Key Words: Postoperative adhesions, surgical barriers, expanded polytetrafiuoroethylene, oxidized regenerated cellulose

Transcription

1 FERTILITY AND STERILITY 1995 American Society for Reproductive Medicine Printed on acid-free paper in U. S. A. Expanded polytetrafluoroethylene (Gore-lex Surgical Membrane*) is superior to oxidized regenerated cellulose (Interceed lc7t) in preventing adhesions:l: A. F. Haney, M.D.II~ John Hesla, M.D.** Bradley S. Hurst, M.D.tt L. Michael Kettel, M.D.:!::!: Anna A. Murphy, M.D.** John A. Rock, M.D. ** Guillermo Rowe, M.D. William D. Schlaff, M.D.tt Duke University Medical Center, Durham, North Carolina; Emory University Medical School, Atlanta, Georgia; University of Colorado Health Sciences Center, Denver, Colorado; University of California at San Diego, San Diego, California; and Women's Hospital of Texas, Houston, Texas Objective: To compare the impact of expanded polytetrafiuoroethylene (PTFE; Gore-Tex Surgical Membrane; W. L. Gore & Associates, Inc., Flagstaff, AZ) and oxidized regenerated cellulose (Interceed TC7, Johnson & Johnson Medical, Inc., Arlington, TX) on the development of postsurgical adhesions. Design: A multicenter, nonblinded, randomized clinical trial. Setting: University medical centers. Interventions: Each barrier was allocated randomly to the left or right sidewall of every patient. Patients: Thirty-two women with bilateral pelvic sidewall adhesions undergoing reconstructive surgery and second-look laparoscopy. Main Outcome Measures: Adhesion score (on a 0- to ll-point scale), the area of adhesion (cm ), and the likelihood of no adhesions. Results: The use of both barriers was associated with a lower adhesion score and area of adhesion postoperatively. However, those sidewalls covered with PTFE had a significantly lower adhesion score (0.97 ± 0.30 versus.76 ± 0.61 points, mean ± SEM) and area of adhesion (0.95 ± 0.35 versus 3.5 ± 0.6 cm ). Overall, more sidewalls covered with PTFE had no adhesions (1 versus 7) and, when adhesions were present on the contralateral sidewall, the number of sidewalls covered with PTFE without adhesions was greater than those covered with oxidized regenerated cellulose (16 versus ). Conclusion: Expanded polytetrafiuoroethylene was associated with fewer postsurgical adhesions to the pelvic sidewall than oxidized regenerated cellulose. Fertil Steril 1995;63:101-6 Key Words: Postoperative adhesions, surgical barriers, expanded polytetrafiuoroethylene, oxidized regenerated cellulose Postoperative pelvic adhesions after gynecologic surgery remain a major clinical concern because ad- Received July 1, 199; revised and accepted December 1,199. * W. L. Gore & Associates, Inc., Flagstaff, Arizona. t Johnson & Johnson Medical, Inc., Arlington, Texas. t Sponsored by a grant from W. L. Gore & Associates, Inc., Flagstaff, Arizona. Presented at the 50th Annual Meeting of The American Fertility Society, San Antonio, Texas, November 8 to 11, 199. II Department of Obstetrics and Gynecology, Duke University Medical Center. ~ Reprint requests: A. F. Haney, M.D., Box 971, Duke University Medical Center, Durham, North Carolina 7710 (FAX: ). nexal adhesions have the potential to impair fertility (1, ). Adhesion reformation after adhesiolysis is a particularly vexing problem when the procedure has been performed specifically to improve or preserve fertility (3-7). Because there are no medical * Department of Obstetrics and Gynecology, Emory University Medical School. tt Department of Obstetrics and Gynecology, University of Colorado Health Sciences Center. tt Department of Obstetrics and Gynecology, University of California at San Diego. Department of Obstetrics and Gynecology, Women's Hospital of Texas. Haney et ai. Barriers and adhesion reformation 101

2 regimens that effectively prevent postsurgical adhesions (8, 9), a strategy has emerged in which a mechanical barrier is placed between two surgically traumatized peritoneal surfaces to allow each surface to heal independently without the development of coalescing adhesions (10-13). There are two surgical barriers currently available, each with very different characteristics. Oxidized regenerated cellulose (Interceed TC7; Johnson & Johnson Medical, Inc., Arlington, TX) is degradable with procoagulant properties, whereas expanded polytetrafiuoroethylene (PTFE; Gore-Tex Surgical Membrane; W. L. Gore & Associates, Inc., Flagstaff, AZ) is permanent. Controversy continues regarding the use of these barriers because there are virtually no data available that directly compare the two materials, and evaluation of either barrier material to no treatment does not address their relative value. This study reports the results of a multicenter, nonblinded, randomized clinical trial conducted to compare directly the two barrier materials, using the extent of adhesion reformation to the pelvic sidewall as the outcome measure. Study Population MATERIALS AND METHODS Women were eligible for inclusion in the study if they were undergoing open reconstructive pelvic surgery, if they had adhesions between the pelvic sidewall and the reproductive organs that were comparable in extent, location, and density on both sides, and if they were candidates for second-look therapeutic laparoscopy. Each center obtained approval for the study from its respective institutional review committee for human experimentation, and all women gave informed consent. Patients with an increased surgical risk based on any metabolic or other condition were excluded from participating in the study, including those with active infectious disease, a history of radiation therapy, inflammation of the peritoneal cavity (e.g., appendicitis), enteric fistula, or thrombosis. In an attempt to reduce intracenter variability, the enrolling surgeons in each center were designated to be responsible for all aspects of surgery and patient care, gathering all data regarding adhesions and ensuring adherence to the study protocol. Thirty-two patients were enrolled in the study between January 1, 1991 and August 31, The patients ranged in age from 18 to 0 years and weighed 6.9 ± 1.0 kg (138. ±.3 pounds; mean ± SEM). Eighty-five percent of the subjects previously had undergone pelvic surgery and 16.% had a history of pelvic infection. 10 Haney et ai. Barriers and adhesion reformation Table 1 Scoring System for Grading Adhesions Description Extent No involvement s5% site involved >5% but s50% site involved >50% but s75% site involved >75% site involved Type None Filmy, transparent, avascular Opaque, translucent, avascular Opaque, capillaries present Opaque, larger vessels present Tenacity None Adhesions fall apart Adhesions lysed with traction Adhesions sharply dissected * Total score = sum of scores. Laparotomy Score* o 1 3 o 1 3 o 1 3 Antibiotic prophylaxis was begun 1 hour before the initial procedure and continued for ;;:: hours. The choice of the antibiotic was left to the surgeon. The adhesiolysis was carried out in conjunction with, but was not necessarily restricted to, salpingostomy, fimbrioplasty, ovarian cystectomy, and conservative resection of endometriosis. The specific surgical technique used was left to the discretion of the surgeon and meticulous hemostasis was achieved in all cases. The status of both pelvic sidewall sites was assessed and the adhesions at each site were graded on a 0- to ll-point scale that evaluated the extent, type, and tenacity of the adhesions (Table 1). After adhesiolysis, the size of the resulting peritoneal defect was measured in square centimeters. At this point in the procedure, if the patient met the criteria for inclusion as a candidate for the study, a randomization envelope (randomization accomplished by RANDOM software; Digital Equipment Corporation, Manard, MA) was opened to reveal which sidewall was to be covered with PTFE and which was to be covered with oxidized regenerated cellulose. Each patient received PTFE on one side and oxidized regenerated cellulose on the other. The PTFE was randomized to the left pelvic sidewall 5% of the time and to the right sidewall 55% of the time with oxidized regenerated cellulose randomized to the inverse percentages. The PTFE barrier was tailored with scissors to provide ;;:: 1 cm of overlap of the defect and was sutured in place with at least one 7-0 or 8-0 nonabsorbable suture of nylon or polypropylene. A variable number of additional absorbable or nonabsorbable sutures were placed at the surgeon's discretion to ensure that the PTFE barrier would remain in place Fertility and Sterility

3 in the postoperative interval. The oxidized regenerated cellulose barrier was cut similarly to cover the defect in the peritoneum and was placed on the sidewall in accordance with the manufacturer's instructions for use, wetting the material after placement. The condition of the pelvic sidewall sites was assessed again upon completing the barrier placement, to ensure specifically that hemostasis had been maintained and that bleeding would not influence adversely the efficacy of oxidized regenerated cellulose, which, unlike PTFE, requires complete hemostasis. No other adhesion prophylaxis measures were used ADHESION SCORE 6 0 LAPAROTOMY ND LOOK LAPAROSCOPY Second-Look Laparoscopy Therapeutic second-look laparoscopy was undertaken from 1 to 6 weeks after the initial reconstructive procedure in a manner consistent with the routine of the operating surgeon. The time interval averaged 30 ±.3 days. On visualization of the peritoneal cavity, the pelvic adhesions at the experimental sidewalls were graded using the scoring system (Table 1) and then both the PTFE barrier and any remnants of the oxidized regenerated cellulose were removed. When adhesions were encountered at the experimental sites, they were lysed and the area of the resultant defect was measured. The PTFE was retrieved from 9 of 3 patients enrolled, and remnants of oxidized regenerated cellulose were removed from patients. Second-look laparoscopy was not possible in three women because of technical difficulties with developing a pneumoperitoneum. Statistical Analysis All data analyses were carried out using the RS/1 statistics package (Digital Equipment Corporation). All comparisons were two-tailed with the a level set at :50.05 and the f3 level set at : A two-way analysis of variance was performed on the data to determine if the operating surgeon was a significant covariable. The analysis rejected this hypothesis (P :5 0.66), and the data for all the centers therefore were pooled. Because the data collected for adhesion score and area of adhesion were not distributed normally, all statistical computations were performed by nonparametric methods (Mann-Whitney U-test and Wilcoxon Ranked Pairs test). A percentage of improvement for adhesion size was calculated for each barrier according to the formula: (initial defect size - size at second look)/initial size x 100. The probability that the site would have no adhesions whatsoever at second-look laparoscopy, with or without selecting for the presence of adhesions on the contralateral sidewall, was assessed for both barriers by X analysis (Fischer's Exact test). Figure 1 Adhesion score. The total adhesion scores from the laparotomy and second-look laparoscopy are presented for both barriers. The graph presents the mean ± SD for the initial and second-look procedures (PTFE, 10.8 ± 0.65 versus 0.97 ± 1.6 cm ; oxidized regenerated cellulose, 10.5 ± 0.86 versus.76 ± 3.8 cm ). The adhesion scores were significantly lower at the second-look laparoscopy compared with those at the initiallaparotomy for both the sidewalls covered with PTFE (P,-;; ) and those covered with oxidized regenerated cellulose (P,-;; ). The adhesion scores at second-look laparoscopy were significantly lower for the sidewalls covered with PTFE compared with those covered with oxidized regenerated cellulose (P,-;; )., PTFE; ~, oxidized regenerated cellulose. Adhesion Score RESULTS Before adhesiolysis, adhesion scores for the sidewalls covered with PTFE and oxidized regenerated cellulose did not differ from each other, demonstrating that the two experimental groups were similar (Fig. 1). Comparison ofthe scores at the second-look laparoscopy for each barrier with the original scores at those same sites revealed that the use of either barrier was associated with an improvement over the original sidewall score (PTFE: median 11.0 versus 0.0 and range 9 to 11 versus 0 to 5; P : ; oxidized regenerated cellulose: 11.0 versus 5.0 and 8 to 11 versus 0 to 10; P : ). However, the adhesion score at second-look laparoscopy for the sidewall sites covered with PTFE was significantly less than the score for those covered by oxidized regenerated cellulose (P : ; Fig. 1). Area of Adhesion Before adhesiolysis, the area of adhesion for the sidewalls covered by PTFE and oxidized regenerated cellulose did not differ from each other, again demonstrating initial comparability of the two experimental groups (Fig. ). When the area of adhesion at the second-look laparoscopy for each barrier was compared with the original area of adhesion at the same site, the use of either barrier was associated with Haney et ai. Barriers and adhesion reformation 103

4 AREA OF Z 10 ADHESION (CM ) 8 6 LAPAROTOMY ND LOOK LAPAROSCOPY roscopy, consistent with the fact that the material is degraded rapidly. The oxidized regenerated cellulose barriers visualizable were soft yellow plaques adherent to the original site of placement on the pelvic sidewall. The PTFE samples were unchanged in appearance and were removed from all of the patients who successfully underwent second-look laparoscopy. Because they were sutured in place, the PTFE barriers remained at the site of placement until the sutures were cut and the barrier was removed. No difficulty was encountered in removing the PTFE via laparoscopy. Figure Area of adhesion. The areas of adhesion noted at laparotomy and second-look laparoscopy are presented for both barriers. The graph presents the mean ± SD for the initial and secondlook procedures (PTFE, 8.16 ± 5.60 versus 0.95 ± 1.88 cm ; oxidized regenerated cellulose, 9.07 ± 7.53 versus 3.5 ± 3.3 cm ). The areas of adhesion were significantly lower at the second-look laparoscopy compared with those at the initial laparotomy for both the sidewalls covered with PTFE (P :=; ) and those covered with oxidized regenerated cellulose (P :=; ). The areas of adhesion at second-look laparoscopy were significantly lower for the sidewalls covered with PTFE compared with those covered with oxidized regenerated cellulose (P :=; ).., PTFE; 1IJil, oxidized regenerated cellulose. an improvement (PTFE: median 6.0 versus 0.0 and range 1.0 to 5.0 versus 0 to 6; P : ; oxidized regenerated cellulose: 6.0 versus 3.0 and.0 to 35.0 versus 0 to 16; P : ; Fig. ). However, the area of adhesion at second-look laparoscopy for the sidewall sites covered with PTFE was significantly less than the area of adhesion when covered by oxidized regenerated cellulose (P : ; Fig. ). The percentage of improvement in area of adhesion was significantly greater (P : ) when PTFE covered the sidewall site than when oxidized regenerated cellulose was used, 86.0% ± 0.06% versus 7.% ± 0.11%, respectively. Adhesion Occurrence The number of sidewalls covered by PTFE without any adhesions whatsoever, 1 of 9 sites (7.%), was significantly less (P < 0.03) than when the sidewalls were covered with oxidized regenerated cellulose, 7 of 9 sites (%). When the number of patients with adhesions on the side with oxidized regenerated cellulose but not on the side with PTFE was compared with the number of sidewalls with adhesions to the side with PTFE and not on the side with oxidized regenerated cellulose (Table ), there were significantly more sidewalls covered with PTFE without adhesions (16 versus ). Retrieval of the Barriers Only of 9 patients had samples of oxidized regenerated cellulose identifiable at second-look lapa- 10 Haney et a1. Barriers and adhesion reformation DISCUSSION The mechanisms of postsurgical peritoneal adhesion formation and reformation remain poorly understood. The experimental evidence suggests that adhesions form between two surgically traumatized surfaces in natural apposition during the healing interval because it is more efficient to combine two sites of tissue repair into a single healing site, resulting in coalescing adhesions between peritoneal surfaces (1). When the pelvic viscera are involved, these adhesions have the potential to impair physiologic functions, such as oocyte collection, and result in infertility (1, ). Our results demonstrate that although the use of either of the currently available barriers is associated with a significant reduction in postsurgical adhesions at the pelvic sidewall after adhesiolysis, PTFE is associated with fewer adhesions than oxidized regenerated cellulose. The clinical data supporting this conclusion is experimentally sound because each barrier was assigned randomly to one sidewall in each patient, allowing for paired statistical evaluation, and because the randomization was successful in creating comparable experimental groups with regard to the adhesions present initially. This study was not designed to evaluate the efficacy of the individual barriers compared with surgery alone but rather their relative likelihood of being associated with postoperative adhesions after reconstructive surgery. However, these findings are Table X Table Based on the Presence or Absence of Adhesions at Experimental Sites* Oxidized regenerated cellulose Adhesions present Adhesions absent * Fischer's Exact test, P :=; Adhesions present 6 PTFE Adhesions absent 16 5 Fertility and Sterility

5 in continued agreement with the preliminary findings of this study reported in 199, in which PTFE showed 89.75% reduction in adhesion formation versus 36.8% for oxidized regenerated cellulose (15). This study design could not address the question of whether the surgery required for removal of the PTFE could result in adhesions. In theory, surgical barriers separate contacting peritoneal surfaces during healing to enable reconstitution ofthe surgically traumatized surfaces separately without coalescing adhesions. This presupposes that the barrier itself is not injurious to the peritoneum or, in some other fashion, adversely affects the wound healing process. The reason for the observed differences between the two barriers may relate to the intrinsic nature of the materials and the tissue response they elicit. Oxidized regenerated cellulose is relatively caustic and causes sloughing of the surface layer of mesothelial cells (16), resulting in an obligatory injury to all contacting peritoneal surfaces. It is uncertain whether this is due to the acidity of the material or the activated leukocytes responsible for degradation, but the peritoneum underlying oxidized regenerated cellulose is broken down and infiltrated with leukocytes. Not surprisingly, in view of its rapid degradation, oxidized regenerated cellulose elicits a peritoneal fluid inflammatory exudate characterized by large numbers of activated macrophages. By contrast, PTFE causes little or no morphological alteration of the adjacent peritoneum and does not elicit a large number of activated peritoneal fluid leukocytes (17). Thus, although both barriers initially may separate opposing peritoneal surfaces mechanically, the oxidized regenerated cellulose injures contacting peritoneum, and the leukocytes responsible for degrading the oxidized regenerated cellulose may function as a cellular bridge leading to a common healing site and coalescing adhesions. With PTFE, the absence of peritoneal injury, its impervious nature, and the lack of leukocyte-mediated degradation may allow a more normal reconstitution of injured peritoneal surfaces separately without coalescing adhesions. Interestingly, in some animal experiments, oxidized regenerated cellulose prevented adhesion formation (18, 19), whereas in others it did not (0-). In mice, it caused de novo adhesions and failed to prevent either adhesion formation or reformation after surgical injury (1, 16). In the only report in which the two barriers were compared directly, PTFE prevented postsurgical adhesion formation and reformation whereas oxidized regenerated cellulose did not (3). Incomplete hemostasis at surgical sites also may have differing consequences for the two barrier materials. Oxidized regenerated cellulose is a procoagu- lant and functions as a surface hemostat. When hemostasis is incomplete, fibrin deposits at the site of injury increases the likelihood of adhesions (). By contrast, failure to achieve complete hemostasis does not seem to alter the effectiveness ofptfe (3). Hemostasis was achieved in all the patients in the present study and could not account for the differences observed. Nevertheless, hemostasis remains a significant clinical concern as the achievement of complete hemostasis at surgery is difficult, particularly with operations such as myomectomy. The major drawback associated with permanent barrier materials such as PTFE is the reluctance of clinicians to leave them in place, preferring to remove the barrier after peritoneal healing is complete. Expanded polytetrafluoroethylene is one ofthe least reactive of several commonly used medicalgrade polymers and, in other applications, such as a pericardial substitute or vascular graft, PTFE has been left in place for many years without any adverse effects. Whether it will be beneficial to remove PTFE from the pelvis when it is placed in such a manner that adnexal anatomy is not altered remains to be determined. Because it is not demonstrably harmful to gametes, as judged by the lack of effect on human sperm motility, hamster egg penetration, and mouse embryo development in vitro (5), it may be just as rational a strategy to leave the PTFE barrier in place and avoid the additional surgery required for removal. REFERENCES 1. Bronson RA, Wallach EE. Lysis of peri adnexal adhesions for correction of infertility. Fertil Steril 1977;8: Trimbos-Kemper T, Trimbos B, van Hall E. Etiological factors in tubal infertility. Fertil Steril 198;37: Trimbos-Kemper TCM, Trimbos JB, van Hall EV. Adhesion formation after tubal surgery: results of the eight-day laparoscopy in 188 patients. Fertil Steril 1985;3: Surrey MW, Freidman S. Second-look laparoscopy after reconstructive pelvic surgery for infertility. J Reprod Med 198; 7 : DeCherney AH, Mezer HC. The nature ofposttuboplasty pelvic adhesions as determined by early and late laparoscopy. Fertil Steril 198;1: Pittaway DE, Daniell JF, Maxson WS. Ovarian surgery in an infertility patient as an indication for a short-interval secondlook laparoscopy: a preliminary study. Fertil Steril 19885;: Diamond MP, Daniell JF, Feste J, Surrey MW, McLaughin DS, Freidman S, et al. Adhesion reformation and de novo adhesion formation after reproductive pelvic surgery. Fertil Steril 1987;7: Jansen RPS. Failure of intraperitoneal adjuncts to improve the outcome of pelvic operations in young women. Am J Obstet Gynecol 1985; 153: di Zerega GS. Contemporary adhesion prevention. Fertil Steril 199;61: Interceed (TC7) Adhesion Barrier Study Group. Prevention of postsurgical adhesions by Interceed, an absorbable Haney et ai. Barriers and adhesion reformation 105

6 adhesion barrier: a prospective, randomized multicenter clinical study. Fertil Steril 1989;51: Interceed (TC7) Adhesion Barrier Study Group II. Pelvic sidewall adhesion reformation: microsurgery alone or with Interceed (TC7) absorbable adhesion barrier. Surg Gynecol Obstet 1993; 177: Sekiba K, The Obstetrics and Gynecology Adhesion Prevention Committee. Use ofinterceed (TC7) absorbable adhesion barrier to reduce postoperative adhesion reformation in infertility and endometriosis surgery. Obstet Gynecol 199; 79: The Surgical Membrane Study Group. Prophylaxis of pelvic sidewall adhesions with Gore-Tex Surgical Membrane: a multicenter clinical investigation. Fertil Steril 199;57: Haney AF, Doty E. The formation of coalescing peritoneal adhesions requires injury to both contacting peritoneal surfaces. Fertil Steril 199;61: The Surgical Membrane Study Group. Prophylaxis of pelvic sidewall adhesions with Gore-Tex Surgical Membrane: a multicenter clinical Investigation. Fertil Steril 199;57: Haney AF, Doty E. Peritoneal injury and de novo adhesion formation caused by oxidized regenerated cellulose (Interceed ) but not expanded polytetrafiuoroethylene (Gore Tex Surgical Membrane) in a murine model. Fertil Steril 199;57: Haney AF, Doty E. Comparison ofthe peritoneal cells elicited by oxidized regenerated cellulose (lnterceed ) and expanded polytetrafiuoroethylene (Gore-Tex Surgical Membrane) in a murine model. Am J Obstet Gynecol 199; 166: Linsky CB, Diamond MP, Cunningham T, Constantine B, DeCherney AH, di Zerega GS. Adhesion reduction in the rabbit uterine horn model using an absorbable barrier, TC-7. J Reprod Med 1987;3: Diamond MP, Linsky CB, Cunningham T, Constantine B, di Zerega GS, DeCherney AH. Development of a model for sidewall adhesions in the rabbit and their reduction by an absorbable barrier. Microsurgery 1987;8: Best CL, Rittenhouse D, Sueldo CEo A comparison of TC7 and 3% dextran 70 for prevention of postoperative adhesions in hamsters. Obstet GynecoI1991;78: Pagidas K, Tulandi T. Effects of Ringer's lactate, Interceed (TC7) and Gore-Tex Surgical Membrane on postsurgical adhesion formation. Fertil Steril 199;57: Maxson WS, Herbert CM, Olfield EL, Hill GA. Efficacy of a modified oxidized cellulose fabric in the prevention of adhesion formation. Gynecol Obstet Invest 1988;6: Haney AF, Doty E. Expanded-polytetrafiuoroethylene but not oxidized regenerated cellulose prevents adhesion formation and reformation in a mouse uterine horn model of surgical injury. Fertil Steril 1993;60: Linsky CB, Diamond MP, Cunningham T, DeCherney AH, di Zerega GS. Effect of blood on the efficacy of barrier adhesion reduction in the rabbit uterine horn model. Infertility 1988; 11: Hunter SK, Scott JR, Hull D, Urry RL. The gamete and embryo compatibility of various synthetic polymers. Fertil Steril 1988;50: Haney et al. Barriers and adhesion reformation Fertility and Sterility