Oliceridine- versus sufentanil-based postoperative PCIA after orthopedic fracture surgery: a target trial emulation study

Comparative evidence for postoperative patient-controlled intravenous analgesia (PCIA) after orthopedic fracture surgery is limited. We compared analgesic adequacy as the primary question and documented postoperative nausea and vomiting (PONV) as a secondary tolerability outcome. Using routinely collected electronic clinical and PCIA records, we emulated a target trial in a single-center retrospective cohort. Adults initiating oliceridine plus tropisetron or sufentanil plus tropisetron after return to the ward were followed for 48 h. The primary outcome was inadequate analgesia, defined as movement numerical rating scale (NRS) ≥ 4 in either the nominal 24-hour or 48-hour routine-care assessment window or rescue analgesia outside PCIA. Assuming conditional exchangeability given measured pre-initiation covariates and empirical positivity, baseline confounding was addressed using propensity-score overlap weighting. The comparison included 632 patients: 255 initiated the oliceridine-based regimen and 377 the sufentanil-based regimen. Overlap-weighted inadequate-analgesia rates were 24.38% and 18.87%, respectively (odds ratio, 1.39; 95% CI, 0.78–2.46; risk difference, 5.50% points; 95% CI, − 4.16 to 15.17). Thus, the primary comparison was inconclusive and did not establish equivalence or non-inferiority. Overlap-weighted documented-PONV rates were 3.12% and 9.65% (odds ratio, 0.30; 95% CI, 0.10–0.92). Calendar-time and prescriber-aware sensitivity analyses produced confidence intervals crossing the null. Comparative analgesic adequacy remained inconclusive. The lower documented PONV associated with the oliceridine-based regimen was a secondary, hypothesis-generating finding given sparse events and residual confounding. Because broader opioid-related safety outcomes were not systematically available, this study does not establish an overall safety advantage or net clinical benefit. Prospective studies with adequate precision and standardized multidomain outcome assessment are needed. These findings compare initial PCIA prescription strategies rather than fully observed opioid exposure or isolated drug efficacy.