NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Collaborating Centre for Women's and Children's Health (UK). Intrapartum Care: Care of Healthy Women and Their Babies During Childbirth. London: National Institute for Health and Care Excellence (UK); 2014 Dec. (NICE Clinical Guidelines, No. 190.)
Update information February 2017: Sections that have been updated (see addendum files) have been marked with dark grey shading.
14.1. Introduction
Birth is an immensely important, often life-changing, event. Not only does the process of labour and birth present challenges to the baby but there are also major rapid physiological changes that take place to enable the baby to adapt to life after birth. These include the establishment of respirations, changes to the cardiovascular system, the regulation of body temperature, digestion and absorption and the development of a resistance to infections.
The vast majority of babies make this transition uneventfully but vigilance on the part of health-care professionals, and timely intervention when necessary, can influence the baby's longer term health and development.
Care of the baby immediately after birth in the intrapartum period is discussed in this chapter. Further care thereafter is discussed in the NICE clinical guideline on Postnatal Care,414 including promotion of breastfeeding, infant and mother bonding, and vitamin K supplementation for newborn babies.
Care of the woman immediately after birth includes assessment of her physical and emotional condition, as well as assessment (and possible repair) of trauma sustained during birth. It is also crucially important that appropriate assessment and treatment of any complications is undertaken, as failure to do so can have long-term consequences for the woman's physical, emotional and psychological wellbeing. As with the immediate care of the newborn baby, this should be balanced between assessing the woman's physical needs (and intervening should that be required) and giving the new mother/parents the opportunity to savour and enjoy this momentous and life-changing event.
14.2. Initial assessment of the newborn baby
14.2.1. Apgar score
14.2.1.1. Introduction
The Apgar score was developed in 1953 and has been widely adopted to assess the baby at the time of birth.415 It was first planned as an indicator for the need for resuscitation. It was not originally intended to predict longer term prognosis and includes assessment of colour, heart rate, tone, respiratory rate and reflex irritability.415–417
14.2.2. Review question
What is the evidence that different methods of initial neonatal assessment and examination influence outcomes?
- Including cardiovascular-respiratory and abnormalities assessment.
14.2.3. Description of included studies
A total of five cohort studies and one systematic review (containing 16 cohort studies) were identified.418–423 Only studies comparing the Apgar score with neonatal death and diagnosis were considered homogeneous enough to provide a new meta-analysis of the data. [EL = 2+]
14.2.4. Review findings
The results of meta-analyses on neonatal mortality and diagnosis of cerebral palsy are shown in Tables 166 and 167. Overall, the Apgar score appeared to be a moderate level predictor for neonatal deaths and the development of cerebral palsy, with the Apgar at 5 minutes having better predictive value than at 1 minute. Surprisingly, only one study was identified that examined predictive values of the Apgar score on longer term neurological development of the infants. There was no high-level study that examined the correlation between Apgar score and immediate neonatal outcomes.
Table 170
Meta-analysis on predictive value of Apgar score (neonatal mortality).
Table 171
Meta-analysis on predictive value of Apgar score (cerebral palsy).
14.2.5. Evidence statement
There is low-level evidence that the Apgar score at 5 minutes is moderately accurate at predicting neonatal death and cerebral palsy with reasonable specificity but low sensitivity. No high-level evidence could be found on immediate or longer term neonatal outcomes.
14.3. Neonatal resuscitation
14.3.1. Timing of cord clamping
14.3.1.1. Review question
When should neonatal resuscitation be instigated with respect to the timing of cord clamping?
14.3.1.2. Description of included studies
No evidence was identified that addressed this question.
14.3.2. Oxygen compared with air
14.3.2.1. Review question
Is air more effective than oxygen when used for neonatal resuscitation (a) initially and (b) after a period of no/poor response?
14.3.2.2. Description of included studies
This review included 8 papers (Bajaj et al., 2005; Ramji et al., 1993; Ramji et al., 2003; Saugstad et al., 1998; Saugstad et al., 2003; Vento et al., 2001; Vento et al., 2003; Vento et al., 2005). These papers represent the same evidence that was available for the 2007 guideline, although in 2007 the trials were included in the form of a systematic review plus 1 additional study, rather than the individual papers.
The papers reported 7 randomised controlled trials – 6 of them were single centre trials conducted in Spain (Vento et al., 2001; Vento et al., 2003; Vento et al., 2005) and India (Bajaj et al., 2005; Ramji et al., 1993; Ramji et al., 2003). The seventh trial was a multicentre trial (Saugstad et al., 1998) with an 18–24 month follow-up study (Saugstad et al., 2003).
All of the included studies compared the use of air (or 21% oxygen) with 100% oxygen for the initial resuscitation of babies. None of the studies identified evaluated the different resuscitation techniques in babies after a period of no/poor response.
Two studies (Vento et al., 2001; Vento et al., 2003) restricted their study populations to term babies, but the remaining studies included some babies born prior to 37 weeks. One study (Saugstad et al., 1998) reports the proportion of preterm babies (24%), but the remaining studies merely report mean gestational age (full details can be found in the evidence table in appendix I). No sub-group analyses were performed by gestational age; therefore, where applicable, the evidence has been downgraded for indirectness. None of the studies restricted their populations to babies born to women at low risk of developing complications, but it was pre-specified in the review protocol that babies born to women at high risk could be included.
14.3.2.3. Evidence profile
Table 172
Summary GRADE profile for comparison of room air with oxygen for neonatal resuscitation.
14.3.2.4. Evidence statements
There was evidence that resuscitation with room air was associated with lower rates of neonatal death overall (n=1367) and a similar trend was demonstrated when only mortality related to asphyxia (n=719) was considered. There was no evidence of a difference in the risk of hypoxic ischaemic encephalopathy (overall [n=1328], or more severe grades [n=897]) following resuscitation with room air or oxygen. Similarly, shorter and longer term neurological and developmental outcomes (n=213) did not show a difference between the 2 groups, but this evidence was mainly from 1 trial. There was evidence from 1 study (n=204) that more babies received intubation as part of the resuscitation procedure following resuscitation with room air, but there were no other differences in terms of the further interventions used in the 2 groups of babies, either in terms of resuscitation failure or the need for adrenaline and/or chest compressions. No difference was found in the mean heart rate at 1 and 5 minutes (n=635). The evidence around 5 minute Apgar score was mixed. Five trials (n=1390) reported no difference in Apgar score between babies resuscitated with room air and 100% oxygen, but 2 trials (n=123) found a higher median Apgar score in the babies resuscitated with room air. The evidence was of moderate to very low.
14.3.2.5. Health economics profile
No published economic evaluations were identified for this question.
14.3.2.6. Evidence to recommendations
14.3.2.6.1. Relative value placed on the outcomes considered
For this review, the guideline development group prioritised outcomes relating to neonatal mortality and morbidity. In particular, they were concerned with serious morbidities such as hypoxic ischaemic encephalopathy and cerebral palsy.
This review formed part of the original 2007 guideline, but for the update the search was expanded to cover air compared with oxygen after a period of no or poor response. However, no additional papers were identified which addressed this issue.
14.3.2.6.2. Consideration of clinical benefits and harms
One relatively large meta-analysis of 5 studies found that there were significantly fewer deaths when using room air compared with using oxygen. The guideline development group agreed that this was an extremely important finding. Although they recognised that there was a statistically significant increase in the need for intubation during resuscitation when using room air compared with oxygen, they agreed that this was outweighed by the mortality finding and so concluded that resuscitation should be performed with air rather than oxygen.
The group's clinical experience was that if a baby requires resuscitation, it will often be taken into a separate room and that this can cause a great deal of anxiety for the mother. The group recognised that in some situations this may be unavoidable, but felt that it was important that where possible this separation should be minimised. Given the anxiety that the need for neonatal resuscitation can cause, the group felt that it was important that in this situation one of the healthcare professionals present should be allocated to stay with the woman to provide ongoing support and be responsible for communicating with the woman and her birth companion(s) throughout the resuscitation.
14.3.2.6.3. Consideration of health benefits and resource uses
There were no specific resource use issues related to the treatment alternatives addressed for this question in terms of staff time and equipment needed. However, use of oxygen may be more expensive, especially where tanks of oxygen are used rather than piped oxygen.
Given the significant difference between the 2 groups in terms of neonatal mortality, it was a straightforward assessment that the benefits from use of air was likely to be cost effective, owing to the gain in quality adjusted life years (QALYs) as a result.
14.3.2.6.4. Quality of evidence
The evidence for this question ranged in quality from moderate to very low, although the evidence for the majority of reported outcomes was low or very low. The group recognised that the quality of the evidence for the neonatal mortality outcome was graded very low, owing to inadequate randomisation and concealment of treatment allocation in the majority of the included studies. However, given the severity of the outcome, the group did not feel that they could disregard the finding.
14.3.2.6.5. Other considerations
In the previous edition of the guideline, it had been recommended that oxygen should be available for all babies who do not respond once adequate ventilation has been established. The group did not feel that the studies provided evidence for when oxygen might be required and so chose to remove the recommendation. With such limited evidence, they did not feel that it was appropriate that midwives attending a homebirth should be required to bring oxygen. However, the guideline development group also acknowledged that there may be situations in which the use of oxygen is clinically appropriate and so did not wish to specifically recommend that oxygen should not be used.
The group agreed that it was important that healthcare professionals be reminded about the importance of calling for help if a baby requires resuscitation.
The group chose to make a recommendation detailing the features that should be assessed in order to determine whether resuscitation is required. They noted that the guideline contained other recommendations detailing the different observations that should be made at birth. However, they felt it important to highlight that respiration, heart rate and tone are particularly relevant when deciding about whether resuscitation is necessary.
14.3.2.7. Recommendations
- 262.
In the first minutes after birth, evaluate the condition of the baby – specifically respiration, heart rate and tone – in order to determine whether resuscitation is needed according to nationally accredited guidelines on neonatal resuscitation. [new 2014]
- 263.
All relevant healthcare professionals caring for women during birth should attend annually a course in neonatal resuscitation that is consistent with nationally accredited guidelines on neonatal resuscitation. [new 2014]
- 264.
In all birth settings:
- bear in mind that it will be necessary to call for help if the baby needs resuscitation, and plan accordingly
- ensure that there are facilities for resuscitation, and for transferring the baby to another location if necessary
- develop emergency referral pathways for both the woman and the baby, and implement these if necessary. [new 2014]
- 265.
If a newborn baby needs basic resuscitation, start with air. [2014]
- 266.
Minimise separation of the baby and mother, taking into account the clinical circumstances. [new 2014]
- 267.
Throughout an emergency situation in which the baby needs resuscitation, allocate a member of the healthcare team to talk with, and offer support to, the woman and any birth companion(s). [new 2014]
14.4. Routine paired cord-blood gas analysis
14.4.1. Review question
Is routine paired cord blood gas analysis predictive of perinatal or longer term outcomes? Does routine paired cord blood analysis improve perinatal outcomes?
For further details on the evidence review protocol, please see appendix E.
14.4.2. Description of included studies
This review includes 8 studies (Hefler et al., 2007; Keski-Nisula et al., 2012; Malin et al., 2010; Svirko et al., 2008; White et al., 2010; Wiberg et al., 2010; Wildschut et al., 2005; Yeh et al., 2012).
One of the included studies is a systematic review (Malin et al., 2011) which comprised 51 observational studies from different countries with a variety of study designs. For the purpose of this review, only the results from the 14 included studies in the systematic review with a term population are reported here. Of the 7 other included studies, 1 is a prospective comparative observational study from the Netherlands (Wildschut et al., 2005) and 3 are retrospective consecutive case control and case series from Sweden (Wiberg et al., 2010), Australia (White et al., 2010) and Finland (Keski-Nisula et al., 2012). The remaining 3 included studies are retrospective comparative observational studies, 2 of which are from the UK (Svirko et al., 2008; Yeh et al., 2012) and one from Australia (Hefler et al., 2007).
The systematic review (Malin et al., 2011) incorporated trials that evaluated the predictive value of either arterial or venous cord blood pH, or base excess for neonatal outcomes. Two studies examined whether umbilical artery cord pH is related to child and adult intelligence (Svirko et al., 2008; Hefler et al., 2007). One study evaluated the impact of introducing universal umbilical cord blood gas analysis at birth on perinatal outcomes (White et al., 2010). The remaining included studies examined the association between umbilical cord pH at birth with short and long term neonatal outcomes.
One study (Wildschut et al., 2005) reported excluding women with multiple pregnancies or who had a complicated pregnancy, but in the remaining studies the inclusion/exclusion criteria and characteristics of the study populations are poorly reported. Therefore, it is not possible to judge whether women would have been classified as low risk prior to the onset of labour. Clinical outcomes for the woman are not reported in any of the included studies for this comparison.
Umbilical artery pH was used as selection criterion in all included studies. It was optimal that both arterial and venous samples be obtained as this allows confirmation of which vessel was sampled. The difference between umbilical arterial pH value and umbilical venous pH measured immediately after birth as an indicator of neonatal outcomes was not investigated in any included studies.
While 4 studies (Hefler et al., 2007; Keski-Nisula et al., 2012; Malin et al., 2010; Wildschut et al., 2005) did not specify the time of cord clamping, 2 studies (Svirko et al., 2008; Yeh et al., 2012) stated that the umbilical cord was clamped immediately after birth and others (White et al., 2010; Wiberg et al., 2010) indicated that the cord was clamped “immediately after birth preferably before the newborn's first breath”.
14.4.3. Evidence profile
Data are reported in GRADE profiles below for the following tests and outcomes:
- The odds ratios of association between umbilical arterial blood pH at birth and neonatal outcomes
- Comparative clinical outcome data for association between umbilical arterial blood pH at birth with short and long term neonatal/child outcomes
- Effect of Introducing universal umbilical cord blood gas analysis at birth on neonatal outcomes
- Correlation of neonatal umbilical arterial blood pH with intellectual function.
Evidence from prospective comparative observational studies or prospective consecutive case series was initially graded as high quality and was then downgraded if there were any issues identified that would undermine the trustworthiness of the findings. Evidence from retrospective comparative observational studies or retrospective consecutive case series started at moderate quality and was then downgraded if there were any issues. Evidence from non-consecutive case series was initially graded as low quality and then downgraded if there were any issues.
14.4.3.1. Predictive accuracy and correlation data
In the following table (table 173), the predictive accuracy data (data to generate a 2×2 table: true positive, false positive, true negative, false negative) were used to calculate the odds of an association between the different pH thresholds and the specific neonatal outcome.
Table 173
Summary GRADE profile showing the odds ratios of association between umbilical arterial blood pH at birth and neonatal outcomes.
Table 174
Summary GRADE profile for association between umbilical arterial blood pH at birth with short and long term neonatal/child outcomes.
Table 175
Summary GRADE profile for the effect of introducing universal umbilical arterial blood gas analysis at birth on perinatal outcome.
Table 176
Summary GRADE profile for correlation of umbilical arterial blood pH and intellectual function.
14.4.4. Evidence statements
14.4.4.1. Predictive accuracy of umbilical arterial blood sampling for neonatal outcomes
There is evidence of a positive association between an arterial cord pH less than 7.00 and neonatal mortality (n=464,663), hypoxic ischaemic encephalopathy and seizure (n=63). However, no association was found between a pH of less than 7.00 and cerebral palsy (n=202). Evidence from a meta-analysis of 3 studies (1,888) found no association between a pH of 7.10–7.20 and neonatal mortality. Using a threshold of a pH of 7.00 or lower (n=14,453), pH less than 7.05 (n=12,929) and pH less than 7.10 (n=12,935) evidence from the studies found a positive association between the cord pH and hypoxic ischaemic encephalopathy. There was evidence from a meta-analysis of 4 studies showed that a pH threshold of less than 7.00 (n=3027) or less than 7.20 (n=1699) had a positive association with seizure. The evidence was of low and very low quality.
14.4.4.2. Association between umbilical arterial blood pH at birth with short and long term neonatal/child outcomes
There was evidence from 1 study (n=51,519) that a higher number of neonates with lower pH (7.0 or less, 7.01–7.10) had encephalopathy with seizures and/or death and admission to a neonatal intensive care unit compared with neonates born with higher pH (7.26–7.30). Evidence from the same study found no difference in the rate of encephalopathy with seizures and/or death in neonates born with a pH of 7.11–7.20 and 7.21–7.25 compared with neonates born with pH 7.26–7.30.
Evidence from 1 study (n=1,236) found no significant effect on overall performance and overall intelligence at the age of 18 in young people born with an umbilical artery pH of less than 7.12 compared with those born with an umbilical cord pH of 7.21 or higher. Another study found no significant effect on the development of asthma, allergic rhinitis and atopic eczema at the age of 5 to 6 years, as a composite outcome, in children born with an umbilical cord pH of less than 7.19 compared with those born with an umbilical artery pH of 7.26–7.29. However, more children born with a pH of less than 7.12 developed asthma at the age of 5 to 6 years compared with children born with a pH of 7.26–7.29.
One small study (n=43) found no association between arterial cord pH (thresholds of less than 7.1 compared with 7.1 or higher) and neurodevelopment and movement at 3 months.
Evidence from 1 study (n=6714) found introducing universal umbilical artery blood gas analysis at birth had no significant effect on rate of neonatal intensive care unit admission at birth. The evidence across all outcomes was of very low quality.
14.4.4.3. Correlation of umbilical arterial blood pH with intellectual function
Evidence from 1 study (n=316) found low and very low correlation between pH (7 or higher) and cognitive measures at age 6 to 8 years. The evidence was of very low quality.
14.4.5. Health economics profile
No published economic evaluations were identified for this question.
14.4.6. Evidence to recommendations
14.4.6.1. Relative value placed on the outcomes considered
For this review, the guideline development group had hoped to find studies investigating the difference between umbilical arterial and venous blood pH as a predictor for poor outcomes.
However, all of the included studies had only looked at the arterial blood results – paired gases had only been obtained in order to ensure that an arterial sample had been taken. As a result, the group focussed on the arterial blood pH values as predictors of poor neonatal outcomes. In particular, the group felt evidence relating to severe neonatal morbidity such as hypoxic ischaemic encephalopathy and seizures was particularly important. In addition, the group also focussed on evidence relating to longer term outcomes such as developmental delay.
14.4.6.2. Consideration of clinical benefits and harms
The studies consistently reported significantly poorer neonatal outcomes for babies with an umbilical arterial blood pH of less than 7.00, including mortality, hypoxic ischaemic encephalopathy, seizures and admission to a neonatal unit. In addition, a number of the studies reported significantly poor outcomes for babies with an arterial blood gas pH of less than 7.10. Given this, the group felt confident that a finding of a low pH is a clear indicator of a baby born in poor condition. The group agreed that identifying when a baby is at increased risk of encephalopathy and seizure is clinically valuable as it then allows for prompt treatment, which may include therapeutic cooling, to try to improve the baby's condition and prevent potentially serious adverse outcomes.
14.4.6.3. Consideration of health benefits and resource uses
As noted, the benefit of taking a cord blood analysis is that it allows clinicians to identify babies who are at risk of poor outcomes, and therefore provides the opportunity for them to perform further interventions to try to prevent those outcomes from occurring. Given the potential QALY gain associated with preventing cases of hypoxic ischaemic encephalopathy and subsequently cerebral palsy, there is therefore a strong health economic case for taking cord blood gases.
The guideline development group discussed whether it would be appropriate to recommend universal cord blood analysis. The evidence for this approach was of poor quality and the group did not feel it was appropriate to place much value on it (see below). They noted that current practice is only to perform cord blood analysis if there has been concern about the baby's condition. Given the limitations of the evidence, the group agreed that it would not be appropriate to recommend performing cord blood analysis in all babies.
14.4.6.4. Quality of evidence
The guideline development group had hoped to find more evidence for outcomes associated with an umbilical cord blood pH in the range 7.00 to 7.10 since this is an area where there is greater clinical uncertainty. The lack of evidence in this area meant the group had to extrapolate from evidence relating to other thresholds plus their own clinical judgement to help draft recommendations.
The group considered that not all the included studies had performed an adjustment for confounding factors, which can lead to a biased estimate of the effect.
The group noted that 1 of the included studies (Hefler et al., 2007) excluded from its analysis babies with a pH of less than 7.00 and babies who had died. In addition, its long-term analysis at 18 years was only conducted in males (army recruits) without severe mental or physical disability. Given all of these issues, the group felt that the findings were biased and unlikely to be generalisable to the population as a whole and so disregarded the study from their discussions.
One review (White et al., 2010) compared various outcomes before and after the introduction of universal umbilical arterial blood gas analysis. This study found a statistically significant reduction in the rate of special care nursery admission. However, the group noted that the proportion of babies admitted to special care was unusually high in this study and felt that although the study had tried to adjust for confounding features, given the time difference between the 2 study periods, it was extremely difficult to be sure that the finding was caused by the universal analysis as opposed to another factor. Given this, the group did not attach much value to the finding and did not feel that it provided compelling evidence to move to a practice of cord blood analysis in all babies.
The group noted the findings from 1 study (Svirko et al., 2008) which showed that there was low or very low correlation between various cognitive measures and pH values of 7.00 or higher. The study excluded neonates with a pH of less than 7.00 and did not report further subgroup analysis by different pH values. Given this, the group felt that the findings were not helpful in identifying whether pH values were predictive of longer term outcomes.
14.4.6.5. Other considerations
The guideline development group discussed whether double cord clamping was necessary in order to obtain a cord blood sample. Some group members were aware of research which has suggested that the blood sample could be taken without clamping the umbilical cord. However, as current UK practice is to double clamp the cord and the group had not formally reviewed the literature looking at whether cord clamping is necessary, they did not feel it was appropriate to recommend a change of practice in this area.
The group also discussed the appropriate time to clamp the cord in light of the updated recommendations for routine management of the third stage, which now state that clamping should not be performed until 1 minute after the birth of the baby (see section 13.3.6). The group agreed that even when there was concern about the condition of the baby, it would be appropriate to wait 1 minute before clamping the cord. Based on their clinical knowledge, they did not think that waiting for 1 minute would affect the blood sample but thought the wait would benefit the baby. Furthermore, they agreed that in the first minute following birth, the healthcare professional should focus on assessing the needs of baby rather than taking a cord blood sample.
14.4.6.6. Key conclusions
The guideline development group agreed that it is clinically valuable to know the umbilical arterial pH at birth in order to provide additional information that can help plan appropriate neonatal care for babies where there is a suspicion that there has been a period of hypoxia.
14.4.7. Recommendations
- 268.
Record the time from birth to the onset of regular respirations. [new 2014]
- 269.
If the baby is born in poor condition (on the basis of abnormal breathing, heart rate or tone):
- follow recommendations 262 to 267 on neonatal resuscitation and
- take paired cord-blood samples for blood gas analysis, after clamping the cord using 2 clamps.
Continue to evaluate and record the baby's condition until it is improved and stable. [new 2014]
- 270.
Do not take paired cord blood samples (for blood gas analysis) routinely. [new 2014]
- 271.
Ensure that a second clamp to allow double-clamping of the cord is available in all birth settings. [2014]
14.5. Care of babies born with meconium-stained liquor
14.5.1. Review question
What is the appropriate care of babies born with meconium-stained liquor?
For further details on the evidence review protocol, please see appendix E.
14.5.2. Description of included studies
Five studies are included in this review (Daga et al., 1994; Liu and Harrington, 1998; Linder et al., 1988; Vain et al., 2004; Wiswell et al., 2000). These papers represent the same evidence that was available for the 2007 guideline.
All of the included studies were randomised controlled trials, although in 1 of the studies it was the paediatricians who were randomised rather than the babies (Linder et al., 1988). Two of the studies were conducted in the USA (Liu and Harrington, 1998; Wiswell et al., 2000) and one was conducted in India (Daga et al., 1994). One further trial was conducted primarily in Argentina but with 1 centre (out of 12) located in the USA (Vain et al., 2004). The last trial did not report the study location but the authors came from Canada and Israel (Linder et al., 1988).
One of the trials compared a policy of suctioning of the oropharynx and nasopharynx before delivery of the shoulders with a policy of no suctioning (Vain et al., 2004). Four of the trials compared a policy of routine intubation, with one of no intubation (Daga et al., 1994; Liu and Harrington, 1998; Linder et al., 1988) or one of intubation only when indicated (Wiswell et al., 2000). In the latter 4 studies, the protocol was for all babies to have oronasopharyngeal suctioning and for only babies who were vigorous at birth to be included (for full details of inclusion and exclusion criteria, see the evidence tables in appendix I).
One trial only included babies born through non-significant (‘thin’) meconium (Liu and Harrington, 1998), 1 trial only included babies born through significant (‘thick’) meconium (Daga et al., 1994) and 3 trials included babies born through meconium of any consistency (Linder et al., 1988; Vain et al., 2004; Wiswell et al., 2000). Four trials restricted their study populations to term babies, but in the fifth trial 45% of babies were born at 34–37 weeks (Daga et al., 1994). No details were reported about cord clamping in the included studies and therefore it was not possible to evaluate what impact this might have had on outcomes.
14.5.3. Evidence profile
The following GRADE profiles are presented below:
- suctioning of the oropharynx and nasopharynx before delivery of the shoulders compared with no suctioning
- intubation compared with no intubation or selective intubation.
Subgroup analyses by the degree of meconium staining have been reported where the data were available in the trials. (Vain et al. [2004] reported a subgroup analysis for 3 outcomes only).
Sensitivity analyses were performed, removing in turn Daga et al. (1994) which had a high proportion of preterm babies and Linder et al. (1988) which randomised paediatricians rather than babies and included babies treated by non-participating physicians in the intubation arm. Unless indicated in a footnote, the sensitivity analyses did not change the direction or significance of the demonstrated effects.
Table 177
Summary GRADE profile for comparison of suctioning of the oropharynx and nasopharynx before delivery of the shoulders with no suctioning.
Table 178
Summary GRADE profile for comparison of routine intubation with no intubation or selective intubation.
14.5.4. Evidence statements
14.5.4.1. Suctioning of the oropharynx and nasopharynx before delivery of the shoulders compared with no suctioning
One large, well-conducted randomised controlled trial (n=2514) did not find a difference in the incidence of neonatal death, meconium aspiration syndrome, other respiratory disorders, need for intubation, suction or positive pressure ventilation in the delivery room, use of mechanical ventilation for meconium aspiration syndrome and pneumothorax between babies randomised to receive suctioning before delivery of the shoulders and those randomised to no suctioning. Similarly, when considering duration of oxygen treatment, mechanical ventilation and hospital care among babies with meconium aspiration syndrome, no difference was found between the 2 groups. Where reported, the evidence was consistent for both subgroups of babies born through non-significant (‘thin’) or ‘moderately’ stained meconium liquor and those born through significant (‘thick’) meconium. The evidence was of high to very low quality.
14.5.4.2. Intubation compared with no intubation or selective intubation
Evidence from 4 studies (n=2884) showed no clinical benefit of a routine intubation policy for meconium-stained liquor for any of the outcomes considered. This was consistent for the subgroup analysis of trials including only babies born through non-significant (‘thin’) meconium and the subgroup analysis of trials including only babies born through significant (‘thick’) meconium, although there were only a small number of babies in the latter group. The evidence was of moderate to very low quality.
14.5.4.3. Health economics profile
No published health economic evaluations were identified for this question.
14.5.5. Evidence to recommendations
14.5.5.1. Relative value placed on the outcomes considered
For this review, the guideline development group prioritised outcomes relating to neonatal mortality and morbidity – in particular, respiratory morbidity. They had hoped that longer term outcomes would also be available but none were reported in the studies. They had also hoped that evidence would be available relating to the timing of cord clamping for these babies. However, despite conducting a broad search, no data were identified. The group was also frustrated that no new evidence had become available since the publication of the previous guideline.
14.5.5.2. Consideration of clinical benefits and harms
The guideline development group recognised that neither suctioning the oropharynx and nasopharynx before delivery of the shoulders nor routine intubation appeared to make a statistically or clinically significant difference to any outcomes for babies born with meconium stained liquor.
The group noted that the overall incidence of poor outcomes was low as the trials all explicitly excluded babies that were born in a depressed state. They felt that this indicated a gap in the evidence as it is not clear whether either suctioning or intubation would be of benefit to babies born in poor condition (that is, babies born with poor tone, having difficulty with respiration or with reduced heart rate). The group acknowledged that there are guidelines produced by the Resuscitation Council (UK) regarding the appropriate methods for neonatal resuscitation. Given the lack of evidence for babies born without normal tone and respiration, the group agreed that it was appropriate that healthcare professionals follow the relevant guideline from the Resuscitation Council UK which acknowledges that, for these babies, early laryngoscopy and suction under direct vision may be appropriate.
14.5.5.3. Consideration of health benefits and resource uses
There were no specific resource use issues addressed for this question and there was limited evidence of effectiveness. However, the guideline development group recognised the general principle that it is not a good use of resources for clinicians to perform interventions unnecessarily. This therefore supported their view to recommend that suctioning should not be performed and that intubation should not be performed routinely.
14.5.5.4. Quality of evidence
Overall, the evidence was of mixed quality. The study from Vain et al. (2004) was relatively large, and was graded high or moderate quality for most reported outcomes. As a result, the guideline development group had confidence in the findings for the comparison of suctioning the upper airways compared with no suctioning (although, as noted above, they recognised that this evidence did not include babies that were born in a depressed state). However, the quality of the evidence for the comparison of routine intubation with no, or selective, intubation was generally of lower quality, with the majority of studies graded as low or very low quality for most outcomes. Ultimately, though, as no differences were observed between the groups for any of the outcomes, the quality of the evidence was less of a concern as the group did not wish to make a positive recommendation in favour of intubation.
14.5.5.5. Other considerations
In updating the recommendations from the 2007 guideline, the guideline development gorup agreed that the definition of ‘significant meconium’ should be clearer and so made a recommendation to this effect.
The previous recommendations had mainly been written from the perspective of giving birth in an obstetric unit. However, the group noted that recommendations were also required for women giving birth outside a hospital setting with significant meconium-stained liquor. The group agreed that in this situation, it would be appropriate to make preparations for transfer but that transfer should only occur if the healthcare professional believes that it can be done before the birth occurs. This is because, in the group's clinical judgement, the woman and baby are likely to be at risk of greater harm if the birth occurs in an ambulance than if the birth occurs at home or in a freestanding midwifery unit.
The group agreed that the recommendations saying not to use suctioning of the upper airways or intubation should specify that this is in babies born with normal tone and heart rate. The group was aware that guidelines from the Resuscitation Council (UK) indicate that for babies without normal tone or respiration, it can be appropriate to perform early laryngosocopy and suction under direct vision. The group felt that this matched their own clinical experience and so agreed that this was a sensible recommendation to make.
14.5.6. Recommendations
- 272.
In the presence of any degree of meconium:
- do not suction the baby's upper airways (nasopharynx and oropharynx) before birth of the shoulders and trunk
- do not suction the baby's upper airways (nasopharynx and oropharynx) if the baby has normal respiration, heart rate and tone
- do not intubate if the baby has normal respiration, heart rate and tone. [new 2014]
- 273.
If there has been significant meconium (see recommendation 164) and the baby does not have normal respiration, heart rate and tone, follow nationally accredited guidelines on neonatal resuscitation, including early laryngoscopy and suction under direct vision. [new 2014]
- 274.
If there has been significant meconium and the baby is healthy, closely observe the baby within a unit with immediate access to a neonatologist. Perform these observations at 1 and 2 hours of age and then 2-hourly until 12 hours of age. [new 2014]
- 275.
If there has been non-significant meconium, observe the baby at 1 and 2 hours of age in all birth settings. [new 2014]
- 276.
If any of the following are observed after any degree of meconium, ask a neonatologist to assess the baby (transfer both the woman and baby if they are at home or in a freestanding midwifery unit, following the general principles for transfer of care described in recommendations 46 to 50):
- respiratory rate above 60 per minute
- the presence of grunting
- heart rate below 100 or above 160 beats/minute
- capillary refill time above 3 seconds
- body temperature of 38°C or above, or 37.5°C on 2 occasions 30 minutes apart
- oxygen saturation below 95% (measuring oxygen saturation is optional after non-significant meconium)
- presence of central cyanosis, confirmed by pulse oximetry if available. [new 2014]
- 277.
Explain the findings to the woman, and inform her about what to look out for and who to talk to if she has any concerns. [new 2014]
14.6. Care of babies born to women with prelabour rupture of the membranes at term
14.6.1. Prolonged rupture of membranes and intrapartum fever as risk factors of neonatal infection
14.6.1.1. Description of included studies
There was one cohort study within a randomised controlled trial300 and six observational studies that were identified.451–456 Among them, two were conducted in the UK.453,455 All the studies, except for one,456 investigated GBS-related disease as an outcome.
14.6.1.2. Review findings
Babies of women with PRoM, who enrolled in the international, multicentre RCT comparing induction of labour and expectant management, were observed to investigate various risk factors for developing neonatal infection.300 [EL = 2+] Multivariate analysis showed the following as risk factors for neonatal infection: clinical chorioamnionitis (OR 5.89, P < 0.001); positive maternal GBS status (versus negative or unknown, OR 3.08, P < 0.001); seven to eight vaginal digital examinations (versus 0 to 2, OR 2.37, P = 0.04); 24 to less than 48 hours from membrane rupture to active labour (versus less than 12 hours, OR 1.97, P = 0.02); 48 hours or less from membrane rupture to active labour (versus less than 12 hours, OR 2.25, P = 0.01); and maternal antibiotics before birth (OR 1.63, P = 0.05).
A UK cross-sectional study was conducted in 2000/2001 involving all babies with GBS disease in the UK and Ireland, younger than 90 days.453 [EL = 3] Among the total of 568 babies, incidence of GBS disease was assumed to be 0.72 per 1000 live births [95% CI 0.66 to 0.78]. Mothers of 140 babies (44%) had prolonged rupture of membranes.
A UK case–control study was conducted between 1998 and 2000.455 [EL = 2+] A total of 37 cases of early onset neonatal GBS sepsis were compared with 147 hospital controls. A logistic regression analysis showed that risk of developing early onset neonatal GBS sepsis for babies from women with prolonged rupture of membranes longer than 18 hours was RR 4.8 [95% CI 0.98 to 23.1], and with rupture of membranes before onset of labour: RR 3.6 [95% CI 0.7 to 17.6].
A Danish cross-sectional study was conducted between 1992 and 2001.454 [EL = 3] A total of 61 babies with blood-culture-positive GBS sepsis/meningitis were investigated (incidence 0.76 per 1000 live births [95% CI 0.0 to 1.91]). Nineteen percent of the babies had a mother with prolonged rupture of membranes (longer than 18 hours) and 16% of those had maternal pyrexia (higher than 38 °C).
A Dutch case–control study was conducted between 1988 and 1995.451 [EL = 2+] A total of 41 neonatal early onset GBS-related cases were compared with 123 hospital controls. A multivariate analysis showed that there was an increased risk of developing early onset GBS-related disease when maternal temperature increased by 0.1 above 37.4 °C (OR 2.0 [95% CI 1.4 to 2.8]), but there was no evidence of association between interval from rupture of membranes to birth (OR per hour between 8 and 24 hours 1.0 [95% CI 0.92 to 1.1]) and prolonged rupture of membranes (OR 2.0 [95% CI 0.47 to 9.6]).
A US cohort study was conducted in 1987/88.456 [EL = 2−] Babies of 205 women with a history of prolonged rupture of membranes were compared with 8586 babies of women without a history of prolonged rupture of membranes. Among 175 out of 205 babies following prolonged rupture of membranes of 24 hours or more, 8.2% yielded positive blood culture. In comparison, 0.1% had positive blood culture from the remaining 8586 babies of women without prolonged rupture of membranes.
A US case–control study was conducted between 1991 and 1992.452 [EL = 2+] Ninety-nine cases of early onset GBS disease were compared with 253 matched hospital controls. A multivariate logistic regression analysis showed strong evidence of association between increased risk of developing early onset GBS disease and prolonged rupture of membranes (OR 8.7, P < 0.001) and intrapartum fever (OR 4.3, P < 0.05).
14.6.1.3. Evidence statement
There is medium-level evidence that risk of developing early onset GBS-related disease, for babies born to women with prolonged rupture of membranes, ranges between 2.0 and 8.7 times higher than those born to women without. The risk of developing fever is about four-fold higher in babies born to women with PRoM when compared with babies born to women without. Up to 40% of babies with early onset GBS-related disease were born to women with prolonged rupture of membranes in the UK.
14.6.2. Clinical manifestation of babies
14.6.2.1. Description of included studies
One cohort study and two case series were identified, all of which were conducted in the USA,456–458 One study compared laboratory test results between symptomatic and asymptomatic babies.456 The other two studies investigated time of onset of symptoms for neonatal infection.
14.6.2.2. Review findings
14.6.2.2.1. Symptoms and laboratory tests
One cohort study was conducted in the USA.456 [EL = 2+] In the 175 babies born to women with prolonged rupture of membranes, using blood culture and complete blood counts results, six symptomatic infants were compared with nine asymptomatic babies. Out of the six symptomatic babies, all had abnormal complete blood counts (two with abnormal white blood cell counts; five with abnormal neutrophil count; four with high band/metamyelocyte count; four with increased immature to total neutrophil ratio). Of the nine asymptomatic babies, seven had abnormal complete blood counts, five with a high white blood cell count, five with a high neutrophil count, two had a high band/metamyelocyte count and one with a high immature to total neutrophil ratio. The sensitivity of the complete blood count was 86% and specificity 66%.
14.6.2.2.2. Onset of symptoms
The other two studies investigated time of onset of symptoms for early onset neonatal GBS disease. The first study was conducted between 1995 and 1996, targeting babies with 2000 g birthweight or more.458 [EL = 3] The study reported that 75.8% of babies with sepsis were first noted to be at risk for sepsis before or at the moment of birth, and 91.2% were identified by 12 hours of age. The second study specifically investigated early onset GBS disease.457 [EL = 3] The population included 37% of preterm babies. The study reported that the median age at onset was 20 minutes ranging from 0 to 77 hours. Sixty-three percent of the babies showed clinical signs within 1 hour of age and 90% were symptomatic within 12 hours.
14.6.2.3. Evidence statement
There is low-level evidence that over 90% of neonatal sepsis presents within 12 hours of age. The majority of babies with sepsis were first noted to be at risk before or at the moment of birth. There is insufficient evidence on the diagnostic value of tests for neonatal sepsis.
14.6.3. Postnatal prophylactic antibiotics for babies
14.6.3.1. Description of included studies
One systematic review with two trials459 and one observational study458 were identified. One of the trials included assessed effectiveness of prophylactic antibiotics on babies born to women with GBS colonisation, hence excluded from this review. The other trial investigated effectiveness of prophylactic antibiotics (intramuscular penicillin and kanamycin for 7 days, n = 24), compared with no prophylactics (n = 25).459 [EL = 1−] The second study, a population-based cohort study in the USA, investigated the relationship between predictors and neonatal bacterial infection.458 [EL = 2+]
14.6.3.2. Review findings
The trial that investigated the effectiveness of prophylactic antibiotics compared with no antibiotics reported no neonatal mortality. It was underpowered to show any differences in incidence of neonatal sepsis (RR 0.12 [95% CI 0.01 to 2.04]).
The US cohort study evaluated 2785 out of 18,299 newborns of 2000 g or more, without major abnormalities for sepsis, with a complete blood count and/or blood culture. Multivariate analysis showed that among 1568 babies whose mothers did not receive antibiotics, initial asymptomatic status was associated with decreased risk of infection (OR 0.27 [95% CI 0.11 to 0.65]). However, there was evidence of an increased risk of neonatal sepsis by antepartum fever (highest ante-partum temperature 101.5 °F (38.6 °C) or higher (OR 5.78 [95% CI 1.57 to 21.29]), rupture of membranes for 12 hours or longer (OR 2.05 [95% CI 1.06 to 3.96]), low absolute neutrophil count for age (OR 2.82 [95% CI 1.50 to 5.34]), and meconium in amniotic fluid (OR 2.24 [95% CI 1.19 to 4.22]).
14.6.3.3. Evidence statement
There is no high-level evidence from trials on prophylactic antibiotics for babies born to women with prolonged rupture of membranes at term.
There is medium-level evidence that, if the baby is asymptomatic at birth, there is a significantly lower risk of it developing neonatal sepsis.
14.6.3.4. Evidence to recommendations
As part of the process of updating the guideline the GDG considered the applicability of all recommendations for each of the four birth settings. The recommendations for care of babies following prolonged rupture of membranes was amended through GDG consensus and it was agreed these should be aligned with the updated recommendations made for observations of the newborn following birth with meconium. In addition the GDG discussed the practicalities of making additional observations in settings outside an obstetric unit and agreed a regimen for making observations that was felt to be both safe and practical.
14.6.3.5. Recommendations
- 278.
Closely observe any baby born to a woman with prelabour rupture of the membranes (more than 24 hours before the onset of established labour) at term for the first 12 hours of life (at 1 hour, 2 hours, 6 hours and 12 hours) in all settings. Include assessment of:
- temperature
- heart rate
- respiratory rate
- presence of respiratory grunting
- significant subcostal recession
- presence of nasal flare
- presence of central cyanosis, confirmed by pulse oximetry if available
- skin perfusion assessed by capillary refill
- floppiness, general wellbeing and feeding.
If any of these are observed, ask a neonatologist to assess the baby (transfer both the woman and baby if they are at home or in a freestanding midwifery unit, following the general principles for transfer of care described in recommendations 46 to 50). [new 2014]
- 279.
If there are no signs of infection in the woman, do not give antibiotics to either the woman or the baby, even if the membranes have been ruptured for over 24 hours. [2007]
- 280.
If there is evidence of infection in the woman, prescribe a full course of broad-spectrum intravenous antibiotics. [2007]
- 281.
Advise women with prelabour rupture of the membranes to inform their healthcare professionals immediately of any concerns they have about their baby's wellbeing in the first 5 days after birth, particularly in the first 12 hours when the risk of infection is greatest. [2007]
- 282.
Do not perform blood, cerebrospinal fluid and/or surface culture tests in an asymptomatic baby. [2007]
- 283.
Refer a baby with any symptom of possible sepsis, or born to a woman who has evidence of chorioamnionitis, to a neonatal care specialist immediately. [2007]
14.7. Mother–infant bonding and promoting breastfeeding
14.7.1. Introduction
Immediate skin-to-skin contact of mothers and babies to promote bonding and breastfeeding was reviewed in the NICE Postnatal Care guideline.414 For ease the relevant recommendations from that guideline are reproduced.
14.7.2. Review question
Are there effective ways of encouraging mother–infant bonding following birth?
- Including skin to skin contact with mothers, breastfeeding.
14.7.3. Description of included studies
There was one systematic review identified that considered intrapartum interventions for promoting the initiation of breastfeeding, although there was no relevant intervention that this guideline covers.424
14.7.4. Recommendations on initial assessment of the baby and mother-infant bonding
- 284.
Record the Apgar score routinely at 1 and 5 minutes for all births. [2007]
- 285.
Encourage women to have skin-to-skin contact with their babies as soon as possible after the birth.u [2007]
- 286.
In order to keep the baby warm, dry and cover him or her with a warm, dry blanket or towel while maintaining skin-to-skin contact with the woman. [2007]
- 287.
Avoid separation of a woman and her baby within the first hour of the birth for routine postnatal procedures, for example, weighing, measuring and bathing, unless these measures are requested by the woman, or are necessary for the immediate care of the baby.v [2007]
- 288.
Encourage initiation of breastfeeding as soon as possible after the birth, ideally within 1 hour.w [2007]
- 289.
Record head circumference, body temperature and birth weight soon after the first hour following birth. [2007]
- 290.
Undertake an initial examination to detect any major physical abnormality and to identify any problems that require referral. [2007]
- 291.
Ensure that any examination or treatment of the baby is undertaken with the consent of the parents and either in their presence or, if this is not possible, with their knowledge. [2007]
14.8. Initial assessment of the mother following birth
14.8.1. Introduction
Appropriate maternal observations immediately after birth are discussed in this section. Advice on further appropriate maternal observations thereafter in the postnatal period are discussed in the NICE Postnatal Care guideline.414
14.8.2. Review question
Is there evidence that the assessment of the following, on admission, and throughout labour and the immediate postnatal period, affect outcomes?
- observation of vital signs.
14.8.3. Description of included studies
There was no relevant study identified to investigate effectiveness of each component of maternal observations immediately following birth.
14.8.4. Evidence statement
There is no high-level study investigating appropriate maternal observations immediately after birth.
14.8.5. Recommendation on initial assessment of the mother
- 292.
Carry out the following observations of the woman after birth:
- Record her temperature, pulse and blood pressure. Transfer the woman (with her baby) to obstetric-led care if any of the relevant indications listed in recommendation 164 are met.
- Uterine contraction and lochia.
- Examine the placenta and membranes: assess their condition, structure, cord vessels and completeness. Transfer the woman (with her baby) to obstetric-led care if the placenta is incomplete.
- Early assessment of the woman's emotional and psychological condition in response to labour and birth.
- Successful voiding of the bladder. Assess whether to transfer the woman (with her baby) to obstetric-led care after 6 hours if her bladder is palpable and she is unable to pass urine.
If transferring the woman to obstetric-led care, follow the general principles for transfer of care described in recommendations 46 to 50. [new 2014]
14.9. Perineal care
14.9.1. Previous guideline
No previous guidelines have considered interventions related to perineal or genital care immediately following childbirth.
14.9.2. Definition of perineal or genital trauma
14.9.2.1. Review question
What is the appropriate definition of perineal or genital trauma?
14.9.2.2. Overview of available evidence and evidence statement
The GDG discussed this and reached consensus to use the following recommendation for the definition of perineal or genital trauma, taken from the Green Top Guideline by the Royal College of Obstetricians and Gynaecologists on methods and materials used in perineal repair.425
14.9.2.3. Recommendation on definition of perineal/genital trauma
- 293.
Define perineal or genital trauma caused by either tearing or episiotomy as follows:
- first degree – injury to skin only
- second degree – injury to the perineal muscles but not the anal sphincter
- third degree – injury to the perineum involving the anal sphincter complex:
- 3a – less than 50% of external anal sphincter thickness torn
- 3b – more than 50% of external anal sphincter thickness torn
- 3c – internal anal sphincter torn.
- fourth degree – injury to the perineum involving the anal sphincter complex (external and internal anal sphincter) and anal epithelium. [2007]
14.9.3. Assessment of perineal trauma
14.9.3.1. Review question
Is there evidence that the type of assessment used to identify perineal or genital trauma affects outcomes?
14.9.3.2. Description of available evidence
Three studies are reviewed in this subsection. The first is an evaluation of a perineal assessment and repair course. The other two prospective intervention studies examine the incidence of third- and fourth-degree perineal trauma and highlight under-diagnosis as a problem in this aspect of care.
14.9.3.3. Review findings
A recent UK before and after study evaluated the effectiveness of a perineal repair course.426 [EL = 2+] The one-day course included lectures, video demonstrations and hands-on teaching of rectal examination and suturing skills using foam pads and models. Participants completed a self-assessment questionnaire prior to the course and 8 weeks afterwards. Findings for the evaluation are based on responses to 147 pairs of pre- and post-course questionnaires (response rate = 71%). Most respondents were midwives (95%), 68% of whom had been qualified for more than 5 years. Seven junior doctors and three students also attended the courses. Following attendance at the course, self-assessed responses showed an improvement in the correct classification of tears depending upon degree of anal sphincter injury: external anal sphincter (EAS) partially torn: 77% versus 85%, P = 0.049; EAS completely torn: 70% versus 85%, P = 0.001; internal anal sphincter (IAS) exposed but not torn: 63% versus 82%, P < 0.001; IAS torn: 45% versus 67%, P < 0.001; anal sphincter and mucosa torn: 80% versus 89%, P = 0.031. There was also a significant change in practice reported with more respondents performing a rectal examination prior to repairing perineal trauma after attending the course: 28% versus 89%, P < 0.001, McNemar's test). There was also a significant shift in favour of a continuous suture to the perineal muscle and skin: continuous suture to muscle: 32% versus 84%, P < 0.001; continuous suture to skin 39% versus 81%, P < 0.001. The paper does not mention two-stage perineal repair as an option.
A prospective intervention study recently conducted in the UK involved re-examination by an experienced research fellow of nulliparous women who sustained perineal trauma in order to ascertain the prevalence of clinically recognisable and true occult anal sphincter injuries.427 [EL = 2+] Women were initially assessed by the attending clinician. Where obstetric anal sphincter injuries (OASIS) were identified, this was confirmed by a specialist registrar or consultant. All participating women (n = 241; response rate = 95%) had an endoanal ultrasound scan performed immediately following birth (prior to suturing). Most of these women (n = 208 (86%)) attended for a repeat ultrasound scan at 7 weeks postpartum. One hundred and seventy-three of the 241 births were attended by midwives, 75% of these births being attended by midwives with at least 5 years of experience. Of the 68 births attended by obstetricians, 63 were instrumental births. The prevalence of OASIS increased significantly from 11% to 24.5% when women were re-examined by the research fellow. Of the 173 births attended by midwives, eight women were diagnosed as having sustained an OASIS. Only four of these were confirmed by the research fellow. Of the remaining 26 women who sustained OASIS, the midwife made a diagnosis of second-degree tear in 25 cases and first-degree tear in one case. All 30 incidents of OASIS were confirmed by the specialist registrar/consultant. Of the 68 births attended by obstetricians, 22 women (32%) had OASIS diagnosed and confirmed by the research fellow. A further seven cases of OASIS were identified by the research fellow, three of these cases had been missed by the duty specialist registrar but were subsequently confirmed by the specialist consultant. Of the 68 births attended by an obstetrician, the midwife caring for the woman was also asked to perform an examination. Only one of the 29 OASIS was identified by a midwife and no midwife performed a rectal examination. All women with OASIS had a defect detected by endoanal ultrasound performed immediately after birth. In addition, there were three defects seen on ultrasound that were not seen clinically. No additional defects were seen at the 7 week follow-up.
A UK prospective observational study was undertaken to assess whether clinical diagnosis of third-degree tears could be improved by increased vigilance in perineal assessment.428 [EL = 3] The study involved assessment of perineal trauma sustained by women having their first vaginal birth at one large teaching hospital. A group of 121 women were assessed initially by the obstetrician or midwife attending the birth and then again by a single independent assessor (a clinical research fellow). Findings from this group were compared with all other women giving birth over the same 6 month period who were assessed by the attending clinician only (i.e. usual care) (n = 362). Both groups were similar for a number of key characteristics, including gestation, mode of birth, analgesia used, duration of labour, birthweight, and head circumference. Episiotomies which extended to involve the anal sphincter were classified as third-degree tears. There were significantly more third-degree tears identified in the assessed group, 14.9%, compared with 7.5% in the control group. The study was underpowered to show statistical significance. In the assessed group, only 11 of the 18 third-degree tears were identified by the clinician attending the birth. Once the diagnosis was made there was no disagreement between attending clinician and research fellow. Third-degree tears were most often associated with instrumental births, especially forceps births. The percentages of women sustaining a third-degree tear for each mode of birth was spontaneous vaginal birth 3.2%, ventouse 14.9% and forceps 22%. Comparing study data with findings for a similar group of women during the 6 months before and after the study period, the overall rates of third-degree tears were before 2.5%, during 9.3%, and after 4.6%, again suggesting that many third-degree tears go undiagnosed.
14.9.3.4. Evidence statement
There is low-level evidence that suggests the systematic assessment of the vagina, perineum and rectum is required to adequately assess the extent of perineal trauma.
There is low-level evidence that current training is inadequate regarding assessment of perineal trauma.
Practitioners who are appropriately trained are more likely to provide a consistent, high standard of perineal care.
14.9.3.5. Recommendations on assessment of perineal trauma
- 294.
Before assessing for genital trauma:
- explain to the woman what is planned and why
- offer inhalational analgesia
- ensure good lighting
- position the woman so that she is comfortable and so that the genital structures can be seen clearly. [2007]
- 295.
Perform the initial examination gently and with sensitivity. It may be done in the immediate period after birth. [2007]
- 296.
If genital trauma is identified after birth, offer further systematic assessment, including a rectal examination. [2007]
- 297.
Include the following in a systematic assessment of genital trauma:
- further explanation of what is planned and why
- confirmation by the woman that tested effective local or regional analgesia is in place
- visual assessment of the extent of perineal trauma to include the structures involved, the apex of the injury and assessment of bleeding
- a rectal examination to assess whether there has been any damage to the external or internal anal sphincter if there is any suspicion that the perineal muscles are damaged. [2007]
- 298.
Ensure that the timing of this systematic assessment does not interfere with mother–baby bonding unless the woman has bleeding that requires urgent attention. [2007]
- 299.
Assist the woman to adopt a position that allows adequate visual assessment of the degree of trauma and for repair. Only maintain this position for as long as necessary for systematic assessment and repair. If it is not possible to adequately assess the trauma, transfer the woman (with her baby) to obstetric-led care, following the general principles for transfer of care described in recommendations 46 to 50. [2007, amended 2014]
- 300.
Seek advice from a more experienced midwife or obstetrician if there is uncertainty about the nature or extent of the trauma. Transfer the woman (with her baby) to obstetric-led care (following the general principles for transfer of care described in recommendations 46 to 50) if the repair needs further surgical or anaesthetic expertise. [2007, amended 2014]
- 301.
Document the systematic assessment and its results fully, possibly pictorially. [2007]
- 302.
All relevant healthcare professionals should attend training in perineal/genital assessment and repair, and ensure that they maintain these skills. [2007]
14.9.4. Perineal repair
14.9.4.1. Review question
Is there evidence that undertaking repair, the timing, analgesia and method and material of perineal repair affect outcomes?
14.9.4.2. Previous guideline
No previous guideline has considered performing perineal repair following childbirth.
14.9.4.3. Undertaking repair
14.9.4.3.1. Description of included studies
Two studies are reviewed under this heading. One RCT compared suturing of first- and second-degree perineal tears with non-suturing, and one qualitative study explored women's experiences of perineal repair.
14.9.4.3.2. Review findings
One UK RCT compared suturing with non-suturing of first- and second-degree perineal tears (SUNS trial).429 [EL = 1+] Randomisation was carried out across two hospital labour wards with stratification for degree of tear to produce a group of nulliparous women who had perineal tears sutured (n = 33) and nulliparous women whose perineal trauma was not sutured (n = 41). Suturing was conducted in accordance with the hospital protocols, which included continuous subcutaneous sutures to the perineal skin. No differences were apparent between trial groups at any time point postnatally regarding level of pain as measured using the McGill Pain Questionnaire. The median total pain scores and point difference in medians for sutured versus unsutured groups were: day 1: 11 [range 0 to 33] versus 10 [range 0 to 44]; 1 [95% CI −2 to 4.999]; day 10: 0 [range 0 to 18] versus 0 [range 0 to 33]; 0 [95% CI 0 to 0.001]; 6 weeks: 0 [range 0 to 28] versus 0 [range 0 to 7]; 0 [95% CI 0 to 0]. Scores obtained using a 10 cm VAS also showed no differences between groups. Healing was measured using a standardised and validated tool, the REEDA scale. Findings showed significantly better wound edge approximation for women in the sutured group (again expressed in terms of median for scores): day 1: 1 [range 0 to 3] versus 2 [range 1 to 3]; -1 [95% CI −1.0001 to 0], P < 0.001; day 10: 1 [range 0 to 2] versus 2 [range 0 to 3]; −1 [95% CI −1.0001 to −0.0003], P = 0.003; 6 weeks: 1 [range 0 to 1] versus 1 [range 0 to 3]; 0 [95% CI −0.9999 to 0.0001], P = 0.001. Total healing scores suggested a tendency towards better wound healing in the sutured group at days 1 and 10: day 1: [range 0 to 9] versus 5 [range 1 to 10]; −1 [95% CI −2 to 0], NS; day 10: 1 [range 0 to 6] versus 2 [range 0 to 8]; 0 [95% CI −1 to 0], NS. At 6 weeks women in the sutured group had significantly better healing scores than those in the unsutured group: 0 [range 0 to 3] versus 1 [range 0 to 3]; 0 [95% CI −1.0001 to −0.0003], P = 0.003. The authors conclude that, despite the small sample size for this trial, the findings show significantly improved healing following perineal suturing compared with non-suturing.
One qualitative study was identified which explored women's experiences of perineal trauma both during its repair and in the immediate postnatal period.430 [EL = 3] This small (n = 6), in-depth, unstructured interview-based study is limited by its reliance on the snowballing technique, which tends to result in a sample of people with similar experiences and/or views. It does, however, highlight the intense and far-reaching effects of bad experiences of care. The importance of interpersonal relationships between women and their carers was illustrated through four emergent themes:
- the importance of communication between women and health professional
- the importance of good pain relief during suturing
- women feeling ‘being patched up’
- women having to endure a procedure that had to be ‘got through’.
Postnatally, women described the feelings associated with coming to terms with perineal trauma. The themes here comprised:
- the severity of negative emotions (anger, upset, frustration)
- concerns about the degree of skill of practitioners
- failing to be heard and taken seriously when there were problems with perineal healing.
14.9.4.3.3. Evidence statement
There is limited high-level evidence that not suturing first- or second-degree perineal trauma is associated with poorer wound healing at 6 weeks.
There is no evidence as to long-term outcomes.
14.9.4.3.4. Recommendations on perineal repair
- 303.
Advise the woman that in the case of first-degree trauma, the wound should be sutured in order to improve healing, unless the skin edges are well opposed. [2007]
- 304.
Advise the woman that in the case of second-degree trauma, the muscle should be sutured in order to improve healing. [2007]
14.9.4.4. Timing of repair
14.9.4.4.1. Description of included studies
No study was identified which considered the timing of perineal repair following childbirth.
14.9.4.4.2. Evidence statement
There is no high-level evidence on timing of perineal repair following childbirth.
14.9.4.4.3. Recommendations on timing of repair
- 305.
Undertake repair of the perineum as soon as possible to minimise the risk of infection and blood loss. [2007]
14.9.4.5. Analgesia used during perineal repair
14.9.4.5.1. Description of included studies
There is no evidence regarding the use of analgesia during perineal repair.
14.9.4.5.2. Evidence statement
There is no high-level evidence on use of analgesia during perineal repair.
14.9.4.5.3. Recommendations on analgesia for perineal repair
- 306.
When carrying out perineal repair:
- ensure that tested effective analgesia is in place, using infiltration with up to 20 ml of 1% lidocaine or equivalent
- top up the epidural or insert a spinal anaesthetic if necessary. [2007]
- 307.
If the woman reports inadequate pain relief at any point, address this immediately. [2007]
14.9.4.6. Method of perineal repair
14.9.4.6.1. Description of included studies
A systematic review of four RCTs plus an additional RCT investigated the effects of continuous subcuticular with interrupted transcutaneous sutures for perineal repair. Two further RCTs compared a two-layer repair technique (leaving the skin unsutured) with a three-layer repair technique.
14.9.4.6.2. Review findings
One systematic review (1998) was identified which compared the effects of continuous subcuticular with interrupted transcutaneous sutures for perineal repair.431 [EL = 1+] Four RCTs were included in the review involving a total of 1864 women. The continuous subcuticular method was found to be associated with less short-term pain (up to day 10 postpartum) compared with interrupted sutures (three trials): 160/789 versus 218/799, OR 0.68 [95% CI 0.53 to 0.86]. No other differences were apparent between the two trials groups for the outcomes tested: analgesia up to day 10 (two trials): 56/527 versus 65/541, OR 0.86 [95% CI 0.58 to 1.26]; reported pain at 3 months (one trial): 58/465 versus 51/451, OR 1.12 [95% CI 0.75 to 1.67]; removal of suture material (up to 3 months) (one trial): 121/465 versus 16/451, OR 0.61 [95% CI 0.46 to 0.80]; failure to resume pain-free intercourse (up to 3 months) (one trial): 157/465 versus 144/451, OR 1.09 [95% CI 0.82 to 1.43]; resuturing (up to 3 months) (two trials, one with no incidents): 3/487 versus 3/531, OR 1.11 [95% CI 0.22 to 5.53]; dyspareunia (up to 3 months) (three trials): 172/775 versus 184/749, OR 0.88 [95% 0.69 to 1.12]. The authors concluded that the continuous subcuticular technique of perineal repair may be associated with less pain in the immediate postpartum period than the interrupted suture technique. The long-term effects are less clear. It is also noted that, while three studies used the same suture material (Dexon) throughout the repair, one trial compared repair using chromic catgut with repair using Dexon. Also, there was considerable heterogeneity between studies regarding skill and training of persons carrying out the repair. The single trial that demonstrated a statistically significant reduction in short-term pain for women in the continuous subcuticular repair group was the trial that also ensured staff were trained and practised in this technique prior to the trial.
A recent UK RCT compared continuous versus interrupted perineal repair with standard or rapidly absorbed sutures.432 [EL = 1+] The study was a 2 × 2 factorial design to allow both comparisons to be made. Findings from the trial relating to method of repair will be reported here (see the next subsection for findings from the materials arm of the trial). A continuous suturing technique for perineal repair (vaginal wall, perineal muscle and skin repaired with one continuous suture) (n = 771) was compared with interrupted sutures (continuous suture to vaginal wall, interrupted sutures to perineal muscle and skin) (n = 771). The trial included women with first- or second-degree tears or an episiotomy following a spontaneous birth. Continuous subcuticular sutures were associated with significantly less short-term perineal pain compared with interrupted sutures: pain at 2 days: 530/770 versus 609/770, OR 0.59 [95% CI 0.44 to 0.79]; pain at 10 days: 204/770 versus 338/769, OR 0.47 [95% CI 0.35 to 0.61]. This reduction in pain at 10 days was noted while sitting, walking, passing urine and opening bowels. No difference was noted between groups regarding long-term pain measures, for example: pain at 3 months: 70/751 versus 96/741, OR 0.70 [95% CI 0.46 to 1.07]; pain at 12 months: 31/700 versus 47/689, OR 0.64 [95% CI 0.35 to 1.16]; dyspareunia at 3 months: 98/581 versus 102/593, OR 0.98 [95% CI 0.72 to 1.33]; dyspareunia at 12 months: 94/658 versus 91/667, OR 1.05 [95% CI 0.77 to 1.43]. Fewer women with continuous sutures reported that the sutures were uncomfortable 2 days post-repair: 273/770 versus 318/770, OR 0.78 [95% CI 0.64 to 0.96]. This difference was slightly more marked at 10 days (OR 0.58 [95% CI 0.46 to 0.74]). Significantly more women in the interrupted group reported tight sutures both at 2 and 10 days, although the numbers were quite small. The need for suture removal was significantly higher in the interrupted group: suture removal between 10 days and 3 months: 22/751 versus 63/741, OR 0.36 [95% CI 0.23 to 0.55]. Wound gaping was more frequent following repair using the continuous technique, although again the numbers were quite small (wound gaping at 10 days: 23/770 versus 50/769, OR 0.46 [95% CI 0.29 to 0.74]). Significantly more women were satisfied with their perineal repair following repair using a continuous suture technique both at 3 months: 628/751 versus 560/741, OR 1.64 [95% CI 1.28 to 2.11] and 12 months: 603/700 versus 542/689, OR 1.68 [95%1.27 to 2.21]. Women in the continuous repair group were also more likely to report that they felt ‘back to normal’ at 3 months postpartum: 414/700 versus 332/689, OR 1.55 [95% CI 1.26 to 1.92]. It is noted that senior midwives (Grade G) were significantly more likely to use the continuous suturing technique compared with Grade E and F midwives. Subsequent analyses were undertaken taking this into consideration.
A UK RCT published in 1998 compared a two-stage perineal repair (n = 890) with the more usual three-stage repair (n = 890).433 [EL = 1+] This trial also employed a 2 × 2 factorial design comparing both the method of repair and suture material used (findings regarding the latter are reported in the following subsection). At 2 days no differences were noted between the trial groups for any of the pain measures investigated: any pain in last 24 hours: 545/885 (62%) versus 569/889 (64%); analgesia in last 24 hours: 400/885 (45%) versus 392/889 (44%); tight stitches: 162/885 (18%) versus 196/889 (22%). Significantly more women in the two-stage repair group had a gaping perineal wound: 203/885 (23%) versus 40/889 (4%), P < 0.00001. At 10 days, while there were no significant differences in reported pain and analgesia use (reported pain in last 24 hours: 221/886 (25%) versus 244/885 (28%); analgesia in last 24 hours: 73/886 (8%) versus 69/885 (8%)), significantly more women in the three-stage repair group reported tight stitches: 126/886 (14%) versus 163/885 (18%), RR 0.77 [95% CI 0.62 to 0.96], P = 0.02. Incidence of perineal gaping was still higher in the two-stage repair group at 10 days: 227/886 (26%) versus 145/885 (16%), P < 0.00001. Women in the two-stage repair group were also significantly less likely to have had suture material removed: 26/886 (3%) versus 67/885 (8%), P < 0.0001. Incidences of repair breakdown were very low and similar for the two groups (n = 5 versus n = 7). At 3 months postpartum there were no differences in most pain measures, for example: any pain in last week: 64/828 (8%) versus 87/836 (10%); resumption of sexual intercourse: 704/828 (85%) versus 712/836 (85%); resumption of pain-free intercourse: 576/828 (70%) versus 551/836 (66%). There was, however, a difference in reported dyspareunia: 128/890 (14.3%) versus 162/890 (18.2%), RR 0.80 [95% CI 0.65 to 0.99], P = 0.04. The difference for removal of suture material was still apparent at 3 months in favour of the continuous method group: 59/828 (7%) versus 98/836 (11%), RR 0.61 [95% CI 0.45 to 0.83]. There was little resuturing required and no difference between groups (n = 4 versus n = 9).
A 1 year postal questionnaire follow-up study was carried out for the above trial, involving 793 women.434 [EL = 1+] The follow-up sample was deliberately biased to include 31% women who had had an instrumental birth (compared with 17% in the original sample). There was no difference between groups regarding persistent pain at 1 year: 28/396 versus 26/396. Women who had undergone the three-stage perineal repair were significantly more likely to report that the perineal area ‘felt different’ than women who had undergone two-stage repair: 17/395 versus 157/396, RR 0.75 [95% CI 0.61 to 0.91]. Subgroup analyses showed this difference to be more marked following spontaneous births compared with instrumental births: instrumental: 45/123 versus 55/124, RR 0.82 [95% CI 0.61 to 1.12]; spontaneous: 72/272 versus 102/272, RR 0.71 [95% CI 0.55 to 0.91]; and more marked following repair using interrupted sutures compared with mixed technique or subcuticular technique: interrupted technique: 57/209 versus 87/202, RR 0.63 [95% CI 0.48 to 0.83]; mixed technique: 46/133 versus 55/136, RR 0.86 [95% CI 0.63 to 1.17]; subcuticular: 14/53 versus 15/58, RR 1.02 [95% CI 0.55 to 1.91]. There were no significant differences between groups for dyspareunia, failure to resume pain-free intercourse or need for resuturing.
A second RCT conducted in Nigeria (2003) also compared two-stage repair with three-stage repair.435 [EL = 1+] The trial was conducted across four sites and recruited 1077 women, 823 of whom were followed up to 3 months postnatally (response rate = 76.4%). As with the UK trial, midwives and labour ward obstetricians were trained in the two-stage repair technique prior to the study. Where skin repair was undertaken, a continuous technique was taught and encouraged. Most repairs were undertaken using chromic catgut. Postnatal assessments of wound healing were carried out by a researcher blinded to the trial allocation of the woman. Compared with three-stage repair, two-stage repair was associated with less pain and fewer reports of tight sutures at 48 hours postnatally (perineal pain: 57% versus 65%, RR 0.87 [95% CI 0.78 to 0.97); tight sutures: 25% versus 38%, RR 0.67 [95% CI 0.54 to 0.82)). Analgesia use and degree of inflammation and bruising were also significantly less in the two-stage group (analgesia use: 34% versus 49%, RR 0.71 [95% CI 0.60 to 0.83]; inflammation/bruising: 7% versus 14%, RR 0.50 [95% CI 0.33 to 0.77]). Wound gaping (skin edges > 0.5 cm apart) was more prevalent in the two-stage repair group: 26% versus 5%, RR 4.96 [95% CI 3.17 to 7.76]. The differences regarding perineal pain and analgesia were still apparent at 14 days and 6 weeks postpartum in favour of the two-stage repair group. The difference in wound gaping was much smaller by 14 days: 21% versus 17%, RR 1.25 [95% CI 0.94 to 1.67]. There was no difference in wound breakdown: 3% versus 2%, RR 1.27 [95% CI 0.56 to 2.85]. At 3 months postpartum, women in the two-stage repair group reported a lower incidence of dyspareunia compared with women in the three-stage repair group: 10% versus 17%, RR 0.61 [95% CI 0.43 to 0.87]. The authors pointed out that the differences in short-term pain found in this study may be due to the fact they used catgut for most of the perineal repairs rather than a synthetic absorbable suture material.
14.9.4.6.3. Evidence statement
There is high-level evidence that a continuous non-locked suturing technique for repair of perineal muscle is associated with less short-term pain. More women who were repaired with a continuous non-locked technique were also satisfied with their perineal repair and felt back to normal at 3 months.
A two-stage repair (where the skin is opposed but not sutured) is associated with no differences in the incidence of repair breakdown but is associated with less dyspareunia at 3 months. There is some evidence that it is also associated with less short-term perineal pain when compared with skin repair undertaken using chromic catgut sutures.
Continuous subcuticular skin repair is associated with less short-term pain when compared with interrupted skin repair.
14.9.4.6.4. Recommendations on methods of perineal repair
- 308.
If the skin is opposed after suturing of the muscle in second-degree trauma, there is no need to suture it. [2007]
- 309.
If the skin does require suturing, use a continuous subcuticular technique. [2007]
- 310.
Undertake perineal repair using a continuous non-locked suturing technique for the vaginal wall and muscle layer. [2007]
14.9.4.7. Materials for perineal repair
14.9.4.7.1. Description of included studies
One systematic review and two additional RCTs have compared the effects of absorbable synthetic suture material with catgut or chromic catgut. An additional UK RCT compared rapidly absorbed synthetic suture material with standard synthetic suture material.
14.9.4.7.2. Review findings
One systematic review (1999) has been conducted to assess the effects of absorbable synthetic suture material compared with catgut on short- and long-term pain experienced by women following perineal repair.436 [EL = 1+] The review included eight trials involving 3642 women. Seven trials used polyglycolic acid (Dexon) and one trial used polyglactin (Vicryl). Women allocated to groups using absorbable synthetic suture material reported significantly less short-term pain compared with those sutured using catgut: day 3 or before: OR 0.62 [95% CI 0.54 to 0.71], eight trials; days 4–10: OR 0.71 [95% CI 0.58 to 0.87], three trials; analgesia use up to day 10: OR 0.63 [95% CI 0.52 to 0.77], five trials. Women allocated to perineal repair using absorbable synthetic suture material also reported less suture dehiscence up to day 10: OR 0.45 [95% CI 0.29 to 0.70], five trials; and need for resuturing of the perineal wound up to 3 months: OR 0.26 [95% CI 0.10 to 0.66], four trials. However, the need for removal of suture material up to 3 months was greater in the absorbable synthetic group: OR 2.01 [95% CI 1.56 to 2.58], two trials. There was no difference reported for long-term pain: OR 0.81 [95% CI 0.61 to 1.08], two trials. The authors of the review noted that the skill level of clinicians may be very different between trials, e.g. suture dehiscence in one trial was 37/71 for the control group and 12/77 for the experimental group, while in another trial there were no incidents of suture dehiscence.
One additional RCT has been conducted in Australia comparing absorbable synthetic suture material (polyglactin) (n = 194) with chromic catgut (n = 197).437 [EL = 1+] Women with a third-degree tear or an instrumental birth were excluded from the trial. Owing to chance imbalance in the proportion of nulliparous women between the two trial groups, parity-adjusted odds ratios were calculated. There was a tendency towards reduced short-term pain in women allocated to the polyglactin group, but differences did not reach statistical significance: perineal pain at 1 day: adjusted OR 0.64 [95% CI 0.39 to 1.06]; perineal pain at 3 days: adjusted OR 0.70 [95% CI 0.46 to 1.08]. No significant differences were seen between groups for any of the longer-term pain outcomes (any perineal pain, resumed intercourse, dyspareunia) at 6 weeks, 3 months or 6 months. At 6 weeks postpartum, eight women repaired with polyglactin reported problems with their sutures compared with three women in the catgut group (one woman in each group reported infection at wound site, the remainder reported tight sutures that required removal) (adjusted OR 2.61 [95% CI 0.59 to 12.41]).
A recent US RCT compared the healing characteristics of chromic catgut with fast-absorbing polyglactin 910.438 [EL = 1+] Although women were recruited and randomised into trial groups during labour, analysis was only performed for those women requiring perineal repair (polyglactin 910: 459/684; chromic catgut: 49/677). This study is unusual in that pain outcomes were measured both for the perineal area (referred to as ‘vaginal’ pain) and uterine cramping. No differences were found between groups for vaginal pain at 24–48 hours, 10–14 days or 6–8 weeks. There were, however, some differences in uterine pain, with significantly more women in the chromic catgut group reporting moderate/severe uterine pain at 24–48 hours: no pain: n = 81 (18%) versus n = 63 (14%), NS; a little/some pain: n = 264 (58%) versus n = 232 (52%), NS; moderate/severe pain: n = 114 (25%) versus n = 154 (34%), P = 0.006. This significant difference was also evident at 6–8 weeks. No differences in uterine pain were noted at 10–14 days. The authors have no explanation for the observed differences in uterine cramping between groups based on suture material used. Given that this difference was only seen at one of the two study sites they conclude that it may simply be an anomaly of the data. At 6–8 weeks no difference was found between groups for persistent suture material (n = 2 women in each group) or perineal wound breakdown (n = 4 versus n = 3).
A UK RCT compared rapidly absorbed synthetic suture material (n = 772) with a standard form of the synthetic suture material (n = 770) within a 2 × 2 factorial study design also comparing suture method.432 [EL = 1+] The study involved women who had sustained either a second-degree tear or an episiotomy. There was no significant difference between the two groups for the primary outcome of pain at 10 days postnatally, although findings favoured the rapidly absorbed suture material: OR 0.84 [95% CI 0.68 to 1.04]. There was, however, a significant reduction in analgesia used in the previous 24 hours reported at 10 days for women in the rapidly absorbed suture material group: OR 0.55 [95% CI 0.36 to 0.83]; and a significant reduction in pain on walking for this group: OR 0.74 [95% CI 0.56 to 0.97]. The need for removal of sutures in the 3 months following birth was also less for women sutured with the rapidly absorbed suture material: OR 0.26 [95% CI 0.18 to 0.37].
14.9.4.7.3. Evidence statement
There is high-level evidence that a rapidly absorbable synthetic suture material is associated with less short-term pain, less suture dehiscence and less need for resuturing of the perineum up to 3 months postpartum.
14.9.4.7.4. Recommendations on material for perineal repair
- 311.
Use an absorbable synthetic suture material to suture the perineum. [2007]
- 312.
Observe the following basic principles when performing perineal repairs:
- Repair perineal trauma using aseptic techniques.
- Check equipment and count swabs and needles before and after the procedure.
- Good lighting is essential to see and identify the structures involved.
- Ensure that difficult trauma is repaired by an experienced practitioner in theatre under regional or general anaesthesia.
- Insert an indwelling catheter for 24 hours to prevent urinary retention.
- Ensure that good anatomical alignment of the wound is achieved and that consideration is given to the cosmetic results.
- Carry out rectal examination after completing the repair to ensure that suture material has not been accidentally inserted through the rectal mucosa.
- After completion of the repair, document an accurate detailed account covering the extent of the trauma, the method of repair and the materials used.
- Give the woman information about the extent of the trauma, pain relief, diet, hygiene and the importance of pelvic-floor exercises. [2007]
14.9.4.7.5. Research recommendation on analgesia during perineal repair
- 30.
Research is needed into the optimum analgesia required during perineal repair.
14.9.4.8. Analgesia for perineal pain following perineal repair
14.9.4.8.1. Description of included studies
A systematic review of three RCTs and one additional RCT were identified which assessed the effectiveness of analgesic rectal suppositories for pain from perineal trauma following childbirth.
14.9.4.8.2. Review findings
A systematic review including 249 women assessed the effectiveness of analgesic rectal suppositories for pain from perineal trauma following childbirth.439 [EL = 1+] All trials used nonsteroidal anti-inflammatory analgesia suppositories, one trial (Saudi Arabia) compared indometacin with a placebo, while the other two trials (UK) compared diclofenac (Voltarol) with a placebo. All trials administered a suppository immediately after perineal repair was complete. In one UK trial, a single dose of 100 mg was given, in the second (Saudi) trial, 2 × 100 mg suppositories were inserted together immediately following perineal repair, and in the third trial (UK), one suppository was given immediately after suturing and another 12 hours later. Findings suggest that nonsteroidal anti-inflammatory drugs (NSAIDs) administered as rectal suppositories provide effective pain relief following perineal repair (two trials). For indometacin the incidence of perineal pain in the first 24 hours was 6/30 versus 30/30, RR 0.20 [95% CI 0.10 to 0.41]; for diclofenac: RR 0.65 [95% CI 0.50 to 0.85]. The meta-analysis of these two findings produces a wide confidence interval that crosses 1 (RR 0.37 [95% CI 0.10 to 1.38] (two trials)). Findings from the other trial are reported as median scores obtained using a VAS. Women in the diclofenac group reported significantly less pain at 24 hours than women in the placebo group (diclofenac: median 1 [range 0 to 2.5], placebo: median 1 [range 0 to 3], P < 0.05 using Mann–Whitney U test). Only one trial included the outcome of any pain experienced 24–72 hours after perineal repair, with the effect of treatment just failing to reach statistical significance: RR 0.73 [95% CI 0.53 to 1.02]. Findings from a stratified analysis for level of pain experienced within the first 24 hours following perineal repair suggest that NSAIDs have their best level of effect for moderate pain compared with mild or severe pain: mild: RR 1.12 [95% CI 0.70 to 1.80] (two trials); moderate: RR 0.13 [95% CI 0.02 to 0.76] (two trials); severe: RR 0.21 [95% CI 0.01 to 4.12] (two trials). Use of additional analgesia was also measured as an outcome in two trials, although not in a way that allows pooling of data. Both trials showed a significant reduction in use of additional analgesia up to 48 hours postpartum. None of the trials reported longer term outcomes such as breastfeeding, effects on mother–infant interactions, postnatal depression or return to pain-free intercourse. All three trials reported that there were no side effects associated with the treatment, although none investigated this as an identified outcome.
An RCT conducted in Australia (2004) also evaluated the effectiveness of rectal diclofenac compared with a placebo.440 [EL = 1+] Women in the treatment group (n = 67) received a diclofenac suppository immediately after perineal repair (of a second-degree tear, third-degree tear or episiotomy). Women randomised to the control group received a placebo (Anusol) suppository. Both groups received a second suppository 12–24 hours later. Pain was measured in three ways – using the Short-form McGill Pain Questionnaire (SF-MPQ), using a 10 cm VAS and using the Present Pain Inventory (PPI). At 24 hours postnatally, women's pain scores were significantly lower for the treatment group compared with the control group, although this was not evident across all measurement scales: at rest: SF-MPQ total score: median 6 [IQR 3 to 11] versus 7 [IQR 3 to 12], NS; VAS: mean 2.8 [SD 0.3] versus 3.9 [SD 0.3]: RR −1.1 [95% CI −1.9 to −0.3], P = 0.01; PPI: mean 31 [SD 53.4] versus 32 [57.1]; RR 0.9 [95% CI 0.7 to 1.3], P = 0.69. For pain scores with movement at 24 hours both the VAS and the PPI score were significantly lower in the treatment group, although this difference was not evident for total SF-MPQ scores. By 48 hours there were no differences in reported pain between the two groups for any of the pain outcome measures. There was also no difference between groups regarding the use of additional analgesia prior to discharge: 81% versus 86%, RR 0.9 [95% CI 0.8 to 1.1] or time from birth to first analgesia (hours): median 6.4 [IQR 3.5 to 10.5] versus 5.8 [IQR 2.9 to 10.2]. Pain outcomes during activities at 10 days and 6 weeks postnatally were also similar for the two groups.
14.9.4.8.3. Evidence statement
There is high-level evidence that rectal NSAIDs reduce immediate perineal pain following repair.
14.9.4.8.4. Recommendation on analgesia for perineal pain following perineal repair
- 313.
Offer rectal non-steroidal anti-inflammatory drugs routinely after perineal repair of first- and second-degree trauma provided these drugs are not contraindicated. [2007]
Footnotes
- u
Recommendations relating to immediate postnatal care (within 2 hours of birth) have been adapted from Routine postnatal care of women and their babies (NICE clinical guideline 37). Please see NICE clinical guideline 37 for further guidance on care after birth.
- v
Recommendations relating to immediate postnatal care (within 2 hours of birth) have been adapted from Routine postnatal care of women and their babies (NICE clinical guideline 37). Please see NICE clinical guideline 37 for further guidance on care after birth.
- w
Recommendations relating to immediate postnatal care (within 2 hours of birth) have been adapted from Routine postnatal care of women and their babies (NICE clinical guideline 37). Please see NICE clinical guideline 37 for further guidance on care after birth.
- Introduction
- Initial assessment of the newborn baby
- Neonatal resuscitation
- Routine paired cord-blood gas analysis
- Care of babies born with meconium-stained liquor
- Care of babies born to women with prelabour rupture of the membranes at term
- Mother–infant bonding and promoting breastfeeding
- Initial assessment of the mother following birth
- Perineal care
- Intrapartum care for healthy women and babies
- Addendum to intrapartum care: care for healthy women and babies
- Addendum to Clinical Guideline CG190, Intrapartum care for healthy women and babies
- 2019 surveillance of intrapartum care for healthy women and babies (NICE guideline CG190)
- Intrapartum Care: Care of Healthy Women and Their Babies During Childbirth (NICE guideline CG55)
- Care of the baby and woman immediately after birth - Intrapartum CareCare of the baby and woman immediately after birth - Intrapartum Care
Your browsing activity is empty.
Activity recording is turned off.
See more...