Objective To analyze the changes of early ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET-CT) metabolic parameters in patients with diffuse large B-cell lymphoma (DLBCL) before and after chemotherapy, and to systematically evaluate the early predictive efficacy of the above prarmeters for residual lesions. Methods Eighty patients with DLBCL admitted to our hospital from January 2021 to January 2025 were randomly selected as study subjects. All patients received ¹⁸F-FDG PET-CT examination. After 3 months of chemotherapy, patients were assessed according to the Lugano 2014 criteria. Patients without residual lesions were assigned to the complete remission group, while those with residual lesions were assigned to the residual lesion group. Clinical data and ¹⁸F-FDG PET-CT parameters were compared between the two groups. A logistic regression model was used to analyze the relationship between ¹⁸F-FDG PET-CT parameters and residual tumor lesions in DLBCL patients after chemotherapy. The receiver operating characteristic curve was used to analyze the predictive value of early ¹⁸F-FDG PET-CT parameters for residual tumor lesions in patients with DLBCL after chemotherapy. Results Compared with the complete remission group, the residual lesion group had a significantly higher proportion of patients with an international prognostic index score>3, as well as significantly higher maximum standardized uptake value (SUV_max), metabolic tumor volume (MTV), and total lesion glycolysis（TLG）. The early percentage change in the maximum standardized uptake value (ΔSUV_max%) was significantly lower in the residual lesion group (P<0.05). There were no significant differences in age, sex, Ann Arbor stage, serum lactate dehydrogenase elevation, Eastern Cooperative Oncology Group performance status score ≥2, and molecular subtype between the two groups (P>0.05). Logistic regression analysis showed that an international prognostic index score >3 and a high MTV were independent risk factors for residual tumor lesions in DLBCL patients after chemotherapy, while a high ΔSUVmax% was a protective factor (P<0.05). Receiver operating characteristic curve analysis showed that the area under the curve for ΔSUV_max% and MTV alone in predicting residual tumor lesions after chemotherapy for DLBCL was 0.918 (95%CI=0.848-0.989) and 0.926 (95%CI=0.866-0.985), with cutoff values of 75.16% and 13.58 cm³, respectively. The combined prediction using both parameters yielded an area under the curve of 0.960 (95%CI=0.891-0.991), with a sensitivity of 91.38% and a specificity of 95.45%. Conclusion The early ¹⁸F-FDG PET-CT parameters ΔSUV_max% and MTV demonstrate high predictive value for residual tumor lesions in DLBCL patients after chemotherapy. Their combined use has improved predictive efficacy, which can provide a reference for clinical evaluation of chemotherapy efficacy in DLBCL.