PURPOSE: This study aimed to elucidate the correlation between the degree of fat infiltration (FI) in thoracic paraspinal muscles and thoracic vertebral degeneration (TVD). METHODS: This cross-sectional study comprised 474 patients who underwent standard thoracic computed tomography (CT) scans. The FI was quantified as the percentage of adipose tissues within the cross-sectional area of thoracic paraspinal muscles. Thoracic vertebra was assessed in terms of osteoporosis, ossification of the anterior longitudinal ligament (OALL), ossification of the posterior longitudinal ligament (OPLL), intervertebral disc calcification, intervertebral disc cavity, and facet joint osteoarthritis (FJO). Logistic regression, linear regression, subgroup, and receiver operating characteristic (ROC) analyses were assessed to evaluate the association between FI and TVD. RESULTS: Multivariate logistic regression revealed that more severe FI was closely associated with more serious osteoporosis (P < 0.001). Furthermore, after adjusting for only age, higher FI was significantly associated with nastier FJO (P < 0.05). In male patients, severe FI was greatly associated with worse osteoporosis (P < 0.05). In female patients, severe FI maintained close correlations with more severe osteoporosis and FJO (P < 0.05). Furthermore, in patients aged < 60 or ≥ 60 years, higher FI had a strong correlation with more severe osteoporosis (P < 0.001). In patients aged < 60 years, higher FI was associated with worse intervertebral disc calcification, OALL, and FJO (P < 0.05). Meanwhile, in patients aged ≥ 60 years, increased FI was only associated with severe OPLL (P < 0.05). Multivariate linear regression showed that FI negatively correlated with bone mineral density in the general population and different sex and age groups (P < 0.001). ROC analysis indicated that FI could predict the occurrence of TVD (P < 0.05). CONCLUSION: Higher FI is associated with more severe TVD. Studies on TVD are currently limited; therefore, this study enriches the related research on TVD, and our findings would facilitate the early prediction and diagnosis of TVD in clinical practice. Furthermore, our findings indicate that thoracic spine pain (TSP) caused by TVD can be prevented, potentially improving the prognosis of patients with TSP.