Methods of repair for obstetric anal sphincter injury
Abstract
Background
Anal sphincter injury during childbirth ‐ obstetric anal sphincter injuries (OASIS) ‐ are associated with significant maternal morbidity including perineal pain, dyspareunia (painful sexual intercourse) and anal incontinence, which can lead to psychological and physical sequelae. Many women do not seek medical attention because of embarrassment. The two recognised methods for the repair of damaged external anal sphincter (EAS) are end‐to‐end (approximation) repair and overlap repair.
Objectives
To compare the effectiveness of overlap repair versus end‐to‐end repair following OASIS in reducing subsequent anal incontinence, perineal pain, dyspareunia and improving quality of life.
Search methods
We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (30 September 2013) and reference lists of retrieved studies.
Selection criteria
Randomised controlled trials comparing different techniques of immediate primary repair of EAS following OASIS.
Data collection and analysis
Trial quality was assessed independently by all authors.
Main results
Six eligible trials, of variable quality, involving 588 women, were included. There was considerable heterogeneity in the outcome measures, time points and reported results. Meta‐analyses showed that there was no statistically significant difference in perineal pain (risk ratio (RR) 0.08, 95% confidence interval (CI) 0.00 to 1.45, one trial, 52 women), dyspareunia (average RR 0.77, 95% CI 0.48 to 1.24, two trials, 151 women), flatus incontinence (average RR 1.14, 95% CI 0.58 to 2.23, three trials, 256 women) between the two repair techniques at 12 months. However, it showed a statistically significant lower incidence of faecal urgency (RR 0.12, 95% CI 0.02 to 0.86, one trial, 52 women), and lower anal incontinence score (standardised mean difference (SMD) ‐0.70, 95% CI ‐1.26 to ‐0.14, one trial, 52 women) in the overlap group. The overlap technique was also associated with a statistically significant lower risk of deterioration of anal incontinence symptoms over 12 months (RR 0.26, 95% CI 0.09 to 0.79, one trial, 41 women). There was no significant difference in quality of life. At 36 months follow‐up, there was no difference in flatus incontinence (average RR 1.12, 95% CI 0.63 to 1.99, one trial, 68 women) or faecal incontinence (average RR 1.01, 95% CI 0.34 to 2.98, one trial, 68 women).
Authors' conclusions
The data available show that at one‐year follow‐up, immediate primary overlap repair of the external anal sphincter compared with immediate primary end‐to‐end repair appears to be associated with lower risks of developing faecal urgency and anal incontinence symptoms. At the end of 36 months there appears to be no difference in flatus or faecal incontinence between the two techniques. However, since this evidence is based on only two small trials, more research evidence is needed in order to confirm or refute these findings.
PICOs
Plain language summary
Methods of repair for obstetric anal sphincter injury
Ways of repairing damage to the muscles of the back passage following tearing during a difficult vaginal birth.
The risk factors for obstetric anal sphincter injuries include a midline cut of the perineum (episiotomy) to facilitate the birth, forceps delivery and the baby’s back presenting posteriorly (occipito‐posterior position). Most women give birth without any significant damage to their perineum or back passage. However, in about 1% to 4% of births, there is tearing and damage which extends to the back passage and the anal sphincter. This can cause considerable problems for some of these women in terms of pain, painful intercourse and faecal incontinence. For a few of these women, the incontinence can be very embarrassing and can impact significantly on their daily lives and relationships. This review of randomised controlled trials compared two different stitching techniques, one where the edges of the tissues were overlapped and the other where they were sewn end‐to‐end. There were 588 women in the six trials analysed. The trials were of moderate quality and differed in the participants, the outcome measures used and the points in time when they were measured. Only three trials with 156 women had 12‐month follow‐up data, when the overlap technique carried out as soon after childbirth as possible appeared to be better in terms of having faecal urgency and incontinence. Further research is needed to address women's views and experiences of surgery to prevent long‐term problems.
Authors' conclusions
Background
Perineal trauma can occur spontaneously during vaginal delivery (Sultan 1994a). However, a cut (episiotomy) may be made to facilitate the birth. Until recently there has been inconsistency in the classification of perineal trauma (Sultan 2002). This caused confusion among the clinicians in identification and management resulting in systematic under‐reporting and misclassification (Sultan 1995). A survey amongst consultant obstetricians in the UK in 2000 revealed that 46% were still classifying a complete or partial injury to the external anal sphincter (EAS) as a second‐degree tear (Fernando 2000). The current classification of perineal trauma was modified by Sultan in 1999 (Sultan 1999b) and has been adopted by the Royal College of Obstetricians and Gynaecologists (RCOG 2007) and the International Consultation on Incontinence since 2002 (Koelbl 2009; Norton 2002).
-
First‐degree: injury to the skin only.
-
Second‐degree: injury to the perineum involving perineal muscles but not involving the anal sphincter.
-
Third‐degree: injury to the perineum involving the anal sphincter complex: 3a: less than 50% of EAS thickness torn; 3b: more than 50% of EAS thickness torn; 3c: internal anal sphincter (IAS) also torn.
-
Fourth‐degree: injury to the perineum involving the anal sphincter complex (EAS and IAS) and ano‐rectal epithelium.
Obstetric anal sphincter injuries (OASIS) include only third‐ or fourth‐degree perineal tears and may cause considerable morbidity when compared to first‐ or second‐degree perineal tears. Over the last five years, there has been an increase in litigation related to anal sphincter injury during childbirth and subsequent complications including faecal incontinence (Eddy 1999).
One of the most distressing immediate complications of any perineal injury is perineal pain which may interfere with the mother's abilities to breast feed and cope with the daily tasks of motherhood (Sleep 1991). Short‐term perineal pain is associated with reactionary oedema, bruising, tight sutures, infection and wound breakdown. Persistent pain and discomfort from perineal trauma may also cause urinary retention and defecation problems. Several studies reported that following OASIS, perineal pain and dyspareunia (painful sexual intercourse) may last for several years and up to 50% of women complained of perineal pain and dyspareunia following OASIS (Haadem 1987; Haadem 1990; Sultan 1994b). Long‐term perineal pain and dyspareunia can have considerable impact on sexual and social well being (Giebel 1993). Wheeless 1998 reported that some women with dyspareunia following OASIS abstained from sexual intercourse for up to 14 years. Sorensen 1988 reported a significant permanent disfigurement of the perineum following OASIS.
Abscess formation, wound breakdown and recto‐vaginal fistulae (abnormal communication between rectum and vagina) have been reported following OASIS (Combs 1990; Corman 1980; Goldberg 1980; Pezim 1987; Rothenberger 1982). Venkatesh 1989 reported a 10% perineal wound disruption rate following primary repair of OASIS. The reported incidence of recto‐vaginal fistulae in women who sustained a fourth‐degree perineal tear ranges from 0.4% to 3.0% (Rogers 2007). Giebel 1993 reported that most recto‐vaginal fistulae were caused by failure to recognise the true extent of OASIS at the time of repair resulting in inadequate sphincter reconstruction and subsequent wound breakdown. Recto‐vaginal fistulae are difficult to treat and may ultimately result in a permanent colostomy (Giebel 1993; Pezim 1987).
In addition to perineal pain, dyspareunia, wound breakdown and recto‐vaginal fistulae formation, OASIS has been reported as the most common cause of anal incontinence following childbirth (Sultan 1997). Anal incontinence is defined by the International Continence Society as involuntary loss of flatus or faeces, which becomes a social or hygienic problem (Milsom 2009). It is estimated that anal incontinence affects one in 20 women up to one year after childbirth, affecting nearly 40,000 mothers each year in the UK (Clarkson 2001; Glazener 1995; Glazener 1997; Glazener 1998; Macarthur 1991). One‐third of these are considered to be due to clinically undiagnosed (occult) OASIS (Sultan 1993). The reported incidence of anal incontinence following primary repair of recognised anal sphincter injury is between 15% and 61% (Koelbl 2009).
Anal incontinence affects women psychologically as well as physically. Many of them do not seek medical attention because of embarrassment engendered by these taboo symptoms (Browning 1983; DOH 2000; Gjessing 1998; Haadem 1988; Sultan 1993; Sultan 1994b; Wood 1998). Wood 1998 reported that the majority of women who sustained OASIS were either unaware of the diagnosis or were given a poor explanation for the injury. Indeed, some women were discouraged from discussing this because they felt that the anal incontinence symptoms were a normal consequence of childbirth (Haadem 1988; Walsh 1996). In one study, only one‐third of women with faecal incontinence had ever discussed the problem with a physician (Johanson 1996). One woman, in a letter to the Continence Foundation, described the eternal shame of being with another person when the worst occurs (Continence Foundation 2000). The impact of this complication on the vulnerable postnatal woman and her baby is potentially catastrophic, affecting her physically and psychosocially. Furthermore, anal incontinence caused by OASIS has been reported to be associated with very high cumulative costs for the health services (Mellgren 1999).
Increased awareness of these complications following vaginal birth has led women to request elective caesarean section without any other medical indication. A survey of female obstetricians reported that 31% would choose an elective caesarean section even in an uncomplicated pregnancy. The reason given by the majority was the potential risk of perineal trauma (Al‐Mufti 1996). However, caesarean delivery is associated with an increased risk of maternal morbidity compared with vaginal birth (NICE 2011).
The reported incidence of clinically detectable OASIS in the world literature is in the range of 0.5% to 7% with medio‐lateral episiotomy (Nordenstam 2008; Sultan 1994b; Tetzschner 1996; Uustal Fornell 1996), but there are some reports with an incidence as high as 17% with midline episiotomy (Fenner 2003). In England the rate of reported third‐ or fourth‐degree perineal tears has tripled from 1.8 to 5.9% between 2000 to 2012 (Gurol‐Urganci 2013).
The risk factors for OASIS include midline episiotomy (Fenner 2003), forceps delivery (Sultan 1994b) and occipito‐posterior position (Fitzpatrick 2001) at delivery and birthweight more than 4 kg (Sultan 1994b). Similar to first‐ and second‐degree tears OASIS is less likely to occur de novo in subsequent pregnancies (Poen 1997; Walsh 1996).
There are two recognised methods for the repair of damaged EAS, end‐to‐end (approximation) and overlap repair. In the end‐to‐end method, the torn ends of the EAS are approximated and sutured. In the overlap method, the torn ends of the EAS are brought together and sutured by overlapping one end of the muscle over the other in a double‐breasted fashion (Sultan 1999a). An overlap repair is only possible when the full length and thickness of the EAS is disrupted and 1 to 1.5 cm of the muscle ends are overlapped (Sultan 1999a, Sultan 2007). Therefore no attempt should be made to overlap Grade 3a tears and partial thickness Grade 3b tears as these should be repaired by the end‐to‐end technique. In 2000 about half of the consultants in the UK used the end‐to‐end method and the other half used the overlap method (Fernando 2000). Repair of OASIS is carried out as soon as possible after childbirth and is defined as a 'primary' repair. This is in contrast to 'secondary' repair which is carried out several months or years after the initial injury which may not be related to childbirth. In the UK, secondary sphincter repair for anal incontinence is usually performed by colorectal surgeons, although this repair can also be performed by gynaecologists with specific training.
The degree of anal incontinence can be measured by incontinence scoring systems (Jorge 1993; Vaizey 1999) or by quality of life assessment (Rockwood 2000). These methods involve using validated questionnaires inquiring about different aspects of incontinence and how this affects the individual's day‐to‐day life. The St Mark’s score (Vaizey 1999) ranges from zero to 24, with zero indicating complete continence and 24 indicating complete incontinence.
The anatomy of the anal sphincter can be assessed by anal endosonography. With this method, the extent of damage of the external and internal anal sphincter can be clearly seen. Function can be assessed by anorectal manometry which measures anal length and the pressure in the anal canal at rest and during voluntary squeeze.
There are several important questions related to the management of OASIS. These include the method of repair, suture materials used for the repair, the ideal clinician to perform the repair (obstetrician or colorectal surgeon), effectiveness of immediate versus delayed repair, prevention of OASIS and mode of delivery in subsequent pregnancies. The authors regard each of these questions as being important for future review. However, the aim of the current review is to examine the available studies to determine whether there is any clear scientific evidence that repair technique has any effect on subsequent continence and quality of life. Use of prophylactic antibiotics has been addressed in a separate Cochrane review (Buppasiri 2005).
Objectives
The objective of this review is to compare the effectiveness of overlap versus end‐to‐end repair following obstetric anal sphincter injuries in terms of preventing subsequent anal incontinence, perineal pain and dyspareunia as well as assessment of general improvement in the quality of life.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials that compared primary overlap versus end‐to‐end repair of obstetric anal sphincter injuries (OASIS) were included in this review. Quasi‐randomised and cluster‐randomised trials were also eligible for inclusion. Cross‐over trials were not considered eligible for inclusion.
Types of participants
All women who sustained OASIS and in whom the repair was performed in the immediate postpartum period (primary repair).
Types of interventions
All randomised controlled comparisons of the overlap versus end‐to‐end technique following OASIS.
Types of outcome measures
Primary outcomes
The main focus is on outcome measures relating to short‐ and long‐term postpartum morbidity.
The main outcome measures were:
-
anal incontinence symptoms (which include one or more of the following symptoms: faecal urgency; flatus incontinence; faecal incontinence);
-
perineal pain;
-
dyspareunia;
-
quality of life assessment.
Secondary outcomes
-
anal incontinence score.
Search methods for identification of studies
For the search methods used for previous versions of this review, please see Appendix 1.
Electronic searches
We searched the Cochrane Pregnancy and Childbirth Group's Trials Register by contacting the Trials Search Co‐ordinator (30 September 2013).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co‐ordinator and contains trials identified from:
-
monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
-
weekly searches of MEDLINE;
-
weekly searches of Embase;
-
handsearches of 30 journals and the proceedings of major conferences;
-
weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co‐ordinator searches the register for each review using the topic list rather than keywords.
Searching other resources
We searched reference lists of retrieved studies.
We did not apply any language restrictions.
Data collection and analysis
For the methods used when assessing the trials identified in the previous version of this review, seeFernando 2006.
For this update we used the following methods when assessing the reports identified by the updated search.
Selection of studies
Three review authors independently assessed and selected the trials for inclusion in this review. It was not possible to assess the relevance of the trials blinded because we knew the authors' names, institution, journal of publication and results, when we applied the inclusion criteria. We resolved any disagreement on eligibility for inclusion by discussion.
Ruwan Fernando, Abdul Sultan and Chris Kettle conducted a randomised controlled trial for overlap and end‐to‐end repair for obstetric anal sphincter injury (Fernando 2006), which was included in this update (2013). A fourth review author, Ranee Thakar, carried out assessments and data extraction of this study (Fernando 2006).
Data extraction and management
We used a standardised proforma to record data from the studies. Three review authors extracted data from eligible studies and any discrepancies were resolved through discussion or, if required, the fourth author was consulted. One review author entered data into Review Manager software (RevMan 2012) and a second author checked for accuracy.
When information regarding any of the included studies was unclear, we attempted to contact the authors of the original papers to provide further details.
Assessment of risk of bias in included studies
Three review authors independently assessed the risk of bias for each study using the criteria outlined in theCochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving the fourth assessor.
(1) Random sequence generation (checking for possible selection bias)
We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the method as:
-
low risk of bias (any truly random process, e.g. random number table; computer random number generator);
-
high risk of bias (any non‐random process, e.g. odd or even date of birth; hospital or clinic record number);
-
unclear risk of bias.
(2) Allocation concealment (checking for possible selection bias)
We described for each included study the method used to conceal allocation to interventions prior to assignment and will assess whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
-
low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
-
high risk of bias (open random allocation; unsealed or non‐opaque envelopes, alternation; date of birth);
-
unclear risk of bias.
(3.1) Blinding of participants and personnel (checking for possible performance bias)
The authors accept that it is impossible to blind the surgeon to the method of repair in these types of studies. However, the authors assessed the blinding of objective assessments such as endoanal ultrasonography and anal manometry at the time of performing the investigation and analysing the data, described below (3.2). We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods as:
-
low, high or unclear risk of bias for participants;
-
low, high or unclear risk of bias for personnel.
(3.2) Blinding of outcome assessment (checking for possible detection bias)
We have described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed methods used to blind outcome assessment as:
-
low, high or unclear risk of bias.
(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)
We have described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We have noted whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. We assessed methods as:
-
low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups);
-
high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation);
-
unclear risk of bias.
(5) Selective reporting (checking for reporting bias)
We have described for each included study how we investigated the possibility of selective outcome reporting bias and what we found.
We assessed the methods as:
-
low risk of bias (where it is clear that all of the study’s pre‐specified outcomes and all expected outcomes of interest to the review had been reported);
-
high risk of bias (where not all the study’s pre‐specified outcomes had been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest were reported incompletely and so could not be used; study failed to include results of a key outcome that would have been expected to have been reported);
-
unclear risk of bias.
(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)
We have noted for each included study any important concerns we had about other possible sources of bias.
We assessed whether each study was free of other problems that could put it at risk of bias:
-
low risk of other bias;
-
high risk of other bias;
-
unclear whether there is risk of other bias.
(7) Overall risk of bias
We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it was likely to impact on the findings. We planned to explore the impact of the level of bias through undertaking sensitivity analyses ‐ seeSensitivity analysis.
Measures of treatment effect
Dichotomous data
For dichotomous data, we have presented the results as summary risk ratio with 95% confidence intervals.
Continuous data
For continuous data, we used the mean difference if outcomes were measured in the same way between trials. We planned to use the standardised mean difference to combine trials measuring the same outcome, but use different methods.
Unit of analysis issues
Cluster‐randomised trials
We had planned to include cluster‐randomised trials in the analyses along with individually‐randomised trials, but we identified no such trials.
In future updates, if identified and eligible, we will include cluster‐randomised trials in the analyses along with individually‐randomised trials. We will adjust their sample sizes using the methods described in the Cochrane Handbook [Section 16.3.4] using an estimate of the intracluster correlation co‐efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster‐randomised trials and individually‐randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely.
We will also acknowledge heterogeneity in the randomisation unit and perform a sensitivity analysis to investigate the effects of the randomisation unit.
Cross‐over trials
We did not consider cross‐over trials would be feasible for this intervention and have not included such trials.
Dealing with missing data
For included studies, we noted levels of attrition in the Characteristics of included studies tables. We explored the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis.
For all outcomes we carried out analyses, as far as possible, on an intention‐to‐treat basis, i.e. we attempted to include all participants randomised to each group in the analyses. The denominator for each outcome in each trial is the number randomised minus any participants whose outcomes are known to be missing.
Assessment of heterogeneity
We examined the forest plots for the analyses visually to assess any obvious heterogeneity in terms of the size or direction of treatment effect between studies. We used the I² and Tau² statistics and the P value of the Chi² test for heterogeneity to quantify heterogeneity among the trials in each analysis. For those outcomes where we identified moderate or high unexplained heterogeneity (I² greater than 40%), we used a random‐effects model. We would advise that all findings where there are high levels of heterogeneity should be interpreted cautiously.
Assessment of reporting biases
In future updates of this review,if there are 10 or more studies included in the meta‐analysis, we plan to investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.
Data synthesis
We planned to carry out statistical analysis using the Review Manager software (RevMan 2012). We used fixed‐effect meta‐analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if substantial statistical heterogeneity was detected, we planned to use random‐effects meta‐analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. The random‐effects summary was treated as the average range of possible treatment effects and we planned to discuss the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we would not combine trials.
Where we used random‐effects analyses, the results were presented as the average treatment effect with 95% confidence intervals, and the estimates of Tau² and I².
Subgroup analysis and investigation of heterogeneity
If we had identified substantial heterogeneity, we planned to investigate it using subgroup analyses and sensitivity analyses. We would have considered whether an overall summary was meaningful, and if so, used a random‐effects analysis to produce it.
We did not carry out any subgroup analyses.
In future updates, if more studies are included, we plan to carry out the following subgroup analyses.
-
Type of tear (3a, 3b, 3c and 4th degree).
-
Type of suture material used.
-
Experience of the surgeon/operator.
The following outcomes will be used in subgroup analysis.
-
Perineal pain.
-
Dyspareunia.
-
Faecal urgency.
-
Flatus incontinence.
-
Faecal incontinence.
-
Anal incontinence score.
-
Quality of life assessment.
We will assess subgroup differences by interaction tests available within RevMan (RevMan 2012). We will report the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value.
Sensitivity analysis
We did not carry out a sensitivity analysis due to too few studies being included within analyses. In future updates, we plan to exclude studies with high risk of bias for allocation concealment or high levels of attrition to see if this has any impact on the results.
Results
Description of studies
Included studies
Six trials (Farrell 2012; Fernando 2006; Fitzpatrick 2000; Garcia 2005; Rygh 2010; Williams 2006) involving 588 women were included in this review. See Characteristics of included studies for details.
There were no quasi‐ or cluster‐randomised trials identified. In future updates, if identified, we would consider these study designs.
There were considerable variations in outcome measures amongst the six included studies. The main outcome measures of the Fitzpatrick 2000 study were symptoms of faecal incontinence, abnormal findings relating to anal manometry investigations and abnormal findings on endoanal ultrasonography at three months postpartum. The primary outcome measure of Fernando 2006 was symptoms of faecal incontinence at 12 months. Secondary outcome measures included faecal urgency, perineal pain, dyspareunia and quality of life at three, six and 12 months; faecal incontinence at six weeks, three and six months, anal manometry, endoanal scan findings and improvement of anal incontinence symptoms at 12 months. Main outcome measures of the Williams 2006 study were suture‐related morbidity at six weeks, bowel symptoms at three, six and 12 months assessed by a validated questionnaire, anorectal physiology at three months and quality of life scores assessed by a validated questionnaire at three and 12 months. The primary outcome of the Rygh 2010 study was leakage of solid stools once a week or more at 12 months. The secondary outcome measures of this study were leakage of liquid stools once a week or more, flatus incontinence once a week or more, dyspareunia once a month or more, mean Wexner score, continence status based on Wexner score, anal manometry and anal sphincter defects detected by endoanal ultrasonography at 12 months. The primary outcome of Farrell 2012 was flatal incontinence at six months whereas there secondary outcome measures were faecal incontinence, quality of life scales based on Rockwood 2000, anal manometry and anal sphincter defects detected by endoanal ultrasound scan at six months. The primary outcome measure of Garcia 2005 was not clear and the study reported anal incontinence symptoms, flatus incontinence and faecal incontinence based on faecal incontinence score (Sangalli 1994).
All six studies compared primary end‐to‐end and overlap repair techniques of external anal sphincter (EAS) performed immediately after obstetric anal sphincter injuries (OASIS). Fitzpatrick 2000 carried out a three‐month follow‐up, Farrell 2012 followed up for six, 12, 24 and 36 months whereas Fernando 2006, Rygh 2010 and Williams 2006 performed a 12‐month follow‐up. Follow‐up of the Garcia 2005 study ranged from four weeks to nine months (mean 3 +/‐ 2.5 months).
In addition to repair techniques, Williams 2006 compared two suture materials (polydioxanone and polyglactin) for the repair of EAS, but details of repair of perineal muscles, vaginal epithelium and perineal skin were not available. Fitzpatrick 2000 used long‐acting monofilament absorbable sutures (2‐0 polyglyconate ‐Maxon) for the repair of EAS and 2‐0 Dexon sutures were used to reconstruct the perineal body and suture the vaginal and perineal skin. Farrell 2012 used 3‐0 polyglyconate sutures for the EAS and internal anal sphincter (IAS) but the suture materials used for the perineal muscles and vaginal epithelium were not reported. Fernando 2006 and Rygh 2010 used 3‐0 polydioxanone for the repair of IAS and EAS. A continuous non‐locking method of repair with fast absorbing 2‐0 polyglactin sutures was used to repair the perineal muscles, vaginal epithelium and perineal skin. Garcia 2005 performed overlap repair using 2‐0 polydiaxanone sutures and end‐to‐end repair using 0 Polyglycolic acid sutures.
There is a considerable variation in inclusion criteria in these studies. Fitzpatrick 2000 and Williams 2006 included all women with partial or complete disruption of EAS, whereas Fernando 2006; Rygh 2010 and Farrell 2012 included only women with disruption of more than 50% thickness of EAS (grade 3b, 3c and 4th degree tears). In five of these women with grade 3b tears, the remaining fibres were divided in order to perform an overlap repair. Fitzpatrick 2000 and Williams 2006 performed the overlap in women with "partial EAS tears", whereas Rygh 2010 performed overlap repair of 3b tears without dividing the remaining fibres. Fitzpatrick 2000 and Williams 2006 did not specify whether the "partial EAS tears" were 3a or 3b tears. Farrell 2012 did not report whether the EAS was divided in 3b tears before the overlap repair.
In the Fitzpatrick 2000, Garcia 2005 and Williams 2006 studies, no attempt was made to identify and repair IAS separately. In contrast, Fernando 2006, Rygh 2010 and Farrell 2012 repaired the IAS if it was torn and recognised at the time of repair.
In all the studies the repairs were performed under regional or general anaesthesia, in the operating theatre under aseptic conditions with antibiotic cover and repairs were carried out by trained clinicians. Following repair, Fitzpatrick 2000 prescribed a codeine‐based constipating agent for three days followed by a laxative regimen for five days or until defecation occurred whereas Fernando 2006 and Williams 2006 prescribed stool bulking agent Ispaghula husk (Fybogel) and a stool softener (Lactulose) for 10 days after the repair. Rygh 2010 prescribed Lactulose during the hospital stay whereas Farrell 2012 did not report the use of laxatives or stool softeners. Garcia 2005 used stool softeners postpartum.
Two studies (Fernando 2006; Fitzpatrick 2000) used questionnaires and modified Wexner scoring system for anal incontinence (Jorge 1993; Vaizey 1999) for subjective assessment of symptoms during follow‐up. In addition, Fernando 2006, Rygh 2010 and Farrell 2012 used the Minneapolis Quality of life scoring system (Rockwood 2000). These scores range from zero to 24 where zero indicates complete continence and 24 indicates complete incontinence. Fitzpatrick 2000 used the questionnaire and scoring system at three months after the repair, whereas, Fernando 2006 used the questionnaires and scoring systems at six weeks, three, six and 12 months after the repair. In addition to the modified Wexner scoring system, Williams 2006 used the Manchester Health Questionnaire (Bugg 2001) to assess quality of life. Rygh 2010 used the quality of life score at 12 months, whereas Farrell 2012 used it at six months. Garcia 2005 used faecal incontinence score (FIS) based on Sangalli 1994 and considered FIS 1‐3 indicating only flatus incontinence FIS greater than three indication of faecal incontinence.
Five studies (Farrell 2012; Fernando 2006; Fitzpatrick 2000; Rygh 2010; Williams 2006) objectively assessed anal sphincter function using endoanal ultrasonography and manometry whereas Garcia 2005 used translabial ultrasound scan to assess the anal sphincter defects at mean follow‐up of three months. Fitzpatrick 2000 and Williams 2006 assessed at three months, Rygh 2010 at 12 months, Farrell 2012 at six, 12, 24 and 36 months and Fernando 2006 assessed at six and 12 months.
One study report (Farrell 2011) is an additional abstract of one included trial (Farrell 2012).
Excluded studies
Three trials were excluded because the repair of EAS was not performed immediately after OASIS but was regarded as secondary repair (Goh 2004; Johansson 2005; Lindqvist 2010). See Characteristics of excluded studies for details.
Ongoing studies
One study report (Ismail 2008) is an abstract of an ongoing study (Johanson 2001). See Characteristics of ongoing studies for details.
Risk of bias in included studies
Details of 'Risk of bias' assessments are summarised in Figure 1 and Figure 2.
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
Allocation
The methodological quality of four of the included studies (Farrell 2012; Fernando 2006; Rygh 2010; Williams 2006) was rated as low risk of bias for random sequence generation and allocation concealment. Fitzpatrick 2000; Rygh 2010; Garcia 2005 and Williams 2006 used sealed‐envelope randomisation, but only Garcia 2005; Williams 2006 and Rygh 2010 stated the envelopes were opaque. However in Garcia 2005, 9/51 (18%) envelopes were opened without randomisation and the authors could not give an explanation for this. Fernando 2006 and Farrell 2012 used a computer‐randomisation package. In Fernando 2006, only two clinicians who carried out the repairs had access to the password‐protected computer‐randomisation package. Once the details of the participants were entered, the computer programme generated the treatment allocation randomly. In Farrell 2012,all investigators and the statistician were blinded to the allocation code until the final analysis.
Blinding
The authors accept that it is impossible to blind the surgeon to the method of repair in these types of studies. However, the authors assessed the blinding of objective assessments such as endoanal ultrasonography and anal manometry at the time of performing the investigation and analysing the data. The blinding of the assessments carried out in the Fitzpatrick 2000 study was not clear. In the studies carried out by Fernando 2006 and Rygh 2010 the individuals who performed and reported the endoanal ultrasound images and the person who performed the anal manometry were blinded to the technique of repair. In Farrell 2012 all investigators and the statistician were blinded to the allocation code until the final analysis. The Williams 2006 study did not describe the blinding of outcomes assessors. In Garcia 2005, postpartum evaluators were blinded for the repair technique.
Incomplete outcome data
Twelve out of 64 (19%) women in the Fernando study (Fernando 2006), 43/103 (42%) women in the Williams study (Williams 2006), 8/174 (4%) in the Farrell 2012 study and 18/119 (15%) in the Rygh 2010 study were lost to follow‐up. Four studies (Farrell 2012; Fernando 2006; Rygh 2010; Williams 2006), clearly documented the steps taken to minimise the loss to follow‐up. In Garcia 2005, 15/41 (37%) women in the study were lost to follow‐up and the authors clearly documented the steps taken to minimise the loss to follow‐up.
Selective reporting
All except Fitzpatrick 2000 and Garcia 2005 have reported the steps taken to minimise the selective reporting bias.
Other potential sources of bias
No evidence of other bias in five of the studies was identified (Farrell 2012; Fernando 2006; Fitzpatrick 2000; Rygh 2010; Williams 2006). However Garcia 2005 has several potential sources of bias including: different types of sutures used in two techniques (overlap repair using 2‐0 polydiaxanone sutures and end‐to‐end repair using 0 Polyglycolic acid sutures); and including four women with flatal incontinence and one woman with solid stool incontinence prior to delivery in the analysis.
Five of the included studies (Farrell 2012; Fernando 2006; Fitzpatrick 2000; Rygh 2010; Williams 2006), have clearly described the method used to calculate the power of the study (number of participants necessary to achieve statistical significance). Fitzpatrick 2000 calculated that 110 women were required based on the assumption of a 30% difference in symptoms between the two techniques with 90% probability. Fernando 2006 calculated the power based on two previous observational studies (Sultan 1994b; Sultan 1999a) and required 64 women to demonstrate a reduction in anal incontinence from 46% to 8% with 90% probability. Williams 2006 calculated the power based on the assumption that absorbable sutures would reduce suture‐related morbidity from 30% to less than 1% compared with non‐absorbable sutures and that the overlapping technique would reduce suture‐related morbidity (defined as above) from 30% to less than 1% compared with end‐to‐end anastomoses. In the Farrell 2012 study, the power calculation was based on previous research carried out in the same centre where 47% of the primiparous women who sustained OASIS experienced flatus incontinence compared to a 7% flatus incontinence rate in the overlap group as reported by Sultan et al (Sultan 1999a). Rygh 2010 calculated the power based on two previous observational studies (Sultan 1994b; Sultan 1999a) to demonstrate a reduction in anal incontinence from 41% to 8%. However Garcia 2005 did not clearly mention the primary outcome measures.
Effects of interventions
Six trials involving 588 women were included in this review.
Comparison ‐ Overlap versus end‐to‐end
Primary outcomes
Faecal urgency
Only Fernando 2006 reported the incidence of faecal urgency at six weeks, three and 12 months. Fitzpatrick 2000 reported faecal urgency at three months. Farrell 2012, Rygh 2010 and Williams 2006 did not report the incidence of faecal urgency. At six weeks (risk ratio (RR) 1.09, 95% confidence interval (CI) 0.48 to 2.46, one trial, 63 women) and three months (RR 0.68, 95% CI 0.42 to 1.09, two trials, 172 women) there was no statistically significant difference in faecal urgency between the two repair techniques (Fernando 2006; Fitzpatrick 2000), Analysis 1.1. However, at six months (RR 0.22, 95% CI 0.05 to 0.94, one trial, 56 women) and 12 months (RR 0.12, 95% CI 0.02 to 0.86, one trial, 52 women) Fernando 2006 reported a statistically significant reduction in the incidence of faecal urgency in the overlap group compared to the end‐to‐end group (Fernando 2006), Analysis 1.1.
Flatus incontinence
Fitzpatrick 2000 and Williams 2006 did not report the incidence of flatus incontinence, whereas Fernando 2006 reported the incidence of flatus incontinence at six weeks (average RR 0.48, 95% CI 0.13 to 1.77, one trial, 63 women) and Fernando 2006 and Garcia 2005 reported flatus incontinence at three months (average RR 1.27, 95% CI 0.56 to 2.90, two trials, 101 women), Analysis 1.2. Both Fernando 2006 and Farrell 2012 reported flatus incontinence at six months (average RR 1.58, 95% CI 1.09 to 2.31, two trials, 205 women), Analysis 1.2. Farrell 2012, Fernando 2006 and Rygh 2010 reported flatus incontinence at 12 months (average RR 1.14, 95% CI 0.58 to 2.23, three trials, 256 women; Heterogeneity: Tau² = 0.20; Chi² = 4.83, df = 2 (P = 0.09); I² = 59%), Analysis 1.2. Farrell 2012 reported flatus incontinence at 24 months (average RR 1.11, 95% CI 0.74 to 1.69, one trial, 95 women) and 36 months (average RR 1.12, 95% CI 0.63 to 1.99, one trial, 68 women), Analysis 1.2. At six months there was a statistically significant difference in flatus incontinence, favouring the end‐to‐end group, but there were no differences in flatus incontinence between the two groups at any of the other time points.
Faecal incontinence, alteration in faecal continence
Fernando 2006 reported the incidence of faecal incontinence at six weeks, three, six and 12 months whereas Farrell 2012 reported faecal incontinence at six, 12, 24 and 36 months and Rygh 2010 reported faecal incontinence at 12 months. Garcia 2005 reported faecal incontinence at 3 months based on faecal incontinence score. Fitzpatrick 2000 reported "alteration in faecal continence" at three months. Hence we analysed the incidence of faecal incontinence from Fernando 2006, Rygh 2010 and Farrell 2012 and the incidence of "alteration in faecal continence" from the Fitzpatrick 2000 and Fernando 2006 studies separately. Williams 2006 did not report on the incidence of faecal incontinence or alteration of faecal continence.
Analysis of the incidence of faecal incontinence at six weeks (average RR 0.65, 95% CI 0.20 to 2.07, one trial, 63 women), three months (average RR 0.84, 95% CI 0.06 to 12.73, two trials, 101 women), six months (average RR 0.48, 95% CI 0.02 to 12.89, two trials, 205 women; Heterogeneity: Tau² = 4.60; Chi² = 4.88, df = 1 (P = 0.03); I² = 80%), 12 months (average RR 0.37, 95% CI 0.03 to 4.68, three trials, 256 women; Heterogeneity: Tau² = 3.68; Chi² = 7.60, df = 2 (P = 0.02); I² = 74%), 24 months (average RR 0.88, 95% CI 0.32 to 2.41, one trial, 95 women) and 36 months (average RR 1.01, 95% CI 0.34 to 2.98, one trial, 68 women), Analysis 1.3, did not show a statistically significant difference between the two repair techniques.
Analysis of the alteration in faecal continence, which included the Fitzpatrick 2000 and Fernando 2006 studies, also revealed similar results. There was no difference in the incidence of alteration in faecal continence at six weeks (RR 0.73, 95% CI 0.41 to 1.27, one trial, 63 women), three months (RR 0.85, 95% CI 0.64 to 1.14, two trials, 172 women), six months (RR 0.82, 95% CI 0.40 to 1.66, one trial, 56 women) and at 12 months (RR 0.46, 95% CI 0.18 to 1.17, one trial, 52 women), Analysis 1.4.
One or more anal incontinence symptoms (Faecal urgency, flatus incontinence, faecal incontinence, alteration in faecal continence) ‐ not pre‐specified outcome
With regards to anal incontinence symptoms, different studies have defined a mixture of anal incontinence symptoms as their outcome measures (e.g. main outcome measures of the Fitzpatrick 2000 study were "symptoms of faecal incontinence"). The primary outcome measure of the Fernando 2006 study was symptoms of faecal incontinence at 12 months. Secondary outcome measures included faecal urgency, perineal pain, at three, six and 12 months; faecal incontinence at six weeks, three and six months. The primary outcome of the Rygh 2010 study was leakage of solid stools once a week or more at 12 months. The secondary outcome measures of this study were leakage of liquid stools once a week or more and flatus incontinence once a week or more. The primary outcome of Farrell 2012 was flatal incontinence at six months whereas the secondary outcome measure was faecal incontinence. Because of this heterogeneity of reported anal incontinence symptoms outcomes, the authors have collated one or more anal incontinence symptoms reported in each trial at six weeks, three, six, 12, 24 and 36 months and reported as a separate outcome measure (One or more anal incontinence symptoms, Analysis 1.5).
The analysis found that there was no difference in one or more anal incontinence symptoms at six weeks (average RR 0.75, 95% CI 0.49 to 1.16, one trial, 252 women), three months (average RR 0.80, 95% CI 0.63 to 1.03, three trials, 505 women), six months (average RR 1.00, 95% CI 0.38 to 2.62, two trials, 522 women; Heterogeneity: Tau² = 0.43; Chi² = 8.54, df = 1 (P = 0.003); I² = 88%), 12 months (average RR 0.73, 95% CI 0.24 to 2.20, three trials, 616 women; Heterogeneity: Tau² = 0.85; Chi² = 19.36, df = 2 (P < 0.0001); I² = 90%), 24 months (average RR 1.06, 95% CI 0.69 to 1.61, one trial, 190 women), or at 36 months (average RR 1.08, 95% CI 0.63 to 1.85, one trial, 136 women), Analysis 1.5.
Deterioration of anal incontinence symptoms
Out of the six included studies only Fernando 2006 collected data regarding improvement or deterioration of anal incontinence symptoms at 12 months. The meta‐analysis showed that statistically significant fewer numbers of participants in the overlap group reported deterioration of anal incontinence symptoms at 12 months postpartum (RR 0.26, 95% CI 0.09 to 0.79, one trial, 41 women), Analysis 1.6.
Perineal pain
It was somewhat difficult to compare the degree of perineal pain in three included studies because of different definitions and different time intervals. Fitzpatrick 2000 sub‐divided perineal pain into perineal discomfort and need for perineal injection with local anaesthetic at three months. Fernando 2006 reported rates of perineal pain at six weeks, three, six and 12 months. Williams 2006 reported perineal pain and dyspareunia together at six weeks, three, six, and 12 months. Farrell 2012 and Rygh 2010 did not report perineal pain as an outcome measure in their studies.
Meta‐analysis of perineal pain at six weeks (RR 0.97, 95% CI 0.42 to 2.26, one trial, 63 women), three months (RR 0.85, 95% CI 0.54 to 1.34, two trials, 172 women), six months (RR 0.25, 95% CI 0.03 to 2.10, one trial, 56 women) and 12 months (RR 0.08, 95% CI 0.00 to 1.45, one trial, 52 women) showed no statistically significant difference between overlap and end‐to‐end techniques, Analysis 1.7. There was also no statistically significant difference in the need for perineal injection with a local anaesthetic between the two repair techniques (RR 0.96, 95% CI 0.48 to 1.91, one trial, 112 women), Analysis 1.8. Combined perineal pain and dyspareunia at six weeks (RR 0.71, 95% CI 0.27 to 1.84, one trial, 103 women), three months (RR 1.79, 95% CI 0.83 to 3.84, one trial, 89 women), six months (RR 0.38, 95% CI 0.11 to 1.36, one trial, 79 women) and 12 months (RR 0.35, 95% CI 0.07 to 1.66, one trial, 60 women) also did not show a statistically significant difference between the two repair techniques, Analysis 1.9.
Dyspareunia
Only three studies reported dyspareunia (Fernando 2006 at three, six and 12 months, Fitzpatrick 2000 at three months and Rygh 2010 at 12 months). There was no statistically significant difference in dyspareunia at three months (average RR 0.82, 95% CI 0.26 to 2.61, two trials, 172 women; Heterogeneity: Tau² = 0.53; Chi² = 4.24, df = 1 (P = 0.04); I² = 76%), six months (average RR 0.86, 95% CI 0.33 to 2.23, one trial, 56 women) and 12 months (average RR 0.77, 95% CI 0.48 to 1.24, two trials, 151 women) between the overlap and end‐to‐end groups, Analysis 1.10.
Quality of life assessment
Only Fernando 2006 reported quality of life assessment following the two repair techniques from six weeks to 12 months based on the scales described by Rockwood (Rockwood 2000). There are four sub scales, addressing lifestyle (effect of incontinence on day‐to‐day activity), coping and behaviour (effect of incontinence on ability to cope with the situation and resulting behaviour), depression and self‐perception (effect of incontinence on feelings and self‐confidence), and embarrassment (incontinence causing embarrassment). Each scale ranges from one to five, one indicating lower functional status of quality of life. Analysis of quality of life at 12 months based on four scales showed no statistically significant difference between overlap and end‐to‐end in mean lifestyle scale (mean difference (MD) 0.06, 95% CI ‐0.23 to 0.35, one trial, 52 women) Analysis 1.11, mean coping and behaviour scale (MD 0.13, 95% CI ‐0.24 to 0.50, one trial, 52 women), Analysis 1.12, mean depression scale (MD 0.00, 95% CI ‐0.26 to 0.26, one trial, 52 women), Analysis 1.13, and mean embarrassment scale (MD 0.20, 95% CI ‐0.14 to 0.54, one trial, 52 women), Analysis 1.14.
Secondary outcomes
Anal incontinence score
Three studies (Fitzpatrick 2000, Fernando 2006 and Rygh 2010) reported modified Wexner anal incontinence scores based on Vaizey 1999. Fernando 2006 reported mean anal incontinence scores at six weeks, three, six and 12 months, whereas Fitzpatrick 2000 reported median anal incontinence scores at three months. Rygh 2010 reported mean anal incontinence score at 12 months but the standard deviations were not reported. Hence anal incontinence scores from Fitzpatrick 2000 and Rygh 2010 were not included in the meta analysis.There was no statistically significant difference in the mean anal incontinence score at six weeks, three and six months between the two repair techniques. However, meta‐analysis at 12 months showed a statistically significant lower anal incontinence score in the overlap group suggestive of better continence (standardised mean difference (SMD) ‐0.70, 95% CI ‐1.26 to ‐0.14), Analysis 1.15. This result was based on the Fernando 2006 study where the confidence intervals were wide suggesting skewness of data, hence this finding should be treated with caution.
When analysing the overall number of participants with an anal incontinence score of more than 10 (that is, clinically significant anal incontinence) at 12 months, there was no difference in anal incontinence scores between the two groups (RR 0.31, 95% CI 0.03 to 2.91, three trials 211 women), Analysis 1.16, (Fernando 2006; Rygh 2010; Williams 2006).
Discussion
There are six included randomised controlled studies comparing immediate primary overlap and end‐to‐end techniques for OASIS. However, there was considerable heterogeneity in the outcome measures, time points and reported results among these studies, which made it difficult to group them into similar outcome groups. Two studies have similar design and outcome measures (Fernando 2006; Rygh 2010).
Meta‐analysis showed that there was no statistically significant difference in perineal pain and dyspareunia, flatus incontinence and faecal incontinence between the two repair techniques at six weeks, three, six and 12 months after the repair. Only one study (Fernando 2006), separately analysed faecal urgency symptoms and showed a statistically significant lower incidence in faecal urgency in the overlap group at 12 months. When addressing the outcome of faecal incontinence by mean anal incontinence scores and deterioration of anal incontinence symptoms at 12 months, the overlap group showed significantly lower relative risks compared to the end‐to‐end group.
The authors noted that the majority of weight in the meta‐analysis was based on only one study where repairs were carried out by only two experienced clinicians (Fernando 2006). Farrell 2012 and Rygh 2010 reported that the surgeons were trained in both techniques and were appropriately supervised. The effect of the surgeon's experience in the outcome of repair technique is not addressed in other included studies.
None of the studies reported the incidence of re‐repair and women's view with regards to the type of repair and subsequent outcome.
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Overlap versus end‐to‐end, Outcome 1 Faecal urgency.
Comparison 1 Overlap versus end‐to‐end, Outcome 2 Flatus incontinence.
Comparison 1 Overlap versus end‐to‐end, Outcome 3 Faecal incontinence.
Comparison 1 Overlap versus end‐to‐end, Outcome 4 Alteration in faecal continence.
Comparison 1 Overlap versus end‐to‐end, Outcome 5 One or more anal incontinence symptoms (Faecal urgency, flatus incontinence, faecal incontinence, alteration in faecal continence) not prespecified outcome.
Comparison 1 Overlap versus end‐to‐end, Outcome 6 Deterioration of anal incontinence symptoms.
Comparison 1 Overlap versus end‐to‐end, Outcome 7 Perineal pain.
Comparison 1 Overlap versus end‐to‐end, Outcome 8 Need for perineal injection at 3 months.
Comparison 1 Overlap versus end‐to‐end, Outcome 9 Perineal pain/dyspareunia.
Comparison 1 Overlap versus end‐to‐end, Outcome 10 Dyspareunia.
Comparison 1 Overlap versus end‐to‐end, Outcome 11 Quality of life scale 1: lifestyle.
Comparison 1 Overlap versus end‐to‐end, Outcome 12 Quality of life scale 2: coping and behaviour.
Comparison 1 Overlap versus end‐to‐end, Outcome 13 Quality of life scale 3: depression and self‐perception.
Comparison 1 Overlap versus end‐to‐end, Outcome 14 Quality of life scale 4: embarrassment.
Comparison 1 Overlap versus end‐to‐end, Outcome 15 Anal incontinence score.
Comparison 1 Overlap versus end‐to‐end, Outcome 16 Anal incontinence score > 10.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Faecal urgency Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Faecal urgency at 6 weeks | 1 | 63 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.09 [0.48, 2.46] |
| 1.2 Faecal urgency at 3 months | 2 | 172 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.68 [0.42, 1.09] |
| 1.3 Faecal urgency at 6 months | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.22 [0.05, 0.94] |
| 1.4 Faecal urgency at 12 months | 1 | 52 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.12 [0.02, 0.86] |
| 2 Flatus incontinence Show forest plot | 4 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 2.1 Flatus incontinence at 6 weeks | 1 | 63 | Risk Ratio (M‐H, Random, 95% CI) | 0.48 [0.13, 1.77] |
| 2.2 Flatus incontinence at 3 months | 2 | 101 | Risk Ratio (M‐H, Random, 95% CI) | 1.27 [0.56, 2.90] |
| 2.3 Flatus incontinence at 6 months | 2 | 205 | Risk Ratio (M‐H, Random, 95% CI) | 1.58 [1.09, 2.31] |
| 2.4 Flatus incontinence at 12 months | 3 | 256 | Risk Ratio (M‐H, Random, 95% CI) | 1.14 [0.58, 2.23] |
| 2.5 Flatus incontinence at 24 months | 1 | 95 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.74, 1.69] |
| 2.6 Flatus incontinence at 36 months | 1 | 68 | Risk Ratio (M‐H, Random, 95% CI) | 1.12 [0.63, 1.99] |
| 3 Faecal incontinence Show forest plot | 4 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 3.1 Faecal incontinence at 6 weeks | 1 | 63 | Risk Ratio (M‐H, Random, 95% CI) | 0.65 [0.20, 2.07] |
| 3.2 Faecal incontinence at 3 months | 2 | 101 | Risk Ratio (M‐H, Random, 95% CI) | 0.84 [0.06, 12.73] |
| 3.3 Faecal incontinence at 6 months | 2 | 205 | Risk Ratio (M‐H, Random, 95% CI) | 0.48 [0.02, 12.89] |
| 3.4 Faecal incontinence at 12 months | 3 | 256 | Risk Ratio (M‐H, Random, 95% CI) | 0.37 [0.03, 4.68] |
| 3.5 Faecal incontinence at 24 months | 1 | 95 | Risk Ratio (M‐H, Random, 95% CI) | 0.88 [0.32, 2.41] |
| 3.6 Faecal incontinence at 36 months | 1 | 68 | Risk Ratio (M‐H, Random, 95% CI) | 1.01 [0.34, 2.98] |
| 4 Alteration in faecal continence Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 4.1 Alteration in faecal continence at 6 weeks | 1 | 63 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.73 [0.41, 1.27] |
| 4.2 Alteration in faecal continence at 3 months | 2 | 172 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.85 [0.64, 1.14] |
| 4.3 Alteration in faecal continence at 6 months | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.82 [0.40, 1.66] |
| 4.4 Alteration in faecal continence at 12 months | 1 | 52 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.46 [0.18, 1.17] |
| 5 One or more anal incontinence symptoms (Faecal urgency, flatus incontinence, faecal incontinence, alteration in faecal continence) not prespecified outcome Show forest plot | 5 | 2221 | Risk Ratio (M‐H, Random, 95% CI) | 0.90 [0.68, 1.17] |
| 5.1 One or more anal incontinence symptoms at 6 weeks | 1 | 252 | Risk Ratio (M‐H, Random, 95% CI) | 0.75 [0.49, 1.16] |
| 5.2 One or more anal incontinence symptoms at 3 months | 3 | 505 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.63, 1.03] |
| 5.3 One or more anal incontinence symptoms at 6 months | 2 | 522 | Risk Ratio (M‐H, Random, 95% CI) | 1.00 [0.38, 2.62] |
| 5.4 All anal incontinence symptoms at 12 months | 3 | 616 | Risk Ratio (M‐H, Random, 95% CI) | 0.73 [0.24, 2.20] |
| 5.5 One or more anal incontinence symptoms at 24 months | 1 | 190 | Risk Ratio (M‐H, Random, 95% CI) | 1.06 [0.69, 1.61] |
| 5.6 One or more anal incontinence symptoms at 36 months | 1 | 136 | Risk Ratio (M‐H, Random, 95% CI) | 1.08 [0.63, 1.85] |
| 6 Deterioration of anal incontinence symptoms Show forest plot | 1 | 41 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.26 [0.09, 0.79] |
| 7 Perineal pain Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 7.1 Perineal pain at 6 weeks | 1 | 63 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.97 [0.42, 2.26] |
| 7.2 Perineal pain at 3 months | 2 | 172 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.85 [0.54, 1.34] |
| 7.3 Perineal pain at 6 months | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.25 [0.03, 2.10] |
| 7.4 Perineal pain at 12 months | 1 | 52 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.08 [0.00, 1.45] |
| 8 Need for perineal injection at 3 months Show forest plot | 1 | 112 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.96 [0.48, 1.91] |
| 9 Perineal pain/dyspareunia Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 9.1 Perineal pain/dyspareunia at 6 weeks | 1 | 103 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.71 [0.27, 1.84] |
| 9.2 Perineal pain/dyspareunia at 3 months | 1 | 89 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.79 [0.83, 3.84] |
| 9.3 Perineal pain/dyspareunia at 6 months | 1 | 79 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.38 [0.11, 1.36] |
| 9.4 Perineal pain/dyspareunia at 12 months | 1 | 60 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.35 [0.07, 1.66] |
| 10 Dyspareunia Show forest plot | 3 | Risk Ratio (M‐H, Random, 95% CI) | Subtotals only | |
| 10.1 Dyspareunia at 3 months | 2 | 172 | Risk Ratio (M‐H, Random, 95% CI) | 0.82 [0.26, 2.61] |
| 10.2 Dyspareunia at 6 months | 1 | 56 | Risk Ratio (M‐H, Random, 95% CI) | 0.86 [0.33, 2.23] |
| 10.3 Dyspareunia at 12 months | 2 | 151 | Risk Ratio (M‐H, Random, 95% CI) | 0.77 [0.48, 1.24] |
| 11 Quality of life scale 1: lifestyle Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 11.1 Mean lifestyle scale at 6 weeks | 1 | 63 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.12, 0.32] |
| 11.2 Mean lifestyle scale at 3 months | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.19, 0.39] |
| 11.3 Mean lifestyle scale at 6 months | 1 | 56 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [‐0.20, 0.20] |
| 11.4 Mean lifestyle scale at 12 months | 1 | 52 | Mean Difference (IV, Fixed, 95% CI) | 0.06 [‐0.23, 0.35] |
| 12 Quality of life scale 2: coping and behaviour Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 12.1 Mean coping and behaviour scale at 6 weeks | 1 | 63 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.19, 0.39] |
| 12.2 Mean coping and behaviour scale at 3 months | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.19, 0.39] |
| 12.3 Mean coping and behaviour scale at 6 months | 1 | 56 | Mean Difference (IV, Fixed, 95% CI) | 0.20 [‐0.06, 0.46] |
| 12.4 Mean coping and behaviour scale at 12 months | 1 | 52 | Mean Difference (IV, Fixed, 95% CI) | 0.13 [‐0.24, 0.50] |
| 13 Quality of life scale 3: depression and self‐perception Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 13.1 Mean depression and self‐perception scale at 6 weeks | 1 | 63 | Mean Difference (IV, Fixed, 95% CI) | 0.30 [0.05, 0.55] |
| 13.2 Mean depression and self‐perception scale at 3 months | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [‐0.23, 0.23] |
| 13.3 Mean depression and self‐perception scale at 6 months | 1 | 56 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.11, 0.31] |
| 13.4 Mean depression and self‐perception scale at 12 months | 1 | 52 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [‐0.26, 0.26] |
| 14 Quality of life scale 4: embarrassment Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 14.1 Mean embarrassment scale at 6 weeks | 1 | 63 | Mean Difference (IV, Fixed, 95% CI) | 0.30 [0.06, 0.54] |
| 14.2 Mean embarrassment scale at 3 months | 1 | 60 | Mean Difference (IV, Fixed, 95% CI) | 0.20 [‐0.08, 0.48] |
| 14.3 Mean embarrassment scale at 6 months | 1 | 56 | Mean Difference (IV, Fixed, 95% CI) | 0.20 [‐0.04, 0.44] |
| 14.4 Mean embarrassment scale at 12 months | 1 | 52 | Mean Difference (IV, Fixed, 95% CI) | 0.20 [‐0.14, 0.54] |
| 15 Anal incontinence score Show forest plot | 1 | Std. Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 15.1 Anal incontinence score at 6 weeks | 1 | 63 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.37 [‐0.87, 0.13] |
| 15.2 Anal incontinence score at 3 months | 1 | 60 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.19 [‐0.70, 0.32] |
| 15.3 Anal incontinence score at 6 months | 1 | 56 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.47 [1.00, 0.07] |
| 15.4 Anal incontinence score at 12 months | 1 | 52 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.70 [‐1.26, ‐0.14] |
| 16 Anal incontinence score > 10 Show forest plot | 4 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 16.1 Anal incontinence score > 10 at 6 weeks | 2 | 166 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.34 [0.06, 2.05] |
| 16.2 Anal incontinence score > 10 at 3 months | 3 | 275 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.43 [0.13, 1.46] |
| 16.3 Anal incontinence score > 10 at 6 months | 2 | 137 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.18 [0.02, 1.53] |
| 16.4 Anal incontinence score > 10 at 12 months | 3 | 211 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.31 [0.03, 2.91] |
