- Email: [email protected]
Use of Ethylene Vinyl Alcohol Copolymer for Tubal Sterilization by Selective Catheterization in Rabbits
Use of Ethylene Vinyl Alcohol Copolymer for Tubal Sterilization by Selective Catheterization in Rabbits Nitamar Abdala, MD, PhD, Abraham Levitin, MD, Andrea Dawson, MD, Romualdo Maffra, Jr, MD, Hector Munoz-Ramirez, DVM, PhD, Karen Godec, RDMS, RVT, and Bart L. Dolmatch, MD
PURPOSE: To assess the efficacy of ethylene vinyl alcohol copolymer (Uryx) in nonsurgically occluding the fallopian tube and achieving tubal sterilization in the rabbit model. MATERIALS AND METHODS: Ten mature virgin female New England rabbits underwent transvaginal selective bilateral fallopian tube cannulation with use of a coaxial catheter system under general anesthesia. Selective salpingography was performed bilaterally to assess patency of the fallopian tubes. Ethylene vinyl alcohol copolymer was injected unilaterally through a microcatheter to completely fill the middle portion of the tube. Three to seven days after injection, each animal was bred. Conception was determined by ultrasonography (US) 7–19 days after effective breeding. If pregnant, the rabbit was killed. Otherwise, it was permitted to rebreed until pregnancy was achieved. Histologic specimens of the fallopian tubes were prepared and analyzed. RESULTS: Patency of the fallopian tubes was demonstrated bilaterally in all animals by the free spillage of contrast material into the peritoneum. The delivery of ethylene vinyl alcohol copolymer into the fallopian tubes was successful in all animals but one, in which most of the plug almost immediately extruded into the uterus. Pregnancy was detected by US in the untreated fallopian tube in the nine rabbits that were receptive to breeding. No pregnancies were detected in the injected side. Histologic analysis demonstrated variable degrees of occlusion, fibrosis, and inflammation, with the majority of specimens demonstrating mild to moderate inflammation and moderate to marked fibrosis. CONCLUSION: Ethylene vinyl alcohol copolymer can reliably be placed nonsurgically via the transvaginal approach into the fallopian tubes with use of a coaxial catheter system. Ethylene vinyl alcohol copolymer appears to result in less fibrosis than previously investigated agents and demonstrates a 100% early sterilization rate in the rabbit model. Index terms:
Alcohol • Fallopian tubes, occlusion • Sterlization
J Vasc Interv Radiol 2001; 12:979 –984 Abbreviations: DMSO ⫽ dimethyl sulfoxide, MCA ⫽ methyl cyanoacrylate
THE most widely used technique of voluntary sterilization in women is tubal ligation via laparoscopy or minilaparotomy. However, these procedures require surgical incision and
occasionally require general anesthesia, along with their adherent risks. In addition, surgical techniques are typically more expensive than less invasive nonsurgical methods. Therefore,
From the Department of Radiology (N.A.), University of Sao Paulo, Sao Paulo; Hospital Santa Helena (R.M.), Goias, Brazil; Departments of Vascular and Interventional Radiology (A.L.), Anatomic Pathology (A.D.), Radiology (K.G.), and Biological Resources Unit, Central Core (H.M.R.), Cleveland Clinic Foundation, Cleveland, Ohio; and Department of Radiology (B.L.D.), University of Texas Southwestern Medical Center, Dallas, Texas. Re-
ceived November 1, 2000; revision requested January 3, 2001; revision received March 28; accepted March 9. From the 2000 SCVIR Annual Meeting. Address correspondence to A.L., Department of Radiology, University of Maryland Medical System, 22 S. Greene St., Baltimore, MD 21201-1595; E-mail: [email protected] © SCVIR, 2001
considerable effort is being devoted to the development of nonsurgical methods of tubal occlusion by means of selective fallopian tube catheterization. This technique requires the use of a suitable mechanical or chemical occlusive agent. Some mechanical devices that have been used with this technique include coils (1), microspindles (2), silicon plugs (3), and hydrogel (4). These devices are inert and occlude or stretch the lumen of the tube, obstructing ovum or sperm passage. Chemical agents include ethanol, tetracycline, quinacrine hydrochloride, and sclerosing agents (5– 8).
979
980
•
Ethylene Vinyl Alcohol Copolymer for Sterilization in Rabbits
Most noninvasive techniques have major drawbacks in that they do not produce satisfactory contraception and/or produce serious side effects. The recent exception is the use of methyl cyanoacrylate (MCA), a sclerosing agent, with which Berkey et al (7) demonstrated a 100% nonpregnancy rate and the absence of significant complications. However, their study demonstrated the development of fallopian tube cysts as well as significant fibrosis and inflammation on histologic analysis. Given the results achieved by Berkey et al (7) and others, it is readily apparent that tubal sterilization is a promising technique and that new material and formulations should be tested to find the one that is most efficient and safe. In this study, we used ethylene vinyl alcohol copolymer (Uryx; Genyx Medical, Aliso Viejo, CA) to plug one of the fallopian tubes in the rabbit model. The rabbit model was chosen because its uterus contains two completely separate symmetric uterine horns and cervices. This enabled us to perform our experiment on one side, leaving the other side as the control. Ethylene vinyl alcohol copolymer is a biocompatible polymer composed of a random mixture of ethylene (hydrophobic) and vinyl alcohol (hydrophilic). It was dissolved in an 8% concentration of anhydrous dimethyl sulfoxide (DMSO) (9). Thirty percent micronized tantalum powder was added to the polymer/solvent mixture to obtain appropriate radiopacity. When this mixture comes into contact with aqueous media, DMSO rapidly diffuses away from the mixture, causing in situ precipitation and solidification of the polymer, with formation of a spongelike plug that is capable of sealing cavities (9). The material can be deployed in a highly controllable manner with use of fluoroscopic guidance. Our goal in this study was to determine the efficacy of ethylene vinyl alcohol copolymer to occlude the fallopian tube and achieve tubal sterilization in the rabbit model.
MATERIALS AND METHODS The Animal Care Protocol was approved by the Institutional Animal Care Committee and complied with the Principles of Laboratory Animal
Care (formulated by the National Society of Medical Research) and the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 80 –23, revised 1985). Twelve mature virgin female and two mature male New Zealand rabbits with an average weight of approximately 3.5 kg were used in the study. The males were used solely for breeding purposes. The females were anesthetized with lidocaine hydrochloride 5 mg/kg and ketamine 35 mg/kg. The vulvar area was scrubbed with an iodine-based cleanser. With use of the technique described by Thurmond et al (10), an 18-F Foley catheter (Sherwood, St. Louis, MO) with a hole punched in its tip was reinforced with an 8.5-F, thick-walled radiopaque Teflon catheter (Cook, Bloomington, IN) and introduced into the vagina. Its balloon was inflated to approximately 10 mL with normal saline solution and a moderate amount of contrast medium (Ioxithalamate; Telebrix Laboratories/ Guerbet; Aulnay-sous-Bois, France) and air were injected to distend and visualize the area of the two cervices. Subsequently, a 5-F Multipurpose catheter (Medi-tech/Boston Scientific, Watertown, MA) was placed into one of the cervices and, with the help of a hydrophilic 0.035-inch Glidewire (Medi-tech/Boston Scientific), advanced to the point of the fallopian tube orifice. A 2.2-F dimethyl sulfoxide (DMSO)-compatible (DMSO is a solvent which is not compatible with many catheters) microcatheter (Easy Rider; MTI Microtherapeutics, Irvine, CA) and a 0.010-inch micro– guide wire (Guidant ACS, Temecula, CA) were advanced to the tip of the Multipurpose catheter. The micro– guide wire, and then the microcatheter, were advanced into the proximal portion of the fallopian tube. To check patency of the fallopian tube, ioxithalamate was injected into the fallopian tube to the point of spillage into the peritoneal cavity. This procedure was repeated for the contralateral fallopian tube. After confirmation of tube patency, the microcatheter was advanced into the isthmic portion of one of the tubes with the aid of the micro– guide wire. One half-milliliter of DMSO was injected to flush the catheter (ethylene vinyl alcohol copolymer solidifies in the presence of water). This was followed by 0.2– 0.5
August 2001
JVIR
mL of ethylene vinyl alcohol copolymer, which was sufficient to completely fill the middle portion of the tube. After injection, all catheters were removed, and the rabbits were allowed to recover and return to regular housing. At this point, two of the rabbits were killed and removed from the study. One was removed because of an anatomic variant, a redundant uterus (the uterine horns had three loops rather than the typical 1–2 loops), which made it impossible to catheterize the fallopian tube as a result of excessive tortuosity of the uterine horns. The other was removed because of perforation of the uterus by the Glidewire during attempts to catheterize the fallopian tube. We believe that this was a result of the small size of the rabbit (the average weight was approximately 3.5 kg), the thin and soft uterine wall, and the anatomy of the rabbit uterus, which is coiled like a spiral, normally containing at least 1–2 loops (Fig 1). The guide wire exerts significant pressure against the wall of the outer curvature of the uterus because the natural tendency of the wire is to straighten. This rabbit was killed immediately after the perforation. Three to seven days after injection of ethylene vinyl alcohol copolymer, each animal was bred. Conception was determined by US after 7–19 days. If pregnant, the rabbit was killed. Otherwise, it was permitted to rebreed until pregnancy was achieved. After the rabbits were killed, their peritoneal cavities were carefully opened and the abdominal and pelvic contents inspected. The presence or absence of embryos in each uterus, presence of ethylene vinyl alcohol copolymer, and presence of any adhesions or other pelvic abnormalities were documented. The distal uterus and attached fallopian tube and ovary from each side were placed separately in formalin. The portion of the tube containing the plug was fixed, sectioned, and stained with hematoxylin and eosin. For the two specimens in which there was no plug in the fallopian tube, the proximal portion of the tube was used. A pathologist analyzed the specimens and graded the degree of inflammation and fibrosis as absent, mild, moderate, or marked according to semiquantitative analysis.
Volume 12
Number 8
Abdala et al
Figure 1. Spot radiograph demonstrates the presence of ethylene vinyl alcohol copolymer within the right fallopian tube (arrow).
RESULTS Patency of the fallopian tubes was demonstrated bilaterally in all animals by the free spillage of contrast material into the peritoneum. The delivery of ethylene vinyl alcohol copolymer into the fallopian tube (six on the right, four on the left) was successful in all animals but one, in which most of the plug almost immediately extruded into the uterus. At that point, no further attempts were made to place additional ethylene vinyl alcohol copolymer into the fallopian tube because a portion of the plug remained in the fallopian tube blocking the catheter. Additionally, no attempts were made to remove the plug with a snare because such maneuvers would have likely resulted in uterine perforation, given the tortuosity and thin wall of the rabbit uterus. The ethylene vinyl alcohol copolymer was left in place and the rabbit was allowed to breed. Fluoroscopic recording and plain film radiography were performed during the procedure for guidance and to confirm the presence of the ethylene vinyl alcohol copolymer within the fallopian tube (Fig 1). There was no spillage of ethylene vinyl alcohol copolymer into the peritoneum. After 3–7 days, the animals were bred. The nonreceptive females were
permitted to rebreed as many times as necessary. The number breeding attempts ranged from 1 to 20 (mean ⫽ 5, median ⫽ 2). Pregnancy was detected by US in the untreated fallopian tube at 7–19 days after effective breeding in nine of the 10 rabbits. The embryos were visualized on US in six of the rabbits, and in the other three, only the gestational sac was detected. These nine rabbits were killed at the time that pregnancy was detected. The remaining rabbit was totally unreceptive and was killed after 20 attempts at breeding. The time between tubal occlusion and sacrifice ranged from 12 to 56 days (mean ⫽ 25.3 d). Gross inspection of the uterus, fallopian tube, ovary, ligaments, and peritoneum confirmed pregnancy in the control side in the nine rabbits that were receptive to breeding. A black plug, representing the ethylene vinyl alcohol copolymer, was demonstrated in the treated fallopian tube in eight rabbits (Fig 2). In two rabbits, the plug was in the distal portion of the uterine horn. There was no evidence of inflammation, adhesions or ethylene vinyl alcohol copolymer in the peritoneum in any of the rabbits. Histologic analysis included rating of the degree of occlusion, fibrosis, and inflammation. The degree of fibro-
•
981
sis and inflammation was graded from none to marked. The occlusion rate within the fallopian tube ranged from 50% to 100% (Fig 3) in eight rabbits. In two rabbits, there was no evidence of occlusion (Fig 4). One of these is the one in which the material extruded into the uterus during injection. The other had pigment (ethylene vinyl alcohol copolymer) in the fallopian tube at sacrifice. The presence of pigment within the fallopian tube was demonstrated in all but two of the rabbits. One of the two is the rabbit in which most of the ethylene vinyl alcohol copolymer extruded into the uterus during injection. The small amount that remained in the fallopian tube was probably extruded into the uterus before sacrifice by fallopian tube peristalsis. The absence of pigment in the other rabbit may also have resulted from expulsion of the pigment into the uterus by fallopian tube peristalsis before the rabbit was killed. Indeed, a small plug was found within the uterine horn at the time the rabbit was killed. Despite this, marked fibrosis, mild inflammation, and 80% occlusion of this rabbit’s fallopian tube was found on sectioning. At the time of the procedure, both these rabbits were noted to have particularly narrow fallopian tubes, making adequate catheterization and injection of the tube very difficult. We were unable to get the microcatheter more than 10 mm into the tube in these two rabbits. The microcatheter did not recoil during injection in either rabbit; rather, in the first case, most of the material refluxed into the uterine horn during injection, and in the latter, we were able to inject only 0.2 mL of the ethylene vinyl alcohol copolymer into the tube because of its small size. The fibrosis was moderate in degree in five (Fig 5), marked in only one and either mild or absent in the remaining four rabbits. In specimens demonstrating fibrosis, it was located submucosally in five and intraluminally in three. The degree of inflammation was moderate in only three rabbits, mild in five, and absent in the two in which there was no evidence of occlusion. None of the specimens demonstrated marked inflammation. Of note, the three rabbits with the longest interval between tubal occlusion and sacrifice had no or only mild
982
•
Ethylene Vinyl Alcohol Copolymer for Sterilization in Rabbits
Figure 2. Photograph of the excised uteri and fallopian tubes after sacrifice demonstrates a small black plug, representing the ethylene vinyl alcohol copolymer, in the right fallopian tube (R) and a single intrauterine pregnancy in the left side (L).
Figure 3. Low-power (40⫻) micrograph (hematoxylin and eosin stain) of cross-section through rabbit fallopian tube 18 days after tubal injection demonstrates occlusion of the lumen by pigment.
fibrosis and inflammation (Fig 4). The predominant inflammatory cells were lymphocytes. Others included histiocytes, eosinophils, and polymorphonucleocytes. Five specimens demonstrated a giant cell reaction (Fig 5).
DISCUSSION Female sterilization is the most commonly used method for permanent contraception throughout the world for those who have completed
August 2001
JVIR
their family (11,12). Currently, the most widely used technique is tubal ligation with use of the laparoscopic or minilaparotomy approach, with the former used most often in developed countries and becoming increasingly available in developing areas (13). Traditionally, these procedures have been performed under general anesthesia, but recently there has been a trend toward performing them under local anesthesia (14,15). Despite the use of local anesthesia, these procedures still require a surgical incision and carry the risk of vascular or other organ injury and the development of peritoneal adhesions. Surgical difficulties resulting in a change of a procedure or prolonged operation time occur in approximately 5% of cases (16). In addition, abdominal pain and dyspareunia may occur in 11.5% and 5% of cases, respectively (17). Therefore, considerable effort is being put into the continued development of nonsurgical methods of tubal occlusion. In attempts to develop a reliable and less expensive nonsurgical sterilization method, many trials have been performed in the past two decades with use of chemical agents (5– 8) or mechanical devices (1–3,5). Most of these studies were performed with the technique of selective fallopian tube catheterization described by Thurmond et al (10). In this study, we used ethylene vinyl alcohol copolymer (Uryx; Genyx Medical), to plug one of the fallopian tubes in rabbits with use of the contralateral fallopian tube as the control. Most previous attempts at nonsurgical selective fallopian tube occlusion have had major drawbacks in that they did not produce satisfactory contraception and/or produced serious side effects. Ross et al (1) placed metal coils into the uterotubal junction in 32 rabbits. In 11 rabbits (34%), the coil was dislodged, and in three rabbits, the coil failed to prevent pregnancy, resulting in an overall failure rate for contraception of 44%. Schmitz-Rode et al (2) demonstrated a contraception failure in one of 10 rabbits after unilateral placement of a microspindle into the fallopian tube. Maubon et al (4) concluded that collagen glue had an insufficient contraceptive effect and should be abandoned. In a separate study, Maubon et al (5) demonstrated that hydrogel placed within the fallo-
Volume 12
Number 8
Abdala et al
Figure 4. Micrograph (200⫻, hematoxylin and eosin stain) of cross-section through rabbit fallopian tube 50 days after ethylene vinyl alcohol copolymer injection demonstrates focal pigment within the lumen and no occlusion. There is preservation of the normal, ciliated epithelium without any evidence of inflammation or fibrosis.
Figure 5. High-power micrograph (400⫻, hematoxylin and eosin stain) of a cross section through rabbit fallopian tube 17 days after ethylene vinyl alcohol copolymer injection demonstrates foreign body giant cell reaction to pigment (single arrow) and fibrosis in the submucosa of the fallopian tube (double arrow).
pian tubes resulted in acute and chronic salpingitis and foreign-body reaction and concluded that a nonreactive material is needed for this procedure. To the contrary, Berkey et
al (7) demonstrated a 100% nonpregnancy rate and the absence of significant complications with selective injection of MCA into the fallopian tubes with use of selective catheterization. It
•
983
is postulated that the method of occlusion of MCA is related to the development of a local inflammatory reaction over an extended period of time, which leads to permanent tubal occlusion (7,18,19). One consequence of the inflammatory reaction elicited by MCA is the development of areas of cystic dilation of the fallopian tube as demonstrated by Berkey et al (7). Theoretically, the development of cystic dilation of the fallopian tube in humans could be painful and may lead to complications such as the development of a tubo-ovarian abscess. In our study, there was mild or no inflammation in seven of 10 rabbits; however, the degree of fibrosis was moderate to marked in six of the 10. There was no evidence of cystic dilation of the fallopian tube. The degree of inflammation and fibrosis appears to be less extensive than that observed by previous investigators. Nonetheless, we achieved a 100% early sterilization rate. Even in the two rabbits in which no evidence of fibrosis, inflammation, or occlusion was demonstrated (Fig 4), pregnancy was achieved only on the control side. In addition, excluding the rabbit that was unreceptive to breeding, the time between injection and sacrifice was the longest for these two rabbits. Another confounding variable is the fact that the three rabbits with the longest time interval between tubal occlusion and sacrifice had mild or no inflammation and fibrosis. This suggests that the method of contraception is not purely mechanical. This is supported by the fact that all of the rabbits but two were shown to have less than 100% occlusion at histologic analysis. Additionally, even in the two rabbits in which the plug was suspected to have been extruded from the fallopian tube by peristalsis, pregnancy occurred in only the control side. Therefore, the method of contraception of this agent requires further study. Half the specimens demonstrated a giant cell reaction, which is indicative of a foreign body response. The lack of a significant fibrotic response may not necessarily be advantageous because, in the presence of peristalsis, there is a theoretical potential for the plug to be extruded from the fallopian tube. As noted previously, this is suspected to have occurred in two of the rabbits.
984
•
Ethylene Vinyl Alcohol Copolymer for Sterilization in Rabbits
The case of extrusion during injection and the case in which the ethylene vinyl alcohol copolymer was found in the uterus at sacrifice were very difficult catheterizations because of the small size of the fallopian tubes in these two specimens. As noted earlier, this resulted in our inability to access more than 10 mm into the fallopian tube. Because of the small lumen, most of the material refluxed into the uterine horn in the former case, and we were able to inject only minimal ethylene vinyl alcohol copolymer in the latter case. One must keep in mind the significant differences in the uterine anatomy between rabbits and humans. The fallopian tube in the rabbit is one third to one half the size of that in humans (10) and the rabbit uterus loops in a spiral fashion (Fig 1) unlike the straightforward anatomy of the human uterus. Our experience is that selective human fallopian tube catheterization can therefore be performed more reliably and safely in the human than in the rabbit. Additionally, the larger human fallopian tube is more likely to accommodate the necessary volume of ethylene vinyl alcohol copolymer without reflux into the uterus. One limitation of this experiment in that it is a relatively short-term study with the rabbits being killed shortly after the detection of pregnancy. Repeating the trial with use of the bilateral occlusion model, as applied by Berkey et al (7), would allow the breeding to be extended over a longer period of time to demonstrate the long-term efficacy of ethylene vinyl alcohol copolymer in achieving sterilization. It would also allow for a more accurate determination of the true incidence of plug extrusion by peristalsis. In conclusion, this study demonstrates that ethylene vinyl alcohol copolymer can be placed nonsurgically into the fallopian tubes with use of the
transcervical approach. At least in the short-term, ethylene vinyl alcohol copolymer appears to be a promising tubal occlusive agent for the purpose of achieving nonsurgical contraception. Nonetheless, further studies are needed to demonstrate the longterm effects of ethylene vinyl alcohol copolymer on the fallopian tube and the incidence of plug extrusion by peristalsis in the animal model before it can be considered for use in the human fallopian tube. However, the potential is there for the use of the technique described herein and by other investigators, whether the agent of choice is ethylene vinyl alcohol copolymer or MCA, to achieve nonsurgical fallopian tube sterilization in humans. Acknowledgment: The authors thank Lisa Jackson for her expert secretarial assistance in the preparation of this manuscript. References 1. Ross PL, Thurmond AS, Uchida BT, et al. Transcatheter tubal sterilization in rabbits: technique and results. Invest Radiol 1994; 29:570 –573. 2. Schmitz-Rode T, Ross PL, Timmermans H, Thurmond AS, Gunther RW, Rosch J. Experimental nonsurgical female sterilization: transcervical implantation of microspindles in fallopian tubes. J Vasc Interv Radiol 1994; 5:905–910. 3. Erb RA, Reed TP. Hysteroscopic oviductal blocking with formed-in-place silicone rubber plugs. J Reprod Med 1979; 23:65– 69. 4. Maubon AJ, Thurmond AS, Laurent A, et al. Tubal sterilization by means of selective catheterization: comparison of a hydrogel and a collagen glue. J Vasc Interv Radiol 1996; 7:733–736. 5. Maubon AJ, Rouanet JP, Laurent A, et al. Tubal sterilization by selective catheterization in an animal model. Invest Radiol 1994; 29:156 –161. 6. Zipper J, Cole LP, Goldsmith A, et al. Quinacrine hydrochloride pellets: preliminary data on a nonsurgical method of female sterilization. Int J Gynecol Obstet 1980; 18:275–279. 7. Berkey GS, Nelson R, Zuckerman AM,
8.
9.
10.
11. 12. 13. 14.
15.
16.
17.
18.
19.
August 2001
JVIR
Dillehay D, Cope C. Sterilization with methyl cyanoacrylate-induced fallopian tube occlusion from a nonsurgical transvaginal approach in rabbits. J Vasc Interv Radiol 1995; 6:669 – 674. Maubon AJ, Thurmond AS, Laurent A, Honiger JE, Scanlan RM, Rouanet JP. Selective tubal sterilization in rabbits: experience with a hydrogel combined with a sclerosing agent. Radiology 1994; 193:721–723. Murayama Y, Vinuela F, Ulhoa A, et al. Nonadhesive liquid embolic agent for cerebral arteriovenous malformations: preliminary histopathological studies in swine rete mirabile. Neurosurgery 1998; 43:1164 –1175. Thurmond AS, Rosch J, Ross PL, Uchida BT, Scanlan RM, Patton PE. Transvaginal fallopian tube catheterization in an animal model. Invest Radiol 1988; 23:818 – 821. Filshie M. Laparoscopic sterilization. Semin Laparosc Surg 1999; 6:112–117. Church CA, Geller JS. Voluntary female sterilization: number one and growing. Popul Rep C 1990; 10:1–23. Wilson EW. Sterilization. Baillieres Clin Obstet Gynaecol 1996; 10:103–119. Jack KE, Chao CR. Female voluntary contraception via minilaparotomy under local anesthesia. Int J Gynecol Obstet 1992; 39:111–116. Bordahl PE, Raeder JC, Nordentoft J, Kirstie U, Refsdal A. Laparoscopic sterilization under local or general anesthesia? A randomized study Obstet Gynecol 1993; 81:137–141. Bhiwandiwala PP, Mumford SD, Feldblum PJ. A comparison of different laparoscopic sterilization occlusion techniques in 24,439 procedures. Am J Obstet Gynecol 1982; 144:319 –331. Dueholm S, Zingenberg HJ, Sandgren G. Late sequelae after laparoscopic sterilization in the pregnant and nonpregnant woman. Acta Obstet Gynecol Scand 1987; 66:227–231. Corfman PA, Richart RM, Taylor HC. Response of the rabbit oviduct to a tissue adhesive. Science 1965; 148:1348 – 1350. Stevenson TC, Taylor DS. The effect of methyl cyanoacrylate tissue adhesive on the human fallopian tube and endometrium. J Obstet Gynecol 1972; 79:1028 –1039.
PURPOSE: To assess the efficacy of ethylene vinyl alcohol copolymer (Uryx) in nonsurgically occluding the fallopian tube and achieving tubal sterilization in the rabbit model. MATERIALS AND METHODS: Ten mature virgin female New England rabbits underwent transvaginal selective bilateral fallopian tube cannulation with use of a coaxial catheter system under general anesthesia. Selective salpingography was performed bilaterally to assess patency of the fallopian tubes. Ethylene vinyl alcohol copolymer was injected unilaterally through a microcatheter to completely fill the middle portion of the tube. Three to seven days after injection, each animal was bred. Conception was determined by ultrasonography (US) 7–19 days after effective breeding. If pregnant, the rabbit was killed. Otherwise, it was permitted to rebreed until pregnancy was achieved. Histologic specimens of the fallopian tubes were prepared and analyzed. RESULTS: Patency of the fallopian tubes was demonstrated bilaterally in all animals by the free spillage of contrast material into the peritoneum. The delivery of ethylene vinyl alcohol copolymer into the fallopian tubes was successful in all animals but one, in which most of the plug almost immediately extruded into the uterus. Pregnancy was detected by US in the untreated fallopian tube in the nine rabbits that were receptive to breeding. No pregnancies were detected in the injected side. Histologic analysis demonstrated variable degrees of occlusion, fibrosis, and inflammation, with the majority of specimens demonstrating mild to moderate inflammation and moderate to marked fibrosis. CONCLUSION: Ethylene vinyl alcohol copolymer can reliably be placed nonsurgically via the transvaginal approach into the fallopian tubes with use of a coaxial catheter system. Ethylene vinyl alcohol copolymer appears to result in less fibrosis than previously investigated agents and demonstrates a 100% early sterilization rate in the rabbit model. Index terms:
Alcohol • Fallopian tubes, occlusion • Sterlization
J Vasc Interv Radiol 2001; 12:979 –984 Abbreviations: DMSO ⫽ dimethyl sulfoxide, MCA ⫽ methyl cyanoacrylate
THE most widely used technique of voluntary sterilization in women is tubal ligation via laparoscopy or minilaparotomy. However, these procedures require surgical incision and
occasionally require general anesthesia, along with their adherent risks. In addition, surgical techniques are typically more expensive than less invasive nonsurgical methods. Therefore,
From the Department of Radiology (N.A.), University of Sao Paulo, Sao Paulo; Hospital Santa Helena (R.M.), Goias, Brazil; Departments of Vascular and Interventional Radiology (A.L.), Anatomic Pathology (A.D.), Radiology (K.G.), and Biological Resources Unit, Central Core (H.M.R.), Cleveland Clinic Foundation, Cleveland, Ohio; and Department of Radiology (B.L.D.), University of Texas Southwestern Medical Center, Dallas, Texas. Re-
ceived November 1, 2000; revision requested January 3, 2001; revision received March 28; accepted March 9. From the 2000 SCVIR Annual Meeting. Address correspondence to A.L., Department of Radiology, University of Maryland Medical System, 22 S. Greene St., Baltimore, MD 21201-1595; E-mail: [email protected] © SCVIR, 2001
considerable effort is being devoted to the development of nonsurgical methods of tubal occlusion by means of selective fallopian tube catheterization. This technique requires the use of a suitable mechanical or chemical occlusive agent. Some mechanical devices that have been used with this technique include coils (1), microspindles (2), silicon plugs (3), and hydrogel (4). These devices are inert and occlude or stretch the lumen of the tube, obstructing ovum or sperm passage. Chemical agents include ethanol, tetracycline, quinacrine hydrochloride, and sclerosing agents (5– 8).
979
980
•
Ethylene Vinyl Alcohol Copolymer for Sterilization in Rabbits
Most noninvasive techniques have major drawbacks in that they do not produce satisfactory contraception and/or produce serious side effects. The recent exception is the use of methyl cyanoacrylate (MCA), a sclerosing agent, with which Berkey et al (7) demonstrated a 100% nonpregnancy rate and the absence of significant complications. However, their study demonstrated the development of fallopian tube cysts as well as significant fibrosis and inflammation on histologic analysis. Given the results achieved by Berkey et al (7) and others, it is readily apparent that tubal sterilization is a promising technique and that new material and formulations should be tested to find the one that is most efficient and safe. In this study, we used ethylene vinyl alcohol copolymer (Uryx; Genyx Medical, Aliso Viejo, CA) to plug one of the fallopian tubes in the rabbit model. The rabbit model was chosen because its uterus contains two completely separate symmetric uterine horns and cervices. This enabled us to perform our experiment on one side, leaving the other side as the control. Ethylene vinyl alcohol copolymer is a biocompatible polymer composed of a random mixture of ethylene (hydrophobic) and vinyl alcohol (hydrophilic). It was dissolved in an 8% concentration of anhydrous dimethyl sulfoxide (DMSO) (9). Thirty percent micronized tantalum powder was added to the polymer/solvent mixture to obtain appropriate radiopacity. When this mixture comes into contact with aqueous media, DMSO rapidly diffuses away from the mixture, causing in situ precipitation and solidification of the polymer, with formation of a spongelike plug that is capable of sealing cavities (9). The material can be deployed in a highly controllable manner with use of fluoroscopic guidance. Our goal in this study was to determine the efficacy of ethylene vinyl alcohol copolymer to occlude the fallopian tube and achieve tubal sterilization in the rabbit model.
MATERIALS AND METHODS The Animal Care Protocol was approved by the Institutional Animal Care Committee and complied with the Principles of Laboratory Animal
Care (formulated by the National Society of Medical Research) and the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 80 –23, revised 1985). Twelve mature virgin female and two mature male New Zealand rabbits with an average weight of approximately 3.5 kg were used in the study. The males were used solely for breeding purposes. The females were anesthetized with lidocaine hydrochloride 5 mg/kg and ketamine 35 mg/kg. The vulvar area was scrubbed with an iodine-based cleanser. With use of the technique described by Thurmond et al (10), an 18-F Foley catheter (Sherwood, St. Louis, MO) with a hole punched in its tip was reinforced with an 8.5-F, thick-walled radiopaque Teflon catheter (Cook, Bloomington, IN) and introduced into the vagina. Its balloon was inflated to approximately 10 mL with normal saline solution and a moderate amount of contrast medium (Ioxithalamate; Telebrix Laboratories/ Guerbet; Aulnay-sous-Bois, France) and air were injected to distend and visualize the area of the two cervices. Subsequently, a 5-F Multipurpose catheter (Medi-tech/Boston Scientific, Watertown, MA) was placed into one of the cervices and, with the help of a hydrophilic 0.035-inch Glidewire (Medi-tech/Boston Scientific), advanced to the point of the fallopian tube orifice. A 2.2-F dimethyl sulfoxide (DMSO)-compatible (DMSO is a solvent which is not compatible with many catheters) microcatheter (Easy Rider; MTI Microtherapeutics, Irvine, CA) and a 0.010-inch micro– guide wire (Guidant ACS, Temecula, CA) were advanced to the tip of the Multipurpose catheter. The micro– guide wire, and then the microcatheter, were advanced into the proximal portion of the fallopian tube. To check patency of the fallopian tube, ioxithalamate was injected into the fallopian tube to the point of spillage into the peritoneal cavity. This procedure was repeated for the contralateral fallopian tube. After confirmation of tube patency, the microcatheter was advanced into the isthmic portion of one of the tubes with the aid of the micro– guide wire. One half-milliliter of DMSO was injected to flush the catheter (ethylene vinyl alcohol copolymer solidifies in the presence of water). This was followed by 0.2– 0.5
August 2001
JVIR
mL of ethylene vinyl alcohol copolymer, which was sufficient to completely fill the middle portion of the tube. After injection, all catheters were removed, and the rabbits were allowed to recover and return to regular housing. At this point, two of the rabbits were killed and removed from the study. One was removed because of an anatomic variant, a redundant uterus (the uterine horns had three loops rather than the typical 1–2 loops), which made it impossible to catheterize the fallopian tube as a result of excessive tortuosity of the uterine horns. The other was removed because of perforation of the uterus by the Glidewire during attempts to catheterize the fallopian tube. We believe that this was a result of the small size of the rabbit (the average weight was approximately 3.5 kg), the thin and soft uterine wall, and the anatomy of the rabbit uterus, which is coiled like a spiral, normally containing at least 1–2 loops (Fig 1). The guide wire exerts significant pressure against the wall of the outer curvature of the uterus because the natural tendency of the wire is to straighten. This rabbit was killed immediately after the perforation. Three to seven days after injection of ethylene vinyl alcohol copolymer, each animal was bred. Conception was determined by US after 7–19 days. If pregnant, the rabbit was killed. Otherwise, it was permitted to rebreed until pregnancy was achieved. After the rabbits were killed, their peritoneal cavities were carefully opened and the abdominal and pelvic contents inspected. The presence or absence of embryos in each uterus, presence of ethylene vinyl alcohol copolymer, and presence of any adhesions or other pelvic abnormalities were documented. The distal uterus and attached fallopian tube and ovary from each side were placed separately in formalin. The portion of the tube containing the plug was fixed, sectioned, and stained with hematoxylin and eosin. For the two specimens in which there was no plug in the fallopian tube, the proximal portion of the tube was used. A pathologist analyzed the specimens and graded the degree of inflammation and fibrosis as absent, mild, moderate, or marked according to semiquantitative analysis.
Volume 12
Number 8
Abdala et al
Figure 1. Spot radiograph demonstrates the presence of ethylene vinyl alcohol copolymer within the right fallopian tube (arrow).
RESULTS Patency of the fallopian tubes was demonstrated bilaterally in all animals by the free spillage of contrast material into the peritoneum. The delivery of ethylene vinyl alcohol copolymer into the fallopian tube (six on the right, four on the left) was successful in all animals but one, in which most of the plug almost immediately extruded into the uterus. At that point, no further attempts were made to place additional ethylene vinyl alcohol copolymer into the fallopian tube because a portion of the plug remained in the fallopian tube blocking the catheter. Additionally, no attempts were made to remove the plug with a snare because such maneuvers would have likely resulted in uterine perforation, given the tortuosity and thin wall of the rabbit uterus. The ethylene vinyl alcohol copolymer was left in place and the rabbit was allowed to breed. Fluoroscopic recording and plain film radiography were performed during the procedure for guidance and to confirm the presence of the ethylene vinyl alcohol copolymer within the fallopian tube (Fig 1). There was no spillage of ethylene vinyl alcohol copolymer into the peritoneum. After 3–7 days, the animals were bred. The nonreceptive females were
permitted to rebreed as many times as necessary. The number breeding attempts ranged from 1 to 20 (mean ⫽ 5, median ⫽ 2). Pregnancy was detected by US in the untreated fallopian tube at 7–19 days after effective breeding in nine of the 10 rabbits. The embryos were visualized on US in six of the rabbits, and in the other three, only the gestational sac was detected. These nine rabbits were killed at the time that pregnancy was detected. The remaining rabbit was totally unreceptive and was killed after 20 attempts at breeding. The time between tubal occlusion and sacrifice ranged from 12 to 56 days (mean ⫽ 25.3 d). Gross inspection of the uterus, fallopian tube, ovary, ligaments, and peritoneum confirmed pregnancy in the control side in the nine rabbits that were receptive to breeding. A black plug, representing the ethylene vinyl alcohol copolymer, was demonstrated in the treated fallopian tube in eight rabbits (Fig 2). In two rabbits, the plug was in the distal portion of the uterine horn. There was no evidence of inflammation, adhesions or ethylene vinyl alcohol copolymer in the peritoneum in any of the rabbits. Histologic analysis included rating of the degree of occlusion, fibrosis, and inflammation. The degree of fibro-
•
981
sis and inflammation was graded from none to marked. The occlusion rate within the fallopian tube ranged from 50% to 100% (Fig 3) in eight rabbits. In two rabbits, there was no evidence of occlusion (Fig 4). One of these is the one in which the material extruded into the uterus during injection. The other had pigment (ethylene vinyl alcohol copolymer) in the fallopian tube at sacrifice. The presence of pigment within the fallopian tube was demonstrated in all but two of the rabbits. One of the two is the rabbit in which most of the ethylene vinyl alcohol copolymer extruded into the uterus during injection. The small amount that remained in the fallopian tube was probably extruded into the uterus before sacrifice by fallopian tube peristalsis. The absence of pigment in the other rabbit may also have resulted from expulsion of the pigment into the uterus by fallopian tube peristalsis before the rabbit was killed. Indeed, a small plug was found within the uterine horn at the time the rabbit was killed. Despite this, marked fibrosis, mild inflammation, and 80% occlusion of this rabbit’s fallopian tube was found on sectioning. At the time of the procedure, both these rabbits were noted to have particularly narrow fallopian tubes, making adequate catheterization and injection of the tube very difficult. We were unable to get the microcatheter more than 10 mm into the tube in these two rabbits. The microcatheter did not recoil during injection in either rabbit; rather, in the first case, most of the material refluxed into the uterine horn during injection, and in the latter, we were able to inject only 0.2 mL of the ethylene vinyl alcohol copolymer into the tube because of its small size. The fibrosis was moderate in degree in five (Fig 5), marked in only one and either mild or absent in the remaining four rabbits. In specimens demonstrating fibrosis, it was located submucosally in five and intraluminally in three. The degree of inflammation was moderate in only three rabbits, mild in five, and absent in the two in which there was no evidence of occlusion. None of the specimens demonstrated marked inflammation. Of note, the three rabbits with the longest interval between tubal occlusion and sacrifice had no or only mild
982
•
Ethylene Vinyl Alcohol Copolymer for Sterilization in Rabbits
Figure 2. Photograph of the excised uteri and fallopian tubes after sacrifice demonstrates a small black plug, representing the ethylene vinyl alcohol copolymer, in the right fallopian tube (R) and a single intrauterine pregnancy in the left side (L).
Figure 3. Low-power (40⫻) micrograph (hematoxylin and eosin stain) of cross-section through rabbit fallopian tube 18 days after tubal injection demonstrates occlusion of the lumen by pigment.
fibrosis and inflammation (Fig 4). The predominant inflammatory cells were lymphocytes. Others included histiocytes, eosinophils, and polymorphonucleocytes. Five specimens demonstrated a giant cell reaction (Fig 5).
DISCUSSION Female sterilization is the most commonly used method for permanent contraception throughout the world for those who have completed
August 2001
JVIR
their family (11,12). Currently, the most widely used technique is tubal ligation with use of the laparoscopic or minilaparotomy approach, with the former used most often in developed countries and becoming increasingly available in developing areas (13). Traditionally, these procedures have been performed under general anesthesia, but recently there has been a trend toward performing them under local anesthesia (14,15). Despite the use of local anesthesia, these procedures still require a surgical incision and carry the risk of vascular or other organ injury and the development of peritoneal adhesions. Surgical difficulties resulting in a change of a procedure or prolonged operation time occur in approximately 5% of cases (16). In addition, abdominal pain and dyspareunia may occur in 11.5% and 5% of cases, respectively (17). Therefore, considerable effort is being put into the continued development of nonsurgical methods of tubal occlusion. In attempts to develop a reliable and less expensive nonsurgical sterilization method, many trials have been performed in the past two decades with use of chemical agents (5– 8) or mechanical devices (1–3,5). Most of these studies were performed with the technique of selective fallopian tube catheterization described by Thurmond et al (10). In this study, we used ethylene vinyl alcohol copolymer (Uryx; Genyx Medical), to plug one of the fallopian tubes in rabbits with use of the contralateral fallopian tube as the control. Most previous attempts at nonsurgical selective fallopian tube occlusion have had major drawbacks in that they did not produce satisfactory contraception and/or produced serious side effects. Ross et al (1) placed metal coils into the uterotubal junction in 32 rabbits. In 11 rabbits (34%), the coil was dislodged, and in three rabbits, the coil failed to prevent pregnancy, resulting in an overall failure rate for contraception of 44%. Schmitz-Rode et al (2) demonstrated a contraception failure in one of 10 rabbits after unilateral placement of a microspindle into the fallopian tube. Maubon et al (4) concluded that collagen glue had an insufficient contraceptive effect and should be abandoned. In a separate study, Maubon et al (5) demonstrated that hydrogel placed within the fallo-
Volume 12
Number 8
Abdala et al
Figure 4. Micrograph (200⫻, hematoxylin and eosin stain) of cross-section through rabbit fallopian tube 50 days after ethylene vinyl alcohol copolymer injection demonstrates focal pigment within the lumen and no occlusion. There is preservation of the normal, ciliated epithelium without any evidence of inflammation or fibrosis.
Figure 5. High-power micrograph (400⫻, hematoxylin and eosin stain) of a cross section through rabbit fallopian tube 17 days after ethylene vinyl alcohol copolymer injection demonstrates foreign body giant cell reaction to pigment (single arrow) and fibrosis in the submucosa of the fallopian tube (double arrow).
pian tubes resulted in acute and chronic salpingitis and foreign-body reaction and concluded that a nonreactive material is needed for this procedure. To the contrary, Berkey et
al (7) demonstrated a 100% nonpregnancy rate and the absence of significant complications with selective injection of MCA into the fallopian tubes with use of selective catheterization. It
•
983
is postulated that the method of occlusion of MCA is related to the development of a local inflammatory reaction over an extended period of time, which leads to permanent tubal occlusion (7,18,19). One consequence of the inflammatory reaction elicited by MCA is the development of areas of cystic dilation of the fallopian tube as demonstrated by Berkey et al (7). Theoretically, the development of cystic dilation of the fallopian tube in humans could be painful and may lead to complications such as the development of a tubo-ovarian abscess. In our study, there was mild or no inflammation in seven of 10 rabbits; however, the degree of fibrosis was moderate to marked in six of the 10. There was no evidence of cystic dilation of the fallopian tube. The degree of inflammation and fibrosis appears to be less extensive than that observed by previous investigators. Nonetheless, we achieved a 100% early sterilization rate. Even in the two rabbits in which no evidence of fibrosis, inflammation, or occlusion was demonstrated (Fig 4), pregnancy was achieved only on the control side. In addition, excluding the rabbit that was unreceptive to breeding, the time between injection and sacrifice was the longest for these two rabbits. Another confounding variable is the fact that the three rabbits with the longest time interval between tubal occlusion and sacrifice had mild or no inflammation and fibrosis. This suggests that the method of contraception is not purely mechanical. This is supported by the fact that all of the rabbits but two were shown to have less than 100% occlusion at histologic analysis. Additionally, even in the two rabbits in which the plug was suspected to have been extruded from the fallopian tube by peristalsis, pregnancy occurred in only the control side. Therefore, the method of contraception of this agent requires further study. Half the specimens demonstrated a giant cell reaction, which is indicative of a foreign body response. The lack of a significant fibrotic response may not necessarily be advantageous because, in the presence of peristalsis, there is a theoretical potential for the plug to be extruded from the fallopian tube. As noted previously, this is suspected to have occurred in two of the rabbits.
984
•
Ethylene Vinyl Alcohol Copolymer for Sterilization in Rabbits
The case of extrusion during injection and the case in which the ethylene vinyl alcohol copolymer was found in the uterus at sacrifice were very difficult catheterizations because of the small size of the fallopian tubes in these two specimens. As noted earlier, this resulted in our inability to access more than 10 mm into the fallopian tube. Because of the small lumen, most of the material refluxed into the uterine horn in the former case, and we were able to inject only minimal ethylene vinyl alcohol copolymer in the latter case. One must keep in mind the significant differences in the uterine anatomy between rabbits and humans. The fallopian tube in the rabbit is one third to one half the size of that in humans (10) and the rabbit uterus loops in a spiral fashion (Fig 1) unlike the straightforward anatomy of the human uterus. Our experience is that selective human fallopian tube catheterization can therefore be performed more reliably and safely in the human than in the rabbit. Additionally, the larger human fallopian tube is more likely to accommodate the necessary volume of ethylene vinyl alcohol copolymer without reflux into the uterus. One limitation of this experiment in that it is a relatively short-term study with the rabbits being killed shortly after the detection of pregnancy. Repeating the trial with use of the bilateral occlusion model, as applied by Berkey et al (7), would allow the breeding to be extended over a longer period of time to demonstrate the long-term efficacy of ethylene vinyl alcohol copolymer in achieving sterilization. It would also allow for a more accurate determination of the true incidence of plug extrusion by peristalsis. In conclusion, this study demonstrates that ethylene vinyl alcohol copolymer can be placed nonsurgically into the fallopian tubes with use of the
transcervical approach. At least in the short-term, ethylene vinyl alcohol copolymer appears to be a promising tubal occlusive agent for the purpose of achieving nonsurgical contraception. Nonetheless, further studies are needed to demonstrate the longterm effects of ethylene vinyl alcohol copolymer on the fallopian tube and the incidence of plug extrusion by peristalsis in the animal model before it can be considered for use in the human fallopian tube. However, the potential is there for the use of the technique described herein and by other investigators, whether the agent of choice is ethylene vinyl alcohol copolymer or MCA, to achieve nonsurgical fallopian tube sterilization in humans. Acknowledgment: The authors thank Lisa Jackson for her expert secretarial assistance in the preparation of this manuscript. References 1. Ross PL, Thurmond AS, Uchida BT, et al. Transcatheter tubal sterilization in rabbits: technique and results. Invest Radiol 1994; 29:570 –573. 2. Schmitz-Rode T, Ross PL, Timmermans H, Thurmond AS, Gunther RW, Rosch J. Experimental nonsurgical female sterilization: transcervical implantation of microspindles in fallopian tubes. J Vasc Interv Radiol 1994; 5:905–910. 3. Erb RA, Reed TP. Hysteroscopic oviductal blocking with formed-in-place silicone rubber plugs. J Reprod Med 1979; 23:65– 69. 4. Maubon AJ, Thurmond AS, Laurent A, et al. Tubal sterilization by means of selective catheterization: comparison of a hydrogel and a collagen glue. J Vasc Interv Radiol 1996; 7:733–736. 5. Maubon AJ, Rouanet JP, Laurent A, et al. Tubal sterilization by selective catheterization in an animal model. Invest Radiol 1994; 29:156 –161. 6. Zipper J, Cole LP, Goldsmith A, et al. Quinacrine hydrochloride pellets: preliminary data on a nonsurgical method of female sterilization. Int J Gynecol Obstet 1980; 18:275–279. 7. Berkey GS, Nelson R, Zuckerman AM,
8.
9.
10.
11. 12. 13. 14.
15.
16.
17.
18.
19.
August 2001
JVIR
Dillehay D, Cope C. Sterilization with methyl cyanoacrylate-induced fallopian tube occlusion from a nonsurgical transvaginal approach in rabbits. J Vasc Interv Radiol 1995; 6:669 – 674. Maubon AJ, Thurmond AS, Laurent A, Honiger JE, Scanlan RM, Rouanet JP. Selective tubal sterilization in rabbits: experience with a hydrogel combined with a sclerosing agent. Radiology 1994; 193:721–723. Murayama Y, Vinuela F, Ulhoa A, et al. Nonadhesive liquid embolic agent for cerebral arteriovenous malformations: preliminary histopathological studies in swine rete mirabile. Neurosurgery 1998; 43:1164 –1175. Thurmond AS, Rosch J, Ross PL, Uchida BT, Scanlan RM, Patton PE. Transvaginal fallopian tube catheterization in an animal model. Invest Radiol 1988; 23:818 – 821. Filshie M. Laparoscopic sterilization. Semin Laparosc Surg 1999; 6:112–117. Church CA, Geller JS. Voluntary female sterilization: number one and growing. Popul Rep C 1990; 10:1–23. Wilson EW. Sterilization. Baillieres Clin Obstet Gynaecol 1996; 10:103–119. Jack KE, Chao CR. Female voluntary contraception via minilaparotomy under local anesthesia. Int J Gynecol Obstet 1992; 39:111–116. Bordahl PE, Raeder JC, Nordentoft J, Kirstie U, Refsdal A. Laparoscopic sterilization under local or general anesthesia? A randomized study Obstet Gynecol 1993; 81:137–141. Bhiwandiwala PP, Mumford SD, Feldblum PJ. A comparison of different laparoscopic sterilization occlusion techniques in 24,439 procedures. Am J Obstet Gynecol 1982; 144:319 –331. Dueholm S, Zingenberg HJ, Sandgren G. Late sequelae after laparoscopic sterilization in the pregnant and nonpregnant woman. Acta Obstet Gynecol Scand 1987; 66:227–231. Corfman PA, Richart RM, Taylor HC. Response of the rabbit oviduct to a tissue adhesive. Science 1965; 148:1348 – 1350. Stevenson TC, Taylor DS. The effect of methyl cyanoacrylate tissue adhesive on the human fallopian tube and endometrium. J Obstet Gynecol 1972; 79:1028 –1039.