| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Castillo 61, Piso 2, Departamento B, (1414) Buenos Aires, Argentina
Editor:
I read with great interest the article by Dr Pelage and colleagues in the August 1998 issue of Radiology (1) describing selective arterial embolization for postpartum hemorrhage.
Selective embolization has proved to be a useful and safe method to control postpartum hemorrhage, but it is not clear why Dr Pelage and colleagues consider uterine artery ligation a technically difficult procedure. Uterine artery embolization was performed in 85% (n = 23) of the 27 patients; 67% (n = 18) of the 27 patients had uterine atony, a condition that according to O'Leary (2) responds to simple ligation in 95% of cases, and underwent uterine artery embolization. Embolization failure in two patients with placenta accreta is not surprising, since in this condition the involved pedicles depend on infracervical vessels (cervicovaginal and vaginal branches of the internal iliac and internal pudendal arteries), which perhaps are not explored with angiography.
It is also noteworthy that both ovarian arteries were embolized to control bleeding after hysterectomy, since in the surgical procedure those vessels are excised. Consequently, it is uncertain which pedicles have been embolized.
I do not believe that therapeutic measures are designed to correct hemodynamic and hemorrhagic disorders; the key issue is to prevent bleeding.
I congratulate Dr Pelage and colleagues for their excellent study. However, to promote bilateral uterine artery ligation or uterine defibrillation for uterine atony, reserving embolization for those cases of placenta accreta or percreta in which a rapid and safe arterial isolation cannot be performed, I suggest first a temporary balloon occlusion of the abdominal aorta at the level of the L4 vertebra (3) in patients with placenta accreta or percreta and second an evaluation of the infracervical pedicles, especially the vaginal branches of the internal pudendal artery, which in pregnant women are larger than the uterine artery.
In my experience (unpublished data, 1998) with 30 cases of uterine atony or placenta accreta or percreta, selective ligation of supra- and infracervical pedicles allows for uterine preservation with an average blood loss of 500 mL. However, I agree with Dr Pelage and colleagues that surgeons are not usually familiar with the origin and variations of these pelvic-subperitoneal vessels and that selective angiography is crucial.
References
Departments of Vascular and Body Imaging
Obstetrics and Gynecology, Hôpital Lariboisière, 2 rue Ambroise Paré, 75475 Paris 10, France
We thank Dr Palacios Jaraquemada for his interest in our study (1) and for his constructive comments. Dr Palacios Jaraquemada has raised several important issues concerning the use of arterial embolization to control primary postpartum hemorrhage. Uterine atony, which to our knowledge represents the main cause of immediate bleeding following delivery, is usually managed by intravenously administering oxytocin and similar drugs and prostaglandin E2 or F2
analogues. Dr Palacios Jaraquemada cites an article (2) that reports a high success rate of vascular ligation to control uterine atony in case of failure of medical treatment. We agree with Dr Palacios Jaraquemada. For patients with moderate bleeding at the time of cesarean delivery, ligation of the uterine artery is probably the simplest treatment (2). Conversely, in cases of vaginal delivery, arterial embolization should be performed first. The reason is that vascular ligation can be technically difficult in a patient with severe coagulopathy.
Because of distorted anatomy, identification of arterial pedicles requires a high degree of surgical skill. In addition, the reported success rate of vascular ligation is extremely variable and depends on the cause of bleeding and the surgical procedure. Proximal internal iliac ligation fails to stop the bleeding in up to 50% of cases (3,4), whereas selective or successive ligation of uterine and ovarian pedicles seems to be more efficient (2,5). Conversely, embolization, which is always possible even in cases of severe coagulopathy, does not preclude further surgical repair if necessary. Disappearance of coagulopathy within the few hours following embolization occurred in all our patients. We disagree with Dr Palacios Jaraquemada in that the key issue of embolization is not only to prevent bleeding but also to treat hemorrhagic disorders.
In our study, failure to control hemorrhage with arterial embolization occurred in one patient with placenta accreta (Dr Palacios Jaraquemada's statement about failure in two women is incorrect). However, delayed hysterectomy was performed after coagulopathy had resolved. Placenta accreta remains one of the major causes of removal of a gravid uterus (6). Thus, Dr Palacios Jaraquemada's experience with conservative treatment for placenta accreta, although the number of women with this condition is not stated, is of real interest. Until now, we have performed uterine embolization in seven women with histopathologically proved placenta accreta (including one woman with placenta percreta). Hysterectomy has been necessary for only one woman. Our success rate for this indication is higher than that previously reported (7). Concerning the prophylactic placement of a temporary balloon (8) or catheter (7) in women with a high risk of postpartum bleeding, we believe that the potential risks associated with these measures (ie, arterial dissection, thrombosis, secondary displacement) do not balance with the benefit of embolization performed only in case of hemorrhage.
We emphasize several technical considerations. During pelvic angiographic procedures, selective catheterization of the internal iliac artery and study of the anterior stem are performed, thus allowing for identification of the internal pudendal, uterine, and vaginal arteries. When needed, superselective catheterization of these branches can then be attempted. Therefore, infracervical vessels can be seen and superselective embolization can be achieved. In addition, a common trunk between the uterine artery and the vaginal artery is a frequent finding; the cervicovaginal artery is a branch that arises from the uterine artery in 50% of cases (9). This is the reason that embolization of the uterine arteries is effective in most cases.
As mentioned in our article, embolization of the anterior division of the internal iliac artery, which maintains flow to the branches of the posterior stem, should be an alternative technique in case of severe vasoconstriction (1). In selected cases (eg, placenta accreta), we add a few pledgets of an absorbable gelatin sponge (Gelfoam; Houde, Compiegne, France) into the anterior stem of the internal iliac artery even after successful superselective embolization of the uterine artery to obtain temporary devascularization of the infracervical vessels mentioned by Dr Palacios Jaraquemada.
Finally, our experience definitely demonstrates the value of arterial embolization in a postoperative setting. In our study, two women were successfully treated with embolization after failure of subtotal hysterectomy (10). Slipped ligature of the ovarian arteries, which probably tend to retract when the ligated pedicle untwists to its normal position, probably accounts for the recurrence of bleeding in our patient.
We thank Dr Palacios Jaraquemada for his interesting surgeon's perspective. Again, multidisciplinary treatment, involving the radiologist, the obstetrician, and the anesthesiologist, is key to a high success rate for this procedure in these women.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
