ObjectiveThis study aimed to compare the diagnostic efficacy of O-(2-18F-fluoroethyl)-l-tyrosine (18F-FET) PET and 2-deoxy-2-[18F]fluoro-d-deoxyglucose (18F-FDG) PET for spinal cord lesions. Materials and methodsPaired preoperative 18F-FDG PET/MRI and 18F-FET PET/MRI scans were conducted on patients with suspected spinal cord tumors. Clinical manifestations and PET performance, including SUVmean, SUVmax, TBRmean, TBRmax, metabolic tumor volume (MTV), and total lesion metabolism (TLM), and tumor volume, were compared using group analysis and receiver operating characteristic (ROC) curves. ResultsThirty-five patients were categorized into three groups based on their pathological diagnosis: high-grade tumors (HGTs, n = 6), low-grade tumors (LGTs, n = 19), and non-tumor diseases (NTDs, n = 10). The background SUVmean of 18F-FET PET was significantly lower than that of 18F-FDG PET (p < 0.0001), while the delineated tumor volumes showed no significant difference (p > 0.05). The mass SUVmean, SUVmax, MTV, and TLM values of both 18F-FDG PET and 18F-FET PET were statistically different between HGTs and LGTs (p < 0.05). Similarly, the mass SUVmax, TBRmax, MTV, and TLM values of both 18F-FDG PET and 18F-FET PET, as well as the mass SUVmean of 18F-FET PET, exhibited statistical differences between HGTs and NTDs (p < 0.05). But none were able to distinguish LGTs and NTDs (p > 0.05). Notably, 18F-FET PET provided valuable supporting diagnostic evidence in 1 case of mixed neuronal-glial tumor (MNGT) and 2 cases of intramedullary inflammatory lesions. Optimal cut-off values of all measured parameters for distinguishing tumors and NTDs were determined through ROC analysis. Conclusion18F-FET PET presented comparable diagnostic performance to 18F-FDG PET in differentiating HGTs, LGTs, and NTDs, but exhibited particular utility in MNGT and inflammatory lesions.