Spinal Implant Density and Postoperative Lumbar Lordosis as Predictors for the Development of Proximal Junctional Kyphosis in Adult Spinal Deformity

OBJECTIVE: To evaluate spinal implant density and proximal junctional kyphosis (PJK) in adult spinal deformity (ASD). METHODS: Consecutive patients with ASD receiving >= 5 level fusions were retrospectively analyzed between 2007 and 2010. Inclusion criteria: ASD, elective fusions, minimum 2-year follow-up. Exclusion criteria: age <18 years, neuromuscular or congenital scoliosis, cervical or cervicothoracic fusions, nonelective conditions (infection, tumor, trauma). Instrumented fusions were classified by the Scoliosis Research Society-Schwab ASD classification. Statistical analysis consisted of descriptives (measures of central tendency, dispersion, frequencies), independent Student t tests, chi(2), analysis of variance, and logistic regression to determine association of implant density [(number of screws + number of hooks)/surgical levels of fusion] and PJK. Mean and median follow-up was 2.8 and 2.7 years, respectively. RESULTS: Eighty-three patients (17 male, 66 female) with a mean age of 59.7 years (standard deviation, 10.3) were analyzed. Mean body mass index (BMI) was 29.5 kg/m(2) (range, 18-56 kg/m(2)) with mean preoperative Oswestry Disability Index of 48.67 (range, 6-86) and mean preoperative sagittal vertical axis of 8.42. The mean levels fused were 9.95 where 54 surgeries had interbody fusion. PJK prevalence was 21.7%, and pseudoarthrosis was 19.3%. Mean postoperative Oswestry Disability Index was 27.4 (range, 0-74). Independent Student t tests showed that PJK was not significant for age, gender, BMI, rod type, mean postoperative sagittal vertical axis, or Scoliosis Research Society-Schwab ASD classification; but iliac fixation approached significance (P = 0.077). Implant density and postoperative lumbar lordosis (LL) were predictors for PJK (P = 0.018 and 0.045, respectively). Controlling for age, BMI, and gender, postoperative LL (not implant density) continued to show significance in multivariate logistic regression model. CONCLUSIONS: PJK, although influenced by a multitude of factors, may be statistically related to implant density and LL.