Chapter 2. Perineal pain and dyspareunia
INTRODUCTION
The majority of women who have a vaginal delivery will experience some degree of perineal pain, which is one of the most commonly reported symptoms in the immediate postnatal period (Sleep 1995). Dyspareunia or painful sexual intercourse, although a different symptom, can be related to perineal pain. Several of the relevant studies, especially those with information on risk factors, have investigated both of these symptoms. If each symptom were to be reviewed and described separately in this guideline, there would be substantial repetition. Some parts of the guideline therefore describe perineal pain and dyspareunia in separate sections and others describe them together. In addition to dyspareunia, there are some recent data on other sexual health problems after childbirth and on the timing of the resumption of sexual intercourse, which are also described in this guideline.
Definitions
• Perineal pain is defined as any pain occurring in the perineal body, an area of muscular and fibrous tissue which extends from the symphysis pubis to the coccyx.
• Dyspareunia refers to pain or discomfort occurring during sexual intercourse, including pain on penetration. Dyspareunia is sometimes categorised as superficial or deep.
Frequency of occurrence: perineal pain
Pain has been assessed using a number of tools developed for use in the general population, including visual analogue scales and visual assessment tools. In one randomised controlled trial (RCT) (Kettle et al 2002), pain measures at 10 days after the birth included levels of pain experienced by women on activity, including when passing urine and opening their bowels. Other indicators of pain severity have included total dosage of analgesia used, weight of total amount of topically applied anaesthesia and/or frequency of treatments. The timing of assessment of perineal pain often differs between studies and there is limited information on the long-term effects of pain.
In a prospective observational study in Scotland, among a representative sample of all deliveries over a defined period, 42% of 1249 women reported a painful perineum when questioned in hospital (Glazener et al 1995). In the West Berkshire Perineal Management Trial (see later), 23% of almost 900 women (inclusion restricted to those expected to have a spontaneous vaginal delivery (SVD)) reported perineal pain at 10 days postpartum (Sleep et al 1984). Among over 5000 spontaneous vaginal births in the ‘hands on or poised’ trial (HOOP), which compared alternative methods of conducting the second stage of labour, 33% of women in the trial as a whole reported perineal pain in the previous 24 hours, when asked at 10 days postpartum (McCandlish et al 1998). A prospective cohort study from Canada, which examined differences in perineal pain relative to the type of trauma sustained among 444 women who had a vaginal birth, found that overall 92% reported pain on the first day after the birth and 61% reported pain on day 7 (Macarthur & Macarthur 2004).
Perineal pain is clearly very common in the early puerperium, but for some women it can also be more persistent and various studies have shown that at least 8–10% of women experience this well past the 6–8-week postnatal discharge from maternity care. In the trial by Sleep et al (1984), when followed up at 3 months, 8% of the women reported perineal pain at that time. In the study by Glazener et al (1995), when questioned again at 8 weeks, 22% of the sample reported experiencing a painful perineum between then and the first week, and when questioned at 12–18 months, 9.8% had experienced perineal pain at some time between 8 weeks and 12–18 months. Brown & Lumley (1998) contacted 1336 women at 6–7 months postpartum in a cross-sectional study of all deliveries occurring over 2 weeks within a region in Australia, and found that for 21% a painful perineum had been a problem at some time since the birth. The proportion who still had this at 6–7 months was not given. Macarthur & Macarthur (2004) reported that 7% of women in their study population had perineal pain when questioned about this at 6 weeks after the birth.
Frequency of occurrence: dyspareunia
Like perineal pain, there is some information on the prevalence of dyspareunia from observational studies and from the trials of perineal management, although again it must be remembered that data on this from samples restricted to particular delivery-types or parity groups will limit the generalisability of the estimates obtained.
Brown & Lumley (1998), in the Australian study described earlier, found that 26% of the women questioned at 6–7 months postpartum had experienced ‘a sexual problem’ at some time since the birth, although further specification of the problem was not given nor what proportion still had it. Glazener (1997) found that at 8 weeks postpartum, 25% of the sample had not attempted intercourse. The remaining 75% had attempted intercourse although for 5% it had been unsuccessful. Questions on problems with intercourse were included, showing that between 1 and 8 weeks, 28% of women had found intercourse to be sore or difficult; between 2–18 months this was reported by 20%. Lack of interest in sex was reported by 9% of women at 8 weeks, rising to 21% between 2–18 months.
A cross-sectional study of all primiparous women delivering in one maternity unit in London was conducted to enquire about a range of sexual health problems (Barrett et al 2000). Postal questionnaires were sent at 6 months postpartum to 796 primiparae asking about sexual problems since the birth and prior to pregnancy; 484 (61%) replied. By 6 weeks (as recalled), 32% of women had resumed intercourse, 62% had done so by 8 weeks and 81% by 3 months. Dyspareunia, which was specified as including painful penetration and/or pain during intercourse or orgasm, was experienced by 62% of women at some time in the first 3 months after birth and 31% still had this at 6 months. Loss of sexual desire was reported by 53% of women in the first 3 months, and by 37% at 6 months. Other problems still experienced at 6 months were vaginal tightness (20%), vaginal looseness or lack of muscle tone (12%) and lack of vaginal lubrication (26%). Enquiry was also made about the same set of sexual problems in relation to the year before pregnancy and, although subject to greater recall bias, all problems were found to have increased significantly in the first 3 months after delivery. Although all problems had declined by 6 months postpartum, they were still substantially greater than pre-pregnancy levels. The response rate in this survey was 61%, which is good given the sensitive nature of the subject, but gives less confidence in relation to prevalence estimates. Even if all of the non-responders were problem free, however, the findings still provide clear evidence that sexual problems are common as well as persistent following childbirth.
Sleep et al (1984), in their trial, found that by 3 months postpartum 90% of women reported having resumed sexual intercourse. Just over half of these had experienced dyspareunia at some time since the birth and almost 20% still had it at 3 months. Klein et al (1994), in a similar trial (see later), found that 64% of the sample had resumed intercourse by 6 weeks and 96% by 3 months. On first intercourse, almost 80% of the women reported some pain, although in most cases this was only mild or discomforting.
Risk factors: perineal pain and dyspareunia
The main risk factors for perineal pain and dyspareunia that have been identified in studies relate to mode of delivery, perineal trauma and primiparity. These factors are all clearly highly interrelated so appropriate multivariate analyses need to be undertaken in order to report on the independent effects of each. Methods of managing perineal trauma, such as the use of different suture methods and materials, have also been investigated with respect to subsequent perineal pain and dyspareunia and are described in this section on risk factors.
Mode of delivery
Mode of delivery is associated with substantial variation in perineal pain (short and long term) with much higher rates occurring after instrumental compared with spontaneous vaginal deliveries and the lowest rates occurring after caesarean sections. There is a similar pattern for reporting of dyspareunia but differences are generally less marked. There has been increasing debate on the benefits and risks of planned caesarean section compared with vaginal birth, with one of the purported reasons for offering women a choice over mode of birth being impact of vaginal birth on perineal trauma and sexual health.
In the study by Glazener et al (1995) described earlier, differences in perineal pain according to delivery mode were clearly documented, occurring whilst still in hospital, between then and 8 weeks, and between 2 and 18 months. The respective prevalences whilst in hospital were 84% after instrumental delivery, 42% after an SVD and 5% after caesarean section; 59%, 19% and 4% between then and 8 weeks; and 30%, 7% and 2% between 2 and 18 months. Data on dyspareunia were also reported for this sample (see earlier), but not separately according to mode of delivery (Glazener 1997).
Johanson et al (1993) obtained data on a number of morbidity indicators from 313 women who were part of an RCT of forceps versus vacuum extraction deliveries, and from 100 consecutive unselected SVD deliveries. A significant difference in perineal discomfort at 24–48 hours was found, which occurred in 45% of the instrumental delivery group and 28% of the SVD group (odds ratio (OR) 2.1, 95% confidence interval (CI) 1.2–3.4). At 15–24 months postpartum, 28% compared with 19% respectively reported a painful perineum (mostly occurring only sometimes), but this difference was not statistically significant. Pain on intercourse (mostly occurring only sometimes) at 15–24 months, however, was significantly more common in the instrumental group, reported by 37%, compared with 21% of the normal deliveries (OR 2.0, 95% CI 1.03–3.80).
Brown & Lumley (1998), in their population-based sample in Australia, found that perineal pain, reported as a problem some time between the birth and 6–7 months postpartum, occurred in 20% of women after SVD, 54% after instrumental birth, 2% after emergency caesarean section and none after elective section (OR for instrumental relative to SVD 4.69, 95% CI 3.2–6.8). Sexual problems, not further specified, differed only for instrumental deliveries, being reported by 24%, 39%, 29% and 27% respectively after the four types of delivery (OR for instrumental relative to SVD 2.06, 95% CI 1.4–3.0). These differences in perineal pain and sexual problems between instrumental compared with spontaneous vaginal births were adjusted to take account of infant birthweight, length of labour and degree of perineal trauma, and remained statistically significant.
Signorello and colleagues (2001), in a study from Boston, Massachusetts which recruited primiparous women, reported an OR of 2.5 (95% CI 1.3–4.8) of dyspareunia at 6 months after giving birth following assisted vaginal delivery compared with spontaneous vaginal delivery. This was after adjusting data for maternal age, method of infant feeding, prior experience of dyspareunia, duration of the second stage of labour, extent of perineal trauma and infant birthweight.
A German study evaluated the impact of mode of delivery on sexual health among primiparous women with responses from 655 of 1613 (41%) women contacted from 6 months to 2.5 years after the birth (Buhling et al 2006). The researchers found that 47% of women had resumed intercourse within 8 weeks of birth, and although 69% (436/633) reported some pain on resumption, the degree of pain differed significantly by mode of delivery. Women who had an operative vaginal delivery had a higher prevalence of pain compared with women who had a caesarean section or spontaneous vaginal birth without perineal injury (p<0.007). The potential for response bias and failure to take account of potential confounding factors in the analysis means findings should be interpreted with caution.
Barrett et al (2000), in their study of sexual problems among primiparae, found that dyspareunia occurring some time during the first 3 months postpartum was reported by 62% of women after an SVD, 78% after forceps/ventouse delivery, 41% after a section with labour and 47% after a section without labour. After adjustment for possible confounders, the difference for forceps/ventouse deliveries relative to SVD remained significant (OR 2.41, 95% CI 1.24–4.69), but the differences for both types of caesarean section did not. Dyspareunia at 6 months postpartum was reported by 30% of women after SVD, 37% after forceps/ventouse and 28% and 21% respectively after section with and without labour, but none of these differences was statistically significant. Continued breastfeeding up to 6 months was one of the two factors in this study found to be significantly associated with dyspareunia at 6 months. The other was a previous history of dyspareunia prior to pregnancy and both factors remained significant after logistic adjustment. Among the women still breastfeeding, 40% reported dyspareunia compared with 25% of those not breastfeeding (OR 2.25, 95% CI 1.42–3.57). The researchers suggest that hormonal profile, loss of libido and vaginal dryness could contribute towards the symptom excess in breastfeeding mothers.
Glazener (1997) found that women who were still breastfeeding at 8 weeks were significantly more likely to report a lack of interest in intercourse than those who were bottle feeding. Tiredness was also more common among the breastfeeding mothers, but even after taking account of this, the relationship with lack of interest in sex remained.
A recent small study compared perineal morbidity and sexual function following vaginal delivery or elective caesarean section (Griffiths et al 2006). Two hundred and eight women were contacted 2 years after delivery of their first baby to record the prevalence of a number of subjective physical and psychological health problems, including sexual satisfaction. Exclusion criteria included women who had had a subsequent pregnancy. Replies were received from 109 women (52%). The proportions of women reporting all symptoms of interest within the 2-year period, including dyspareunia, were higher among women who had given birth vaginally. Dyspareunia was reported by 33% of women in the spontaneous vaginal delivery group, 36% in the ventouse group and 52% in the forceps group, with no cases reported among the elective caesarean section group. Women who had dyspareunia were asked about the severity of this (information on how questions were asked was not provided), which showed that severity was highest among women who had a forceps delivery. Findings should be interpreted with caution due to the low response rate, small numbers, potential for recall bias in relation to when symptoms were first experienced and lack of analysis to enable independent risk factors to be identified.
Perineal trauma
Perineal trauma is generally considered to be more likely to result in perineal pain and dyspareunia than if the perineum is intact, but there has been considerable debate on the benefits and adverse effects of episiotomy compared with spontaneous laceration (Sleep 1995). Several trials have examined this issue. The West Berkshire Perineal Management Trial (Sleep et al 1984), referred to earlier, of a restricted versus a liberal episiotomy policy, was the first to provide good evidence on the effects of episiotomy use among women who, towards the end of second-stage labour, were expected to have a spontaneous vaginal delivery. For women allocated to the restrictive group, the midwife was asked to avoid episiotomy and only perform an incision if fetal indications (bradycardia, tachycardia, meconium-stained liquor) warranted this (n = 498). In the liberal group the midwife was asked to try to prevent a perineal tear (n = 502), the intention being that she should use episiotomy more liberally to do this. The different policies did result in more tears and more intact perinea in the restrictive group, but no differences were found in perineal pain at 10 days or 3 months, nor in dyspareunia at 3 months. These authors concluded that the findings provide little support for either the liberal use of episiotomy or for claims that its reduced use would decrease postpartum morbidity (Sleep et al 1984).
Klein et al (1994) carried out a similar trial in Canada, which again found that although significantly fewer episiotomies were performed in the restrictive policy group, there were no differences at 1 and 10 days postpartum in perineal pain, or in sexual problems (pain and sexual satisfaction) at 3 months.
A Cochrane systematic review (Carroli et al 2000) including six RCTs comparing policies of the restrictive use of episiotomy versus routine/liberal use confirmed that there were no differences in perineal pain or dyspareunia according to episiotomy policy, although other differences (reduced risk of posterior perineal trauma, less suturing and fewer healing complications) led the reviewers to conclude that a restrictive episiotomy policy was beneficial and should be recommended.
Since these trials were comparing perineal management policies, some women in the restricted episiotomy group still had an episiotomy, and vice versa. In the study by Klein et al (1994), it was felt that this justified undertaking further subgroup analyses and all women in the trial were recategorised (irrespective of trial group allocation) according to the type of perineal trauma (intact, laceration, episiotomy, third/fourth-degree tears). Perineal pain between days 1–10 and at 3 months was found to be least in the intact group and greatest in those with third/fourth-degree tears. Episiotomy and spontaneous tears were intermediate and differed from each other for early perineal pain and for later pain classed as occurring most or all of the time, both a little more common after episiotomy. Dyspareunia on first resumption of intercourse followed a similar pattern of relationship with trauma. The type of episiotomy in this trial was midline, which is rarely practised in the UK, and almost all women in the trial had spontaneous vaginal deliveries.
In a subsequent analysis of data from the study by Glazener (1997) to determine the independent predictors of pain or difficulty with intercourse at 2 months postpartum, logistic regression was undertaken and episiotomy was found to be an independent predictor of this (Glazener 1998).
The prospective cohort study from Canada (Macarthur & Macarthur 2004) which collected data on perineal pain outcomes from 447 women of all parities who had given birth vaginally, at 1 day, 7 days and 6 weeks after the birth, included 84 women who had an intact perineum, 220 who had a first- or second-degree tear, 97 women who had an episiotomy (including women who had either a midline or mediolateral episiotomy) and 46 who had a third- or fourth-degree tear. The primary study outcome was the incidence of perineal pain on the day of interview, with secondary outcomes including pain score measurements and impact of pain on daily activities. There were high response rates at each stage of the study (95% or above). Multivariable regression analysis was undertaken to calculate adjusted relative risks, after taking account of differences between the groups on potential confounding variables. Acute postpartum perineal pain was common among all women but pain severity was more frequent and more severe for women who had more extensive perineal trauma. On day 1, the three groups of women who had sustained trauma were 30% more likely to report pain, and on day 7 the adjusted RR of perineal pain in the trauma groups compared with the intact group ranged from 1.5 (first- or second-degree tear) to 2.1 (third- or fourth-degree tear).
Parity
Perineal pain and dyspareunia have both been documented as more common after first compared with subsequent births (Barrett et al., 1998 and Glazener et al., 1995). However, the extent to which this is influenced by the greater proportion of instrumental deliveries and perineal trauma that occur among first births is difficult to assess. The parity association found by Glazener et al (1995) for painful perineum was based on univariate analysis, not taking other factors into account. For pain or difficulty with intercourse at 2 months postpartum, multivariate analysis was undertaken to take account of possible interrelated factors and showed that primiparity remained associated with the symptoms, although forceps delivery did not. In the perineal management trial by Sleep et al (1984), data were presented separately for first and later births showing that mild and moderate perineal pain at 10 days were reported by almost twice as many primiparae as multiparae. This study was of anticipated spontaneous vaginal births so should not be confounded by any parity-related instrumental delivery effect, but in both trial arms the primiparae had more perineal trauma (episiotomy or laceration) than the multiparae. Klein et al (1994), in their similar study, stated that primiparae experienced more perineal pain and sexual problems than multiparae, but presented no data to show this.
Suture materials and methods
The effects of suture materials and methods on subsequent perineal pain and dyspareunia have been examined in a number of systematic reviews, some of which have had clear findings relevant to practice.
A Cochrane systematic review, including eight RCTs, compared synthetic suture materials with catgut (Kettle & Johanson 1999). Catgut is manufactured from collagen from mammals and has been a commonly used suture material, whilst synthetic suture materials, polyglycolic acid (Dexon) and polyglactin (Vicryl), are more recent. All the trials in the review were consistent in showing lower rates of short-term (3 days) perineal pain in the synthetic suture groups (OR 0.62, 95% CI 0.54–0.71). Two of the trials examined perineal pain and three examined dyspareunia at 3 months, with no differences found in either of these longer term outcomes according to type of suture material. Other outcomes, the need for analgesia and suture dehiscence, were reduced in the groups sutured with synthetic material, but two trials that included as secondary outcomes the removal of suture material at 10 days and 3 months found this was undertaken more often in the synthetic materials group. The reviewers concluded that the use of absorbable synthetic suture material (polyglycolic acid and polyglactin sutures) appears to decrease short-term perineal pain. The length of time taken for synthetic material to be absorbed, however, is of concern (Kettle & Johanson 1999).
The effects of continuous subcuticular versus interrupted transcutaneous sutures for closure of perineal skin on long- and short-term perineal pain and dyspareunia, and a number of other outcomes, were examined in a recently updated Cochrane systematic review (Kettle et al 2007). Seven trials, which provided data on 3822 women, were included. Meta-analysis showed that continuous suture techniques (all layers or perineal skin only) were associated with less pain for up to 10 days postpartum (relative risk (RR) 0.70, 95% CI 0.64–0.76) when compared with interrupted sutures for perineal closure. Subgroup analysis showed greater reduction in pain when continuous suturing techniques were used for all layers (RR 0.65, 95% CI 0.60–0.71). There was an overall reduction in analgesia use associated with the continuous subcutaneous technique compared with interrupted sutures for repair of perineal skin (RR 0.70, 95% CI 0.58–0.84). There was some evidence of reduction in dyspareunia up to 3 months after the birth experienced by women in the groups who had continuous suturing for all layers of their perineal trauma (RR 0.83, 95% CI 0.70–0.98). There was also a reduction in suture removal in the continuous suturing groups versus interrupted (RR 0.54, 95% CI 0.45–0.65) but no significant differences in the need for resuturing of wounds or long-term pain.
A stratified RCT (Ipswich Childbirth Study — included in the systematic review described above of type of suture material), used a 2 × 2 factorial design and also compared the effects of the standard three-stage suturing with a two-stage method, in which the perineal skin was left unsutured (Gordon et al 1998). The trial was carried out in a single centre and involved 1780 women of all parities who sustained episiotomy or first- or second-degree tears following spontaneous or instrumental vaginal deliveries. Women were assessed by questionnaire at 24–48 hours, 10 days and 3 months postpartum. No differences were found in perineal pain at 24–48 hours or 10 days between the two- and three-stage repair groups, but at 3 months fewer women in the two-stage group reported perineal pain and more had resumed pain-free intercourse. Among women who had resumed intercourse, reduced rates of dyspareunia in the two-stage group just reached statistical significance (RR 0.80, 95% CI 0.65–0.99).
In a large RCT undertaken at a hospital in North Staffordshire, which also used a 2 × 2 factorial design (included in the above review of suturing methods, but not included in the review of suture materials), Kettle et al (2002) randomised 1542 women who had a spontaneous vaginal delivery with a second-degree tear or episiotomy to either a continuous (n = 771) or interrupted (n = 771) suturing method, and to be sutured using either a rapidly absorbed polyglactin 910 suture material (n = 772) or standard polyglactin 910 (n = 770). Primary outcomes were pain at 10 days after giving birth, and superficial dyspareunia at 3 months. There was significantly less perineal pain amongst women who had a continuous technique at 10 days (OR 0.47, 95% CI 0.38–0.58), the researchers postulating that the pain reduction may be a consequence of the transfer of tension along the whole of the single suture and avoidance of nerve endings as skin sutures are inserted into subcutaneous tissue. There were no differences in the reporting of pain according to the type of suture material used. At 3 months there were no differences in the reporting of superficial dyspareunia between women whose trauma had been sutured using the continuous or interrupted methods (OR 0.84, 95% CI 0.72–1.33) or among the rapidly absorbed suture material or standard suture material groups (OR 1.13, 95% CI 0.84–1.54). Suture material was less likely to have been removed among women in the rapidly absorbed material group and in the continuous suture method group. The researchers concluded that use of the continuous repair methods could prevent one in six women from experiencing perineal pain at 10 days after birth, and use of the rapidly absorbed polyglactin 910 material could obviate the need for suture removal within the first 3 months for one in 10 women sutured.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
