Efficacy and Safety of Oliceridine for Patient-Controlled Intravenous Analgesia in Elderly Patients Undergoing Laparoscopic Radical Resection Surgery for Gastrointestinal Malignant Tumors: A Prospective Randomized Controlled Non-Inferiority Clinical Trial

Introduction

Laparoscopic radical resection has become the standard approach for the treatment of gastrointestinal malignancies, offering advantages over open surgery, including reduced operative trauma, smaller incisions, and accelerated postoperative recovery.^1,2 Nevertheless, factors inherent to laparoscopic surgery, such as pneumoperitoneum, transient splanchnic ischemia, and procedural stress, frequently elicit acute postoperative pain (APSP), which, if inadequately managed, can lead to adverse outcomes including delayed wound healing, increased risk of infection, and prolonged recovery.^3,4

Patient-controlled intravenous analgesia (PCIA) remains the cornerstone of postoperative pain management.⁵ Sufentanil, a potent synthetic opioid with rapid onset and high analgesic efficacy, is commonly employed in this context. However, its clinical utility is constrained by opioid-related adverse events (ORAEs), including respiratory depression, excessive sedation, nausea, vomiting, pruritus, and reduced gastrointestinal motility.^6,7 These concerns are especially pronounced in elderly patients, who often exhibit diminished organ reserve, slower drug metabolism, and heightened sensitivity to sedative effects, thereby increasing their vulnerability to postoperative complications.⁸

Oliceridine, a novel G protein-biased μ-opioid receptor (MOP) agonist, has emerged as a promising alternative for the management of moderate-to-severe acute pain requiring intravenous opioid therapy.⁹ Unlike conventional opioids, oliceridine preferentially activates G-protein signaling while attenuating β-arrestin recruitment, a pathway implicated in many ORAEs.¹⁰ Preliminary studies have demonstrated a favorable safety profile for oliceridine compared with morphine, and recent trials have begun to evaluate its use against other opioids such as sufentanil. However, comparative data remain limited, particularly in elderly patients undergoing laparoscopic surgery for gastrointestinal malignancies, a population at increased risk for opioid-related complications.

To address this knowledge gap, we conducted a randomized, double-blind, non-inferiority clinical trial comparing the efficacy and safety of oliceridine versus sufentanil for PCIA in elderly patients undergoing laparoscopic radical resection of gastrointestinal malignancies. The non-inferiority margin was prespecified as 20% of the mean AUC in the sufentanil group, a threshold based on prior studies^11,12 and deemed to represent a clinically acceptable difference in analgesic effect. Our objective was to evaluate whether oliceridine offers effective analgesia with a reduced incidence of opioid-related adverse effects in this high-risk surgical population.

Methods

Trial Design and Oversight

We conducted a randomized, double-blind, controlled, non-inferiority clinical trial at Tangdu Hospital, Fourth Military Medical University. Eligible participants were patients undergoing laparoscopic radical surgery for gastrointestinal malignant tumors. The trial protocol was approved by the Ethics Committee of Tangdu Hospital (Approval No. K202411-39) and prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2400090780; https://www.chictr.org.cn/showprojEN.html?proj=242515) on October 13, 2024. The trial was conducted in accordance with the principles of the Declaration of Helsinki. All participants provided written informed consent. Trial conduct and reporting followed the Consolidated Standards of Reporting Trials (CONSORT) guidelines. The first patient was enrolled on December 2024.

Patients

Patients were eligible for inclusion if they were 60 years of age or older, were scheduled to undergo elective laparoscopic radical resection of gastrointestinal tumors under general anesthesia, had an American Society of Anesthesiologists (ASA) physical-status classification of I to III, and had a body-mass index (BMI) of 18 to 30 kg/m². Key exclusion criteria were serious systemic or hematologic disease or clinically significant hepatic or renal dysfunction; the presence of another malignant tumor or organ metastases; known allergy or contraindication to opioid drugs or the study drugs; preexisting chronic pain or neurologic disorders; long-term use or abuse of analgesic or psychiatric medications; use of sedatives, antiemetics, or antipruritic drugs within 24 hours before surgery or antidepressants within 15 days before surgery; and inability to participate because of language barriers or severe cognitive impairment.

Randomization and Blinding

Patients were assigned in a 1:1 ratio to receive oliceridine or sufentanil by means of a random-number table. Allocation assignments were placed in sequentially numbered, sealed, opaque envelopes. Before induction of anesthesia, a study coordinator who was not involved in trial assessments opened the envelope in recruitment order and prepared a PCIA pump containing the assigned study drug; the pumps were identical in appearance. Patients, anesthesiologists, perioperative care staff, and investigators responsible for data collection and follow-up remained unaware of treatment assignments throughout the trial.

Perioperative Management

On the day before surgery, an anesthesiologist assessed each patient, provided an explanation of the trial procedures to patients and their families, and obtained written informed consent. On arrival in the operating room, peripheral venous access was established, and standard monitoring was initiated, including electrocardiography, pulse oximetry, invasive arterial blood pressure, and bispectral index (BIS).

General anesthesia was induced with etomidate (0.3 mg per kilogram of body weight), midazolam (0.1 mg per kilogram), rocuronium (0.6 mg per kilogram), and sufentanil (0.4 to 0.5 μg per kilogram), administered intravenously. After the BIS value decreased to 45 to 55, tracheal intubation was performed with video laryngoscopy and mechanical ventilation was initiated. Ventilation was delivered with a tidal volume of 6 to 8 mL per kilogram, a respiratory rate of 12 to 15 breaths per minute, and an inspiratory-to-expiratory ratio of 1:1.5 to 1:2; settings were adjusted to maintain an end-tidal partial pressure of carbon dioxide of 35 to 45 mm Hg and an oxygen saturation of 98 to 100%.

After induction, bilateral ultrasound-guided transversus abdominis plane block was performed with an in-plane technique; a 20-gauge needle was advanced to the plane between the internal oblique and transversus abdominis muscles and 0.25% ropivacaine was injected (20 mL per side).¹³ Anesthesia was maintained with a continuous infusion of remifentanil (0.1 to 0.2 μg per kilogram per minute) and dexmedetomidine (0.2 μg per kilogram per minute) and inhaled sevoflurane (1 to 3%), with dose adjustment to maintain a BIS value of 40 to 60. Rocuronium and sufentanil were administered intermittently as needed. Hemodynamic stability was maintained throughout the procedure, and vasoactive agents were administered when clinically indicated. The intraoperative doses of sufentanil and remifentanil were recorded for all patients to ensure balance between groups.

Interventions

All patients received ketorolac tromethamine (30 mg intravenously) 30 minutes before wound closure. In the postanesthesia care unit (PACU), patients received a blinded PCIA pump according to trial assignment. In the oliceridine group (Group O), the PCIA solution contained ketorolac tromethamine (120 mg), tropisetron (5 mg), and oliceridine (0.4 mg per kilogram of body weight), diluted with normal saline to a total volume of 150 mL. In the sufentanil group (Group S), the PCIA solution contained ketorolac tromethamine (120 mg), tropisetron (5 mg), and sufentanil (2 μg per kilogram), diluted with normal saline to 150 mL. The PCIA settings were identical in the two groups: a loading dose of 5 mL, background infusion at 2.5 mL per hour, patient-initiated bolus dose of 2.5 mL, a lockout interval of 15 minutes, a maximum delivery of 10 mL per hour, and a treatment duration of 48 hours.

Patients were transferred from the PACU to the ward after full recovery of consciousness with an Aldrete score of more than 9. Rescue analgesia consisted of intravenous flurbiprofen axetil (50 mg in 100 mL of saline) when the VAS pain score was 4 or higher despite appropriate use of the PCIA device. For severe nausea or vomiting, intravenous ondansetron (4 mg) was permitted, and adjustment of the infusion rate or temporary suspension of PCIA was allowed when clinically indicated.

Criteria for withdrawal from the trial were a patient request to discontinue participation, refusal of follow-up, postoperative transfer to the intensive care unit or delayed recovery precluding assessment, conversion to open surgery, or a surgical duration of less than 3 hours or more than 6 hours.

Clinical Outcomes and Assessments

The primary endpoint was the cumulative area under the curve (AUC) for resting visual analogue scale (VAS) scores from T₀ through 48 hours (T₅). Pain intensity was assessed with an 11-point VAS (range, 0 to 10, with 0 indicating no pain and 10 the worst pain imaginable), recorded at extubation (T₀) and at 6 hours (T₁), 12 hours (T₂), 24 hours (T₃), 36 hours (T₄), and 48 hours (T₅) after surgery, both at rest and during movement.

The key secondary efficacy endpoint was the cumulative AUC for movement-evoked VAS pain over the same interval. AUC values were calculated with the trapezoidal rule by summing adjacent measurements over time; for consecutive time points ti < tj with corresponding VAS scores VAS_i and VAS_j, each interval contributed (VAS_i + VAS_j) × (t_j − t_i) / 2 to the total AUC.

Serum markers of the pain-related stress response - substance P (SP), prostaglandin E2 (PGE2), tumor necrosis factor alpha (TNF-α), high-sensitivity C-reactive protein (hs-CRP), and cortisol (Cor) - were measured by radioimmunoassay. Peripheral venous blood samples were obtained before induction of anesthesia and at 6 hours (T₁), 24 hours (T₃), and 48 hours (T₅) after surgery.

Safety was assessed by the incidence of adverse reactions within 48 hours after surgery, including postoperative nausea and vomiting (PONV), pruritus, respiratory depression, hypotension, dizziness or headache, abdominal distention, and drowsiness. Sleep quality was assessed with the Richards–Campbell Sleep Questionnaire (RSCQ) before surgery and at 24 hours (T₃) and 48 hours (T₅). Symptoms of anxiety and depression were assessed with the Hamilton Anxiety Scale (HAMA) and Hamilton Depression Scale (HAMD), respectively, at the same time points.

Analgesic utilization was quantified by recording the number of effective and total PCIA demands, total analgesic consumption, and use of rescue analgesia. Recovery outcomes included time to first flatus, time to start ambulation, time to drainage-tube removal, and length of hospital stay. Patient satisfaction with postoperative analgesia was assessed at 24 hours (T₃) and 48 hours (T₅) on a 4-point scale ranging from 1 (dissatisfied) to 4 (very satisfied).

Sample-Size Determination

This trial was designed to test non-inferiority with respect to the primary endpoint, the cumulative AUC of pain intensity at rest, measured by the VAS pain score, during the first 48 hours after surgery. On the basis of preliminary data, the mean (±SD) 48-hour VAS AUC at rest was 105.84±7.77 in the oliceridine group and 91.48±5.68 in the sufentanil group. The non-inferiority margin was prespecified as 20% of the mean value in the sufentanil group (18.3 points). This margin corresponds to an average difference of less than 0.5 points on the 0–10 VAS scale per assessment point over the 48‑hour period, which is substantially smaller than the minimal clinically important difference of approximately 1.5 points reported for APSP after abdominal surgery.¹⁴ The choice of a 20% threshold is also consistent with previous non‑inferiority trials in postoperative pain management and represents a clinically acceptable difference in analgesic effect.^11,12 Sample-size calculations were performed with PASS software, version 15.0.5, assuming a one-sided type I error of 0.025 and 80% power. Under these assumptions, 76 patients were required. To allow for an anticipated dropout rate of 15%, the target enrollment was 90 patients, assigned in a 1:1 ratio (45 per group).

Statistical Analysis

All statistical analyses were performed with SPSS software, version 27.0, and GraphPad Prism, version 10.1.2. The distribution of continuous variables was assessed with the Kolmogorov–Smirnov test. Normally distributed continuous variables are presented as means (±SD) and were compared between groups with the independent-samples t-test, after confirming homogeneity of variance using Levene’s test. When the assumption was violated, Welch’s t-test was used. For normally distributed outcomes measured repeatedly within participants, comparisons over time were performed with repeated-measures analysis of variance. Sphericity was assessed using Mauchly’s test; when the sphericity assumption was violated, the Greenhouse–Geisser correction was applied.

Continuous variables that were not normally distributed are presented as medians with interquartile ranges and were compared between groups with the Mann–Whitney U-test. For repeated measures with nonnormal distributions, comparisons were performed with generalized estimating equations and the Friedman nonparametric test, as appropriate. Categorical variables are presented as counts and percentages and were compared with the chi-square test or Fisher’s exact test, as appropriate.

Statistical analyses were performed on four predefined populations. Baseline characteristics were compared using the intention‑to‑treat (ITT) population (all 90 randomized patients). The safety population (patients who received at least one dose of the study drug, n=88) was used for all safety analyses. For the primary efficacy analyses, the per‑protocol (PP) population (patients who completed the trial without major protocol violations, n=88) served as the primary analysis set, while a modified intention‑to‑treat (mITT) population (all patients who received the study drug and had at least one post‑baseline pain assessment, n=88) was employed as a sensitivity analysis. The ITT analysis including all 90 randomized patients was not feasible for the primary efficacy endpoint, as the two patients excluded from the PP population never received the study drug and therefore had no post‑baseline pain assessments. As recommended by ICH E9 guidelines, we used a mITT approach that included all patients who received the study drug and had at least one post‑baseline pain assessment. All other secondary efficacy outcomes, including VAS scores at individual time points, analgesic utilization, sleep quality, anxiety and depression scores, and recovery outcomes, were analyzed in the PP population.

For the primary endpoint, non-inferiority was assessed by calculating the between-group difference in the cumulative AUC of resting pain over 48 hours and the corresponding 95% CI. The prespecified noninferiority margin was 20% of the mean value in the sufentanil group, and noninferiority was concluded if the upper bound of the 95% CI was below this margin.

A two-sided P value of less than 0.05 was considered to indicate statistical significance for secondary analyses. For multiple comparisons of secondary endpoints, including the analysis of serum markers and repeated measures, Bonferroni correction was applied to control the Type I error rate. The study was powered for the primary efficacy endpoint; therefore, analyses of safety outcomes and other secondary endpoints should be considered exploratory.

Results

Patients

A total of 90 elderly patients underwent randomization. Two patients discontinued the intervention before receiving the study drug: one in Group S due to intraoperative massive bleeding requiring intensive care, and one in Group O due to conversion to open surgery. These two patients did not receive postoperative patient-controlled analgesia and were excluded from all drug-related analyses. Accordingly, 88 patients received the study drug and were included in the safety and efficacy analyses (44 in each group) (Figure 1). Baseline characteristics of all 90 randomized patients (the ITT population) were well balanced between the two groups (Table 1).

Table 1 Baseline Patient Characteristics and Perioperative Variables

Figure 1 Patient flowchart.

Primary Outcome

Cumulative AUC of Pain at Rest (0 to 48 Hours)

In the PP analysis (88 patients who completed the trial), the mean (±SD) cumulative AUC of VAS pain scores at rest from 0 to 48 hours after surgery was 73.02±22.59 in Group S and 72.55±23.54 in Group O (between-group difference, 0.48; 95% CI, −9.30 to 10.26; P=0.923) (Table 2). The upper bound of the 95% CI for the between-group difference was below the pre-specified non-inferiority margin (18.3), thus establishing non-inferiority of oliceridine versus sufentanil in analgesic efficacy. A mITT analysis including all 88 patients who received the study drug (the same patients as the PP population, since the two excluded patients never received the drug) yielded identical results, further supporting the robustness of the non-inferiority conclusion.

Table 2 Comparison of the Cumulative AUC of the VAS Scores Within 48 h After Surgery, Mean (SD), Score

Cumulative AUC of Pain during Movement (0 to 48 Hours)

The mean (±SD) cumulative AUC during movement was 115.16±24.18 in Group S and 113.39±25.33 in Group O (between-group difference, 1.77; 95% CI, −8.72 to 12.27; P=0.738) (Table 2). No significant between-group differences were observed in the cumulative AUC of pain during movement.

Pain Scores at Prespecified Time Points

VAS pain scores at rest increased at T₂ as compared with T₀ in both groups (each P<0.001). No significant between-group differences were observed in VAS pain scores at any assessed time point either at rest or during movement (all P>0.05) (Table 3).

Table 3 Comparison of the Pain VAS Scores Between the Two Groups at Different Time Points, M (IQR), Score

Postoperative Analgesia and Patient-Reported Outcomes

The effective pressing frequency and total pressing frequency of the PCIA pump were lower in Group O than in Group S (P=0.045 and P=0.030, respectively). Pain-satisfaction scores at T₃ and T₅, total analgesic consumption, and the rate of rescue analgesia did not differ significantly between groups (P>0.05) (Table 4).

Table 4 Comparison of Indicators of Postoperative Analgesia Between the Two Groups

Serum Pain-Stress and Inflammatory Markers

At T₁, serum SP, Cor, and hs-CRP levels were lower in Group O than in Group S (P=0.041, P=0.037, and P<0.001, respectively); at T₃, Cor and hs-CRP levels were lower in Group O than in Group S (P=0.017 and P=0.030, respectively). Relative to preoperative values, SP levels increased in Group S at T₁, T₃, and T₅ (P=0.002, P=0.008, and P=0.011, respectively); PGE2 levels increased in both groups at T₁, T₃, and T₅ (P<0.001 for all comparisons); hs‑CRP levels increased in both groups at T₁, T₃, and T₅, with P values<0.001 for all comparisons except for Group O at T₁ (P=0.018); Cor levels increased in both groups at T₁ (P<0.001 for Group S and P=0.040 for Group O). No significant between-group or within-group differences were observed in TNF-α levels (P>0.05 for all comparisons) (Figure 2).

Figure 2 Serum levels of pain-stress and inflammatory markers at four time points: preoperative baseline (T0), 6 hours (T1), 24 hours (T3), and 48 hours (T5) after surgery. (a) Substance P (SP); (b) Prostaglandin E2 (PGE2); (c) Cortisol (Cor); (d) Tumor necrosis factor-α (TNF-α); (e) High-sensitivity C-reactive protein (hs-CRP). Data are presented as mean (SD). *P < 0.05, vs. Group S; #P < 0.05, vs. preoperative baseline within groups, All pairwise comparisons have been corrected by Bonferroni.

Adverse Events and Use of Rescue Antiemetics

PONV was defined as nausea, retching, or vomiting.¹⁵ Respiratory depression was defined as a respiratory rate of less than 8 breaths per minute and/or an oxygen saturation of less than 90%.¹⁶ Hypotension was defined as a mean arterial pressure of less than 65 mm Hg.¹⁷ Adverse events were assessed in the safety population, which included all 88 patients who received the study drug. The incidence of respiratory depression and hypotension was lower in Group O than in Group S (P=0.035 and P=0.034, respectively). The incidences of PONV, pruritus, dizziness, headache, abdominal distention, and somnolence, as well as the use of rescue antiemetic therapy, did not differ significantly between the two groups (P>0.05 for all comparisons) (Table 5).

Table 5 Comparison of Adverse Reactions Incidence and Remedial Antiemetic Rate Between the Two Groups, n (%)

Sleep Quality

Sleep quality, assessed with the RCSQ and analyzed with repeated-measures analysis of variance, was higher at T₅ in Group O than in Group S (P=0.044). In within-group analyses, RCSQ scores increased from baseline to T₅ in Group O (P=0.002). No significant between-group difference was observed at T₃ (P=0.988) (Table 6).

Table 6 Comparison of RSCQ Scores Between the Two Groups, Mean (SD), Score

Anxiety and Depression

Symptoms of anxiety and depression were assessed with the HAMA and HAMD scores. In analyses using generalized estimating equations and the Friedman test, no significant between-group differences were observed in HAMA or HAMD scores at any time point (P>0.05 for all comparisons) (Table 7). In within-group comparisons, HAMA scores decreased from baseline to T₅ in Group O (P=0.003), whereas HAMD scores increased from baseline to T₃ in Group S (P=0.049).

Table 7 Comparison of Anxiety and Depression Negative Emotions Between the Two Groups, M (IQR), Score

Postoperative Recovery

No significant between-group differences were observed in time to start ambulation, time to first flatus, time to removal of the drainage tube, or length of hospital stay (P>0.05 for all comparisons) (Table 8).

Table 8 Comparison of Postoperative Recovery Between the Two Groups

Discussion

This prospective, randomized, double-blind, non-inferiority trial demonstrated that oliceridine, when administered via PCIA, provides non-inferior analgesia compared to sufentanil in elderly patients undergoing laparoscopic radical resection for gastrointestinal malignancies. Acute postoperative pain after abdominal surgery is largely attributable to abdominal-wall pain and typically persists for 1 to 2 days,¹² therefore, we selected the 48-hour cumulative AUC of resting VAS pain scores as the primary efficacy endpoint. The primary outcome of 48-hour cumulative AUC for resting pain met the prespecified non-inferiority margin in both the PP and mITT analyses, with the two analysis populations being identical (n=88, as the excluded patients never received the study drug), confirming the robustness of this finding. Beyond comparable efficacy, oliceridine was associated with significantly lower incidences of respiratory depression and hypotension, two clinically critical adverse events in this vulnerable population. Importantly, intraoperative opioid exposure was balanced between groups, further supporting that the observed differences reflect the PCIA regimens rather than confounding from intraoperative analgesia.

This trial was designed using an equianalgesic-dose approach, guided by the prescribing information for oliceridine and preliminary experimental findings. The planned doses were 0.4 mg/kg for oliceridine and 2 μg/kg for sufentanil (with sufentanil estimated to be approximately 200 times as potent as oliceridine). To balance efficacy and safety, the PCIA regimen was standardized with a 5-mL loading dose, 2.5-mL patient-controlled boluses, a continuous infusion rate of 2.5 mL/h, and a lockout interval of 15 minutes. The effective and total PCIA demand counts were lower in Group O than in Group S, suggesting that patients required fewer supplemental doses to achieve comparable pain control. Pain scores at rest increased in both groups at T₂ as compared with T₀, indicating a pain peak in the early postoperative period. However, no significant between-group differences were observed in VAS pain scores at rest or during movement, cumulative AUC values over 48 hours, analgesia-satisfaction scores at T₃ and T₅, total analgesic consumption, or the proportion of patients receiving rescue analgesia. These findings indicate that oliceridine provided analgesia comparable to sufentanil when administered by PCIA in elderly patients undergoing laparoscopic resection of gastrointestinal malignant tumors.

The observed analgesic profile may be related to the biased agonism of oliceridine. Opioid analgesics exert their effects through opioid receptors in the central nervous system and peripheral tissues, including MOP, δ-opioid receptors (DOP), κ-opioid receptors (KOP), and nociceptin/orphanin FQ receptors.¹⁸ The MOP receptor is a G protein–coupled receptor that signals through both Gi-protein–mediated and β-arrestin–mediated pathways. Activation of Gi protein inhibits adenylate cyclase, promotes opening of receptor-gated potassium channels, and inhibits voltage-gated calcium channels, thereby reducing presynaptic neurotransmitter release and hyperpolarizing postsynaptic membranes, with consequent attenuation of nociceptive transmission. In contrast, engagement of the β-arrestin pathway leads to recruitment of β-arrestin proteins and may dampen analgesic signaling.¹⁰ Oliceridine preferentially activates Gi-protein signaling and has been reported to produce substantially less β-arrestin recruitment than morphine (approximately 14% of the morphine effect), a signaling bias that may help preserve analgesia while potentially improving tolerability.

Intraoperative tissue injury, ischemia–reperfusion, and acute postoperative pain can sustain a state of oxidative stress, characterized by excess generation of reactive oxygen species and disruption of the balance between oxidant and antioxidant defenses. These processes may activate complement and cytokine pathways and contribute to a heightened systemic stress response.¹⁹ SP, a key mediator of nociceptive transmission, can activate nociceptors and amplify pain perception.²⁰ hs-CRP is an acute-phase reactant that increases in response to tissue injury and promotes inflammatory cascades.²¹ Cor is central to maintenance of physiologic homeostasis during stress and is widely used as an index of stress responsivity.²² In our study, several serum pain-stress indicators increased during the first 48 hours after radical surgery for gastrointestinal malignant tumors and peaked at 6 hours and 24 hours after surgery, a pattern consistent with the time course of VAS pain scores. As compared with Group S, Group O had lower SP, Cor, and hs-CRP levels at 6 hours and lower Cor and hs-CRP levels at 24 hours after surgery. Notably, cortisol and hs‑CRP are not merely stress markers—they are biologically active molecules with direct roles in metabolism and immunity. Cortisol regulates glucose homeostasis, and its persistent elevation can induce insulin resistance.²³ hs‑CRP and TNF‑α are integral to the inflammatory cascade and reflect immune activation.^24,25 Thus, the lower levels of these markers in the oliceridine group raise the possibility of broader benefits, including improved glycemic control and preserved immune function. This aligns with evidence that μ‑opioid receptors are expressed on immune cells²⁶ and that biased agonism may attenuate pro‑inflammatory signaling.²⁷ However, given the absence of direct correlation analyses, these mechanistic interpretations should be viewed as hypothesis-generating and require confirmation in future studies.

The most clinically significant findings of this trial relate to safety. We observed lower incidences of postoperative respiratory depression and hypotension in Group O than in Group S, whereas the incidences of postoperative nausea and vomiting, pruritus, dizziness, headache, abdominal distention, somnolence, and the use of rescue antiemetics were similar in the two groups. These findings may be related to the pharmacologic profile of oliceridine as a biased MOP agonist, since β-arrestin recruitment has been implicated in opioid-related adverse events in addition to modulating analgesic signaling. Our results align with previous studies: a Phase II trial reported a lower incidence of postoperative respiratory depression with oliceridine than morphine among patients undergoing abdominoplasty,²⁸ and a recent trial comparing oliceridine (11.8%) with sufentanil (18.2%) for gastrointestinal endoscopy reported a lower incidence of hypotension during the procedure with oliceridine.²⁹ In summary, oliceridine appears to offer a safety advantage over sufentanil in this vulnerable population, particularly with respect to respiratory and hemodynamic outcomes. However, given that this trial was designed and powered primarily to assess efficacy, these safety findings—while clinically meaningful—should be considered exploratory and require confirmation in larger studies specifically powered for safety endpoints.

We did not observe a significant between-group difference in the incidence of PONV, a finding that contrasts with reports in which oliceridine was associated with a lower incidence of PONV than conventional opioid analgesics.³⁰ One potential explanation is that laparoscopic surgery is an established risk factor for PONV, and the operative context in our study may have attenuated any drug-related differences. During laparoscopic resection for gastrointestinal malignant tumors, instrumentation may injure the gastric mucosa or stimulate vagal afferents. In addition, the high-pressure pneumoperitoneum can increase intraabdominal and intracranial pressures, activate central emetic pathways, and contribute to hypercapnia, which may further stimulate chemoreceptor trigger zones and increase the likelihood of PONV.³⁰ Given the modest sample size, larger studies will be needed to better define the effect of oliceridine on PONV in elderly patients undergoing laparoscopic radical surgery for gastrointestinal malignant tumors.

Postoperative sleep disturbance, often manifested by fragmented sleep and reduced total sleep time, is common after surgery. Pain is an important contributor to postoperative sleep disruption, and sleep disturbance can, in turn, exacerbate pain, creating a reciprocal relationship.³¹ In our study, RCSQ scores at 48 hours after surgery were higher in Group O than in Group S and were improved relative to preoperative values, suggesting better early postoperative sleep quality with oliceridine. This finding may reflect effective analgesia and a favorable feedback loop between pain control and sleep restoration.

Symptoms of anxiety and depression are also frequent in older surgical patients and have been associated with postoperative pain and functional decline.³² Regarding negative affect, although we found no significant between-group differences in HAMA or HAMD scores at prespecified time points, within-group comparisons suggested divergent trajectories: HAMA scores decreased from baseline to 48 hours in Group O, whereas HAMD scores increased from baseline to 24 hours in Group S. These observations raise the possibility that oliceridine may be associated with less postoperative anxiety and less worsening of depressive symptoms than sufentanil, potentially mediated through improved pain control. However, comparative evidence regarding the effect of oliceridine on negative affect is limited, and these findings should be interpreted cautiously pending confirmation.

This study has limitations. First, it was conducted at a single center and involved a relatively small sample, which may limit the generalizability of the findings. Second, two randomized patients did not receive the study drug due to intraoperative complications and were excluded from all drug‑related analyses. Although this exclusion is methodologically justified, it necessarily reduces the intention‑to‑treat population for efficacy and safety assessments. Third, the study was powered for the primary efficacy endpoint, not for safety outcomes; therefore, the differences in adverse events, while clinically meaningful, should be interpreted with appropriate caution. Fourth, to minimize confounding, we enrolled only elderly patients undergoing laparoscopic radical surgery for gastrointestinal malignancies; the generalizability of these findings to younger patients or to other surgical procedures is uncertain. Fifth, the mechanistic interpretations based on serum biomarkers remain speculative, as direct correlation analyses were not performed; these findings require confirmation in studies designed to investigate mechanisms directly. Therefore, multicenter studies with larger sample sizes, specifically designed to assess safety outcomes and investigate underlying mechanisms, are warranted to validate and extend our findings.

In summary, among elderly patients undergoing laparoscopic radical resection for gastrointestinal tumors, PCIA with oliceridine provided analgesia that was comparable to that with sufentanil. Notably, oliceridine was associated with significantly lower incidences of respiratory depression and hypotension, two critical safety advantages in this vulnerable population. Exploratory findings also suggested potential benefits in attenuating postoperative stress responses and improving sleep quality, with the observed reductions in cortisol and inflammatory markers raising the possibility of broader effects on metabolism and immune function. These findings support oliceridine as an alternative opioid option for PCIA when respiratory and hemodynamic safety are paramount. Larger multicenter studies are warranted to confirm these results and further elucidate the full spectrum of oliceridine’s effects on postoperative recovery.