Endometriosis. Is IVF the Answer for Infertility? 
Stage-related Success Rates in 1417 Consecutive IVF-ET Cycles

by S.D. Spandorfer and Z. Rosenwaks
The Center for Reproductive Medicine and Infertility, Weill Medical College of Cornell University, New York, New York, U.S.A.

Introduction
Endometriosis is a very common disease affecting up to 10% of reproductive age women (Wheeler, 1992). A higher prevalence of endometriosis is reported in infertile women, with estimates ranging from 25 - 48% (Olive and Schwartz, 1993; Strathy et al., 1982). Endometriosis, when refractory to medical or surgical therapies, is a frequent indication for in vitro fertilization and embryo transfer (IVF-ET). 

The effect of endometriosis on pregnancy outcome after IVF-ET is controversial. Many authors have reported a detrimental effect (Wardle et al., 1985; Matson and Yovich, 1986; Mills et al., 1992; Pellicer et al., 1995; Arici et al., 1996), but not all have concurred with this finding (French National IVF Registry, 1993; Dmowski et al., 1995; Olivennes et al., 1995). Most of these studies have been relatively small or multicenter in design, and have not been able to specifically assess the impact of endometriosis stage on IVF-ET outcome. 

Our purpose is to report the experience at The New York Presbyterian Hospital/ Weill Medical of Cornell University in 1417 consecutive treatment cycles of IVF-ET in women with a history of endometriosis, specifically analyzing the effect of the stage of endometriosis on pregnancy outcome. We also attempted to assess whether the presence of an endometrioma at the initiation of stimulation affected IVF outcome.

Materials and Methods
All patients with the diagnosis of endometriosis that underwent IVF-ET at The New York Presbyterian Hospital-Weill Medical College of Cornell University from January 1989 to December 1997, were evaluated in this retrospective study.

The diagnosis and stage of endometriosis was determined by laparoscopic evaluation prior to the IVF cycle. The stage of endometriosis was based on the ASRM classification system (The American Society of Reproductive Medicine, 1997). All medical records of women with endometriosis were reviewed to verify the diagnosis and stage of endometriosis as described in the operative report.

Patients were treated with standard ovulation induction protocols and underwent IVF-ET according to previously published guidelines (Davis and Rosenwaks, 1996). In brief, women treated with luteal phase leuprolide acetate were desensitized with this gonadotropin-releasing hormone (GnRH-a; Lupron; Tap pharmaceuticals, Deerfield, IL, USA), 1 mg s.c. daily until ovarian suppression was achieved. Women not treated with luteal leuprolide acetate began stimulation on day 2 of their treatment cycle. Ovarian stimulation was then carried out using a combination of gonadotropins [human menopausal gonadotrophin (hMG); pure follicle stimulating hormone (FSH), following our standard step-down protocol (Davis and Rosenwaks, 1996). Human chorionic gonadotropin (hCG) was administered (3300-10000 IU) when a minimum of two follicles reached at least 16-18 mm in mean diameter as measured by transvaginal ultrasound. Oocytes were harvested by transvaginal ultrasound-guided puncture 35 hours after hCG administration, under intravenous sedation.

Oocyte insemination or the use of micromanipulation were applied on a case by case basis. Morphologically normal embryos were transferred back into the uterine cavity approximately 72 hours after the retrieval. The number of embryos transferred was dependent on maternal age according to our standard protocol (Davis and Rosenwaks, 1996). In general, patients under 35 years of age had 3 embryos transferred, patients 35 to 40 years of age had 4 embryos transferred, and patients over 40 had up to 5 embryos transferred. Starting on the day of oocyte retrieval, methylprednisolone (16 mg/day) and tetracycline (250 mg every 6 hours) were administered for 4 days to all patients. Progesterone was administered on day 3 after hCG administration (25-50 mg i.m./day) and was continued until the assessment of pregnancy (Davis and Rosenwaks, 1996).
Parameters analyzed included demographics, stimulation protocols, peak estradiol levels, Number of oocytes retrieved and fertilized, number of embryos transferred and pregnancy outcomes.

All pregnancy outcomes were evaluated on a per transfer basis. A clinical pregnancy was defined as the detection of the presence of fetal heart motion by ultrasound. An ongoing pregnancy was defined as a pregnancy continuing beyond the 20th week of gestation. Implantation rates refer to the number of sonographically identified embryos with positive fetal heart movements per number of embryos transferred. 
Statistical analysis was performed by analysis of variance (ANOVA), Chi-square test, and Student's t-test when appropriate. P values < 0.05 were considered significant.

Results
During the 8-year study period, a total of 1417 stimulation cycles for IVF-ET were initiated. Of these, 1196 (84.4%) underwent retrieval and 1105 (92.3% of those that underwent retrieval) had an embryo transfer. Overall, the clinical pregnancy rate/transfer was 44.7% (495/1105) and the ongoing pregnancy rate/transfer was 37.1% (410/1105).

Outcomes by stage of disease are presented in Table 1. Patients with a lower stage of disease were older, although by less than a year. Otherwise, there were no differences in the cycle characteristics of these patients. Furthermore, there were no differences in pregnancy outcomes by stage of disease.

In order to analyze the impact of endometriosis stage on prognosis, we combined patients with milder disease (stages I and II; n = 803) and compared this group to patients with more severe disease (stages III and IV; n = 302). The patients with milder disease were older (36.1 vs. 35.5; p < 0.05), but exhibited higher peak estradiol levels (1311 vs. 1194; p < 0.05) and a greater number of mature oocytes recovered (10.1 vs. 9.1; p < 0.05). There was no difference in the number of embryos transferred (3.3 vs. 3.3; p = NS). Despite the higher response to stimulation, there was no difference in either clinical or ongoing pregnancy rates between the two groups -- clinical pregnancies/transfer of 44.4% vs. 45.6% and ongoing pregnancies/transfer of 35.8% vs. 40.3%, respectively.

In order to reduce the potential bias of patients undergoing multiple attempts of IVF-ET, we also separately analyzed the first attempt for each patient (Table 2). No differences were noted in stimulation or pregnancy outcome. However, a comparison of patients with milder disease (stages I and II; 512 cycles at transfer) to patients with more severe disease (stages III and IV; 177 cycles at transfer), reveal that mild disease was associated with higher peak estradiol levels (1389 vs. 1253, P < 0.05) and recovery of more mature oocytes per retrieval (10.8 vs. 9.6; P < 0.05). Yet, the number of embryos transferred and pregnancy outcomes were not different (Table 3).

It is well known that patients with endometriosis benefit from down-regulation with GnRH agonists before gonadotropin stimulation and that they represent a group with a more favorable prognosis. Thus, we endeavored to separately analyze the effect of endometriosis stage on this favorable prognosis group (n = 892; Table 4). Patients with milder disease had higher estradiol levels and obtained more mature oocytes, but no differences were detected in pregnancy outcomes.

The impact of endometriomas -- found at time of cycle initiation -- on IVF outcome was also analyzed. This study was limited to patients with moderate or severe endometriosis. Twelve cycles were not included in this analysis because it was unclear if an endometrioma was present at the beginning of the stimulation cycle. A total of 286 cycles with Stage III or IV endometriosis were included in this analysis, of which 89 (31.1%) were noted to have an endometrioma. Stimulation cycle characteristics and IVF outcome for patients with or without endometriomas are depicted in Table 5. There were no differences in patients' age, peak estradiol levels, oocytes obtained at retrieval, fertilized oocytes, or number of embryos transferred. A non-significant trend towards a decreased clinical pregnancy rate was found in patients with endometriomas.

Discussion
In this large analysis, we have demonstrated that the stage of endometriosis in patients undergoing IVF-ET does not appear to effect pregnancy outcome. Patients with mild disease had similar pregnancy outcomes to patients with more severe endometriosis. Interestingly, patients with more severe endometriosis (stages III or IV), despite being younger, exhibited a poorer ovarian response to stimulation (lower peak estradiol levels and fewer oocytes retrieved) than patients with milder disease. This remained true after controlling for stimulation type and cycle attempts. Two potential explanations for this finding is offered, both based on the fact that patients with more severe disease usually have more ovarian involvement. First, ovarian endometriosis may affect follicular recruitment. Perhaps elaboration of cytokines specific to endometriosis affects the ovarian response to medications. Secondly, patients with more severe endometriosis have often undergone multiple surgical procedures involving endometriomas. Invariably, endometrioma resection may compromise or destroy adjacent normal ovarian tissue, resulting in subsequent diminished ovarian reserve and reduced oocyte harvests. This important observation suggests that even with severe endometriosis surgical treatment should be as conservative as possible. 

Furthermore, we have demonstrated for the first time, that in patients with severe endometriosis, the presence of an endometrioma at the inception of an IVF-ET cycle has no adverse affect on stimulation, but is associated with a trend towards a decreased pregnancy rate. This merits further investigation, as this would possibly warrant endometrioma removal prior to commencing an IVF cycle.

The retrospective nature of the study, may limit some of our conclusions. For example, our reliance on operative reports to stage endometriosis could be suboptimal: as the study spanned a period of several years, improvements in IVF success rates and an uneven distribution of stage-specific patients year-by-year could also affect our interpretation. However, the analysis of the data presented indicates that there was a relatively equal patient distribution over time, particularly as it relates to specific disease stage.

In summary, we have demonstrated in this large study that the stage of endometriosis may have a slight effect on ovarian stimulation response (milder stage disease in good prognosis patients had better responses), but, more importantly, no differences in pregnancy outcomes were detected. However, patients with endometriomas may indeed exhibit diminished pregnancy rates, thus, removing endometriomas prior to an IVF cycle should be given consideration.

References

1. 
   Arici A, Oral E, Bukulmez O, Duleba A, Olive D, Jones EE. (1996) The effect of endometriosis on implantation: results from the Yale University in vitro fertilization and embryo transfer program. Fertil Steril 65, 603-7.
2. Davis OK, Rosenwaks Z. In Vitro Fertilization. In: Reproductive Endocrinology, Surgery, and Technology. Eli Adashi, John A Rock, Z Rosenwaks (editors), Philadelphia, Lippincott-Raven, 1996, 2319-34.
3. Dmowski WP, Rana N, Michalowska J, Friberg J, Papierniek C, El-Roeiy A. (1995) The effect of endometriosis, its stage and activity, and of autoantibodies on in vitro fertilization and embryo transfer success rates. Fertil Steril 63, 555-62.
4. FIVNAT (1993) French National IVF Registry: Analysis of 1986 to 1990 data. Fertil Steril 59, 587-95.
   Matson PL and Yovich JL. (1986) The treatment of infertility associated with endometriosis by in vitro fertilization. Fertil Steril 46, 432-4.
5. Mills MS, Eddowes HA, Cahill Dj, et al. (1992) A prospective controlled study of in vitro fertilization, gamete intrafallopian transfer and intrauterine insemination combined with superovulation. Hum Reprod 7, 490-4.
6. Olive DL and Schwartz LB. (1993) Endometriosis. New Engl J Med 328, 1759.
   Olivennes F, Feldberg D, Liu H-C, Cohen J, Moy F, Rosenwaks Z. (1995) Endometriosis: a stage by stage analysis - the role of in vitro fertilization. Fertil Steril 64, 392-8.
7. Pellicer A, Oliveira N, Ruiz A, Remohi J, Simon C. (1995) Exploring the mechanisms o endometriosis-related infertility: an analysis of embryo development and implantation in assisted reproduction. Hum Reprod 10, 91-97.
8. Strathy JH, Molgaard CA, Coulam CB, Melton LJ III. (1982) Endometriosis and infertility: a laparoscopic study of endometriosis among fertile and infertile women. Fertil Steril 38, 667-72.
9. The American Society of Reproductive Medicine (1997) The American Society of Reproductive Medicine Classification of Endometriosis. Fertil Steril 67, 819-21.
10. Wardle PG, McLaughlin EA, McDermott A, Mitchell JD, Ray BD, Hull MGR. (1985) Endometriosis and ovulatory disorder: reduced fertilization in vitro compared with tubal and unexplained infertility. Lancet 2, 236-9.
11. Wheeler, JM (1992) Epidemiology and prevalence of endometriosis. Infertil Reprod Med Clin North Am 3, 345-9.