Recently, robot-assisted thoracic surgery has been increasingly performed for mediastinal disease. However, appropriate postoperative analgesic methods have not been evaluated. We retrospectively studied patients who underwent robot-assisted thoracic surgery for mediastinal disease at a single university hospital between January 2019 and December 2021. Patients were performed either general anesthesia alone, general anesthesia combined with thoracic epidural anesthesia, or general anesthesia combined with ultrasound-guided thoracic block. Patients were divided into three groups [non-block (NB), thoracic epidural analgesia (TEA), and thoracic paraspinal block (TB)] according to postoperative analgesic methods, and they compared with terms of postoperative pain scores by using numerical rating scale (NRS) at 0, 3, 6, 12, 18, 24, and 48 h. Additionally, rescue supplemental analgesic within 24 h, side effects of anesthesia such as respiratory depression, hypotension, postoperative nausea and vomiting, pruritus and urinary retention, time to ambulation after surgery, and hospital stay after surgery were also compared among the three groups. Data from 169 patients (Group NB: 25, Group TEA: 102, and Group TB: 42) were progressed to the analysis. Postoperative pain scale at 6 and 12 h was significantly lower in Group TEA than NB (1.2±1.6 vs. 2.4±1.8, P<0.01; and 1.2±1.5 vs. 2.2±1.7, P=0.018, respectively). There were no differences in pain scores between Groups TB and TEA at any point. The incidence of patients using rescue analgesics within 24 h was significantly different between groups [Group NB: 15/25 (60%), Group TEA: 30/102 (29.4%), Group TB: 25/42 (59.5%), P=0.01]. For postoperative side effects, only the number of patients complaining of postoperative nausea and vomiting for 24 h after surgery differed significantly between groups [Group NB: 7/25 (28%), Group TEA: 19/102 (18.6%), Group TB: 1/42 (2.4%), P=0.01]. TEA provided better analgesia after robot-assisted thoracic surgery for mediastinal disease than NB as indicated by lower pain scores and fewer rescue analgesic requirements. However, the frequency of postoperative nausea and vomiting was lowest in Group TB of all the groups. Thus, TBs might also provide adequate postoperative analgesia following robot-assisted thoracic surgery for mediastinal disease.