of their primary thoracic curve; however, only 8 of 89 patients (9%) fused to the LSTV developed adding-on (P = 0.005). Three variables were found to be significant predictors of adding-on: LIV proximal to LSTV (odds ratio, OR 3.63; P = 0.01), Risser zero (OR 4.93; P = 0.02), and C7-CSVL distance <2 cm (OR 3.97; P = 0.01). The risk of adding-on increased as the number of predictors increased from 16% with one risk factor to 80% when all three preoperative risk factors were present (P < 0.001).
Conclusion: Choosing the LSTV as the LIV in Lenke 1 and 2 curve patterns with an AR lumbar modifier significantly decreases the risk of distal adding-on. Skeletally immature patients, those fused short of LSTV, and those with relative coronal balance preoperatively are at increased risk of distal adding-on between the initial postoperative visit and 2-year follow up.
Level of Evidence: 4.