Cervical Paraspinal Muscle Morphometry and Mechanical Failure: A Propensity-Matched Analysis

Hardware-related complications are rare but important issues after posterior cervical fusion. Current risk stratification relies on bone quality parameters, but other variables may also contribute to mechanical complications. In lumbar spine surgery, paraspinal muscle morphometry has emerged as a predictor of mechanical failure, though cervical-specific evidence is limited. This study evaluates preoperative cervical paraspinal muscle morphometry in predicting mechanical failure following posterior cervical fusion.

This retrospective propensity score-matched study included 26 patients who underwent multilevel posterior cervical decompression and fusion. Mechanical failure was defined as proximal junctional kyphosis, distal junctional kyphosis, adjacent segment disease, or pseudoarthrosis. Patients with mechanical failure were matched 1:1 with controls based on age, sex, body mass index, smoking status, osteoporosis status, indication, the upper instrumented vertebrae, and lower instrumented vertebrae. Preoperative cervical magnetic resonance imagings from C2 to T1 were analyzed using an automated segmentation. Bilateral measurements included volume and fatty infiltration percentage for semispinalis capitis, semispinalis cervicis/multifidus, and splenius capitis muscles. Paired t-tests were used to compare measurements between groups. Statistical significance was set at P < 0.05.

The mechanical failure group demonstrated significantly lower preoperative total paraspinal volume (188 ± 38 mL vs. 224 ± 46 mL, P = 0.043) and total muscle volume (153 ± 33 mL vs. 184 ± 41 mL, P = 0.044). Individual muscle groups demonstrated volume reductions without reaching significance.

This study showed that preoperative cervical paraspinal muscle volume is associated with increased risk of mechanical failure following posterior cervical decompression and fusion. Routine preoperative magnetic resonance imaging might play an important role in assessing possible hardware outcomes.