[RELATED FACTORS ANALYSIS OF DISTAL ADDING-ON PHENOMENON IN MAJOR THORACIC CURVE AFTER POSTERIOR SELECTIVE FUSION]

To identify the prevalence of distal adding-on phenomenon after posterior selective fusion in type Lenke 1A idiopathic scoliosis, to analyze its risk factors so as to find the reasonable choice for lowest instrumented vertebra (LIV).

A retrospective study was made on the clinical data of 43 patients with type Lenke 1A idiopathic scoliosis undergoing posterior selection fusion with pedicle screw instrumentation between July 2011 and December 2015. There were 15 males and 28 females, aged 12-18 years (mean, 16 years). The preoperative Cobb angle was (50.1±11.3)°. The anteroposterior and lateral radiographs were taken at preoperation, immediate after operation, and last follow-up to measure the radiographic parameters. Forty-three patients were divided into adding-on group and control group according to whether or not the occurrence of distal adding-on phenomenon on anteroposterior radiographs of the spine at last follow-up. All the factors that maybe cause distsal adding-on were evaluated by statistical analysis.

The patients were followed up 12-50 months (mean, 26.5 months). At last follow-up, distal adding-on was observed in 10 of 43 patients (23.3%), including 2 males (13.3%) and 8 females (28.6%). Univariate analyses showed following several factors associated with adding-on:preoperative lumbar flexibility, preoperative pelvic tilt in coronal plane, preoperative LIV+1 deviation from center sacral vertical line, L₄ subtype, the difference between LIV and last touching vertebra (LTV) (LIV-LTV), the difference between LIV and lower end vertebra (LEV) (LIV-LEV), and the difference between LIV and stable vertebra (SV) (LIV-SV). The risk factors above were brought into Logistic regression model, the results showed that preoperative LIV+1 deviation (deviation>10 mm, odds ratio=10.812, P=0.026), LIV-LTV (LIV-LTV<1, odds ratio=9.017, P=0.04), and L₄ subtype (1A-R, odds ratio=9.744, P=0.047) were significantly associated with adding-on.

Preoperative LIV+1 deviation of >10 mm, L₄ subtype (1A-R) and LIV-LTV of <1 are independent predictive factors of adding-on after surgery. As high risk of adding-on was closely related with the incorrect fusion level, it plays an important role for surgeons to decide the level of the fusion before surgery thoroughly. The LIV should be extended at least to LTV+1 to avoid adding-on phenomenon.