Background and Objectives: Epidural analgesia after total knee arthroplasty (TKA) is effective, but its side effects are largely dose-dependent. This study evaluated whether adding intraoperative periarticular multimodal infiltration to epidural patient-controlled analgesia (PCA) reduces epidural drug consumption and side effects. Materials and Methods: In this prospective, single-centre randomised controlled trial (ClinicalTrials.gov NCT06872528, registered 24 February 2025), 72 patients undergoing unilateral TKA under combined spinal–epidural anaesthesia were randomised to epidural PCA plus periarticular infiltration with a modified Ranawat solution (Group EP) or epidural PCA alone (Group E). Eleven patients were withdrawn because of catheter dislodgement, and 61 were analysed (Group EP, n = 31; Group E, n = 30). The primary outcomes were epidural local anaesthetic consumption and the incidence of complications and rescue analgesia. Pain (visual analogue scale, VAS), PCA bolus number, haemodynamics and time to first voiding were also recorded for 48 h. Results: VAS scores did not differ between groups at any time point. Cumulative epidural volume was lower in Group EP at 24 h (188.4 ± 43.4 vs. 213.2 ± 44.5 mL; p = 0.032) but not at 48 h (344.0 ± 52.5 vs. 361.8 ± 57.8 mL; p = 0.213). Epidural volume and bolus number were lower in Group EP during the 2–6 h and 12–24 h intervals. Vomiting was less frequent in Group EP (9.7% vs. 40.0%; p = 0.006). Rescue analgesia, nausea, pruritus, urinary retention, haemodynamics and time to first voiding were similar. Conclusions: Adding periarticular infiltration to epidural PCA did not further reduce pain scores, but it reduced epidural consumption during the first postoperative day and the incidence of vomiting; the difference in epidural consumption was modest and should be confirmed in larger trials.