FormalPara What does this study add to the clinical work

Issue : How effective is PFMT in preventing perineal trauma?

What is already known : PFMT can be effective in preventing and treating pelvic floor muscle dysfunction, but its preventive effect on perineal injury has yet to be verified.

What this paper adds : With moderate certainty of evidence, “PFMT combine with perineal massage” and PFMT showed more superiority for preventing perineal trauma.

Introduction

Most women suffer from perineal trauma during childbirth, whether it is natural tears or episiotomy, affecting 53–79% of women [1]. In Queensland, 59.16% women required surgical repair because of perineal trauma in 2020 [2]. Injury to the perineum without involvement of the anal sphincter does not generally cause long term problems for women. In contrast, injury to the anal sphincter can result in long term sequelae, such as fecal incontinence, and can significantly affect women’s quality of life [1]. Episiotomy and obstetric anal sphincter injuries (OASIS) rates are commonly used as quality indicators of health systems and health care. [3]

The prevention of perineal trauma through exercise during pregnancy was feasible and effective, which has been verified and recommended by a number of guidelines and clinical studies [4,5,6,7,8,9,10]. However, the results concerning the relationship between different antenatal PFMT relevant prevention strategies and perineal trauma seemed to be inconclusive. Therefore, we intended to use network meta-analysis method to combine results of RCTs to assess the relative effectiveness of different PFMT relevant strategies in the prevention of perineal trauma during childbirth.

Methods

Our systematic review and NMA manuscript were written in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension statement for reporting systematic reviews incorporating NMA [11].

Search strategy

We searched PubMed, Embase, the Cochrane Library, CINAHL, CNKI, CBM, WANFANG DATA and ClinicalTrials.gov for citations published in any language from inception to 1 July 2020. We also reviewed reference lists of included studies and related systematic reviews. The search strategy was provided in Appendix Table 2

Study selection and criteria

We included RCTs of different strategies (except antenatal perineal massage) used to prevent perineal trauma in pregnant women. We followed the PICO (population, strategies, comparison, outcome) framework to determine the research question and to facilitate literature search: (1) Population: single pregnant women, there were no other restrictions; (2) Strategies: PFMT relevant strategies; (3) Comparison: usual care or other strategies. Usual care means that pregnant women only receive routine guidance and nurse during pregnancy; (4) Outcome: perineal trauma; (5) Study design: RCTs. We excluded: (1) in vitro and animal studies, (2) studies whose data were unable for extraction and analysis. Appendix Table 3 shows the definitions of different prevention strategies.

Data extraction

Pairs of reviewers (XC, QY) independently screened all citations and extracted data from the included studies. Discrepancies in study inclusion were resolved by deliberation within the reviewer pairs or with input from a third reviewer (MYS). The following items were extracted from each identified study: (1) basic information: title, authors, publication year, etc.; (2) risk of bias domains; and (3) study outcomes: the rate of intact perineal, the rate of episiotomy, the rate of perineal tear, etc.

Risk of bias assessment

Two reviewers independently assessed risk of bias with adjudication by a third reviewer, using a Cochrane Collaboration tool [12]. This tool included seven domains: random sequence generation, allocation concealment, blinding of participants and personal, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias.

Statistical analysis

We conducted network meta-analysis with R version 4.1.2 and Stata 17.0 [13, 14]. For the dichotomous outcomes, we calculated risk ratios (RRs) and 95% confidence intervals, and mean differences (MDs) and 95% confidence intervals for the continuous outcomes. We assessed the heterogeneity in strategies effects among studies using visual inspection of forest plots and the I2 We assessed the transitivity (similarity) assumption by comparing the distribution of the population, the strategies, and the methodological characteristics of the studies across strategies comparisons [15]. To assess incoherence, we assessed the global incoherence for the entire network for each outcome under the assumption of a full design-by-strategies interaction random effects model, and then local incoherence for each comparison using the node-splitting model [16, 17]. We ranked strategies according to their P score, which is between 0 and 1 [18]. We used the egger’s test and the funnel chart test to publish bias.

Certainty of evidence assessment

We used the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) approach to assess the certainty of evidence [19]. Regarding the certainty of evidence in estimates: high certainty means that further research is very unlikely to change our confidence in the estimate of effect; moderate certainty means that further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate; low certainty means that further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate; and very low certainty means that any estimate of effect is very uncertain. [20]

Results

The search strategies generated 12,632 citations, 21 RCTs were eligible [21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41]. Figure 1 presents the study flow diagram.

Fig. 1
figure 1

Study selection flow diagram

Full size image

Characteristics of included studies

Twenty-one RCTs with sample sizes from 20 to 596, enrolled a total of 4931 patients. All RCTs included pregnant women, and 12 of these included primiparous women, one of these included primiparous women with bladder neck mobility (BNM), one of these included pregnant women ≥ 35 years, one of these included pregnant women with pelvic floor dysfunction.

Figure 2 presents the network plot including all studies and demonstrated that the most common comparisons were between PFMT relevant strategies and usual care. Ten RCTs compared PFMT with usual care; five RCTs compared Epi-NO with usual care; three RCTs compared “PFMT combine perineal massage” with usual care; one RCT compared “PFMT combine perineal massage” with perineal massage; two RCTs compared yoga with usual care (one Pilates, one yoga). Table 1 presents detailed characteristics of individual RCTs.

Fig. 2
figure 2

Network plot of comparisons among Usual (usual care), PFMT, Epi-NO, PM (PFMT combine with perineal massage), Massage (perineal massage) or Yoga

Full size image
Table 1 Characteristics of included RCTs
Full size table

Risk of bias assessment

Most of the included studies showed a low or ambiguous risk of bias in seven areas, with one study (4.76%) showed a high risk of bias in random sequences generation and two studies (9.52%) showed a high risk of bias in the blindness. Because pelvic floor muscle exercise required patient participation, it was difficult to implement blinding method. All outcome indicators were measurable objective, we considered the impact of blinding method was small. Overall, only four studies were assessed as to be low risk of bias, and the remaining 17 studies were all high risk of bias. Appendix Fig. 5 presents the risk of bias.

Primary outcome

In this network meta-analysis, none of the outcomes formed a closed loop. Therefore, we did not need to test the inconsistency. Appendix Figs. 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 presents forest plots for each outcome. No publication bias was observed, see Appendix Figs. 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33.

The rate of intact perineum

Eight RCTs [25, 27, 30, 31, 35, 36, 38, 39] reported the rate of intact perineum, included usual care, PFMT, Epi-NO, “PFMT combine with perineal massage”, perineal massage, and Yoga. Compared with usual care, moderate certainty of evidence suggested that “PFMT combine with perineal massage” (RR = 5.37, 95% CI 3.79–7.60), Yoga (RR = 3.81, 95% CI 2.55–5.69), PFMT (RR = 2.58, 95% CI 1.34–4.97), Epi-NO (RR = 1.77, 95% CI 1.30–2.40) could significantly improve the rate of intact perineum. There were no differences between strategies except the comparison between “PFMT combine with perineal massage” and Epi-NO. Figure 3 shows the GRADE summary of findings.

Fig. 3
figure 3

GRADE summary of findings for the rate of intact perineum and episiotomy

Full size image

The rate of episiotomy

Twenty RCTs [21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37, 39,40,41] reported the rate of episiotomy, included usual care, PFMT, Epi-NO, “PFMT combine with perineal massage”, perineal massage, and Yoga. Compared with usual care, low certainty of evidence suggested that “PFMT combine with perineal massage” (RR = 0.26, 95% CI 0.14–0.49) and PFMT (RR = 0.63, 95% CI 0.45–0.90) could significantly reduce the rate of episiotomy, and combined strategy was better than PFMT alone (RR = 0.41, 95% CI 0.20–0.85; low certainty). There were no significant differences between other strategies (Fig. 3).

The rate of perineal tear

Ten RCTs [21,22,23, 25, 30, 36,37,38,39,40] reported the rate of perineal tear, included usual care, PFMT, Epi-NO and “PFMT combine with perineal massage” strategies. Compared with usual care, low certainty of evidence suggested that “PFMT combine with perineal massage” could significantly reduce the rate of perineal tear (RR = 0.52, 95% CI 0.31–0.88). There were no significant differences between strategies (Fig. 4).

Fig. 4
figure 4

GRADE summary of findings for the rate of perineal tear and OASIS

Full size image

The rate of OASIS

Nine RCTs [21, 23, 26, 27, 29, 30, 34, 38, 41] reported the rate of OASIS, included usual care, PFMT, Epi-NO and “PFMT combine with perineal massage” strategies. Compared with usual care, high certainty of evidence suggested that PFMT (RR = 0.49, 95% CI 0.28–0.85) and moderate certainty of evidence suggested that “PFMT combine with perineal massage” (RR = 0.35, 95% CI 0.16–0.78) could significantly reduce the rate of OASIS. Results suggested no significant differences between any of strategies (Fig. 4).

Secondary outcomes

Compared with usual care, “PFMT combine with perineal massage” (RR = 0.70, 95% CI 0.56–0.88, moderate certainty) and perineal massage (RR = 0.60, 95% CI 0.39–0.93, low certainty) could significantly reduce the rate of slight perineal tear; Epi-NO improved the rate of assisted vaginal delivery (RR = 1.59, 95% CI 1.04–0.42, moderate certainty); PFMT combine with perineal massage (RR = 1.38, 95% CI 1.04–1.82, low certainty) and yoga (RR = 1.75, 95% CI 1.47–2.08, moderate certainty) could significantly improve the rate of natural childbirth; PFMT combine with perineal massage (RR = 0.31, 95% CI 0.12–0.83, low certainty) and yoga (RR = 0.47, 95% CI 0.32–0.71, low certainty) could significantly reduce the rate of caesarean section; PFMT combine with perineal massage (RR = 0.35, 95% CI 0.16–0.78, low certainty) and PFMT (RR = 0.49, 95% CI 0.28–0.85, low certainty) could significantly reduce the rate of fetal distress; PFMT (MD = − 20.27, 95% CI − 38.63–1.90, low certainty) could significantly reduce the duration of the second stage of labor. Figure 5 presents the GRADE summary of findings for secondary outcomes.

Discussion

Main findings

Our meta-analysis showed that both “PFMT combine with perineal massage” and PFMT showed advantages in terms of the primary outcomes compared with usual care. This suggested that “PFMT combine with perineal massage” and PFMT might be the most effective first line strategy for the prevention of perineal trauma.

Some systematic review found that exercise could not prevent perineal trauma, but one found that perineal massage could prevent perineal trauma [12,13,14,15]. In the present network meta-analysis, we focused on the relationship between different antenatal PFMT relevant strategies and perineal trauma. Therefore, we did not deeply compare PFMT relevant strategies with perineal massage. This network meta-analysis provided moderate certainty evidence for the use of “PFMT combine with perineal massage” and PFMT to prevent perineal trauma.

Strengths and limitations

The strength of our study was that we considered most current PFMT relevant strategies for the prevention of perineal trauma. Application of the network meta-analysis framework allowed to compare multiple relative strategies and rank the best agents for preventing perineal trauma given the evidence to date from RCTs. However, there were several limitations in our study. First, we did not consider all antenatal strategies; therefore, our conclusion was limited to these strategies we focused on. Second, we did not consider cost effectiveness, scalability, and sustainability of these strategies. Third, we have not found any research about patient’s preference; therefore, the most effective strategy might not be accepted by pregnant women.

Interpretation

The strategies of PFMT and perineal massage were popular in many countries. However, now, PFMT is more used in the PFM recovery, and perineal massage is more used in intrapartum. Several effective antenatal strategies, including perineal massage and exercise were already accessible in some clinical and community settings. However, the decision to use any of these strategies, and the order in which they might be used, was based on choices made by physician and patient, and to some extent was influenced by many factors.

In view of the results, with moderate certainty of evidence, antenatal “PFMT combine with perineal massage” and PFMT were effective in the prevention of perineal trauma. Clinicians and healthcare professionals could consider to recommend these strategies based on the condition and preference of pregnant women.