We read with interest the study by Maassen et al.1 concerning uterine artery embolisation (UAE) for postpartum haemorrhage (PPH) and have several comments to make about this work.

First, we were surprised to learn that in the Netherlands, PPH is defined by a blood loss in excess of 1000 ml, whatever the method of delivery, whereas the usual threshold for vaginal delivery is 500 ml. Current recommendations2 concerning UAE were respected in this study (usage after failure of uterotonics and after transfusion of packed red cells and platelets), but the delay in the diagnosis, because of the excessively high threshold, may have been the main cause of the failure of the treatment for PPH.

Moreover, concerning the technique for embolisation, we would like to underline the need to create an angiographic map of the vascular bed before the intervention for occlusion. Eventhough it may be difficult to establish such a map because of vasospasm induced by the haemodynamic state, it does allow detection of the zone responsible for the bleeding as well as anastomoses notably utero-ovarian that may exist. It is then possible to obturate these anastomoses proximally, using microcoils, even before UAE to maintain ovarian function.

Moreover, several cases of uterine necrosis secondary to UAE have been reported,3 and this despite the use of resorbable material. In the present study, the vesico-vaginal fistula, secondary to the vesical necrosis, is a further example to confirm sometimes irreversible occlusion caused by the gelatin particles. The size of the particles and the degree of selectivity in the site of embolisation seem to play a major role in the genesis of ischaemic necrosis. The smaller the particles, the more distal the embolisation and the lower the risk of recurrence of bleeding, at the price, however, of a greater risk of ischaemia (impossibility to develop collateral circulation). In contrast, the bigger the particles, the more proximal the embolisation and the lower the risk of ischaemia. The risk of recurrent haemorrhage, however, is greater because of the development of collaterals.

Maassen et al.1 also reported the risk of gelatin particles migrating to the femoral artery in the case of iterative interventions. Thus, it seems irresponsible to continue to perform re-embolisations, sometimes sacrificing the two ovarian arteries in the process, in patients who will be able to preserve the uterus, but at the price of definitive menopause.

In cases of severe PPH, Touboul et al.4 confirmed the interest of UAE in uterine atony in vaginal delivery. In contrast, the efficacy seems to be less certain in cases of haemodynamic instability or for cesarean section. They propose a clinical trial to clarify the interest of UAE in comparison with surgery (ligature of the hypogastric arteries or hysterectomy) in this life-threatening situation.

Uterine artery embolisation has certainly become an essential tool in the management of PPH. However, we must advise his reasoned use in second-line treatment, by avoiding re-embolisations, as its efficacity compared with surgery has not yet been clearly established in cases of extremely severe PPH.


  1. Top of page
  2. References