Two-period demand phenotypes of oliceridine-based patient-controlled intravenous analgesia after gynecologic surgery: perioperative correlates of persistent high demand

BackgroundPostoperative opioid requirements vary substantially despite standardized analgesic pathways, and cumulative consumption may obscure temporal heterogeneity. We characterized two-period demand phenotypes of oliceridine-based patient-controlled intravenous analgesia (PCIA) after gynecologic surgery and examined perioperative correlates of consumption.MethodsThis single-center retrospective cohort included adult women who underwent elective gynecologic surgery and received oliceridine-based PCIA in 2025. The protocol-specified primary outcome was cumulative 48-h oliceridine consumption. Multivariable linear regression with heteroscedasticity-consistent type 3 (HC3) robust standard errors evaluated its perioperative correlates. Exploratory post hoc phenotypes were defined from cumulative 0–24 h and incremental 24–48 h consumption using cohort-specific third-quartile thresholds (19.6 and 18.2 mg): low/intermediate, early high, late-increasing, and persistent high demand. Phenotype-specific and high 48-h consumption (>38.1 mg) logistic models and sensitivity analyses assessed consistency.ResultsAmong 325 patients, 224 (68.9%) had low/intermediate demand, 20 (6.2%) early high demand, 20 (6.2%) late-increasing demand, and 61 (18.8%) persistent high demand. In the primary continuous-outcome model (n = 319), each 10-year increase in age was associated with 2.73 mg lower 48-h consumption, whereas each 60-min increase in surgical duration, open surgery, and hysterectomy were associated with 2.73, 6.85, and 5.26 mg higher consumption, respectively. Regional block and intraoperative dexmedetomidine were each associated with reductions of 2.58 and 3.25 mg, respectively. In the exploratory phenotype model, younger age, longer surgery, open surgery, and hysterectomy were associated with persistent high demand, whereas regional block and intraoperative dexmedetomidine were each inversely associated. Compared with neither intervention, the combination of regional block and intraoperative dexmedetomidine was associated with lower odds of persistent high demand (adjusted odds ratio, 0.17; 95% confidence interval, 0.05–0.50) and an adjusted 48-h consumption that was 5.9 mg lower.ConclusionPerioperative correlates of cumulative 48-h oliceridine consumption were broadly concordant with those identified using the exploratory two-period phenotype framework. These findings may support risk-stratified acute pain surveillance, but they require prospective external validation.Study registrationRetrospectively registered with the Chinese Clinical Trial Registry (ChiCTR2600128515) after completion of data collection.