Association of peripartum management and high maternal blood loss at cesarean delivery for placenta accreta spectrum (PAS): A multinational database study.
Schwickert A., van Beekhuizen HJ., Bertholdt C., Fox KA., Kayem G., Morel O., Rijken MJ., Stefanovic V., Strindfors G., Weichert A., Weizsaecker K., Braun T., International Society for Placenta Accreta Spectrum (IS-PAS) None.
INTRODUCTION: Placenta accreta spectrum (PAS) carries a high burden of adverse maternal outcomes, especially significant blood loss, which can be life-threatening. Different management strategies have been proposed but the association of clinical risk factors and surgical management options during cesarean delivery with high blood loss is not clear. MATERIAL AND METHODS: In this international multicenter study, 338 women with PAS undergoing cesarean delivery were included. Fourteen European and one non-European center (USA) provided cases treated retrospectively between 2008 and 2014 and prospectively from 2014 to 2019. Peripartum blood loss was estimated visually and/or by weighing and measuring of volume. Participants were grouped based on blood loss above or below the 75th percentile (>3500 ml) and the 90th percentile (>5500 ml). RESULTS: Placenta percreta was found in 58% of cases. Median blood loss was 2000 ml (range: 150-20 000 ml). Unplanned hysterectomy was associated with an increased risk of blood loss >3500 ml when compared with planned hysterectomy (adjusted OR [aOR] 3.7 [1.5-9.4], p = 0.01). Focal resection was associated with blood loss comparable to that of planned hysterectomy (crude OR 0.7 [0.2-2.1], p = 0.49). Blood loss >3500 ml was less common in patients undergoing successful conservative management (placenta left in situ, aOR 0.1 [0.0-0.6], p = 0.02) but was more common in patients who required delayed hysterectomy (aOR 6.5 [1.7-24.4], p = 0.001). Arterial occlusion methods (uterine or iliac artery ligation, embolization or intravascular balloons), application of uterotonic medication or tranexamic acid showed no significant effect on blood loss >3500 ml. Patients delivered by surgeons without experience in PAS were more likely to experience blood loss >3500 ml (aOR 3.0 [1.4-6.4], p = 0.01). CONCLUSIONS: In pregnant women with PAS, the likelihood of blood loss >3500 ml was reduced in planned vs unplanned cesarean delivery, and when the surgery was performed by a specialist experienced in the management of PAS. This reinforces the necessity of delivery by an expert team. Conservative management was also associated with less blood loss, but only if successful. Therefore, careful patient selection is of great importance. Our study showed no consistent benefit of other adjunct measures such as arterial occlusion techniques, uterotonics or tranexamic acid.