Cipepofol versus Conventional Sedation for Pneumonia Patients Receiving Mechanical Ventilation: A Real-World Retrospective Study of Effectiveness and Hemodynamic Safety

Abstract

Purpose

Severe pneumonia often necessitates invasive mechanical ventilation (MV) and prolonged sedation in the ICU. This real-world retrospective cohort study aimed to compare cipepofol vs conventional sedation regimens in this vulnerable population, with MV duration as the primary endpoint and hemodynamic safety as a key safety outcome.

Patients and Methods

This retrospective cohort study evaluated mechanically ventilated patients with severe pneumonia who received continuous sedation for more than 48 hours. Patients were classified into cipepofol or conventional sedation (propofol-, midazolam- or dexmedetomidine-based) groups. Propensity score matching (PSM) was performed at a 1:1 ratio based on age and APACHE II scores.

Results

After PSM, 198 patients (99 per group) were analyzed. The primary endpoint–MV duration did not differ significantly between the cipepofol and conventional groups (median 218.23 vs 227.25 hours, P = 0.740). Subgroup analysis further confirmed that there was no significant difference in MV duration between propofol- and midazolam-based regimens compared to the cipepofol-regimen. Notably, cipepofol demonstrated superior hemodynamic safety, showing significantly lower incidences of early hypotension within the first hour (11.1% vs 27.3%, P=0.007) and overall bradycardia (19.1% vs 36.1%, P=0.014). Post-hoc subgroups analysis revealed that cipepofol reduced the risks of early hypotension and overall bradycardia compared to propofol- or midazolam-based regimens in hospital-acquired pneumonia patients. Multivariate analysis confirmed cipepofol use as an independent protective factor against early hypotension (adjusted OR 0.300, 95% CI: 0.136–0.665, P=0.003) and bradycardia (adjusted OR 0.414, 95% CI: 0.212–0.808, P=0.010). No significant differences were found regarding sedation success, length of ICU stay, 28-day ventilator-free days, 28-day mortality, vasopressor requirements, or organ function parameters (all P>0.05).

Conclusion

In mechanically ventilated patients with severe pneumonia requiring prolonged sedation, cipepofol did not shorten MV duration compared with propofol- or midazolam-based conventional regimens, but it significantly reduced risks for early hypotension and overall bradycardia events.

Trial Registration

ClinicalTrials.gov (NCT06514781).

Introduction

Severe pneumonia is a life-threatening infection characterized by acute hypoxemic respiratory failure and systemic inflammation. Pneumonia can be categorized into community-acquired pneumonia (CAP) and hospital-acquired pneumonia (HAP) based on the site of infection onset and clinical features, with significant difference in pathogen profiles, antimicrobial resistance risks and clinical outcomes.1 Among mechanically ventilated patients, HAP manifests as ventilator-associated pneumonia (VAP). Progression of pneumonia can result in the development of acute respiratory distress syndrome, potentially leading to respiratory failure, sepsis, shock, multiple organ failure and even death.2,3 A large retrospective cohort study in the United States involving 4,543 patients with microbiologically confirmed pneumonia found that in-hospital mortality rates were as high as 29.3% for VAP, 18.8% for HAP and only 10% for CAP.4 Regardless of the type of pneumonia, timely, appropriate, and adequate initial antibiotic therapy remains the core strategy for reducing mortality and mitigating antimicrobial resistance. However, etiological diagnosis in critically ill patients is clinically challenging, particularly in those with a history of prior antibiotic exposure or receiving mechanical ventilation (MV).5 It is a given that a substantial proportion of patients with severe pneumonia require invasive MV as a vital supportive measure.6 Accordingly, beyond targeted antimicrobial therapy, effective sedation is a cornerstone of the management of mechanically ventilated critically ill patients, as it plays an important role in ensuring patient comfort, safety and ventilator synchrony, and may exert a significant impact on the clinical trajectory of severe pneumonia.

Achieving adequate sedation while maintaining circulatory stability is particularly challenging for patients with severe pneumonia due to the coexistence of acute hypoxemic respiratory failure, systemic inflammation and, in many cases, hemodynamic compromise, which collectively heighten the patient’s sensitivity to the cardiovascular and metabolic effects of sedatives. Midazolam and propofol remain the most commonly administered sedatives for mechanically ventilated patients in intensive care units (ICUs).7 However, both of these drugs have notable limitations in this vulnerable population. Midazolam carries a substantial risk of drug accumulation in critically ill patients due to its high lipophilicity and active metabolites, which can lead to delayed success in spontaneous breathing trials and prolongation of MV duration.8,9 Propofol is characterized by a rapid onset and offset and is associated with a shorter wake-up time, length of ICU stay and duration of MV compared to midazolam.10,11 Dexmedetomidine provides light sedation with preserved respiratory drive, which may facilitate earlier extubation, but it consistently fails to achieve target deep sedation (Richmond Agitation-Sedation Scale [RASS] −1 to −4) when used as a monotherapy in severely ill patients.12 However, dexmedetomidine can shorten the length of ICU stay and duration of MV compared to benzodiazepines and propofol,13,14 although this drug is also associated with a high incidence of hypotension and bradycardia.15,16 In patients with severe pneumonia, pulmonary infection serves as the initiating factor of respiratory failure, and the weaning time directly determines the duration of ventilator exposure, which is in turn closely associated with the risk of increased risk of VAP, airway injury and sedative drug accumulation.17–19 It has been reported that excessive or inappropriate sedation has been associated with prolonged ventilation,20,21 and lighter or protocolized sedation strategies have been studied specifically to reduce the MV duration, risks of delirium and other extubation-related adverse outcomes.7,22 These inherent trade-offs between sedation depth, hemodynamic safety and MV duration highlight a critical unmet clinical need.

An ideal sedative for mechanically ventilated patients with severe pneumonia should exhibit rapid onset and recovery, minimal cardiovascular depression and no drug accumulation during prolonged infusion. Cipepofol, a newly developed sedative with a four- to five-fold higher potency than propofol, has exhibited a favorable hemodynamic profile in previous studies involving a variety of patients.23–25 The use of cipepofol in the ICU setting has gradually increased in recent years.26,27 A recently published study that compared sedation efficacy between cipepofol and propofol in mechanically ventilated patients found that cipepofol appeared to be associated with better hemodynamic stability, while achieving comparable sedative efficacy.28 However, the real-world data of cipepofol for patients with severe pneumonia, who are exceptionally vulnerable to hypotension, has not been specifically investigated. In patients with severe pneumonia, sedation management necessitates a deliberate balance among sedation depth, MV duration and hemodynamic tolerance, given the presence of infection-driven systemic inflammatory response, risk of circulatory instability, and elevated weaning failure rates. In light of the aforementioned research gaps, this study adopted a real-world retrospective cohort design to compare the effects of cipepofol vs conventional sedation regimens (propofol-, midazolam- or dexmedetomidine-based sedation) on the duration of MV in patients with severe pneumonia (including CAP/HAP) receiving prolonged sedation in the ICU. Additionally, the present study also assessed differences in circulatory adverse events, such as hypotension and bradycardia, across different sedation strategies, thereby providing real-world evidence to support sedative drug selection for these high-risk patients.

Materials and Methods

Study Design and Patients

This was a retrospective, single-center cohort study conducted in the Respiratory ICU of Xiangya Hospital, Central South University. The study protocol adhered to the Declaration of Helsinki and was approved by our institutional ethics committee (Approval No. 2024020151), which waived the requirement for informed consent due to the retrospective nature of the study. All patient data were strictly maintained with confidentiality and identifying information was anonymized prior to analysis. This study was registered with ClinicalTrials.gov (NCT06514781). Data were retrospectively extracted from the Hospital Information System and nursing records for patients admitted between February 1, 2023, and December 31, 2024.

Patients were eligible if they met all of the following inclusion criteria: (1) a diagnosis of severe pneumonia was made according to the Chinese expert consensus;29 (2) patients undergoing invasive MV requiring continuous sedation; and (3) aged ≥ 18 years. Patients were excluded if any of the following criteria applied: (1) known allergy to commonly used sedatives or analgesics; (2) no sedation‑agitation assessment documented; (3) transfer to another hospital during treatment; and (4) withdrawl from active treatment or were discharged from the ICU within 72 hours after initiation of sedation.

Patients were stratified into two groups based on whether they initially received cipepofol sedation for more than 48 hours (cipepofol group) or conventional sedation including propofol-, midazolam-, dexmedetomidine-based sedation (conventional group).

Outcomes and Exposure

The primary outcome was the duration of invasive MV. For patients intubated after ICU admission, MV duration was calculated from intubation to extubation or death, whichever occurred first. For patients already intubated at ICU admission, MV duration was calculated from ICU admission to extubation or death, whichever occurred first. Follow-up for this outcome was limited to 90 days. Secondary outcomes were sedation success, ICU length of stay, and 28-day mortality. Sedation success was defined as achieving a RASS score of −1 to −4 in more than 90% of recorded assessments for each patient. RASS assessments were evaluated throughout continuous sedation. Observation ended at sedation discontinuation, ICU discharge, or death, whichever occurred first, and was limited to 90 days. ICU length of stay was calculated from ICU admission to ICU discharge. For patients who died in the ICU, the date of death was used as the endpoint. The prespecified 28-day survival endpoint was reported as 28-day mortality in the present analysis, as complete 28-day vital status was available for all included patients. Accordingly, 28-day mortality was defined as death from any cause within 28 days after ICU admission. Pre-specified safety endpoints included the incidences of hypotension (systolic blood pressure [SBP] < 90 mmHg, or diastolic blood pressure [DBP] < 60 mmHg, or mean arterial pressure [MAP] < 70 mmHg or > 30% drop from baseline) and bradycardia (heart rate [HR] < 40 bpm or > 30% drop from baseline). These events were assessed from the first sedative dose through 7 days after the last dose. Exploratory safety analyses included the total duration (hours) and cumulative dose (mg) of vasoactive drugs administered during this period.

Post hoc exploratory hemodynamic analyses additionally evaluated early hypotension and bradycardia, defined as events meeting the above criteria within 1 hour after initiation of the index sedative regimen. Other post hoc exploratory outcomes included 28-day ventilator-free days (VFD) and laboratory safety outcomes comprising abnormal triglyceride elevation, liver or renal function deterioration, and severe creatine kinase elevation. The 28-day VFD was defined as the number of days alive and free from invasive MV within 28 days after ICU admission. Patients who died within 28 days were assigned 0 VFD. For survivors, days requiring invasive MV, including periods after reintubation, were not counted as ventilator-free days. Abnormal triglyceride elevation was defined as a triglyceride level > 400 mg/dL (approximately 4.5 mmol/L), or an increase of ≥ 50% from baseline with an absolute level > 200 mg/dL (approximately 2.3 mmol/L). Liver and renal function deterioration were defined as an increase of at least one grade from baseline in the respective function classification during treatment; the grading criteria are described in the Data Collection and Baseline Assessments section. Severe creatine kinase elevation was defined as a maximum creatine kinase level >1000 U/L together with an increase of ≥50% from baseline during treatment.

Data Collection and Baseline Assessments

Data were retrospectively extracted from the Hospital Information System and nursing records of the Respiratory Critical Care Medicine Ward at Xiangya Hospital of Central South University. The baseline characteristics analyzed included age, sex, body mass index (BMI), Acute Physiology and Chronic Health Evaluation II (APACHE II) score, oxygenation index (PaO₂/FiO₂), hemodynamic data (lactic acid concentration, HR, SBP, DBP and MAP), types of infected pathogens (bacterial, fungal, viral infections or multi-drug resistant [MDR]-bacterial infections), the adequacy of initial antibiotic regimens (appropriate and inappropriate) and laboratory parameters (eg, blood lipids and liver/renal function tests). Pneumonia is categorized based on the site and timing of initial pneumonia onset:30 CAP was defined as pneumonia contracted in a community setting or manifesting within 48 hours of hospital admission. HAP was defined as newly developed pneumonia occurring more than 48 hours post-admission, with no relevant medical history of the condition at admission and no evidence of being in the incubation period at that time. The initial antibiotic regimen was deemed appropriate if it aligned with current clinical practice guidelines, was initiated within 24 hours following the first confirmed diagnosis of pneumonia, and all identified causative pathogens demonstrated in vitro susceptibility to more than one antibiotic included in the regimen;31 in all other cases, the regimen was considered inappropriate. Liver function was classified using the Albumin-Bilirubin (ALBI) score:32 (1) grade 1 (mild liver impairment), ALBI score ≤ −2.60; (2) grade 2 (moderate liver impairment), ALBI score > −2.60 and ≤ −1.39; and (3) grade 3 (severe liver impairment), ALBI score > −1.39. Renal function was classified based on the estimated creatinine clearance rate (Ccr):33 (1) normal or compensated renal function; Ccr ≥ 90 mL/min; (2) mild renal impairment, Ccr 60–89 mL/min; (3) moderate renal impairment, Ccr 30–59 mL/min; and (4) severe renal impairment or renal failure, Ccr < 30 mL/min. Other organ supports such as continuous renal replacement therapy (CRRT) and extracorporeal membrane oxygenation (ECMO) were collected. Treatment details included sedation/analgesia duration, sedative agents administered, vasoactive drug requirements and MV parameters.

Statistical Analysis

All analyses were performed using R version 4.5.1. Normally distributed continuous variables are presented as the mean ± standard deviation and were compared using an independent t-test. Non-normally distributed continuous data are given as the median (interquartile range [IQR]) and were analyzed using the Mann–Whitney U-test. Categorical variables are presented as n (%), and intergroup comparisons were made using the chi-squared test or Fisher’s exact test, when the expected frequency in any cell was < 5. To reduce selection bias, 1:1 nearest-neighbor propensity score matching (PSM) was performed with a caliper width of 0.3 (a common choice for establishing covariate balance with a standardized mean difference [SMD] < 0.25), using age and APACHE II scores as the matching variables. All primary efficacy and safety analyses, including comparisons of study outcomes between groups, were conducted on the matched cohort using the same statistical tests as described above. Post hoc subgroup analyses were conducted for the associations between initial sedation regimen (cipepofol, propofol and midazolam) and clinical outcomes across all patients as well as CAP and HAP patients. Subsequently, a generalized linear model (GLM) was employed to calculate the odds ratios (ORs) or regression coefficient (β) relative to the initial cipepofol sedation regimen, along with their corresponding 95% confidence intervals (CIs). Univariate regression analyses were performed to identify factors associated with clinical outcomes. Subsequently, multivariate regression models were employed to assess the independent effect of the sedation regimen, with calculation of ORs or regression coefficient (β) and corresponding 95% CIs. Specifically, multivariate logistic regression adjusted for age, sex, APACHE II score and COVID-19 status was used for binary outcomes (eg, early hypotension and bradycardia), while multivariate linear regression was applied for continuous outcomes. P < 0.05 was taken to indicate statistical significance.

Data were missing for several variables. There were no missing data for the primary outcome (MV duration), 28-day VFD, 28-day mortality or variables used for PSM (age and APACHE II score). The missing data rate was low for blood pressure (3.8%, 10/260) and analgesic duration (1.5%, 4/260), so a complete-case analysis was performed for these variables. BMI (missing data rate: 40.4%, 105/260) and liver/renal function grades (0.4%, 1/260) were not included as covariates in any regression model and were reported only as baseline characteristics in Table 1. Changes in liver and renal function during treatment were analyzed separately as post hoc exploratory laboratory safety outcomes. Patients with no documentation of RASS score (27/260, 10.4%) were excluded from the analysis of sedation success rate (a secondary outcome); all available RASS records were used for the remaining patients. Available-case analysis without imputation was applied to triglyceride, liver/renal function, and creatine kinase outcomes because the availability of these laboratory measurements varied across patients in this retrospective cohort.

Table 1.

Patient Characteristics Before and After Propensity Score Matching

Variables	Before PSM (N = 260)				After PSM (N = 198)
	Cipepofol Group (n = 140)	Conventional Group (n = 120)	P-Value	SMD	Cipepofol Group (n = 99)	Conventional Group (n = 99)	P-Value	SMD
Demographic data
Age (years)	68.00 [58.00–75.25]	67.00 [58.00–75.00]	0.948	0.034	68.00 [58.50–74.50]	67.00 [58.00–74.00]	0.946	0.032
Female, n (%)	35 (25.0)	30 (25.0)	0.757	0.004	25 (25.5)	25 (25.3)	1.000	0.006
BMI (kg/m2)	22.19 ± 3.07	22.80 ± 4.14	0.363	0.168	22.30 ± 2.92	22.60 ± 4.06	0.660	0.086
Hemodynamic data
Lactic acid concentration (mmol/L)	1.20 [0.90–1.70]	1.30 [1.00–2.02]	0.072	0.303#	1.20 [0.90–1.70]	1.30 [1.00–1.85]	0.431	0.146
HR (beats/min)	98.00 [82.00–108.50]	95.00 [79.00–109.00]	0.552	0.061	98.00 [80.00–108.00]	94.00 [77.00–107.00]	0.514	0.056
SBP (mmHg)	129.19 ± 20.77	126.73 ± 26.50	0.413	0.103	128.83 ± 22.03	127.98 ± 26.50	0.810	0.035
DBP (mmHg)	66.61 ± 12.09	66.53 ± 15.64	0.965	0.006	66.30 ± 11.61	67.58 ± 15.64	0.523	0.093
MAP (mmHg)	87.47 ± 12.89	86.60 ± 16.97	0.646	0.058	87.14 ± 13.08	87.71 ± 17.19	0.797	0.037
Baseline characteristics
APACHE II score	17.00 [12.75–21.00]	22.00 [16.00–27.00]	< 0.001*	0.587#	17.00 [13.00–23.00]	19.00 [15.00–24.00]	0.124	0.199
Oxygenation index (PaO2/FiO2)	127.85 [84.93–206.63]	131.44 [83.50–205.62]	0.791	0.007	160.00 [88.26–232.92]	124.44 [81.38–202.36]	0.107	0.212
Types of pneumonia, n (%)			0.263	0.204			0.329	0.213
CAP	92 (65.7)	68 (56.7)			56 (56.6)	54 (54.5)
HAP	48 (34.3)	52 (43.3)			43 (43.4)	45 (45.5)
COVID-19 infection, n (%)	15 (10.7)	20 (16.7)	0.223	0.173	11 (11.1)	17 (17.2)	0.308	0.174
Types of infected pathogens, n (%)
Viral infection	47 (33.6)	34 (28.3)	0.438	0.113	31 (31.3)	29 (29.3)	0.877	0.044
Bacterial infection	107 (76.4)	95 (79.2)	0.704	0.066	80 (80.8)	78 (78.8)	0.860	0.050
Fungal infection	80 (57.1)	57 (47.5)	0.153	0.194	52 (52.5)	50 (50.5)	0.887	0.040
MDR-bacterial infection	56 (40.0)	42 (35.0)	0.483	0.103	45 (45.5)	37 (37.4)	0.313	0.165
Liver function grade, n (%)			0.442	0.160			0.524	0.162
1	11 (7.9)	8 (6.7)			8 (8.1)	6 (6.1)
2	91 (65.0)	70 (58.8)			64 (64.6)	59 (59.6)
3	38 (27.1)	41 (34.5)			27 (27.3)	34 (34.3)
Renal function grade, n (%)			0.393	0.218			0.330	0.266#
1	70 (50.0)	67 (55.8)			60 (60.6)	51 (51.5)
2	30 (21.4)	28 (23.3)			15 (15.2)	25 (25.3)
3	37 (26.4)	21 (17.5)			21 (21.2)	21 (21.2)
4	3 (2.1)	3 (2.5)			3 (3.0)	2 (2.0)
Treatment and organ support
Initial antibiotic regimen, n (%)			0.514	0.144			0.575	0.150
Appropriate	52 (37.1)	53 (44.2)			36 (36.4)	43 (43.4)
Inappropriate	88 (62.9)	67 (55.8)			63 (63.6)	56 (56.6)
Organ supports, n (%)
CRRT	22 (15.7)	35 (29.2)	0.014*	0.327#	17 (17.2)	21 (21.2)	0.588	0.103
ECMO	1 (0.7)	6 (5.0)	0.081	0.259#	1 (1.0)	3 (3.0)	0.613	0.144
Treatment details
Analgesia duration (hours)	162.00 [98.38–241.00]	151.00 [115.90–265.95]	0.788	0.017	153.47 [96.55–240.88]	145.83 [113.25–254.53]	0.541	0.021
Sedation duration (hours)	149.26 [98.12–225.29]	147.33 [115.10–244.00]	0.582	0.042	143.50 [95.11–235.94]	143.00 [111.50–236.67]	0.573	0.015

Notes: Data are presented as the mean ± standard deviation, median [interquartile range] or number (percentage). *P < 0.05 indicates statistical significance. Covariates were considered balanced if the absolute value of the SMD was < 0.25. ^# SMD exceeded 0.25. Pneumonia is categorized based on the site and timing of initial pneumonia onset. CAP was defined as pneumonia contracted in a community setting or manifesting within 48 hours of hospital admission. HAP was defined as newly developed pneumonia occurring more than 48 hours post-admission, with no relevant medical history of the condition at admission and no evidence of being in the incubation period at that time. The initial antibiotic regimen was deemed appropriate if it aligns with current clinical practice guidelines, is initiated within 24 hours following the first confirmed diagnosis of pneumonia, and all identified causative pathogens demonstrate in vitro susceptibility to more than one antibiotic included in the regimen; in all other cases, the regimen is considered inappropriate. Liver function was classified using the Albumin-Bilirubin score: grade 1 (≤ −2.60), grade 2 (> −2.60 and ≤ −1.39), or grade 3 (> −1.39). Renal function was classified based on the estimated creatinine clearance rate: grade 1 (≥ 90 mL/min), grade 2 (60–89 mL/min), grade 3 (30–59 mL/min), or grade 4 (< 30 mL/min). Missing data before PSM: BMI, 105/260 (40.4%); liver function grade, 1/260 (0.4%); renal function grade, 1/260 (0.4%); analgesia duration, 4/260 (1.5%); and blood pressure, 10/260 (3.8%). All baseline comparisons were based on available cases.

Abbreviations: APACHE II, Acute Physiology and Chronic Health Evaluation II; CAP, community acquired pneumonia; CRRT, continuous renal replacement therapy; ECMO, extracorporeal membrane oxygenation; DBP, diastolic blood pressure; HAP, hospital acquired pneumonia; MAP, mean arterial pressure; MDR, multidrug-resistant; PSM, propensity score matching; SBP, systolic blood pressure; BMI, body mass index; HR, heart rate; SMD, standardized mean difference.

Results

Patients and Baseline Characteristics

A total of 349 patients were screened between February 2023 and December 2024. After applying the exclusion criteria, 260 patients were included in the initial analysis (Figure 1). Among these 260 patients, 140 patients initially received sedation with cipepofol for more than 48 hours. Most demographic data (age, sex and BMI), hemodynamic data (lactic acid, HR, SBP, DBP and MAP) and baseline characteristics (eg, oxygenation index, types of infected pathogens and the adequacy of initial antibiotic regimens) were comparable between the cipepofol and conventional groups (all P > 0.05), although the conventional group had a higher APACHE II score (median: 22.00 [IQR: 16.00–27.00] vs 17.00 [IQR: 12.75–21.00], P < 0.001) than the cipepofol group. After conducting 1:1 PSM using age and APACHE II scores as the matching variables, 62 patients were excluded from the analysis due to failure to find suitable matches, resulting in a final matched cohort of 198 patients (99 in the cipepofol group and 99 in the conventional group). Accordingly, APACHE II scores were well balanced between the cipepofol and conventional groups (median: 17.00 [IQR:13.00–23.00] vs 19.00 [IQR: 15.00–24.00], P = 0.124, SMD = 0.199; SMD < 0.25). The proportion of patients receiving CRRT in the conventional group was significantly higher than that in the cipepofol group before PSM (29.2% vs 15.7%, P = 0.014), and well balance between groups was achieved after PSM (17.2% vs 21.2%, P = 0.588, SMD = 0.103; SMD < 0.25). The proportion of patients receiving ECMO was comparable between the cipepofol and conventional groups before (0.7% vs 5.0%, P = 0.081) and after PSM (1.0% vs 3.0%, P = 0.613). After PSM, the proportions of patients with CAP (56.6% vs 54.5%) and HAP (43.4% vs 45.5%) were comparable between the cipepofol and conventional groups (all P > 0.05), among which there were 2 cases of VAP in the cipepofol group (Table 1). Baseline data relating to organ function and metabolic parameters are summarized in Supplementary Table S1. No significant differences were observed in baseline blood lipids, liver, renal or cardiac functions between the two groups after PSM.

Figure 1.

Study flow diagram.

Sedation Regimens

The sedation/analgesia duration were comparable between the cipepofol and conventional groups before and after PSM (P > 0.05) (Table 1). After PSM, 99 patients initially received more than 48 hours of cipepofol sedation, with a median cipepofol dosage of 293.76 (IQR: 218.40–406.08) mg per day (Supplementary Table S2). In addition to cipepofol sedation, 81 (81.8%), 5 (5.1%) and 3 (3.0%) patients in the cipepofol group received sedation with midazolam, dexmedetomidine, and remimazolam, respectively. The corresponding median daily dosages were 80.40 (IQR: 59.04–125.28), 0.48 (IQR: 0.24–0.72) and 384.00 (range: 55.68–864.00) mg, respectively.

Before PSM, among the 120 patients in the conventional group, 50 (41.7%), 68 (56.7%) and 2 (1.7%) patients initially received sedation with propofol, midazolam or dexmedetomidine for a duration of more than 48 hours, respectively. Following PSM, 21 patients (17.5%) were excluded from the analysis, resulting in a final matched cohort consisting of 50 (50.5%), 47 (47.5%) and 2 (2.0%) patients who had initially received midazolam-, propofol- and dexmedetomidine-based sedation regimens, respectively. Overall, there were 88 (88.9%), 47 (47.5%) and 9 (9.1%) patients in the conventional group who received sedation with midazolam, propofol and dexmedetomidine, with a median daily dosage of 112.56 (IQR: 82.32–159.12), 1054.08 (IQR: 300.24–1561.44) and 0.48 (IQR: 0.24–0.48) mg, respectively (Supplementary Table S2). In addition, there was 1 patient (1.0%) sedated with cipepofol at a dose of 20.88 mg per day, while 2 patients (2.0%) were sedated with remimazolam at daily doses of 112.80 and 254.16 mg, respectively.

Efficacy Endpoints

Primary Outcome

After PSM, there was no significant difference in the duration of MV between the cipepofol and conventional groups (median [IQR]: 218.23 [145.80–380.30] vs 227.25 [140.64–356.26] hours, P = 0.740, Table 2). Subgroup analysis also exhibited no significant difference in the MV duration between the propofol- or midazolam-based regimens compared with cipepofol sedation, across all populations as well as in HAP and CAP patients (Supplementary Tables S3 and S4).

Table 2.

Comparisons of Efficacy Endpoints for the Enrolled Patients Before and After Propensity Score Matching

Variable	Before PSM (N = 260)			After PSM (N = 198)
	Cipepofol Group (n = 140)	Conventional Group (n = 120)	P-Value	Cipepofol Group (n = 99)	Conventional Group (n = 99)	P-Value
Primary endpoint
MV duration (hours)	219.57 [150.50–380.75]	225.83 [143.16–360.29]	0.652	218.23 [145.80–380.30]	227.25 [140.64–356.26]	0.740
Secondary endpoints
Sedation success rate, n/N (%)	105/126 (83.3)	95/107 (88.8)	0.262	74/89 (83.1)	78/87 (89.7)	0.299
Length of ICU stay (days)	9.99 [7.79–16.91]	12.71 [8.16–18.07]	0.124	10.08 [7.13–16.94]	12.09 [8.09–17.07]	0.282
28-day VFD (days)	11.14 [0.00–19.25]	10.15 [0.00–19.48]	0.975	12.08 [0.00–19.52]	8.22 [0.00–19.26]	0.536
28-day mortality, n/N (%)	61/140 (43.6)	48/120 (40.0)	0.615	39/99 (39.4)	41/99 (41.4)	0.885

Notes: Data are presented as the median [interquartile range] or number (percentage). N indicated numbers of patients who had the available data.

Abbreviations: ICU, intensive care unit; MV, mechanical ventilation; PSM, propensity score matching; VFD, Ventilator-Free Days.

Secondary Outcomes

After PSM, no significant differences were found between groups in any of the secondary outcomes (Table 2). Specifically, the sedation success rate was 83.1% (74/89) in the cipepofol group and 89.7% (78/87) in the conventional group (P = 0.299). The median (IQR) length of ICU stay was 10.08 (7.13–16.94) days in the cipepofol group vs 12.09 (8.09–17.07) days in the conventional group (P = 0.282). The median (IQR) 28-day VFD was 12.08 [0.00–19.52] days in the cipepofol group vs 8.22 [0.00–19.26] days in the conventional group (P = 0.536). The 28-day mortality rate was 39.4% (39/99) in the cipepofol group and 41.4% (41/99) in the conventional group (P = 0.885). Compared to the cipepofol regimen, subgroup analysis revealed no significant differences in 28-day VFD and 28-day mortality between patients who received initial sedation with propofol or midazolam for more than 48 hours, across all populations as well as in HAP and CAP patients (Supplementary Tables S3 and S4).

Safety Endpoints

After PSM, the cipepofol group had lower incidences of hypotension within the first hour (11.1% vs 27.3%, P = 0.007) and overall bradycardia events (19.1% vs 36.1%, P = 0.014) than the conventional group. No significant between-group differences were found for the total dosage and duration of vasoactive drug (norepinephrine) administration (Table 3).

Table 3.

Comparisons of Safety Endpoints for the Enrolled Patients Before and After Propensity Score Matching

Variable	Before PSM (N = 260)			After PSM (N = 198)
	Cipepofol Group (n = 140)	Conventional Group (n = 120)	P-Value	Cipepofol Group (n = 99)	Conventional Group (n = 99)	P-Value
Hypotension within the first hour, n (%)	15 (10.7)	31 (25.8)	0.034*	11 (11.1)	27 (27.3)	0.007*
Bradycardia within the first hour, n (%)	0	2 (1.7)	0.211	0 (0.0)	2 (2.1)	0.491
Overall hypotension, n (%)	84 (62.2)	75 (65.2)	0.720	58 (61.7)	59 (60.8)	1.000
Overall bradycardia, n (%)	30 (22.2)	43 (37.4)	0.012*	18 (19.1)	35 (36.1)	0.014*
Duration of vasoactive drug use (hours)	127.00 [75.00–196.00]	131.00 [72.00–224.50]	0.744	122.22 [73.25–202.75]	122.00 [65.50–220.70]	0.924
Dosage of vasoactive drugs (mg)	114.0 [42.00–252.00]	174.00[48.00–353.00]	0.034*	102.00 [36.00–247.50]	162.00 [36.00–311.00]	0.108

Notes: Data are presented as the median [interquartile range] or number (percentage). * P < 0.05 indicates statistical significance.

Abbreviation: PSM, propensity score matching.

Subgroup analysis further revealed that, compared with the cipepofol regimen, the initial sedation of propofol for more than 48 hours was significantly associated with an increased risk of hypotension within the first hour (OR = 3.394, 95% CI = 1.400–8.226, P = 0.007) and overall bradycardia events (OR = 2.621, 95% CI = 1.200–5.722, P = 0.016) (Supplementary Table S3). Herein, patients undergoing initial sedation with midazolam for more than 48 hours exhibited a significantly increased risk of hypotension within the first hour (OR = 2.667, 95% CI: 1.098–6.475, P = 0.030). In contrast, the association between initially midazolam regimen for more than 48 hours and an increased risk of overall bradycardia events did not achieve statistical significance (OR = 2.175, 95% CI = 0.998–4.739, P = 0.050). Notably, HAP patients who initially received sedation with propofol or midazolam for more than 48 hours exhibited a statistically significantly increased risk of hypotension within the first hour (propofol: OR = 6.222, 95% CI = 1.421–27.252, P = 0.015; midazolam: OR = 5.833, 95% CI = 1.339–25.410, P = 0.019) and overall bradycardia events (propofol: OR = 4.000, 95% CI = 1.115–14.354, P = 0.033; midazolam: OR = 4.308, 95% CI = 1.191–15.586, P = 0.026) compared to those receiving the cipepofol regimen. In contrast, no such statistical significance was detected in patients with CAP (Supplementary Table S4).

In addition, there were no significant differences between groups in the incidences of liver/renal function deterioration, hypertriglyceridemia or severe creatine kinase elevation (all P > 0.05) (Table 4; for detailed data, please refer to Supplementary Table S1). Further subgroup analysis also revealed that there was no significant difference in the drug-related elevation of triglycerides in patients who initially received sedation regimen of cipepofol and propofol for more than 48 hours (P = 0.541) (Supplementary Table S3).

Table 4.

Incidences of Abnormal Elevation of Blood Lipids After Sedation and Organ Function Deterioration During Treatment

Variable, n/N (%)	Before PSM (N = 260)			After PSM (N = 198)
	Cipepofol Group (n = 140)	Conventional Group (n = 120)	P-Value	Cipepofol Group (n = 99)	Conventional Group (n = 99)	P-Value
Abnormal elevation of triglycerides after sedation	5/19 (26.3)	3/17 (17.6)	0.823	3/11 (27.3)	3/15 (20.0)	1.000
Hepatic function deterioration during treatment	47/140 (33.6)	33/120 (27.5)	0.356	35/99 (35.4)	29/99 (29.3)	0.447
Renal function deterioration during treatment	5/139 (3.6)	9/120 (7.5)	0.261	3/99 (3.0)	7/98 (7.1)	0.330
Severe creatine kinase elevation during treatment	6/133 (4.5)	6/107 (5.6)	0.942	3/94 (3.2)	5/89 (5.6)	0.671

Notes: Data are presented as n/N (%). N indicated the numbers of patients who had the available data. Abnormal elevation of triglycerides after sedation was defined as a triglyceride level > 400 mg/dL (≈ 4.5 mmol/L), or an increase of ≥ 50% compared with the baseline level with an absolute value > 200 mg/dL (≈ 2.3 mmol/L). Hepatic function deterioration during treatment was defined as a higher hepatic function grade during treatment than at baseline. Renal function deterioration during treatment was defined as a higher renal function grade during treatment than at baseline (see the Data Collection and Baseline Assessments section for the definitions of hepatic and renal function grades). Severe elevation of creatine kinase during treatment was defined as a maximal creatine kinase level during treatment that was ≥ 50% higher than the baseline level with an absolute value > 1000 U/L. Inter-group comparisons were performed using the χ²-test or Fisher’s exact test, as appropriate.

Abbreviation: PSM, propensity score matching.

Exploratory Prognostic Factor Analyses

The results of the univariate and multivariate regression analyses for the sedation regimen are presented in Table 5, while those for other covariates are provided in Supplementary Table S5. The univariate analysis revealed that, compared to conventional regimens, the use of cipepofol was associated with significantly reduced odds of hypotension within the first hour (OR = 0.350, 95% CI = 0.166–0.737, P = 0.006) and overall bradycardia (OR = 0.419, 95% CI = 0.217–0.810, P = 0.010) (Table 5). After multivariate adjustment for age, sex, APACHE II score and COVID-19 infection status, cipepofol use remained independently associated with significantly reduced odds of early hypotension (OR = 0.300, 95% CI = 0.136–0.665, P = 0.003) and overall bradycardia (OR = 0.414, 95% CI = 0.212–0.808, P = 0.010) (Table 5).

Table 5.

Univariate and Multivariate Analysis of the Associations Between Sedation Regimen and Key Outcomes After PSM

Outcome	Univariate Analysis		Multivariate Analysis
	OR (95% CI)	P-Value	Adjusted OR (95% CI)	P-Value
Hypotension within the first hour	0.350 (0.166–0.737)	0.006*	0.300 (0.136–0.665)	0.003*
Hypotension (overall)	1.045 (0.584–1.868)	0.882
Bradycardia (overall)	0.419 (0.217–0.810)	0.010*	0.414 (0.212–0.808)	0.010*
Sedation success rate	0.560 (0.233–1.347)	0.196
28-day mortality rate	0.920 (0.523–1.614)	0.772

Notes: Odds ratios (with 95% CIs) are for cipepofol group vs conventional (reference) group. * P < 0.05 indicates statistical significance.

Abbreviations: CI, confidence interval; OR, odds ratio.

As shown in Supplementary Table S5, multivariate analysis revealed that increasing age was an independent risk factor for hypotension within the first hour of drug administration (OR = 1.052, 95% CI = 1.016–1.089, P = 0.004), whereas female sex was identified as a protective factor (OR = 0.285, 95% CI = 0.097–0.842, P = 0.023). Furthermore, COVID-19 infection was strongly associated with a longer invasive MV duration (regression coefficient = 54.517, 95% CI = 23.831–85.204, P < 0.001), revealed by multivariate analysis.

Discussion

Based on real-world clinical data from the ICU, the present single-center, retrospective cohort study used PSM to construct comparable cohorts of 99 patients each for cipepofol and conventional sedation (propofol-, midazolam- or dexmedetomidine-based sedation) groups, with a primary aim to investigate the effects on the duration of MV for patients with severe pneumonia requiring prolonged sedation. The main findings showed that there was no significant difference in the primary endpoint of MV duration between the two groups.

The duration of MV was selected as the primary endpoint because this study aimed to evaluate the overall clinical effectiveness of prolonged ICU sedation in real-world practice rather than the isolated pharmacodynamic performance of a sedative agent. Prior ICU sedation trials have commonly used the duration of MV, time to successful extubation, or weaning time as primary outcomes, supporting the clinical relevance and acceptability of this endpoint in studies of sedation strategy.34–36 This choice is also biologically plausible, because sedation strategy can affect arousability, ventilator synchrony, delirium and readiness for extubation, all of which may influence the course of MV. The SCEPTER Expert Panel also designates the duration of MV as a key efficacy endpoint in its recommendations for sedation clinical trial design.37 It argues that this endpoint is more feasible than mortality and better captures the comprehensive performance of sedative agents within real-world clinical pathways, as compared to single-dimensional sedation depth scores. However, the duration of MV is a downstream endpoint and not a direct measure of sedative efficacy. It is modulated by a multitude of factors beyond sedation, including patient case-mix, ICU clinical protocols, predefined sedation targets, extubation procedures, and other organ support interventions. Prior comprehensive analyses have demonstrated substantial heterogeneity in MV duration outcomes across studies.38,39 Importantly, major baseline factors that could strongly affect MV duration, including types of infected pathogens, the adequacy of initial antibiotic regimens and organ support modalities, were comparable between groups, reducing concern that the primary comparison was driven by these observed baseline differences. Moreover, MV duration exhibits high data accessibility and straightforward clinical interpretability in real-world retrospective studies. Therefore, the primary endpoint in the present study should be interpreted as a composite clinical outcome reflecting the net consequence of sedation strategy rather than merely serving as a direct surrogate for sedation quality.

Of note, for patients who died while still receiving MV, MV duration in the primary analysis was calculated as the time from intubation (or ICU admission, if intubation occurred before ICU admission) to the time of death. For patients who were successfully weaned before death, MV duration was calculated up to the time of extubation. To address the potential bias that death may artificially shorten the observed MV duration, we added a post hoc sensitivity analysis using 28-day VFDs. Patients who died within 28 days were assigned 0 VFDs, regardless of how many days they had been ventilator-free before death. Patients who remained on MV for the entire 28 days also received 0 VFDs. This approach assigns death the worst possible VFD outcome and reduces the interpretive problem arising from death-truncating the observed duration of MV. The results of the 28-day VFD sensitivity analysis were consistent with the primary analysis, supporting the robustness of our findings. Moreover, there were no significant differences between groups in the sedation success rate, ICU length of stay and 28-day mortality. The sedation efficacy and clinical outcome findings align with previous studies conducted in broader ICU populations.34,40 However, these earlier trials had relatively short observation periods and so may not have fully captured the hemodynamic profiles during prolonged infusion. By focusing on a high-risk cohort of mechanically ventilated patients with severe pneumonia who required prolonged sedation, our study provides the clinical evidence that the pharmacological advantage of cipepofol over conventional sedatives translates into superior hemodynamic stability in a population vulnerable to cardiovascular depression, thus extending the findings from studies in lower-risk, hemodynamically stable settings.

The impact of sedative agents on the circulatory system holds notable clinical significance in patients with severe pneumonia. Such patients frequently present with systemic inflammation, increased vascular permeability and impaired cardiac compliance, making them extremely susceptible to sedation-related hypotension. Although sedative choice was not randomized in the present retrospective cohort study, PSM was performed using age and baseline APACHE II scores, as the matching variables and our findings demonstrate that well balance between groups was achieved after PSM. Moreover, the baseline hemodynamic data including lactic acid, HR, SBP, DBP and MAP were both comparable between groups before and after PSM. The consistency of these findings in the unmatched and matched cohorts suggests that the observed hemodynamic safety signals were not driven solely by measured baseline severity or clinician preference. We compared the incidences of early hypotension (within the first hour of drug administration) between groups to explore whether cipepofol might have an advantage over conventional sedation regimens. In our cohort, patients sedated with cipepofol had significantly lower incidences of early hypotension (11.1% vs 27.3%, P = 0.007) and overall bradycardia (19.1% vs 36.1%, P = 0.014) than those sedated with conventional regimens, which is consistent with previous studies that compared cipepofol with conventional sedatives. A meta-analysis of three randomized controlled trials that involved mechanically ventilated ICU patients reported a numerically lower risk of hypotension with cipepofol vs conventional sedatives (risk ratio [RR] = 0.668, 95% CI = 0.397–1.124), although statistical significance was not reached.28 Another meta-analysis across sedation and anesthesia settings found that the risk of intraoperative hypotension was significantly lower for cipepofol than for propofol (RR = 0.64, 95% CI = 0.52–0.77, P < 0.001).28 Similarly, Yang et al41 reported a significantly lower incidence of hypotension with cipepofol than with propofol in otherwise healthy patients undergoing gastrointestinal endoscopy who had a robust cardiovascular reserve. The magnitude of cipepofol’s hemodynamic advantage was even more pronounced in our cohort of patients with sepsis, multiorgan dysfunction and severely impaired hemodynamic reserve, with the cipepofol group exhibiting a markedly lower incidence of early hypotension than the conventional sedative group. Moreover, the comparable 28-day mortality rate implies that cipepofol’s hemodynamic benefits do not come at a cost of reduced survival.

The observed lower incidences of early hypotension and bradycardia in the cipepofol group compared with the conventional sedative group can be explained by cipepofol’s distinct pharmacological properties. From a mechanistic perspective, the affinity of cipepofol for γ-aminobutyric acid type A receptors is approximately 4–5 times that of propofol, which means that the clinically equivalent dose of cipepofol to achieve the same sedative depth is only 1/4 to 1/5 that of propofol.42 The blood concentration of cipepofol is substantially lower than that of propofol at the equivalent sedative dose, resulting in cipepofol having a milder inhibitory effect on the cardiovascular system, including weaker myocardial depression and less pronounced inhibition of sympathetic vascular tone. This mechanism likely underlies the superior hemodynamic stability observed in the cipepofol group than in the conventional sedative group. The present findings are consistent with those reported in previous studies26 and are particularly relevant for critically ill patients with sepsis, impaired vascular reactivity or a limited cardiovascular reserve.

In addition to its hemodynamic benefits, cipepofol also had comparable organ and metabolic safety profiles to conventional sedatives, presumably because the changes in the measured indicators (lipid parameters, hepatic and renal functions, and the creatine kinase level) were mainly attributable to severe infection and systemic inflammation rather than the choice of sedative drug. Regarding vasoactive drug requirements, the total dosage of norepinephrine used was lower in the cipepofol group than in the conventional sedative group before PSM (P = 0.034), but this difference did not persist after matching. This observation suggests that a higher baseline APACHE II score in the conventional group was the main determinant of vasoactive drug requirement, rather than the choice of sedative itself. However, this finding contrasts with a recent prospective study by He et al,26 which reported a significantly lower norepinephrine requirement in patients sedated with cipepofol than in those sedated with propofol. The apparent discordance likely stems from differences in the comparator (propofol alone vs mixed sedatives used in our conventional group), baseline vasopressor requirement and statistical power. Overall, the results of the present study provide an important clinical reference for sedation management in mechanically ventilated patients with severe pneumonia, especially with regard to balancing sedation outcomes and circulatory safety.

In the present study, propofol-, midazolam- and dexmedetomidine-based regimens were grouped into a conventional sedation cohort, which not only mirrors the real-world clinical context of the ICU but also enhances the external validity of the research. However, the combined analysis may mask the drug-specific effects among sedative agents. For instance, propofol offers advantages in rapid onset, recovery and extubation, but it is associated with well-documented risks of hypotension (up to 42%)43 and hyperlipidemia. Midazolam tends to accumulate due to its lipophilic nature and active metabolites, leading to prolonged sedation and delayed awakening.44 Dexmedetomidine at FDA-approved doses often fails to achieve deep sedation and frequently requires additional agents to reach target RASS goals.45 To address this limitation in the present study, the conventional group was further stratified into propofol- and midazolam-based sedation subgroups, with only two cases of dexmedetomidine-based sedation excluded from the subgroup analysis. Subgroup analysis results revealed no statistically significant differences between cipepofol and propofol or midazolam in terms of MV duration, 28-day VFD and 28-day mortality, which aligned with the direction of the primary analysis conclusions. Moreover, these findings were consistent across both the CAP and HAP subsets. Subgroup analysis further confirmed that cipepofol produced a statistically significant difference in the reduced risk of early hypotension compared to propofol- and midazolam-based regimens, which were highly consistent with the conclusions of the primary analysis. However, a statistically significant reduction in the risk of overall bradycardia events was observed exclusively in comparison with the propofol-based regimen; in contrast, the comparison with the midazolam-based regimen yielded only marginal statistical significance (P = 0.050). Notably, the significant differences of reduced risks of early hypotension and overall bradycardia events persisted in HAP patients but did not reach statistical significance in CAP patients. This observed inter-subgroup discrepancy may be attributed to the inherent distinctions in clinical characteristics between patients with CAP and HAP. Patients with HAP, particularly those with VAP, often experience prolonged ICU lengths of stay, more frequent exposure to vasoactive agents, and higher proportions of MDR infections. Compared with CAP patients, they exhibit poorer cardiopulmonary reserve and autonomic nervous system regulatory capacity, rendering them more sensitive to sedative-induced vasodilation and myocardial depression, thus facilitating the detection of cipepofol’s hemodynamic advantages. A large-scale analysis by Woodhead et al using the ICNARC database demonstrated that pneumonia patients admitted to the ICU ≥ 7 days after hospital admission had significantly elevated mortality,46 suggesting that HAP patients generally present with more complex clinical conditions and more severe concurrent organ dysfunction, which provides a larger effect size to enable the detection of intergroup differences. In contrast, CAP patients typically require ICU admission earlier in their hospital course; despite high acute severity, they have relatively sufficient time for organ function compensation, potentially enhancing their tolerance to acute hemodynamic perturbations induced by sedatives. Consequently, intergroup differences are reduced to a magnitude below the threshold of statistical detectability. Therefore, cipepofol may be a favorable option for achieving optimal sedation with good circulatory protection in patients with a fragile circulation (eg, due to septic shock or underlying cardiovascular disease) or those requiring prolonged sedation. However, our findings are hypothesis-generating and require confirmation in larger prospective studies.

The major strengths of the present study included the use of PSM to reduce confounding bias from differences in baseline disease severity and its focus on a high-risk population of mechanically ventilated patients with severe pneumonia who require prolonged sedation. However, the present study has a few limitations. First, this was a single-center retrospective study, and patient selection and treatment regimens may have been influenced by the clinical practice preferences of the participating medical center, thus the generalizability of the findings to other healthcare settings should be interpreted with caution. Second, a residual imbalance in renal function classification persisted (SMD = 0.266, slightly above the conventional threshold of 0.25) after PSM. However, the proportion of patients with moderate-to-severe renal dysfunction (Ccr stage 3–4) was comparable between the two groups (24.2% vs 23.2%), and no significant differences in baseline creatinine clearance or changes in creatinine clearance were observed after PSM (Supplementary Table S1), indicating this minor imbalance did not notably affect creatinine metabolism or core outcome comparisons. In addition, although PSM balanced measured baseline characteristics, including APACHE II, sedation strategies were not randomly allocated. Clinician preferences and other unmeasured confounders may still have biased treatment assignment. Accordingly, the hemodynamic safety outcomes should be interpreted as associations rather than causal relationships, and the potential influence of residual confounding cannot be ruled out. Third, we acknowledge the inherent heterogeneity associated with pooling propofol, midazolam, and dexmedetomidine into a single conventional sedation group. Thus, we incorporated a post hoc subgroup analysis stratified by initial sedation regimens, partially mitigated the potential masking of drug-specific effects that may have arisen from pooling propofol, midazolam and dexmedetomidine in the primary analysis. However, the reduced sample size in subgroup analyses compromises statistical power, warranting cautious interpretation of the results. Further multicenter, prospective controlled studies should be carried out to compare cipepofol head-to-head with existing commonly used ICU sedatives (either as monotherapy or in combination regimens), clarify the differences in efficacy and safety between cipepofol and various conventional drugs, and provide a more evidence-based foundation for personalized sedation strategies.

Conclusion

The present real-world retrospective analysis found that cipepofol was not associated with a statistically significant reduction in MV duration compared with conventional sedation regimens among mechanically ventilated patients with severe pneumonia requiring prolonged sedation. Subgroup analyses were consistent with the primary analysis. Nevertheless, cipepofol was associated with a lower risk of early hypotension and overall bradycardia events than propofol- and midazolam-based conventional regimens across the overall population and HAP patients. These hemodynamic findings suggested a favorable profile of cipepofol in this vulnerable population. Given the study’s retrospective, single-center design and the heterogeneity of the conventional group, the results should be interpreted cautiously. Confirmation in large prospective multicenter studies is needed to establish unequivocal clinical recommendations regarding the choice of sedative regimen in this critically ill patient group.

Funding Statement

This study was supported by the Medical and Health Research and Development Fund Project - Clinical and Basic Research Special Project (Grant No. S182), National Natural Science Foundation of China (82100099), and National Clinical Research Center for Geriatric Disorders (2023LNJJ08).

Data Sharing Statement

The datasets used or analyzed during the current study are available from the corresponding author Dr. Juan Jiang upon reasonable request via email: drjiangjuan0923@163.com.

Author Contributions

All authors made a significant contribution to the research 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

The authors declare no competing interests.