Efficacy and safety of pain management in perioperative period of TACE in patients with HCC: a single-center, parallel-group, randomized trial.

OBJECTIVE: To evaluate the efficacy and safety of proactive perioperative pain management with hydromorphone patient-controlled analgesia (PCA) versus standard-care subcutaneous morphine (SCM) in patients with hepatocellular carcinoma undergoing transarterial chemoembolization (TACE). METHODS: This single-center, parallel-group randomized trial enrolled 110 patients, who were assigned 1:1 to hydromorphone PCA or as-needed SCM. The modified intention-to-treat population included 105 patients (PCA, n = 51; SCM, n = 54). Numerical rating scale (NRS) pain scores were assessed intraoperatively and at 2, 12, 24, and 48 h after TACE. The prespecified primary outcome was the 24-h NRS score. Secondary outcomes included clinically relevant pain events, 48-h pain area under the curve (AUC), rescue opioid use, adverse events, quality of life, and factors associated with breakthrough pain. RESULTS: At 24 h, the mean NRS score was lower in the PCA group than in the SCM group (0.39 ± 0.67 vs. 1.04 ± 1.43; mean difference, -0.64; 95% CI, -1.07, -0.22; Hedges' g = -0.570; p = 0.004). NRS scores ≥4 at any assessment occurred in 21.6% versus 50.0% of patients (risk difference, -28.4 percentage points; 95% CI, -45.9, -11.0; Holm-adjusted p = 0.005). The PCA group also had a lower 48-h pain AUC (20.71 ± 28.11 vs. 46.54 ± 52.33 NRS·h; mean difference, -25.83; 95% CI, -41.93, -9.73; Holm-adjusted p = 0.005) and less rescue opioid use (3.9% vs. 50.0%; risk difference, -46.1 percentage points; 95% CI, -60.4, -31.7; Holm-adjusted p < 0.001). Dizziness was more frequent with PCA (31.4% vs. 3.7%; p < 0.001), whereas overall adverse-event incidence was similar. Quality of life and postoperative hospital stay did not differ significantly. Complete embolization and SCM were associated with breakthrough pain. CONCLUSION: Hydromorphone PCA reduced short-term pain burden and rescue opioid requirements after TACE but increased dizziness. Because absolute differences in mean NRS scores were modest, these findings should be interpreted alongside clinically oriented secondary outcomes and the need for individualized safety monitoring.