Objective To investigate the influencing factors of poor prognosis after transcatheter arterial chemoembolization (TACE) in patients with hepatocellular carcinoma (HCC) based on propensity score matching (PSM). Methods The clinical data of 496 patients with HCC admitted to First Hospital, Zhengzhou University from January 2017 to January 2020 were retrospectively analyzed; there were 305 males and 191 females; they were (58.19±7.24) years old; their body mass index (BMI) was (24.67±2.37) kg/m2. According to the patients' survival after 3 years' follow up, they were divided into a good prognosis group (82 cases) and a poor prognosis group (414 cases). PSM was used to perform 1 : 1 matching between the two groups. The patients' general data [tumor node metastasis (TNM) stage, Child-Pugh classification, maximum tumor diameter, portal vein tumor thrombus, cirrhosis, distant metastasis, arteriovenous fistula, TACE treatment times, lipiodol deposition classification, neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), alanine transaminase (ALT), aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT), and alpha fetoprotein (AFP)] after matching were compared. The independent sample t test was used to compare the measurement data, and the χ2 test was used to compare the count data. The indicators with statistical differences were included into the multivariate COX regression analysis to screen the influencing factors of the poor prognosis after TACE in the HCC patients. Results Before matching, there were no statistical differences in gender, BMI, smoking history, diabetes history, hypertension history, and tumor distribution between the good prognosis group and the poor prognosis group (all P>0.05), while there were statistical differences in age and drinking history (both P<0.05). After PSM matching, 78 cases of HCC patients were obtained from both the good prognosis group and the poor prognosis group, and there were no statistical differences in gender, age, BMI, smoking history, drinking history, diabetes history, hypertension history, and tumor distribution between the two groups (all P>0.05). After matching, there were no statistical differences in liver cirrhosis, TACE treatment times, NLR, PLR, ALT, and AST between the poor prognosis group and the good prognosis group (all P>0.05), but there were statistical differences in the TNM stage Ⅲa [58.97% (46/78) vs. 26.92% (21/78)], Child-Pugh B [62.82% (49/78) vs. 29.49% (23/78)], maximum tumor diameter > 5 cm [47.44% (37/78) vs. 29.49 % (23/78)], multiple tumors [48.72% (38/78) vs. 32.05% (25/78)], portal vein tumor thrombus [41.03% (32/78 ) vs. 17.95%(14/78)], distant metastasis [21.79% (17/78) vs. 8.97% (7/78)], arteriovenous fistula [21.79% (17/78) vs. 5.13% (4/78)], lipiodol deposition type Ⅲ + Ⅳ [38.46% (30/78) vs. 14.10% (11/78)], GGT > 50 U/L [88.46% (69/78) vs. 73.08% (57/78)], and AFP > 400 μg/L [55.13% (43/78) vs. 28.21% (22/78)] between the poor prognosis group and the good prognosis group (χ2=16.351, 17.437, 5.308, 4.500, 9.989, 4.924, 9.299, 11.944, 5.943, and 11.631; all P < 0.05). The COX regression analysis showed that TNM stage Ⅲa (HR=3.504, 95%CI 1.548-7.935), Child-Pugh grade B (HR=4.345, 95%CI 1.889-9.994), maximum tumor diameter > 5 cm (HR=2.084, 95%CI 1.115-3.893), portal vein tumor thrombus (HR=2.547, 95%CI 1.217-5.333), distant metastasis (HR=1.900, 95%CI 1.078-3.348), arteriovenous fistula (HR=2.291, 95%CI 1.163-4.514), lipiodol deposition Ⅲ+Ⅳ type (HR=3.089, 95%CI 1.359-7.023), GGT > 50 U/L (HR=2.206, 95%CI 1.137-4.278), and AFP > 400 μg/L (HR=2.821, 95%CI 1.283-6.202) all were risk factors for the poor prognosis of the HCC patients after TACE (all P<0.05). Conclusion After PSM balanced covariates, TNM stage Ⅲa, Child-Pugh grade B, maximum tumor diameter > 5 cm, portal vein tumor thrombus, distant metastasis, arteriovenous fistula, lipiodol deposition Ⅲ+Ⅳ type, GGT > 50 U/L, and AFP > 400 μg/L all are factors affecting the poor prognosis after TACE in HCC patients.