31. Evolution of proximal junctional kyphosis and proximal junctional failure rates over 10 years of enrollment in a prospective multicenter adult spinal deformity database

Abstract

<h3>BACKGROUND CONTEXT</h3> Proximal junctional kyphosis (PJK) is one of the most common complications following adult spinal deformity (ASD) surgery and has been intensively studied over the last decade. <h3>PURPOSE</h3> This study aims to investigate the evolution of PJK rate over a 10-year enrollment period within a prospective database of ASD patients. The hypothesis is that the rate of PJK and PJF will decrease across time. <h3>STUDY DESIGN/SETTING</h3> Retrospective review of prospective multicenter database. <h3>PATIENT SAMPLE</h3> A total of 641 ASD patients with minimum 2-year follow-up. <h3>OUTCOME MEASURES</h3> Rate of PJK and PJF. <h3>METHODS</h3> ASD patients with a fusion extended to pelvis and a minimum of 2-year follow-up were included. To investigate a continuous evolution of the PJK/PJF rate, a moving average of 321 patients was calculated across the entire enrollment period, allowing for 95% CI on rate for an incidence of 30% or less. Radiographic PJK was defined as a PJA>10° at 2Y and ΔPJA>10° between preoperative and 2-year follow-up. PJF was defined as either revised for PJK before 2-year visit or Severe radiographic PJK at 2-year follow-up (PJA>28° & ΔPJA>22°). Logistic regression was used to investigate association between date of surgery and PJK and/or PJF. Comparison of PJK rate, demographics and surgical strategies was performed between the first and second half of the cohort. <h3>RESULTS</h3> A total of 641 patients met inclusion criteria (age: 64±10yo, 78.4% F, BMI: 28.3±5.7). The overall rate of radiographic PJK at 2-year was 48.2% (N=309/641); 12.9% (N=83/641) of the patients developed a PJF, with 31.3% (N=26/83) being revised within 2-year. The stratification by date of surgery (DOS) yielded 2 timeframes: First=Oct 2008 to Jul 2014 (with a median DOS of Sep 2011), and Second=Jun 2014 to Apr 2018 (with a median DOS of Aug 2016). Between these 2 periods, the rate of radiographic PJK demonstrated a nonsignificant decrease from 50.3% to 45.5% (p=0.22) and PJF from 15.0% to 10.9% (p=0.12). Linear interpolation of the rate of PJK and PJF by enrollment using a moving average suggested a decrease of 1.1% PJK per year and 1.0% PJF per year. Over the study period, patients became continuously older (from 62.6±10.3 to 65.5±8.7 p<0.001) and were more likely be classified as a pure sagittal deformity (SRS Type N from 31.2% to 47.4% p<0.001). Across the enrollment period, there was a shift in surgical strategy, with a significant reduction of 3CO use (35.1% vs 20.8% p<0.001), an increase in ALL release (0% vs 18.4% p<0.001), a decrease in the number of levels fused (12.2±3.5 vs 11.7±3.9 p=0.016), and an increase in the use of PJK prophylaxis (31.3% vs 55.1%). Logistical regression analysis demonstrated no significant association between date of surgery and radiographic PJK (p=0.19) or PJF (p=0.39). <h3>CONCLUSIONS</h3> Despite extensive research examining risk factors for PJK/PJF and increasing utilization of intraoperative PJK prophylaxis techniques, the rate of radiographic PJK and/or PJF did not significantly decrease across the 10-year enrollment period of this ASD database. The increase in patient risk factors (older) was probably balanced by changes in surgical strategy and preoperative optimization. Continuous enrollment and follow-up will allow ongoing study which may eventually lead to significant decreases in the rate of PJK and PJF. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.