A randomized trial evaluating leuprolide acetate before hysterectomy as treatment for leiomyomas Thomas G. Stovall, MD, Frank W. Ling, MD, Louie C. Henry, MD, and Marie R. Woodruff, RN Memphis, Tennessee Fifty premenopausal patients requiring hysterectomy as treatment for symptomatic uterine leiomyomas, which were the size of 14 to 18 weeks' gestation, were randomized into two groups to determine whether preoperative gonadotropin-releasing hormone agonist would increase the feasibility of vaginal rather than abdominal hysterectomy. The control group (group A; n = 25) did not receive preoperative gonadotropin-releasing hormone agonist, but patients in Group B (n = 25) received 2 months of gonadotropin-releasing hormone agonist before undergoing hysterectomy. Patients in the two groups were similar with respect to age, gravidity, parity, pretreatment uterine size, and hemoglobin and hematocrit levels. Patients in group B had an increase in hemoglobin levels (10.75 to 12.12 gm/dl, p < 0.05) and a decrease in uterine volume (1086.7 to 723.4 ml, p < 0.05) after 8 weeks of agonist therapy and were more likely to undergo vaginal hysterectomy (76.0% vs 16%). Patients in group B also had shorter hospitalizations (5.2 vs 3.8 days, p < 0.05). We conclude that the administration of gonadotropin-releasing hormone agonist for 2 months followed by vaginal hysterectomy is preferable to abdominal hysterectomy in selected patients with uterine leiomyomas. (AM J OBSTET GVNECOL 1991 ;164:1420-5.)

Key words: Leuprolide acetate, hysterectomy

When compared with abdominal hysterectomy, vaginal hysterectomy is associated with significantly fewer complications, less febrile morbidity, and the need for fewer transfusions. In addition, patients require shorter hospital stays and have a shorter convalescent period.' Uterine morcellation or intramyometrial coring techniques allow vaginal removal of uteri >700 gm.2.3 Despite the advantages of vaginal hysterectomy, some leading authorities in gynecologic surgery do not recommend vaginal hysterectomy when the uterus is larger than 10 to 14 weeks' gestational size! Many studies now show that administration of gonadotropin-releasing hormone (GnRH) agonist reduces both leiomyoma size and uterine volume and that the majority of this reduction occurs within the first 8 weeks of treatment. 5-9 The purpose of this study was to determine whether short-course preoperative GnRH agonist therapy before planned hysterectomy as treatment for uterine leiomyomas could result in increased From the DivisIOn of Gynecology, Department of Obstetncs and Gynecology, University of Tennessee, Memphis. Thts study was funded In part by TAP Pharmaceuticals. Certificate of Merit Award, presented at the Fifty-eighth Annual Meeting of the Central Assoctation of Obstetricians and Gynecologtsts, Louisville, Kentucky, October 11-13,1990. Reprint requests: Thomas G. Stovall, MD, Assistant Professor, DIvisIOn of Gynecology, 853 Jefferson Ave., Room E-102, Memphts,

TN 38163.

6/6/28578

1420

performance of vaginal hysterectomy. We also wished to determine if this treatment would be associated with a decrease in morbidity, hospital stay, or convalescent period.

Material and methods Premenopausal women with symptomatic uterine leiomyomas who did not desire future fertility voluntarily consented to participate in this study, which was approved by the University of Tennessee Institutional Review Board. Inclusion criteria included: (1) premenopausal status (follicle-stimulating hormone level 600 ml. The design of this study does not include placebo controls but does include a group of untreated patients (group A). Group B, the treated group, includes patients receiving a GnRH agonist, leuprolide acetate, by one of two routes. The groups were not evaluated over the same time period and did not have the same number of clinical or ultrasonographic studies before surgery. The patients, the ultrasonographer, and the surgeon were not blinded, so that an element of bias cannot be eliminated. This study could have been strengthened by incorporating blinding or placebo procedures. The primary thesis cannot be tested because all patients in group B were randomized not to determine whether they were candidates for vaginal or abdominal hysterectomy before therapy but, rather receive therapy regardless of uterine size. Reference to Table II shows that in group B uterine size varied from 12 to 18 weeks' gestational size. The authors have pointed out in the Material and methods section that all patients with uterine size :5 14 weeks' gestational size were considered candidates for vaginal hysterectomy; therefore we should conclude that at least one patient (if not more) in group B had uterine size :514 weeks' size before therapy and that preoperative therapy made no difference in qualifying those patients for vaginal hysterectomy as a result of uterine size only. Similarly, in group A (the untreated patients) the mean uterine size was 14.4 weeks, yet only 4 of 25 patients underwent vaginal hysterectomy. The authors provide us with no standard deviations or measures of variance; consequently, it is not possible to determine how many patients in each group should have qualified for vaginal hysterectomy as a result of initial uterine size. Nor is it possible to determine if Student's t test is appropriate for assessing significance. To make meaningful comments about the blood loss in surgery, hospital stay, and period of convalescence in relation to the use of agonist therapy, it may be more appropriate to stratify the comparison between untreated and treated groups by the route of the hysterectomy. It is possible that agonist therapy could improve these end points irrespective of the route of surgery, but we are unable to assess this possibility with the data presented. No information correlating clinical estimates of uterine size, ultrasonographic volume determination of uterine size, and uterine weights in individual patients was provided. Such information might help us evaluate the role of these variables in clinical decision-making. I hope the authors will include their observations in a final manuscript.

June 1991 Am J Obstet Gynecol

A reasonable question is, what do ultrasonographically determined uterine volumes mean? In both groups A and B the uterine weights were approximately 45% of the uterine volume. Does that mean that the uterus has a density one half that of water? Using Archimedes' principle, we recently measured the volume of water displaced by extirpated uteri from 10 different patients and found that the true uterine volume closely approximates the weight of the uterus in grams. In an animal study a group of Japanese investigators determined ultrasonographic volume of the canine prostate and subsequently compared the ultrasonographic determinations with specific gravity, actual volume, and measured weights of the canine prostates determined after prostatectomy. They showed the specific gravity of the canine prostate to be approximately 1.06 and that an accurate weight could be determined by multiplying volume by specific gravity.s Ultrasonographic evaluation of the pelvis may help to assess adnexal structures; however, I can find no evidence in this study or elsewhere that uterine volume measured by ultrasonography can be translated into estimates of uterine weight or gestational size. It is likely that the irregular contours of some uteri do not correspond to any single geometric formula for volume determination; therefore use of a single formula should produce an overestimation of true uterine volume, particularly in women with leiomyomas. The authors concluded that preoperative therapy with GnRH agonist "allows selected patients safely to undergo vaginal hysterectomy." In group B 9% of patients who had vaginal hysterectomy attempted required abdominal completion of the surgery. In severallarge series approximately I % of patients who had vaginal hysterectomy attempted required abdominal completion.' The current report suggests that the authors' incidence of failed vaginal hysterectomy is a significant problem that could outweigh any benefits that may be ascribed to GnRH agonist therapy. In summary, the addition of placebo controls and the use of blinded observations would have provided meaningful data regarding the possible effects of preoperative GnRH agonist therapy. In addition, the correlation of clinical examination, ultrasonographic volume determination, and uterine weights in individual patients might have provided valuable information for the clinician who is trying to determine which patient with leiomyomas is a candidate for vaginal hysterectomy. REFERENCES 1. Schlaff WD, Zerhoun EA, Huth JAM, Chen J, Dame-

wood MD, Rock RA. A placebo-controlled trial of depot gonadotropin-releasing hormone analogue (leuprolide) in the treatment of uterine leiomyomata. Obstet Gynecol 1989;74:856-62.

2. Friedman AJ, Rein MS, Harrison-Atlas D, Garfield JM,

Doubilet PM. A randomized, placebo-controlled, doubleblind study evaluating leuprolide acetate depot treatment before myomectomy. Ferti! Steril 1989;52:728-33. 3. Miyashita H, Watanabe H, Ohe H, Saitoh M, Oogama Y,

Leuprolide before hysterectomy

Volume 164 Number 6, Part 1

Tijiasa S. Transrectal ultrasonotomography of the canine prostate. Prostate 1984;5:453-7. 4. Dicker RC, GreenspanJR, Strauss LT, et al. Complications of abdominal and vaginal hysterectomy among women of reproductive age in the United States: the collaborative review of sterilization. AM J OBSTET GYNECOL 1982; 144:841-8.

DR. PEDRO A. POMA, Melrose Park, Illinois. I want to state my bias. I prefer vaginal hysterectomy if possible. If the authors are going to continue with their studies, I would like them to consider a group in whom progestin could be used. On many occasions, using progestin, which is probably one-tenth of the cost of this other fancy new product, would produce the same results. DR. DAVID V. FOLEY, Wauwautosa, Wisconsin. I stand, not to comment on this article, although I thought it was a very meaningful one, but to challenge academicians to get back to teaching vaginal surgery. I have been privileged over the last 14 years to be a board examiner and have been agonized by the lack of vaginal surgery that has been taught to and done by the candidates. At the last examination I examined 20 candidates. In their entire case list there were two vaginal hysterectomy cases. I asked each one of these candidates what he or she would do with a traumatic cystotomy, and not one of them could handle the situation independently. I think that this is a travesty, and I think that we should get back to doing pelvic surgery and teaching pelvic surgery. DR. ROBERT K. ZURAWIN, Houston, Texas. I guess that (Dr. Foley's comment) stole a little bit of the thunder from what I was going to say. The authors mentioned that in the treatment study 16 of the 19 patients who underwent hysterectomy required morcellation. That also indicates that a thorough knowledge of vaginal operative technique is necessary. When I was taught vaginal surgery, the decision to perform a hysterectomy abdominally as opposed to vaginally had to do with your clinical judgment. In other words, the success of a vaginal hysterectomy had to do with the proper selection of patients. I did not hear any mention of whether the gravidity of the patient affected selection, nor was there any mention of pelvic anatomy, such as the narrowness of the arch, the width of the ischial spines, or those things that might also influence a surgeon to attempt a hysterectomy vaginally. But, as the previous questioners mentioned, I think that if you're going to embark on a hysterectomy with a uterus of this size, especially those that either before or after treatment are going to require morcellation, you're going to need a certain amount of surgical skill that medical treatment is not going to affect one way or the other. DR. STOVALL (Closing). As most of you are probably aware, the United States Food and Drug Administration is currently requiring all agonist manufacturers

1425

to perform blinded placebo-controlled trials to determine the efficacy of preoperative GnRH agonist prior to myomectomy or hysterectomy. We are currently involved in three such trials, and we believe that these additional data will confirm the data presented today. It is very difficult, however, when doing clinical trials with agonists to do placebo-controlled, blinded trials because of the effects of the agonists, such as uterine size reduction, the hypoestrogenic effects with hot flashes, and vaginal dryness. Thus it is nearly impossible to perform a truly "blinded" trial. All patients in the groups initially had uteri that were 2:: 14 weeks' size. The size was confirmed before treatment or surgery by two examiners. To calculate uterine volume ultrasonographically, we used the formulas that have been shown in the past to correlate gestational size with uterine volume; this was done by Sauer et al. 1 and others from several institutions. Progestins, danazol, and other agents have been used in an attempt to reduce uterine size, but these have not been nearly as successful as GnRH agonists, or they have not been successful at all. This study was also done in an attempt to increase the number of vaginal hysterectomy candidates at our institution so that our residents would have increased training in vaginal surgery. As the number of general gynecology patients decreases and as the indications for hysterectomy decline, we have been faced with a decrease in the number of surgical candidates for resident training, as have other institutions. An outgrowth of this has been to increase the number of vaginal hysterectomies that our residents perform. To attempt vaginal hysterectomy for large uteri, you must be comfortable with pelvic anatomy, uterine morcellation, and other techniques used to remove large uteri. I think that more than that, you must be persistent in what you are doing and must have good operative assistance. We believe that there are probably three roles for GnRH agonist therapy for myomas. The first is for patients who have a low preoperative hematocrit, to increase the hemoglobin and hematocrit before an intended operative procedure and thereby to decrease the chance that the patient will require either intraoperative or postoperative blood transfusions. Second, we believe that selected patients can undergo conversion from abdominal hysterectomy to vaginal hysterectomy, which decreases their overall operative and postoperative morbidity. Finally, there are patients who will undergo myomectomy who might benefit from preoperative GnRH agonist therapy. Although these patients have not been clearly identified in the literature, certainly patients who have very large uteri probably do benefit from GnRH agonist therapy before myomectomy. REFERENCE 1. Sauer MV, Agnew C, Worthen N, et al. Reliability of ultrasound in predicting uterine leiomyoma volume. J Reprod Med 1988;33:612-4.

A randomized trial evaluating leuprolide acetate before hysterectomy as treatment for leiomyomas.

Fifty premenopausal patients requiring hysterectomy as treatment for symptomatic uterine leiomyomas, which were the size of 14 to 18 weeks' gestation,...
663KB Sizes 0 Downloads 0 Views