To explore: (1) the antenatal and postnatal morbidity experienced by women in relation to operative vaginal delivery (OVD); and (2) the impact of restrictive versus routine use of episiotomy.
To explore: (1) the antenatal and postnatal morbidity experienced by women in relation to operative vaginal delivery (OVD); and (2) the impact of restrictive versus routine use of episiotomy.
Longitudinal prospective cohort study embedded within a two-centre randomised controlled trial (RCT).
Two UK tertiary-level maternity units.
Antenatally recruited participants of an RCT of restrictive versus routine use of episiotomy at OVD.
A self-completing questionnaire was administered antenatally, before hospital discharge, at 6 weeks and at 1 year postpartum.
Urinary and anal incontinence, dyspareunia, perineal pain and psychological morbidity.
Longitudinal data have revealed that morbidities historically associated with OVD were often as prevalent, if not more prevalent, in the third trimester of pregnancy than postpartum. Restrictive episiotomy use was associated with: a higher incidence of perineal pain in the immediate postpartum period (98.9% restrictive versus 87.8% routine, RR 1.10, 95% CI 1.01–1.21); greater psychological morbidity in the immediate postpartum period (mean scores on the Edinburgh Postnatal Depression Scale, Edinburgh Postnatal Depression Score (EPDS) 6.7 restrictive versus 5.1 routine; P = 0.01 ); and more stress urinary incontinence at 6 weeks postpartum (42.2% restrictive versus 27.2% routine, RR 1.55, 95% CI 1.00–2.40); however, this had resolved by 1 year. No other differences were found between the groups at 6 weeks and 1 year postpartum.
Morbidities previously attributed to OVD may in fact be present antenatally, to a greater or similar degree. A restrictive approach to the use of episiotomy at OVD may increase rates of urinary morbidity, in particular stress incontinence and perineal pain, in the immediate postpartum period.
Historically, episiotomy has been an element of vaginal delivery, with the rationale of preventing extensive perineal tearing extending to include the involvement of the anal sphincter and damage to the pelvic floor. It was thus thought to be protective of maternal morbidities such as perineal pain, sexual dysfunction, and urinary and fecal incontinence following a difficult vaginal delivery.[1-4] These morbidities may have a profound impact on a woman's recovery in the puerperium, and on subsequent health and psychological wellbeing.
The routine use of episiotomy at spontaneous vaginal delivery has been discredited, with little benefit being demonstrated in pelvic floor morbidity; however, episiotomy is still frequently routinely employed at operative vaginal deliveries (OVDs).
Operative vaginal delivery and extensive perineal tears have long been recognised as major risk factors for subsequent pelvic floor morbidity, but little attention has been given to the specific role of episiotomy in this context. Our previously reported two-centred randomised controlled trial (RCT) of restrictive versus routine approach to episiotomy at in OVD suggested that a restrictive approach neither reduced nor greatly increased the incidence of anal sphincter tears. In light of this tentative evidence, we carried out a prospective cohort study nested within this RCT to: (1) examine longitudinal data collected prospectively from the antenatal period up to 1 year postpartum, investigating the incidence of pelvic floor morbidities and the psychological impact on women's lives following forceps and vacuum delivery; and (2) investigate the impact of employing restrictive versus routine use of episiotomy at OVD on these morbidities.
The cohort comprised 200 nulliparous women recruited antenatally and randomised within the previously described RCT of restrictive versus routine use of episiotomy at OVD. Women were either randomised to have an episiotomy performed restrictively (only if tearing became apparent or the operator considered an episiotomy necessary) or routinely (in all cases). Right mediolateral episiotomy was universally employed, and was performed in 46% of cases in the restrictive arm versus 80% of cases in the routine arm.
Questionnaires were devised using validated tools identified in the literature. Urinary continence was assessed using a shortened Bristol Lower Urinary Tract Symptom (BLUTS) questionnaire. Bowel function was assessed using the Wexner score. Fecal urgency was defined as an inability to defer defecation for longer than 5 minutes. Perineal pain was assessed by the McGill pain questionnaire. Psychological wellbeing was assessed using the Edinburgh Postnatal Depression Scale (EPDS). The recommended cut-off of 13 on the EPDS was taken to signify possible depression. Perineal assessment using the REEDA scale (redness, edema, ecchymosis, discharge and approximation) was made visually prior to hospital discharge.
Baseline data on urinary and bowel morbidity, dyspareunia and psychological wellbeing were collected by self-completed questionnaire in the third trimester, usually at the time of consenting to participation in the RCT. Clinical follow-up of the woman was completed prior to hospital discharge by a visual inspection of the perineum and the administration of a self-completed questionnaire investigating perineal pain and psychological wellbeing. At 6 weeks and 1 year postpartum a postal questionnaire was sent to elicit maternal health outcomes. Data were obtained on urinary continence, bowel function, dyspareunia, perineal pain and psychological wellbeing. A reminder letter and enclosed questionnaire were sent after 4 weeks to non-responders.
Urinary outcome measures included: urinary urgency and frequency; urge urinary incontinence; stress urinary incontinence; and reduced urinary sensation. Outcome measures of anal incontinence included: incontinence of flatus; incontinence of liquids; and incontinence of solids. In addition, the incidence of dyspareunia, perineal pain and psychological morbidity was investigated.
The statistical package spss 17.0 (SPSS, Chicago, IL, USA) and Confidence Interval Analysis (CIA) were used for analysis. Descriptive statistics were used to investigate the comparability of the respondents at baseline and at 1 year to establish if the smaller number of women returning the questionnaire at this time point was unrepresentative of the study population as a whole.
The first analyses comprised a longitudinal comparison of urinary, anal and sexual outcomes, perineal pain and psychological morbidity at appropriate time points for the entire cohort, irrespective of the use of episiotomy. The second analyses investigated the morbidity outcomes according to the approach to episiotomy use at delivery. Analysis was performed using McNemar's test (a paired-sample equivalent of the chi-square test for differences in proportions) for data from women who provided data at each time point and the Student's t-test for differences in means. Significance is taken as P < 0.05. Results are presented as relative risks (RRs) or differences in means and 95% confidence intervals (95% CIs).
Baseline data were obtained in the antenatal period from 99% of women. Follow-up in the immediate postpartum period was achieved in 93% of cases. Response rates were 83% at 6 weeks postpartum and 55% at 1 year postpartum (Figure 1). Response rates were balanced between the randomised study arms at each time point. No statistical differences were found in the maternal, labour, delivery and neonatal characteristics of respondents at baseline, and at 1 year postpartum (Table 1).
|Antenatal n = 198||1st/2nd day postpartum n = 184||6 weeks postpartum n = 164||1 year postpartum n = 108||P a|
|Maternal age > 35 years (%)||22 (11.1)||21 (11.4)||19 (11.6)||13 (12.0)||0.61|
|Body mass index > 30 (%)b||31 (15.9)||29 (15.9)||28 (17.3)||18 (17.0)||0.60|
|Birthweight (g) Mean (SD) (range)||3570 (514) (2250–5060)||3563 (513) (2250–5060)||3587 (518) (2250–5060)||3560 (529) (2250–5060)||0.78|
|Head circumference (cm) Mean (SD) (range)||35.1 (1.3) (31.5–38.1)||35.1 (1.3) (31.5–38.1)||35.1 (1.3) (31.5–38.1)||35.1 (1.3) (31.5–38.1)||0.97|
|Fetal malposition (%)c||73 (36.9)||66 (35.9)||56 (34.1)||41 (38.0)||0.64|
|Vacuum delivery (%)||47 (23.7)||45 (24.4)||33 (20.1)||20 (18.6)||0.07|
|Forceps delivery (non-rotational) (%)||115 (58.1)||106 (57.6)||97 (59.2)||64 (59.3)||0.70|
|Forceps delivery (rotational) (%)||12 (6.1)||11 (6.0)||12 (7.3)||9 (8.3)||0.13|
|SVD (%)||12 (6.1)||11 (6.0)||12 (7.3)||6 (5.6)||0.84|
|Caesarean section (%)||12 (6.1)||11 (6.0)||10 (6.1)||9 (8.3)||0.13|
Of the 104 women who had reported urinary incontinence of any type during pregnancy, 55 (66%) had persistent urinary incontinence of any type at 6 weeks postpartum, whereas new urinary incontinence of any type in the puerperium was reported by 24 (30%) women who had been continent in pregnancy. Urgency of micturition was reported more often in pregnancy than at 6 weeks postpartum (74.7% antenatal versus 50.0% 6 weeks; P < 0.001), with no further decrease at 1 year postpartum. Urge urinary incontinence was experienced by up to 37% of women, but with no significant differences across the time points examined. Stress urinary incontinence was reported with a similar prevalence at all time points (41.4% antenatal versus 34.8% 6 weeks; P = 0.34). Reduced urinary sensation was more prevalent in pregnancy than at either 6 weeks or 1 year postpartum, but differences did not reach statistical significance (13.2% versus 9.8 and 5.7%, respectively; Table 2).
|Antenatal n = 198||6 weeks postpartum n = 164||P b , c||1 year postpartum n = 108||P b , d|
|Urgency of micturition (%)||148 (74.7)||82 (50.0)||<0.001a||52 (49.5)||1.00|
|Urge urinary incontinence (%)||70 (35.4)||57 (34.8)||0.67||39 (37.1)||0.82|
|Stress urinary incontinence (%)||82 (41.4)||57 (34.8)||0.34||49 (46.7)||0.02a|
|Reduced urinary sensation (%)||26 (13.2)||16 (9.8)||0.83||6 (5.7)||0.23|
|Anal incontinence of flatus (%)||104 (52.5)||70 (42.9)||0.04a||38 (35.2)||0.54|
|Anal incontinence of liquids (%)||9 (4.5)||10 (6.1)||0.29||6 (5.5)||0.69|
|Anal incontinence of solids (%)||2 (1.0)||8 (4.9)||0.02a||0 (0)||0.03a|
|Urgency of defecation (%)e||14 (7.1)||13 (7.9)||1.00||8 (7.5)||1.00|
|Moderate/severe dyspareunia (%)||8 (4.1)||12 (9.6)||0.18||5 (4.7)||0.34|
|Dyspareunia preventing intercourse (%)||18 (9.3)||29 (17.8)||0.02a||9 (9.1)||0.01a|
Incontinence of flatus was more prevalent in pregnancy than the postpartum period (52.5% antenatal versus 42.9% 6 weeks; P = 0.04). Anal incontinence of liquids was found to be uncommon but constant across time points, ranging from 4.5 to 6.0%. Anal incontinence of solids was reported by 1% of women in pregnancy, but was a significantly greater problem at 6 weeks postpartum (4.9%; P = 0.02); however, this had resolved in all cases by 1 year.
Moderate or severe dyspareunia was most reported at 6 weeks postpartum (4.1% antenatal versus 9.6% 6 weeks; P = 0.18), but by 1 year postpartum the incidence had decreased to baseline levels. This finding was reflected in the significantly higher levels of dyspareunia having prevented intercourse at 6 weeks postpartum (9.3% antenatal versus 17.8% 6 weeks; P = 0.02), which had also returned to baseline levels by 1 year postpartum.
All measures of perineal pain were found to reduce significantly with time (Table S1). Perineal pain was present for nearly all women in the immediate postpartum period (96.7%), but was reported for less than half of respondents at 6 weeks (43.9%). For nearly one in five women perineal pain persisted to 1 year postpartum (17.8%). In addition, the intensity of the pain was found to be more severe on the first or second day than at 6 weeks postpartum [mean Visual Analogue Score (VAS) 35 versus 6, respectively; mean McGill pain rating index (PRI) 11.9 versus 2.0 and mean McGill present pain intensity (PPI) score 1.8 versus 0.4]. On average, women chose seven words from the McGill questionnaire to describe their pain in the immediate postpartum period, compared with only one at 6 weeks postpartum.
The mean EPDS score, a screening test for psychological ill health, was significantly higher in pregnancy than on the first or second day postpartum (7.4 versus 5.0; P < 0.001), then remained constant up to 1 year postpartum (Table S2). Likewise, the prevalence of an EPDS score ≥13, suggestive of clinical depression, was significantly higher antenatally compared with the first or second day postpartum (14.5 versus 6.5%, respectively, P = 0.02), remaining constant at 6 weeks and 1 year postpartum.
No differences were found between the restrictive and routine episiotomy groups for any urinary, anal or sexual morbidity at baseline. New urinary incontinence postpartum was reported by 17 women (17%) randomised to restrictive use of episiotomy, compared to seven women (7%) randomised to routine use of episiotomy (RR 2.37, 95% CI 1.04–5.41). No differences were found between groups in the rates of urinary morbidities at 6 weeks or 1 year postpartum, other than stress urinary incontinence, which was found to be significantly more likely with restrictive use of episiotomy than with routine use at 6 weeks postpartum (42.2% restrictive versus 27.2% routine, RR 1.55, 95% CI 1.00–2.40; Table 3); however, this difference was no longer present at 1 year postpartum. At 6 weeks postpartum, women complaining of urinary morbidities ‘most or all of the time’ were uncommon: urgency of micturition or urge urinary incontinence postpartum (≤2%) and stress urinary incontinence (5.5% restrictive use versus 1.2% routine use).
|Antenatal||6 weeks postpartum||1 year postpartum|
|Routine n = 98||Restrictive n = 100||RR (95% CI)b||Routine n = 81||Restrictive n = 83||RR (95% CI)b||Routine n = 52||Restrictive n = 56||RR (95% CI)b|
|Urgency of micturition (%)||78 (79.6)||70 (70.0)||0.88 (0.75–1.04)||39 (48.1)||43 (51.8)||1.08 (0.79–1.46)||26 (52.0)||26 (47.3)||0.93 (0.63–1.37)|
|Urge incontinence(%)||32 (32.7)||38 (38.0)||1.16 (0.80–1.70)||25 (30.9)||32 (38.6)||1.25 (0.82–1.91)||21 (42.0)||18 (32.7)||0.80 (0.48–1.32)|
|Stress incontinence (%)||40 (40.8)||42 (42.0)||1.03 (0.74–1.43)||22 (27.2)||35 (42.2)||1.55 (1.00–2.40)a||25 (50.0)||24 (43.6)||0.89 (0.59–1.35)|
|Reduced urinary sensation (%)||12 (12.2)||14 (14.0)||1.14 (0.56–2.35)||6 (7.4)||10 (12.0)||1.63 (0.62–4.27)||4 (8.0)||2 (3.6)||0.46 (0.09–2.43)|
|Incontinence of flatus (%)||57 (58.2)||47 (47.0)||0.80 (0.62–1.06)||37 (45.7)||33 (40.2)||0.87 (0.61–1.24)||18 (34.6)||20 (35.7)||1.03 (0.62–1.72)|
|Incontinence of liquids (%)||3 (3.0)||6 (6.0)||1.96 (0.50–7.62)||4 (4.9)||6 (7.2)||1.46 (0.43–5.00)||4 (7.7)||2 (3.6)||0.46 (0.09–2.43)|
|Incontinence of solids (%)||1 (1.0)||1 (1.0)||0.98 (0.06–15.45)||5 (6.1)||3 (3.6)||0.59 (0.15–2.37)||0 (0)||0 (0)||–|
|Urgency of defecation (%)c||9 (9.2)||5 (5.0)||0.54 (0.19–1.57)||5 (6.2)||8 (9.6)||1.56 (0.53–4.57)||6 (11.5)||2 (3.7)||0.31 (0.07–1.47)|
|Moderate/severe dyspareunia (%)||2 (2.1)||6 (6.1)||2.94 (0.61–14.21)||6 (7.5)||6 (7.2)||0.98 (0.33–2.90)||1 (1.9)||4 (7.1)||3.71 (0.43–32.16)|
|Dyspareunia preventing intercourse (%)||6 (6.3)||12 (12.2)||1.96 (0.77–5.02)||11 (13.8)||18 (21.7)||1.60 (0.81–3.17)||3 (6.1)||6 (12.0)||1.86 (0.49–7.05)|
No differences were found between the two groups for any outcome measures of anal incontinence or dyspareunia at either time point postpartum (Table 3).
A visual perineal inspection using the REEDA tool was performed on 75% of women (25% were not assessed because of early discharge from hospital prior to physical examination). No differences were found between groups in the mean REEDA score. Perineal pain was reported more often in the immediate postpartum period following restrictive use of episiotomy than with routine use (98.9% restrictive versus 87.8% routine, RR 1.10, 95% CI 1.01–1.21), but this difference did not persist at 6 weeks or 1 year postpartum. Although the differences in other measures of perineal pain at 6 weeks postpartum were small, and did not meet statistical significance, the restrictive use of episiotomy was consistently associated with more perineal pain than routine use (Table S3).
The mean EPDS score was higher following restrictive use of episiotomy compared with routine use in the immediate postpartum period (6.7 restrictive versus 5.1 routine, P = 0.01). This difference did not persist in the longer term. No differences were found in the incidence of an EPDS ≥ 13, indicative of possible depression, at any time point (Table S4).
This longitudinal follow-up of women randomised to restrictive versus routine use of episiotomy at OVD in a two-centred RCT provides a comprehensive assessment of maternal morbidities from the third trimester of pregnancy to 1 year postpartum. Pelvic floor morbidities historically associated with OVD were often as prevalent, if not more prevalent, in the third trimester of pregnancy than postpartum. Postpartum urinary incontinence was reported by up to half of the respondents, but new urinary incontinence was reported by one in three women. Urgency of micturition was most prevalent in pregnancy, and reduced significantly in the puerperium. Anal incontinence was most prevalent at 6 weeks postpartum, but had resolved in all cases by 1 year. Moderate or severe dyspareunia was reported by almost 5% of the cohort antenatally, which doubled at 6 weeks postpartum, but returned to baseline levels by 1 year. As expected, perineal pain was found to be present in nearly all women immediately postpartum, but resolved over time, but with one in five women still affected to some degree at 1 year postpartum. Psychological morbidity was found to be more prevalent in the third trimester of pregnancy than in the postpartum period, which may be an indication of the emotional adjustments experienced by women, particularly in a first pregnancy. A restrictive approach to episiotomy at OVD was associated with more perineal pain in the immediate postpartum period, and more stress urinary incontinence at 6 weeks postpartum than a routine approach; however, these differences had resolved by 1 year postpartum.
An important strength of this study is the longitudinal methodology used, with data collected prospectively from the third trimester up to 1 year postpartum. This has rarely been achieved in either clinical trials or observational studies of OVD. Having baseline data available for comparison with postpartum data has revealed that although the morbidities studied are associated with OVD, they are in fact as prevalent and in some cases more prevalent in the third trimester of pregnancy. It may be that morbidities previously attributed to OVD have in fact been overestimated. Validated assessment tools were used to gain an in-depth understanding of maternal morbidities and how they impact on the lives of women in the postpartum period. The questionnaires were relatively easy to complete and the percentage of missing data was small (<5%), despite the respondents being instructed to miss out any question they did not feel comfortable completing. Participating numbers were high in the antenatal, immediate postpartum period and at 6 weeks postpartum (99, 92 and 83%, respectively). The main limitation of the study was the loss to follow-up at 1 year postpartum (46%). The data from the descriptive analysis reassured us that although the 1 year postpartum response rates were less than at baseline, there was no specific population from whom data were missing.
One longitudinal study of the prevalence of maternal morbidities following OVD with data collected in the antenatal period was identified in the literature. Our study supports the authors’ findings with regard to the antenatal prevalence of urinary and anal incontinence. They reported postpartum urinary incontinence to be more likely if urinary incontinence was present during pregnancy, which is also consistent with our findings. The reported rates of postnatal urinary morbidity in our study were in line with previously published studies,[15-17] but were higher than those reported by Chalihaet et al.
No differences in the incidence of urinary incontinence were found between a restrictive and routine approach to episiotomy in studies of spontaneous vaginal delivery.[15, 19, 20] In contrast to this, our findings suggest that routine use of episiotomy at OVD may in fact offer some protection against urinary morbidity at 6 weeks postpartum, especially stress urinary incontinence. There is limited evidence available on the relationship between episiotomy at OVD and anal incontinence; however, in their quasi-randomised trial Fritel et al. reported that a policy of routine episiotomy was associated with a greater risk of anal incontinence, compared with a restrictive policy, 4 years after first delivery if incontinence of flatus was included (16 versus 11%; P = 0.04). This difference persisted in a subgroup analysis for incontinence of flatus alone (13 versus 8%; P = 0.02), but not for anal incontinence of solids alone (3% for both groups, P = 0.94). Our results do not support this finding, as no differences were identified between the two approaches to episiotomy with regard to anal incontinence of flatus, liquids or solids; however, this may reflect limitations of the sample size.
Similarly, most studies of episiotomy use at spontaneous vaginal delivery report no differences in perineal pain or healing complications between a restrictive and a routine approach to episiotomy,[19-21] whereas the Argentine Episiotomy Trial Collaborative Group reported that a selective approach to episiotomy use at vaginal delivery was associated with less perineal pain at hospital discharge than a routine approach (RR 0.72, 95% CI 0.65–0.81). Similarly, previously published data by the authors reported no difference in the use of moderate/strong analgesia (a proxy measure of perineal pain) in either the inpatient or outpatient period up to 10 days postpartum between women randomised to either a restrictive or routine use of episiotomy. Findings from this study, in contrast, suggest that the restrictive use of episiotomy at OVD may increase perineal pain in the immediate postpartum period.
The longitudinal approach to data collection in this study commencing in the antenatal period is strongly recommended for future studies of OVD and episiotomy, as significant levels of pre-existing maternal morbidity have been identified that could be inappropriately attributed to delivery procedures. The findings of the randomised analyses should be interpreted with caution, as this was a large two-centre pilot study known to be under-powered to detect significant differences in anal incontinence morbidity, although with an adequate sample size to address urinary morbidity and perineal pain.[1, 8, 23] Further large-scale studies are required to provide women and their carers with high-quality information regarding episiotomy use at OVD in terms of anal sphincter damage and subsequent anorectal symptoms.
Obstetricians vary greatly in their approach to OVD and use of episiotomy. Although our initial RCT findings did not provide definitive evidence of benefit or harm in relation to restrictive versus routine use of episiotomy, the contemporaneous cohort study of 1360 women reported higher rates of postpartum haemorrhage with the use of episiotomy.[8, 24] In contrast, the current study suggests potential benefits with routine use of episiotomy in terms of less perineal pain and urinary stress incontinence at 6 weeks postpartum. As with many intrapartum interventions there are both benefits and risks, and for now clinicians are justified in taking either a routine or restrictive approach to episiotomy at OVD, depending on their perception of the balance of morbidities.
No conflicts of interest to declare.
D.J.M. had the original idea for the study. DJM, MM and B.S. carried out the study design. DJM, MM and B.S. obtained funding. MM, RB, KG and LH were responsible for data collection. MM and DJM carried out the analysis. MM drafted the article, which was later revised by all authors.
This study was given a favourable ethical opinion by the Tayside Committee on Medical Research Ethics A (06/S1401/55).
We gratefully acknowledge funding received from an Anonymous Trust Grant, Tayside Universities Hospitals. R.B. was supported by a Wellbeing of Women Research Training Fellowship.
We are very grateful to all of the women who participated in the study.
Third- or fourth-degree lacerations are important obstetric complications to understand, for clinicians and researchers alike. These severe lacerations are associated with both fecal and flatal incontinence, chronic pain and dyspareunia. There is evidence to suggest that routine episiotomy at the time of operative vaginal delivery is associated with more third- and fourth-degree lacerations (Hartmann et al., JAMA 2005;293:2141–2148); however, when this evidence is dissected, the association appears to be with midline episiotomy as opposed to mediolateral episiotomy, which may actually be protective (Revicky et al., EJOG 2010;150:142–146).
The current trial was conducted to examine whether mediolateral episiotomy is associated with short- or long-term maternal outcomes. The authors prospectively randomised women to restrictive episiotomy versus routine mediolateral episiotomy. In the short term, they found higher rates of perineal pain or urinary incontinence in women randomised to restrictive episiotomy. In the long term, they found no differences in outcomes. Furthermore, the authors found a relatively high rate of incontinence at baseline prior to the delivery.
This high rate of symptoms of pelvic floor dysfunction suggest that when such symptoms are experienced postpartum, they may not be the result of the mode of delivery itself. Further study to characterise the rates in women prior to pregnancy, their development throughout pregnancy and the impact of the mode of delivery is important work that, although easy to do, requires the development of a pre-pregnancy cohort, which can be expensive.
The findings of the randomised trial are quite intriguing, particularly in the USA, where mediolateral episiotomy is rarely employed. One common concern regarding the use of mediolateral episiotomy is postpartum perineal pain, but in this trial, the opposite was found. Thus, with a potential reduction in both pain and short-term urinary incontinence, it seems reasonable to use mediolateral episiotomy in the setting of an operative vaginal delivery; however, is this study enough to support a change in practice? I do not believe so. I think if one is already conducting routine mediolateral episiotomy, then persisting makes sense, as this study did not suggest particular harms. Alternatively, if one was not a practitioner of routine episiotomy in the setting of operative vaginal delivery, it is fair to say that there was no improvement in long-term outcomes, and the confidence intervals were quite wide, even with the short-term outcomes. Thus, I think that although interesting, this study needs follow-up with a much larger prospective multicentre trial that can address the uncertainty produced by the current study.
Disclosure of interests
Dr Caughey serves as a medical advisor to three companies – Mindchild, Ariosa and Cellscape – none of which have any relation to the issues covered in the current commentary.
Department of Obstetrics and Gynecology, Oregon Health and Science University, Portland, OR, USA