FERTILITY AND STERILITY VOL. 80, NO.

5, NOVEMBER 2003
Copyright 2003 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A.

IN VITRO FERTILIZATION

A double-blind, randomized, placebocontrolled study to assess the efcacy of ketoconazole for reducing the risk of ovarian hyperstimulation syndrome
Mohammad Ebrahim Parsanezhad, M.D.,a Saeed Alborzi, M.D.,a Mahnaz Pakniat, M.D.,a and Ernst Heinrich Schmidt, M.D.b
Shiraz University of Medical Sciences, Shiraz, Iran, and Evang. Diakonie Teaching Hospital of Go ttingen University, Bremen, Germany

Objective: To evaluate the role of ketoconazole in prevention of ovarian hyperstimulation syndrome (OHSS) in women with the polycystic ovary syndrome (PCOS) undergoing ovarian stimulation with gonadotropins. Design: Prospective, randomized, double-blind, placebo-controlled study. Setting: University hospitals. Patient(s): One hundred nine women with PCOS who were referred for treatment with gonadotropins. Intervention(s): Fifty patients were randomly assigned to receive two ampoules of hMG beginning on day 2 or 3 of the cycle and ketoconazole (50 mg every 48 hours) starting on the rst day of hMG treatment. Fifty-one patients received the same amount of hMG plus one tablet of placebo every 48 hours. Main Outcome Measure(s): Follicular development, E2 level, and pregnancy rate. Result(s): The total number of hMG ampoules and duration of treatment to attain ovarian stimulation were higher among ketoconazole recipients. The serum E2 level and number of patients with dominant follicles on day 9 of the cycle were greater in placebo recipients. Serum E2 level and total number of follicles at the time of hCG administration did not differ between the two groups. The cancellation rate and OHSS rate were similar in the two groups. Conclusion(s): Ketoconazole does not prevent OHSS in patients with PCOS who are undergoing ovarian stimulation. It may reduce the rate of folliculogenesis and steroidogenesis. (Fertil Steril 2003;80: 11515. 2003 by American Society for Reproductive Medicine.) Key Words: Polycystic ovary, gonadotropins, hyperstimulation, ketoconazole

Received December 5, 2002; revised and accepted April 2, 2003. Reprint requests: Mohammad Ebrahim Parsanezhad, M.D., P.O. Box 71345-1657, Shiraz, Iran (FAX: 98-711229-6486; E-mail: parsame@sums.ac.ir). a Department of Obstetrics and Gynecology, School of Medicine, Shiraz University of Medical Sciences. b Department of Obstetrics and Gynecology, Evang. Diakonie Teaching Hospital of Go ttingen University.
0015-0282/03/$30.00 doi:10.1016/S0015-0282(03) 01177-4

Ovarian hyperstimulation syndrome (OHSS) is a relatively common and potentially life-threatening complication of ovarian stimulation with gonadotropins (1). In its severe form, the syndrome is characterized by extreme ovarian multifollicular enlargement, ascites, hydrothorax, hypovolemia, and hemoconcentration (2). The underlying mechanisms leading to OHSS remain to be elucidated, but excessive quantities of peptides regulating the growth and permeability of blood vessels are likely to be involved (3). Some authors have suggested that augmentation of steroids and peptide production from the multifollicular hyperstimulated ovaries may be responsible for several features of the OHSS (4).

Vascular endothelial growth factor has been shown to exert a twofold effect: It serves as a potent promoter of neovasculogenesis and increases vascular permeability (57). Early identication of patients at risk and prevention of OHSS is clinically important. Several methods have been recommended for minimizing the probability of OHSS in highrisk patients, including canceling the cycle before administration of hCG or reducing the ovulatory hCG dose (8), withholding luteal support with hCG (9), cryopreserving all embryos for further use in a nonstimulated cycle (10), repeating aspiration of ovarian follicles and early corpus luteum cysts (11), administering human albumin at the time of hCG injec1151

tion (12), using GnRH analogue to trigger ovulation (13), and performing vaginal aspiration of ascites (14). Ketoconazole interferes with cytochrome P450 enzyme systems in testis, ovary, adrenal gland, and liver. Steroidogenesis is inhibited by its action on the C17-20 lyase, the cholesterol side-chain cleavage enzyme, and the 17-hydroxylase. In gonads, it inhibits aromatase, and adrenocortical steroid biosynthesis is inhibited at the 11-hydroxylation and 18-hydroxylation steps. The antiandrogenic effect of ketoconazole may be useful in the management of metastatic prostate carcinoma and hirsutism. Its anticortisolic effect may be useful in the treatment of most patients with Cushings syndrome (1517). Ketoconazole, a broad-spectrum imidazole antimycotic agent, was recently reported to reduce the incidence of OHSS in women with PCOS undergoing superovulation with gonadotropins (18). Because this study was not placebo controlled, however, a placebo effect could not be ruled out. We performed a double-blind, placebo-controlled study to assess the role of ketoconazole in prevention of multifollicular development, excess ovarian steroidogenesis, and moderate and severe OHSS in patients with PCOS undergoing superovulation with gonadotropins.

their allocated intervention from the rst menstrual cycle after randomization. Serum concentrations of liver aminotransferase and bilirubin were measured by using kinetic and colour anylin methods, respectively. E2 was assayed by using the Coat-aCount recombinant immunoassay (Diagnostics Products Co., Los Angeles, CA) 2 to 3 days after the start of menstrual bleeding or progesterone-induced bleeding between days 1 and 3 of menstrual cycle. During this time, transvaginal ultrasonography was performed by using a 5-MHz transvaginal transducer (Medison 600, Korea). All patients had sonographic characteristics of PCOS. Liver aminotransferase and bilirubin levels were normal in all patients. The serum E2 level was normal for the early follicular phase. Before receiving the allocated intervention and undergoing work-up, all patients were randomized by using a random table. A pharmacist who did not take part in this study administered medication and placebo. Neither the patients nor the physician and laboratory staff knew the treatment protocol. The ketoconazole administration protocol was similar to that used as alternate protocol by Gal et al. (18). Fifty patients received two ampoules of hMG (75 IU of FSH and 75 IU of LH per ampoule) (Organon, Oss, The Netherlands) i.m. beginning on day 2 or 3 of the cycle and the minimal dose of ketoconazole (50 mg) starting on the rst day of hMG treatment. Ketoconazole was given at a dose of 50 mg every 48 hours, and its administration was limited up to the last day of hMG stimulation. Fifty-one patients received the same hMG protocol but received one placebo tablet every 48 hours. Ovarian stimulation was monitored by ultrasonographic assessment of total number of growing follicles, mean diameter of the follicles, and the full endometrial thickness and measurement of E2 every 2 days. Ovulation was triggered by i.m. injection of 10,000 IU of hCG (Organon) when the leading follicle reached a diameter greater than 16 mm and the E2 concentration was greater than 300 pg/mL for each main follicle (16 mm in diameter). Liver aminotransferase and bilirubin measurement was repeated on the last day of ketoconazole and hMG administration. Ovarian hyperstimulation syndrome was predicted by using the criteria described by Rabe et al. (19). The syndrome was graded by using established criteria (20). If a patient was at risk of OHSS, the hCG dose was reduced to 5,000 IU or her cycle was cancelled (i.e., administration of hCG was withheld). Patients were monitored during the luteal phase 6 2 days and 12 1 days after injection of hCG. They were asked about symptoms of OHSS (abdominal pain, nausea, vomiting, diarrhea, and weight gain).
Vol. 80, No. 5, November 2003

MATERIALS AND METHODS

The study was performed at Shiraz University of Medical Sciences, Shiraz, Iran. From September 2000 to November 2002, we enrolled 637 women with PCOS from our infertility division into this prospective, randomized, doubleblind, placebo-controlled study. Patients were interviewed, the study protocol was described to them, and patients charts were carefully reviewed. Patients had undergone a complete infertility evaluation that included hormonal assay, hysterosalpingography, postcoital testing, semen analysis, and endometrial biopsy. Inclusion criteria were infertility, an elevated serum LH level, a normal or elevated FSH level, an LH-to-FSH ratio greater than 2, elevated testosterone and DHEAS levels, oligomenorrhea or amenorrhea, and at least 10 follicles smaller than 8 mm in diameter in the subcapsular region around the hyperechogenic central stroma on ultrasonography. Except for PCOS-related anovulatory infertility, women with other infertility factors were excluded. A total of 528 patients were excluded (217 women did not fulll the inclusion criteria and 311 women declined to participate). Thus, 109 women were randomly allocated after giving informed consent. The ethics review committee for human research of the university approved the study. Women had been unsuccessfully treated with clomiphene citrate at daily doses of up to 200 mg for 5 days. All were being considered for treatment with hMG. They received 1152 Parsanezhad et al.
Ketoconazole in prevention of OHSS

TABLE 1 Effects of ketoconazole and placebo in patients undergoing hMG superovulation.

Variable No. of cycles Duration of hMG therapy Total no. of hMG ampoules/patient E2 level on day 9 of the cycle No. of patients with dominant follicles on day 9 (%) E2 level before hCG injection (pg/ml) Endometrial thickness before hCG administration (mm) No. of patients with E2 level 1,500 pg/mL at hCG injection (%) No. of patients who received 5,000 IU of hCGa No. of patients with 13 lead follicles (%) No. of successful stimulation cycles (%)
a

Ketoconazole group 50 13.5 0.98 19.44 1.5 465.27 134.24 13 (25.6%) 1349.58 381.71 10.7 1.45 36 (72.1%) 5 (10%) 43 (86%) 36 (72.1%)

Placebo group 51 9.6 1.9 15.18 1.9 1001.54 552.28 30 (59.1%) 1288.65 473.64 10.4 1.4 38 (75%) 7 (13%) 45 (88.6%) 38 (75%)

P value .0001 .0001 .0001 .0001 .05 .05 .05 .05 .05 .05

Because of the risk of ovarian hyperstimulation syndrome.

Parsanezhad. Ketoconazole in prevention of OHSS. Fertil Steril 2003.

Transvaginal ultrasonography was performed to measure peritoneal uid, ovarian size, total number of cysts, and maximum cyst diameter. Blood samples were obtained for measurement of E2, -hCG, plasma proteins, and electrolytes and for a blood count. If any evidence of moderate OHSS was present, the patient was hospitalized for further management. Clinical and laboratory variables were compared between the treatment and placebo groups. Statistical analysis was performed by using the Student t-test and 2 test. Correlations were calculated at a 95% level of condence. P.05 was considered signicant.

(1001.54 552.28 pg/mL vs. 465.27 134.24 pg/mL among ketoconazole recipients; P.0001). The number of hMG ampoules that was needed to attain superovulation was signicantly higher among ketoconazole recipients (P.0001) (Table 1). Serum E2 level, endometrial thickness at the time of hCG administration, and number of patients who received 5,000 IU of hCG were similar in the two groups (Table 1). Table 1 shows the number of patients with an E2 value less than the critical level of 1,500 pg/mL and the number of lead follicles at the time of hCG administration. The E2 level and number of lead follicles were similar in the two groups. The number of pregnancies, cases of OHSS, and number of cancelled cycles did not differ between the ketoconazole and placebo group (Table 2).

RESULTS
One hundred one women completed the study protocol. Three ketoconazole recipients and ve placebo recipients were lost to follow-up. Ketoconazole and placebo recipients did not statistically differ in mean (SD) age (26.3 3.1 years vs. 28.33 4.0), body mass index (27.12 2.1 kg/m2 vs. 28.62 4.8 kg/m2), duration of infertility (3.7 1.8 years vs. 3.8 1.4 years), and percentage of participants with primary infertility (81.4% vs. 75%). Ketoconazole had no side effects with the treatment protocol. The duration of treatment with hMG was signicantly longer in ketoconazole recipients than placebo recipients (P.0001). Although the total numbers of follicles on day 9 of the cycle were similar in ketoconazole and placebo recipients (11.13 3.37 and 10.6 3.63, respectively), the number of the patients with dominant follicles (mean diameter 14 mm) was signicantly higher among placebo recipients (P.0001). Serum E2 concentrations on day 9 of the cycle were signicantly higher among placebo recipients
FERTILITY & STERILITY

DISCUSSION
Gal et al. (18) rst attempted use of ketoconazole to treat OHSS. They administered 50 mg of ketoconazole every 24 to 48 hours beginning on the rst day of hMG administration

TABLE 2 Outcome of treatment in patients receiving ketoconazole or placebo.

Variable No. No. No. No. of of of of pregnancies (%) multifetal pregnancies cancelled cycles (%) patients with OHSS (%) Ketoconazole group 9 (18) 0 14 (27.9) 4 (7) Placebo group 11 (21.5) 0 16 (31) 5 (9) P value .05 .05 .05

Parsanezhad. Ketoconazole in prevention of OHSS. Fertil Steril 2003.

1153

and reported a signicant reduction in OHSS after treatment with ketoconazole. The regimen that we used does not differ from that used as alternate protocol by Gal et al. (18) in terms of drug, dosage, and follow-up. We found a slow rate of E2 production during the follicular phase (on day 9 of the cycle) in the ketoconazole group compared with placebo recipients. Similar results were reported when a different dose of ketoconazole was given (21). On the 9th day of the cycle, equal numbers of follicles were growing in the two groups. However, signicantly more placebo recipients had dominant follicles. This nding suggests that the drug did not affect the total number of growing follicles, but the rate of follicular growth was reduced. Ketoconazole was expected to prevent multiple follicular growth and excess steroidogenesis despite continuing gonadotropin administration to attain acceptable numbers of dominant follicles, an appropriate serum E2 level, and a successful stimulation cycle. When administration of gonadotropins was continued to reach an acceptable number of dominant follicles to trigger ovulation by hCG, some patients developed hyperstimulated ovaries. Zelinski-Wooten et al. (22) and Moudgal et al. (23) also reported a similar nding. They discovered that trilostane, a 3-hydroxysteroid dehydrogenase inhibitor, and fadrozole, a nonsteroidal aromatase inhibitor, administered throughout the follicular phase to rhesus monkeys undergoing ovarian stimulation with human gonadotropins had no effect on the total number of antral follicles and their distribution, the number of retrieved oocytes, and ovulation, despite a profound reduction in E2 levels. In contrast, Gal et al. (18) found a substantial reduction in the number of mediumsized follicles and reported a reduction in cancellation rate from 37% to 7%. In our study, the incidence of OHSS in both groups was similar to that reported by Gal et al. (18) in their untreated cycles. We also found no signicant difference in the rates of cycle cancellation and hyperstimulated cycles between ketoconazole and placebo recipients. The numbers of successful stimulation cycles were also similar in the two groups. Our results suggest that during ovulation induction with gonadotropins, ketoconazole may decrease the rate of folliculogenesis and steroidogenesis. The total number of follicles, including small, mediumsized, and leading follicles, and serum E2 levels were similar in the ketoconazole and placebo groups at the time of hCG administration. These ndings may explain why ketoconazole does not prevent OHSS. Gal et al. (18) reported that low-dose ketoconazole did not affect progesterone production. Because progesterone does not play a critical role in our results, we did not consider these values in our analysis. 1154 Parsanezhad et al.
Ketoconazole in prevention of OHSS

Zelinski-Wooten et al. (22) reported a reduced fertilization rate among trilostane-treated monkeys. Similar results were reported by Moudgal et al. (23) when aromatase inhibitor was administered. Although we did not evaluate the fertilization process, the similar pregnancy rate per ovulatory cycle in our two groups may support the idea that low-dose ketoconazole does not adversely affect fertilization. In conclusion, low-dose ketoconazole during stimulation cycles with gonadotropins may reduce the rate of folliculogenesis and steroidogenesis. Ketoconazole has no benecial effect on the development of multiple follicles, nal serum levels of E2 at the time of HCG administration, OHSS development, and cycle cancellation. Further study to evaluate the effects of various protocols of ketoconazole on vasoactive agents affecting OHSS may be helpful.

References
1. Schenker JG, Ezra Y. Complications of assisted reproductive techniques. Fertil Steril 1994;61:41122. 2. Aboulghar MA, Mansour RT, Serour GI, Elhelw BA, Shaarawy M. Elevated concentrations of angiogenin in serum and ascitic uid from patients with severe ovarian hyperstimulation syndrome. Hum Reprod 1998;13:2068 71. 3. Pellicer A, Albert C, Mercader A, Musoles F, Remohi J, Simon C. The pathogenesis of ovarian hyperstimulation syndrome: in vivo studies investigating the role of interleukin-1, interleukin-6, and vascular endothelial growth factor. Fertil Steril 1999;71:4829. 4. Neulen J, Yan Z, Raczek S, Wiendel K, Keck C, Weich HA, et al. Human chorionic gonadotropin dependent expression of vascular endothelial growth factor/vascular permeability factor in human granulosa cells: importance in ovarian hyperstimulation syndrome. J Clin Endocrinol Metab;80:196771. 5. Yan Z, Weich HA, Bernart W, Breckwoldt M, Neulen J. Vascular endothelial growth factor (VEGF) messenger ribonucleic acid (mRNA) expression in luteinized human granulose cells in vitro. J Clin Endocrinol Metab 1993;77:17235. 6. McLure N, Healy DL, Rogers PA, Sullivan J, Beaton L, Haning RV, et al. Vascular endothelial growth factor as capillary permeability agent in ovarian hyperstimulation syndrome. Lancet 1994;344:2356. 7. Shweieki D, Itin A, Neufeld G, Gitay-Goren H, Keshet E. Patterns of expression of vascular endothelial growth factor (VEGF) and VEGF receptors in mice suggest a role in hormonally regulated angiogenesis. J Clin Invest 1993;91:223543. 8. Sher G, Salem R, Feinman M, Dodge S, Zouves C, Knutzen V. Elimination the risk of life-endangering complications following overstimulation with menotropin fertility agents; a report on women undergoing in vitro fertilization and embryo transfer. Obstet Gynecol 1993; 81:1009 11. 9. Asch RH, Li HP, Balmaceda JP, Weckstein LN, Stone SC. Severe ovarian hyperstimulation syndrome in assisted reproductive technology; denition of high risk groups. Hum Reprod 1991;6:13959. 10. Amso N, Ahuja K, Morris N, Shaw RW. The management of predicted ovarian hyperstimulation involving gonadotropin releasing hormone analog with elective cryopreservation of all pre-embryos. Fertil Steril 1990;53:108790. 11. Amit A, Yaron Y, Yovel I, Peyser MR, David MP, Botchan A, et al. Repeated aspiration of ovarian follicles and early corpus luteum cysts in an in-vitro fertilization program reduced the risk of hyperstimulation syndrome in high responders. Hum Reprod 1993;8:1184 6. 12. Asch RH, Ivery G, Goldsman M, Fredrick JL, Stone SC, Balmaceda JP. The use of intravenous albumin in patients at high risk for severe ovarian hyperstimulation syndrome. Hum Reprod 1993;8:101520. 13. Ron ER. Complications of ovulation induction. Baillieres Clin Obstet Gynecol 1993;7:43553. 14. Aboulghar MA, Mansour RT, Serour GI, Amin YM. Ultrasonically guided vaginal aspiration of ascites in the treatment of ovarian hyperstimulation syndrome. Fertil Steril 1990;53:9335. 15. Sonino N. The use of ketoconazole as an inhibitor of steroid production. N Engl J Med 1987;317:8128.

Vol. 80, No. 5, November 2003

16. Pepper G, Brenner SH, Gabrilove JL. Ketoconazole use in the treatment of ovarian hyperandrogenism. Fertil Steril 1990;54:438 44. 17. Isik AZ, Gokmen O, Zeyneloglu HB, Senoz S, Zorlu CG. Low dose ketoconazole is an effective and a relatively safe alternative in the treatment of hirsutism. Aust N Z J Obstet Gynaecol 1996;36:4879. 18. Gal M, Edlar-Geva T, Margalioth EJ, Barr I, Oryl J, Diamanti Z. Attenuation of ovarian response by low-dose ketoconazole during superovulation in patients with polycystic ovary syndrome. Fertil Steril 1999;27:26 31. 19. Insler V, Lunenfeld E, Lunenfeld B. Ovarian hyperstimulation syndrome. In: Rabe T, Diedrich K, Runnebaum B (eds). Manual on assisted reproduction. Berlin: Springer, 1997:154 69. 20. Lunenfeld B, Insler V, Rabau E. Induction de lovulation par les

gonadotrophines. In: Oricard R, Ferrin J (eds). Lovulation. Paris: Masson, 1969:27. 21. Gal M, Orly J, Barr I, Algur N, Boldes R, Diamant YZ. Low dose ketoconazole attenuates serum androgen levels in patients with polycystic ovary syndrome and inhibits ovarian steroidogenesis in vitro. Fertil Steril 1994;61:82332. 22. Zelinski-Wooten MB, Hess DL, Wolf DP, Stouffer RL. Steroid reduction during ovarian stimulation impairs oocyte fertilization, but not folliculogenesis, in rhesus monkeys. Fertil Steril 1994;61:114755. 23. Moudgal NR, Shetty G, Selvaraj N, Bhatnagar AS. Use of a specic inhibitor for determining whether there is a role for estrogen in follicle/ oocyte maturation, ovulation and preimplantation embryo development. J Reprod Fertil Suppl 1996;50:69 81.

FERTILITY & STERILITY

1155