Fatty infiltration in the multifidus predicts screw-loosening following short-segment decompression and fusion: proof of why we should protect and rehabilitate the paraspinal muscles

Abstract Purpose Screw-loosening is a common instrumentation-related complication following fusion. Patients who present with pain and neurological symptoms due to screw-loosening require revision. It has been reported that fat-infiltrated and/or atrophied paraspinal muscles were associated with low back pain, disability, radiculopathy, and instrumentation-related failures. However, there is limited and conflicting knowledge regarding the association of paraspinal muscles with screw-loosening. In the present study, we aimed to identify whether fatty infiltration in the paraspinal muscles was associated with screw-loosening. Methods A retrospective analysis of the clinical and radiological data of the patients who underwent short-segment decompression and fusion for lumbar spinal stenosis (LSS) at a tertiary spine clinic between 2013 and 2023. Goutallier’s classification system was used for grading fatty infiltration in the paraspinal muscles. Results Patients with screw-loosening had fattier multifidus at the upper lumbar spine (particularly L2-L3, cephalad to the upper instrumented level of L3-L4) compared to those without screw-loosening. In univariate analysis fatty multifidus at L2-L3 level, elder age, and male sex had ORs of 1.509 ( p = 0.008 ), 1.116 ( p = 0.001 ) and 4.702 ( p = 0.004 ), respectively. In multivariate analysis fatty multifidus at L2-L3 level, elder age and male sex had ORs 1.428 ( p = 0.043 ), 1.109 ( p = 0.003 ), and 5.911 ( p = 0.004 ), respectively. Conclusion Fatty infiltration in the multifidus muscle (particularly in the one at the cranial end of the fusion mass) is predictive for screw-loosening following short-segment lumbar decompression and fusion for LSS. Preserving multifidus in subjects is essential to prevent future long-term complications of spine surgery.