Tao Feng,* Shengyang Jin,* Junjie Niu, Qi Yan, Dawei Song, Jinning Wang

Department of Orthopaedic Surgery, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Dawei Song, Department of Orthopaedic Surgery, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, People’s Republic of China, Tel +86 138 0155 0577, Email sdw_333@163.com Jinning Wang, Department of Orthopaedic Surgery, The First Affiliated Hospital of Soochow University, Suzhou, Jiangsu Province, People’s Republic of China, Tel +86 158 5005 7066, Email wangjinning@suda.edu.cn

Background: Residual low back pain (LBP) is frequently reported after percutaneous kyphoplasty (PKP) for osteoporotic vertebral fractures (OVFs), yet its underlying mechanisms remain unclear. Paravertebral muscles (PVMs) degeneration, particularly fat infiltration and atrophy may contribute to persistent postoperative pain.
Objective: To evaluate the association between PVMs degeneration and residual LBP after PKP and identify imaging-based predictors for risk stratification.
Methods: This retrospective cohort study included 213 patients (mean age 70.88 ± 8.58 years; 82.2% female) with single-level OVFs who underwent PKP between January 2021 and June 2023. Patients with multiple-level fractures, chronic LBP, neurological deficits, prior spinal surgery, incomplete imaging, or inadequate follow-up were excluded. Fat infiltration percentage (FI%) and cross-sectional area of the multifidus (MF), erector spinae (ES), and psoas major (PS) were measured at the L4 level using transverse T2-weighted MRI. Residual LBP was defined as postoperative VAS ≥ 3.5 at 12-month follow-up. Logistic regression and ROC analyses were conducted and appropriate univariate tests (t-test or Mann–Whitney U-test) were performed.
Results: Residual LBP occurred in 13.6% of patients and was associated with higher VBQ scores (3.14 ± 0.38 vs 2.57 ± 0.25, P=0.001), greater postoperative kyphosis (16.03 ± 6.69° vs 6.70 ± 4.80°, P=0.001), increased FI% of ES/MF (57.28 ± 5.63% vs 43.40 ± 14.93%, P=0.001), reduced PS area (10.74 ± 4.23 cm² vs 16.15 ± 3.71 cm², P=0.001), and concentrated cement distribution (11.5% vs 73.6%, P=0.001). Independent predictors included elevated VBQ (OR=85.2, 95% CI 7.006– 1036.458), kyphosis (OR=1.14, 95% CI 1.017– 1.276), FI% of ES/MF (OR=1.082, 95% CI 1.008– 1.160), and PS area (OR=0.509, 95% CI 0.285– 0.910). ROC analysis identified FI% ≥ 49.78% and PS area ≤ 11.937 cm² as optimal cutoffs.
Conclusion: Preoperative magnetic resonance imaging assessment of paravertebral muscle may help identify patients at risk for residual low back pain after kyphoplasty. Incorporating preoperative imaging and postoperative physical therapy referral may improve patient outcomes.

Keywords: paravertebral muscle degeneration, low back pain, osteoporotic vertebral fracture, percutaneous kyphoplasty