Comparison of different interventions for the reduction of labor pain: A systematic review and network meta-analysis

Abstract

Objective:

This systematic review and network meta-analysis were performed to compare different interventions for the reduction of labor pain.

Methods:

PubMed, Embase, Cochrane Library, Web of Science and ScienceDirect databases were searched for the randomized controlled trials (RCTs) meeting prespecified inclusion criteria up to January, 2023. Interventions including electrical acupoint stimulation (TEAS), epidural analgesia (EA) and control treatments. The primary outcomes included pain scores, failure rate of natural delivery, adverse events and Apgar scores. The methodological quality was assessed by the Cochrane risk of bias tool. Meta-analysis was performed by R software with gemtc package. Surfaces under the cumulative ranking curves (SUCRA) were used to rank the intervention.

Results:

Twelve studies met the inclusion criteria and were included in the network meta-analysis. TEAS (WMD −3.1, 95% CrI −3.8, −2.5) and EA (WMD −2.1, 95% CrI −2.8, −1.3) was more effective than the control in decreasing VAS. TEAS ranked first (SUCRA, 90.9%), EA ranked second (SUCRA, 74.0%) and control ranked last (SUCRA, 35.0%) for reducing VAS. For patients with labor pain, with respect to the most effective treatment for reducing failure rate of natural delivery, TEAS ranked first (SUCRA, 96.6%), EA ranked second (SUCRA, 50.4%) and control ranked last (SUCRA, 3.0%). With regard to the Apgar scores, there was high probability that TEAS ranked first (SUCRA, 80.7%), compared to control (SUCRA, 41.4%) and EA (SUCRA 27.9%). With regard to the adverse events, there was high probability that TEAS ranked first (SUCRA, 99.9%), compared to control (SUCRA, 33.2%) and EA (SUCRA 17.6%).

Conclusion:

TEAS has the potential to serve as a viable alternative for women in labor, offering a simple, noninvasive, and non-pharmacological intervention that surpasses EA in terms of both analgesic effectiveness and safety for both mothers and neonates.

1. Introduction

Childbirth pain is regarded as one of the most intense and demanding experiences for many women.^[1,2] This demanding process induces significant physiological changes in mothers,^[2,3] leading to increased oxygen consumption, blood pressure, cardiac output, and impacting their psychodynamic behavior.^[3,4] Furthermore, uncontrolled pain during childbirth is linked to higher rates of cesarean sections and adverse outcomes for newborns, including an increased incidence of fetal hypoxemia.^[5]

Epidural analgesia (EA) involves the combination and injection of local anesthetics and opioid analgesics into the epidural space of the spinal cord, targeting the spinal nerve roots to block sensory nerves and reduce endogenous catecholamines.^[6] EA is widely recognized as the most effective and commonly used intervention for alleviating childbirth pain, often resulting in high maternal satisfaction.^[7] However, it may not be suitable for patients with spinal deformities, hemodynamic instability, or allergies to anesthetics, and some mothers may decline this treatment.^[8] Furthermore, concerns persist about potential side effects of EA, including hypotension, motor block, fever, itching, and urinary retention.^[9,10] As a result, identifying effective, safe, and non-pharmacological alternative therapies becomes of utmost importance.

Acupuncture is gaining popularity worldwide among both doctors and patients.^[11,12] A 2020 systematic review^[13] revealed that compared to sham acupuncture, genuine acupuncture may enhance the satisfaction of parturients with pain relief, but its effectiveness is closely linked to the acupuncturist skill. One significant challenge for doctors is to maintain continuous manual needle manipulation throughout the entire labor process to sustain the parturient “Qi” sensation.

Transcutaneous electrical nerve stimulation (TENS) is a recognized noninvasive and non-anesthetic analgesic treatment successfully employed in managing labor and delivery pain.^[14,15] Building on traditional acupuncture principles, transcutaneous electrical acupoint stimulation (TEAS), also known as non-needle acupuncture or acupuncture-like TENS, involves placing electrode patches or pens with regulating stimulation patterns on specific acupoints, resulting in twice the improvement in efficacy compared to acupoint therapy and TENS alone. The analgesic effect of TEAS is believed to be mediated through various mechanisms, including the inhibition of endogenous pain-causing substances, the promotion of opioid-like peptides production, and interference with the normal transduction of mitogen-activated protein kinase signaling pathways.^[16] A pilot study suggested that TEAS could serve as a supplementary intervention for managing post-abortion pain.^[17] Moreover, TEAS has demonstrated significant reductions in visual analogue scale (VAS) scores for postoperative pain treatment.^[18] Although TENS has been proposed as an alternative method for relieving labor pain, a systematic review investigating the effectiveness and safety of TEAS in pain relief has not been conducted. Nevertheless, several randomized controlled trials (RCTs) have indicated that TEAS effectively alleviates labor pain.^[19–22]

The objective of this network meta-analysis was to assess different treatments including, TEAS, EA and control groups for reducing labor pain. We hypothesize that TEAS outperformed EA for pain mitigation during labor without increasing adverse events.

2. Methods

2.1. Search strategy

RCTs that compared TEAS, EA, and placebo for reducing labor pain were searched in electronic databases including PubMed, EMBASE, Web of Science, and Cochrane library All relevant RCTs will be collected from the inception of each database to June 2023. The search strategy included the use of MESH or Emtree terms and their free text words. Boolean operators AND and OR were used to combine search terms. Search terms used were: “Acupuncture Points,” “Electric Stimulation,” “Transcutaneous Acupoint Electrical Stimulation,” “Anesthesia, Epidural,” “Parturition,” “Delivery, Obstetric”[Mesh] AND “Obstetric Labor.” Ongoing trials or unpublished studies were also searched in ClinicalTrials.gov. Two reviewers (Mengxian Yu and Haiyan Qian) independently read titles and abstracts for preliminary screening and reviewed full-text eligibility. Any disagreements were resolved by discussion with a third party or by consultation with the investigator (Miao Gan). Ethics approval was not required for this systematic review and network meta-analysis since no patient contact took place.

2.2. Inclusion criteria and exclusion criteria

Studies were included if the following inclusion criteria were met: Population (P): healthy laboring parturients planning natural birth; Interventions (I): TEAS and EA placebo for reducing labor pain; Control (C): Sham TEAS group that received very low electrical stimulation (<5 mA), saline placebo, blank control or routine care; Outcomes (O) VAS; failure to progress natural delivery, adverse events and neonatal Apgar score; Study (S): All relevant randomized controlled trials were included. Exclusion criteria were as follows: (1) case reports and comments; (2) studies with insufficient data; (3) reviews or meta-analyses studies; (4) studies with only case group; (5) no follow-up after discharge.

2.3. Data extraction

Two authors (MY and HQ) independently screen literature, extract data, and cross-check them. In case of disagreements, a third reviewer is consulted to assist with the judgment, and any missing information is supplemented by contacting the author as much as possible. When screening literature, the title and abstract are read first, and after excluding obviously irrelevant literature, the full text is further read to determine whether it should be included. The main content of data extraction includes: basic information of the included study, such as title, trial name, first author, and publication year; baseline characteristics of the study subjects, such as the number of participants, gender, and age; specific details of the intervention measures and follow-up time; key elements of bias risk assessment; outcome indicators of interest and data on outcome measurements, such as VAS; failure to progress natural delivery, adverse events and neonatal Apgar score.

2.4. Quality assessment

The risk of bias in RCTs was assessed by 2 reviewers (Mengxian Yu and Haiyan Qian) using the Cochrane Collaboration risk of bias tool as depictive in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of interventions, version 5.1.0), which comprised items such as random sequence generation, allocation concealment, blinding of participants and outcome assessors, incomplete outcome data, reporting bias, and other bias. Each domain was assessed as low, unclear and high according to the instructions. In case of any discrepancies in the evaluations between the 2 reviewers, a third reviewer (Miao Gan) was consulted to resolve them.

2.5. Statistical analysis

A network meta-analysis was conducted to assess the labor pain of 2 treatments (TEAS, EA and control groups) using a random-effect model within a Bayesian framework. The analysis was carried out using the “gemtc” and “rjags” packages in R software (version 3.5.1, https://www.r-project.org/). Convergence was evaluated through Markov chain Monte Carlo methods, employing 4 chains initialized with over-dispersed values and Gibbs sampling. The total number of iterations was 50,000, with a burn-in phase of 20,000 iterations (Sutton and Abrams). Median values from the posterior distribution were used to calculate estimated outcomes, represented as mean differences or odds ratios, along with their corresponding 95% confidence intervals. Statistically significant differences were considered when the 95% confidence intervals did not encompass 1 for odds ratios or did not encompass 0 for mean differences. The significance level was set at P < .05. Furthermore, Surface under the cumulative ranking curve (SUCRA) values were employed to rank the different interventions, where higher SUCRA values indicated more effective interventions. To evaluate heterogeneity, the I² test was used, with a threshold of 50% indicating low heterogeneity. Global inconsistency was assessed by comparing the fit of consistency and inconsistency models using the deviance information criterion (DIC), with similar DIC values across different models indicating good consistency. For local inconsistency, node-splitting analysis was utilized, and a P value >0.05 suggested no significant inconsistency between the direct pairwise results and the indirect results. In order to evaluate the confidence in the results comparing different treatment for labor pain, we employed the Confidence in Network Meta-Analysis web application. Confidence in network meta-analysis serves as a user-friendly tool specifically designed to aid in the assessment of confidence levels in the findings derived from a network meta-analysis.

3. Results

3.1. Study selection

Of the 885 relevant publications that were identified through the electronic search and additional sources, 251 were excluded because they were duplicates. After the title and abstract were screened, further 619 studies were excluded. After verifying the remaining 15 full-text reports, 12 studies met the inclusion criteria and were included in the network meta-analysis.^[19,21–31] The study selection process is shown in Figure 1.

Figure 1.

Flow diagram of the literature selection process.

3.2. Study characteristics

Only one-arm-based study, and the 11 2-arm-based study were included. The included studies were published between 1990 and 2022. Six studies compared TEAS versus control, 3 compared EA versus control, 2 compared TEAS versus EA and one compared TEAS versus EA versus control. Number of patients ranged from 30 to 200. Gestational age ranged from 28.0 to 40.3. Acupoints including LI4, PC6, SP6, EX-B2 (T10-L3), BL32 and HT7. Waveforms including continuous and dilatational. Electrical frequency including 100 Hz, 2/100 Hz, 2-4 Hz, 5-10 Hz and once an hour, for 30 minutes each time. Details of EA and duration of intervention can be seen in Table 1.

Table 1.

General characteristic of the included studies.

Author	No. of patients (I/C)	Interventions	Age (yr)	Gestational age (yr)	Details of TEAS			Details of EA	Duration of intervention
					Acupoints	Waveforms	Electrical frequency
Aghamohammadi 2013	32/32	TEAS/Control	28.4/27.6	39.0/39.1	LI4,SP6(bilateral)	Continuous	100Hz		Active phase
BASTANI 2020	76/75	TEAS/EA	28/27	39.6/39.5	LI4/PC6	Dilatational	2/100Hz	1% lidocaine 5 mL + the mixture of 0.125% ropivacaine and 2 µg/mL sufentanil (8 mL)	Active phase
Chao 2007	50/50	TEAS/Control	27.9/28.8	39.0/38.9	LI4,SP6(bilateral)	Dilatational	2/100Hz		From the time that analgesia methods are required to the end of the first stage of labor
Chestnut 1990	29/34	EA/Control	27.9/28.8	39.7/40.1				0.0625% bupivacaine + 0.0002% fentanyl	From the second stage to the end of the delivery
Kamali 2018	60/60	EA/Control	25.4/25.9	40.2/40.3				0.125% Marcaine 4–6 cc? a single dose of 5ug fentanyl
Liu 2015	30/30	TEAS/Control	27.5/26.6	39.0/39.0	EX-B2 (T10-L3), BL32		2/100Hz		Active phase
Mehri 2022	64/66	TEAS/Control	24.1/24.8	38.8/39.2	SP6,EX-LF29		2-4Hz		From active phase to the end of delivery
Miao 2020	76/75	TEAS/EA	28/27	39.6/39.5	LI4, PC6, EX-B2(T10-L1), BL32 (Bilateral)	Dilatational	2/100Hz	Maintenance: the mixture of 0.125% ropivacaine + 2 µg/mL sufentanil (100 mL) Speed: 10 mL/h
Mucuk 2013	39/39	TEAS/Control	23.8/23.8	39.1/39.4	LI4 (bilateral)	Dilatational	5-10Hz		20 min (from the beginning of the active phase)
Shen 2017	200/200	EA/Control	28.0/28.1	28.0/39				0.08% ropivacaine with 0.4 mg/mL sufentanil 8–10 mL	The second stage of labor
Chen 2021	98/100	TEAS/Control	22.6/23.4	39.5/39.8	bilateral Hegu (LI4) and Quchi (LI11)		once an hour, for 30 min each time		The second stage of labor
Qi 2021	76/75/78	TEAS/EA/Control	27.6/27	NS	L2 and L3		2/100Hz	mL of 0.125% ropivacaine and 0.5 μg/mL sufentanil	The second stage of labor

Acupoints: Hegu (LI4), Shenmen (HT7), Sanyinjiao (SP6), Neiguan (PC6), Ciliao (BL32), Jiaji (EX-B2), extra meridian point Neimadian (EX-LE29); I = intervention, C = comparator, EA = epidural analgesia, TEAS = transcutaneous electrical acupoint stimulation.

3.3. Risk of bias

The risk of bias summary and risk of bias graph for the included trials is shown in Figures 2 and 3, respectively. Two trials were graded as having a low risk of bias, 9 trials were at an unclear risk of bias and 1 was at high risk of bias. All included trials reported an appropriate random sequence generation and allocation concealment. Six studies were graded as having an unclear risk of bias for blinding of participants and personnel. One study was graded as having high risk of bias for blinding of participants and personnel. Only one study was graded as having unclear risk of bias for blinding of participants and personnel and selective reporting. Four studies were graded as unclear risk of bias for other bias.

Figure 2.

Risk of bias graph of the included studies.

Figure 3.

Risk of bias summary of the included studies.

3.4. VAS

A total of 8 studies involving 557 patients, including 3 treatments (TEAS, EA and control) contributed to the clinical outcome of the VAS. As displayed in Figure 4A, the network structure diagrams detailed the direct comparisons between different treatments in the VAS. Network meta-analysis showed considerable heterogeneity with global (EA vs control, 75.6%, TEAS vs control, 92.9%, TEAS vs EA, 82.2%, Fig. 4B, Table 2). In a head-to-head comparison, TEAS (WMD −3.1, 95% CrI −3.8, −2.5, Fig. 4C) and EA (WMD −2.1, 95% CrI −2.8, −1.3, Fig. 4C) was more effective than the control in decreasing VAS. The confidence intervals from direct and indirect evidence are in general consistent, with minor differences (P > .05).

Figure 4.

(A) Network structure diagrams of VAS; (B) Heterogeneity plot of this network meta-analysis (constant score); (C) Forest plot of the VAS as compared with placebo; (D) Inconsistency plot of this network meta-analysis (constant score); (E) Surface under the cumulative ranking curve (SUCRA) probabilities of different treatments for VAS. VAS = visual analogue scale.

Table 2.

Efficacy of different treatments for VAS by WMDs and corresponding 95% CrIs.

Control	−2.05 (−2.85, −1.26)	−3.13 (−3.8, −2.49)
2.05 (1.26, 2.85)	EA	−1.08 (−1.91, −0.25)
3.13 (2.49, 3.8)	1.08 (0.25, 1.91)	TEAS

EA = epidural analgesia, TEAS = electrical acupoint stimulation, VAS = visual analogue scale.

In network meta-analysis, one of the fundamental assumptions is the coherence between direct evidence and indirect evidence. The node-splitting method is employed to assess the level of consistency between indirect and direct evidence. In our network, the evidence seems to exhibit coherence in most comparisons (P > .05, Fig. 4D).

The SUCRA shows that TEAS ranked first (SUCRA, 90.9%), EA ranked second (SUCRA, 74.0%) and control ranked last (SUCRA, 35.0%, Fig. 4E).

3.5. Failure rate of natural delivery

A total of 10 studies including 3 treatments (TEAS, EA and control) contributed to the clinical outcome of the failure rate of natural delivery. As displayed in Figure 5A, the network structure diagrams detailed the direct comparisons between different treatments in the failure rate of natural delivery. Network meta-analysis showed considerable heterogeneity with global I² = 0% (Fig. 5C).

Figure 5.

(A) Network structure diagrams of failure rate of natural delivery; (B) Heterogeneity plot of this network meta-analysis (constant score); (C) Forest plot of the failure rate of natural delivery as compared with placebo; (D) Surface under the cumulative ranking curve (SUCRA) probabilities of different treatments for failure rate of natural delivery.

In a head-to-head comparison, only TEAS (OR 0.38, 95% CrI 0.21, 0.65, Fig. 5B) was more effective than the placebo in decreasing the failure rate of natural delivery. However, there was no statistically significant between EA versus placebo in terms of the failure rate of natural delivery (OR 0.63, 95% CrI 0.35, 1.1, Table 3).

Table 3.

Efficacy of different comparisons of different treatments for failure rate of natural delivery by ORs and corresponding 95% CrIs.

Control	0.63 (0.35, 1.13)	0.38 (0.21, 0.65)
1.59 (0.88, 2.88)	EA	0.6 (0.29, 1.18)
2.66 (1.54, 4.76)	1.67 (0.85, 3.41)	TEAS

EA = epidural analgesia, TEAS = electrical acupoint stimulation.

The SUCRA shows that TEAS ranked first (SUCRA, 96.6%), EA ranked second (SUCRA, 50.4%) and control ranked last (SUCRA, 3.0%, Fig. 5D).

3.6. Apgar scores

A total of 10 studies involving 618 patients, including 3 treatments (TEAS, EA and control) contributed to the clinical outcome of the Apgar scores. As displayed in Figure 6A, the network structure diagrams detailed the direct comparisons between different treatments in the Apgar scores. Network meta-analysis showed considerable heterogeneity with global I² = 0% (Fig. 6B).

Figure 6.

(A) Network structure diagrams of Apgar scores; (B) Heterogeneity plot of this network meta-analysis (constant score); (C) Forest plot of the Apgar scores as compared with placebo; (D) Inconsistency plot of this network meta-analysis (constant score); (E) Surface under the cumulative ranking curve (SUCRA) probabilities of different treatments for Apgar scores.

In a head-to-head comparison, there was no statistically significant between EA versus placebo (WMD = −0.095, 96% CrI: −0.76 to 0.58, Fig. 6C), TEAS versus placebo (WMD = 0.22, 96% CrI: −0.35 to 0.77, Fig. 6C) in terms of the Apgar scores (P > .05, Table 4).

Table 4.

Efficacy of different comparisons of different treatments for failure rate of natural delivery by WMDs and corresponding 95% CrIs.

Control	−0.09 (−0.76, 0.58)	0.22 (−0.35, 0.77)
0.09 (−0.58, 0.76)	EA	0.31 (−0.4, 1.01)
−0.22 (−0.77, 0.35)	−0.31 (−1.01, 0.4)	TEAS

EA = epidural analgesia, TEAS = electrical acupoint stimulation.

One of the main assumptions of the network meta-analysis is the consistency between direct evidence and indirect evidence. The degree of indirect evidence is consistent with direct evidence by the node splitting method. The evidence in the network seems to be consistent with most comparisons (P > .05, Fig. 6D).

The SUCRA shows that TEAS ranked first (SUCRA, 80.7%), control ranked second (SUCRA, 41.4%) and EA ranked last (SUCRA, 27.9%, Fig. 6E).

3.7. Adverse events

A total of 10 studies involving 618 patients, including 3 treatments (TEAS, EA and control) contributed to the clinical outcome of adverse events.

As displayed in Figure 7A, the network structure diagrams detailed the direct comparisons between different drugs in the adverse events. Network meta-analysis showed considerable heterogeneity with global I² = 0% (Fig. 7B).

Figure 7.

(A) Network structure diagrams of adverse events; (B) Heterogeneity plot of this network meta-analysis (constant score); (C) Forest plot of the adverse events as compared with placebo; (D) Surface under the cumulative ranking curve (SUCRA) probabilities of different treatments for adverse events.

In head-to-head comparison, TEAS (OR 1.5, 95% CrI 0.63, 3.3, Fig. 7C) and EA (OR 1.6, 95% CrI 0.59, 4.2, Fig. 7C, Table 5) were not associated with an increase of the adverse events. The SUCRA shows that control ranked first (SUCRA, 99.9%), TEAS ranked second (SUCRA, 32.3%) and EA ranked last (SUCRA, 17.6%, Fig. 7D).

Table 5.

Efficacy of different comparisons of different treatments for adverse events by ORs and corresponding 95% CrIs.

Control	1.56 (0.59, 4.25)	1.45 (0.63, 3.31)
0.64 (0.24, 1.71)	EA	0.93 (0.34, 2.5)
0.69 (0.3, 1.58)	1.08 (0.4, 2.97)	TEAS

EA = epidural analgesia, TEAS = electrical acupoint stimulation.

3.8. Publication bias and grading evidence

We used a comparison-adjusted funnel plot to determine whether there was a small sample effect or publication bias in the network. The funnel plot of the studies was found to be symmetrical (Fig. 8), indicating that there was no publication bias or small sample size study effects. Most of the confidence in our comparisons of interest was rated as low or very low (Supplement S1, http://links.lww.com/MD/L769).

Figure 8.

The funnel plot used for publication bias for VAS (A), failure rate of natural delivery (B), Apgar scores (C) and adverse events (D). VAS = visual analogue scale.

4. Discussion

In this study, we conducted an NMA to compare the efficacy and safety of 2 interventions: TEAS and EA for relieving labor pain compared to control groups. Our findings found that both TEAS and EA significantly reducing the VAS, failure rate of natural delivery and Apgar scores. Moreover, adverse events between TEAS and EA groups were not statistically significant. SURCA value indicated that TEAS ranked first in reducing VAS and failure rate of natural delivery.

4.1. Compared with previous meta-analyses

Only one systematic review and meta-analysis about TEAS for reducing labor pain was published.^[32] The weakness of the study is the small number of subjects enrolled. Furthermore, an inconsistency analysis was not carried out in this study, which is important for a meta-analysis. Lei et al^[33] conducted a meta-analysis about EA versus patient-controlled analgesia for reducing pain in labor patients. Patient-controlled analgesia can be an optional alternative to EA with similar pain relief and less risk of intrapartum maternal fever. In a systematic review and meta-analysis conducted by Anim-Somuah et al,^[7] the comparison between EA and non-epidural or no analgesia for pain management during labor was investigated. The findings, based on low-quality evidence, suggest that EA might be more effective in reducing labor pain and enhancing maternal satisfaction with pain relief compared to non-epidural methods.

Childbirth is often considered one of the most painful experiences for women, varying in duration from a few hours to several days. Unfortunately, this fear of childbirth pain has led to a high rate of non-medical cesarean sections as some pregnant women opt for this delivery method. Currently, EA is the most reliable and widely used approach for labor pain relief. However, due to the absence of comprehensive systematic evaluations, the clinical impact of EA remains a subject of intense debate. Based on the aforementioned results, no significant differences were observed between TEAS and EA concerning the VAS score, failure rate of natural delivery, or Apgar scores. This indicates that TEAS demonstrated comparable effectiveness to EA in terms of reducing labor pain, decreasing the failure rate of natural delivery, and improving newborn well-being. Importantly, TEAS achieved the highest SUCRA score among the 3 interventions or controls, indicating its superior ranking in all aspects. Consequently, we propose that TEAS may be a preferable choice for parturients due to its relatively effective analgesic properties, higher rate of natural delivery, and favorable Apgar scores. In conclusion, addressing the fear of childbirth pain and providing reliable and effective labor pain relief options, such as TEAS, may help reduce the unnecessary prevalence of non-medical cesarean sections and improve overall childbirth experiences for women.

Network meta-analysis results found that both TEAS and EA will not associate with an increase of the adverse events. Two RCTs involving TEAS interventions reported on adverse events.^[22,29] Liu et al^[22] revealed that no clear differences were seen between groups for adverse effects between TEAS and control groups. Other study reported that adverse events in the TEAS group was lower than that in the EA group (P < .01).^[29] These studies suggested that TEAS was a relative safe method to relieve pain without increasing the adverse events.

According to a review,^[34] nonpharmacologic approaches such as acupuncture and TENS show promise in providing safe and harmless alternatives, but further research is needed to substantiate their effectiveness. While TEAS has not shown specific adverse events in our study, it would be premature to claim that it is entirely risk-free for both parturients and neonates. Being a nonpharmacological intervention, TEAS offers the advantage of avoiding potential drawbacks like hemodynamic instability, intoxication,^[35,36] subtle depression, and neonatal respiratory depression associated with narcotic drugs.^[36] Consequently, TEAS presents an additional potential alternative for parturients and doctors, especially in unique circumstances where these issues are a concern, making it a more acceptable option for parturients. However, careful evaluation and continued research are essential to establish its safety conclusively.

5. Limitations

This study does have several limitations that need to be considered when interpreting the findings. Firstly, the major concern of this network meta-analysis is the inclusion of drugs with different doses and treatment duration, which undermines the robustness and reliability of the results and conclusions. Secondly, subgroup analysis was not investigated due to the number of the included studies, including drug dose and duration. Thirdly, potential confounding factors were not stratified and excluded, and might influence the results. Besides, a wide ranged mean age and follow-up time data also increased heterogeneity between studies.

6. Conclusions

In summary, our NMA suggests that TEAS, a simple intervention, maybe a viable option for parturients in providing effective and safe pain relief to both mothers and neonates when compared to EA. This study provides new evidence supporting TEAS as a noninvasive and non-narcotic approach to manage labor pain, particularly in cases of hemodynamic instability and anesthetic allergies. However, it is important to note that the evaluation of TEAS safety is not entirely comprehensive due to a lack of specific adverse event data. To address this crucial aspect, further RCTs are needed. However, the overall quality of evidence is low, the overall certainty of the evidence synthesis is low. In the future, there is a need for more high-quality RCTs to reassess or confirm this conclusion.

Author contributions

Formal analysis: Haiyan Qian.

Software: Miao Gan.

Supervision: Miao Gan.

Validation: Haiyan Qian.

Visualization: Mengxian Yu.

Writing – review & editing: Mengxian Yu.