Rationale and impact
- Benefits and risks of caesarean and vaginal birth
- Placenta accreta spectrum
- Maternal choice for caesarean birth
- Prevention and management of hypothermia and shivering
- Methods to reduce infectious morbidity and wound care after caesarean birth
- Surgical opening technique
- Closure of the uterus
- Monitoring after caesarean birth
- Pain management after caesarean birth
These sections briefly explain why the committee made the recommendations and how they might affect practice. They link to details of the evidence and a full description of the committee's discussion.
There was some evidence for a selected number of outcomes on the short- and long-term effects of caesarean birth compared with vaginal birth, although there were some limitations with the quality of the evidence, and not all evidence was from a comparison of planned mode of birth. The committee used this evidence, along with their clinical expertise, to update the advice comparing the relative benefits and risks of these 2 modes of birth.
For some outcomes there was conflicting or limited evidence, and there were also a number of outcomes for which no evidence was identified for inclusion, so the committee highlighted these uncertainties.
As the evidence was limited for this review the committee made a research recommendation.
There were also 3 outcomes included in the 2011 guideline which had not been included in this current review (vaginal tears, length of stay and pain) but the committee agreed that the advice was still appropriate and should be carried forward into the updated guideline.
The committee considered that their recommendations would reinforce best practice. It is already current practice to discuss the risks and benefits of alternative modes of birth during the antenatal period and this review has simply led to an update of the information that should be discussed with women.
The routine 20-week ultrasound scan offered to women and pregnant people is used to determine placental location, so the committee agreed that identification of a placenta praevia or low-lying placenta at this scan in those with a previous caesarean scar, or a scar from previous uterine surgery, should be the trigger for further investigations to check for the presence of placenta accreta. There was evidence from a number of studies that greyscale ultrasound with colour doppler had moderate sensitivity and high specificity for detecting placenta accreta. However, the committee were aware that the sensitivity and specificity of ultrasound scans is very operator-dependent, and that if a case of placenta accreta is missed, there is a risk of severe morbidity or mortality at birth. They therefore recommended that the scan should be done by a senior clinician with the necessary expertise and experience to detect placenta accreta.
The committee discussed the timing of this scan and noted that in most of the included studies, it was carried out between 28 and 37 weeks. The committee agreed that an earlier scan would allow more time for planning the birth, particularly if an early birth was thought to be needed, but that scanning too early would not be an advantage as the placenta may move. They therefore recommended that the scan be done around 28 weeks, but not later than 29 weeks.
The committee discussed that, since 2020, NHS England has commissioned a specialised maternity service for women and pregnant people diagnosed with placenta accreta spectrum (which they refer to as abnormally invasive placenta, or AIP). This service enables diagnosis and care, including the birth, to be given in a centre with an appropriate multidisciplinary team, access to adult intensive care, level 3 neonatal care and access to blood products. The committee therefore recommended that, if placenta accreta is suspected at the scan carried out around 28 weeks, women and pregnant people should be referred to a specialist placenta accreta centre.
The committee discussed evidence showing that the sensitivity and specificity of MRI (without contrast or with contrast unspecified) for assessing placenta accreta spectrum was no better than that for ultrasound with colour doppler, and therefore agreed that it was not necessary to offer an MRI scan to confirm the presumed diagnosis of placenta accreta. However, the committee also agreed that MRI was a useful imaging technique when planning surgical management of placenta accreta to identify if, for example, placenta percreta had impacted on other organs, and so updated the recommendation on this.
Although the committee were aware that women and pregnant people with placenta accreta are now referred to specialist placenta accreta spectrum centres, they discussed that some people may present very late in pregnancy or in labour and need to be cared for at their local maternity unit, and so the recommendations on care in this setting were retained. The committee agreed that the discussion on birth options should be carried out by a senior obstetrician, and they also updated the advice on who may need to be consulted or present for the caesarean birth, including their roles. The committee also advised that a critical care bed should be available for the woman or pregnant person and for the baby in case they are needed, but that emergency surgery should not be delayed if a bed was not immediately available.
The committee amended the existing recommendation on use of blood products to clarify that this should only be if the use of such products is acceptable to the woman or pregnant person, and included a link to the NICE guideline on blood transfusion as this provides some useful advice on alternatives to blood transfusions and also cell salvage and tranexamic acid.
The committee agreed that as placenta accreta was no longer managed by local hospitals, a local protocol was no longer appropriate, but that specialist placenta accreta spectrum centres should develop protocols with their local maternity units about how placenta accreta should be managed within the network, such as methods for referral and ongoing management. The committee also agreed that the protocol should include details of how local maternity units should provide care if placenta accreta was identified very late, which may include, for example, emergency transfer to a specialist regional centre or support from the specialist centre being provided to the local maternity unit.
The 20-week ultrasound scan is routinely offered to all women and pregnant people already, so this will not be a change in practice. The use of a 28‑week ultrasound scan with colour doppler to assess for the presence of placenta accreta is also not a change in practice, and although the timing is now earlier than previously recommended, this will not lead to additional scans. Advising that the scan is done by a senior clinician with expertise in placenta accreta is a change to the previous advice, but in practice, the scan is already done by a consultant in most cases so it is not thought this will result in a major change in practice. Using a more experienced operator for when looking for placenta accreta is likely to lead to the best sensitivity and specificity from the scan. This in turn is likely to correctly identify more cases, reducing the risk of serious morbidity or mortality for the mother and baby, and optimising resource use by preventing complications at birth that may otherwise need expensive critical care. This should also reduce the number of false positives referred to specialist centres where the presumed diagnosis of placenta accreta is later found to be incorrect.
The recommendation to consider MRI only when planning surgery may reduce the use of MRI for the diagnosis of placenta accreta, in turn reducing costs. The discussion about birth options being carried out by a senior obstetrician is not a change in practice and so will not have a resource impact.
Specialist regional placenta accreta spectrum centres are already established, so recommending appropriate referral to them will not have a resource impact. Ensuring that all maternity centres (both local units and specialist centres for placenta accreta spectrum) understand their responsibilities will enable safe care to be provided.
The committee discussed the fact that some women or pregnant people may request a caesarean birth because they have concerns about aspects of the birth and believe that a caesarean birth would be the best way to alleviate these concerns. However, there may be cases where the concerns can be addressed in other ways, such as choosing an alternative place of birth, opting for a birth which will provide greater continuity of midwifery care, or by planning adequate pain relief. The committee therefore expanded the advice to include these aspects in the discussion.
The committee were aware that this discussion would be best held in a clinic where there was time to explore the different options and preferences, and so suggested this should be in a birth options clinic. The committee were aware that consultant midwives were often involved in such discussions and so included them in the list of healthcare professionals who should be involved.
The committee revised the wording of the recommendation on perinatal mental health support to clarify that the access to the planned place of birth was needed during the antenatal period, and not that all antenatal support had to be provided at the planned place of birth.
The committee revised the wording of the recommendation on offering caesarean birth to be more person-centred, and to ensure the woman or pregnant person's choice to have a caesarean birth was supported.
The committee agreed that women and pregnant people should not have to move to a different obstetric unit for a caesarean birth, and so recommended that caesarean birth should be offered within their obstetric unit. The committee discussed the potential rare situations where there was a clinical reason behind a reluctance to perform a maternal request caesarean birth, but agreed that in this situation a full multidisciplinary team discussion would be needed during the pregnancy to agree a plan for the woman or pregnant person.
This change to add more factors into the discussions around requests for caesarean birth may mean more women and pregnant people can be supported to have their preferred mode of birth. The change may also increase the number of women or pregnant people being seen by a consultant midwife or senior midwife for a longer 'birth options' appointment, but is unlikely to have a resource impact as these conversations would previously have been held across multiple midwife appointments.
This change to recommend that caesarean birth should be offered within the woman or pregnant person's obstetric unit will reduce the number of people who have to move to a different obstetric unit in order to have a caesarean birth. This will benefit the following groups in particular:
women or pregnant people with disabilities who find it difficult to travel or feel anxious about change
women or pregnant people from lower socioeconomic groups where the increased travel costs are a concern
younger women or pregnant people who may not feel confident enough to transfer to another unit
women or pregnant people from certain racial groups who may experience bias surrounding decisions relating to mode of birth and who may have less favourable maternity outcomes
women or pregnant people who do not speak English as a first language, or those from groups such as migrants or refugees who may not be familiar with navigating the healthcare system and who therefore may have more difficulty changing their provider or travelling to another unit.
There was evidence for the effectiveness of active warming measures (for example, forced air warming, under body pads, warmed IV fluids) to prevent shivering and hypothermia in women having a caesarean birth, and there was some evidence for improved thermal comfort and maternal temperature. The committee recommended the use of warmed IV fluids and irrigation fluids for all women having caesarean birth, but because of the low incidence of hypothermia and shivering during caesarean birth, the physiological differences between women having caesarean birth and the general surgical population, the lack of beneficial effect on wound infections, and the fact that warming methods are likely to be as effective at managing hypothermia and shivering as they are at preventing it, the committee recommended that other warming measures should only be used in women who were shivering, said they felt cold or were hypothermic, and not in all women for prevention. The committee recommended forced air warming as the method of choice as this was already widely available, easier to use and could be easily moved with the woman.
There was evidence that pethidine was also effective at reducing shivering, but the committee did not recommend this because of the possible adverse effects on breastfeeding.
The recommendation to use forced air warming will standardise practice across the NHS. There could be resource implications for units to purchase the disposable 'blankets' used, but this could be offset by earlier discharge of women from recovery to the postnatal ward.
The use of warmed intravenous fluids, blood and irrigation fluids is already standard practice, so this recommendation will not change this.
There was evidence that alcohol-based chlorhexidine solution skin preparations reduce the risk of surgical site infections, compared with alcohol-based iodine solutions.
There was also evidence that aqueous iodine vaginal preparations reduce the risk of endometritis in women with ruptured membranes. Although there was some evidence on chlorhexidine vaginal preparations, overall the evidence indicated that that iodine vaginal preparations might be more effective.
There was some evidence that negative pressure wound therapy (NPWT) reduces the risk of wound or surgical site infections for women with a body mass index (BMI) of 30 kg/m2 or more but economic evidence indicated that this would not be cost effective in those with a BMI of less than 35 kg/m2 and only borderline cost effective in the group with a BMI of 35 kg/m2 or more.
The evidence showed no difference in wound infection or readmissions into hospital when the dressing was removed either 6 hours or 24 hours after surgery.
There was very limited evidence on the use of 2 different types of dressing, but the committee agreed it was not enough to recommend a specific type.
There was no evidence on the use of incise drapes, diathermy or body hair removal, so the committee did not make recommendations about these, but noted that the NICE guideline on surgical site infections (which covers general surgery rather than caesarean birth) has recommendations on some of these interventions.
The recommendations on skin preparation are broadly in line with current best clinical practice. The committee agreed that the recommendation to use aqueous iodine vaginal preparation will be a change in clinical practice, because the use of vaginal preparation is not routine across England.
The committee identified that considering the use of NPWT for women with a BMI of 35 kg/m2 will be a change of practice for many units (some units do not use it at all, or only at higher BMI thresholds), and could have resource implications, particularly in areas where a higher proportion of pregnant women will meet the criteria.
The committee agreed that vertical midline incisions were no longer routinely used for caesarean birth, and so the recommendation advising that a transverse incision should be used instead was withdrawn. The committee noted that in some rare clinical situations (for example, previous abdominal surgery) a vertical midline incision may still be necessary, and so added this to their recommendation on the type of incision to be used.
There was evidence that a low transverse straight skin incision with blunt dissection of subsequent tissue layers (such as described in the Joel–Cohen or modified Joel–Cohen techniques) led to a reduction in post-operative febrile morbidity and use of post-operative analgesia, reduced decreases in haemoglobin and reduced total operating time compared to the use of a very low transverse curved skin incision and sharp dissection (as described in the Pfannenstiel technique). The committee agreed that in some women and pregnant people, such as those who have had previous caesarean births, the presence of scar tissue may mean that blunt dissection was not always possible and so in some cases, (which would be decided by the surgeon at the time of operation), sharp dissection would be necessary for extension.
There was evidence that in women and pregnant people with class 3 obesity (BMI of 40 kg/m2 or more) there was an increased risk of wound complications with the Pfannenstiel technique (which uses a very low incision) compared to a higher transverse abdominal incision. The committee discussed the fact that in very obese women and pregnant people the presence of a panniculus led to the wound being covered which increased the risk of infection. Based on this evidence, and their knowledge of the complications caused by a panniculus, the committee therefore recommended that in these situations a higher incision may be needed.
The committee agreed not to use the names of the surgical techniques in the recommendations as the number of techniques, including the modified techniques, and the slight variations between them may lead to confusion. They therefore agreed that it was preferable to describe the details of the incision and subsequent opening.
The recommendations on method of skin incision and opening of subsequent layers will standardise care across the NHS while still allowing room for surgeon modification on a case-by-case basis, for example, in women and pregnant people who are obese or have had previous caesarean births. Use of a technique which reduces operating time, blood loss, pain and infections may lead to savings in resource use to treat these complications.
There was evidence showing that there was no difference in any outcomes when comparing single and double layer closure of the uterus. There was some evidence of the reduced need for blood transfusions with single layer compared with double layer closure, as part of a comparison of different caesarean birth techniques, but this could have been confounded by other differences in the techniques.
Current practice is to use a double layer uterine closure technique, except in occasional circumstances when there is a specific reason for using single layer closure. This recommendation will allow surgeons to choose single or double layer closure, depending on the individual clinical circumstances at the time of the surgery.
There was no evidence found on the best monitoring schedule for women, but the committee used their knowledge and expertise of current best practice to develop recommendations on the monitoring schedule.
The factors which may increase the risk of respiratory depression listed in the Society for Obstetric Anesthesia and Perinatology consensus statement include class 3 obesity (BMI 40 kg/m2 or more), cardiovascular disease, magnesium administration, and obstructive sleep apnoea. The committee discussed whether listing these factors would be useful but agreed, based on feedback from stakeholders, that leaving the risk factors unspecified would encourage healthcare professionals to consider all potential risk factors for respiratory depression when deciding on appropriate levels of monitoring.
The committee agreed that continuous pulse oximetry monitoring may interfere with the ability of a woman or person who has given birth to mobilise and care for their baby. They agreed that it was therefore more appropriate to use hourly monitoring of oxygen saturations in combination with hourly monitoring of sedation and respiratory rate to detect respiratory depression.
The committee agreed that heart rate, blood pressure, temperature and pain monitoring would be carried out as in usual postnatal protocols, but were not necessary every hour to detect respiratory depression.
The committee agreed to add information to the recommendation about overestimation of oxygen saturation levels using pulse oximetry in people with dark skin to advise healthcare professionals of the potential implications and what action could be taken to overcome this limitation.
The committee agreed that as a result of the different pharmacokinetic properties of neuraxial morphine compared with neuraxial diamorphine, it may lead to an increased risk of respiratory depression and over a longer period of time, and that this fact should be highlighted. Because of the long-lasting effects of neuraxial morphine and the possible increased risk of respiratory depression, the committee discussed whether to adopt the monitoring recommendations developed by the Society for Obstetric Anesthesia and Perinatology but agreed that healthcare staff should be advised that monitoring could be the same as that recommended for diamorphine, unless there were concerns about the woman or person's respiratory status, in which case more intensive respiratory monitoring would need to continue.
The recommendations should standardise the frequency and duration of monitoring of most women who have received intrathecal or epidural opioids at the time of caesarean birth, but will mean women need to be assessed for risk factors or other concerns about respiratory status to determine if they need a more intensive monitoring schedule. The recommendations may increase the number of people assessed as being at risk of respiratory depression, as the determination of risk has now been left to clinical judgement instead of including a list of suggested risk factors in the recommendation. An increase in the number of people monitored is likely to increase costs. However, these costs will be offset to some extent as the revised recommendations will reduce the use of continuous pulse oximetry and the number of observations being carried out every hour for the purpose of detecting respiratory depression. The frequency of monitoring for respiratory depression has not changed.
Additional ways of monitoring for respiratory depression may be used more for people with dark skin. No resource impact is anticipated from this change.
The recommendations will increase awareness of the differences in risk and duration of respiratory depression between neuraxial morphine and diamorphine, and women and people with risk factors for respiratory depression who have given birth may need increased monitoring after neuraxial morphine, compared with those who have received neuraxial diamorphine, if there are concerns about respiratory depression after 12 hours of hourly monitoring. As diamorphine shortages are unpredictable, it is difficult to quantify the resource impact. Any increased monitoring required for people receiving morphine may increase the staff resources needed, although this could be mitigated as the checks could be undertaken by a maternity support worker or healthcare assistant (typically Agenda for Change Band 3). Although the use of morphine is not expected to lead to longer hospital stays, any more intensive monitoring that is necessary after 12 hours may require the woman or person who has given birth to stay in a setting where this monitoring can be carried out.
The committee developed separate recommendations for women receiving regional or general anaesthesia, based on their knowledge of the likely differences in analgesia requirements. For all women, the committee agreed that any postoperative analgesia should be suitable for use while breastfeeding, but that women should be made aware of any potential adverse effects on their baby.
The committee agreed to retain the previous NICE recommendation to offer diamorphine (delivered intrathecally or by epidural) for women who have regional anaesthesia. Giving spinal or epidural diamorphine in this way reduces the need for additional opioids and other rescue medications during surgery, and it remains effective for up to 12 hours (when pain is likely to be most severe). The committee agreed that it was not necessary to specify a range of doses for diamorphine. The committee were aware of evidence that suggested a ceiling effect in neuraxial opioid doses, above which there is no more analgesic effect but there is an increased risk of side effects. In most cases, a maximum intrathecal dose of 300 micrograms and a maximum epidural dose of 3 mg are considered to be adequate. These are the doses suggested by the Obstetric Anaesthetists' Association and are the doses used in current clinical practice. In individual cases, and based on clinical judgement, anaesthetists may choose to use alternative doses of intrathecal or epidural diamorphine.
As there may be intermittent shortages of diamorphine, the committee agreed that preservative-free morphine could be used as an alternative and that advising this alternative will provide clear guidance for healthcare professionals to follow in this situation. This will reduce the likelihood of incorrect preparations or dosages of morphine being substituted for diamorphine. The committee agreed that fentanyl has a rapid onset of action and will help ensure rapid onset of pain relief until the intrathecal morphine takes full effect. The dose of intrathecal morphine plus fentanyl and of epidural morphine are based on those recommended by the Obstetric Anaesthetists' Association. In individual circumstances, and based on clinical judgement, anaesthetists may choose to use alternative doses of these drugs.
The committee were aware that neuraxial morphine was more likely to lead to side effects than neuraxial diamorphine, and so advised that these side effects may need treatment to reduce the severity of symptoms. Based on stakeholder feedback, the committee made an additional recommendation to advise on the safe storage and use of preservative-free morphine, particularly in settings where both preservative-free and preservative-containing morphine are available.
The committee agreed that women receiving regional anaesthesia should be offered oral morphine sulfate, as the evidence showed it to be effective.
The evidence on pain relief for women after general anaesthesia was sparse, but the committee agreed that intravenous patient-controlled analgesia (PCA) using morphine should be offered as these women will likely have a higher level of pain. Monitoring requirements for women and people having PCA morphine now includes pulse oximeter oxygen saturation as this is one of the respiratory monitoring modalities used in conjunction with respiratory rate and sedation assessment. If PCA morphine is not acceptable to the woman, then oral morphine should be considered as a less invasive alternative.
From their knowledge and experience, the committee agreed that paracetamol and a non-steroidal anti-inflammatory drug (NSAID) such as ibuprofen should be offered in combination to all women to limit the amount of opioids needed, and to allow opioids to be stopped. Based on the evidence on the benefits of fixed interval pain management timing, the committee recommended that these are prescribed to be taken regularly to maintain good pain control, in preference to on-request administration, which had lower rates of satisfaction reported by women.
Some women will have contraindications to NSAIDs (for example, inflammatory bowel disease, gastric ulcer or pre-eclampsia) and will not get sufficient pain relief from paracetamol alone. Based on their experience, the committee suggested an alternative of dihydrocodeine in addition to paracetamol, or co-dydramol, as these are also suitable for use while breastfeeding.
There was evidence for the effectiveness of oxycodone, and some evidence for tramadol, but the committee were aware both of these drugs can cause neonatal sedation and respiratory depression if used when breastfeeding. However, in women with severe pain, the committee agreed that a short course of tramadol or oxycodone could be considered as long as the woman was informed of the risks and chose to use them. The length of the course was not defined as there was no evidence for a specific period or dosage.
The committee were aware that there were general recommendations in the BNF on the use of opioids in breastfeeding women and so included these as part of the recommendations. The committee were also aware of an MHRA warning on the risk of serious neonatal respiratory depression and sedation with codeine in some women. Because of this, they recommended that codeine, or medications that include codeine (such as co‑codamol), should not be used, and that women should be advised not to use codeine‑containing medicines while breastfeeding.
Based on their knowledge and experience, the committee recommended that anti-emetics could be prescribed if needed for nausea and vomiting, and that laxatives should be considered for the prevention of constipation.
The dosing of intrathecal and epidural diamorphine will be simpler for staff and no resource impact is anticipated from this change. These recommendations will lead to increased use of preservative-free morphine if there is a shortage of diamorphine. Preservative-free morphine is more expensive than diamorphine so there may be an increased cost to the NHS, but as the frequency and duration of diamorphine shortages is not known, it is difficult to predict the resource impact.
The treatment of the possible side effects from morphine are not expected to have a resource impact as these will not affect all women, some women will already receive anti-emetics after a caesarean birth, and the medicines are generics.
The separate storage and labelling of different morphine preparations may require additional staff time, but as the frequency and duration of diamorphine shortages is not known, it is difficult to predict the resource impact.
The committee agreed that the recommendations on postoperative pain management would reinforce current practice. However, there may be a reduction in the use of intravenous PCA opioids for pain management after caesarean birth, and an increase in the use of oral morphine. The committee agreed that the recommendations relating to dihydrocodeine and codeine‑containing medicines would provide greater clarity and increase safety.