DISPENSABILITY OF FIMBRIAE: OVUM PICKUP BY TUBAL FISTULAS IN THE RABBIT

Transcription

1 , I FERTIUTY AND STERIIJTY Copyright" 1979 The American Fertility Society Vol. 32, No.3, September 1979 Printed in U.SA. DISPENSABILITY OF FIMBRIAE: OVUM PICKUP BY TUBAL FISTULAS IN THE RABBIT KAREL G. P. METZ, M.D.* LUIGI MASTROIANNI, JR., M.D. Department of Obstetrics and Gynecology, University Hospital, Utrecht, The Netherlands, and Division of Reproductive Biology, Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania This study was carried out to test the general belief that fimbriae should be considered almost indispensable to ovum capture. Surgical procedures, including resection of the mesotubarium superius and infundibulum tubae together with the creation of a fistula, are described. An attempt to enhance tubal patency by temporarily using estrogens is discussed. Ovum pickup was determined by flushing both tubes and the uterus 2 days following ovulation induction. The prior use of estrogens did not appear to have increased subsequent ovum pickup in control animals; on the contrary, it even seems possible that estrogens had negatively influenced pickup by the fistulas. With five fistulas, however, three recovered ova represented a pickup rate of 7% to 14% of all ova available following one human chorionic gonadotropin-induced ovulation. These figures suggest that ovum pickup by terminal ampullary fistulas is not negligible; they might indicate that absence of fimbriae offers neither good protection against pregnancy following sterilization nor a hopeless prognosis for fimbriectomy reversal when a distal patent tube is present. Fertil Steril 32:329, 1979 It is generally thought that fimbriae have an almost indispensable role in ovum capture, 1 resulting in a widespread belief that fimbriectomy is a very reliable sterilization method. It is even postulated that fimbriectomy is not likely to produce any failures as fimbriated ends are excised,2 that preservation of the fimbriated end is essential if tubal sterilization is to be considered reversible,3 that fimbriectomy is absolutely irreversible,4 and that an attempt at reversal of sterilization is contraindicated in patients who have undergone fimbriectomy.5 The hypothesis that fimbriectomy is a very reliable sterilization method with virtually no potential for reversal implies that ovum pickup will never or rarely occur when tubal patency is achieved following attempted fimbriectomy rever- Received March 21, 1979; accepted April 23,1979. *Reprint requests: Karel G. P. Metz, M.D., Department of Obstetrics and Gynecology, University Hospital, Catharijnesignel101, Utrecht, The Netherlands. 329 sal or when it occurs spontaneously following sterilization. Restitution of fertility following fimbriectomy has been reported to be unsuccessful, but experience with such attempts thus far has been too limited 6, 7 to justify this hypothesis. On the other hand, unexpectedly high numbers of unwanted pregnancies have been reported following fimbriectomy as a result of fistula formation.s, 9 It was therefore decided to investigate ovum pickup by tubal fistulas following fimbriectomy in an animal model in order to test the postulated indispensability of fimbriae for ovum pickup. MATERIALS AND METHODS Animals. Sexually mature New Zealand White does weighing 3 to 4.5 kg were caged separately and had access to Purina chow and water ad libitum. They were exposed to 12 hours oflight and 12 hours of darkness each day. Surgical Procedures. Fimbriectomy was carried

2 330 METZ AND MASTROIANNI September 1979 out in consonance with the normal anatomy of the rabbit fallopian tube. The mesotubarium superius (MTS) extends as a thin membrane from the antimesosalpingeal border of the curved tube, terminating in a free border along which a fimbria (fimbria mesotubaria) extends. Because the fatty mesotubarium inferius (MTI) curves medially to the upper pole of the ovary, the distal part of the tube forms a loop, while the two mesotubaria create an open peri toneal pocket around the ovary. The fimbria mesotubaria and the distal ampulla must first be dissected free of the MTS to enable fimbriectomy. As the partially dissected MTS might become adherent and thus could interfere with subsequent ovum pickup, it was decided to resect the entire MTS completely at the time of fimbriectomy. Food was withheld from the animals for approximately 24 hours before surgery. Anesthesia was induced and maintained with intravenous sodium pentobarbital. All surgery was performed by a single individual (K. G. P. M.). The abdomen was entered through a midline incision. A wet gauze was placed between the adnexa and the posterior abdominal wall to bring the tubes into the operative field. A beveled, pliant, polyethylene splint was then inserted through the tubal ostium into the ampulla and held in place with a mosquito forceps placed around the fimbriated end to permit easy tissue handling and thus minimize trauma. The MTS was first completely resected using a fine diathermic needle. The fimbria ovarica was then coagulated with bipolar current at its attachment to the ovary and transected. The blood vessels coursing along the fatty MTI to the 3 cm of distal ampulla, visualized by transillumination, were subsequently ligated using atraumatic 6-0 polyglycolic acid sutures (Dexon). A fine curved clamp was then placed on the MTI between the ligatures and the tube, from the upper pole of the ovary to a point about 2 em distant from the ostium where the MTI met the tube, and the MTI was incised adjacent to the clamp. The clamp was allowed to remain in situ for about 30 minutes to obtain better coaptation of the wound edges, thus keeping extrusion of fat to a minimum. If bleeding then occurred, hemostasis was achieved either by bipolar coagulation or with a crosstitch of 6-0 Dexon. Meanwhile, an atraumatic 3-0 catgut suture was applied through the MTI near the tip of the clamp and gently tied around the splinted ampulla, both to decrease blood flow in a distal direction temporarily and to prevent the remaining ampullary splint from slipping out following fimbriectomy. The ampulla and splint were then cut transversely 0.5 cm distal to this suture, and the fimbriated end was removed. The largest blood vessels within the tubal wall of the 0.5 em of tube distal to the catgut suture, easily visualized because the ampulla was stretched around the remaining splint, were subsequently coagulated adjacent to the resection site with fine bipolar forceps. The catgut suture was left in place for approximately 30 minutes to allow hemostasis. If bleeding from the resection site then occurred, careful bipolar coagulation was used. Finally, the splint was removed and one atraumatic 6-0 Dexon suture was used to bring the fistula to a point within 1 cm of the ovary. Thus, the distal part of the ampulla covered part of the resection surface at the MTI. The above procedure was carried out on both sides. During the procedure, tissues were frequently irrigated with normal saline to avoid drying and to locate bleeding sites. Before closing the abdomen, the peritoneal cavity was gently washed with saline, taking care to remove any blood clots. After closing the peritoneum, 100 ml of normal saline were instilled by needle intra peritoneally to further impair adhesion formation. 10 This procedure was carried out in seven animals. Ovum pickup was finally determined in four of the seven (group I). Additional Use of Estrogens. As inadvertent closure of fistulas frequently occurred during a pilot study, an additional measure was tried. The basis for this was the observation that the mean volume of accumulated tubal fluid is much greater 24 hours following ligation at both proximal and distal ends than it is when the ligature is placed at the distal end only,ll suggesting that tubal fluid flows mainly in the direction of the uterus under conditions where flow to the peritoneal cavity is impaired. It was therefore postulated that, if the luminal fluid were prevented from flowing into the uterus following fimbriectomy, an accumulation of fluid might aid in disrupting a possible occlusion of the distal extremity by blood or fibrin. In addition, if such a reversal of flow were to take place, this might also promote ultimate patency by further preventing adherence between the wound edges by enhancing extrusion of the endosalpinx. As estrogens are also believed to provoke an occlusion at the uterine end of the tube and at the same time maximize luminal fluid production, their (temporary) use was preferred to ligation at the uterotubal junction. High doses of estrogens have been reported to prevent the usual disappearance of hydrosalpinx in a distally ligated tube as late as

3 Vol. 32, No.3 DISPENSABILITY OF FIMBRIAE 331 4lh days following mating. 12 Thus, 0.2 mg of estradiol (17.a-estradiol; Sigma Chemical Co., St. Louis, Mo.) in 0.6 ml of sesame oil was administered daily subcutaneously for 5 days beginning preoperatively on the day of surgery. Control Animals. Four animals were used to determine the rate of ovum pickup following laparotomy, without additional surgical procedures, including temporary insertion of an ampullary splint and peritoneal instillation of 100 ml of normal saline. These animals received 0.6 ml of sesame oil daily subcutaneously for 5 days beginning preoperatively on the day of surgery (group II). Four animals were used to determine a possible influence of estrogens on ovum pickup in the presence of intact genital organs. These animals underwent laparotomy, temporary insertion of a splint, and instillation of saline as in group II; 0.2 mg of estradiol in 0.6 ml of sesame oil was administered daily subcutaneously for 5 days beginning preoperatively on the day of surgery (group III). In four animals the MTS was resected, the ampullary splint was placed temporarily, and peritoneal instillation of 100 ml of normal saline was carried out. These animals received sesame oil as did those in group II (group IV). Four animals were used to determine a possible influence of estrogens on ovum pickup in the absence of the MTS. In these animals, the MTS was excised, a splint was temporarily inserted, and saline was instilled as in group IV. Estradiol in sesame oil was administered using the same regimen as in groups I and III (group V). Determination of Ovum Pickup. An arbitrary period of 3 weeks was chosen before ovulation was induced with 75 IU of human chorionic gonadotropin (Sigma Chemical Co.). Two days following ovulation induction, the animals were killed with an overdose of 95% alcohol injected intravenously. The tubes, ovaries, and uterus were excised en block. The tubes and uterus were dissected free of supportive ligaments and ovaries. Each tube was then flushed twice with 2 ml of normal saline, using a small needle passed through the tubal wall. The first flushing was carried out from the proximal isthmus in the direction of the distal ampullary end. If the distal end was proved patent, the isthmus was flushed again, this time from the proximal ampulla, following separation of the tube at the uterotubal junction in order to recover any ova which might still be located at the uterine end of the tube. Finally, each uterus was flushed once from the cervix in the direction of the uterotubal junction, using 5 ml of normal saline. Each flushing was collected separately in a watch glass and examined for ova with a binocular dissecting microscope. The ovaries were examined under the dissecting microscope and the number of fresh corpora lutea (CL) was determined; this number was assumed to correspond with the number of ova available for pickup. DataAnalysis. The number of recovered ova was recorded for each tube together with its connecting uterus and expressed as a percentage of the number of ova available for pickup. The mean percentage of recovered ova was then determined by averaging these percentages within each group. Differences between means were tested by Student's t-test. RESULTS From inspection of the ovaries during surgery, all animals were presumably in estrus with the TABLE 1. Results in Group I: Resection of the Mesotubarium Superius and Infundibulum Tubae with Administration of Estrogens No. of ova avail- Fistulas able for pickup Rabbit no. Side No. of ova Diameter Endosalpinx Location recovered Exclusive of Inclusive of extrusion external external migration migration 1 L <lmm Minimal Adherent midway on ovary 2 R <lmm Minimal Adherent ± 3 cm from ovary" L ±lmm Slight Adherent 0.5 cm from ovary" 3 R ±lmm Obvious Adherent midway on ovary 4 L ±2mm Obvious No adherence Total % Recovered ova (approximately) 14 7 "Upper pole.

4 332 METZ AND MASTROIANNI September 1979 TABLE 2. Results of Group II (Laparotomy), Group III (Laparotomy with Administration of Estrogens), Group IV (Resection of Mesotubarium Superius), and Group V (Resection of Mesotubarium Superius with Administration of Estrogens) No. of ova recovered No. of ova available for pickup Group IV Group V Rabbit no. Side Group Group Group Group Group Group II III IV V II III -EM" +EM" -EM" +EM' 1 R (3") 2 L 2 5 2d I d R 2 2 2d L R 3 5 d 2 6" L d Od, R L Total a_em, exclusive of external migration (except for one ovum). The figures correspond to the numbers of corpora lutea, except in group IV (rabbit 3, L), where five CL were present. It is assumed that all recovered ova originated from the ipsilateral ovaries, except one in group IV (rabbit 3, L), which was assumed to have been also available to the right side. b + EM, inclusive of external migration. Within groups IV and V it is assumed that only the nonrecovered ova might also have been available to the contralateral tube following external migration, in addition to one ovum in group IV (rabbit 3, R). C Although three corpora lutea were present, their ova were believed not to have reached the peritoneal cavity, as the ovary was completely embedded in adhesions. din seven cases, the end ofthe fimbria mesotubaria was adherent to the ampulla. In all other cases the fimbria mesotubaria was not attached. exception of two in group III (animals 2 and 4), which. apparently were pseudopregnant. In group I, three animals of a total of seven were discarded. In one, hemostasis could not be obtained because of the abundance of fat at the resection site, and the experiment was terminated. Another died 1 hour postoperatively and a third 19 days postoperatively. In the latter, a fistula 2 mm in diameter with obvious extrusion of the endosalpinx was observed on one side, and the other tube was distally occluded. In the four remaining animals, five fistulas of varying size with varying degrees of extrusion of the endosalpinx were observed (Table 1) and three tubes proved to be distally occluded. Twenty-one fresh CL were present in the ovaries ipsilateral to the fistulas. Two, six, and five fresh CL were present in the ovary contralateral to the fistula in animals 1, 3, and 4, respectively. Three ova were recovered. In groups II and III, 33 and 38 ova were recovered, while 36 and 40 fresh CL, respectively, were present (Table 2). In groups IV and V, with 41 and 39 fresh CL, 29 and 15 ova, respectively, were recovered. An obvious mucin layer was observed around each ovum. Four ova in group IIIand one in group V were recovered from the second tubal flushing, while three ova in group II, three in group IV, and one in group V were recovered from the uterus. All other ova were recovered from the first tubal flushing. DISCUSSION The creation of five fistulas with only three tubal occlusions suggests that the surgical approach used was reasonably satisfactory. Tubal patency might have been enhanced by the use of estrogens, and it would seem useful to conduct additional experiments to establish the importance of such treatment. In all five fistulas, the endosalpinx appeared grossly to have been extruded in a manner which has been observed previously in postfimbriectomy patients. 8, 9 In a pilot study, a similar variabili ty in the extrusion of endosalpinx and in the diameter of the ostia was observed with four fistulas 5 to 7 days postoperatively in a series in which estrogens were not used. This suggests that the gross morphology of the fistulas had not in fact been altered by the use of estrogens, but this point merits further investigation. In group I, the number of CL ipsilateral to the fistulas corresponds to 21 ova available for pickup if the possibility of external migration is not considered (Table 1). However, as pickup of ova has been reported following the removal of the ovary on one side and of the tube on the other, 13 the possibility of pickup of ova from an ovary contralateral to the fistula should also be considered. If one includes all of the ova available for pickup, including those that might have reached the fistulas by external migration, the total becomes maximally 44. Thus, the three recovered ova con-

5 Vol. 32, No.3 DISPENSABILITY OF FIMBRIAE 333 TABLE 3. Comparison of the Mean Percentage of Recovered Ovaa between Groups II and III and between Groups N and V Excluding external migration" Including external migration' Group II Group III Difference 91 ± 6% 96± 2% Not significant Group IV Group V Difference Group IV Group V Difference 65± 9% 29 ± 13% < P < ± 9% 19± 8% 0.01 < P < 0.02 asee "Materials and Methods," data analysis, for definition. Values are means ± standard error of the mean (n = 8; group V, excluding external migration, n = 7). bsee footnote a, Table 2. csee footnote b, Table 2. stitute between 7% and 14% of all ova available for pickup. Around ovulation time, the fimbriae are in intimate contact with the ovarian surface as a result of the combined action of intermittent muscular contractions ofthe tube and ofthe tubal and ovarian membranes. It then seems reasonable to suppose that the ova are-in general-captured by the ipsilateral tube and that pickup by a contralateral tube following external migration does not generally occur in a normal situation. This supposition can be supported by comparing the number of CL to the number of ipsilaterally recovered ova within groups II and III (Table 2). Consequently, the availability of ova for pickup within these groups has been analyzed without considering the possibility of external migration. When comparing the mean percentages of recovered ova between groups II and III (Table 3), a significant difference was not revealed, suggesting that the prior use of estrogens did not negatively influence subsequent ovum pickup, at least under normal anatomical conditions. Pickup of ova from a contralateral ovary, however, might have occurred within groups IV and V, as suggested by at least one ovum in group IV (animal 3), resulting in at least 42 and 39 ova available for pickup, respectively (Table 2). Whenever external migration of additional ova is considered, it should be remembered that direct contact between the fimbriae and the ipsilateral ovary remains a possibility in the absence of the MTS13 and that ova remain adherent to the ovary for many hours following ovulation if they are not captured.14 It seems therefore reasonable to suppose that ovum pickup by the ipsilateral tube occurs preferentially in these groups, and thus that possible external migration should be limited only to nonrecovered ova from an ipsilateral ovary. Consequently, a maximum of 54 and 63 ova seems to be the best estimation when including external migration to calculate the number of ova available for pickup. When comparing the mean percentages of recovered ova in groups IV and V, a significant difference is revealed both excluding and including external migration (Table 3). The prior use of estrogens therefore seems to reduce subsequent ovum pickup in the absence of the MTS. Although the mechanism of this possible estrogenic effect is unknown, it might well be that this effect also contributed in a negative sense to the ability ofthe fistulas to pick up ova, as the MTS was also absent here. The figure of 7% to 14% might therefore be considered too low to represent the real capability of ovum pickup by fistulas. The 7% to 14% (or greater) rate of ovum pickup by fistulas does not favor the concept that the presence of normal fimbriae is almost indispensable for ovum pickup. In any event, this finding undermines the general belief that pregnancy will hardly occur in the presence of a distal patent tube following fimbriectomy or its reversal, considering the number of ovulations occurring before the end of reproductive life. With two of four rabbits, a pregnancy would have been theoretically possible following one multiple ovulation only, assuming that captured ova are as well-fertilized, developed, and transported as in the intact tube. Acknowledgment. The statistical analysis by Mr. H. J. M. Beyer is gratefully acknowledged. REFERENCES 1. Editorial comment: Re Metz KGP: Failures following fimbriectomy. Obstet Gynecol Survey 32:397, Soonawala RP: Vaginal sterilization. In Advances in Voluntary Sterilization: Proceedings of the Second International Conference, Geneva, Switzerland, February 25 to March 1, 1973, Edited by ME Schima, I Lubell, JE Davis, E Connel, DWK Cotton. Amsterdam, Excerpta Medica Int Congr Ser 284, 1974, P Garcia CoR: Oviduct anastomosis procedures. In Human Sterilization, Edited by RM Richart, DJ Prager. Springfield, Ill, Charles C Thomas, 1972, p 116

6 334 METZ AND MASTROIANNI September Mehta PV, Bhatia DL, Pai DN: Evaluation of single-stitch tubal ligation in postpartum Women. Obstet Gynecol 51:567, Diamond E: Microsurgical reconstruction of the uterine tube in sterilized patients. Fertil Steril 28:1203, Siegler AM, Perez RJ:. Reconstruction offallopian tubes in previously sterilized patients. Fertil Steril 26:383, Williams GFJ: Fallopian tube surgery for reversal of sterilization. Br Med J 1:599, Metz KGP: Failur$! following fimbriectomy. Fertil Steril 28:66, Metz KGP: Failures following fimbriectomy: a further report. Fertil Steril 30:269, Seitz HM Jr, Schenker JG, Epstein S, Garcia C-R: Postoperative intraperitoneal adhesions: a double-blind assessment of their prevention in the monkey. Fertil Steril 24:935, Hafez ESE: The uterotubal junction and the luminal fluid of the uterine tube in the rabbit. Anat Rec 145:7, Black DL, Asdell SA: Mechanism controlling entry of ova into rabbit uterus. Am J Physiol 197:1275, Westman A: A contribution to the question ofthe transit of the ovum from the ovary to uterus in rabbits. Acta Obstet Gynecol Scand [Suppl]5:1, Blandau RJ: Gamete transport--comparative aspects. In The Mammalian Oviduct, Edited by ESE Hafez, RJ Blandau. Chicago, University of Chicago Press, 1969, p 129