Group A received hydromorphone hydrochloride combined with flurbiprofen axetil via PCIA. Hydromorphone hydrochloride 6 mg (2 mL:2 mg, Hubei Yichang Humanwell Pharmaceutical) + flurbiprofen axetil 50 mg (5 mL:50 mg, Beijing Tide Pharmaceutical) were diluted with normal saline to 100 ml and loaded into a PCIA pump (48-h cycle). The pump was initiated 15 min before surgery, with settings of background infusion 2 ml/h, bolus dose 2 ml, and lockout interval 15 min. Patients could self-administer boluses when the numerical pain rating scale (NRS) score was ≥4, with repeated presses during lockout having no effect. For patients with an NRS score ≥4 after two effective boluses, hydromorphone 1 mg diluted in normal saline to 10 ml was administered as a slow push over 2-3 min.

Group B received on-demand analgesia. When the NRS score was ≥4, flurbiprofen axetil 50 mg was administered intravenously, with intramuscular tramadol 100 mg added if analgesia was inadequate.

Group A received hydromorphone hydrochloride combined with flurbiprofen axetil via PCIA. The pump was initiated 15 min before surgery, with settings of background infusion 2 ml/h, bolus dose 2 ml, and lockout interval 15 min.

The NRS was used to assess pain intensity intraoperatively, immediately postoperatively, and at 1 h, 4 h, 8 h, 12 h, and 24 h postoperatively, with scores ranging 0-10 (0=no pain, 1-3=mild pain, 4-6=moderate pain, 7-10=severe pain).

Total time from puncture initiation to completion of dressing (min).

Adverse events within 24 h postoperatively were recorded, including elevated ALBI scores, ALT and AST >2×ULN, nausea and vomiting, drowsiness and dizziness, pruritus, and urinary difficulty, with related indicator changes and incidence documented.

Fasting venous blood was collected preoperatively and on the second postoperative morning to measure procalcitonin (PCT)(ng/ml) and interleukin-6 (IL-6)(pg/ml) levels.