Comparison of Ciprofol and Propofol on Postoperative Quality of Recovery and Intraoperative Cerebral Oxygen Saturation in Patients Undergoing Thoracoscopic Thoracic Surgery: A Randomized, Double-Blind, Non-Inferiority Trial

Abstract

Purpose

Ciprofol provides more stable induction, though its impact on patients’ quality of recovery (QoR) and brain oxygenation requires clarification. This trial assessed whether propofol anesthesia is non-inferior to ciprofol regarding postoperative recovery quality, while additionally comparing their effects on intraoperative cerebral oxygenation.

Patients and Methods

This single-center, randomized, double-blind, non-inferiority study involved 94 patients scheduled for elective surgery, randomized equally to receive ciprofol (Group C) or propofol (Group P). The primary outcome was the global QoR-15 score measured 24 hours after surgery, with a non-inferiority threshold set at −6.0. Secondary outcomes included 48-hour QoR-15 scores, injection pain frequency, vasopressor requirements, and regional cerebral oxygen saturation (rScO₂).

Results

Eighty- eight patients completed the study (Group C: 44; Group P: 44). In ITT analysis, the mean QoR-15 score at 24 hours for Group C was non-inferior to Group P (P_{non-inferiority} < 0.001), with consistent findings in the PP analysis. Similar non-inferiority was observed at 48 hours. However, Group C experienced a notably lower rate of injection pain (4.5% vs 59.0%, P < 0.001) and required less phenylephrine (P = 0.009). Although Group P had higher heart rates after induction of anesthesia (P < 0.05), its average arterial pressure was notably reduced (P < 0.05). Notably, rScO₂ in group C was higher than that in group P after induction of anesthesia (P = 0.046) and during the anesthesia maintenance phase (P < 0.05).

Conclusion

Ciprofol is non-inferior to propofol regarding early postoperative recovery quality. Furthermore, ciprofol mitigates injection pain, reduces vasoconstrictor requirements, and better maintains cerebral oxygenation.

Introduction

Maintaining cerebral oxygen saturation during thoracic surgery is clinically significant, as perioperative cerebral hypoperfusion and tissue hypoxia are closely associated with poor neurological outcomes and prolonged hospital stays. The one-lung ventilation (OLV) technique achieved through double-lumen bronchial intubation in lung surgeries is widely adopted. Substantial evidence suggests that the incidence of reductions in rScO₂ during OLV ranges from 25% to 82%.1–6 Despite the adoption of lung-protective ventilation strategies during anesthesia, hypoxemia may still occur due to a variety of factors, rendering elderly patients especially susceptible to postoperative neurological complications.4 Traditional monitoring techniques are limited in their ability to assess the functional state of the brain. To address this clinical vulnerability, rScO₂ monitoring via near-infrared spectroscopy (NIRS) has emerged as a vital tool. Cerebral perfusion can be continuously monitored by this technology and cerebral ischemia can be detected early. Therefore, the technology has been widely applied in clinical practice.7–9 Notably, rScO₂ monitoring is characterized by its high efficiency, accuracy, non-invasiveness, and low sensitivity to temperature fluctuations. It is capable of early identification of cerebral hypoxia and imbalances between cerebral oxygen supply and demand, thereby helping to prevent perioperative neurocognitive disorders.10 Intraoperatively, cerebral oxygen monitoring can effectively reflect alterations in tissue oxygenation under conditions such as hypotension, hypovolemia, embolism, and shock.11 Consequently, integrating cerebral oxygen monitoring into thoracic surgery serves as a critical strategy to optimize perioperative anesthetic management, mitigate the risk of neurological complications, and ultimately enhance the overall quality of recovery.

Ciprofol is a new short-acting intravenous anesthetic, a γ-aminobutyric acid (GABA) receptor agonist. Ciprofol originates from propofol by adding the cyclopropyl group to the isopropyl chain, thus improving its affinity with GABA type A (GABA_A) receptors.12 The affinity of ciprofol with GABA_A receptors is about four to five times as that of propofol, and this specific characteristic of ciprofol enables it to reach similar anesthetic effect in a small dose.13 Clinical evidence suggests that ciprofol improves hemodynamic stability, with a lower incidence of hypotension during anesthesia, while maintaining comparable anesthetic efficacy.14,15 The favorable safety profile of ciprofol, including reduced respiratory depression, enhanced hemodynamic stability, absence of injection pain, and lower lipid load, may contribute to the improvement of postoperative recovery.16 Furthermore, rScO₂ reflects the balance between cerebral oxygen supply and consumption and is influenced by cerebral perfusion, arterial blood pressure, cardiac output, arterial oxygen content, carbon dioxide partial pressure, and depth of anesthesia.17 Since anesthetic-induced circulatory suppression may impair cerebral perfusion, drugs offering superior hemodynamic stability may better maintain cerebral oxygenation during surgery. Previous reports have indicated that ciprofol causes less circulatory suppression than propofol. Thus, we hypothesize that ciprofol may be associated with greater stability in intraoperative rScO₂. Based on the above, the present study aims to evaluate whether ciprofol is non-inferior to propofol with respect to early postoperative recovery quality. Additionally, we compare the effects of ciprofol and propofol on intraoperative rScO₂ and explore their implications for optimizing perioperative management.

Materials and Methods

Study Design

This is a single-center, randomized, double-blind clinical trial. Propofol was used as the control drug. The study was conducted at the Sixth Affiliated Hospital of South China University of Technology from January 2025 to January 2026. Before the patient screening began, the study had been approved by the Medical Ethics Committee of the Sixth Affiliated Hospital of South China University of Technology (NYKY-2024-157-01). The trial was officially registered on December 26, 2024. The official records can be accessed publicly through the Chinese Clinical Trial Registry (number: ChiCTR2400094687). All participants, or their legally appointed representatives, provided written informed consent prior to enrollment.

Study Participants

Overall, 94 patients from January 2025 to January 2026 undergoing thoracoscopic surgery were recruited, with the following inclusion criteria: 1) The subjects have a full understanding of the purpose and significance of this trial, voluntarily participate in this clinical trial, and sign the informed consent; 2) Patients undergoing thoracoscopic thoracic surgery under general anesthesia; 3) Age 18–65 years; 4) 18 kg/m²≤BMI≤30 kg/m²; and 5) American Society of Anesthesiologists (ASA) Grade I–III. The exclusion criteria were 1) People who are allergic to propofol, ciprofol and other ingredients or components that may be used in the study; 2) Patients with long-term use of narcotic analgesics, sedatives, or NSAIDs drugs; 3) Patients with systemic hypotension despite the use of vasoactive drugs; 4) Heart rate <55 beats/min; 5) High atrioventricular block without pacemaker; 6) Complicated with craniocerebral injury and intracranial hypertension; 7) Patients with severe heart, lung, liver and renal dysfunction; 8) Patients with a history of allergic disease, malignant hyperthermia in the past, and epilepsy in the past; 9) Subjects with a history of mental illness and cognitive dysfunction; 10) Fat metabolism disorders; 11) Predictable difficulty with intubation; 12) Patients with myasthenia gravis; 13) Subjects with sensory disorders such as hyperalgesia; 14) History of alcoholism; 15) Have severe visual or hearing impairment; 16) Unable to cooperate to complete the test after surgery; 17) Other circumstances in which the investigator determines that the subject is not suitable to participate in the clinical trial; 18) Pregnant or lactating women; and 19) HIV or RPR positive.

Interventions

Patients were randomly assigned to one of two groups: the ciprofol group (Group C) or the propofol group (Group P). In Group C, ciprofol was the sedative for anesthesia induction and maintenance; in Group P, it served as the standard control drug for both induction and maintenance. Apart from this, all perioperative management strategies were identical between the two groups. After randomization, the specific anesthetic protocol was implemented in the following sequence: Patients were restricted from drinking (including water) within 1 h before surgery and needed to fast for 6–8 h prior to surgery. After entering the operating room, patients were connected to multi-parameter anesthesia monitoring equipment. Peripheral venous and arterial access was then established. Continuous monitoring was performed for electrocardiogram (ECG), respiratory rate (RR), heart rate (HR), non-invasive and invasive blood pressure (BP), pulse oxygen saturation (SpO₂), bispectral index (BIS), and rScO₂. For anesthesia induction, patients in Group C (n = 47) received 0.4 μg/kg sufentanil, 0.4 mg/kg ciprofol (Haisco Pharmaceutical Group), and 2 mg/kg cisatracurium besilate; patients in Group P (n = 47) received 0.4 μg/kg sufentanil, 2 mg/kg propofol (Sichuan Guorui Pharmaceutical), and 2 mg/kg cisatracurium besilate. Two minutes after drug administration, airway access was established via endotracheal intubation. Mechanical ventilation followed a standardized lung-protective strategy: tidal volume was maintained at 8 mL/kg of ideal body weight during double-lung ventilation, and adjusted to 5–6 mL/kg during OLV. Respiratory rate was adjusted between 12 and 15 breaths per minute to strictly maintain end-tidal carbon dioxide (EtCO₂) and target partial pressure of arterial carbon dioxide (PaCO₂) between 35 and 45 mmHg. All patients received a standard thoracic paravertebral nerve block with 30 mL of 0.33% ropivacaine before the start of surgery. During anesthesia maintenance, the sedation dose was strictly adjusted based on anesthesia depth. Ciprofol was administered at an initial dose of 0.8 mg/kg/h, and propofol at an initial dose of 5 mg/kg/h, to maintain the BIS value between 40 and 60. Intraoperative neuromuscular blockade was standardized by administering intermittent boluses of cisatracurium based on train-of-four (TOF) monitoring to maintain a count of 1–2. Hemodynamic and fluid management followed a goal-directed strategy. Intraoperative hypotension, defined as a >20% decrease in mean arterial pressure (MAP) from baseline or MAP < 65 mmHg, was immediately managed with phenylephrine, combined with a goal-directed crystalloid/colloid regimen to maintain stable cardiac output and cerebral perfusion. Hypoxemia (SpO₂ < 90%) was managed with temporary lung recruitment maneuvers and an increase in fraction of inspired oxygen (FiO₂). A standard antiemetic regimen (intravenous injection of 4 mg ondansetron) was administered 30 minutes before the end of surgery. Delirium prevention measures were uniformly implemented throughout the procedure, including orientation disclosure at multiple time points, use of a multimodal analgesia regimen, avoidance of drugs with high delirium risk, and correction of electrolyte and fluid imbalances.

Randomization and Blinding

Random numbers were generated by the computer, and these numbers were used to evenly distribute 94 participants into two different treatment groups. Group C received the ciprofol treatment regimen, while Group P received the propofol control treatment. To ensure the strict confidentiality of the allocation, the grouping information for these groups was sealed in opaque, sequentially numbered envelopes, and these seals were not opened until before the induction of anesthesia. To minimize errors, we strictly divided the clinical responsibilities of each researcher. An independent anesthesia nurse, who was entirely insulated from subsequent clinical management, data collection, and statistical analysis, opened the envelope and prepared the study drugs. The study syringes were prepared outside the actual operating room in a dedicated central anesthetic preparation unit. Because both ciprofol and propofol are identical white lipid emulsions, they were drawn into identical, unlabeled 20-mL syringes to ensure visual indistinguishability. The prepared syringes were then delivered to the operating room. One anesthesiologist was responsible for administering all the drugs during the surgery, and another researcher was responsible for the assessment after each anesthesia and the subsequent follow-up interviews. The strict blinding procedure ensured the integrity of our final data. Therefore, patients, surgeons, and data collectors were completely unaware of the specific drug allocation throughout.

Data Collection and Monitoring

Patients were preliminarily screened the day before surgery, underwent relevant clinical examinations during the screening period, and signed informed consent forms. Baseline demographic data including gender, age, BMI, medical history (coronary heart disease, hypertension, diabetes, central nervous system disorders, and psychiatric or psychological conditions), alcohol abuse history, and drug dependence history were systematically collected. After entering the operating room, continuous physiological monitoring was performed. MAP, HR, and SpO₂ were recorded at specific time points: upon arrival in the operating room (T0), before anesthesia induction (T1), after anesthesia induction (T2), before endotracheal intubation (T3), one minute after intubation (T4), five minutes after intubation (T5), and at the start of surgery (T6). rScO₂ was measured not only at the same T0–T6 time points but also one hour after surgery began (T7) and at the end of surgery (T8). Total intraoperative phenylephrine consumption was recorded and archived at the conclusion of surgery. During postoperative follow-up, patients were assessed using the QoR-15 at 24 hours and 48 hours after surgery, and their overall recovery time was accurately documented on the day of discharge. Record any postoperative complications that occur from the end of surgery until discharge, including postoperative hypotension, postoperative hypoxemia, and postoperative nausea and vomiting.

Outcomes

Primary outcome was the QoR-15 score at 24 h after surgery. Secondary outcomes were rScO₂ measurements obtained at the following times: on T0; T1; T2; T3; T4; T5; T6; T7; T8. Furthermore, from time point T0 to T6, the researchers also recorded MAP, HR, and SpO₂. Additionally, the total dosage of vasoactive drugs used during the surgery was recorded at the end of the surgery. The QoR-15 score was evaluated 48 hours after surgery, and the patients’ length of stay, defined as the period from the surgery date to discharge, was recorded at discharge. Adverse events such as bradycardia, arrhythmia, vomiting, nausea, other heart problems, injection site pain, and postoperative hypotension were also recorded.

Sample Size

This clinical study was designed as a non-inferiority trial. The primary outcome was the QoR-15 score evaluated within 24 hours after surgery. The noninferiority margin was set at 6 points on the QoR-15 scale based on the latest research guidelines, with a minimum clinically important difference (MCID).18 A 6-point change on the QoR-15 scale represents the smallest meaningful difference patients perceive in their recovery experience. Patients and anesthesiologists might accept a potential decline in recovery quality equivalent to the MCID in exchange for hemodynamic stability and potentially reduced cerebral oxygenation impact associated with propofol. Furthermore, this margin has been adopted in previous non-inferiority trials comparing anesthesia techniques in similar surgical populations.19 Assuming a standard deviation of 10, To control for type I error (false positive error), we set the significance level at 0.025 (one-sided) and aimed for a statistical power of 80%, resulting in an initial calculation requiring 42 subjects per group. Additionally, we included a 10% buffer for dropouts to account for sample loss due to withdrawal or withdrawal of informed consent. Ultimately, 47 patients were recruited per group, yielding a total sample size of 94. The team then randomly assigned the entire cohort to the corresponding groups for propofol and ciprofol treatment.

Statistical Analysis

SPSS 25.0 was used for the statistical analysis of the experimental data of all individuals. For continuous variables, the Kolmogorov–Smirnov test was carried out first to test for their distribution, and whether such data follow a normal curve distribution or not; indicators conforming to normal distribution were presented by the mean ± SD, and compared by an independent sample t-test between these values. Skewed data were processed non-parametrically and presented by median (interquartile range [IQR]), and compared by the Mann–Whitney U-test. Categorical variables were given as frequency (%); the chi-square test or Fisher test was used for categorical comparisons in accordance with specific situations. The main endpoint, the 24-h global QoR-15 score, was analysed for noninferiority. The primary non-inferiority analysis was performed using both the Intention-to-Treat (ITT) and Per-Protocol (PP) frameworks to rigorously account for missing data and dropouts. Both ITT and PP analyses assessed whether the lower limit of the 95% confidence interval of the mean difference between Group C and Group P was still above −6.0 (the pre-specified non-inferiority boundary). MAP, HR, and rScO₂ were followed dynamically at seven time points (T0 – T6) in linear mixed models (LMM) in analyses according to CONSORT guidelines for non-inferiority trials to confirm robustness of the main conclusion. The LMM included time and the group × time interaction as fixed effects and subjects as random effects to account for the correlation of repeated measurements. Bonferroni correction was used for post hoc pairwise comparisons between groups at each time point. Non-inferiority tests used a one-sided P < 0.025, while two-sided P values < 0.05 were considered significant.

Results

Between Jan. 2025 and Jan. 2026, we screened 154 patients for trial eligibility. Among these, 60 patients were excluded (29 patients did not meet the inclusion criteria, 9 met exclusion criteria, 22 patients did not want to participate). 94 patients were randomized into the two groups: ciprofol group (Group C, n = 47) and propofol group (Group P, n = 47). In the Group C, 2 patients were lost to follow-up and 1 was lost to drop out, thus, 44 patients were analysed in final analysis. In the Group P, 3 patients were lost to follow-up and 44 patients were analysed. All randomized patients (n = 94) were included in ITT analysis and 88 patients who completed study protocols were included in PP analysis. This trial complies with the applicable Consolidated Standards of Reporting Trials (CONSORT) guidelines (Figure 1).

Figure 1.

CONSORT flow diagram of patient recruitment and retention. A total of 154 patients were assessed for eligibility, of whom 94 were randomized to receive either ciprofol (n = 47) or propofol (n = 47). Reasons for exclusion and the flow of participants through each stage of the trial, including the numbers analyzed for both Intention-to-Treat (ITT) and Per-Protocol (PP) populations, are detailed.

General Characteristics of the Two Groups

No statistically significant differences were observed in terms of gender, age, body mass index, prevalence of underlying diseases (covering diabetes, coronary heart disease, and hypertension), or ASA classification (P > 0.05 for all, Table 1).

Table 1.

Comparison of General Information Between the Two Groups

Parameters		P Group	C Group	P
Age (years)		55.61±7.77	52.39±8.75	0.25
Gender	Male	17 (38.6%)	15 (34.1%)	0.20
	Female	27 (61.4%)	29 (65.9%)
BMI (kg/m2)		22.96±3.08	22.58±3.08	0.84
Hypertension		10 (22.7%)	6 (13.6%)	0.41
Diabetes		2 (4.5%)	8 (18.2%)	0.08
Coronary heart disease		2 (4.5%)	3 (6.8%)	>0.99
ASA I		4 (9.1%)	3 (6.8%)	0.37
ASA II		39 (88.6%)	37 (84.1%)
ASA III		1 (2.3%)	4 (9.1%)

Note: Data were presented as the deviation or n (%).

Intraoperative and Postoperative Data of the Two Groups

Regarding the operation time, anesthesia time, intraoperative blood loss, and length of stay, there was no statistically significant difference between the C group and the P group (P > 0.05), as shown in Table 2. Regarding adverse reactions in the two groups of patients, injection pain was more severe in group P compared to group C (59.1% vs 4.5%, P < 0.001). However, there were no statistically significant differences between the two groups regarding postoperative hypotension, postoperative hypoxemia, and postoperative nausea and vomiting (P > 0.05).

Table 2.

Intraoperative and Postoperative Data of the Two Groups

Parameters	P Group	C Group	P
Operation time (min)	131.84±60.29	126.5±57.12	0.45
Anesthesia time (min)	189.39±57.78	188.57±56.537	0.64
Intraoperative blood loss (mL)	20 (10, 50)a	20 (10, 50)a	0.83
Length of stay (d)	5.36±2.01	5.36±1.95	0.63
Phenylephrine consumption (μg)	1150 (600, 1500)a	500 (0, 1273)a	0.009b
Injection pain	26 (59.1%)	2 (4.5%)	<0.001b
Postoperative hypotension	5 (11.4%)	3 (6.8%)	0.46
Postoperative hypoxemia	1 (2.3%)	1 (2.3%)	>0.999
Postoperative nausea and vomiting	3 (6.8%)	2 (4.5%)	0.65

Notes: Data were presented as the deviation or n (%); ^aData are presented as median (Q1, Q3) due to non-normal distribution; ^bP<0.05.

Primary Outcome

As shown in the ITT analysis (Figure 2) and PP analysis (Figure 3) in Table 3, the QoR-15 score at 24 hours after surgery, regardless of the ITT analysis (mean [95% CI], 95.2 [93.0~97.5] vs 94.1 [89.6~98.7]; mean difference [95% CI], 1.14 [−0.34~2.61]; P<0.001 for noninferiority) or PP analysis (mean [95% CI], 95.3 [93.0~97.6] vs 94.1 [89.6~98.7]; mean difference [95% CI], 1.15 [−0.35~2.66]; P<0.001 for noninferiority), was not inferior in the C group compared to the P group. The same result was observed for the QoR-15 score on QOR2 at 48 hours, including both the ITT analysis (mean [95% CI], 106.4 [102.0–110.8] vs 105.1 [102.5–107.6]; mean difference [95% CI], 1.34 [−0.13~2.79]; P<0.001 for noninferiority) and PP analysis (mean [95% CI], 106.3 [102.2–110.4] vs 105.3 [103.0–107.6]; mean difference [95% CI], 1.04 [−0.36~2.44]; P<0.001 for noninferiority). Additionally, the linear mixed model analysis of the QoR-15 scores between the two groups showed no significant difference (P=0.14). Moreover, there was no statistically significant difference in the changes that occurred between the two groups over time (P=0.85).

Figure 2.

Non-inferiority analysis of Quality of Recovery-15 (QoR-15) scores in the Intention-to-treat (ITT) analysis. Data are presented as mean differences between the Ciprofol and Propofol groups, with error bars representing the two-sided 95% confidence intervals (CIs). The endpoints include the global QoR-15 score at 24 h (QoR1) and 48 h (QoR2) postoperatively. The QoR-15 is a validated 15-item instrument measuring health-related quality of recovery (range: 0–150; higher scores indicate superior recovery). The vertical dashed line at −6 represents the pre-specified non-inferiority margin, corresponding to the minimal clinically important difference (MCID) for the QoR-15 scale. Non-inferiority is established as the lower limit of the 95% CI for the mean difference (Ciprofol minus Propofol) lies to the right of this margin. The 95% CIs for both time points did not cross the margin, confirming that Ciprofol-based anaesthesia is non-inferior to Propofol-based anaesthesia.

Figure 3.

Non-inferiority analysis of Quality of Recovery-15 (QoR-15) scores in the Per-protocol (PP) analysis. Data are presented as mean differences between the Ciprofol and Propofol groups, with error bars representing the two-sided 95% confidence intervals (CIs). The endpoints include the global QoR-15 score at 24 h (QoR1) and 48 h (QoR2) postoperatively. The QoR-15 is a validated 15-item instrument measuring health-related quality of recovery (range: 0–150; higher scores indicate superior recovery). The vertical dashed line at −6 represents the pre-specified non-inferiority margin, corresponding to the minimal clinically important difference (MCID) for the QoR-15 scale. Non-inferiority is established as the lower limit of the 95% CI for the mean difference (Ciprofol minus Propofol) lies to the right of this margin. The 95% CIs for both time points did not cross the margin, confirming that Ciprofol-based anaesthesia is non-inferior to Propofol-based anaesthesia.

Table 3.

QoR-15 Score of the Two Groups

Parameters		P Group	C Group	Mean Difference [95% CI]	P
ITT	QOR1 (24 hours after surgery)	94.1 [89.6~98.7]	95.2 [93.0~97.5]	1.14 [−0.34~2.61]	<0.001*
	QOR2 (48 hours after surgery)	105.1 [102.5–107.6]	106.4 [102.0–110.8]	1.34 [−0.13~2.79]	<0.001*
PP	QOR1 (24 hours after surgery)	94.1 [89.6~98.7]	95.3 [93.0~97.6]	1.15 [−0.35~2.66]	<0.001*
	QOR2 (48 hours after surgery)	105.3 [103.0–107.6]	106.3 [102.2–110.4]	1.04 [−0.36~2.44]	<0.001*

Notes: Data are presented as mean [95% CI]; *P<0.05.

Hemodynamic Profiles and Cerebral Oxygenation

Regarding intraoperative hemodynamic management, the cumulative dose of phenylephrine administered was significantly higher in Group P than in Group C [median (Q1, Q3), 1150μg (600, 1500) vs 500μg (0, 1273), P = 0.009], as shown in Table 2. The perioperative hemodynamic trends, including HR and MAP, are illustrated in Figures 4 and 5, respectively. Both parameters exhibited significant group-by-time interactions P_Group*Time = 0.035 for HR and P_Group*Time < 0.001 for MAP). Specifically, patients in group P had significantly higher heart rates than patients in group C at T2 and T3 (P < 0.05) and lower mean arterial pressure at the same time points (P < 0.05). Regarding cerebral monitoring, the changes in rScO2 are shown in Figure 6. Significant interaction between group and time was found P_Group*Time = 0.042. Independent samples t-tests were done to compare rScO2 values between group P and group C at each individual time point. No differences were found at baseline and early induction (T0–T1). rScO2 was significantly lower in Group P at T2 P = 0.046, T6 P = 0.022, T7 P = 0.011, and T8 P = 0.032 compared with Group C.

Figure 4.

Comparison of perioperative heart rate (HR) trends between the two groups. Data are presented as Mean ± SD. The red solid line represents Group P (n=44), and the blue dashed line represents Group C (n=44). Time points: In the operating room (T0), before induction (T1), after induction (T2), before intubation (T3), 1 minute after intubation (T4), 5 minutes after intubation (T5), and at the start of surgery (T6). Statistical analysis: Intergroup and longitudinal differences were analyzed using two-way repeated measures ANOVA. P_Group*Time denotes the interaction effect between group and time. Significant differences between groups at specific time points (T2 and T3) were identified using independent samples t-tests, with P-values indicated in the figure (P < 0.05).

Figure 5.

Changes in mean arterial pressure (MAP) throughout the perioperative period. Data are presented as Mean ± SD. Group identification: The red solid line indicates Group P (n=44), and the blue dashed line indicates Group C (n=44). Time points: In the operating room (T0), before induction (T1), after induction (T2), before intubation (T3), 1 minute after intubation (T4), 5 minutes after intubation (T5), and at the start of surgery (T6). Statistical analysis: Two-way repeated measures ANOVA was used to evaluate group and time effects. P_Group*Time < 0.001 indicates a highly significant interaction between group and time factors. Significant differences between groups at T2 and T3 were identified using independent samples t-tests, with the corresponding P-values displayed in the figure (P < 0.05).

Figure 6.

Perioperative changes in regional cerebral oxygenation saturation (rScO2) between the two groups. Data are presented as Mean ± SD. Group identification: The red solid line represents Group P (n=44), and the blue dashed line represents Group C (n=44). Time points: Upon entering the operating room (T0), before induction (T1), after induction (T2), before intubation (T3), 1 minute after intubation (T4), 5 minutes after intubation (T5), at the start of the surgery (T6), 1 hour after the start of the surgery (T7), and at the end of the surgery (T8). Statistical analysis: A two-way repeated measures ANOVA was first used to assess the interaction between group and time P_Group×Time = 0.042. Independent samples t-tests were then performed to compare rScO₂ values between the two groups at each individual time point. Statistically significant differences were observed at T2, T6, T7, and T8, with the corresponding P-values displayed above the error bars (P < 0.05).

Discussion

This study represents the first systematic comparison of the effects of ciprofol and propofol on rScO₂ and postoperative recovery in the context of thoracoscopic surgery. Both ITT and PP analyses revealed that at 24 and 48 hours postoperatively, the QoR-15 scores for the ciprofol group were non-inferior to those of the propofol group. Notably, the ciprofol group exhibited significant advantages, including higher rScO2, more stable hemodynamic parameters MAP, and a reduced incidence of injection pain and vasopressor requirements. Ciprofol reduced heart rate more effectively than propofol after induction, as shown by a greater heart rate decrease in the group C.

Modern medical models increasingly emphasise patient-centred perioperative care: perioperative management seeks not just surgical success but also the patient’s subjective experience, speed of functional recovery and long-term quality of life.20 Thoracoscopic surgery, a key technique in thoracic surgery, requires special measures in anaesthesia management (eg, during single-lung ventilation) in order to balance oxygen supply and demand in the brain. Ciprofol, a short-acting intravenous sedative based on modification of the propofol structure, is a new type of sedative drug and has been increasingly applied clinically.21,22 It may also be helpful for perioperative management in thoracoscopic surgery. In a meta-analysis conducted by Pan et al involving 18 randomized controlled trials, the oxygen index corresponding to total intravenous anesthesia (TIVA) using propofol was significantly higher than that observed with inhalation anesthetics.22 Moreover, TIVA has always been superior to contemporary inhalation anesthetics in terms of recovery quality. In this context, we conducted a non-inferiority test to compare the effects of ciprofol and propofol on patients undergoing elective thoracoscopic lung surgery. Our study particularly focused on the comparison of intraoperative rScO₂ and postoperative recovery, which is crucial for improving postoperative recovery quality, surgical satisfaction, and quality of life of patients.

The quality of postoperative recovery is a core assessment indicator in the Enhanced Recovery After Surgery (ERAS) concept.23 The main objective of our study was the 24-hour QoR-15 score after surgery. The QoR-15 scale is highly recognized for its strong reliability, validity, responsiveness, and clinical feasibility, and has been established as a standardized tool for evaluating the quality of postoperative recovery.24 The results of this study indicate that, based on the most recent annual MCID definition of the QoR-15 scale (set at −6 points), the average difference in QoR-15 scores between the C group and the P group at 24 hours and 48 hours (approximately 1.1 to 1.3 points) is far above −6 points at the lower limit of the 95% confidence interval, which is the threshold.18 The P value of the non-inferiority test is less than 0.001, strongly confirming that the C group does not perform worse than the P group in promoting early postoperative recovery. This non-inferiority result means that while maintaining the quality of postoperative recovery unchanged, the protocol adopted by the C group may bring additional benefits to patients, such as improving the intraoperative experience or reducing complications, which is reflected in the secondary results of this study. Additionally, the analysis using a linear mixed model shows that there is no significant difference in the trend of QoR-15 scores over time between the two groups (P = 0.85), indicating that the effects of the two protocols on the recovery trajectory are comparable. The difference in the 24-hour postoperative recovery process is not statistically significant, and this conclusion is supported by previous clinical trials and the known pharmacokinetic characteristics of the two drugs.25 As a novel positive allosteric agonist of GABA_A receptors, ciprofol has the properties of ultra-short effect, rapid clearance and no accumulation, and its elimination kinetics is extremely similar to that of propofol. Therefore, once drugs are completely excreted from the body in the early postoperative period, their residual central sedative effects will also subside uniformly, resulting in a patient-reported overall recovery quality at 24 hours that is consistent with that reported in recent trials.

One of our major findings is that the rScO₂ in group C could be better maintained, possibly because of a more stable curve of MAP in this group. Our results indicated that in comparison with group P, MAP in group C was more stable during anesthesia induction (T2, T3) in our study, which was statistically significant. In addition, group C required fewer doses of exogenous vasoconstrictor during surgery than group P. This advantage in physiological status might better explain the different rScO₂ in these two groups. Since direct measurements of cerebral perfusion, cerebral metabolic rate, or drug exposure were not performed, any definitive causal explanation for these rScO₂ differences remains speculative. However, several potential homeostatic and pharmacological mechanisms warrant consideration. Anesthetics, surgical stress, and comorbidities may disturb cerebral homeostasis (mean arterial pressure 60–150 mmHg typically) and thus impair cerebral blood flow (CBF): this is in contrast to other organs, where a homeostatic mechanism ensures regulation of cerebral circulation.26 In Group P, the mean arterial pressure reduction at time points T2 and T3 probably placed patients at or even below their lower threshold of cerebral homeostatic reserve. Therefore, the reduced cerebral perfusion pressure (CPP), or mean arterial pressure minus intracranial pressure (ICP), directly led to a reduction of CBF, as reflected in the significantly lower rScO2 values recorded at T2 and T3. Furthermore, the quality of blood pressure management and the systemic impact of vasoconstrictors must be considered. In Group P, the use of a higher dose of phenylephrine for counteracting hypotension has a contaminating effect. Phenylephrine was able to normalize systemic MAP by enhancing the systemic vascular resistance, but it may not always increase cardiac output (CO).27,28 Instead, it may lower CO by increasing the left ventricle afterload and inducing reflex bradycardia. Since cerebral oxygen delivery (DO2) is a function of the product of arterial oxygen content and cerebral blood flow19,29 and cerebral blood flow is particularly sensitive to changes in CO,30,31 the mean rScO2 shown in group P indicates that MAP maintained by vasopressor agents is not sufficient to counterbalance a possible global reduction in CO. Conversely, group C achieved higher “autonomous” MAP with very little vasopressor aid. This could reflect that the anaesthetic protocol of group C is associated with less sympathetic inhibition or less direct inhibitory effect on the myocardium and, thus, preserves the intrinsic relation between whole-body perfusion and cerebral oxygen delivery.

Previous studies have shown that ciprofol due to its enhanced affinity for GABA_A receptors and reduced clinical dosage can effectively alleviate the dose-dependent non-target effects typically associated with propofol. This is demonstrated by the lower incidence of hypotension during the induction period and the significant reduction in intraoperative vasopressor drug requirements when using ciprofol.32 These results indicate that ciprofol has a relatively mild direct negative inotropic effect on the myocardium, while maintaining sympathetic autonomic reflexes more effectively at the same sedative depth. In high-risk surgical stages with significant fluctuations in hemodynamics, such as during the initiation of pneumoperitoneum and extreme body position changes, these pharmacological properties provide patients with a wider hemodynamic safety margin.33 Interrelation between brain oxygen saturation and POCD-related outcomes indicates that during the intraoperative maintenance phase (T6 - T8) it is worthwhile maintaining rScO2 levels in group C patient. Optimizing cerebral oxygenation during this critical surgical period may help mitigate the risk of mild postoperative neurocognitive fluctuations (as shown in 48-hour QoR-15 scores).34 The latest randomized controlled trials and meta-analyses show that declines in intraoperative brain oxygen saturation are closely related to the occurrence of POCD and postoperative hospitalization, with higher degrees and longer durations of regional rScO2 reduction increasing risk for early postoperative cognitive decline.35–37 However, inconsistent results with regard to the effect of interventions directed at brain oxygen saturation reductions on POCD incidence have also been reported.34,38 Thus, POCD-reducing treatments might do better than before with some subsets of patients in terms of early postoperative cognitive decline prevention. Anesthetic protocols maintaining myocardial contractility and intrinsic sympathetic nervous system activity may afford better neurocognitive protection than those involving high-dose vasoconstrictors for maintaining systemic blood pressure,30 because a decrease in brain oxygen saturation might reflect potential hemodynamic vulnerabilities such as heart-brain coupling disorders or insufficient CO.

Studies have shown that about 70% of patients are likely to develop injection pain with propofol (without preventive measures),39 which might cause hemodynamic instability and threaten patient safety during the induction stage.40,41 Pain at the time of injection, therefore, is of clinical importance. Our results show that incidence of injection pain in group C (4.5%) was significantly lower than that in group P (59.0%, P < 0.001). Local harmful stimulation might cause temporary discomfort and pre-induction anxiety and even the sympathetic adrenal response in some patients and thus poses a risk to patient safety in the induction stage. The injection pain in group C was significantly reduced after using ciprofol, attributed to its unique physical and chemical properties and higher efficacy. Compared to propofol, the clinical dose of ciprofol required was significantly reduced, and its optimized formulation significantly reduced the free active drug concentration in the aqueous phase, effectively alleviating the direct stimulation of the venous endothelium and the activation of the kinin release enzyme - bradykinin system. These results are consistent with previous multicenter studies and further demonstrate the clinical advantages of ciprofol in improving the quality of anesthesia induction.42,43

While ciprofol exhibits potential advantages in certain areas, it is important to recognize several limitations of our study. Firstly, the limited availability of prior research necessitated that our study population primarily consisted of patients with relatively lower systemic risks. Although group C demonstrated robust non-inferiority in this population, the physiological disturbances observed in group P particularly the transient reduction in rScO2 and the increased demand for vasopressor drugs may have more profound clinical significance in the “vulnerable” population. In elderly patients or those with compromised cerebral autoregulation, even a brief decrease in blood oxygen saturation could potentially surpass critical thresholds and have clinical implications. This warrants further investigation in high-risk subgroups.

Secondly, given the challenges associated with the predefined primary endpoint and the multiplicity of statistical comparisons, the observed inter-group hemodynamic differences should be interpreted as exploratory findings. Future research should adopt a more rigorous design, with a particular emphasis on prioritizing hemodynamic parameters as the primary endpoint.

Third, continuous CO monitoring was not performed in this study. In our analysis, we attributed the lower rScO2 values observed in group P to a reduction in CO, potentially induced by the higher dose of phenylephrine administered. Considering that cerebral DO2 is determined by the product of arterial oxygen content and cerebral blood flow, both of which are highly sensitive to changes in CO, the lack of direct CO measurements limits our ability to definitively elucidate the hemodynamic mechanism responsible for the observed decrease in cerebral oxygen saturation.

Fourth, this study concentrates on the acute phase of postoperative recovery, specifically the first 48 hours. Although the normal interval for the QoR-15 assessment is 48 hours, future large-scale longitudinal studies may be able to examine the recovery conditions under longer intervals to explore whether the early physiological benefits brought by ciprofol anesthesia can be transformed into long-term rehabilitation effects for patients, or whether it can reduce the consumption of medical resources.

Conclusion

Our study demonstrated that ciprofol was non-inferior to propofol regarding the primary endpoint of 24 h QoR-15 scores, with comparable recovery profiles maintained at 48 h. Regarding secondary physiological indexes, ciprofol was associated with a significantly lower incidence of injection pain. Concurrently, distinct profiles in intraoperative rScO₂ and vasoconstrictor reliance were observed between the two groups, with Group P exhibiting lower rScO₂ values and a higher requirement for intraoperative phenylephrine. These results indicated that although the recovery of patients in terms of outcome was basically equivalent, ciprofol was more stable and perhaps more patient-friendly in the perioperative period, especially in alleviating the discomfort of anesthesia induction and maintaining brain tissue oxygen.

Funding Statement

This study was funded by Clinical and Basic Research Special Project of the Medical and Health Research Development Fund (S212).

Abbreviations

QoR, quality of recovery; rScO₂, Regional cerebral oxygen saturation; NIRS, near-infrared spectroscopy; OLV, one-lung ventilation; GABA, γ-aminobutyric acid; BMI, Body Mass Index; ASA, American Society of Anesthesiologists; HIV, Human Immunodeficiency Virus; RPR, Rapid Plasma Reagin; ECG, electrocardiogram; RR, respiratory rate; HR, heart rate; BP, blood pressure; SpO₂, pulse oxygen saturation; BIS, bispectral index; EtCO₂, end-tidal carbon dioxide; PaCO₂, partial pressure of arterial carbon dioxide; TOF, train-of-four; MAP, Mean Arterial Pressure; FiO₂, fraction of inspired oxygen; MCID, minimum clinically important difference; IQR, interquartile range; ITT, Intention-to-Treat; PP, Per-Protocol; LMM, linear mixed models; TIVA, total intravenous anesthesia; ERAS, Enhanced Recovery After Surgery; CBF, cerebral blood flow; CPP, cerebral perfusion pressure; ICP, intracranial pressure; CO, cardiac output; POCD, Postoperative cognitive dysfunction; DO₂ oxygen delivery.

Data Sharing Statement

This study has been registered in the Chinese Clinical Trial Registry (ChiCTR) with the registration number [ChiCTR2400094687]. The research data involved in this thesis (including the raw data and the analyzed data set) can all be obtained through the registration center platform. Readers can visit the official website of ChiCTR (http://www.chictr.org.cn), enter the registration number of this study in the search bar, and then review and apply for obtaining the relevant data.

Ethical Approval and Consent to Participate in the Study

This study was conducted in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For studies involving human participants, informed consent was obtained from all individual participants included in the study. Ethical approval was provided by the Medical Ethics Committee of the Sixth Affiliated Hospital of South China University of Technology, and the study was approved on December 26, 2024. The ethics committee approval number is NYKY-2024-157-01. Informed consent was obtained from all individual participants included in the study. Additional information can be found in the study protocol. Any potentially identifying patient details have been anonymized. All authors had access to the study data and played a role in the writing of the manuscript.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

All authors report no conflicts of interest in this work.