Although aromatase inhibitors are not approved by the United States Food and Drug Administration (FDA) for any indication except postmenopausal breast cancer, they have been used as a clinical treatment for uterine leiomyomas, resulting in a reduction in symptoms such as heavy menstrual bleeding and pelvic pain, as well as a decrease in leiomyoma volume1. Another gynecological benign condition for which aromatase inhibitors are being investigated is endometriosis, and their use in combination with gonadotropin-releasing hormone (GnRH) analogs can prevent the initial estrogen flare effect that they cause2. Aromatase inhibitors are usually well tolerated and cause no changes in systemic hormonal status3. It is well known that GnRH analogs promote a decrease in uterine volume and an increase in uterine artery resistance indices4. However, there are no previous studies analyzing uterine Doppler velocimetry after use of aromatase inhibitors; we therefore conducted this pilot study. The protocol was approved by the local research ethics committee. We included patients from 40 to 60 years of age who were scheduled for vaginal hysterectomy for the treatment of leiomyomas, with uterine volume ranging from 200 to 500 cm3. Exclusion criteria were patients with renal and/or hepatic dysfunction and hypersensitivity to anastrozole.
After providing written informed consent, eight patients received 1 mg oral anastrozole once daily for 12 consecutive weeks. Transvaginal ultrasound examinations were performed using an ATL HDI 3000CV (Universal Diagnostic Solutions, Oceanside, CA, USA) ultrasound machine. Conventional color Doppler imaging was used to evaluate blood flow in the uterine arteries and in the largest leiomyoma volume if the patient presented with several tumors. The resistance index (RI), pulsatility index (PI) and peak systolic velocity (PSV) were calculated for right and left uterine arteries and for the largest leiomyoma. For each Doppler variable, two measurements were taken for the leiomyoma and for the uterine artery on each side, with the mean of the two measurements used for further analysis. Scans were performed by the same examiner before and after treatment. The dimensions of the largest leiomyoma and uterus were measured and their volumes were estimated using the formula for an ellipsoid. The power of our data to detect an increase of 0.1 in RI was 90% and 70% to detect an increase of 0.5 in PI.
The median age of the women included in the study was 45 years. All were premenopausal and all were parous. Three patients had each undergone two previous Cesarean sections, and the others had delivered vaginally only. Regarding the location of fibroids, all were intramural, except in one patient who presented with a subserosal leiomyoma. All of the patients had heavy menstrual bleeding and two patients reported pelvic pain. After administration of anastrozole, mean ( ± SD) uterine volume decreased from a basal level of 415.82 ± 82.35 cm3 to 331.65 ± 100.98 cm3 (20% reduction; P = 0.04 (Wilcoxon test)) and mean volume of the largest leiomyoma decreased from 44.22 ± 21.09 cm3 to 26.10 ± 25.03 cm3 (40.9% reduction; P = 0.01 (Wilcoxon test)). Blood flow index measurements are presented in Table 1. No significant changes were noted in any Doppler parameters. Heavy menstrual bleeding was reduced in all patients, but no change was reported in patients with pelvic pain.
|Parameter||Before anastrozole||After anastrozole||Change||P*|
|Largest leiomyoma||0.66 ± 0.10||0.43 ± 0.29||− 0.23||0.12|
|Right uterine artery||0.82 ± 0.08||0.81 ± 0.09||− 0.01||1.00|
|Left uterine artery||0.80 ± 0.09||0.82 ± 0.15||+ 0.02||0.62|
|Largest leiomyoma||1.46 ± 0.63||0.75 ± 0.50||− 0.71||0.25|
|Right uterine artery||2.17 ± 0.71||2.06 ± 0.55||− 0.11||0.87|
|Left uterine artery||1.94 ± 0.65||2.14 ± 0.96||+ 0.20||0.87|
|Peak systolic velocity (cm/s)|
|Largest leiomyoma||21.57 ± 10.29||5.8 ± 4.19||− 15.77||0.12|
|Right uterine artery||43.57 ± 3.29||46.4 ± 15.40||+ 2.83||1.00|
|Left uterine artery||28.2 ± 13.86||21.3 ± 8.17||− 6.9||1.00|
Our results confirm the effect of aromatase inhibitors on leiomyoma volume3 and suggest a non-vascular mechanism of action, as indicated by the absence of statistically significant changes in Doppler measurements. In order to investigate this, samples of fibroids should be histologically analyzed and the vascular system of leiomyomas should be further evaluated. It is likely that aromatase inhibitors do not interfere with vascularization of leiomyomas, but rather that their effect is due to their inhibition of local and peripheral estrogen synthesis. Our group intends to compare this aromatase inhibitor with the GnRH analog gosereline in a trial investigating surgical and sonographic outcomes (NCT01280045), similar to a previous randomized controlled trial with another aromatase inhibitor (letrozole)5. Further research is necessary to investigate whether Doppler flow indices are predictive of response to treatment with aromatase inhibitors.