- Email: [email protected]
Healing and restoration of fertility following microsurgical transection and anastomosis of the rabbit oviduct
Vol. 38, No.3, September 1982 Printed in U.S.A.
FERTILITY AND STERILITY Copyright c 1982 The American Fertility Society
Healing and restoration of fertility following microsurgical transection and anastomosis of the rabbit oviduct i
Carlton A. Eddy, Ph.D.* Virendraa K. Bajpai, Ph.D. t Department of Obstetrics and Gynecology, The University of Texas Health Science Center at San Antonio, San Antonio, Texas
The postoperative time course of healing following tubal microsurgery was examined in the rabbit by transmission electron microscopy (TEM) and scanning electron microscopy (SEM) and was correlated with changes in fertility. Groups of animals were bred or used for morphologic studies on days 1,2,3,4, 7, and 14 after surgery. During the first 4 postoperative days, a major loss of ciliary and secretory activity occurred and was characterized by disruption of subcellular organelles. Tissuedamage was restricted to within several millimeters of the anastomosis. Healing was well advanced by day 7 and was largely complete by day 14. Corresponding changes in fertility were observed. A significant decline in fertility was noted on the first postoperative day. Thereafter, fertility progressively improved and was normal by day 14. Fertil Steril 38:354, 1982
Tuboplastic microsurgery has become a widely accepted technique for the restoration offertility, ideally in women who have undergone previous elective tubal sterilization and who subsequently seek its surgical reversal. Among such patients, the success rate may approach 80%.1, 2 Despite a great deal of interest in improving the surgical technique itself, little is known regarding the time course of healing. As a consequence, no standardized scheme of postoperative management has evolved either with regard to appropriate medication or to when pregnancy should be attempted. Such decisions are at present largely subjective. At one extreme are those who favor prolonged use of intratubal stents and adjunctive therapy to
Received February 22, 1982; revised and accepted May 18; 1982. *Reprint requests: Dr. C. A. Eddy, Department of Obstetrics and Gynecology, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, Texas 78284. tWorld Health Organization Postdoctoral Fellow. 354
promote healing, avoid adhesion formation, and maintain tubal patency prior to allowing the patient to attempt pregnancy.3, 4 Alternatively, little emphasis has been given toward encouraging early resumption of coitus, due primarily to the lack of information regarding the time course of healing. It has been previously shown that normal fertility is possible in rabbits within 2 weeks of fimbriectomy5 and end-to-end anastomosis. 6 We undertook the present study to examine the morphologic characteristics of healing at various postoperative intervals following a standardized microsurgical end-to-end anastomosis performed in the mid-ampulla of the rabbit oviduct and to correlate these changes with fertility. MATERIALS AND METHODS
Sexually mature New Zealand White rabbits were used in the study. The animals were caged individually and maintained under conditions of 24° to 27° C and exposed to a fluorescent light photoperiod of 12 hours light:12 hours darkness.
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
Figure 1 Scanning electron micrograph of the mid-ampulla. Approximately half of the cells are ciliated. The remainder are secretory (original magnification, x 2850).
Free access to water and commercial rabbit chow was maintained throughout the experiment. The animals were isolated for a minimum of 3 weeks prior to surgery.
dissection into the mesosalpinx in order to sever the oviduct completely. End-to-end anastomosis was then performed with the use of 10-0 monofilament nylon suture. We placed interrupted sutures around the periphery of the tube, commencing at the mesosalpinx. The sutures were placed to include the serosa and myosalpinx but avoid the endosalpinx. We took care to replace the everted mucosa within the lumen, using a jet of saline prior to the placement and tying of each suture so that, upon completion of the anastomosis, no mucosa protruded and peritoneal continuity, interrupted by six to eight symmetrically placed sutures, was restored. The oviduct and ovary were returned to the abdominal cavity. The peritoneum was carefully repaired with interrupted 3-0 chromic suture. The flank incision was closed in two layers, with the use of interrupted 2-0 chromic suture on the fascia and continuous 3-0 polyglycolic acid suture on the skin.
MICROSURGICAL TECHNIQUE
Anesthesia was induced and maintained with intravenous (IV) sodium pentobarbital. Access to the reproductive tract was obtained through a 3-cm incision in the left flank; using an aseptic surgical technique. The left oviduct and ovary were exteriorized and supported on saline-moistened gauze sponges. Unilateral end-to-end anastomosis of the ampulla was performed-with the use of microsurgical techniques. A Zeiss OPMI 6 operating microscope (Carl Zeiss, Inc., New York, NY) fitted with a 250-mm objective lens and equipped with an auxiliary fiberoptic light source provided visualization of the reproductive tract during surgery. Under low magnification, the ampulla was examined. An area of mid-ampulla bounded on both sides by major vascular arborizations originating in the mesosalpinx was identified. Under high magnification, blood vessels perfusing the area of ampulla to be divided were doubly ligated with 10-0 monofilament nylon sutures placed immediately adjacent to the intended transection site. This generally entailed the placement of from three to five pairs of hemostatic sutures, the location of which around the periphery of the ampulla defined the plane of dissection. In this fashion, electrocoagulation and any attendant thermal trauma to the oviduct was avoided. The ampulla was severed perpendicular to its long axis with Vannas scissors. We continued the Vol. 38, No.3, September 1982
Figure 2 Transmission electron micrograph of the mid-ampulla. Ciliated cells (Cc) contain numerous kinocilia arranged along the apex, some of which have been cross-sectioned, revealing the typical arrangement of microtubules. Secretory cells (Sc) possess an abundance of secretory granules (Sg) (original magnification, x 4800).
Eddy and Bajpai Healing and fertility following anastomosis
355
JEOL JSM-35 scanning electron microscope (JEOL USA, Peabody, MA) at a power setting of 14 kv. For transmission electron microscopy (TEM), tissue was postfixed in osmium tetroxide (2% in 0.1 M cacodylate buffer) for 2 hours, washed overnight in buffer, dehydrated with sequential concentrations of alcohol, and embedded in Spurr's plastic. Thin sections were cut on an LKB 8800A Ultratome III microtome (LKB Instruments, Inc., Rockville, MD), stained with 1% uranyl acetate and 1% lead citrate, and examined with a Siemens 101 electron microscope (Siemens Corp., Iselin, NJ) at a power setting of 80 kv. FERTILITY STUDIES
Groups of five animals each were mated to a buck of proven fertility at days 1,2,3,4,7, and 14 after anastomosis. An IV injection of 100 IU hu-
Figure 3 Disruption of cytoplasmic organelles seen at time O. A ciliated cell is seen flanked between two secretory cells. The former retains a normal arrangement of cilia, but the latter are depleted of secretory material (TEM, original magnification, x 7600).
MORPHOLOGIC STUDIES
Groups of two or three animals were killed, and tissue immediately adjacent to the anastomosis was examined with the electron microscope upon completion of the anastomosis at days 0,1,2,3,4, 7, and 14. Tissues were quickly excised, and a segment of mid-ampulla approximately 1 cm long, containing the anastomosis, was resected and immediately immersed in 0.1 M sodium cacodylate buffer (pH 7.4). The tissue was trimmed of all mesenteries, cut open longitudinally, and pinned flat. Tissues were fixed overnight at room temperature in a mixture of 4% glutaraldehyde and 1% paraformaldehyde in 0.1 M cacodylate buffer. Tissue to be examined by scanning electron microscopy (SEM) was dehydrated in graded concentrations of ethyl alcohol, placed in acetone, critical-point-dried with CO 2 , coated with goldpalladium under vacuum, and examined in a 356
Figure 4 Day O. Submucosa showing a cross-section of a capillary. Polymorphonuclear leukocytes (Lu) are within the lumen, having been transported to the anastomosis. Red blood cells (RBC) are present intercellularly, collagen fibers are absent, and edema is apparent (TEM, original magnification, x 3700).
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
·. ~ ~ «11-« .. t
•
~~
,.
II
•
~
Figure 5 Ciliated cell undergoing apocrinelike deciliation on day l. Adjacent secretory cells lack secretory material and exhibit severe subcellular disruption (TEM, original magnification, x 4500).
Histopathology Control. The endosalpinx of the rabbit oviduct consisted of columnar epithelium. In the mid-ampulla portion, approximately half of the epithelial cells were ciliated, and the remainder were secretory (Fig. 1). Ciliated cells exhibited an electronlucent cytoplasm with a large, irregularly shaped, ovoid nucleus located in the apical cytoplasm (Fig. 2). Numerous polyribosomes and elongated, tubular mitochondria were present, the latter located predominantly in the supranuclear and apical portions of the cytoplasm. Little endoplasmic reticulum was present, and the Golgi complex, when visualized, was found in the supranuclear region. The apex of ciliated cells characteristically contained numerous long and erect kinocilia interspersed with fingerlike microvilli. The cilia demonstrated the typical 9 + 2 arrangement of microtubules with prominent kinetosomes just below the cell surface.
man chorionic gonadotropin (hCG) was given after mating to ensure ovulation. Two weeks later, the animals underwent exploratory laparotomy. Both uterine horns were examined for the presence of normal implantations, and these were correlated with the numbers of normal corpora lutea present in respective ovaries and were expressed as the mean ± standard error.
RESULTS MORPHOLOGY
Gross Pathology In no instance was the normal gross anatomy of the oviduct severely disturbed. No oviducts were palpably fibrosed. The location of the anastomosis was often revealed solely by the presence of the microsutures in the serosa. All anastomosed oviducts were functionally patent, as revealed by pregnancy, free passage of dye, or microdissection with the operating microscope ' during specimen preparation. Adhesions, when present, were mild and nondistorting and were restricted to the serosa and mesosalpinx at the level of the anastomosis. The endosalpinx was continuous across the anastomosis, which often could only be located by floating the epithelial folds in a jet of buffer and identifying their ends, some of which had fused with other folds or simply interdigitated with them. Vol. 38, No.3, September 1982
Figure 6 Deciliation and flattening of cells adjacent to the anastomosis site. The cobblestone topography and pleomorphism are typical of day 2 (SEM, original magnification, x 1290).
Eddy and Bajpai Healing and fertility following anastomosis
357 '
Figure 7 Extensive cellular disruption is seen in this day 2 specimen. Secretory cells (Sc) exhibit widespread vacuolization with diminished or abse~t secretory activity. Ciliated cells (Cc) devoid of cilia are seen. Apical microvilli persist. Numerous free ribosomes and centrioles (arrows) can be seen in the cytoplasm of ciliated cells, suggestive of cellular viability (TEM, original magnification, x 4600).
Interspersed among the ciliated cells were an approximately equal number of secretory cells, the nuclei of which were in size and shape similar to ciliated cells but more basally located. Electron-dense cytoplasm contained an abundance of rough endoplasmic reti~ulum, oval mitochondria, and free ribosomes dispersed throughout the cell, but particularly in the apical, infranuclear, and supranuclear regions. The Golgi complex was prominent in most cells and was located in the supranuclear region. Secretory cells characteristically exhibited the presence of numerous spherical membrane-bound secretory granules containing electron-dense material. Secretory granules tended to be most numerous in the distal portion of the cell and in the apical bulges that protruded into the lumen. The apex of each cell contained numerous microvilli (Fig. 2 ). The submucosal stroma consisted of connective tissue cells interspersed in a matrix of collagen fibers arranged in parallel bundles oriented in various directions. Time O. Oviducts examined immediately after completion of the anastomosis showed marked changes in response to the trauma of surgery. Cilia were matted and no longer erect. There was also a loss of membrane-bound secretory granules. Both ciliated and secretory cells experienced a generalized disorganization of cytoplasmic organelles, which included dilation of the endoplasmic reticulum and increased density of mi tochondrial membranes and chromatin (Fig. 3). Cross-sections of capillaries within stromal tissue 358
showed the presence of numerous neutrophils that had been transported to the damaged zone. These cells generally had not yet infiltrated the tissue but instead were still confined to the capillary lumen. Numerous red blood cells were present within the connective tissue, which showed early signs of edema, including enlarged intercell ular space and the disruption and loss of collagen fibers (Fig. 4). Day 1. Twenty-four hours after surgery, disruptive changes continued. Ciliated cells adjacent to the anastomosis either exhibited ciliary agglutination or had undergone frank deciliation. Apocrinelike loss of cilia could occasionally be seen (Fig. 5). Cells that had not undergone deciliation tended to have fewer cilia per cell. Secretory cells were totally depleted of secretory material and exhibited a lack of synthesis and storage. Both cell types showed signs of atrophy and often assumed a cuboidal shape. Marked tissue edema was seen, with massive infiltration ofneutrophils and monocytes. Day 2. The degenerative changes seen on day 1 continued on day 2. Deciliated cells and depleted secretory cells often showed a pleomorphic surface (Fig. 6). Raw areas denuded of epithelium were covered with neutrophils and fibrin networks, suggesting an early stage of granulation formation. Secretory and ciliated cells showed widespread vacuolization and continued disruption of subcellular organization, making it difficuI t for us to distinguish the two cell types (Fig. 7). Stromal edema with associated disruption of blood vessels continued.
Figure 8 Persistent patchy deciliation seen on day 4. Intact ciliated cells are present among an expanse of cobblestone-shaped cells (SEM, original magnification, x 70, inset, x 350).
Eddy and Bajpai Healing and fertility following anastomosis
.
Fertility and Sterility
Figure 9 Area undergoing early restoration of cellular function. Despite uniform surface topography, ciliated cells (Cc) and secretory cells (Sc) exhibit characteristic differences. Note active ciliogenesis and formation of centrioles (arrows) (original magnification, x 7200).
Day 3. Patchy deciliation remained common. It was noteworthy that intact isolated ciliated cells could be seen within deciliated areas and that large numbers of contiguous ciliated cells, apparently unaffected by surgical trauma, were also seen. Unlike the variability seen among ciliated cells, the absence of secretory activity was quite uniform throughout the area of the anastomosis. Day 4. By day 4 after surgery, degenerative changes appeared to have stabilized, and active tissue repair was underway. The cilia that had persisted were erect, and the epi theli um appeared less flattened, with a large number of microvilli v.isible on the cell surface (Fig. 8). Hypertrophy of the endosalpinx had started to transform the epithelium into a more normal columnar shape. Marked subcellular disorganization still existed, with the concomitant presence of noticeable intercellular space. An occasional secretory granule could be seen in some secretory cells. Day 7. By day 7, ciliogenesis and secretory activity were observed in some zones but were absent in others. In areas that had previously experienced deciliation and loss of secretory activity, it was difficult to differentiate ciliated cells from secretory cells solely on the basis of the morphologic characteristics of the cell surface. The majority of cells in such areas showed a unified appearance, which included a microvilli-covered, dome-shaped apex. Upon examination by TEM (Fig. 9), cellular differentiation was quite apparent. Many deciliated cells showed active ciliogenesis with the presence of numerous, newly synthesized basal bodies in the cell apex migrating Vol. 38, No.3, September 1982
toward the plasma membrane of the cell. The electron-lucent cytoplasm contained numerous free ribosomes, probably engaged in protein synthesis for cell repair. Elongated mitochondria were present but were fewer in number than normal. Secretory cells in such areas exhibited an electron-dense cytoplasm with numerous mitochondria and an extensive network of rough endoplasmic reticulum. Although the capacity for synthesis appeared to be present, secretory activity was generally not observed. In more normal-appearing areas, ciliated and secretory cells showed characteristic differences in surface form. Numerous ciliated cells were present, bearing cilia ranging in size from buds to mature forms protruding above the surface of adjacent microvilli-covered secretory cells (Fig. 10). Secretory cells in such areas contained numerous secretory granules: primarily in the apical cytoplasm (Fig. 11). The submucosal stroma appeared normal. The edema had resolved, vascular integrity had been restored, and fibroblastic activity had reestablished a normal complement of collagen fibers arranged singly and in bundles (Fig. 12). Day 14. By day 14, healing was complete or well under way. While it was still possible to find ciliogenesis in some areas, dense ciliation throughout the area of the anastomosis (Fig. 13), together with abundant secretory activity, was the predominant finding.
Figure 10 Area of advanced recovery seen on day 7. Numerous ciliary buds of varying lengths are seen within cells actively undergoing ciliogenesis (SEM, original magnification, x 1850).
Eddy and Bajpai Healing and fertility following anastomosis
359
The ratio of implantations to numbers of ovulations demonstrates more clearly the effect of surgery on fertility. On the transected side, only 12% of ovulated ova were fertilized and implanted when mating occurred 1 day after surgery. This rate remained depressed on day 2 but thereafter steadily improved, reaching 57% on day 7 and 64% in animals mated on day 14. The systemic effects of surgery on fertility depressed the numbers of ova that became implanted in the unoperated control side during the first 4 postoperative days. These effects were not resolved until 7 days after surgery, at which time the ratio of ova ovulated to those that became implanted attained levels consistent with normal fertility. These effects were not mediated through alteration of ovulation rates. The mean numbers of ovulations in control and anastomosed sides were 6.8 ± 0.4 and 6.5 ± 0.3, respectively. DISCUSSION
Figure 11 Similar area seen in Figure 10 viewed by TEM. Centrioles and kinetosomes (arrows) have migrated to the cell apex of a ciliated cell and are lining up beneath the cell membrane. Active synthesis is under way within adjacent secretory cells (original magnification, x 7200).
The immediate response of the oviduct to surgical trauma was a rapid release of stored secretory material and degenerative cellular changes that led ultimately to cessation of secretory activity and deciliation, but not widespread cellular death. Presumably the fact that the lesion produced by transection was highly localized and the anastomosis meticulously and accurately reopposed all tissue layers served to limit tissue destruction and promote healing. During the first 3 postoperative days, loss of cellular function and integri ty and the occurrence of tissue edema reached a maximum. By day 4, degenerative ·
FERTILITY
Table 1 presents the fertility data of animals mated at various postoperative intervals. The chronology of healing and its impact on fertility are shown. Because of the nonparametric nature of the data and the large standard errors of the mean, statistical analysis, despite transformation of the raw data, failed to reveal statistically significant differences either between or within means. However, the effects of surgery and their gradual resolution with time are biologically evident. On the anastomosed side, the pregnancy rate ranged from 40% 1 day after surgery to 100% in animals mated 14 days after anastomosis. In the unoperated, contralateral side, pregnancy rates ranged from 80% following mating on day 1 to 100% in animals mated 7 and 14 days after surgery. 360
'-....
Figure 12 Submucosal stroma, day 7. Edema and cellular infiltration have subsided. Numerous well-organized collagen fibers and bundles are present among normal-appearing connective tissue cells (TEM, original magnification, x 4500).
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
Figure 13 Dense, normal ciliation seen on day 14 (SEM, original magnification, x 2300).
changes had stabilized. Thereafter, the healing process became dominant. Resolution of tissue edema, active ciliogenesis, and resumed secretory activity were present by day 7. Two weeks after surgery, a high degree of functional integration, consistent with normal fertility, had been restored. This chronology of healing was quite rapid and distinct from that in the literature. In previous studies in which groups of rabbits underwent macrosurgical transection and repair of the uterotubal junction with the use of chronic indwelling polyethylene stents, a noticeable incidence of anatomic disturbance occurred. 7 , 8 These included marked epithelial irregularities, loss of mucosal folds, and absence of or diminution in the number of ciliated and secretory cells. These changes persisted up to 6 weeks postoperatively and improved thereafter. Concomitantly, degenerative changes in the myosalpinx were observed 2 and 6 weeks postoperatively, although normal muscle continuity was present 2 weeks after surgery.
Thereafter, progressive fibrous replacement of damaged muscle tissue occurred and was maximal at 6 months. Only three pregnancies occurred in 30 animals, all within the group in which stents had been removed 2 weeks postoperatively. This finding led the authors to suggest that widespread deciliation, loss of secretory activity, and irregular epithelium at the site of anastomosis were not critical for pregnancy. In contrast, the loss of muscle and its progressive replacement by fibrotic tissue may have interfered with contractile activity and normal gamete transport and, hence, with the establishment of pregnancy. Anastomosis was performed with the placement of two serosal sutures of 4-0 catgut. The resultant lack of opposition of the two cut ends would tend to promote fibroblastic proliferation and remodeling of the raw surfaces of unopposed cut ends with nonfunctional connective tissue. Moreover, the use of highly reactive catgut suture and prolonged use of stents would be expected to exacerbate this reaction, since their use has been shown to have a deleterious effect on tissue repair and subsequent fertility.9 In the present study, the microsurgical technique used presumably avoided or minimized these problems through the use of relatively atraumatic surgery, nonreactive microsuture, meticulous apposition of tissue layers, and judicious avoidance of stents. Healing was rapid and occurred primarily by first intention. The ability of such tissue to generate and propagate action potentials across the anastomosis site has been demonstrated 3 to 6 weeks following ampullary and isthmic transection and end-to-end anastomosis.lO This finding is in opposition to the presumed lack of such function in earlier studies. 7 , 8 In the present study, we did not observe delayed tissue repair and prolonged absence of ciliation or secretory activity, nor did we observe the
Table 1. Effects of Unilateral Transection of the Ampulla on Fertility Time after operation
No. of animals
Pregnancy rate
Fecundity"
Control side
Transected side
80 60 60 60 100 100
40 60 60 60 80 100
Control side
Transected side
days
1 2 3 4 7 14 aNumber of implants
5 5 5 5 5 5
27.6 48.4 51.4 30.0 70.6 85.7
± 9.8 ± 20.5 ± 22.4 ± 12.6 ± 16.9 ± 7.3
12.0 14.1 29.9 46.6 57.4 63.9
± 9.7 ± 6.6 ± 13.3 ± 20.6 ± 14.7 ± 7.6
x 100. Mean ± standard error.
Number of ovulations Vol. 38, No.3, September 1982
Eddy and Bajpai Healing and fertility following anastomosis
361
presence of, or healing mediated by, "indifferent" cells thought to be precursors of secretory or ciliated cells. l l There was no evidence for the transformation of ciliated cells into secretory cells. Secretory granules were occasionally seen in ciliated cells and cilia in secretory cells. Similar observations 12 ,13 have been interpreted as evidence for such interconversion. 14 While the loss of secretory activity and deciliation, together with attendant loss of subcellular organization, produced cellular forms consistent with dedifferentiation when viewed by SEM, examination of tissue by TEM always revealed retention of subcellular organelles characteristic of ciliated or secretory cells. Most affected cells appear to have retained sufficient integrity to allow healing and resumption of normal function after a relatively short period of time. Ciliogenesis was consistently observed in deciliated cells. Although ciliogenesis itself is an ordered sequential process within individual cells, it was not synchronized between adjacent cells or in different regions. It was not possible, due to lack of appropriate staining and serial sectioning, to determine whether ciliogenesis was occurring via the centriolar15 or acentriolar pathway.16 Secretory activity resumed coincident with ciliogenesis. The occurrence of pregnancy following mating at all postoperative intervals reflects the rapidity of healing and the importance of meticulous repair with precise alignment of tubal segments. Exudation of fibrinogen onto raw surfaces begins almost immediately. If tissue trauma is severe and widespread, cellular death will be extensive and will be followed by repair through fibroblastic proliferation and connective tissue formation. If tubal anatomy is not surgically restored with great accuracy, the resultant disruption may lead to loss of patency and function. Ironically, the rapidity of fibrinogen formation and resultant loss of patency following massive tissue trauma has been clinically regarded as a prime cause for failure of tuboplasty to restore fertility17 and, hence, for the historical and widespread use of stents until tissue death, fibrous connective tissue replacement, and long-term healing are completed. The results of the present study indicate that when tissue trauma is minimal and repair precise, tubal function returns rapidly and healing is not associated with extensive disruption of normal anatomy. The progressive improvement in fertility reflected the restoration of function as healing proceeded. The rapidity of that healing 362
was noteworthy in view of the trauma of transection and repair and the deleterious effects of anesthesia and the abdominal operation which manifested themselves systemically, leading to depressed fertility in contralateral unoperated sides during the first 4 postoperative days. Healing following end-to-end anastomosis of the human oviduct has not been well documented. The mean interval between microsurgical reversal of elective tubal sterilization and the occurrence of pregnancy has been reported to range from 7 18 to 19 months. 2 However, pregnancy has occurred as early as the first cycle following reconstructive surgery2 and, in some cases, during the same cycle in which surgery was performed, since menses did not occur between operation and pregnancy. 19 The present model is imperfect insofar as it did not entail resection of tissue followed by an intervening period between transection and subsequent repair. It is nevertheless analogous to clinical reversal of previous elective midtubal sterilization. In such cases, anastomosis of healthy tubal segments is performed immediately after the resecting of nonfunctional fibrotic tissue. The rapid healing, lack of adhesions, and early return of fertility seen in this study, in the absence of aggressive postoperative adjunctive therapy, is noteworthy. Clinically, delayed attempts at pregnancy may not be necessary when one is dealing with a patient with midtubal sterilization who presents with a pelvis free of adhesions and oviducts amenable to repair with minimal resection and manipulation of tissue. REFERENCES 1. Winston RML: Microsurgery of the fallopian tube: from
fantasy to reality. Fertil Steril 34:521, 1980 2. Gomel V: Microsurgical reversal offemale sterilization: a reappraisal. Fertil Steril 33:587, 1980 3. Shirodkar VN: Further experiences in tuboplasty. Aust NZ J Obstet Gynaecol 5:1, 1965 4. Roland M, Leisten D: Tuboplasty in 130 patients: improved results due to stents and preoperative endoscopy. Obstet Gynecol 39:1, 1972 5. Beyth Y, Winston RML: Ovum capture and fertility following microsurgical fimbriectomy in the rabbit. Fertil Steril 35:464, 1981 6. Perez L, Rajkumar K, Eddy CA: Fertility and ovum transport after microsurgical removal of the uterotubal junction in rabbits. Fertil Steril 36:803, 1981 7. Khoo SK, Mackay EV: Reactions in rabbit fallopian tube after plastic reconstruction. I. Gross pathology, tubal patency, and pregnancy. Fertil Steril 23:201, 1972 8. Mackay EV, Khoo SK: Reactions in the rabbit fallopian tube after plastic reconstruction. II. Histopathology. Fertil Steril 23:207, 1972
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
9. Winston RML: Microsurgical reanastomosis of the rabbit oviduct and its functional and pathological sequelae. Br J Obstet Gynaecol 82:513, 1975 10. Archer DR, Eddy CA, Pauerstein CJ: Electrophysiology of the rabbit oviduct following tubal microsurgery. Fertil Steril 31:423, 1979 11. Clyman MJ: Electronmicroscopy of human fallopian tube. Fertil Steril 17:281, 1966 12. Nilsson 0: Electronmicroscopy of the fallopian tube epithelium of rabbits in oestrus. Exp Cell Res 14:341, 1958 13. Merchant H: Secretory granules in ciliated cells of the rabbit oviduct. Exp Cell Res 56:171, 1969 14. Martinek J, Kraus R, Jirsova Z: Cytology of tubal epithelium. Folia Morphol (Prahal 15:241, 1967
Vol. 38, No.3, September 1982
15. Dirksen ER: Centriole morphogenesis in developing ciliated epithelium of the mouse oviduct. J Cell Bioi 51:286, 1971 16. Brenner RM: Ciliogenesis during the menstrual cycle in rhesus monkey oviduct. J Cell Bioi 35:16a, 1967 17. Weisman AI: Rapidity of human fallopian tube wound healing as cause for poor results from reparative tubal surgery. Am J Surg 82:278, 1951 18. Gomel V: Microsurgery for reversal of female sterilization. In Reversal of Sterilization, Edited by JJ Sciarra, GI Zatuchni, JJ Speidel. Hagerstown, MD, Harper & Row, 1978, p 195 19. Winston RML: Unpublished data
Eddy and Bajpai Healing and fertility following anastomosis
363
FERTILITY AND STERILITY Copyright c 1982 The American Fertility Society
Healing and restoration of fertility following microsurgical transection and anastomosis of the rabbit oviduct i
Carlton A. Eddy, Ph.D.* Virendraa K. Bajpai, Ph.D. t Department of Obstetrics and Gynecology, The University of Texas Health Science Center at San Antonio, San Antonio, Texas
The postoperative time course of healing following tubal microsurgery was examined in the rabbit by transmission electron microscopy (TEM) and scanning electron microscopy (SEM) and was correlated with changes in fertility. Groups of animals were bred or used for morphologic studies on days 1,2,3,4, 7, and 14 after surgery. During the first 4 postoperative days, a major loss of ciliary and secretory activity occurred and was characterized by disruption of subcellular organelles. Tissuedamage was restricted to within several millimeters of the anastomosis. Healing was well advanced by day 7 and was largely complete by day 14. Corresponding changes in fertility were observed. A significant decline in fertility was noted on the first postoperative day. Thereafter, fertility progressively improved and was normal by day 14. Fertil Steril 38:354, 1982
Tuboplastic microsurgery has become a widely accepted technique for the restoration offertility, ideally in women who have undergone previous elective tubal sterilization and who subsequently seek its surgical reversal. Among such patients, the success rate may approach 80%.1, 2 Despite a great deal of interest in improving the surgical technique itself, little is known regarding the time course of healing. As a consequence, no standardized scheme of postoperative management has evolved either with regard to appropriate medication or to when pregnancy should be attempted. Such decisions are at present largely subjective. At one extreme are those who favor prolonged use of intratubal stents and adjunctive therapy to
Received February 22, 1982; revised and accepted May 18; 1982. *Reprint requests: Dr. C. A. Eddy, Department of Obstetrics and Gynecology, The University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, Texas 78284. tWorld Health Organization Postdoctoral Fellow. 354
promote healing, avoid adhesion formation, and maintain tubal patency prior to allowing the patient to attempt pregnancy.3, 4 Alternatively, little emphasis has been given toward encouraging early resumption of coitus, due primarily to the lack of information regarding the time course of healing. It has been previously shown that normal fertility is possible in rabbits within 2 weeks of fimbriectomy5 and end-to-end anastomosis. 6 We undertook the present study to examine the morphologic characteristics of healing at various postoperative intervals following a standardized microsurgical end-to-end anastomosis performed in the mid-ampulla of the rabbit oviduct and to correlate these changes with fertility. MATERIALS AND METHODS
Sexually mature New Zealand White rabbits were used in the study. The animals were caged individually and maintained under conditions of 24° to 27° C and exposed to a fluorescent light photoperiod of 12 hours light:12 hours darkness.
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
Figure 1 Scanning electron micrograph of the mid-ampulla. Approximately half of the cells are ciliated. The remainder are secretory (original magnification, x 2850).
Free access to water and commercial rabbit chow was maintained throughout the experiment. The animals were isolated for a minimum of 3 weeks prior to surgery.
dissection into the mesosalpinx in order to sever the oviduct completely. End-to-end anastomosis was then performed with the use of 10-0 monofilament nylon suture. We placed interrupted sutures around the periphery of the tube, commencing at the mesosalpinx. The sutures were placed to include the serosa and myosalpinx but avoid the endosalpinx. We took care to replace the everted mucosa within the lumen, using a jet of saline prior to the placement and tying of each suture so that, upon completion of the anastomosis, no mucosa protruded and peritoneal continuity, interrupted by six to eight symmetrically placed sutures, was restored. The oviduct and ovary were returned to the abdominal cavity. The peritoneum was carefully repaired with interrupted 3-0 chromic suture. The flank incision was closed in two layers, with the use of interrupted 2-0 chromic suture on the fascia and continuous 3-0 polyglycolic acid suture on the skin.
MICROSURGICAL TECHNIQUE
Anesthesia was induced and maintained with intravenous (IV) sodium pentobarbital. Access to the reproductive tract was obtained through a 3-cm incision in the left flank; using an aseptic surgical technique. The left oviduct and ovary were exteriorized and supported on saline-moistened gauze sponges. Unilateral end-to-end anastomosis of the ampulla was performed-with the use of microsurgical techniques. A Zeiss OPMI 6 operating microscope (Carl Zeiss, Inc., New York, NY) fitted with a 250-mm objective lens and equipped with an auxiliary fiberoptic light source provided visualization of the reproductive tract during surgery. Under low magnification, the ampulla was examined. An area of mid-ampulla bounded on both sides by major vascular arborizations originating in the mesosalpinx was identified. Under high magnification, blood vessels perfusing the area of ampulla to be divided were doubly ligated with 10-0 monofilament nylon sutures placed immediately adjacent to the intended transection site. This generally entailed the placement of from three to five pairs of hemostatic sutures, the location of which around the periphery of the ampulla defined the plane of dissection. In this fashion, electrocoagulation and any attendant thermal trauma to the oviduct was avoided. The ampulla was severed perpendicular to its long axis with Vannas scissors. We continued the Vol. 38, No.3, September 1982
Figure 2 Transmission electron micrograph of the mid-ampulla. Ciliated cells (Cc) contain numerous kinocilia arranged along the apex, some of which have been cross-sectioned, revealing the typical arrangement of microtubules. Secretory cells (Sc) possess an abundance of secretory granules (Sg) (original magnification, x 4800).
Eddy and Bajpai Healing and fertility following anastomosis
355
JEOL JSM-35 scanning electron microscope (JEOL USA, Peabody, MA) at a power setting of 14 kv. For transmission electron microscopy (TEM), tissue was postfixed in osmium tetroxide (2% in 0.1 M cacodylate buffer) for 2 hours, washed overnight in buffer, dehydrated with sequential concentrations of alcohol, and embedded in Spurr's plastic. Thin sections were cut on an LKB 8800A Ultratome III microtome (LKB Instruments, Inc., Rockville, MD), stained with 1% uranyl acetate and 1% lead citrate, and examined with a Siemens 101 electron microscope (Siemens Corp., Iselin, NJ) at a power setting of 80 kv. FERTILITY STUDIES
Groups of five animals each were mated to a buck of proven fertility at days 1,2,3,4,7, and 14 after anastomosis. An IV injection of 100 IU hu-
Figure 3 Disruption of cytoplasmic organelles seen at time O. A ciliated cell is seen flanked between two secretory cells. The former retains a normal arrangement of cilia, but the latter are depleted of secretory material (TEM, original magnification, x 7600).
MORPHOLOGIC STUDIES
Groups of two or three animals were killed, and tissue immediately adjacent to the anastomosis was examined with the electron microscope upon completion of the anastomosis at days 0,1,2,3,4, 7, and 14. Tissues were quickly excised, and a segment of mid-ampulla approximately 1 cm long, containing the anastomosis, was resected and immediately immersed in 0.1 M sodium cacodylate buffer (pH 7.4). The tissue was trimmed of all mesenteries, cut open longitudinally, and pinned flat. Tissues were fixed overnight at room temperature in a mixture of 4% glutaraldehyde and 1% paraformaldehyde in 0.1 M cacodylate buffer. Tissue to be examined by scanning electron microscopy (SEM) was dehydrated in graded concentrations of ethyl alcohol, placed in acetone, critical-point-dried with CO 2 , coated with goldpalladium under vacuum, and examined in a 356
Figure 4 Day O. Submucosa showing a cross-section of a capillary. Polymorphonuclear leukocytes (Lu) are within the lumen, having been transported to the anastomosis. Red blood cells (RBC) are present intercellularly, collagen fibers are absent, and edema is apparent (TEM, original magnification, x 3700).
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
·. ~ ~ «11-« .. t
•
~~
,.
II
•
~
Figure 5 Ciliated cell undergoing apocrinelike deciliation on day l. Adjacent secretory cells lack secretory material and exhibit severe subcellular disruption (TEM, original magnification, x 4500).
Histopathology Control. The endosalpinx of the rabbit oviduct consisted of columnar epithelium. In the mid-ampulla portion, approximately half of the epithelial cells were ciliated, and the remainder were secretory (Fig. 1). Ciliated cells exhibited an electronlucent cytoplasm with a large, irregularly shaped, ovoid nucleus located in the apical cytoplasm (Fig. 2). Numerous polyribosomes and elongated, tubular mitochondria were present, the latter located predominantly in the supranuclear and apical portions of the cytoplasm. Little endoplasmic reticulum was present, and the Golgi complex, when visualized, was found in the supranuclear region. The apex of ciliated cells characteristically contained numerous long and erect kinocilia interspersed with fingerlike microvilli. The cilia demonstrated the typical 9 + 2 arrangement of microtubules with prominent kinetosomes just below the cell surface.
man chorionic gonadotropin (hCG) was given after mating to ensure ovulation. Two weeks later, the animals underwent exploratory laparotomy. Both uterine horns were examined for the presence of normal implantations, and these were correlated with the numbers of normal corpora lutea present in respective ovaries and were expressed as the mean ± standard error.
RESULTS MORPHOLOGY
Gross Pathology In no instance was the normal gross anatomy of the oviduct severely disturbed. No oviducts were palpably fibrosed. The location of the anastomosis was often revealed solely by the presence of the microsutures in the serosa. All anastomosed oviducts were functionally patent, as revealed by pregnancy, free passage of dye, or microdissection with the operating microscope ' during specimen preparation. Adhesions, when present, were mild and nondistorting and were restricted to the serosa and mesosalpinx at the level of the anastomosis. The endosalpinx was continuous across the anastomosis, which often could only be located by floating the epithelial folds in a jet of buffer and identifying their ends, some of which had fused with other folds or simply interdigitated with them. Vol. 38, No.3, September 1982
Figure 6 Deciliation and flattening of cells adjacent to the anastomosis site. The cobblestone topography and pleomorphism are typical of day 2 (SEM, original magnification, x 1290).
Eddy and Bajpai Healing and fertility following anastomosis
357 '
Figure 7 Extensive cellular disruption is seen in this day 2 specimen. Secretory cells (Sc) exhibit widespread vacuolization with diminished or abse~t secretory activity. Ciliated cells (Cc) devoid of cilia are seen. Apical microvilli persist. Numerous free ribosomes and centrioles (arrows) can be seen in the cytoplasm of ciliated cells, suggestive of cellular viability (TEM, original magnification, x 4600).
Interspersed among the ciliated cells were an approximately equal number of secretory cells, the nuclei of which were in size and shape similar to ciliated cells but more basally located. Electron-dense cytoplasm contained an abundance of rough endoplasmic reti~ulum, oval mitochondria, and free ribosomes dispersed throughout the cell, but particularly in the apical, infranuclear, and supranuclear regions. The Golgi complex was prominent in most cells and was located in the supranuclear region. Secretory cells characteristically exhibited the presence of numerous spherical membrane-bound secretory granules containing electron-dense material. Secretory granules tended to be most numerous in the distal portion of the cell and in the apical bulges that protruded into the lumen. The apex of each cell contained numerous microvilli (Fig. 2 ). The submucosal stroma consisted of connective tissue cells interspersed in a matrix of collagen fibers arranged in parallel bundles oriented in various directions. Time O. Oviducts examined immediately after completion of the anastomosis showed marked changes in response to the trauma of surgery. Cilia were matted and no longer erect. There was also a loss of membrane-bound secretory granules. Both ciliated and secretory cells experienced a generalized disorganization of cytoplasmic organelles, which included dilation of the endoplasmic reticulum and increased density of mi tochondrial membranes and chromatin (Fig. 3). Cross-sections of capillaries within stromal tissue 358
showed the presence of numerous neutrophils that had been transported to the damaged zone. These cells generally had not yet infiltrated the tissue but instead were still confined to the capillary lumen. Numerous red blood cells were present within the connective tissue, which showed early signs of edema, including enlarged intercell ular space and the disruption and loss of collagen fibers (Fig. 4). Day 1. Twenty-four hours after surgery, disruptive changes continued. Ciliated cells adjacent to the anastomosis either exhibited ciliary agglutination or had undergone frank deciliation. Apocrinelike loss of cilia could occasionally be seen (Fig. 5). Cells that had not undergone deciliation tended to have fewer cilia per cell. Secretory cells were totally depleted of secretory material and exhibited a lack of synthesis and storage. Both cell types showed signs of atrophy and often assumed a cuboidal shape. Marked tissue edema was seen, with massive infiltration ofneutrophils and monocytes. Day 2. The degenerative changes seen on day 1 continued on day 2. Deciliated cells and depleted secretory cells often showed a pleomorphic surface (Fig. 6). Raw areas denuded of epithelium were covered with neutrophils and fibrin networks, suggesting an early stage of granulation formation. Secretory and ciliated cells showed widespread vacuolization and continued disruption of subcellular organization, making it difficuI t for us to distinguish the two cell types (Fig. 7). Stromal edema with associated disruption of blood vessels continued.
Figure 8 Persistent patchy deciliation seen on day 4. Intact ciliated cells are present among an expanse of cobblestone-shaped cells (SEM, original magnification, x 70, inset, x 350).
Eddy and Bajpai Healing and fertility following anastomosis
.
Fertility and Sterility
Figure 9 Area undergoing early restoration of cellular function. Despite uniform surface topography, ciliated cells (Cc) and secretory cells (Sc) exhibit characteristic differences. Note active ciliogenesis and formation of centrioles (arrows) (original magnification, x 7200).
Day 3. Patchy deciliation remained common. It was noteworthy that intact isolated ciliated cells could be seen within deciliated areas and that large numbers of contiguous ciliated cells, apparently unaffected by surgical trauma, were also seen. Unlike the variability seen among ciliated cells, the absence of secretory activity was quite uniform throughout the area of the anastomosis. Day 4. By day 4 after surgery, degenerative changes appeared to have stabilized, and active tissue repair was underway. The cilia that had persisted were erect, and the epi theli um appeared less flattened, with a large number of microvilli v.isible on the cell surface (Fig. 8). Hypertrophy of the endosalpinx had started to transform the epithelium into a more normal columnar shape. Marked subcellular disorganization still existed, with the concomitant presence of noticeable intercellular space. An occasional secretory granule could be seen in some secretory cells. Day 7. By day 7, ciliogenesis and secretory activity were observed in some zones but were absent in others. In areas that had previously experienced deciliation and loss of secretory activity, it was difficult to differentiate ciliated cells from secretory cells solely on the basis of the morphologic characteristics of the cell surface. The majority of cells in such areas showed a unified appearance, which included a microvilli-covered, dome-shaped apex. Upon examination by TEM (Fig. 9), cellular differentiation was quite apparent. Many deciliated cells showed active ciliogenesis with the presence of numerous, newly synthesized basal bodies in the cell apex migrating Vol. 38, No.3, September 1982
toward the plasma membrane of the cell. The electron-lucent cytoplasm contained numerous free ribosomes, probably engaged in protein synthesis for cell repair. Elongated mitochondria were present but were fewer in number than normal. Secretory cells in such areas exhibited an electron-dense cytoplasm with numerous mitochondria and an extensive network of rough endoplasmic reticulum. Although the capacity for synthesis appeared to be present, secretory activity was generally not observed. In more normal-appearing areas, ciliated and secretory cells showed characteristic differences in surface form. Numerous ciliated cells were present, bearing cilia ranging in size from buds to mature forms protruding above the surface of adjacent microvilli-covered secretory cells (Fig. 10). Secretory cells in such areas contained numerous secretory granules: primarily in the apical cytoplasm (Fig. 11). The submucosal stroma appeared normal. The edema had resolved, vascular integrity had been restored, and fibroblastic activity had reestablished a normal complement of collagen fibers arranged singly and in bundles (Fig. 12). Day 14. By day 14, healing was complete or well under way. While it was still possible to find ciliogenesis in some areas, dense ciliation throughout the area of the anastomosis (Fig. 13), together with abundant secretory activity, was the predominant finding.
Figure 10 Area of advanced recovery seen on day 7. Numerous ciliary buds of varying lengths are seen within cells actively undergoing ciliogenesis (SEM, original magnification, x 1850).
Eddy and Bajpai Healing and fertility following anastomosis
359
The ratio of implantations to numbers of ovulations demonstrates more clearly the effect of surgery on fertility. On the transected side, only 12% of ovulated ova were fertilized and implanted when mating occurred 1 day after surgery. This rate remained depressed on day 2 but thereafter steadily improved, reaching 57% on day 7 and 64% in animals mated on day 14. The systemic effects of surgery on fertility depressed the numbers of ova that became implanted in the unoperated control side during the first 4 postoperative days. These effects were not resolved until 7 days after surgery, at which time the ratio of ova ovulated to those that became implanted attained levels consistent with normal fertility. These effects were not mediated through alteration of ovulation rates. The mean numbers of ovulations in control and anastomosed sides were 6.8 ± 0.4 and 6.5 ± 0.3, respectively. DISCUSSION
Figure 11 Similar area seen in Figure 10 viewed by TEM. Centrioles and kinetosomes (arrows) have migrated to the cell apex of a ciliated cell and are lining up beneath the cell membrane. Active synthesis is under way within adjacent secretory cells (original magnification, x 7200).
The immediate response of the oviduct to surgical trauma was a rapid release of stored secretory material and degenerative cellular changes that led ultimately to cessation of secretory activity and deciliation, but not widespread cellular death. Presumably the fact that the lesion produced by transection was highly localized and the anastomosis meticulously and accurately reopposed all tissue layers served to limit tissue destruction and promote healing. During the first 3 postoperative days, loss of cellular function and integri ty and the occurrence of tissue edema reached a maximum. By day 4, degenerative ·
FERTILITY
Table 1 presents the fertility data of animals mated at various postoperative intervals. The chronology of healing and its impact on fertility are shown. Because of the nonparametric nature of the data and the large standard errors of the mean, statistical analysis, despite transformation of the raw data, failed to reveal statistically significant differences either between or within means. However, the effects of surgery and their gradual resolution with time are biologically evident. On the anastomosed side, the pregnancy rate ranged from 40% 1 day after surgery to 100% in animals mated 14 days after anastomosis. In the unoperated, contralateral side, pregnancy rates ranged from 80% following mating on day 1 to 100% in animals mated 7 and 14 days after surgery. 360
'-....
Figure 12 Submucosal stroma, day 7. Edema and cellular infiltration have subsided. Numerous well-organized collagen fibers and bundles are present among normal-appearing connective tissue cells (TEM, original magnification, x 4500).
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
Figure 13 Dense, normal ciliation seen on day 14 (SEM, original magnification, x 2300).
changes had stabilized. Thereafter, the healing process became dominant. Resolution of tissue edema, active ciliogenesis, and resumed secretory activity were present by day 7. Two weeks after surgery, a high degree of functional integration, consistent with normal fertility, had been restored. This chronology of healing was quite rapid and distinct from that in the literature. In previous studies in which groups of rabbits underwent macrosurgical transection and repair of the uterotubal junction with the use of chronic indwelling polyethylene stents, a noticeable incidence of anatomic disturbance occurred. 7 , 8 These included marked epithelial irregularities, loss of mucosal folds, and absence of or diminution in the number of ciliated and secretory cells. These changes persisted up to 6 weeks postoperatively and improved thereafter. Concomitantly, degenerative changes in the myosalpinx were observed 2 and 6 weeks postoperatively, although normal muscle continuity was present 2 weeks after surgery.
Thereafter, progressive fibrous replacement of damaged muscle tissue occurred and was maximal at 6 months. Only three pregnancies occurred in 30 animals, all within the group in which stents had been removed 2 weeks postoperatively. This finding led the authors to suggest that widespread deciliation, loss of secretory activity, and irregular epithelium at the site of anastomosis were not critical for pregnancy. In contrast, the loss of muscle and its progressive replacement by fibrotic tissue may have interfered with contractile activity and normal gamete transport and, hence, with the establishment of pregnancy. Anastomosis was performed with the placement of two serosal sutures of 4-0 catgut. The resultant lack of opposition of the two cut ends would tend to promote fibroblastic proliferation and remodeling of the raw surfaces of unopposed cut ends with nonfunctional connective tissue. Moreover, the use of highly reactive catgut suture and prolonged use of stents would be expected to exacerbate this reaction, since their use has been shown to have a deleterious effect on tissue repair and subsequent fertility.9 In the present study, the microsurgical technique used presumably avoided or minimized these problems through the use of relatively atraumatic surgery, nonreactive microsuture, meticulous apposition of tissue layers, and judicious avoidance of stents. Healing was rapid and occurred primarily by first intention. The ability of such tissue to generate and propagate action potentials across the anastomosis site has been demonstrated 3 to 6 weeks following ampullary and isthmic transection and end-to-end anastomosis.lO This finding is in opposition to the presumed lack of such function in earlier studies. 7 , 8 In the present study, we did not observe delayed tissue repair and prolonged absence of ciliation or secretory activity, nor did we observe the
Table 1. Effects of Unilateral Transection of the Ampulla on Fertility Time after operation
No. of animals
Pregnancy rate
Fecundity"
Control side
Transected side
80 60 60 60 100 100
40 60 60 60 80 100
Control side
Transected side
days
1 2 3 4 7 14 aNumber of implants
5 5 5 5 5 5
27.6 48.4 51.4 30.0 70.6 85.7
± 9.8 ± 20.5 ± 22.4 ± 12.6 ± 16.9 ± 7.3
12.0 14.1 29.9 46.6 57.4 63.9
± 9.7 ± 6.6 ± 13.3 ± 20.6 ± 14.7 ± 7.6
x 100. Mean ± standard error.
Number of ovulations Vol. 38, No.3, September 1982
Eddy and Bajpai Healing and fertility following anastomosis
361
presence of, or healing mediated by, "indifferent" cells thought to be precursors of secretory or ciliated cells. l l There was no evidence for the transformation of ciliated cells into secretory cells. Secretory granules were occasionally seen in ciliated cells and cilia in secretory cells. Similar observations 12 ,13 have been interpreted as evidence for such interconversion. 14 While the loss of secretory activity and deciliation, together with attendant loss of subcellular organization, produced cellular forms consistent with dedifferentiation when viewed by SEM, examination of tissue by TEM always revealed retention of subcellular organelles characteristic of ciliated or secretory cells. Most affected cells appear to have retained sufficient integrity to allow healing and resumption of normal function after a relatively short period of time. Ciliogenesis was consistently observed in deciliated cells. Although ciliogenesis itself is an ordered sequential process within individual cells, it was not synchronized between adjacent cells or in different regions. It was not possible, due to lack of appropriate staining and serial sectioning, to determine whether ciliogenesis was occurring via the centriolar15 or acentriolar pathway.16 Secretory activity resumed coincident with ciliogenesis. The occurrence of pregnancy following mating at all postoperative intervals reflects the rapidity of healing and the importance of meticulous repair with precise alignment of tubal segments. Exudation of fibrinogen onto raw surfaces begins almost immediately. If tissue trauma is severe and widespread, cellular death will be extensive and will be followed by repair through fibroblastic proliferation and connective tissue formation. If tubal anatomy is not surgically restored with great accuracy, the resultant disruption may lead to loss of patency and function. Ironically, the rapidity of fibrinogen formation and resultant loss of patency following massive tissue trauma has been clinically regarded as a prime cause for failure of tuboplasty to restore fertility17 and, hence, for the historical and widespread use of stents until tissue death, fibrous connective tissue replacement, and long-term healing are completed. The results of the present study indicate that when tissue trauma is minimal and repair precise, tubal function returns rapidly and healing is not associated with extensive disruption of normal anatomy. The progressive improvement in fertility reflected the restoration of function as healing proceeded. The rapidity of that healing 362
was noteworthy in view of the trauma of transection and repair and the deleterious effects of anesthesia and the abdominal operation which manifested themselves systemically, leading to depressed fertility in contralateral unoperated sides during the first 4 postoperative days. Healing following end-to-end anastomosis of the human oviduct has not been well documented. The mean interval between microsurgical reversal of elective tubal sterilization and the occurrence of pregnancy has been reported to range from 7 18 to 19 months. 2 However, pregnancy has occurred as early as the first cycle following reconstructive surgery2 and, in some cases, during the same cycle in which surgery was performed, since menses did not occur between operation and pregnancy. 19 The present model is imperfect insofar as it did not entail resection of tissue followed by an intervening period between transection and subsequent repair. It is nevertheless analogous to clinical reversal of previous elective midtubal sterilization. In such cases, anastomosis of healthy tubal segments is performed immediately after the resecting of nonfunctional fibrotic tissue. The rapid healing, lack of adhesions, and early return of fertility seen in this study, in the absence of aggressive postoperative adjunctive therapy, is noteworthy. Clinically, delayed attempts at pregnancy may not be necessary when one is dealing with a patient with midtubal sterilization who presents with a pelvis free of adhesions and oviducts amenable to repair with minimal resection and manipulation of tissue. REFERENCES 1. Winston RML: Microsurgery of the fallopian tube: from
fantasy to reality. Fertil Steril 34:521, 1980 2. Gomel V: Microsurgical reversal offemale sterilization: a reappraisal. Fertil Steril 33:587, 1980 3. Shirodkar VN: Further experiences in tuboplasty. Aust NZ J Obstet Gynaecol 5:1, 1965 4. Roland M, Leisten D: Tuboplasty in 130 patients: improved results due to stents and preoperative endoscopy. Obstet Gynecol 39:1, 1972 5. Beyth Y, Winston RML: Ovum capture and fertility following microsurgical fimbriectomy in the rabbit. Fertil Steril 35:464, 1981 6. Perez L, Rajkumar K, Eddy CA: Fertility and ovum transport after microsurgical removal of the uterotubal junction in rabbits. Fertil Steril 36:803, 1981 7. Khoo SK, Mackay EV: Reactions in rabbit fallopian tube after plastic reconstruction. I. Gross pathology, tubal patency, and pregnancy. Fertil Steril 23:201, 1972 8. Mackay EV, Khoo SK: Reactions in the rabbit fallopian tube after plastic reconstruction. II. Histopathology. Fertil Steril 23:207, 1972
Eddy and Bajpai Healing and fertility following anastomosis
Fertility and Sterility
9. Winston RML: Microsurgical reanastomosis of the rabbit oviduct and its functional and pathological sequelae. Br J Obstet Gynaecol 82:513, 1975 10. Archer DR, Eddy CA, Pauerstein CJ: Electrophysiology of the rabbit oviduct following tubal microsurgery. Fertil Steril 31:423, 1979 11. Clyman MJ: Electronmicroscopy of human fallopian tube. Fertil Steril 17:281, 1966 12. Nilsson 0: Electronmicroscopy of the fallopian tube epithelium of rabbits in oestrus. Exp Cell Res 14:341, 1958 13. Merchant H: Secretory granules in ciliated cells of the rabbit oviduct. Exp Cell Res 56:171, 1969 14. Martinek J, Kraus R, Jirsova Z: Cytology of tubal epithelium. Folia Morphol (Prahal 15:241, 1967
Vol. 38, No.3, September 1982
15. Dirksen ER: Centriole morphogenesis in developing ciliated epithelium of the mouse oviduct. J Cell Bioi 51:286, 1971 16. Brenner RM: Ciliogenesis during the menstrual cycle in rhesus monkey oviduct. J Cell Bioi 35:16a, 1967 17. Weisman AI: Rapidity of human fallopian tube wound healing as cause for poor results from reparative tubal surgery. Am J Surg 82:278, 1951 18. Gomel V: Microsurgery for reversal of female sterilization. In Reversal of Sterilization, Edited by JJ Sciarra, GI Zatuchni, JJ Speidel. Hagerstown, MD, Harper & Row, 1978, p 195 19. Winston RML: Unpublished data
Eddy and Bajpai Healing and fertility following anastomosis
363