Ciprofol versus propofol and the risk of hemodynamic adverse events: a meta-analysis with trial sequential analysis of randomized controlled trials

Abstract

Introduction

Intraoperative hypotension increases postoperative risks. Ciprofol, a novel anesthetic, has an onset and recovery profile comparable to propofol, but its potential superiority in mitigating intraoperative hypotension is unclear. This meta-analysis compares the incidence of intraoperative hypotension between ciprofol and propofol.

Methods

A comprehensive search of PubMed, Embase, and the Cochrane Library was performed up to December 10, 2025, to identify randomized controlled trials comparing ciprofol and propofol. Data were analyzed using RevMan and Stata. Subgroup analyses were conducted according to age, anesthetic dose, and type of procedure. Trial sequential analysis (TSA) was used to assess the robustness and sufficiency of the evidence.

Results

This meta-analysis of 34 trials (5,162 patients) found that ciprofol was associated with a significantly lower risk of intraoperative hypotension compared to propofol (RR = 0.65, 95% CI 0.57–0.73). This benefit was consistent across various subgroups, including different age groups, dose ranges, and procedure types. Ciprofol also demonstrated lower incidences of respiratory depression (RR = 0.44), injection pain (RR = 0.19), and hypoxemia (RR = 0.62), but was associated with a very small, likely clinically insignificant increase in awakening time;however, these findings are based on secondary outcomes and should be interpreted as exploratory. Trial sequential analysis confirmed sufficient sample size for the intraoperative hypotension and injection pain outcomes. The certainty of the evidence, assessed using the GRADE approach, was rated as low to moderate.

Conclusions

Ciprofol was associated with a reduced incidence of intraoperative hypotension compared with propofol, with comparable anesthetic efficacy. Nevertheless, due to heterogeneity and overall low-to-moderate certainty of evidence, these findings should be interpreted with caution. Further large-scale, well-designed randomized controlled trials are warranted to validate these results and delineate the clinical benefits of ciprofol.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-026-03648-8.

Key Messages

This meta-analysis suggests that ciprofol is associated with a significantly lower risk of hypotension compared to propofol during anesthesia induction and maintenance.

Ciprofol also demonstrates a substantial benefit in reducing injection pain and shows favorable profiles for respiratory depression and hypoxemia.

The overall certainty of evidence is low to moderate, indicating a need for further high-quality trials to confirm these hemodynamic advantages.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12871-026-03648-8.

Introduction

It is well established that intraoperative hypotension increases the incidence of postoperative complications and prolongs hospital stay [1, 2]. Moreover, sustained or severe intraoperative hypotension can exacerbate injury to vital organs such as the heart, kidneys, and brain, and is associated with delayed postoperative recovery, postoperative delirium in critically ill patients, and even increased mortality [3–5].

With the continuous improvement of living standards, there has been an increasing emphasis on comfort-oriented medical care. Consequently, general anesthesia techniques are being more widely adopted, particularly in painless diagnostic and therapeutic procedures [6]. Propofol, one of the most commonly used intravenous anesthetics, is favored for its rapid onset, short half-life, high clearance rate, good tolerability, lack of accumulation, and rapid recovery [7, 8]. However, its use is limited by adverse effects such as injection pain, anaphylaxis, respiratory depression, intraoperative hypotension, a narrow therapeutic window, the absence of a specific antagonist, infusion syndrome, and dose dependency. These issues are particularly concerning in patients with circulatory dysfunction, such as the elderly or those with cardiac disease [9–12]. Such adverse effects substantially reduce patient comfort during anesthesia. Therefore, identifying sedative agents with fewer side effects has become a key focus of current anesthetic research.

Ciprofol (HSK3486) is a novel intravenous anesthetic agent structurally related to propofol. It acts as a positive allosteric modulator of the GABA_A receptor, enhancing GABA-mediated chloride influx and producing sedative–hypnotic effects [13–15]. Preclinical studies have shown that ciprofol is approximately four to five times more potent than propofol, with improved pharmacokinetic stability and a wider safety margin [16–18]. Clinically, it demonstrates rapid onset and recovery, with good tolerability in procedures such as gastrointestinal endoscopy and general anesthesia [19, 20].

Compared with propofol, ciprofol appears to exert a milder influence on hemodynamics, with a lower incidence of intraoperative hypotension and respiratory depression reported in recent randomized trials [21, 22]. However, these findings remain inconsistent, and quantitative evidence on its cardiovascular safety is limited [23–25].

Importantly, intraoperative hypotension should not be viewed merely as an isolated adverse event, but rather as an intermediate surrogate marker on the causal pathway leading to postoperative organ injury, including myocardial injury, acute kidney injury, stroke, and increased mortality. Accumulating evidence suggests that even brief or moderate reductions in mean arterial pressure are associated with clinically relevant end-organ dysfunction. Accordingly, we formulated the a priori hypothesis that, compared with propofol, ciprofol is associated with a lower incidence of intraoperative hypotension, thereby potentially reducing the risk of downstream organ injury. In this systematic review and meta-analysis, intraoperative hypotension was therefore prespecified as the primary outcome, reflecting its role as a mechanistically and clinically meaningful surrogate endpoint rather than a simple adverse event. To test this hypothesis and to assess the robustness of the accumulated evidence, we further applied TSA in addition to conventional meta-analytic methods.

Methods

Ethical approval is not necessary for this study as it is a systematic review.

Study registration

We conducted a systematic review and meta-analysis to compare the effects of ciprofol and propofol on the incidence of intraoperative hypotension following anesthesia, adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The study is registered in the International Prospective Register of Systematic Reviews (PROSPERO) on 14 October 2023 (Registration No: CRD42023469127).

Search approach and eligibility standards

A comprehensive search was conducted in PubMed, Embase, and the Cochrane Library databases up to December 10, 2025, with no language restrictions. Keywords and MeSH terms, including “ciprofol”, “propofol”, and “hypotension”, were combined using Boolean operators (AND, OR). The complete, database-specific and reproducible search strategies for PubMed, Embase, and the Cochrane Library are provided in Supplementary Table 4. Manual searches of grey literature (e.g., conference abstracts, dissertations) and reference lists of original studies, case reports, and reviews were also performed to ensure comprehensiveness.

Study selection and data extraction

The data search included author names, publication years, anesthesia and surgery details, interventions, intraoperative hypotension incidence, and total patients. Two independent reviewers (Z.C. and J.J.J.) screened eligible studies and extracted key data: study specifics (authors, publication year, location), patient demographics (sample size, age, sex ratio, surgery type), and intervention details (ciprofol and propofol dosages). Primary outcomes focused on intraoperative hypotension, while secondary outcomes included respiratory depression, hypoxemia, injection pain, bradycardia, and awakening times. The inclusion/exclusion criteria were assessed by the reviewers, with any disputes resolved through discussion among all authors, ensuring rigorous and reliable selection.

Inclusion criteria

Studies were eligible for inclusion if they met all of the following criteria:

1. RCTs directly comparing ciprofol and propofol, and reporting the incidence of intraoperative hypotension, defined as a ≥ 20% reduction in mean arterial pressure or systolic blood pressure.

2. Studies involving patients undergoing procedural sedation or general anesthesia.

3. Studies reporting explicit data on intraoperative hypotension incidence or providing sufficient information to allow accurate data extraction or calculation.

4. Studies in which the intervention group received ciprofol and the control group received propofol, with dosages administered within clinically recommended ranges.

Exclusion criteria

Studies were excluded if they met any of the following criteria:

1. Non-randomized studies, including reviews, non-clinical studies, case reports, editorials, conference abstracts, or observational designs.

2. Studies with incomplete, unavailable, or irretrievable outcome data.

3. Duplicate publications.

4. Studies not directly comparing ciprofol with propofol.

5. Studies with inappropriate outcome definitions or clearly flawed statistical analyses.

6. Studies reporting outcomes unrelated to intraoperative hypotension or other predefined endpoints.

Information extraction and evaluation of bias risk

Two authors (Z.M.M. and Z.J.F) independently assessed the risk of bias for each randomized controlled trial using the revised Cochrane risk-of-bias tool for randomized trials (RoB 2.0), released in 2019. This tool evaluates bias across five domains: (1) bias arising from the randomization process, (2) bias due to deviations from intended interventions, (3) bias due to missing outcome data, (4) bias in measurement of the outcome, and (5) bias in selection of the reported result. Each domain was judged as “low risk of bias,” “some concerns,” or “high risk of bias,” and an overall risk-of-bias judgment was derived for each study following the RoB 2.0 guidance.

In addition, we extracted information on concomitant medications administered during induction and maintenance of anesthesia, such as opioids, benzodiazepines, neuromuscular blockers, and adjunctive agents. These data are summarized in Supplementary Table 3.

Trial sequential analysis

The TSA was performed using version 0.9.5.10 beta software to assess sample size adequacy and determine the need for further research. We calculated the required information size considering trial heterogeneity, which reflects study variability, by summing differences and accounting for sampling errors. TSA was conducted with a 5% type I error risk, a standard in meta-analyses. The essential data size was set with a 5% alpha and 20% beta error. Surpassing the monitoring threshold suggests strong evidence, while maintaining it may lead to more trials.

For TSA, the required information size was calculated assuming a two-sided type I error of 5% and a type II error of 20% (power 80%). A conservative anticipated relative risk reduction of 20% was prespecified, consistent with prior trial sequential analyses of perioperative hemodynamic outcomes. The control event proportion was derived from the pooled event rate in the propofol groups of included trials. Diversity-adjusted required information size was calculated to account for between-trial heterogeneity. A conventional two-sided O’Brien–Fleming-type monitoring boundary was applied.

TSA was performed only for intraoperative hypotension and injection pain, as intraoperative hypotension was the primary outcome of our study and injection pain was considered an important clinical outcome. In addition, both endpoints had a sufficient number of included studies to allow meaningful statistical evaluation.

Handling of outcome definition heterogeneity and sensitivity analysis

To address the potential heterogeneity in the definitions of intraoperative hypotension among included studies, we systematically extracted the original definitions from each study (see Supplementary Table 1 in the Supplementary Material). We conducted a sensitivity analysis restricted to studies that defined intraoperative hypotension as systolic blood pressure (SBP) < 90 mmHg or a decrease of 20–30% from baseline, to assess the robustness of the pooled results.

Quality analysis of evidence

With a deep commitment to rigor and transparency, we assessed the quality of evidence for each outcome using the GRADE system. This respected framework enabled us to evaluate the certainty of the evidence with careful consideration of various factors: the study design, risk of bias, inconsistency, indirectness, imprecision, and the potential for publication bias. By systematically weighing these elements, we were able to classify the evidence from high to very low certainty, ensuring that our conclusions were both robust and grounded in the highest standards of scientific evaluation. This thoughtful approach reflects our dedication to providing the most reliable, evidence-based insights possible. Because injection pain is a subjective patient-reported outcome and safety outcomes were variably defined across trials, these outcomes were evaluated with additional scrutiny of outcome measurement and ascertainment and were downgraded when definitions or measurement methods were unclear or inconsistent.

Statistical analysis

Statistical analyses were performed using Review Manager 5.4 and Stata 17.0. Dichotomous variables like the incidence of intraoperative hypotension, injection pain, respiratory depression, hypoxemia, bradycardia, body movement, dizziness, hypertension, PONV, rash, and tachycardia were summarized as RR with 95% CI. For binary outcomes, pooled analyses were primarily based on unadjusted event counts extracted from each trial. Although some studies reported adjusted effect estimates, these were not consistently available and were therefore not pooled. For binary outcomes with low event rates or zero-event arms, RevMan applies Mantel–Haenszel methods with continuity corrections, which may yield unstable variance estimates. Although more advanced methods (e.g., REML τ² estimation, Hartung–Knapp adjustment, or beta-binomial models) may be preferable in such contexts, these approaches were not implementable within the primary software used. To mitigate this limitation, we conducted prespecified sensitivity analyses excluding very small trials with fragile event proportions, which did not materially alter the main conclusions. Pooled RRs were calculated to estimate the differences in these incidences between ciprofol and propofol groups. For awakening time, which is a continuous variable, the pooled MD was used for assessment. The overall effect was determined by the Z test, with a P-value < 0.05 considered statistically significant. All pooled estimates were calculated using a random-effects model to account for expected clinical and methodological heterogeneity across trials, regardless of the magnitude of statistical heterogeneity. Statistical heterogeneity was assessed using τ² and I² statistics but was not used to determine the choice of pooling model. Subgroup analyses were conducted in an exploratory manner to investigate potential sources of heterogeneity.

Sensitivity analysis was conducted to evaluate the impact of individual studies on the overall results, with a focus on studies with low or unclear risk of bias and another sensitivity analysis was conducted to assess whether the pooled estimate for intraoperative hypotension was unduly influenced by very small trials with potentially fragile event proportionsPublication bias was assessed using funnel plots and Egger’s test. Subgroup analyses were performed based on patient age (old vs. not old), dosage of administration (≥ 0.4 mg/kg vs. < 0.4 mg/kg), and type of procedure(outpatient examination vs. surgical procedures) to explore potential sources of heterogeneity and to further understand the relationship between ciprofol, propofol, and the various outcomes. For subgroup analyses, “elderly” was defined as patients aged ≥ 65 years, consistent with the definitions used in the majority of included trials, while “non-elderly” referred to patients aged < 65 years. If individual trials used different age thresholds, subgroup classification followed the original study definitions.

For studies with missing, unclear, or inconsistently reported outcome data, attempts were made to extract usable information from the full text, Supplementary materials, or trial registries. Trials with irretrievable or incomplete outcome data were excluded a priori. When outcome definitions varied across studies, we retained the original definitions and explored their impact through prespecified sensitivity analyses. No outcome data were imputed.

All analyses were prespecified and conducted according to a random-effects framework, reflecting the assumption that true intervention effects may vary across clinical settings rather than representing a single common effect.

Results

Study selection

As shown in the flow diagram (Fig. 1), the search of PubMed, Embase, and Cochrane Library and initially, 105 records were identified from the databases. After removing 35 duplicates, 70 records remained for further screening. Upon reviewing the titles and abstracts, 11 records were excluded. Efforts to retrieve the full text of the remaining 59 records were made, but 12 could not be obtained due to unavailability. Of the 47 records assessed for eligibility, 13 were excluded: 5 for incomplete data, 4 for being re-analyses of the same data, and 4 due to their study design. In the end, 34 studies [17, 18, 20, 26–56] comprising 37 reports, were included in the final review.

Fig. 1.

Flow diagram of the inclusion and exclusion process

Study characteristics

This meta-analysis included 34 randomized [17, 18, 20, 26–56] controlled trials comprising 5,162 patients, with individual sample sizes ranging from 6 to 220 participants. Participants’ ages spanned from early adulthood to advanced age (approximately 31–79 years), and both sexes were represented, although sex distribution varied across studies. Most patients were classified as ASA physical status I-II, with a minority of ASA III patients included, particularly in studies involving major surgery or intensive care unit sedation.

The trials encompassed a wide range of clinical settings, including gastrointestinal endoscopy, gynecological procedures, fiberoptic bronchoscopy, elective and major surgeries, thoracoscopic and urological surgery, orthopedic procedures, and ICU sedation. Ciprofol and propofol were administered for induction alone or for induction and maintenance. Induction doses of ciprofol typically ranged from 0.3–0.5 mg/kg, whereas propofol doses ranged from 1.0–4.0 mg/kg. All included studies reported intraoperative hypotension as an outcome. Detailed study characteristics are summarized in Table 1.

Table 1.

General information of patients with incidence of intraoperative hypotension

Study	Age	Sex (Male/Female)	ASA Class I/II/III (n, %)	Comparisons (Group)	Operation	Hypotension	Total
Chen, B. Z. 2022 [39]	33.9 ± 9.1	0/60	I:32(53.3), II:28(46.7), III:0(0),	Ciprofol 0.4 mg/kg	Gynecological surgery	25	60
	33.8 ± 9.6	0/60	I:34(56.7), II:26(43.3), III:0(0),	Propofol 2 mg/kg		36	60
Deng, J.0.4 2023 [40]	-	-	I:6(100), II:0(0), III:0(0), IV:0(0)	Ciprofol 0.4 mg/kg	Healthy subjects	5	6
			I:6(100), II:0(0), III:0(0)	Propofol 2 mg/kg		4	6
Deng, J.0.6 2023 [40]	-	-	I:6(100), II:0(0), III:0(0)	Ciprofol 0.6 mg/kg	Healthy subjects	6	6
			I:6(100), II:0(0), III:0(0)	Propofol 3 mg/kg		6	6
Deng, J.0.8 2023 [40]	-	-	I:6(100), II:0(0), III:0(0)	Ciprofol 0.8 mg/kg	Healthy subjects	5	6
			I:6(100), II:0(0), III:0(0)	Propofol 4 mg/kg		6	6
Ding, G. 2024 [41]	73.6 ± 7.2	73/69	I:0(0), II:98(69.0), III:44(31.0)	Ciprofol 0.3–0.4 mg/kg maintenance 0.8–1.2 mg/kg/h	ERCP	27	142
	73.7 ± 7.2	72/70	I:0(0), II:95(66.9), III:47(33.1)	Propofol 1.5–2.0 mg/kg maintenance 4.0–12.0 mg/kg/h		31	142
Gao, S. H. 2024 [42]		34/48	I:16(19.5), II:66(80.5), III:0(0)	Ciprofol 0.4 mg/kg	Colonoscopy	54	82
		32/50	I:20(24.4), II:62(75.6), III:0(0)	Propofol 2 mg/kg		66	82
Lan, H. 2023 [18]	41.7 ± 11.5	0/75	I:43(57.3), II:32(42.7), III:0(0)	Ciprofol 0.4 mg/kg maintenance 0.6–1.2 mg/kg/h	Hysteroscopy	30	75
	43.7 ± 11.1	0/74	I:36(48.6), II:38(51.4), III:0(0)	Propofol 2.0 mg/kg maintenance 3.0–6.0 mg/kg/h	Hysteroscopy	51	74
Li, J. 2022 [43]	43.8 ± 11.8	55/89	I:115(79.9), II:29(20.1), III:0(0)	Ciprofol 0.4 mg/kg	Gastroscopy and colonoscopy	18	144
	44.1 ± 11.3	63/82	I:118(81.4), II:27(18.6), III:0(0)	Propofol 1.5 mg/kg	Gastroscopy and colonoscopy	11	145
Li, T. 2024 [44]	46.36 ± 12.3	48/60	I:70(64.8), II:38(35.2), III:0(0)	Ciprofol 0.5 mg/kg	Gastroscopy	32	108
	47.34 ± 11.20	49/60	I:71(65.1), II:38(34.9), III:0(0)	Propofol 2 mg/kg		44	109
Liang, P. 2023 [45]	38.5 ± 10.1	23/63	I:48(55.8), II:38(44.2), III:0(0)	Ciprofol 0.4 mg/kg mamaintenance 0.8 mg/kg	Nonemergency, noncardiothoracic, and nonneurosurgical elective surgery	32	86
	40.5 ± 10.1	10/32	I:22(52.4), II:20(47.6), III:0(0)	Propofol 2.0 mg/kg mamaintenance 5.0 mg/kg/h		17	42
Liang, Z. 2024 [46]	-	34/38	I:4(5.6), II:37(51.4), III:31(43.1)	Ciprofol 0.2–0.5 mg/kg maintenance 0.40 −3 mg/kg/h	Laparoscopic major abdominal surgery	24	72
	-	40/32	I:6(8.3), II:35(48.6), III:28(38.9)	Propofol 1.0–2.0 mg/kg maintenanc 4.0–12.0 mg/kg/h		39	72
Liao, J. 2023 [20]	44.98 ± 11.74	87/98	I:79(42.7), II:106(57.3), III:0(0)	Ciprofol 0.4 mg/kg	Gastroscopy	23	185
	45.35 ± 11.12	77/106	I:62(33.9), II:121(66.1), III:0(0)	Propofol 4 mg/kg		39	183
Liu Y. 2022 [17]	-	11/15	I:13(50.0), II:13(50.0), III:0(0)	Ciprofol 0.1–0.2 mg/kg maintenance 0.30 mg/kg/h,	Intensive care unit patients receiving mechanical ventilation	2	26
	-	8/5	I:6(46.2), II:7(53.8), III:0(0)	Propofol 0.5–1.5 mg/kg maintenance 1.50 mg/kg/h		3	13
Liu, Y. 2024 [47]	55.91 ± 12.98	69/45	Not mentioned	Ciprofol 0.1–0.2 mg/kg maintenance 0.30 mg/kg/h	Eligible ICU patients who were expected to require sedation	6	114
	53.16 ± 12.73	34/24	Not mentioned	Propofol 0.5–1.5 mg/kg maintenance 1.50 mg/kg/h		8	58
Lu, Y. F. 2024 [48]	78.7 ± 3.2	10/20	I:0(0), II:22(73.3), III:8(26.7)	Ciprofol 0.3 mg/kg	Hip Fracture Surgery	11	30
	79.3 ± 3.9	14/16	I:0(0), II:23(76.7), III:7(23.3)	Propofol 1.5 mg/kg		19	30
Luo, Z. 2022 [49]	46.60 ± 15.31	62/72	I:63(47.0), II:68(50.7), III:3(2.2)	Ciprofol 0.4 mg/kg	Fberoptic bronchoscopy	30	134
	46.90 ± 13.98	73/60	I:50(37.6), II:82(61.7), III:1(0.8)	Propofol 2 mg/kg		37	133
Man, Y. 2023 [50]	42.2 ± 9.46	0/64	I:18(28.1), II:46(71.9), III:0(0)	Ciprofol 0.5 mg/kg	Gynecological day surgery	25	64
	44.1 ± 9.4	0/64	I:14(21.9), II:50(78.1), III:0(0)	Propofol 2 mg/kg		36	64
Wu, B. 2022 [51]	58.02 ± 5.47	26/20	I:10(21.7), II:36(78.3), III:0(0)	Ciprofol 0.3 mg/kg	Fiberoptic bronchoscopy	5	46
	57.48 ± 5.28	24/22	I:8(17.4), II:38(82.6), III:0(0)	Propofol 1.2 mg/kg		12	46
Zeng, Y. 2022 [52]	42.5 ± 10.3	11/19	I:16(53.3), II:14(46.7), III:0(0)	Ciprofol 0.4 mg/kg	-	14	30
	46.4 ± 11.2	3/7	I:4(40.0), II:6(60.0), III:0(0)	Propofol 2 mg/kg	-	7	10
Zhang, J. 2023 [53]	-	-	Not mentioned	Ciprofol 0.3 mg/kg	Bidirectional endoscopy	12	93
	-	-	Not mentioned	Propofol 1.2 mg/kg		17	92
Zhong, J. 2023 [74]	57.3 ± 12.9	37/32	I:14(20.3), II:49(71.0), III:6(8.7)	Ciprofol 6 mg/kg/h	Endoscopic procedure(ESD, ERCP, or FB)	12	69
	57.6 ± 13.3	33/36	I:14(20.3), II:48(69.6), III:7(10.1)	Ciprofol 8 mg/kg/h		19	69
	56.9 ± 13.1	41/28	I:10(14.5), II:53(76.8), III:6(8.7)	Propofol 40 mg/kg/h		24	69
Zhou, R. 2024 [55]	48.02 ± 12.05	52/68	Not mentioned	Ciprofol 0.3–0.5 mg/kg	Gastrointestinal Endoscopy	22	120
	48.72 ± 9.97	66/54	Not mentioned	Propofol 1.5–2.5 mg/kg		32	120
Zou, H. 2024 [56]	68.8 ± 5.9	35/22	I:0(0), II:49(86.0), III:8(14.0)	Ciprofol 0.3 mg/kg	Elective surgery	15	57
	68.6 ± 6.2	36/22	I:0(0), II:50(86.2), III:8(13.8)	Propofol 1.5 mg/kg		28	58
Chen, J. 2025 [26]	45.6 ± 11.8	54/34	I:9(10.2%), II:79(89.8%), III:0(0%)	Ciprofol 0.4 mg/kg	Gastrointestinal endoscopy	22	88
	44.8 ± 10.8	59/29	I:11(12.5%), II:77(87.5%), III:0(0%)	Propofol 2 mg/kg		37	88
Cheng, S. 2025 [27]	35.313 ± 7.519	0/62	I:48(77.4%), II:14(22.6%), III:0(0%)	Ciprofol 0.4 mg/kg (induction) + 1–1.5 mg/(kg·h) (maintenance)	Hysteroscopy	4	62
	37.253 ± 7.508	0/62	I:49(79.0%), II:13(21.0%), III:0(0%)	Propofol 2.0 mg/kg (induction) + 5–7 mg/(kg·h) (maintenance)		7	62
Chi, X. 2025 [28]	31.45 ± 7.85	37/69	I:0(0%), II:41(38.68%), III:65(61.32%)	Ciprofol 0.5 mg/kg (induction)	Laparoscopic sleeve gastrectomy	15	106
	32.65 ± 7.33	30/76	I:0(0%), II:46(43.40%), III:60(56.60%)	Propofol 2.5 mg/kg (induction)		38	106
Fan, G. F. 2025 [29]	68.4 ± 5.2	13/17	I:0(0%), II:7(23.3%), III:23(76.7%)	Ciprofol 0.3–0.4 mg/kg (induction) + 1.0–1.5 mg/(kg·h) (maintenance)	Spinal surgery	5	30
	67.1 ± 4.5	12/18	I:0(0%), II:8(26.7%), III:22(73.3%)	Propofol 1.5–2 mg/kg (induction) + 4–8 mg/(kg·h) (maintenance)		12	30
Jian, C. F. 2025 [30]	49.54 ± 11.02	22/13	I:8(22.9%), II:20(57.1%), III:2(5.7%)	Ciprofol 0.2 mg/kg (induction)	Gastrointestinal endoscopy	9	35
	50.89 ± 10.69	20/15	I:6(17.1%), II:21(60.0%), III:3(8.6%)	Propofol 1 mg/kg (induction)		2	35
Lan, L. F. 2025 [31]	53(45.5–57)	15/15	I-II:30(100%), III:0(0%)	Ciprofol 0.4 mg/kg (induction); 1–1.5 mg/kg/h (maintenance)	Thoracoscopic surgery	12	30
	53(43–58)	13/17	I-II:30(100%), III:0(0%)	Propofol 2.5–4.5 mg/ml (induction); 2.5–3.5 mg/ml (maintenance)		16	30
Li, L. 2025 [32]	47.99 ± 16.21	117/103	I:107(48.6%), II:104(47.3%), III:9(4.1%)	Ciprofol 0.4 mg/kg (induction)	Gastroscopy	27	220
	46.57 ± 15.83	106/98	I:85(41.1%), II:112(54.1%), III:7(3.4%)	Propofol 1.5 mg/kg (induction)		44	207
Nie, J. 2025 [33]	43.5 ± 12.6	46/30	I:3(3.9%), II:72(94.7%), III:0(0%)	Ciprofol 0.4 mg/kg (induction)	Fiberoptic bronchoscopy	5	76
	44.1 ± 11.8	45/31	I:2(2.6%), II:74(97.4%), III:0(0%)	Propofol 2.5 mg/kg (induction)		53	76
Shi, S. Q. 2025 [34]	56.3 ± 11.0	31/24	I:3(5.5%), II:49(89.1%), III:3(5.5%)	Ciprofol 0.4 mg/kg (induction) + 0.8–1.5 mg/(kg·h) (maintenance)	Ureteroscopy	18	55
	55.1 ± 11.7	33/21	I:3(5.6%), II:44(81.5%), III:7(13.0%)	Propofol 2.0 mg/kg (induction) + 4–10 mg/(kg·h) (maintenance)		31	54
Tan, C. 2025 [35]	65.0 ± 9.0	17/22	I:0(0%), II:39(100%), III:0(0%)	Ciprofol 0.5 mg/kg (induction) + 1.5–2 mg/(kg·h) (maintenance)	Meningioma resection	23	39
	67.6 ± 6.5	16/23	I:0(0%), II:39(100%), III:0(0%)	Propofol 1.5–2 mg/kg (induction) + 5–8 mg/(kg·h) (maintenance)		37	39
Wang, Y. 2025 [36]	55.98 ± 9.59	23/37	I:4(6.7%), II:56(93.3%), III:0(0%)	Ciprofol 0.4 mg/kg (induction) + 0.4–2.4 mg/(kg·h) (maintenance)	Thoracoscopic surgery	2	60
	54.67 ± 10.07	21/39	I:3(5.0%), II:57(95.0%), III:0(0%)	Propofol 2.0 mg/kg (induction) + 4.0–12 mg/(kg·h) (maintenance)		19	60
Zhan, L. 2025 [37]	61.8 ± 9.2	64/21	I:38(44.7%), II:42(49.4%), III:5(5.9%)	Ciprofol 0.3–0.4 mg/kg (induction) + 1.0–1.5 mg/(kg·h) (maintenance)	Urological surgery	18	85
	61.1 ± 10.5	62/19	I:38(46.9%), II:39(48.2%), III:4(4.9%)	Propofol 1.5–2.0 mg/kg (induction) + 4–8 mg/(kg·h) (maintenance)		42	81
Zhang, Y. J. 2025 [38]	44.5 ± 8.0	26/14	I:19(47.5%), II:21(52.5%), III:0(0%)	Ciprofol 0.4 mg/kg	Gastroscopy	12	40
	48.0 ± 9.8	26/14	I:15(37.5%), II:25(62.5%), III:0(0%)	Propofol 2.0 mg/kg		22	40

Definitions of intraoperative hypotension varied slightly among the included studies. Most studies used a threshold of SBP < 90 mmHg or a decrease of 20–30% from baseline, while a minority used other criteria or did not specify their definition. The details are summarized in Supplementary Table 1.

Concomitant medications varied across studies. Commonly reported agents included opioids (e.g., sufentanil, remifentanil, fentanyl), benzodiazepines (e.g., midazolam), muscle relaxants (e.g., rocuronium, cis-atracurium), and adjunctive agents such as lidocaine, esketamine, and sevoflurane. Details of concomitant medications used in each trial are summarized in Supplementary Table 3. Several studies did not report specific information, which may contribute to heterogeneity.

The methodological quality of the included studies

The methodological quality of the included randomized controlled trials was assessed using the revised Cochrane Risk of Bias tool for randomized trials (RoB 2.0). Most trials were rated as either “low risk of bias” or “some concerns,” with the most frequent issues related to blinding of participants and outcome assessors, and allocation concealment. Overall, outcome data were largely complete across studies, reducing the risk of attrition bias. A detailed risk-of-bias judgment across the five RoB 2.0 domains is shown in Fig. 2.

Fig. 2.

Graph of the risk of bias of the included studies

Results of meta-analysis

Ciprofol versus propofol on intraoperative hypotension

A total of 34 trials [17, 18, 20, 26–56], involving 5,162 patients, assessed the incidence of intraoperative hypotension by comparing ciprofol to propofol. The pooled analysis revealed a significantly lower incidence of intraoperative hypotension in the ciprofol group (RR = 0.65, 95% CI: 0.57–0.73), as illustrated in Fig. 3. Both Begg’s test (P = 0.398) and Egger’s test (P = 0.782) confirmed the absence of significant publication bias in these comparisons (Fig. 4). Additionally, subgroup analyses were conducted to explore factors influencing the occurrence of intraoperative hypotension, providing further insights into this outcome.

Fig. 3.

Results of the incidence of intraoperative hypotension

Fig. 4.

Results of the Begg’s test and Egger’s test

Subgroup analysis

Age of patients

Ciprofol significantly reduced the incidence of intraoperative hypotension (pooled RR of 6 trials [17, 29, 35, 37, 41, 56]: 0.59, 95% CI: 0.48 to 0.72) when the age of patients is old, and also not old (pooled RR of 23 trials [18, 20, 26–28, 30–34, 36, 38, 43–45, 47, 49–52]: 0.61, 95% CI: 0.51 to 0.72) (Figs. 5A).

Fig. 5.

Results of subgroup analysis of the incidence of intraoperative hypotension by age of patients (A), dosage of administration (B), and type of procedure (C)

Dosage of administration

Ciprofol significantly reduced the incidence of intraoperative hypotension (pooled RR of 18 [18, 20, 26–28, 31–36, 38–40, 42–45, 49, 50, 52] trials: 0.59, 95% CI: 0.49 to 0.72) when the dosage of ciprofol is more than or equal to 0.4 mg/kg, but also less than 0.4 mg/kg (pooled RR of 5 trials [29, 30, 37, 47, 51]: 0.56, 95% CI: 0.28 to 1.12) (Figs. 5B).

Type of procedure

Ciprofol reduced the incidence of intraoperative hypotension significantly (pooled RR of 16 trials [20, 26, 30, 32, 33, 38, 41–44, 49, 51, 53–55]: 0.67, 95% CI: 0.54 to 0.83) when the type of procedure is outpatient examination, but also surgical procedures (pooled RR of 15 trials [18, 27–29, 31, 34–37, 39, 45, 46, 48, 50, 56]: 0.57, 95% CI: 0.49 to 0.66) (Figs. 5C).

The incidence of respiratory depression, hypoxemia, injection pains and bradycardia

Ciprofol significantly reduced the incidence of injection pains (pooled RR of 28 trials [18, 20, 26–31, 33, 35–46, 48, 49]: 0.19, 95% CI: 0.14 to 0.25) (Figs. 6C), respiratory depression (pooled RR of 8 trials [17, 18, 30, 40, 42, 43, 51, 53]: 0.44, 95% CI: 0.29 to 0.69) (Figs. 6A), hypoxemia (pooled RR of 15 trials [20, 26, 27, 30, 32, 33, 38, 41, 43–45, 49, 51, 54]: 0.62, 95% CI: 0.49 to 0.79) (Figs. 6B), but not the incidence of and bradycardia(pooled RR of 26 trials [17, 18, 20, 27–31, 36–40, 42–54]: 1.12, 95% CI: 0.93 to 1.36)(Figs. 6D). These secondary outcomes were variably defined and were not primary endpoints in most trials; therefore, the pooled estimates should be interpreted as exploratory.

Fig. 6.

Results of the incidence of respiratory depression, hypoxemia, injection pains and bradycardia

The incidence of body movement, dizzness, hypertension, PONV, rash and tachycardia

Ciprofol significantly did not reduce the incidence of body movement (pooled RR of 6 trials [18, 38–40, 44, 50]: 0.98, 95% CI: 0.60 to 1.61) (Supplementary Figure S1A), dizzness (pooled RR of 8 trials [20, 30, 38, 42–44, 49, 55]: 0.93, 95% CI: 0.67 to 1.29) (Supplementary Figure S1B), hypertension (pooled RR of 14 trials [18, 27, 28, 31, 36, 38, 39, 46, 48, 50, 51, 53, 56]: 1.04, 95% CI: 0.76 to 1.43) (Supplementary Figure S1C), PONV (pooled RR of 5 trials [20, 30, 34, 44, 46]: 0.75, 95% CI: 0.46 to 1.21) (Supplementary Figure S1D), rash(pooled RR of 3 trials [45, 49, 52]: 2.06, 95% CI: 0.35 to 12.02)(Supplementary Figure S1E) and tachycardia (pooled RR of 11 trials [18, 27, 28, 31, 36, 39, 42, 48, 50, 53, 56]: 0.74, 95% CI: 0.52 to 1.06) (Supplementary Figure S1F).

Awakening time, the times to anesthesia induction success, duration of operation

Ciprofol was associated with a statistically significant but very small increase in awakening time (pooled MD of 9 trials [18, 26, 36, 38, 44, 45, 50, 51, 53]: 0.36, 95% CI: 0.09 to 0.64), which is unlikely to be clinically meaningful(Supplementary Figure S2A). No statistically significant difference was observed in time to anesthesia induction success(pooled MD of 9 trials [26, 31, 35, 36, 38, 41, 44, 51, 52]: 1.47, 95% CI: −1.43 to 4.37), with wide confidence intervals indicating substantial imprecision. (Supplementary Figure S2B). Similarly, there was no statistically significant difference in the duration of operation between ciprofol and propofol (pooled MD of 10 trials [18, 26, 30, 35, 36, 39, 45, 51, 52, 54]: −0.29, 95% CI: −1.69 to 1.10) (Supplementary Figure S2C).

Trial sequential analysis

Using data from 34 RCTs (5,162 patients) for intraoperative hypotension and 28 RCTs (4,320 patients) for injection pain, TSA was performed to evaluate the robustness of the findings. The required information size was calculated with adjustments for diversity, targeting a 5% type I error and a 20% type II error. Remarkably, the cumulative Z-curve surpassed both the futility and monitoring boundaries before achieving the required information size. This strongly supports the conclusion that ciprofol demonstrates a significant advantage over propofol in reducing the incidence of intraoperative hypotension and injection pain, providing robust statistical support for these findings, while acknowledging that the overall certainty of evidence remains limited(Supplementary Figure S3).

Sensitivity analysis

Sensitivity analysis restricted to studies using the strict definition (SBP < 90 mmHg or 20–30% reduction from baseline) showed a pooled RR of 0.66 (95% CI: 0.59–0.74; I² = 28.8%), consistent with the overall findings (see Supplementary Figure S4, Supplementary Table 2). Another sensitivity analysis excluding trials with fewer than 10 participants per group yielded consistent results (Supplementary Figure S5).

Quality of evidence

Certainty of evidence was summarized in Table 2. Briefly, certainty was low for intraoperative hypotension because of study limitations and potential reporting bias. For injection pain and respiratory depression, certainty was downgraded because outcome ascertainment and definitions were variably reported across trials (and injection pain is inherently subjective), despite large observed effect sizes. Other outcomes were generally supported by low to moderate certainty evidence.

Table 2.

GRADE summary between ciprofol and propofol after anesthesia

Quality assessment							Quality	Importance
No of studies	Design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations
Hypotension
34	randomised trials	Seriousa	Seriousi	no serious indirectness	no serious imprecision	reporting biasb	⊕⊕ОО LOW	CRITICAL
Bradycardia
26	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	no serious imprecision	reporting biasb	⊕⊕⊕О MODERATE	IMPORTANT
Hypoxemia
15	randomised trials	no serious risk of bias	no serious inconsistency	Seriousc	no serious imprecision	none	⊕⊕ОО LOW	CRITICAL
Injection pains
28	randomised trials	Seriousg	Seriousd	Serioush	no serious imprecision	very strong associatione	⊕⊕ОО LOW	CRITICAL
Respiratory depression
9	randomised trials	no serious risk of bias	no serious inconsistency	no serious indirectness	Seriousc	strong associationf	⊕⊕⊕О MODERATE	CRITICAL

^aLack of blinding in Lan H. 2022, Liu Y. 2022 [17], Zeng Y. 2022 [52]; Fiftheen studies lacked allocation concealmen

^bThere were manufacturer-sponsored studies

^cTotal number of events < 300

^dI² > 50%

^eRR < 0.2

^fRR < 0.5

^g Injection pain is a subjective patient-reported outcome; blinding and outcome measurement were often insufficiently described

^h Injection pain ascertainment varied across studies (e.g., binary incidence vs. scale-based assessment), contributing to indirectness

ⁱ Inconsistency was downgraded due to variability in effect estimates across studies, including differences in direction and magnitude of effects, despite a modest I² value under random-effects modeling

Discussion

Intraoperative hypotension is a common postoperative complication, particularly among high-risk elderly patients [57]. The main findings of this study were as follows: (1) Our meta-analysis indicates that ciprofol is associated with a significantly lower incidence of intraoperative hypotension compared with propofol; however, the certainty of evidence was rated as low. (2) When the type of procedure was outpatient examination, but also surgical procedures, ciprofol reduced the incidence of intraoperative hypotension. (3) Ciprofol was associated with lower incidences of injection pain, hypoxemia, and respiratory depression; however, these findings are based on secondary outcomes and should be interpreted as exploratory rather than confirmatory. (4) Although ciprofol was associated with a statistically significant but very small increase in awakening time, the absolute difference was minimal and unlikely to be clinically relevant.

Intraoperative hypotension represents a critical hemodynamic perturbation rather than a benign or isolated adverse event. Substantial evidence indicates that hypotension during anesthesia acts as an intermediate surrogate on the causal pathway linking anesthetic exposure to postoperative organ injury, including myocardial injury, acute kidney injury, cerebral ischemia, and increased mortality. From a pathophysiological perspective, inadequate organ perfusion during hypotensive episodes may precipitate ischemia–reperfusion injury and inflammatory cascades, ultimately contributing to adverse postoperative outcomes. In this context, the primary finding of our meta-analysis—that ciprofol is associated with a lower incidence of intraoperative hypotension compared with propofol—may have implications beyond short-term hemodynamic stability. Although the included trials were not designed or powered to detect differences in hard clinical endpoints such as organ injury or mortality, the observed reduction in hypotension suggests a potential downstream protective effect that warrants further investigation.

Very small trials with fragile event proportions may yield unstable estimates for binary outcomes. Although the sensitivity analysis supports the robustness of our findings, this methodological issue warrants cautious interpretation. Importantly, exclusion of these very small trials did not materially alter the pooled effect estimate, suggesting that the primary findings were not driven by micro trials.

A recent meta-analysis by Yang et al. [58] published in BMC Anesthesiology in 2024 evaluated the efficacy and safety of ciprofol versus propofol across procedural sedation settings both inside and outside the operating room. While their analysis focused on a broad range of sedation-related outcomes, the present study differs in several important aspects. First, we specifically concentrated on intraoperative hemodynamic outcomes, with hypotension as the primary endpoint, and applied trial sequential analysis and GRADE methodology to assess the robustness and certainty of evidence. Second, our updated search incorporated a larger number of randomized trials, enabling more precise estimation of cardiovascular safety outcomes. Together, these differences allow the present study to complement and extend existing evidence by providing a focused and methodologically rigorous assessment of hemodynamic risk during anesthesia.

Intraoperative hemodynamic stability is crucial for protecting vital organs such as the heart, brain, and kidneys [59]. Perioperative intraoperative hypotensionsignificantly increase the risk of myocardial injury, ischemic stroke, cognitive dysfunction, acute kidney injury, and mortality [60–63]. Studies show that blood pressure management strongly affects postoperative outcomes [64]. Turan et al. [65] found a link between intraoperative hypotension and 30-day mortality, major cardiac events, and kidney injury. Vernooij et al. [66] reported myocardial injury in 14.9% and acute kidney injury in 14.8% of cases, with mean arterial pressure (MAP) < 80 mmHg for over 10 min increasing mortality risk. Propofol's circulatory suppression limits its anesthesia use [67]. Propofol inhibits myocardial contraction by reducing intracellular calciumion concentration through a protein kinase C-dependent pathway [68]. It has also been suggested that propofol causes intraoperative hypotension during induction and maintenance of anesthesia in part by inhibiting vascular tone, with longer duration of propofol sedation and larger doses of propofol associated with longer duration and more severe intraoperative hypotension [69, 70]. Because all pooled estimates in this meta-analysis were derived using random-effects models, the reported relative risk reduction should be interpreted as the mean of a distribution of true effects rather than a single common effect. This implies that the magnitude of benefit associated with ciprofol may vary across different clinical contexts, with some patient populations or procedural settings experiencing greater benefit and others less. The observed overall effect therefore represents an average treatment effect across heterogeneous trials rather than a uniform effect applicable to all patients. Notably, individual trials reported effect estimates that differed substantially in both magnitude and direction, with some studies favoring ciprofol and others favoring propofol. Under a random-effects framework, such variability reflects genuine differences in underlying true effects across clinical contexts rather than statistical anomaly. This further supports interpreting the pooled estimate as an average effect across heterogeneous populations, rather than evidence of a single common treatment effect.

Our study demonstrated that ciprofol is associated with a lower incidence of intraoperative hypotension compared with propofol. This difference may be explained by ciprofol’s weaker suppression of myocardial contractility and cardiac output, resulting in more stable hemodynamics. In addition, ciprofol exhibits a higher binding affinity for GABA-A receptors, allowing effective sedation to be achieved at lower plasma concentrations and with smaller effective doses. Furthermore, ciprofol shows less interindividual variability in plasma concentrations and a more predictable clearance profile. In contrast, the rapid redistribution and metabolism of propofol may cause significant blood pressure fluctuations, especially during induction in critically ill elderly patients [71, 72]. From a pharmacokinetic and pharmacodynamic perspective, ciprofol and propofol share a common mechanism of action as positive allosteric modulators of the GABA_A receptor; however, important differences in potency, dose requirements, and cardiovascular effects may partly explain the observed differences in hemodynamic stability. Ciprofol demonstrates higher receptor affinity and achieves comparable anesthetic depth at lower plasma concentrations, which may reduce dose-dependent myocardial depression and vasodilation. In addition, its more predictable clearance and reduced interindividual variability may contribute to smoother hemodynamic profiles during induction and maintenance of anesthesia.

Our findings indicate that ciprofol is associated with a lower incidence of intraoperative hypotension compared with propofol, consistent with its pharmacological profile, suggesting reduced cardiovascular depression. However, the overall certainty of this evidence is limited due to significant heterogeneity across studies, small event numbers, and potential publication bias. Accordingly, the GRADE certainty of evidence for intraoperative hypotension was rated as low to moderate, indicating that further well-designed trials may substantially influence both the estimated effect size and its clinical interpretation.The observed heterogeneity may arise from varying surgical types, anesthetic doses, and definitions of intraoperative hypotension across studies. While these results suggest that ciprofol may offer advantages in patients at increased risk of hemodynamic instability, further large-scale, rigorously designed randomized controlled trials are required to confirm these findings and strengthen the evidence base.

For injection pain, ciprofol demonstrated a significantly lower incidence compared to propofol. Nevertheless, injection pain is a patient-reported outcome and many trials did not clearly describe blinding procedures or the exact pain assessment method, which increases susceptibility to measurement bias. We therefore downgraded the certainty of evidence for injection pain despite the large effect The soybean oil content in a 1% ciprofol solution is half that of propofol. Additionally, ciprofol exhibits greater potency due to its lower concentration in the aqueous phase compared to propofol [43, 73]. So the injection pain caused by ciprofol was less than that caused by propofol, which may be due to the lower drug concentration in the aqueous phase of the emulsion.

The wide range of reported event rates across trials (including studies reporting zero events) raises concerns regarding heterogeneity in outcome ascertainment and reporting thresholds, which further limits the interpretability of pooled estimates for these secondary safety outcomes.

The recovery time for patients administered ciprofol was slightly longer compared to those given propofol. The reason for this may be the very short half-life of the initial distribution of propofol and the high clearance rate [51, 74]. However, although the time to full alertness and recovery was significantly longer in the ciprofol group, it was within 10 to 15 min, a delay that may be clinically irrelevant. Further large-scale, double-blind RCTs with standardized definitions of intraoperative hypotension are warranted to confirm these findings.

Trial sequential analysis was applied in this study to reduce the risk of random errors and false-positive findings that may arise from repetitive testing in cumulative meta-analyses. By establishing monitoring boundaries analogous to interim analyses in randomized trials, TSA helps to determine whether the available evidence is sufficient to draw conclusions with respect to a prespecified effect size. However, it is important to emphasize that TSA does not overcome fundamental limitations related to study quality, small sample sizes, or clinical and methodological heterogeneity. The majority of included trials were small, single-center studies with variable definitions of intraoperative hypotension and heterogeneous clinical settings. Consequently, although TSA suggests that the cumulative sample size for intraoperative hypotension may be sufficient to reduce the risk of false-positive conclusions, it does not eliminate bias arising from these underlying limitations. Therefore, the TSA findings should be interpreted as supporting the statistical robustness of the observed association, rather than as definitive proof of a causal effect.

Limitations

First, age significantly influences blood pressure, with notable fluctuations occurring across different age groups, particularly in the elderly. As a result, age-related differences can have varying effects on outcomes. Second, the concomitant medications administered during anesthesia induction and maintenance, such as opioids, benzodiazepines, muscle relaxants, or adjuncts, may have influenced the incidence of intraoperative hypotension. While we summarized the reported regimens in Supplementary Table 3, incomplete reporting in several studies limits the ability to fully assess their impact. Third, Blood pressure monitoring methods, such as cuff measurement versus direct arterial monitoring, can similarly influence the results. Fourth, our analysis included trials conducted in both sedation and general anesthesia settings. Although this broadens applicability, it may also introduce heterogeneity, as the baseline risk of intraoperative hypotension and the concomitant medication regimens may differ between sedation and anesthesia procedures. Fifth, detailed reporting of ASA physical status was inconsistent across included trials, with most studies enrolling ASA I–II patients. Therefore, the generalizability of these findings to higher-risk populations, such as patients with ASA III–IV status, remains uncertain. Lastly, most included studies were small, single-center trials from one country, which may limit generalizability and introduce bias. In addition, pooled analyses were based on unadjusted event data, as adjusted effect estimates were not consistently reported across trials. In smaller studies, unadjusted estimates may be more susceptible to residual confounding, particularly for secondary outcomes. For binary outcomes, very small trials with extreme event proportions may produce fragile estimates that are sensitive to minor changes in event counts, representing an inherent limitation of proportion-based meta-analyses. Although alternative variance estimators and small-sample corrections (e.g., REML τ² estimation or Hartung–Knapp adjustment) may yield more conservative confidence intervals, these methods were not uniformly applicable across all outcomes and software platforms. Therefore, we focused on robustness checks via sensitivity analyses and cautious interpretation rather than relying on a single modeling assumption. Importantly, most included trials were small, single-center studies conducted within one country, which limits external validity and contributes to the overall low-to-moderate certainty of evidence. Moreover, the predominance of trials conducted within a limited geographic region further restricts external validity, underscoring the need for large, multinational randomized controlled trials to determine whether the observed hemodynamic advantage translates into clinically meaningful benefits across diverse healthcare systems and patient populations.

Conclusion

Ciprofol may be associated with a lower average risk of intraoperative hypotension compared with propofol across heterogeneous clinical settings. Findings for secondary outcomes, including injection pain, should be interpreted cautiously as exploratory. However, given the overall low to moderate certainty of evidence, these findings should be interpreted with caution. High-quality, large-scale, multinational randomized controlled trials are required to confirm the hemodynamic advantages of ciprofol and to determine their real-world clinical relevance and significance. As ciprofol receives regulatory approval and becomes available for use in broader geographic regions, large-scale, multicenter randomized controlled trials conducted across diverse healthcare settings will be essential to confirm these findings and to further define its pharmacodynamic and safety profile.

Abbreviations

ASA

American Society of Anesthesiologists classification

RCTs

Randomized controlled trials

RR

Risk Ratio

CI

Confidence Interval

GABA

Gamma-aminobutyric acid

GRADE

Grading of Recommendations Assessment, Development and Evaluation

TSA

Trial Sequential Analysis

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

International Prospective Register of Systematic Reviews

SBP

Systolic Blood Pressure

MAP

Mean Arterial Pressure

PONV

Postoperative Nausea and Vomiting

ERCP

Endoscopic Retrograde Cholangiopancreatography

ICU

Intensive Care Unit

MeSH

Medical Subject Headings

CNKI

China National Knowledge Infrastructure

SMD

Standardized Mean Difference

Authors’ contributions

Z.J.F. designed and supervised the study. J.J.J. and Z.M.M. performed all database searches. Z.C. extracted the data. J.J.J. conducted qualitative analysis. All authors participated in manuscript writing, contributed to critical comments and revised the manuscript.

Funding

Not applicable.

Data availability

The datasets supporting the conclusions of this article are included within the Article.

Declarations

Ethics approval and consent to participate

The article is in accordance with ethical standards.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.