Buserelin acetate versus expectant management in the treatment of infertility associated with minimal or mild endometriosis: A randomized clinical trial

Buserelin acetate versus expectant management in the treatment of infertility associated with minimal or mild endometriosis: A randomized clinical trial

Buserelin acetate versus expectant management in the treatment of infertility associated with minimal or mild endometriosis: A randomized clinical tri...

786KB Sizes 0 Downloads 64 Views

Buserelin acetate versus expectant management in the treatment of infertility associated with minimal or mild endometriosis: A randomized clinical trial Luigi Fedele, MD,. Fabio Parazzini MD,b Enrico Radici, MD,e Luca Bocciolone, MD,., b Stefano Bianchi, MD,. Cosetta Bianchi, SCD,b and Giovanni Battista Candiani, MD" Milan and Bergamo, Italy OBJECTIVE: We performed a randomized clinical trial to evaluate the efficacy of intranasal 400 ,...g buserelin three times daily for 6 months versus expectant management in the treatment of infertile women with pelvic endometriosis stage I or II of the revised American Fertility Society classification. STUDY DESIGN: Seventy-one consecutive patients (mean age 32 years) were studied at the First Department of Obstetrics and Gynecology, University of Milan, and the Department of Obstetrics and Gynecology, Ospedali Riuniti, Bergamo, between February 1988 and June 1989. Thirty-five women were randomly allocated to buserelin treatment and 36 to expectant management. The baseline distribution of subjects for age, disease stage, and reproductive history was similar in the two groups. All patients were followed regularly; median follow-up was 17 months in the buserelin group and 18 months in the women given expectant management. If pregnancy did not occur within 12 months of randomization, cycles were monitored by Ultrasonography and hormone measurements, and when abnormalities were detected clomiphene citrate and human chorionic gonadotropin were administered. RESULTS: A total of 17 pregnancies were observed both in the buserelm-treated patients and in the expectant management group. The 1- and 2-year actuarial overall pregnancy rates were similar in the two groups, 30% and 61% in the former and 37% and 61% in the latter group, respectively. Spontaneous abortion occurred in five of the 17 pregnancies in the women treated with buserelin and in one of the 17 in those managed expectantly; this difference was, however, not statistically significant (x~ adjusted for disease stage and use of clomiphene citrate and human chorionic gonadotropin treatment = 3.01, P = 0.08). No fetal death or stillbirth was observed. CONCLUSIONS: Our findings suggest that treatment with gonadotropin-releasing hormone agonists is unlikely to have a marked influence on the reproductive outcome of infertile women with minimal or mild endometriosis. (AM J OBSTET GVNECOL 1992;166:1345-50.)

Key words: Buserelin acetate, infertility, endometriosis

Pelvic endometriosis is a common gynecologic disease. Because it is often found in patients with otherwise unexplained infertility, it has been commonly implicated as a cause of infertility. Surgical ablation of endometriotic lesions is the most widely performed therapy in advanced stages, but in minimal and mild endometriosis the usual therapeutic approach is laser vaporization of the lesion or pharmacologic suppression of ovarian activity. Currently, four basic hormonal regimens are available: danazol, progestins, gestrinone, and gonadotropin-releasing hormone (GnRH) agonists. From the Centro per lo Studio e la Terapia dell'Endometriosi, Prima Clinica Ostetrica Ginecologica, Uniuersita di Milano,' the Istituto di Ricerche Farmacologiche "Mario Neg;ri,"b and the Divisione di Ostetricia e Ginecologia, Ospedali Civili Bergamo.' Supported by the Ministero della Sanua (Rome, Italy) Convenzione Mmistero della Sanita-Universita di Milano, "Studio Territoriale Multicentrico della Malattia Endometnosica." Received for publication May 20, 1991; revised October 10, 1991; accepted November 30,1991. Reprint requests: Fabio Parazzini, lstuuio "Mario Negri," via Eritrea, 62, 20157 Milan, Italy.

In spite of the large amount of published clinical data, there is still considerable uncertainty regarding the efficacy of these hormonal treatments for increasing conception rates in early endometriosis." 2 Published series were generally small and subjects had different stages of the disease. Furthermore few randomized trials included untreated or placebo controls. In particular few data are available on the role of GnRH agonists to improve reproductive prognosis in infertile women with minimal or mild pelvic endometriosis. In this paper we report the results of a randomized clinical trial comparing the efficacy of buserelin versus expectant management in the treatment of infertile women with pelvic endometriosis stage I or II of the revised American Fertility Society classification.'

Subjects and methods Eligible patients were women aged :538 years who were trying to conceive and had a laparoscopically confirmed diagnosis of minimal or mild endometriosis

6/1/35389

1345

1346 Fedele et al.

May 1992 Am J Obsret Gynecol

80

60 r--·-~

f~---------··--··,

40

r-------··.r---' ,._.. __ .. _.... -.. -

.--------'

______.. ._.J.---J

20

,,

r-·--.J

r---··-----'

r 2

4

6

8

10

12

14

16

18

20

22

24

months Fig. 1. Overall 24-month probability of becoming pregnant in 71 subjects with minimal or mild endometriosis according to treatment allocation, Milan, Italy, 1988 to 1990. - - , Buserelin-treated group; - - - - " expectant management group. xi adjusted for stage = 0.69. P = 0.41 for 12-month analysis; xl = 0.95, P = 0.33 for 24-month analysis.

(stage I or II of the revised American Fertility Society classification') and otherwise unexplained infertility of ~2 years. Women were eligible if they had normal results on standard medical and gynecologic examination and hysterosalpingogram, luteal-phase endometrial biopsy, hormone profile (two follicle-stimulating hormone, luteinizing hormone, and estradiol assays in the follicular phase and three progesterone and prolactin assays in the luteal phase), and postcoital test. Further, all partners had two semen analyses to exclude severe dyspermia (> 10 x 106 sperm per milliliter, quantity > 1.5 ml, >30% progressive motility at the third hour, > 30% typical forms, absence of agglutination, and leukocyte count <2,000,000 ml). Women with a previous clinical or laparoscopic-laparotornic diagnosis of endometriosis or with any other disease that might affect fertility (for example, uterine malformation or uterine myomas) were specifically excluded from the trial. No women had had previous therapy for endometriosis or infertility. The investigators obtained approval of the protocol from their institutional review board, and each patient gave informed consent. Four eligible women refused to enter the study.

A total of 71 consecutive women observed between February 1988 and June 1989 at the First Department of Obstetrics and Gynecology, University of Milan, and Department of Obstetrics and Gynecology of Ospedali Riuniti, Bergamo, entered the study. The diagnostic laparoscopy was performed under the supervision of L.F. for cases diagnosed in Milan and E.R. for those diagnosed in Bergamo. Patients were assigned by phone according to a computer-generated randomization list to one of the following treatments: intranasal buserelin (Suprefact, Hoechst, Milan, Italy) 400 "",g three times daily for 6 months (35 subjects) and expectant management (36 subjects). If dysmenorrhea or pelvic pain occurred, the patients were asked to use naproxen sodium as firstline treatment. All patients were examined regularly every 4 months. The median follow-up was 17 months for the patients who received buserelin and 18 months for those who did not. At each follow-up visit a standard gynecologic examination was performed and the occurrence of pregnancy recorded. If pregnancy did not occur within 12 months of randomization, cycles were monitored for

Buserelin vs expectant management in treatment of infertility 1347

Volume 166 Number 5

Table I. Distribution of 71 infertile women with pelvic endometriosis according to selected characteristics and treatment allocation, Milan, Italy, 1988 to 1990 Buserelin

No.

I

Expectant management

%

No.

29 57 14

II

I

%

Age (yr)

<30 30-34 >34

10

20 5

xi trend Disease stage 1 II

31 50 19

21 15

58 42

32 4

89

8 28

22 78

0.04 (p = 0.84)

21 14

60 40

xi

History of spontaneous abortion No Yes

18 7

0.20 (p = 0.65)

33 2

94 6

II

0.66 (p = 0.42)

xi

Years of infertility

2 23

10

25

29 71 0.37 (p = 0.54)

xi

two cyles by ultrasonography and hormone measurements. In 29 women the following abnormalities, according to Eissa et al.," were detected: luteinized unruptured follicle cycle (seven cases), luteal phase failure (eight cases), and asynchronous follicle cycle (14 cases). In all these cases clomiphene citrate, 100 mg / day, was given from day 3 to 7 of the cycle, and human chorionic gonadotropin (hCG), 10,000 IV, was administered intramuscularly on the basis of ultrasonographic findings. Induction of ovulation was performed until six cycles. The present analysis is based on information obtained untilJune 1990. The outcome of pregnancy with the last menstrual period before June 1990 was checked and data on deliveries collected until February 1991. Data analysis. We calculated with the actuarial method the cumulative proportion of women who became pregnant, and the curves obtained were compared by the log-rank test." The event data used in the computation of the probability of becoming pregnant were the date of laparoscopy (i.e., of randomization) and the date of the last menstrual period or last followup visit. Tests of statistical significance for the differences in baseline characteristics of the patients and in the frequency of spontaneous abortions on the total of observed pregnancies in relation to treatment allocation were based on the usual X2 value, comparing observed and expected numbers of events with the outcome of pregnancy. To allow for the potential confounding effect of stage and ovulation-induction treatment on the outcome of pregnancy, Mantel-Haenszel procedure was used." When indicated, the significance in the linear trend was assessed with the Mantel test.'

Results Table I shows the distribution of the patients according to age, disease stage, history of spontaneous abortion, years of infertility, and treatment allocation. The distribution for these variables was similar in both groups. All patients who entered the study actively tried to become pregnant during the study period. All the buserelin-treated patients took the drug regularly. Amenorrhea was observed in 24 subjects after 1 month of treatment and in all cases after 2 months of treatment. A total of 15 patients in the buserelin (43%) and 14 in the no-treatment group (39%) received drugs to stimulate ovulation (see Subjects and methods). A total of 17 pregnancies was observed both in the buserelin-treated patients and in the expectant management group (Table II). Of the 34 pregnancies, 13 (seven in the buserelin and six in the no-treatment group) occurred after treatment with clomiphene citrate and hCG. The mean number of cycles in which ovulation was induced in women who conceived after the treatment with clomiphene citrate and hCG was 3.8, with no important difference between the buserelintreated and the expectant management group. The 12-month (i.e., before clomiphene citrate + hCG therapy) and the 24-month actuarial overall pregnancy rates were, respectively, 30% and 61 % in the buserelin group and 37% and 61 % in the expectant management group (xi adjusted for stage = 0.69, P = 0.41 for 12-month analysis; xi = 0.95, P = 0.33 for 24-month analysis; Fig. I). Spontaneous abortion occurred in five (29%) of the 17 pregnancies observed in the buserelin-treated group and in one (6%) of the 17 in the expectant management

1348 Fedele et al. Am

J

May 1992 Obstet Gyne col

Table II. Number of patients at risk and number of observed pregnancies according to month of follow-up and treatment allocation, Milan, Italy, 1988 to 1990 Month of follow-up Trial entry

Buserelin No. of pat ients at risk (entering interval) No. of pregnancies observed in interval Expectant management No. of patients at risk (entering interval) No. of pregnancies observed in interval

35

36

11

35

35

36

34

2

3

group, but this difference was not statistically significant (xi adjusted for disease stage and use of clomiphene citrate and hCG treatment = 3.01; P = 0.08 ).

Comment This study shows that in infertile women with min imal or mild endometriosis the short-term reproductive outcome was not markedly affected by treatment with buserelin. Before discussing these results, potential biases should be carefully considered. Our study compares the reproductive outcome in women treated with buserelin and untreated controls in a nonblinded way. However, this should not represent a problem with regard to reproductive prognosis. Considering the frequency of adverse effects related to GnRH agonist treatment, it is, in our opinion, impossible to conduct a "true" double-blind trial. As indicated in th e data analysis section, we computed reproductive rates with the date of randomization as an event date. Thus when we computed the I-year pregnancy rate patients randomized to the buserelin group had actually tried to conceive for only 6 months (because the y were under treatment for the first 6 months). We chose to present our results in this way because we think that the 6 months of treatment during which the patients could not conceive should be considered in an overall riskbenefit analysis of the effect of the treatment. In any case life-table analysis performed on onl y the months of active attempts at conceiving (i.e., considering as starting date the date of randomization in the no -treatment group and the date of stopping treatment in the buserelin group) did not show any statisticall y significant differences in reproductive outcome at 6 months (data not shown). With regard to other sources of bias, all randomized patients were regularly observed, compliance with the stu d y protocol was generally complete,

35

31

35

31

35

28

35

27

35

26

35

28

6

2

26

26

25

24

25

23

3

and the two groups were comparable in terms of age, reproductive history, and disease stage. Another potential limitation of this study is the small sample size and hence its low statistical power. However, with this sample size we can exclude, at the usual level of a study power (~ = 0.20), a double reproductive rate in the treated versus the not-treated women . According to the study protocol, if a patient had not become pregnant 1 year from randomization, she un derwent ultrasonography and endocrine monitoring of two cycles. When we detected abnormalities of folliculogenesis or ovu lation as in the patterns described by Eissa et al.,' the patients were treated with clomiphene citrate and hCG. This situation must be considered carefully in interpreting the 2-year pregnancy rates and in the generalization of the results of the study. In this regard, the observation that ovulation ind uction improved fertility rates similarly in both the gr ou ps suggests that previous treatment with a GnRH agonist does not affect the efficacy of ovulation induction tr eatment. Of course, this study cannot evaluate the effect of ovulation induction in improving the prognosis of infertile endometriotic women. It is conceivable that endometrosis impairs folliculogenesis , and ovulation induction may overcome this obstacle, increasing the likelihood of a pregnancy. In our treated group the abortion rate was greater than in the women given expectant management. This difference was not statistically significant, and in the absence of any biologic explanation it should be largely attributable to chance. In spite of these potential limitations this is, to our knowledge, the first randomized study reported in the literature that compares GnRH agonist treatment of endometriosis-associated infertility with expectant management. T he fertility rates obtained after buserelin treatment were comparable to those observed in

Volume 166 Number 5

Buserelin vs expectant management in treatment of infertility

1349

Month offollow-up

12

13

14

15

16

17

18

19

20

21

22

23

23

23

23

23

22

21

19

16

13

12

10

9

4

4

2

22

17

14

13

11

11

11

10

5

4

2

the main clinical studies published so far on the use of GnRH agonists in the treatment of endometriosis. In a large randomized multicenter study" comparing danazol, 800 mg/day, and intranasal nafarelin, 400 or 800 fl.g/day , for 6 months in 149 infertile women there were no statistically significant differences among the 12month pregnancy rates in the three groups. The pregnancy rates were, respectively, 36%, 30%, and 52%. A randomized study comparing danazol, 600 mg/day, and intranasal buserelin, 1200 fl.g /day, for 6 months in about 60 patients showed 18-month cumulative pregnancy rates of, respectively, 43 % and 38%." The 12month cumulative pregnancy rates in the randomized study of Dmowski et al." comparing subcutaneous or intranasal buserelin with danazol were 44 % and 62 %, respectively. In an open study performed by Franssen et aLII in which buserelin, 900 fl.g/day, was given for 6 months the cumulative pregnancy rate after 6 months of follow-up was 32 % in 22 previously infertile women . Until now only four randomized clinical trials ":" comparing hormonal treatment with expectant management or placebo have had reproductive outcome as one of the end points. None of these studies reported significant differences between the treatment groups, which received danazol, gestrinone, or medroxyprogesterone acetate or were not treated . In the latter group, the cumulative pregnancy rates at 12 to 30 months were 24% and 57%, respectively. The failure of bu serelin treatment, like that of other hormonal treatments, to improve the infertility associated with minimal and mild endometriosis warrants consideration . First , evidence exists that endometriotic lesions regress after treatment with GnRH agonists," 10 . II . 16 but microscopic ex amination of the apparently cured pelvic areas has demonstrated the persistence of ectopic endometrium" and mitotic activity.!" Further, there is no evidence of the presumed

relation between minimal and mild endometriosis and subfertiliry, 19. 20 In conclusion , these results and those of previous studies provide evidence that trearnent with GnRH agonists is unlikely to have a marked influence on the reproductive outcome in infertile women with minimal or mild endometriosis. REFERENCES 1. Olive DL, Haney AF. Endometriosis-associated infertility:

2.

3. 4.

5.

6. 7. 8.

9.

10.

II.

a critical review of therapeutic approaches . Obstet Gynecol Surv 1986;4 :538-55. Cand iani GB, VerceIlini P, Fedele L, Bocciolone L, Bianchi C. Medical treatment of mild endometriosis in infertile women: analysis of a failure. Hum Reprod 1990;5: 90 1-5. American Fertilit y Society. Revised American Fertilit y Society classification of endometriosis: 1985. Fertil Steril 1985 ;43:351-2. Eissa MK, Sawers RS, Docker MF, Lynch SS, Newton JR. Cha racteristics and incidence of d ysfunctional ovulation patterns detected by ultrasound. Fertil Steril 1987;47:603-12 . Peto R, Pike Me. Armitage P, et al. Design and anal ysis of randomized clinical trials requiring prolonged observation of each patient. II. Anal ysis and examples. Br J Can cer 1977;35 :1-39. Mantel N, Haenszel W. Statistical aspe cts of the anal ysis of d ata from retrospective studies of d isease.J Natl Cancer Inst 1959;22:719-48. Mantel N. Chi-square tests with one d egree of freedom : extension of the Mantel-Haenszel proc edure. J Am Stat Assoc 1963;58:690-700. Henzl MR, Corson SL, Moghissi K, Buttram VA, Berqvist C, Jac obson J. Administration of nasal nafarelin as compared with oral danazol for endometriosis. N Engl J Med 1988;318:485-9. Fedele L, Bianchi S, Arcaini L, Vercellin i P, Candiani GB. Buserelin versus danazol in the tr eatment of endometriosis-associated infertility. AM J OBSTET GY NECOL 1989;161:871-6. Dmowsky WP, Tumrnon I, Radwanska E, Pepping P, Binor Z. Ovarian suppression induced with buserelin or danazol in the management of endometriosis: a randomized , comparative study. Fertil Steril 1989 ;51 :395-400. Franssen AMHW, Zijlstra .lA, Kauer FM, Rolland R,

Fedele et al,

12. 13.

14. 15.

Chadha DR. Endometriosis: treatment with gonadotropin-releasing hormone agonist buserelin. Fertil Steril 1989;51 :401-8. Seibel MM, Berger MJ, Weinstein FG, Taymoz ML. The effectiveness of danazol on subsequent fertility in minimal endometriosis. Fertil Steril 1982;38:534-7. Bayer SR, Seibel MM, Saffan DS, Berger MJ, Taymor ML. Efficacy of danazol treatment for minimal endometriosis in infertile women. A prospective randomized study. J Reprod Med 1988;33:179-83. Thomas EJ, Cooke lD. Successful treatment of asymptomatic endometriosis: does it benefit infertile women? BMJ 1987;294:1117-9. Telimaa S. Danazol and medroxyprogesterone acetate inefficacious in the treatment of infertility in endometriosis. Fertil Steril 1988;50:872-5.

May 1992 Am J Obstet Gynecol

16. Kennedy SH, Barlow DH, Williams lA, Shaw R, Brodribb J. A comparison of nafarelin acetate and danazol in the treatment of endometriosis. Fertil Steril 1990;53:9981003. 17. Schriock E, Monroe SE, Henzl M, Jaffe RB. Treatment of endometriosis with a potent agonist of gonadotropinreleasing hormone (nafarelin). Fertil Steril 1985;44: 583-8. 18. Nisolle-Pochet M, Casanas-Roux F, Donnez J. Histologic study of ovarian endometriosis after hormonal therapy. Fertil Steril 1988;49:423-6. 19. Bancroft K, Vaughan Williams CA, Elstein M. Minimal/mild endometriosis and infertility. A review. Br J Obstet Gynaecol 1989;96:454-60. 20. Metzger DA, Luciano A. Hormonal therapy of endometriosis. Clin Obstet Gynecol North Am 1989; 16: 105-22.

Magnetic cell sorting and the transferrin receptor as potential means of prenatal diagnosis from maternal blood Dorothee Ganshirt·Ahlert, PhD,. Monika Burschyk, MS,. Henk S.P. Garritsen, MD," Lisa Helmer, BS,. Peter Miny, MD,eJiirgen Horst, MD,c Hermann P.G. Schneider, MD,. and Wolfgang Holzgreve, MD" Munster, Germany OBJECTIVE: We wanted to test whether the recently described method of using the transferrin receptor system for fluorescence-activated cell-sorter enrichment of nucleated red blood cells can be used for prenatal diagnosis from maternal blood. STUDY DESIGN: Instead of the laborious, expensive fluorescence-activated cell-sorter system, we used the newly described magnetic-activated cell sorter. RESULTS: An effective enrichment could be achieved with separation of lymphocyte subsets. With the transferrin receptor, however, the enrichment was very inefficient because of the poor specificity of the antibody itself. Even in umbilical cord blood only 25% of nucleated red blood cells were labeled as demonstrated by immunogold silver enhancement of transferrin receptor-labeled cells. CONCLUSION: In spite of the availability of a fast and effective separation method (magnetic-activated cell sorter) the use of the transferrin receptor antigen alone is not likely to enable a reliable identification of fetal cells in maternal circulation. (AM J OSSTET GVNECOL 1992;166:1350-5.)

Key words: Transferrin receptor, fluorescence-activated cell sorter, magnetic-activated cell sorter, polymerase chain reaction The occurrence of fetal nucleated cells within maternal circulation has raised hopes for a noninvasive method of prenatal diagnosis.!" By application of the polymerase chain reaction, a powerful technique for From the Departments of Obstetrics and Gynecology: Haematology,! and Human Genetics,' University of Micnster. Received for publication July 19,1991,. revised November 18,1991,. accepted December 30, 1991. Reprint requests: Wolfgang Holzgreve, MD, Zentrum fur Frauenheilkunde, Albert Schweitzer Strasse 33, D-4400 Munster, Germany.

6/1 /35996

1350

amplification of specific deoxyribonucleic acid sequences from samples as small as one cell, this goal appears to be accessible. Attempts to amplify Y-specific sequences from the deoxyribonucleic acid of pregnant women have shown that the number of fetal cells in maternal circulation is extremely small, ranging within the lowest detection limit even of the polymerase chain reaction.' Yet the risk of contamination-a well-known limitation of the polymerase chain reaction techniqueis most prevalent at the highest level of sensitivity and hence often leads to false-positive results!' 5