In vitro fertilization outcome relative to embryo transfer difficulty: a novel approach to the forbidding cervix

FERTILITY AND STERILITY VOL. 72, NO. 2, AUGUST 1999 Copyright 1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. In vitro fertilization outcome relative to embryo transfer difficulty: a novel approach to the forbidding cervix Nicole Noyes, M.D., Frederick Licciardi, M.D., James Grifo, M.D., Ph.D., Lewis Krey, Ph.D., and Alan Berkeley, M.D. New York University Program for In Vitro Fertilization, Infertility, and Reproductive Surgery, New York University Medical Center, New York, New York Objective: To assess the impact of ET difficulty on IVF outcome and to optimize the ET procedure. Design: Retrospective analysis of IVF outcome by ET catheter type and ET difficulty. Prospective treatment and follow-up of patients with a history of extremely difficult cervical passage. Setting: Large university-based IVF program. Patient(s): All patients 40 years of age undergoing IVF-ET from September 1995 to May 1998. Intervention(s): Surgical correction of cervical stenosis. Main Outcome Measure(s): Pregnancy and embryo implantation rates. Result(s): Only 0.6% of ETs were extremely difficult. Pregnancy rates were not statistically significantly different among ETs graded easy, moderate, and difficult. In contrast, no pregnancies occurred in the rare extremely difficult ET group. Eight patients with a history of extremely difficult cervical passage underwent surgical correction of their cervical stenosis. Twelve postoperative IVF-ET in these women resulted in eight clinical pregnancies, six of which were multiple gestations. The embryo implantation rate of these cycles was 42.2%. Conclusion(s): Patients with a history of extremely difficult ET may benefit from hysteroscopic evaluation and possible modification of their cervical canal before a future IVF attempt. (Fertil Steril 1999;72:261 5. 1999 by American Society for Reproductive Medicine.) Key Words: IVF, embryo transfer difficulty, hysteroscopy, pregnancy Received November 12, 1998; revised and accepted March 22, 1999. Reprint requests: Nicole Noyes, M.D., New York University Medical Center, 660 First Ave, 5th Floor, New York, New York 10016 (FAX: 212-263-7853; E-mail: Nnoyes01@aol.com). 0015-0282/99/$20.00 PII S0015-0282(99)00235-6 Most unsuccessful IVF cycles are caused by embryo implantation failure. Factors influencing IVF implantation can be divided into three categories: [1] embryonic, [2] endometrial, and [3] method of ET. Today, most embryos created with assisted reproductive technologies reach the uterine cavity via a transcervical approach using one of a variety of catheters designed to achieve this goal. Limited data exist about the influence of the ET procedure on IVF success. The results of one large series suggest that a more difficult ET results in a lower chance for pregnancy (1). Several investigators (2, 3) reported that frank cervical blood at the time of IVF-ET lowers the chance for pregnancy, reinforcing the hypothesis that a traumatic transfer is not optimal for IVF success. The IVF-ET difficulty may result from operator inexperience or from other variables beyond the operator s control. For instance, cervical stenosis can lead to an embryo replacement procedure that is extremely difficult or impossible even in the best of hands. Various techniques have attempted to circumvent difficult cervical passage. Cervical dilation at oocyte retrieval has been used in cases of known cervical stenosis. Although this usually results in easier embryo replacement, the resultant pregnancy rate is dismal (4). Ultrasound-directed transmyometrial embryo placement has been suggested for patients with difficult ETs or known cervical stenosis. Although some investigators have achieved excellent success with this technique (5, 6), others have not (7). Embryo replacement by zygote intrafallopian transfer (ZIFT) is an option in patients with difficult cervical ET, but this 261

alternative requires laparoscopy. We discuss the outcome of ETs based on procedural difficulty and present a novel approach to difficult embryo replacement procedures. MATERIALS AND METHODS From September 1995 to May 1998, a total of 1,138 ETs were performed in our university-based IVF program in women 40 years of age. Embryo replacements (n 1,117) performed with two standard catheter choices, the Wallace (H.G. Wallace Limited, Colchester, England) (first-line) and Tomcat (Sherwood Medical, St. Louis, MO) (second-line) catheters, were considered for evaluation. Transfers performed with the Embryon catheter (n 21) were excluded because this represented a small subset of cycles in which the catheter was being tested. All patients accepted into the IVF program had undergone a precycle assessment of cervical accessibility using an empty Tomcat catheter. Embryo transfers were performed by one of four physicians affiliated with the IVF program. The operating physician graded ET difficulty at the time the procedure was completed, and a list was compiled for later evaluation. ET grades included [1] easy, [2] moderate, [3] difficult, and [4] extremely difficult. Easy transfers were those where embryos were placed without effort using a Wallace catheter. Moderate transfers included all other replacements performed with a Wallace catheter, i.e., those requiring a tenaculum to straighten the cervical canal and/or significant sheath-threading into the cervix, as well as transfers performed with a Tomcat catheter without difficulty. A speck of blood was considered acceptable in this group. Difficult transfers were those performed with a Tomcat catheter requiring cervical manipulation similar to that of moderate transfers performed with a Wallace catheter, but whose end result was smooth placement of the embryos into the uterine cavity. A speck of blood or the use of anesthesia or sedation was acceptable in this group. Extremely difficult transfers included those performed in women with cervical stenosis where the Tomcat catheter was threaded into the cervix with extreme difficulty, with or without cervical dilation, or where frank bleeding from the cervix was seen after the procedure. These transfers were not believed to have a smooth end. Pregnancy and embryo implantation rates relative to the type of catheter used and the technical difficulty of the ET were calculated. In addition, patients whose cervical canals were extremely difficult to negotiate and who wanted to undergo further IVF cycles were offered surgical correction of the stenosis. Institutional review board approval was not required for this procedure. Hysteroscopic evaluation with surgical creation of a smooth cervical tract was performed where necessary. This TABLE 1 Parameters and outcome of IVF-ET performed with the Wallace and Tomcat catheters. Variable Wallace (n 1,025) involved use of the operative hysteroscope with a 90 wire loop using 70 W of current. A smooth, straight tract was created in the posterior midline of the cervix measuring approximately 8 mm in width and 5 mm in depth beginning at the internal cervical os and extending at least half way down the cervical canal toward the external os. All patients who underwent this procedure had a size 12 French pediatric rubber catheter (Bard Urological, Covington, GA) placed transcervically into the uterine cavity at the completion of the procedure with concomitant oral antibiotics for 3 14 days. Postoperatively, outcomes of mock ET and IVF cycles were reported. Clinical pregnancy was defined as visualization of a gestational sac on ultrasound. Statistical analysis was performed with a standard computerized statistics program using 2 and Student s t-test where applicable. Statistical significance was considered at P.05. RESULTS Type of catheter Tomcat (n 92) Mean ( SEM) age 34.7 0.1 35.0 0.1 Mean ( SEM) no. of oocytes 12.5 0.2 13.9 0.7 Mean ( SEM) no. of embryos for ET 3.4 0.03 3.6 0.1 Ease of ET 847 E;178 M 18 M;67 D;7 ED No. (%) of clinical pregnancies 577 (56.2%)* 31 (33.7%)* Embryo implantation rate (no. [%]) 999/3,522 (28.4%) 54/328 (16.5%) Note: Ease of ET: E easy; M moderate; D difficult; ED extremely difficult. Embryo implantation rate no. of embryos that implanted per no. of embryos transferred. * P.0001. P.0001. The overall clinical pregnancy rate of the study group was 54.4%. The embryo implantation and multiple pregnancy rates were 27.5% and 54.6%, respectively. Clinical parameters and pregnancy and embryo implantation rates of cycles where different ET catheter types were used are shown in Table 1. Most ETs (92%, 1,025 of 1,117) were performed with the Wallace catheter, our program s transfer instrument of choice. Thus, any Tomcat catheter transfer was preceded by a trial Wallace catheter failure. The clinical pregnancy and embryo implantation rates of cycles performed with a Wallace catheter were statistically significantly higher than those where a Tomcat catheter was used (P.0001). 262 Noyes et al. Embryo transfer difficulty and IVF success Vol. 72, No. 2, August 1999

TABLE 2 Clinical pregnancy rate by ET difficulty. FIGURE 1 Projecting ridge (arrow) of tissue in cervical canal. ET difficulty No. of pregnancies/ no. of ETs (%) Easy Wallace 474/847 (56) Moderate no tenaculum 85/154 (55) tenaculum 19/42 (45) Difficult no tenaculum 13/25 (52) tenaculum 17/42 (41) Extremely difficult 0/7 (0) Clinical pregnancy rates relative to difficulty of embryo replacement were tabulated (Table 2): 75.8% of ETs were considered easy, 17.6% moderate, 6.0% difficult, and 0.6% extremely difficult. No pregnancies occurred in the extremely difficult group, whereas clinical pregnancy rates were not statistically significantly different from each other when the ET was graded easy, moderate, or difficult. This suggests that even if a transfer requires considerable skill, the pregnancy rate is not affected as long as the ET is completed in a clean, smooth fashion. Within a particular grade, the pregnancy rate of cycles requiring a tenaculum (to straighten the cervical canal) was lower, although not significantly so, than those where a tenaculum was not used (Table 2). A total of eight patients 40 years of age underwent hysteroscopic evaluation of the cervix (Table 3). Seven of the patients had a history of an extremely difficult embryo TABLE 3 Characterization of eight patients with cervical stenosis. replacement, and one patient (patient 2) had undergone three horrible attempts at IUI. At hysteroscopy, three of the eight patients (patients 1 3) had projecting ridges within the cervical canal (Fig. 1), two of which also had a tortuous cervical path. The protruding ridges were resected hysteroscopically with a wire loop at 70 W of cutting current. Two patients (patients 4 and 5) had a nonelastic, tight internal cervical os, whereas two others (patients 6 and 7) were acutely ante- FIGURE 2 Relatively obstructive posterior internal os tissue (two shorter arrows) in cervical passage of an acutely anteverted uterus. Note old perforation site (longer arrow) below obstructing ridge. Patient no. Age (y) No. of prior IVF cycles Postoperative mock ET No. of embryos transferred Result 1 29 2 Easy 3 Triplets 2 35 3 (IUI) Easy 4 Twins 3 34 1 Easy 3 #1 Miscarriage 4 #2 Quadruplet (1) 4 35 2 Easy 3 Twins 5 38 1 Moderate 4 #1 Negative 4 #2 Negative 6 32 3 Easy 4 #1 Negative 4 #2 Triplets (FET) 7 34 2 Easy 4 #1 Negative (FET) 4 #2 Triplets 8 38 1 Easy 3 Singleton Note: FET frozen embryo transfer. FERTILITY & STERILITY 263

FIGURE 3 Posterior cervical canal during shaving of obstructing tissue. The perforation site (arrow) was subsequently cauterized closed. TABLE 4 In vitro fertilization data of patients with cervical stenosis. Variable Value No. of patients 8 Mean ( SEM) age 34.4 1.1 Mean ( SEM) no. of prior IVF cycles 1.71 0.3 No. of postoperative IVF ET 12 Easy postoperative IVF ET (%) 10/12 (83%) Mean ( SEM) no. of embryos transferred 3.7 0.1 Clinical pregnancy rate No. pregnancies/no. of cycles (%) 8/12 (67) No. pregnancies/no. of patients (%) 7/8 (86) Ongoing pregnancy rate No. pregnancies/no. of cycles (%) 7/12 (58) Multiple pregnancy rate No. of multiple births/no. of cycles (%) 6/12 (50): 2 twin 3 triplet 1 quadruplet Embryo implantation rate (no. [%]) 19/45 (42.2%) Note: Embryo implantation rate no. of embryos that implanted/total no. of transferred embryos; clinical pregnancy gestational sac on ultrasound; ongoing pregnancy 18 weeks of gestation. flexed with the posterior internal os tissue relatively obstructive to catheter passage (Fig. 2). In one of the latter two cases, the patient had an old perforation site inferior to the internal os tissue (Figs. 2 and 3). In all of the above cases, a smooth posterior cervical tract was created by surgically shaving away approximately 0.5-mm depth of cervical tissue starting at the internal cervical os extending caudad to at least the midportion of the cervical canal. One of the eight patients (patient 8) had synechiae at the internal cervical os as the cause of her cervical obstruction. These were resected with use of hysteroscopic scissors and alligator forceps. All patients had a Foley catheter placed in the uterus postoperatively. Prophylactic antibiotics were administered for the duration of the Foley catheter placement. Postsurgical mock ETs were performed in all patients. They were graded easy in six of eight and moderate in two of eight patients after the hysteroscopic surgery. Results of subsequent IVF attempts are presented in Tables 3 and 4. A total of 12 ETs were performed in the eight postoperative patients. Clinical pregnancy was achieved in 8 of 12 (66.7%) cycles. Only one patient had a miscarriage. The embryo implantation rate of the cycles was 42.2%, and the multiple pregnancy rate was 50%. DISCUSSION Technical skill and method of ET are important variables in IVF success. Our program consists of four physicians who perform all of the embryo replacements, and each has performed 1,000 ETs procedures in his/her career. There is no statistically significant difference among pregnancy rates between the physicians. Of embryo replacements evaluated, 76% were considered easy, and only 0.6% were considered extremely difficult, verifying the skill of the clinicians performing the procedures. The data presented in this article reveal that if a transfer is performed smoothly, despite its difficulty, pregnancy rates are excellent. It also suggests that use of a tenaculum, if necessary, during the transfer procedure does not significantly lower pregnancy rates, although a larger number of patients are needed to ensure this lack of statistical significance. Use of a Tomcat catheter did result in a lower pregnancy rate compared with the Wallace catheter, but the Tomcat group was biased because the latter group represented technically more difficult ET procedures. Patients with cervical stenosis or a history of extremely difficult ET procedures represent a small but significant group of IVF failures. This study supports the use of hysteroscopic evaluation of these patients and presents a novel approach to surgical correction of their cervical impedance. After hysteroscopic resection, most patients had an easy embryo replacement, and 67% of cycles resulted in a clinical pregnancy. One patient experienced two ETs that were moderate and difficult after hysteroscopic repair and did not achieve pregnancy despite two IVF attempts. Of all the patients, her postoperative intrauterine Foley catheter was left in for the shortest duration (3 days). Perhaps, 3 days is not long enough for optimal cervical healing. 264 Noyes et al. Embryo transfer difficulty and IVF success Vol. 72, No. 2, August 1999

Multiple pregnancy and embryo implantation rates were high in patients who became pregnant. We tended to be lenient with embryo number due to multiple prior IVF failures and thus often transferred one more embryo than is customary for a given age. These data make the point that if the ET procedure has been significantly improved, the number of embryos replaced should be limited to the age-appropriate number for one s IVF program to avoid high-order multiple gestations. References 1. Mansour R, Aboulghar M, Serour G. Dummy embryo transfer: a technique that minimizes the problems of embryo transfer and improved the pregnancy rate in human in vitro fertilization. Fertil Steril 1990;54:678 81. 2. Goudas V, Hammitt D, Damario M, Session D, Singh A, Dumesic D. Blood on the embryo transfer catheter is associated with decreased rates of embryo implantation and clinical pregnancy with the use of in vitro fertilization-embryo transfer. Fertil Steril 1998;70:878 82. 3. Wisanto A, Janssens R, Deschacht J, Camus M, Devroey P, Van Steirteghem A. Performance of different embryo transfer catheters in a human in vitro fertilization program. Fertil Steril 1989;52:79 84. 4. Groutz A, Lessing J, Wolf Y, Yovel I, Azem F, Amit A. Cervical dilation during ovum pick-up in patients with cervical stenosis: effect of pregnancy outcome in an in vitro fertilization-embryo transfer program. Fertil Steril 1997;67:909 11. 5. Kato O, Takatsuka R, Asch R. Transvaginal-transmyometrial embryo transfer: the Towako method experiences of 104 cases. Fertil Steril 1993;59:51 3. 6. Sharif K, Afnan M, Lenton W, Bilalis D, Hunjan M, Khalaf Y. Transmyometrial embryo transfer after difficult immediate mock transcervical transfer. Fertil Steril 1996;65:1071 4. 7. Groutz A, Lessing J, Wolf Y, Azem F, Yovel I, Amit A. Comparison of transmyometrial and transcervical embryo transfer in patients with previously failed in vitro fertilization-embryo transfer cycles and/or cervical stenosis. Fertil Steril 1997;67:1073 6. FERTILITY & STERILITY 265