INTRODUCTION
Although the lower uterine segment usually heals well after Caesarean section (CS), any loss of muscle during the repair, which can be substantial even after one CS, leads to niche formation (1). Niche formation has been linked to a number of modifiable surgical factors including single layer closure and use of locking sutures. Both niche prevalence and depth increase with number of CS, prevalence rising from 62% after one to 78% after three CS depending on method of evaluation (2). The nearer the scar is to the level of the internal os the higher the likelihood of a niche and the larger the defect (2). Niche development may be accompanied by formation of new vascular connections within the scar area, which along with in-migration of a variety of inflammatory cells contributes to healing (3).
During future pregnancies the niche has the potential to become an isthmocele leading to complications including uterine rupture and placenta accreta spectrum disorders (PASD). As the niche deepens, the residual myometrial thickness (RMT) gets less, with the RMT further thinning during pregnancy as the niche widens. The longer and deeper the initial scar area, the larger the decrease in RMT seen in pregnancy and the larger the resultant defect (4). The likelihood of developing complications in future pregnancy is related to the RMT, as it reflects the amount of muscle remaining. An RMT of less than 2mm carries a significant risk of scar rupture (5).
The frequency of PASD has markedly increased over the last few years because of the increased Caesarean section rate. The risk is higher when placenta praevia complicates a pregnancy following previous CS, especially if the primary indication mandated a higher uterine scar, and rises with each subsequent CS (6). Historically treatment options centred round intentional retention of the placenta (IRP) with or without hysterectomy but latterly several techniques have been developed aimed at conserving the uterus including the Triple P (7) and Naussica (8) procedures.
Recent evidence of PASD being a disorder of defective decidua and uterine dehiscence rather than destructive trophoblastic invasion (9) suggests that a conservative approach with effective manual removal of placenta, when combined with peri-operative prophylactic haemostatic measures can be an optimal approach.
We have previously described a technique (Siraj et al) for repair of the myometrial defect or isthmocele at the site of the previous scar at the time of repeat Cesarean section (10). The repair increases the RMT and reinforces the posterior uterine wall which is often thin in these cases. We now describe how this technique can be extended for use in cases of suspected PASD with focus on control of the major hemorrhage, in part due to the neovascularization around the scar which can be offered as an alternative to peripartum hysterectomy in women who request uterine conservation. This approach has the added advantage of offering opportunity to reconstitute the uterine wall by contemporaneous correction of the niche, potentially allowing future pregnancy and avoiding other niche-related complications.