Need for blood transfusion

• Three randomised controlled trials (N=13171) provided very low quality evidence to show no clinically important difference in the need for blood transfusion between single layer and double layer uterine closure.

Additional surgical procedures

• One randomised controlled trial (N=9286) provided very low quality evidence to show no clinically important difference in the incidence of any further operative procedures between single layer and double layer uterine closure.
• One randomised controlled trial (N=9286) provided very low quality evidence to show no clinically important difference in the incidence of any further operative procedures on the wound between single layer and double layer uterine closure.

Uterine rupture (in subsequent pregnancy)

• One randomised controlled trial (N=3234) provided very low quality evidence to show no clinically important difference in the incidence of uterine rupture in a subsequent pregnancy between single layer and double layer uterine closure.
• Two randomised controlled trials (N=3378) provided very low quality evidence to show no clinically important difference in the incidence of uterine scar dehiscence in a subsequent pregnancy between single layer and double layer uterine closure.

Use of antibiotics

• Five randomised controlled trials (N=12713) provided low quality evidence to show no clinically important difference in the use of antibiotics for wound infection (and wound sepsis) between single layer and double layer uterine closure.
• Two randomised controlled trials (N=12265) provided very low quality evidence to show no clinically important difference in the use of antibiotics for febrile morbidity between single layer and double layer uterine closure.
• Two randomised controlled trials (N=10192) provided very low quality evidence to show no clinically important difference in the use of antibiotics for endometritis between single layer and double layer uterine closure.

Morbidly adherent placenta/abnormal invasion of placenta (in subsequent pregnancy)

• One randomised controlled trial (N=3233) provided very low quality evidence to show no clinically important difference in the incidence of morbidly adherent placenta in a subsequent pregnancy between single layer and double layer uterine closure.

Peri-partum hysterectomy (in subsequent pregnancy)

• One randomised controlled trial (N=3234) provided very low quality evidence to show no clinically important difference in the incidence of hysterectomy (during the 6 weeks postpartum) in a subsequent pregnancy between single layer and double layer uterine closure

Caesarean birth (in subsequent pregnancy)

• Two randomised controlled trials (N=3421) provided low quality evidence to show no clinically important difference in the incidence of caesarean birth in a subsequent pregnancy between single layer and double layer uterine closure

Need for blood transfusion

• Three randomised controlled trials (N=1324) provided very low quality evidence to show a clinically important reduction in the need for blood transfusion when using a caesarean birth (CB) technique that included single layer closure, as compared to a technique that included double layer closure.

Additional surgical procedures

• No evidence was available for this outcome.

Uterine rupture (in subsequent pregnancy)

• No evidence was available for this outcome.

Use of antibiotics

• Two randomised controlled trials (N=196) provided very low quality evidence to show no clinically important difference in the use of antibiotics (for an unspecified reason) between CB techniques that included single or double layer uterine closure.
• Three randomised controlled trials (N=1324) provided very low quality evidence to show no clinically important difference in the use of antibiotics (for wound infection or wound sepsis) between CB techniques that included single or double layer uterine closure.

Morbidly adherent placenta/abnormal invasion of placenta (in subsequent pregnancy)

• No evidence was available for this outcome.

Peri-partum hysterectomy (in subsequent pregnancy)

• No evidence was available for this outcome.

Caesarean birth (in subsequent pregnancy)

• No evidence was available for this outcome.

The outcomes that matter most

As double layer closure is currently standard practice, the committee wished to determine if single layer closure is as effective and safe. The committee therefore specified three critical outcomes, which were of primary importance for this review. These were the need for blood transfusion as this is an indication of how successful the surgical closure is, the need for additional surgical procedures in the short term as failure of the closure may require the patient to return to theatre, and the occurrence of uterine rupture in a subsequent pregnancy.

In addition, four important outcomes were identified. These were the use of antibiotics within 7 days of the caesarean birth which may be an indicator of wound infection, and three outcomes related to future pregnancies - the presence of morbidly adherent/abnormally invasive placenta, the need for peri-partum hysterectomy, and caesarean birth in a future pregnancy.

The quality of the evidence

Despite a number of large, well conducted trials in this area, the evidence was downgraded in all studies for a high risk of performance bias (due to the inability to blind the surgeon to allocation). Some studies did not blind outcome assessors to the allocated intervention, therefore were also at high risk of detection bias.

Studies comparing different caesarean surgical methods (comparison 2) were downgraded for indirectness as they compared other differences in technique as well as uterine closure (differences in uterine incision, exteriorisation of the uterus, peritoneal closure, skin closure, and suture material). Finally, a number of rare events were included as relevant outcomes in this evidence review (such as uterine rupture and peri-partum hysterectomy). The small number of events that occurred led to a wide confidence interval around the result, meaning that the data was downgraded for imprecision.

Overall the data was considered to be low to very low quality.

Benefits and harms

The committee reviewed the evidence presented as two separate comparisons. For the trials that specifically compared single and double layer closure, no clinically important difference was identified for any of the outcomes. These trials were considered to most accurately reflect the difference between single and double layer closure. When assessing evidence from the trials which randomised women to different caesarean surgical techniques, the committee noted that the only difference in outcomes was an increased chance of requiring a blood transfusion when a double layer uterine closure technique was used. This comparison had multiple confounding factors as it compared completely different caesarean techniques, and not just uterine closure technique. The committee agreed that the additional blood loss would also be anticipated as a result of the difference in method of opening the abdomen and uterus: in the arm that included double layer closure this involved cutting using scissors/scalpel, compared to the blunt entry used in the arm that included single layer closure.

The committee discussed the low event rate of uterine rupture in a subsequent pregnancy using either technique, as historically this concern was used as the rationale for double layer closure, and agreed that the low incidence and lack of difference between the techniques was reassuring, and indicated that either method could be used safely.

The committee discussed the length of the caesarean procedure, and the desire for clinicians to close the abdomen as quickly as possible, in order to minimise the potential for infection. The committee agreed that the lack of difference in infection rates (antibiotic use) between single and double layer closure was encouraging as it suggested the additional time taken to close the uterus using a second layer of sutures did not give rise to an increased chance of infection.

As there was no difference between single and double layer closure for the majority of outcomes the committee agreed that either technique could be used. The committee added the information about the similar risks of bleeding or uterine rupture in a subsequent pregnancy to the recommendation to provide further context for surgeons who may not be familiar with single layer closure and who may be concerned that single layer closure could increase the risk of these adverse events.

The committee discussed the differing levels of experience of those performing a caesarean procedure, and whether a separate recommendation should be made for those with less experience, but agreed that by recommending that either closure method could be used, the decision to use single or double layer could be made on an individual basis for each woman. The committee discussed which factors should be taken into consideration when deciding which closure to use, and agreed that surgeons would make an individual choice based on the clinical circumstance which would include an assessment of the woman’s clinical presentation. The committee discussed the fact that surgeons may also have different preferences for single or double closure, and that this is in turn may be influenced by their level of experience.

The committee reiterated that where additional layers of suturing were required (for example, due to ongoing bleeding), the surgeon or treating clinician would continue to respond appropriately, as is current practice.

The committee discussed the fact that as double layer uterine closure is currently recommended, surgeons who, for clinical reasons, decide to carry out single layer closure currently document this in the notes. As the recommendation now allows the choice of either single or double layer closure, it will no longer be seen as necessary to justify why a single layer uterine closure has been carried out.