Reply: Possibility of hidden damages with temporary uterine artery occlusion device

Sir, Dr Palacios Jaraquemada raises concerns of ureteric and ovary damage following the treatment of women with symptomatic fibroids with temporary, Doppler-directed, transvaginal uterine artery occlusion with the flostat system (Vilos et al. 2006). At this stage in the system’s development, we have limited knowledge about the risks of either ureteric or ovarian damage. We are, however, pleased to share what we do know. From development work completed before the two pilot trials we have performed, it is known that the flostat clamp does not come into contact or apply pressure to the pelvic ureter. During direct laparoscopic observation, it was seen that the pelvic ureters are lateral to the closed clamp (Lichtinger et al., 2005). The flostat clamp is a vascular clamp that exerts closing pressures in the range of pressures exerted by other Food and Drug Agency (FDA) cleared vascular clamps. It is not a tissue-crushing clamp. When closed, the flostat clamp folds vaginal tissue and the uterine arteries against the lateral walls of the uterus. When the bladder is empty, tissue in the trigone of the bladder could be distorted as the vaginal and vascular tissue is folded. To mitigate this possibility, we now apply the flostat clamp only with the urinary bladder full. Bilaterally occluding the uterine arteries at the level of their junction with the lateral walls of the isthmus does not entirely stop blood flow to the uterus from the uterine arteries. When temporary, transvaginal uterine artery occlusion is performed with the flostat clamp, small, unnamed arteries within the broad ligament immediately dilate proximal to the clamp and supply a trickle of flow to the ascending branch of the uterine artery. With this degree of ischaemia, no woman whom we have treated has experienced amenorrhoea, and most women treated have had relief from their fibroid related symptoms. Although we agree that amenorrhoea may be seen as we treat more and more women in the future, we do not believe that the frequency of ovarian damage will be anywhere near the 10% range suggested by Palacios Jaraquemada. Our belief rest upon the following observations. Women have two ovaries, not one. To render a woman amenorrhoeic following uterine artery occlusion, both ovaries would have to be damaged. Harming one ovary, although undesirable, is not the same as harming two. For a woman wanting to bear a child, harming both ovaries would be a tragedy. In most women, each ovary has a dual, or redundant, arterial blood supply. Most ovaries receive blood from an ipsilateral ovarian artery and from a communicating artery that connects with the ascending branch of the ipsilateral uterine artery (Binkert et al., 2001). Palacios Jaraquemada is concerned about women with vascular anatomy variants who have both ovaries supplied exclusively or ‘mainly’ by their uterine arteries. Vascular anatomy variations are random events that occur during embryology. In general, they are not correlated with one another and the probability of observing two vascular anatomy variations in the same women is less than 5% (Razavi et al., 2004). Although the unilateral arterial supply of one ovary from its ipsilateral uterine artery has been described at a frequency between 4 and 6.6%, no article in the world’s medical literature has ever described both ovaries being primarily supplied by its ipsilateral uterine artery. It is, therefore, only a theoretical possibility that such an event can occur, at all. And if it does occur, the probability of this event is less than half the number proposed by Palacios Jaraquemada, somewhere between 1.6 and 4.4% (Razavi et al., 2002). Consequently, though an interesting theoretical concern, we cannot find data to support Palacios Jaraquemada statement that ‘in approximately 10% of cases, the ovary circulation depends mainly on the uterine artery’, nor can we find a single published case example of a woman whose ovaries are bilaterally ‘mainly’ dependent on circulation from the uterine arteries. It is now well established that a woman can become amenorrhoeic following uterine artery embolization (UAE). However, UAE is an entirely different procedure than temporary, transvaginal uterine artery occlusion with the flostat system. During UAE, permanent, flow-directed particles form clot in arteries, wherever the particles ultimately land. Because communicating arteries connect the ascending branch of the uterine artery with the ovarian circulation, UAE particles can and do reach the ovaries. Pathology studies have demonstrated UAE particles in and around the ovaries following UAE (Payne et al., 2002). Despite these pathological findings, the frequency of ovarian failure following UAE in women below the age of menopause is much less than the 10% predicted by Palacios Jaraquemada. Pron et al. (2003) tabulated the frequency of cessation of menstruation following UAE by age. In women 44 years and younger, 5 of 305 (1.6%) ceased menstruating following UAE. In peri-menopausal women (women 50 years and older), the frequency of amenorrhoea following UAE was much higher, 26%. The average age of menopause in the United States is 51.4 years. Others have confirmed the findings of Pron et al. (2003) in younger women. Tropeano et al. (2004) measured day 3 FSH levels, ovarian volumes and follicle numbers in 20 normally cycling women aged 33–39 prior to and at 3, 6 and 12 months following UAE. UAE had no short- or mid-term effects on ovarian reserve. This information, taken together, leads us to predict that following transvaginal uterine artery occlusion, some younger women may be at risk of amenorrhoea, but the risk would be small, in the 1% range, not the 10% range as suggested by Palacios Jaraquemada. We thank Palacios Jaraquemada for his interest in our work and for his thoughtful opinions. We anticipate that as further clinical studies are performed, clinical data will accumulate to address his concerns, more directly.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.