FERTILITY AND STERILITY Copyright <> 987 The American Fertility Society Printed in U.S.A. Results of microsurgical reconstruction in patients with combined proximal and distal tubal occlusion: double obstruction Phillip E. Patton, M.D. Tiffany J. Williams, M.D.* Carolyn B. Coulam, M.D.t Mayo Clinic and Mayo Foundation, Rochester, Minnesota Before the advent of in vitro fertilization and embryo transfer (IVF-ET), both macrosurgical and microsurgical techniques were used to correct postinfl.ammatory tubal occlusion. Despite the operative success in many forms of tubal disease, pregnancy rates have been poor in patients with multiple sites of tubal occlusion. In patients having extensive tubal disease, which carries a poor postoperative prognosis, IVF-ET may be a better option than surgery. A poor operative prognosis might be expected in patients with evidence of both proximal and distal tubal occlusion, but the predicted pregnancy rate in these patients has not been determined adequately. Patient selection, length of follow-up, methods of pregnancy reporting, and the rarity of this disorder make it difficult to determine even a theoretical pregnancy rate. Since pregnancy rates after microsurgical reconstruction of both the proximal and the distal portions of the oviduct are unknown, preoperative counseling is limited. The purpose of this article is to report the microsurgical results in patients having both proximal and distal oviductal occlusion and to discuss the preoperative evaluation and management options. Received November 3, 986; revised and accepted June 6, 987. *Reprint requests: Tiffany J. Williams, M.D., Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905. t Present address: Carolyn B. Coulam, M.D., Methodist Center for Reproduction and Transplantation Immunology, 604 North Capitol Avenue, Indianapolis, Indiana 46202. Copyright 987 The American Fertility Society. 670 Patton et al. Communications-in-brief MATERIALS AND METHODS Records of 295 patients with inflammatory oviductal disease who underwent microsurgical tuboplasty at the Mayo Clinic from 977 through 983 were reviewed. Thirty-one patients who had at least two distinct and separate anatomic sites of both proximal and distal occlusion were included in the study. Proximal occlusion was defined as occlusion of the interstitial or isthmic portion (or both portions) of the oviduct. Distal occlusion was defined as occlusion distal to the isthmic segment, primarily at the fimbriated end. Patients who had bilateral occlusion of only a single anatomic site (for example, bilateral isthmic occlusion) were excluded from the study. No patients were excluded on the basis of age, extent of tubal disease, duration of infertility, tubal length, history of prior surgery, or the extent and degree of concurrent pelvic disease. Thirteen patients in the study were nulligravid and 8 were multigravid. The mean and range of age were 27.5 years and 20 to 35 years, respectively. The mean duration and range of infertility were 3.8 years and 0.4 to years, respectively. The mean length of follow-up was 250 days, and the range was 29 to 375 days. An infertility investigation consisted of a complete history and physical examination and documentation of ovulatory status by basal body temperature, luteal phase, or timed endometrial biopsy. Male factor was assessed by a semen analysis. A preoperative hysterosalpingogram was obtained on all patients. Factors other than tubal ones were corrected postoperatively as indicated.
All microsurgical procedures were done with the use of a Zeiss operating microscope (OPMI-6; Carl Zeiss, Inc., Wellesley, MA). Atraumatic tissue-handling, a fine suture material (7.0 and 8.0 Vicryl; 8.0 Ethilon [Ethicon, Somerville NJ]), bipolar cautery for hemostasis, unipolar microcautery for lysis of adhesions, and continuous irrigation with a physiologic balanced salt solution containing heparin were used. No intraoperative adjuncts other than heparin irrigation were used. Antibiotics were given prophylactically in most cases; ovarian suspension and hydrotubations (as described by Grant 2 ) were used in selected cases. No other medical or surgical adjuncts were used. Sixteen of the 3 patients had had previous pelvic surgery, and a total of 9 operations had been performed (Table ). Twenty-one patients had preoperative laparoscopy, and 9 operative records were available for review. All but one patient had laparoscopy performed within 2 years of surgery, and 7 (89.4%) had laparoscopy performed within 0 months of surgery. The sites of tubal occlusion predicted at laparoscopy were compared with the sites of obstruction found at the time of microsurgery. Sensitivity and specificity of laparoscopy in predicting proximal and distal occlusion were determined. Sensitivity was defined as the number of occluded sites correctly predicted by laparoscopy divided by the total number of occluded sites found at microsurgery. Specificity was defined as the number of patent oviducts predicted by laparoscopy divided by the total number of patent sites found at microsurgery. In each patient, the anatomic site of oviductal occlusion was determined at the time of microsurgery. Eighteen patients had proximal bilateral occlusion and distal bilateral occlusion (PBO-DBO). Five patients who had a single oviduct with both proximal and distal occlusion of the remaining oviduct were included in this category. Seven patients had proximal unilateral occlusion and distal bilateral occlusion (PUO-DBO). Six patients had prox- Table Previous Surgical Procedures in Patients with Oviductal Occlusion Table 2 Grading of Tubo-ovarian Adhesions and Hydrosalpinges Adhesions Hydrosalpinx Best Worst Best Worst Grade" tube tube Gradeb tube tube I 4 0 0 6 0 II 5 3 I 8 6 III 6 9 II 8 5 IV 6 4 III 8 0 IV 5 Total 3 26 3 26 "Grading scale for adhesions: I, <25% of ovarian surface involved; II, 25%-50%; III, 50%-75%; IV, 75%-00%. b Grading scale for hydrosalpinx: 0, no occlusion; I, occlusion only; II, <5 mm diameter of tube; Ill, 5-30 mm; IV, >30 mm. imal bilateral occlusion and distal unilateral occlusion (PBO-DUO). The degrees of tubo-ovarian adhesions and fimbria! disease were graded for each oviduct by review of operative reports and from photographs obtained at each operation. The grading scale for adhesions and fimbrial disease for the fallopian tube with the least disease (best tube) and the tube with the most extensive disease (worst tube) is presented in Table 2. Significant histologic findings noted at the time of microsurgery are presented in Table 3. Sixteen of the 3 patients had postoperative hysterosalpingograms available for review; 68.8% had PBO-DBO, 2.5% had PUO-DBO, and 8.8% had PBO-DUO preoperatively. In each case, the radiographic reports were reviewed for evidence of recurrent proximal and distal disease. Statistical analysis of conception and pregnancy rates was determined by the use of the Kaplan Meier survival curve. RESULTS Only three patients in the series conceived. Two were in the group that had PUO-DBO and one was in the group that had PBO-DBO at the time of Salpingectomy (with or without oophorectomy) Ovarian cystectomy Tuboplasty Ovarian wedge resection Appendectomy Laparotomy for small bowel obstruction Partial salpingectomy 5 4 5 2 Table 3 Histologic Diagnosis in Patients with Proximal and Distal Obstruction Salpingitis isthmica nodosa Chronic salpingitis Intratubal endometriosis Noninflammatory fibrous adhesions 4 3 3 Patton et al. Communications-in-brief 67
microsurgery. There were two ectopic pregnancies, one spontaneous abortion, and no full-term pregnancies in the study population. The conception rate at 2.5 years of observation was 2.% (95% confidence interval± 0.2). Most patients in the study had significant tuboovarian adhesions and hydrosalpinges. Although the degree of tubo-ovarian adhesions was variable, all patients in the study had at least grade I adhesions, and 38.7% had at least grade III adhesions in the fallopian tube with the least disease. The degree of tubo-ovarian adhesions was similar bilaterally, and no patients in the study had tubo-ovarian adhesive disease that differed by more than 2 degrees. The degree of hydrosalpinx ranged from occlusion with minimal dilatation to hydrosalpinges greater than 30 mm in diameter. Although 9.4% had occlusion only, 54.8% had hydrosalpinges that were at least 5 mm in diameter in the fallopian tube with the least disease. In patients with bilateral disease, the degree of damage was within degree in all but four patients. Three patients had grade III hydrosalpinges, and one patient had a grade IV hydrosalpinx, compared with a grade zero hydrosalpinx in the contralateral oviduct. In the 9 patients who had records available for review, the anatomic sites of occlusion predicted at laparoscopy and found at microsurgery were compared. A total of 38 observations of proximal and distal tubal patency were possible. (A unilateral tube was present in five cases, but the surgeon's interpretation of proximal and distal disease in the absent oviduct was included in the analysis.) The sensitivity of laparoscopy in predicting proximal and distal occlusion was 66 and 55%, respectively. The specificity of laparoscopy for detecting proximal and distal disease was 67 and 78%. The positive predictive value for proximal and distal occlusion was 89 and 96%, and the negative predictive value for proximal and distal occlusion was 35 and 4%, respectively. In only three patients did laparoscopy correctly predict both proximal and distal tubal status. Of the 6 patients who had a postoperative hysterosalpingogram, 4 had bilaterally patent oviducts as a measure of surgical success. Three patients were in the group with preoperative PBO-DBO and patient had preoperative PUO-DBO. Nine patients had radiographic evidence of proximal occlusion in one oviduct. Proximal reocclusion occurred in 8 patients (6 patients with PBO-DBO, patient with PBO-DUO, and patient with PUO DBO). Occlusion of a previously patent oviduct occurred in patient (PUO-DBO). Only 2 patients had distal reocclusion in a single oviduct, and both had preoperative PBO-DBO. DISCUSSION The operative success in the correction of both proximal and distal oviductal occlusion has been reported poorly. Operative success rates may be extremely variable because of the different methods of pregnancy reporting, patient age and selection, oviductal and pelvic disease, the use of surgical adjuncts, different surgical techniques, the extent of follow-up, and small numbers of cases. In this study, all patients with postinflammatory proximal and distal obstruction who underwent a microsurgical repair were evaluated. Patients were not excluded on the basis of age, tubal disease, or prior surgical procedures. All microsurgery was directed by one of the authors using a standard surgical technique. The mean length of follow-up from surgery was 250 days. The probability of conception at 2.5 years after surgery was 2% (95% confidence limits ± 0.2), but there were no live births. The results of the present study suggest that patients with multiple sites of oviductal occlusion have an extremely poor prognosis with surgical management. In an early series by Frantzen and Schlosser, operative repair of concomitant proximal and distal occlusion resulted in an extremely poor pregnancy rate. The same authors reported an 8.7% live birth rate in a larger series 3 ; however, the method of patient selection and degree of pelvic disease are unknown. Surprisingly, pregnancy rates were highest in patients with bilateral proximal and distal occlusion and were lower in patients with less extensive disease. Many factors other than tubal occlusion could have contributed to the poor results found in our study. Previous surgical history, associated pelvic disease, and fertility factors in addition to tubal disease may have had a significant effect on surgical outcome. The theoretical effect of each of these variables on surgical outcome is unknown. In fact, seven patients were treated with clomiphene citrate for oligo-ovulation (n = 4) and luteal phase defect (n = 3) postoperatively, and one conception resulted. Although ovulatory disorders were thought to be corrected in most cases, they could 672 Patton et al. Communications-in-brief
have contributed to the poor outcome. Only one patient in the series was thought to have a significant male factor. In 5.6% of the patients in this series, microsurgery had been preceded by pelvic operation, which could have adversely affected attempts at microsurgical reconstruction. However, pregnancy and conception rates were equally poor in patients without prior pelvic surgery. In contrast to patients with only proximal oviductal occlusion, 4 most patients in the series had evidence of significant tubo-ovarian adhesions and oviductal hydrosalpinges, regardless of a history of previous pelvic surgery. It is probable, then, that the extent of tubo-ovarian disease in this study population had a significant adverse effect on conception and pregnancy rates. The prognostic significance of various intratubal pathologic processes on microsurgical results is unknown. In patients in the study population, salpingitis isthmica nodosa (SIN) was identified in the proximal oviduct at the time of microsurgery. In this study, areas of diseased oviduct were removed even though there was tubal patency, because previous studies have shown that reocclusion is more likely if diseased areas are not removed. SIN has been reported to be present in nearly 40% of patients with proximal oviductal occlusion. 5 The prognostic value of SIN in predicting tubal reocclusion, conception rate, and pregnancy rate has not been determined. In the present study, four patients with SIN had a postoperative hysterosalpingogram and three had evidence of proximal oviductal reocclusion. Although the results suggest that SIN is associated with tubal reocclusion, further studies are needed to determine whether SIN has a significant effect on the success of microsurgical reconstruction. A critical factor in the assessment of surgical results is the length of patient follow-up. Although the mean length of follow-up was 250 days, it is possible that, with extended follow-up of selected cases, a higher pregnancy rate may result.* However, only one patient in the series had a follow-up rate of less than 30 days, and all other patients had a postoperative follow-up duration of at least year. Although additional pregnancies may result * An additional ectopic pregnancy and an ongoing intrauterine pregnancy in a single patient have been recorded since submission of this manuscript. with extended observation after surgery, many patients will not have the option of extended observation because of age and possibly other factors. The poor pregnancy rate found in the study population after microsurgical repair strongly suggests that another option, such as IVF-ET, should be considered. It is surprising that patients with unilateral occlusion of either the proximal or distal portion of the oviduct associated with bilateral occlusion of the contralateral oviduct do not have a higher conception or live birth rate than patients with more extensive disease. Indeed, at our institution, a live birth rate of 53% has been found in patients with bilateral proximal oviduct occlusion. 4 The effects of concomitant proximal and distal disease may be more than additive. Undetected but significant tubal disease other than that found in the occluded tubal segment may be present in the oviduct with only unilateral occlusion. Because of the small sample size in each of the groups described, the possibility of an increased pregnancy rate in any one group cannot be excluded. Additional studies will be needed to determine whether any one group has a more favorable operative prognosis. However, the results suggest that a poor operative prognosis may depend on the finding of both proximal and distal occlusion and not necessarily on the number of anatomic sites occluded. Clearly, our study was not designed to assess the predictive value of laparoscopy prior to microsurgery. Although prospective studies are needed, the results of our laparoscopic data deserve mention. The sensitivity of laparoscopy in predicting proximal and distal occlusion was relatively poor (66 and 55%, respectively). In only 3 of 9 patients was laparoscopy completely accurate in predicting findings at the time of microsurgery. It is possible that, during the laparoscopy-microsurgery interval, progressive tubal disease occurred, which could account for the poor sensitivity of laparoscopy. However, only two patients in the series had a laparoscopy-microsurgery interval of longer than 0 months. In 3 instances, distal disease was incorrectly identified. It is possible that distal occlusion could occur with a patent proximal oviduct during the interval, but it seems unlikely that distal occlusion could occur in patients with preoperative proximal occlusion. In part, the degree of tubaovarian adhesions in this study may have limited the surgeon's ability to accurately assess distal disease. It is also possible that sensitivity rates might Patton et al. Communications-in-brief 673
have improved if all patients had had preoperative laparoscopy. Nevertheless, the poor conception and pregnancy rates found in patients with multiple sites of occlusion confirm the necessity of a concentrated effort to accurately assess proximal and distal tubal patency at the time of laparoscopy. IVF-ET resulted in pregnancy rates per patient of up to 33% in a large series, 6 and IVF-ET may be a superior option when the success rate of IVF-ET exceeds that of microsurgical reconstruction. In patients with multiple sites of oviductal occlusion, conception and live birth rates were extremely poor. In contrast, live birth rates reported from our institution were 53% in patients with bilateral proximal tubal occlusion 4 and 32% in patients with bilateral distal tubal occlusion. 7 Although the choice between IVF-ET and a microsurgical procedure depends on many factors, the present study suggests that IVF-ET should be considered strongly in patients who have evidence of preoperative proximal and distal disease and significant tubo-ovarian adhesions. Since conception rates and pregnancy rates appear to be a function of the sites of occlusion, an accurate laparoscopic evaluation is mandatory for preoperative counseling. SUMMARY Thirty-one patients underwent microsurgery for correction of postinflammatory occlusion of both the proximal and the distal portions of the oviduct. Preoperative staging of tubo-ovarian adhesions, hydrosalpinges, and sites of oviductal occlusion was completed in each patient. Laparoscopic records were reviewed to correlate laparoscopic sites of occlusion with those found at the time of microsurgery. The conception rate at 2.5 years of observation was 2%, and there were no live births. The sensitivity of laparoscopy to predict proximal and distal disease was 65.6 and 55.2%, respectively. The poor surgical outcome in these patients suggests that IVF-ET should be strongly considered and that a careful preoperative laparoscopic examination is necessary for prospective counseling in these patients. REFERENCES. Frantzen C, Schlosser HW: Microsurgery and postinfectious tubal infertility. Fertil Steril 38:397, 982 2. Grant A: Infertility surgery of the oviduct. Fertil Steril 22:496, 97 3. Verhoeven HC, Berry H, Frantzen C, Schlosser H-W: Surgical treatment for distal tubal occlusion: a review of 67 cases. J Reprod Med 28:293, 983 4. Patton PE, Williams TJ, Coulam CB: Microsurgical reconstruction of the proximal oviduct. Fertil Steril 4 7:35, 987 5. Fortier KJ, Haney AF: The pathologic spectrum of uterotubal junction obstruction. Obstet Gynecol 65:93, 985 6. Jones HW Jr, Acosta AA, Andrews MC, Garcia JE, Jones GS, Mayer J, McDowell JS, Rosenwaks Z, Sandow BA, Veeck LL, Wilkes CA: Three years of in vitro fertilization at Norfolk. Fertil Steril 42:826, 984 7. Thie JL, Williams TJ, Coulam CB: Repeat tuboplasty compared with primary microsurgery for postinflammatory tubal disease. Fertil Steril 45:784, 986 674 Patton et al. Communications-in-brief