- Email: [email protected]
Adhesion formation after ovarian wound repair in New Zealand white rabbits: A comparison of ovarian microsurgical closure with ovarian nonclosure
Adhesion formation after ovarian wound repair in New Zealand white rabbits: A comparison of ovarian microsurgical closure with ovarian nonclosure Anne K. Wiskind, MD, Andrew A. Toledo, MD, A. Gatewood Dudley, MD, and Kathryn Zusmanis, BS Atlanta, Georgia Thirty female New Zealand white rabbits underwent standard laparotomy. Each ovary was bivalved and hemostasis was achieved with bipolar electrocautery. One ovary was then randomized to the closure group, whereas the other ovary was placed in the nonclosure group. In the closure group, the ovarian capsule was closed with a continuous suture of 8-0 Vicryl absorbable surgical suture material with microsurgical technique. In the nonclosure group, the ovaries were left open. Three weeks later the animals were killed and the ovarian adhesions were graded with a standardized scale by an observer (A. A. T.) blinded to the closure status of the animals. Five control animals underwent sham operations with minimal adhesion formation. Statistical analysis of the study animals by the paired Student t test showed a significantly higher adhesion score on the ovaries that were microsurgically closed compared with the ovaries not closed (p = 0.02). (AM J OesTET GVNECOL 1990;163:1674-8.)
Key words: Ovarian wound, nonclosure, adhesion formation
Postoperative adhesion formation is one of the most unfortunate sequelae of gynecologic surgery and results in mechanical infertility, pelvic pain, and bowel obstruction. Adhesion formation after reconstructive ovarian surgery is particularly frustrating because it may have an important impact on future fertility. Adhesions coating the ovary may interfere with ovulation mechanics and distort tuboovarian relationships, thereby disrupting ovum transport. I Many of the studies that have evaluated both the pathophysiology of adhesions and methods to deter their formation have involved the adhesiogenesis of peritoneal defects with relatively little attention focused on the ovary. However, procedures resulting in ovarian surgical wounds are commonly encountered in women of reproductive age who are then most likely to be adversely affected by the subsequent adhesions. Examples of such procedures include ovarian cystectomy, resection of endometriosis, wedge resection because of polycystic ovarian disease, and rarely, resection of an ovarian pregnancy. The least adhesiogenic method of managing ovarian From the Department of Gynecology and Obstetrics, Emory University School of Medicine. Funding was provided by a grant from the Emory Medical Care Foundation of Emory University. Received for publication April 9, 1990; revised June 29, 1990; accepted July 17, 1990. Reprint requests: Anne K. Wiskind, MD, Department of Gynecology and Obstetrics, Emory University School of Medicine, Atmnta, GA 30303. 6/1 /23863
1674
surgical defects remains to be determined. Previous studies showed that microsurgical closure with fine synthetic suture is superior to macrosurgical closure with chromic catgut suture. 2 Another stud y3 showed that bipolar electrocautery of ovarian surgical defects is less adhesiogenic than macrosurgical closure. However, bipolar electrocautery of ovaries alone has been shown to cause significant adhesions. 3 , 4 Thus far there has been no study that compares adhesion formation after ovarian closure using microsurgical technique with ovarian nonclosure. This was the purpose of our study. Material and methods
The study was approved by the Institutional Animal Care and Use Committee of Emory University, PROTOCOL 125-89, and the animals were housed at Emory University under the supervision of university veterinary staff. Thirty female New Zealand white rabbits of reproductive age that weighed 2500 to 4300 gm were anesthetized with ketamine hydrochloride 40 mg/kg and xylazine hydrochloride 10 mg/kg body weight. The abdomen was shaved and prepared with povidoneiodine solution, and the animal was draped with sterile sheets for abdominal surgery. Talc was carefully removed from the surgeon's gloves. The animals then underwent laparotomy through a midline abdominal incision approximately 4 to 5 cm long beginning just inferior to the umbilicus. The bowels were carefully packed in the upper abdomen with a moistened lap sponge and a pediatric Balfour retractor. With an op-
Volume 163 Number 5, Part I
erating microscope (modelI402,J.C. Hoppl Corp., distributed by Edward Weck Co., Research Triangle Park), the ovaries were each bivalved longitudinally to the level of the hilum with a No. 15 blade scalpel. Hemostasis of the ovary was achieved with bipolar electrocautery provided by a surgical cautery unit (Micro 120, Aspen Labs, Inc., Littleton, Colo.) on setting five. Once this was completed on both ovaries, a card was pulled from a sealed envelope and one ovary was assigned to the closure group and the other ovary was assigned to the nonclosure group. In the closure group, the ovarian capsule was closed with a continuous, nonlocking suture of 8-0 Vicryl absorbable surgical suture material (Ethicon, Somerville, N.J.) on a V130-S tapercut needle. Care was taken to include just the ovarian capsule with minimal cortex in each bite of suture. The rabbit ovarian surface was quite friable, requiring that some cortex be included to achieve a smooth closure. The suture bites were no more than 1 mm in depth and were spaced about 1 mm apart for approximately seven to nine bites per ovary. In the nonclosure group the ovary was left open. The randomization of closure and nonclosure was equally distributed between the right and left ovaries. Careful attention to meticulous microsurgical technique including gentle tissue handling, excellent hemostasis, and continuous irrigation with sterile saline solution was performed throughout the procedure. The Balfour retractor and laparotomy sponge were removed, the abdominal wall was closed with a continuous suture of 3-0 Vicryl absorbable surgical suture material, and the skin was closed with a subcuticular stitch of 4-0 Prolene sutures (Ethicon, Somerville, N.J.) reinforced by simple interrupted sutures of 4-0 Prolene. Descriptions of all operative procedures and notes of special problems were kept in a laboratory diary. Five control animals also underwent laparotomy in the same fashion with visualization, mobilization, and handling of the ovaries with microforceps, but ovarian incisions, electrocautery, or sutures were not used. All animals were allowed to recover routinely. Three weeks after the surgery, the animals were killed with a thiopental preparation (Pentothal) 60 mg/kg body weight administered intravenously. A second laparotomy was performed by an investigator blinded to the group status (A. A. T.) who had not been involved in any of the initial surgeries, and the pelvic adhesions were graded by a method suggested by Michael P. Diamond, MD (personal communication). A score of 0 to 1.0 was given for the percentage of ovarian surface involved with adhesions. Separate scores were given to the incision line, the lateral ovarian surface, and the medial ovarian surface, which were then added for a total adhesion score for each ovary (Table I, Fig. I). Standard analysis of the data was performed with the paired Student t test.
Adhesion formation after ovarian wound repair
1675
Fig. 1. Ovarian adhesion scale. A, Represents medial surface of rabbit ovary; B, incision line, and C, lateral surface. Percentage of each of these surfaces involved with adhesions is scored individually, then added for total adhesion score for ovary.
Table I. Adhesion score scale Description of ovary
I
Potential score
Per.centage of incision covered with adheSlons
0-1.00
Percentage of lateral ovarian surface covered with adhesions
0-1.00
Percentage of medial ovarian surface covered by adhesions
O-l.OO
Total adhesion score for each ovary
0-3.00
(Diamond MP. Unpublished data.)
Results For statistical analysis the animals were divided into two groups: group A (ovarian microsurgical closure) and group B (ovarian nonclosure). Comparison oftotal
1676 Wiskind et al.
adhesion scores in group A (closure) and group B (nonclosure), summarized in Table II, shows a higher mean adhesion score in group A than that in group B, which is statistically significant to a p value of 0.02. The five control animals were evaluated separately with no adhesion formation in three animals and minimal adhesions in the other two, suggesting that opening and closing the peritoneum alone contributed little to adhesion formation (Table III). There were also 14 ovaries (23%) in the study animals with an adhesion score of zero. Eight of these were in the nonclosure group, whereas six were in the closure group. Four animals (13%) had no adhesions on either ovary. Graphic plotting of the ovarian adhesion scores from each animal (Fig. 2) shows the higher adhesion scores in the ovarian closure group, with the exception of five animals in which ovarian nonclosure resulted in more adhesions that that of ovarian closure. After reviewing these cases it was found that in one animal the adhesions in both ovaries were minimal; however, the ovary that was closed underwent severe atrophy with subsequently fewer adhesions being formed than in the opposite normal-sized ovary that was not closed. One of the nonclosure ovaries in another animal was fixed deep in the pelvic sidewall. The mobilization necessary to position the ovary for bivalving and bipolar cautery was very difficult and traumatic. In fact, the ovary was separated into three pieces during the procedure, eliciting severe postoperative adhesion formation. In the remaining three animals, there was nothing unusual about the operations to indicate why nonclosure was more adhesiogenic than ovarian closure. It was the opinion of the statistician that these animals should be included in the study group, inasmuch as five of 30 were too many to exclude as outlying data points from a statistical standpoint. Furthermore, it was not considered statistically sound to exclude only one from this group even though excessive trauma was noted. Marked atrophy was an interesting finding at second laparotomy in eight ovaries. These were evenly distributed among ovaries that were closed and those that were not closed. It was suspected to have occurred as a result of excessive bipolar cautery compromising the ovarian blood supply. However, it was also noted in cases in which only minimal cautery was used. The depth of the hilar incision was kept fairly consistent, but may be indirectly related to ovarian atrophy, inasmuch as the deeper the incision the more cautery required to achieve hemostasis. In this study 11 ovaries were found encased in dense bowel adhesions. All but one of these ovaries had undergone microsurgical closure. The remaining ovary encased in dense adhesions was not closed, but had been mutilated by excessive trauma during the operative procedure.
November 1990 Am J Obstet Gyneco1
Four animals had superficial wound separations. There was no evidence of postoperative evisceration or peritonitis in any of the animals, and they responded to treatment of local wound care. The operating time was approximately 1 hour for most animals, with the exception of the control animals that had an operating time of about 30 minutes.
Comment In this study we showed that nonclosure of ovarian surgical defects is less adhesiogenic than microsurgical closure in the rabbit model. Most of the adhesions in the closure group involved the incision line; with 100% of the incision line involved in one third of the closed ovaries, and> 50% of the incision line was involved in over halfthese ovaries. All but one case of bowel adhesions to the adnexa involved ovaries in the closure group. In contrast, the incision line was much less frequently involved in adhesions in the nonclosure ovaries. This is consistent with previous studies that showed suture to elicit a foreign body tissue reaction with subsequent adhesiogenesis.!>-7 The paucity of adhesions to the incision line in the nonclosure ovaries suggests that the cauterized ovarian parenchyma did not elicit a significant inflammatory response. Trauma to tissue during the operative procedure also appeared to be a significant adhesiogenic factor. Most of these adhesions in the nonclosure ovaries involved the lateral and medial surfaces of the ovary and were most likely a result of trauma to the ovarian surface during manipulation. Instances in which excessive trauma to the tissue was noted during the operative procedure resulted in extensive postoperative adhesion formation. In fact, a partial explanation for the greater adhesions scores in the closure group may be the increased amount of manipulation required for surgical closure. Thus gentle, atraumatic tissue handling is of paramount importance in decreasing adhesion formation. 1.7-10 Preservation of the ovarian blood supply also is critical to ensure satisfactory results after ovarian reconstructive surgery. Compromise of the blood supply can result from certain types of surgical closure 5 • 8 and from excessive electrocautery, as we suspect may have occurred in some of our animals with ovarian atrophy. Failure to preserve a functioning ovary negates the purpose of conservative ovarian surgery. In applying this study in the rabbit model to the human patient, several factors must be considered. First, the rabbit ovary is much smaller and more friable than that in human beings, measuring approximately 0.8 X 0.5 X 0.3 cm, making subcapsular closure technically difficult, if not impossible, even with the operating microscope. A subcapsular closure that is possible on human ovaries may be less adhesiogenic than the
Adhesion formation after ovarian wound repair
Volume 163 Number 5, Part I
1677
line of equality
3.0 2.8 2.6 2.4 2.2
e
:l til
2.0
0
1.8
C
1.6
Z
1.4
U, 0
C
as .;: as > 0
1.2 1.0 0.8 0.6 0.4 0.2
0.0 0.2 0.4
0.6 0.8 1.0 1.2
1.4
1.6
1.8 2.0 2.2 2.4
2.6 2.8 3.0
Ovarian Closure Fig. 2. Graphic representation of adhesion scores for each study animal. !:J. Represents four animals with adhesion score of zero on both ovaries.• Represents 21 animals with adhesion score of microsurgically closed ovary greater than adhesion score of nonclosure ovaries .• Illustrates five animals with nonclosure ovary adhesion score greater than adhesion score of ovary microsurgically closed.
continuous suture through the ovarian capsule used in this study.5 Furthermore, after the surgical defects were created in the rabbit ovaries, the two halves of the ovaries naturally fell together, restoring their natural shape and appearance. In contrast, after many ovarian procedures in human beings, especially after resection of a cyst or an endometrioma, the remaining ovary is often a markedly distorted, thin, elongated shell. Whether adequate blood perfusion, hormone production, and tuboovarian relationships can be reestablished by this distorted ovary is yet to be determined. In conclusion, our data show that ovarian microsurgical closure is significantly more adhesiogenic than ovarian nonclosure (p = 0.02). On the basis of these results, we suggest that for smaller surgical defects in which the ovarian architecture is not significantly altered, nonclosure of the defect is the management of choice. In larger defects in which the ovary is markedly distorted with a large denuded surface, the most appropriate management is unclear. At this time we recommend surgical reconstruction of the ovary by internal closure with a fine synthetic suture. With this technique the edges of the ovarian capsule can be reapproximated and left open. An additional layer of
Table II. Adhesion scores in ovarian microsurgical closure (group A) versus ovarian nonclosure (group B) Mean adhesion score
Group A Group B Mean difference
p Value
=
1.21 (SD ± 1.12) 0.67(SD ± 0.79) 0.543 (SD ± 1.25)
0.02.
Table III. Adhesion scores in control animals Control animal
I
2
3
4
5
Adhesion score
o o
0.10 0.20
o
suture to close the ovarian surface is not only unnecessary, but more adhesiogenic. Special emphasis should be placed on atraumatic tissue handling to further minimize adhesion formation.
678
Wiskind et al.
November 1990 Am J Obstet Gynecol
REFERENCES 1. Levinson CJ, Swolin K. Postoperative adhesions: etiology,
2. 3.
4. 5.
prevention and therapy. Clin Obstet Gynecol 1980;23: 1213-20. Eddy CA, Asch RH, Balmaceda JP. Pelvic adhesions following microsurgical and macrosurgical wedge resection of the ovaries. Fertil Steril 1980;33:557-61. Rewis JR, Smith RP, Gallup DG. A comparison of severity of adhesion formation after two methods of ovarian wound repair in the rabbit model. In: Proceedings of the Society of Gynecologic Surgeons Meeting, New Orleans, Louisiana, March 1989. Awadalla SG, Mattox JH, Slickenmyer WJ. The effect of cauterization of the rabbit model on adhesion formation. Fertil Steril 1986;46:696-700. Oeslner G, Grabe RA, Boyers SP, Pan SB, Barnes ER, DeCherney AH. A comparison of three techniques for
6.
7. 8. 9. 10.
ovarian reconstruction. AM J OBSTET GYNECOL 1986; 154:569-72. van Ryssel EJC, Brand R, Admiraal C, Smith I, Trimbos JB. Tissue reaction and surgical knots: the effect of suture size, knot configuration, and knot volume. Obstet Gynecol 1989;74:64-8. Stangel JJ, Nisbet JD, Settles H. Formation and prevention of postoperative abdominal adhesions. J Reprod Med 1983;29: 143-56. Ellis H. The cause and prevention of postoperative intraperitoneal adhesions. Surg Gynecol Obstet 1971; 133:497511. Jackson BB. Observations on intraperitoneal adhesions. Surgery 1958;44:507-14. Buttram VC, Vaquero C. Post-ovarian wedge resection adhesive disease. Ferti! Steril 1975;26:874-6.
Key words: Ovarian wound, nonclosure, adhesion formation
Postoperative adhesion formation is one of the most unfortunate sequelae of gynecologic surgery and results in mechanical infertility, pelvic pain, and bowel obstruction. Adhesion formation after reconstructive ovarian surgery is particularly frustrating because it may have an important impact on future fertility. Adhesions coating the ovary may interfere with ovulation mechanics and distort tuboovarian relationships, thereby disrupting ovum transport. I Many of the studies that have evaluated both the pathophysiology of adhesions and methods to deter their formation have involved the adhesiogenesis of peritoneal defects with relatively little attention focused on the ovary. However, procedures resulting in ovarian surgical wounds are commonly encountered in women of reproductive age who are then most likely to be adversely affected by the subsequent adhesions. Examples of such procedures include ovarian cystectomy, resection of endometriosis, wedge resection because of polycystic ovarian disease, and rarely, resection of an ovarian pregnancy. The least adhesiogenic method of managing ovarian From the Department of Gynecology and Obstetrics, Emory University School of Medicine. Funding was provided by a grant from the Emory Medical Care Foundation of Emory University. Received for publication April 9, 1990; revised June 29, 1990; accepted July 17, 1990. Reprint requests: Anne K. Wiskind, MD, Department of Gynecology and Obstetrics, Emory University School of Medicine, Atmnta, GA 30303. 6/1 /23863
1674
surgical defects remains to be determined. Previous studies showed that microsurgical closure with fine synthetic suture is superior to macrosurgical closure with chromic catgut suture. 2 Another stud y3 showed that bipolar electrocautery of ovarian surgical defects is less adhesiogenic than macrosurgical closure. However, bipolar electrocautery of ovaries alone has been shown to cause significant adhesions. 3 , 4 Thus far there has been no study that compares adhesion formation after ovarian closure using microsurgical technique with ovarian nonclosure. This was the purpose of our study. Material and methods
The study was approved by the Institutional Animal Care and Use Committee of Emory University, PROTOCOL 125-89, and the animals were housed at Emory University under the supervision of university veterinary staff. Thirty female New Zealand white rabbits of reproductive age that weighed 2500 to 4300 gm were anesthetized with ketamine hydrochloride 40 mg/kg and xylazine hydrochloride 10 mg/kg body weight. The abdomen was shaved and prepared with povidoneiodine solution, and the animal was draped with sterile sheets for abdominal surgery. Talc was carefully removed from the surgeon's gloves. The animals then underwent laparotomy through a midline abdominal incision approximately 4 to 5 cm long beginning just inferior to the umbilicus. The bowels were carefully packed in the upper abdomen with a moistened lap sponge and a pediatric Balfour retractor. With an op-
Volume 163 Number 5, Part I
erating microscope (modelI402,J.C. Hoppl Corp., distributed by Edward Weck Co., Research Triangle Park), the ovaries were each bivalved longitudinally to the level of the hilum with a No. 15 blade scalpel. Hemostasis of the ovary was achieved with bipolar electrocautery provided by a surgical cautery unit (Micro 120, Aspen Labs, Inc., Littleton, Colo.) on setting five. Once this was completed on both ovaries, a card was pulled from a sealed envelope and one ovary was assigned to the closure group and the other ovary was assigned to the nonclosure group. In the closure group, the ovarian capsule was closed with a continuous, nonlocking suture of 8-0 Vicryl absorbable surgical suture material (Ethicon, Somerville, N.J.) on a V130-S tapercut needle. Care was taken to include just the ovarian capsule with minimal cortex in each bite of suture. The rabbit ovarian surface was quite friable, requiring that some cortex be included to achieve a smooth closure. The suture bites were no more than 1 mm in depth and were spaced about 1 mm apart for approximately seven to nine bites per ovary. In the nonclosure group the ovary was left open. The randomization of closure and nonclosure was equally distributed between the right and left ovaries. Careful attention to meticulous microsurgical technique including gentle tissue handling, excellent hemostasis, and continuous irrigation with sterile saline solution was performed throughout the procedure. The Balfour retractor and laparotomy sponge were removed, the abdominal wall was closed with a continuous suture of 3-0 Vicryl absorbable surgical suture material, and the skin was closed with a subcuticular stitch of 4-0 Prolene sutures (Ethicon, Somerville, N.J.) reinforced by simple interrupted sutures of 4-0 Prolene. Descriptions of all operative procedures and notes of special problems were kept in a laboratory diary. Five control animals also underwent laparotomy in the same fashion with visualization, mobilization, and handling of the ovaries with microforceps, but ovarian incisions, electrocautery, or sutures were not used. All animals were allowed to recover routinely. Three weeks after the surgery, the animals were killed with a thiopental preparation (Pentothal) 60 mg/kg body weight administered intravenously. A second laparotomy was performed by an investigator blinded to the group status (A. A. T.) who had not been involved in any of the initial surgeries, and the pelvic adhesions were graded by a method suggested by Michael P. Diamond, MD (personal communication). A score of 0 to 1.0 was given for the percentage of ovarian surface involved with adhesions. Separate scores were given to the incision line, the lateral ovarian surface, and the medial ovarian surface, which were then added for a total adhesion score for each ovary (Table I, Fig. I). Standard analysis of the data was performed with the paired Student t test.
Adhesion formation after ovarian wound repair
1675
Fig. 1. Ovarian adhesion scale. A, Represents medial surface of rabbit ovary; B, incision line, and C, lateral surface. Percentage of each of these surfaces involved with adhesions is scored individually, then added for total adhesion score for ovary.
Table I. Adhesion score scale Description of ovary
I
Potential score
Per.centage of incision covered with adheSlons
0-1.00
Percentage of lateral ovarian surface covered with adhesions
0-1.00
Percentage of medial ovarian surface covered by adhesions
O-l.OO
Total adhesion score for each ovary
0-3.00
(Diamond MP. Unpublished data.)
Results For statistical analysis the animals were divided into two groups: group A (ovarian microsurgical closure) and group B (ovarian nonclosure). Comparison oftotal
1676 Wiskind et al.
adhesion scores in group A (closure) and group B (nonclosure), summarized in Table II, shows a higher mean adhesion score in group A than that in group B, which is statistically significant to a p value of 0.02. The five control animals were evaluated separately with no adhesion formation in three animals and minimal adhesions in the other two, suggesting that opening and closing the peritoneum alone contributed little to adhesion formation (Table III). There were also 14 ovaries (23%) in the study animals with an adhesion score of zero. Eight of these were in the nonclosure group, whereas six were in the closure group. Four animals (13%) had no adhesions on either ovary. Graphic plotting of the ovarian adhesion scores from each animal (Fig. 2) shows the higher adhesion scores in the ovarian closure group, with the exception of five animals in which ovarian nonclosure resulted in more adhesions that that of ovarian closure. After reviewing these cases it was found that in one animal the adhesions in both ovaries were minimal; however, the ovary that was closed underwent severe atrophy with subsequently fewer adhesions being formed than in the opposite normal-sized ovary that was not closed. One of the nonclosure ovaries in another animal was fixed deep in the pelvic sidewall. The mobilization necessary to position the ovary for bivalving and bipolar cautery was very difficult and traumatic. In fact, the ovary was separated into three pieces during the procedure, eliciting severe postoperative adhesion formation. In the remaining three animals, there was nothing unusual about the operations to indicate why nonclosure was more adhesiogenic than ovarian closure. It was the opinion of the statistician that these animals should be included in the study group, inasmuch as five of 30 were too many to exclude as outlying data points from a statistical standpoint. Furthermore, it was not considered statistically sound to exclude only one from this group even though excessive trauma was noted. Marked atrophy was an interesting finding at second laparotomy in eight ovaries. These were evenly distributed among ovaries that were closed and those that were not closed. It was suspected to have occurred as a result of excessive bipolar cautery compromising the ovarian blood supply. However, it was also noted in cases in which only minimal cautery was used. The depth of the hilar incision was kept fairly consistent, but may be indirectly related to ovarian atrophy, inasmuch as the deeper the incision the more cautery required to achieve hemostasis. In this study 11 ovaries were found encased in dense bowel adhesions. All but one of these ovaries had undergone microsurgical closure. The remaining ovary encased in dense adhesions was not closed, but had been mutilated by excessive trauma during the operative procedure.
November 1990 Am J Obstet Gyneco1
Four animals had superficial wound separations. There was no evidence of postoperative evisceration or peritonitis in any of the animals, and they responded to treatment of local wound care. The operating time was approximately 1 hour for most animals, with the exception of the control animals that had an operating time of about 30 minutes.
Comment In this study we showed that nonclosure of ovarian surgical defects is less adhesiogenic than microsurgical closure in the rabbit model. Most of the adhesions in the closure group involved the incision line; with 100% of the incision line involved in one third of the closed ovaries, and> 50% of the incision line was involved in over halfthese ovaries. All but one case of bowel adhesions to the adnexa involved ovaries in the closure group. In contrast, the incision line was much less frequently involved in adhesions in the nonclosure ovaries. This is consistent with previous studies that showed suture to elicit a foreign body tissue reaction with subsequent adhesiogenesis.!>-7 The paucity of adhesions to the incision line in the nonclosure ovaries suggests that the cauterized ovarian parenchyma did not elicit a significant inflammatory response. Trauma to tissue during the operative procedure also appeared to be a significant adhesiogenic factor. Most of these adhesions in the nonclosure ovaries involved the lateral and medial surfaces of the ovary and were most likely a result of trauma to the ovarian surface during manipulation. Instances in which excessive trauma to the tissue was noted during the operative procedure resulted in extensive postoperative adhesion formation. In fact, a partial explanation for the greater adhesions scores in the closure group may be the increased amount of manipulation required for surgical closure. Thus gentle, atraumatic tissue handling is of paramount importance in decreasing adhesion formation. 1.7-10 Preservation of the ovarian blood supply also is critical to ensure satisfactory results after ovarian reconstructive surgery. Compromise of the blood supply can result from certain types of surgical closure 5 • 8 and from excessive electrocautery, as we suspect may have occurred in some of our animals with ovarian atrophy. Failure to preserve a functioning ovary negates the purpose of conservative ovarian surgery. In applying this study in the rabbit model to the human patient, several factors must be considered. First, the rabbit ovary is much smaller and more friable than that in human beings, measuring approximately 0.8 X 0.5 X 0.3 cm, making subcapsular closure technically difficult, if not impossible, even with the operating microscope. A subcapsular closure that is possible on human ovaries may be less adhesiogenic than the
Adhesion formation after ovarian wound repair
Volume 163 Number 5, Part I
1677
line of equality
3.0 2.8 2.6 2.4 2.2
e
:l til
2.0
0
1.8
C
1.6
Z
1.4
U, 0
C
as .;: as > 0
1.2 1.0 0.8 0.6 0.4 0.2
0.0 0.2 0.4
0.6 0.8 1.0 1.2
1.4
1.6
1.8 2.0 2.2 2.4
2.6 2.8 3.0
Ovarian Closure Fig. 2. Graphic representation of adhesion scores for each study animal. !:J. Represents four animals with adhesion score of zero on both ovaries.• Represents 21 animals with adhesion score of microsurgically closed ovary greater than adhesion score of nonclosure ovaries .• Illustrates five animals with nonclosure ovary adhesion score greater than adhesion score of ovary microsurgically closed.
continuous suture through the ovarian capsule used in this study.5 Furthermore, after the surgical defects were created in the rabbit ovaries, the two halves of the ovaries naturally fell together, restoring their natural shape and appearance. In contrast, after many ovarian procedures in human beings, especially after resection of a cyst or an endometrioma, the remaining ovary is often a markedly distorted, thin, elongated shell. Whether adequate blood perfusion, hormone production, and tuboovarian relationships can be reestablished by this distorted ovary is yet to be determined. In conclusion, our data show that ovarian microsurgical closure is significantly more adhesiogenic than ovarian nonclosure (p = 0.02). On the basis of these results, we suggest that for smaller surgical defects in which the ovarian architecture is not significantly altered, nonclosure of the defect is the management of choice. In larger defects in which the ovary is markedly distorted with a large denuded surface, the most appropriate management is unclear. At this time we recommend surgical reconstruction of the ovary by internal closure with a fine synthetic suture. With this technique the edges of the ovarian capsule can be reapproximated and left open. An additional layer of
Table II. Adhesion scores in ovarian microsurgical closure (group A) versus ovarian nonclosure (group B) Mean adhesion score
Group A Group B Mean difference
p Value
=
1.21 (SD ± 1.12) 0.67(SD ± 0.79) 0.543 (SD ± 1.25)
0.02.
Table III. Adhesion scores in control animals Control animal
I
2
3
4
5
Adhesion score
o o
0.10 0.20
o
suture to close the ovarian surface is not only unnecessary, but more adhesiogenic. Special emphasis should be placed on atraumatic tissue handling to further minimize adhesion formation.
678
Wiskind et al.
November 1990 Am J Obstet Gynecol
REFERENCES 1. Levinson CJ, Swolin K. Postoperative adhesions: etiology,
2. 3.
4. 5.
prevention and therapy. Clin Obstet Gynecol 1980;23: 1213-20. Eddy CA, Asch RH, Balmaceda JP. Pelvic adhesions following microsurgical and macrosurgical wedge resection of the ovaries. Fertil Steril 1980;33:557-61. Rewis JR, Smith RP, Gallup DG. A comparison of severity of adhesion formation after two methods of ovarian wound repair in the rabbit model. In: Proceedings of the Society of Gynecologic Surgeons Meeting, New Orleans, Louisiana, March 1989. Awadalla SG, Mattox JH, Slickenmyer WJ. The effect of cauterization of the rabbit model on adhesion formation. Fertil Steril 1986;46:696-700. Oeslner G, Grabe RA, Boyers SP, Pan SB, Barnes ER, DeCherney AH. A comparison of three techniques for
6.
7. 8. 9. 10.
ovarian reconstruction. AM J OBSTET GYNECOL 1986; 154:569-72. van Ryssel EJC, Brand R, Admiraal C, Smith I, Trimbos JB. Tissue reaction and surgical knots: the effect of suture size, knot configuration, and knot volume. Obstet Gynecol 1989;74:64-8. Stangel JJ, Nisbet JD, Settles H. Formation and prevention of postoperative abdominal adhesions. J Reprod Med 1983;29: 143-56. Ellis H. The cause and prevention of postoperative intraperitoneal adhesions. Surg Gynecol Obstet 1971; 133:497511. Jackson BB. Observations on intraperitoneal adhesions. Surgery 1958;44:507-14. Buttram VC, Vaquero C. Post-ovarian wedge resection adhesive disease. Ferti! Steril 1975;26:874-6.