Can Discharge Radiographs Predict Junctional Complications? A Decision Tree Analysis

Area of Science:

• Spine surgery
• Orthopedic research
• Radiographic analysis

Background:

• Junctional complications like proximal junctional kyphosis (PJK) and proximal junctional failure (PJF) are significant concerns after long spinal fusion surgeries.
• Predictive markers for these complications are crucial for improving patient outcomes and surgical planning.

Purpose of the Study:

• To evaluate the efficacy of standing pre-discharge radiographs in predicting the development of PJK and PJF in adult spinal deformity patients.
• To identify specific radiographic parameters on pre-discharge images that correlate with increased risk of junctional complications.

Main Methods:

• A retrospective cohort study included adult spinal deformity patients who underwent lumbar fusion (≥5 levels, LIV pelvis).
• Standing full-length radiographs were obtained before discharge.
• Chi-square automatic interaction detection (CHAID) decision tree analysis was used to identify thresholds for proximal junctional angle (PJA) and its change (ΔPJA) associated with PJK/PJF risk.

Main Results:

• The study analyzed 117 patients, stratifying them into No PJK (54.7%), PJK (33.3%), and PJF (12.0%) groups.
• No significant differences in discharge alignment or pre- to post-operative changes were found between groups.
• Decision tree analysis identified upper instrumented vertebra (UIV) and ΔPJA as key predictors. Patients with a lower thoracic UIV and ΔPJA < 4.3° had a low risk of PJK/PJF.
• The highest risk group had a lower thoracic UIV, ΔPJA ≥4.3°, and a discharge PJA > 15.5°, with 57.1% developing PJF and 28.6% PJK.

Conclusions:

• Standing pre-discharge radiographs, particularly PJA and ΔPJA, can predict the risk of junctional complications after spinal fusion.
• A specific combination of factors—lower thoracic UIV, ΔPJA ≥4.3°, and discharge PJA > 15.5°—identifies a high-risk group for PJF.