Effect of remifentanil on parturients and newborns undergoing cesarean section with general anesthesia: a systematic review and a meta-analysis

Introduction

Intrathecal anesthesia is preferred for cesarean sections due to its rapid effectiveness, but general anesthesia is needed if intrathecal is unsuitable or fails. The use of opioids, particularly with general anesthesia, is debated as they can depress newborn respiration, with studies indicating maternal opioid use may cause respiratory issues in newborns (1, 2). Effective pain management during general anesthesia is vital to stabilize maternal heart rate and blood pressure, especially in those with cardiovascular problems, making research on opioid use in cesarean sections under general anesthesia clinically significant.

Remifentanil, a fast-acting synthetic opioid, is increasingly used in cesarean anesthesia for its rapid metabolism and stable hemodynamic effects, which aid in quick recovery and reduce neonatal respiratory depression risks (3). While generally showing minimal impact on neonatal respiratory outcomes and Apgar scores, some studies highlight potential fetal respiratory risks, especially at high doses (4). A systematic review found remifentanil does not significantly affect Apgar scores, but safety concerns persist (5). Recent randomized controlled trials, including those comparing dexmedetomidine, have been published since earlier meta-analyses. To overcome individual study limitations and offer fresh insights, we conducted this systematic review and meta-analysis. It assesses the effects of remifentanil versus placebo or other anesthetics on maternal and neonatal outcomes during general anesthesia induction for cesarean delivery.

Method

Search strategies

This study was conducted in accordance with the guidelines specified in the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement and registered it in the PROSPERO network (registration number: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261342473, CRD420261342473, 16 March 2026). A thorough literature search was executed across four databases—PubMed, Web of Science, Embase, and Cochrane—covering publications up to December 31, 2025. The search strategy utilized the keywords “remifentanil” combined with “c-section,” “caesarean section,” or “caesarean delivery.” The inclusion criteria were restricted to studies published in the English language.

Study selection

The initial screening involved assessing the titles and abstracts, followed by a detailed review of the full texts. Studies included in the meta-analysis had to: (1) be randomized controlled trials (RCTs); (2) examine intravenous remifentanil for general anesthesia induction in cesarean sections with a control group; (3) provide intraoperative and postoperative data for neonates and mothers. Exclusion criteria were: (1) non-RCT studies like reviews, case reports, or observational studies; (2) maternal prenatal exams showing fetal distress or congenital issues; (3) lack of access to detailed study information.

Data extraction and quality assessment

Researchers ZJ and CY independently selected studies and extracted data using a standardized form to ensure consistency. They collected information on the first author, publication year, country, patient characteristics, interventions, sample sizes, and outcomes. Key outcomes included Apgar scores, neonatal ventilation and intubation rates, umbilical blood gas analysis, and maternal MAP and HR during surgery. They assessed inter-researcher agreement and resolved discrepancies through discussion, consulting a senior author (HY) if needed. The risk of bias in each RCT was evaluated using the Cochrane Collaboration's tool. Each domain was judged as low risk, unclear risk, or high risk across the following domains: 1. Random sequence generation (selection bias) 2. Allocation concealment (selection bias) 3. Blinding of participants and personnel (performance bias) 4. Blinding of outcome assessment (detection bias) 5. Incomplete outcome data (attrition bias) 6. Selective reporting (reporting bias) 7. Other potential bias.

Statistical analysis

The effect size was measured using odds ratios for dichotomous variables and mean ± standard deviation for continuous ones. The study included both placebo and dexmedetomidine-fentanyl groups, combining them to enhance sample size, statistical power, and clinical relevance, despite increased heterogeneity. To avoid broad conclusions, subgroup analyses compared remifentanil to placebo and dexmedetomidine, providing an overall effect trend while minimizing confounding bias. Heterogeneity was assessed using the Q-test and I² statistic, with significant heterogeneity indicated by a P-value < 0.05 or I² > 50%, prompting the use of a random-effects model; otherwise, a fixed-effects model was used. Sensitivity analysis was performed using the leave-one-out method to test the robustness of pooled outcomes. Each eligible RCT was excluded one at a time, and we recalculated the weighted mean difference, 95% CI, I² and Q-test P value of primary maternal and neonatal indicators after each exclusion. Results were considered stable if the direction of effect estimates, statistical significance and heterogeneity level did not change remarkably after removing any single study. Analyses were performed with Review Manager 5.4.1 and STATA 19.0.

Outcomes

Literature search

A total of 267 studies were initially identified (47 from PubMed, 85 from Cochrane, 109 from Web of Science, and 26 from Embase). After screening titles and abstracts, most were excluded for reasons like lack of randomized controlled trials, non-compliance with intervention criteria, or duplication. A full-text evaluation of the remaining 19 studies found 11 eligible, involving 323 cases and 322 controls (3, 6–15). The screening process is shown in Figure 1.

Studies' characteristics

Table 1 provides details of the 11 studies analyzed: three from China, two from Egypt, and the rest from South Korea, the Czech Republic, Turkey, Iran, Italy, and France. Inclusion criteria differed, with three studies focusing on mothers with severe preeclampsia and seven on ASA grades I-II mothers. Five studies compared remifentanil with a placebo, four with dexmedetomidine, one concurrently compared dexmedetomidine with a placebo, and one compared remifentanil with fentanyl. Some studies lacked clear randomization and blinding procedures.

Table 1. Study characteristics of included studies.

Neonatal outcome

We gathered data on neonatal outcomes, such as Apgar scores at 1 and 5 min, ventilation and intubation needs, and umbilical blood gas analyses, as illustrated in Figures 2, 3. The control group, labeled “Control” in the forest plots, included placebo, dexmedetomidine, and fentanyl. The 1-minute Apgar score was significantly lower in the remifentanil group compared to the control (P < 0.05, WMD: −0.51, 95% CI: −0.84–−0.19), but there was no significant difference at 5 min. No significant differences were observed in neonatal respiratory support or blood gas analyses between the groups (P > 0.05).

Maternal outcome

Maternal indicators, including mean arterial pressure (MAP) and heart rate (HR), were measured at five key stages: operating room admission, anesthesia induction, tracheal intubation, uterine incision, and fetal delivery. As shown in Figures 4, 5, the remifentanil group had significantly lower MAP during intubation compared to the control group (P < 0.05), with a weighted mean difference of −11.37 and a 95% confidence interval of −21.46–−2.00. No significant MAP or HR differences were found at other stages (P > 0.05).

Subgroup analysis

To compare outcomes between remifentanil and placebo, as well as other anesthetics like dexmedetomidine and fentanyl, we conducted subgroup analyses. Excluding dexmedetomidine, remifentanil showed a significant difference in the 1-minute Apgar score compared to placebo (P < 0.05, WMD: −0.52, 95% CI: −1.02, −0.03, Figure 6). Excluding placebo, remifentanil differed significantly from dexmedetomidine (P < 0.05, WMD: −0.59, 95% CI: −1.01, −0.16, Figure 7). No significant differences were found in the 5-minute Apgar score, mean arterial pressure, heart rate, neonatal respiratory support, or blood gas analysis. Additionally, preeclampsia status did not affect outcomes. Due to variations in remifentanil dosage and administration across studies, only 11 studies were included, limiting the statistical power of a meta-regression. We categorized the bolus doses as low (≤0.5 μg/kg), medium (0.5–1 μg/kg), and high (>1 μg/kg). Figure 8 shows subgroup analysis results based on these levels. A significant difference in 1-minute Apgar scores was found in the medium-level group (P < 0.05, WMD: −0.66, 95% CI: −0.95, −0.37), but not in the low or high-level groups.

Risk of bias

Figure 9 presents domain-level risk of bias assessments. Most studies showed a low risk of bias for incomplete data and selective reporting, as all outcomes were fully reported. Adequate randomization methods were used in most studies, indicating low risk for random sequence generation. However, concerns were noted in allocation concealment and blinding due to insufficient details, leading to unclear risk ratings. No study was rated at high risk across all domains. After sequential exclusion of each trial, the pooled WMD of 1-minute Apgar score fluctuated between −0.30 and −0.54, and the pooled WMD of intubation MAP ranged from −6.95 to −13.17; all re-calculated 95% CIs did not cross the null value, and heterogeneity I² values only had minor fluctuations without crossing the 50% threshold. No single study was found to be the dominant source of overall heterogeneity, which indicated that our main meta-analysis conclusions were robust. Funnel plots and Begg's and Egger's tests indicated no publication bias (P > 0.05), but results should be cautiously interpreted due to the small sample size (n = 11).

Discussion

This meta-analysis of 11 randomized controlled trials assessed the effects of remifentanil during general anesthesia for cesarean delivery on maternal and neonatal outcomes. Remifentanil effectively reduced maternal mean arterial pressure increases due to tracheal intubation and caused a minor, statistically significant decrease in neonatal 1-minute Apgar scores without clinical harm. There were no differences between groups in 5-minute Apgar scores, neonatal ventilation rates, or umbilical cord blood gas measures. The slight early Apgar score drop did not affect neonatal respiratory function or cause acidosis, aligning with remifentanil's rapid clearance after placental transfer. Further subgroup analyses by control type confirmed that remifentanil lowered the 1-minute Apgar score compared to placebo and dexmedetomidine, but this effect disappeared by the 5-minute mark across all groups. While Apgar scoring is subjective, umbilical blood gas provides an objective measure of perinatal asphyxia. Our analysis of intact cord blood parameters supports that the slight drop in the 1-minute Apgar score does not indicate actual neonatal physiological depression. The Apgar score is often used to assess newborns quickly, but its reliability in predicting asphyxia and complications varies (16).A study found moderate agreement between the Apgar score and umbilical cord blood pH (κ = 0.44), indicating that the Apgar score may not fully replace biochemical markers (17). Blood gas analysis from the umbilical artery and vein provides more precise information on physiological status, especially in assessing neonatal acidosis and asphyxia severity (18, 19). Previous research highlights umbilical artery pH and lactate levels as key indicators for short-term neonatal prognosis (20). Remifentanil, a fast-metabolizing opioid, did not cause abnormal changes, suggesting it doesn't lead to intrauterine hypoxia-acidosis. However, further studies are needed to confirm its long-term safety due to neonatal physiological immaturity.

To reduce fetal exposure to anesthetics and aspiration risk after maternal unconsciousness, rapid sequence induction is often used in general anesthesia for cesarean sections. Opioids are typically avoided before fetal delivery to prevent newborn respiratory depression, but this can lead to insufficient anesthesia during intubation, causing maternal hemodynamic changes and increased cardiovascular and cerebrovascular risks. This is especially concerning for women with preeclampsia, as it may trigger eclampsia and increase the risk of fetal hypoxia and brain injury (21). Stroke risk is also higher in preeclampsia patients undergoing general anesthesia for cesarean sections (22). Maintaining stable hemodynamics is crucial for placental perfusion and fetal health (23). Our study noted that parturients given remifentanil had significantly lower mean arterial pressure during intubation than the control group, though heart rates were similar. This aligns with a prior study on remifentanil in severely preeclamptic patients, showing it effectively reduces hypertension from intubation without affecting heart rate (24). However, more data is needed to fully understand the clinical significance of the MAP reduction.

This meta-analysis enrolled RCTs with three distinct control arms (placebo, dexmedetomidine and fentanyl), grouped according to whether remifentanil was used for anesthesia induction. Despite heterogeneous comparator regimens, core intervention parameters were consistent across trials, supporting pooled analysis of remifentanil's neonatal safety. Earlier meta-analyses have completed separate pairwise comparisons for individual controls (5, 25), so repeated isolated subgroup pooling would yield no novel evidence. Integrating multiple routinely-used clinical comparators and applying predefined subgroup stratification enables evaluation of the overall efficacy and safety of remifentanil against non-remifentanil induction strategies, matching real-world obstetric anesthesia practice. Although combining varied controls brings inherent clinical heterogeneity, preplanned subgroup analyses by comparator maintain reliable result interpretation (26). Network meta-analysis could not be performed due to the limited number of included trials and lack of direct head-to-head comparisons among active control agents.

Initially, subgroup analyses based on the control group demonstrated that the 1-minute Apgar score remained significantly different between remifentanil and both placebo and dexmedetomidine controls. In contrast, subgroup analyses for the 5-minute Apgar score revealed no significant differences. These findings suggest that remifentanil significantly impacts the 1-minute Apgar score when compared to placebo or dexmedetomidine, although this effect may be transient (7, 27, 28). Regarding comparisons involving other anesthetic agents in general anesthesia for cesarean sections, the absence of experimental data due to limited clinical application precluded further discussion of these findings. The subgroup analyses of other outcome measures in this study revealed no significant differences, aligning with previous clinical studies that suggest remifentanil does not substantially impact fetal respiratory function or blood gas levels at birth. The observed decrease in Apgar scores is transient and lacks clinical significance (13). A prior meta-analysis examining remifentanil use in general anesthesia for cesarean sections found no suppression of neonatal respiratory function, despite a transient reduction in Apgar scores at 1 and 5 min (29). In the present study, the 5-minute Apgar scores did not differ significantly; however, when considered alongside other neonatal outcome indicators, this difference did not notably influence clinical outcomes post-birth. Further clinical research is warranted to assess long-term prognoses.

The impact of remifentanil dosage on neonatal Apgar scores remains inconclusive due to insufficient clinical trial data. Typically, remifentanil is administered via single injection or 10-minute continuous infusion during cesarean deliveries, with doses ranging from 0.5–1 μg/kg, considered safe (30). This study categorized doses into low (below 0.5 μg/kg), medium (0.5–1 μg/kg), and high (above 1 μg/kg). Results indicated that medium doses significantly lowered 1-minute Apgar scores compared to the control group, while low doses had no significant effect. High doses showed no significant difference, possibly due to the use of continuous infusion in the studies analyzed. No significant differences were observed in 5-minute Apgar scores. Analysis indicates that variations in remifentanil dosing can influence the 1-minute Apgar score of newborns, though the doses remain within the safe range. Further research is necessary to examine the effects of single-dose versus continuous infusion and specific dosages on Apgar scores.

Several limitations of this meta-analysis should be acknowledged. Firstly, the research was restricted to clinical studies sourced from only four databases and restricted to publications in the English language, thereby excluding studies from other databases and those published in other languages. Secondly, there were variations in the dosage and administration of remifentanil among the included studies. Despite conducting a sensitivity analysis on remifentanil dosage, our method for differentiating dosages is still subjective and lacks strict administration guidelines. Further clinical research is warranted to explore this issue. Thirdly, the scarcity of non-placebo-controlled studies precluded the possibility of conducting a network meta-analysis to compare the effects of various anesthetic agents. Consequently, subgroup analyses were conducted solely to compare the effects of dexmedetomidine, a non-placebo control. While this approach expanded the study population and enriched the evidence base, offering a more accurate reflection of clinical outcomes, it also compromised the homogeneity of the included studies. Performing a network meta-analysis after more relevant clinical studies become available is the most practical strategy to reduce such heterogeneity. Moreover, the choice of neonatal outcome indicators is a key limitation of this meta-analysis. While umbilical cord blood gas analysis is a more precise marker for neonatal asphyxia, we used Apgar scores as the main endpoint due to their practical clinical value. Apgar scores allow immediate bedside assessment without invasive procedures, making them a standard tool in obstetrics. All 11 RCTs reported Apgar data, but many lacked umbilical blood gas parameters, so using Apgar scores increased the sample size and reliability of estimates. However, Apgar scoring is subjective and only moderately correlates with umbilical arterial pH, which weakens the evidence for evaluating neonatal safety. Additionally, the reliability of publication bias assessment was limited by the small number of included studies (n = 11), as Begg's and Egger's tests have low statistical power when fewer than 10–15 studies are available. Thus, conclusions regarding publication bias should be interpreted cautiously. Lastly, the research incorporated a limited number of high-quality studies. Future investigations necessitate more detailed and comprehensive high-quality studies with rigorous research protocols.

Conclusion

Remifentanil for cesarean general anesthesia provides favorable maternal hemodynamic stabilization during intubation. The mild reduction in 1-minute Apgar score is statistically significant but lacks clinical importance given unchanged late Apgar score, intact neonatal respiratory function and normal umbilical blood gas values. Further large-scale standardized RCTs with unified drug dosage and long-term neonatal follow-up are required to supplement existing evidence.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fanes.2026.1854181/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Tao Xu, International Peace Maternity and Child Health Hospital, China

Reviewed by: Rui Ma, Shanghai Jiao Tong University School of Medicine, China

Suzana Sobot, University Clinical Center of Republika Srpska, Bosnia and Herzegovina