- Email: [email protected]
Long-term analysis of peritoneal plasminogen activator activity and adhesion formation after surgical trauma in the rat model
FERTILITY AND STERILITY® Copyright
Vol. 66, No.6, December 1996 Printed on acid-free paper in U. S. A.
1996 American Society for Reproductive Medicine
long-term analysis of peritoneal plasminogen activator activity and adhesion formation after surgical trauma in the rat model
Erica A. Bakkum, M.D., Ph.D. *t+ Jef J. Emeis, Ph.D.§ Remco A. J. Dalmeijer, B.Sc.*
Clemens A. van Blitterswijk, Ph.D. * J. Baptist Trimbos, M.D., Ph.D.t Trudy C. M. Trimbos-Kemper, M.D., Ph.D.t
Leiden University Medical Center, Leiden, The Netherlands
Objective: Recent literature has shown that a common pathway in postsurgical adhesion formation is a transient reduction in local plasminogen activator activity, shortly after peritoneal trauma. This deficit in fibrinolysis permits deposited fibrin to become organized into fibrous, permanent adhesions. Although adhesion formation is a process that continues beyond the first postoperative days, long-term analysis of this theory has not been performed. Design: A standardized peritoneal adhesion model in the rat. Main Outcome Measure(s): Long-term analysis of the peritoneal fibrinolytic activity (extraction technique) was related to the extent of postsurgical adhesion formation, up to 1 year postoperatively. Result(s): Total and tissue plasminogen activator activity were significantly increased at days 3 and 8, and 1 month postoperatively. A mean adhesion percentage of 75% per peritoneal defect was found to persist throughout all evaluation times, which was directly related to the increase of fibrinolysis. Conclusion(s): In contrast to the classical concept that adhesion formation is related to a reduction in fibrinolysis, an impressive increase of the fibrinolysis was found to be associated with the persistence of adhesions. Fertil Steril® 1996;66:1018-22 Key Words: Postsurgical adhesions, fibrinolytic activity, peritoneal side wall model, rat
Injury to the peritoneum leads to the formation of a fibrinous exudate. This fibrinous exudate is part of the hemostatic process and aids in tissue repair by providing a matrix for invading fibroblasts and new vessels. Fibrin deposition is an essential component of normal tissue repair while resolution of this fibrin deposit is required to restore the preoperative conditions. An imbalance between these two processes is thought to lead to adhesion formation when the deposited fibrin is not removed effectively and fibrinous adhesions become organized into fibrous, permanent adhesions (1-3). Normal peritoneum has
Received October 16, 1996; revised and accepted July 23, 1996. * Biomaterials Research Group, Laboratory for Biocompatibility. t Biomaterials Research Group, Department of Gynecology. :j: Reprint requests: Erica A. Bakkum, M.D., Ph.D., Department of Gynecology, Groene Hart Hospital, GraafFlorisweg 77-79, 2805 HH Gouda, The Netherlands (FAX: 31-182-566449). § Gaubius Laboratory TNO-PG, Dutch organization for applied scientific research: prevention and health. 1018
Bakkum et al. Fibrinolysis and adhesions
an inherent fibrinolytic activity, as first described by Hartwell (4) and confirmed by other authors (5, 6). Damaged peritoneum possesses a reduced fibrinolytic capacity. As a result, plasminogen is not converted sufficiently into plasmin and lysis of fibrin will not take place; this process is described as being related to postsurgical adhesion formation (7 -11). The time course and return of this fibrinolytic capacity after peritoneal trauma was studied first by Raftery (9), who found a reduction in fibrinolytic capacity immediately after the initial trauma. In that study, fibrinolysis was measured in the layer of mesothelial cells located at the surface of the peritoneum, which means that fibrinolysis was detected only at the surface ofthe wound area. Therefore, this assay method measures only part of the fibrinolytic capacity, which is known to reside not only in the mesothelium but also in the submesothelial blood vessels (3, 6). In addition, it has been reported that regeneration of the mesothelium starts at the base of the wound (12). Evaluation of the fibrinolytic parameters therefore should be performed in the total Fertility and Sterility®
peritoneal layer. This was performed in a later study by Buckman et al. (8), who found a reduction in fibrinolysis with full-thickness peritoneal biopsies, although they only measured fibrinolysis during the first 96 hours postoperatively. The present study was undertaken to determine whether a standard peritoneal sidewall trauma, known to produce uniform and consistent adhesion formation (13), was associated with a depression of fibrinolysis in the peritoneal layer, especially at longer postoperative intervals. Full-thickness biopsies were taken to ensure measurement of the fibrinolytic capacity of the entire depth of the peritoneal layer. Fibrinolytic capacity was evaluated at different postoperative time intervals, from directly postoperative to :=;1 year. Total plasminogen activator activity, tissue plasminogen activator activity, and urokinase plasminogen activator activity were chosen as parameters to detect fibrinolytic capacity. In the presence of fibrin, tissue plasminogen activator activity and plasminogen will bind to fibrin. Subsequently, plasminogen is converted into plasmin, which can disintegrate fibrin. Urokinase plasminogen activator activity has a similar working mechanism but does not require fibrin as a cofactor and is probably more important in extracellular proteolytic processes. Tissue plasminogen activator activity has a significant role in the modulation of intravascular thrombi. The observed changes in the fibrinolytic capacity were related to the extent of postsurgical adhesion formation. MATERIALS AND METHODS
Twenty-four female Wistar rats of reproductive age (Harlan CPB, Zeist, The Netherlands), weighing 180 to 200 g, were anesthetized with ether. A standard adhesion model was used, as described previously, which comprised bilateral excision and suturing of an oval-shaped parietal peritoneal defect. The defect measured 1.5 x 1.2 cm, with a depth of 3 mm, and encompassed the mesothelial and both underlying muscular layers (13). After 1, 3, and 8 days and 4, 12, 24, and 52 weeks, the rats were killed. By means ofthe three Vicryl sutures (Johnson and Johnson, Amersfoort, The Netherlands) with which the defect was closed, the defect was subdivided into eight areas of 12.5%. Adhesions were scored according to extent, as a percentage of the peritoneal defect covered with adhesions (13). At each interval, three rats were killed, thereby allowing six peritoneal defects per survival time to be evaluated. Three rats received an abdominal incision without additional peritoneal trauma and thus served as controls (n = 6). After scoring of adhesions, full-thickness biopsies Vol. 66, No.6, December 1996
from each defect (6 mm in diameter), including the mesothelial and the underlying muscular layers, were dissected with a dermal punch instrument, at a standard localization between the first and second suture. In the control animals without peritoneal defects, biopsies were taken at the same locations and are referred to as normal peritoneal fibrinolytic activity. After excision, the biopsies were frozen immediately and stored at -70°C. Subsequently, they were triturated under liquid nitrogen, resuspended in Camiolo's buffer (40 mg of wet tissue weight/mL) and extracted according to Padro et al. (14, 15). In the extracts, three parameters were determined spectrophotometrically, as described by Verheijen et al. (16); these parameters were the total plasminogen activator activity, tissue plasminogen activator activity, and urokinase plasminogen activator activity. Tissue plasminogen activator activity is the plasminogen activator activity quenched by anti-rat tissue plasminogen activator activity immunoglobulin (Ig)G as described by Padro et al. (15). Urokinase plasminogen activator activity is the activity quenched by anti-mouse urokinase plasminogen activator activity IgG (American Diagnostics, Greenwich, CT), which cross-reacts with rat urokinase plasminogen activator activity. All total plasminogen activator activities were corrected for the plasmin-inhibitory activity present in the extract (15) and will be expressed as units (U) of activity of 1 ng recombinant human tissue plasminogen activator activity) per g wet tissue weight. Total plasminogen activator activity represents tissue plasminogen activator activity and urokinase plasminogen activator activity together. In each individual peritoneal defect, the total plasminogen activator activity and tissue plasminogen activator activity subsequently were related to the extent of postsurgical adhesion formation. Statistical analysis was performed with the oneway analysis of variance and the Newman-Keuls multiple-range-test. Statistical significance was defined as P < 0.01. Data were expressed per peritoneal defect as mean adhesion percentage, mean total plasminogen activator activity, mean tissue plasminogen activator activity, and mean urokinase plasminogen activator activity (±SD). RESULTS Fibrinolytic Parameters
Normal peritoneal plasminogen activator activity as measured in the control animals, averaged 11.8 ± 4.5 U/g. The effect of peritoneal trauma on the plasminogen activator activity is depicted in Figure 1. At days 1 and 3, an increase was found (33.3 ± 9.2 Bakkum et a1. Fibrinolysis and adhesions
1019
250 200
~ .s
150
~
100
«
---f --- -
- ---
--- {
A mean adhesion percentage of 89.6% ± 5.1% was found at day 1 postoperatively, which increased slightly to 93.8% ± 6.8% on day 3. After day 3, a minimal decline was seen (Figs. 1 and 2). At 1 year, a mean adhesion percentage of 75% ± 13.7% was found per peritoneal defect. No statistical significant difference was found between any of the postoperative intervals.
50
Od
DISCUSSION
0"---...1----'-----'---"------'-----'--1d 3d ad 1m 3m 8m 1y Od time after surgery
Figure 1 Mean total plasminogen activator activity (U/g) and adhesion percentage (%) at different postoperative intervals after a standard peritoneal trauma. (- -) Mean total plasminogen activator activity; (- - -) mean adhesion percentage. P AA, total plasminogen activator activity. **p < 0.01 compared with control peritoneum (sham-operated animals).
and 99.9 ± 31.8 U/g; P < 0.01), with a further increase to 173.2 ± 69.7 U/g (P < 0.01) at day 8 postoperatively. From this interval onward, the activity decreased until the control values were reached at 3 months postoperatively (Fig. 1). The tissue plasminogen activator activity followed a similar time course (Fig. 2). Normal control peritoneal tissue plasminogen activator activity was 6.2 ± 3 U/g. Peritoneal trauma caused the tissue plasminogen activator activity to increase, with the highest level being 165.8 ± 69.4 U/g found at day 8 (P < 0.01). After day 8, a decline was observed, similar to the total plasminogen activator activity (Fig. 2). The urokinase plasminogen activator activity showed a different course. Normal peritoneal urokinase plasminogen activator activity was 5.8 ± 2.4 U/g. This activity fell to 3.6 ± 3.7 U/g on day 1 and increased to 7.9 ± 6.7 U/g on day 3, after which a decline was seen. Because urokinase plasminogen activator activity represents only an insignificant amount of the total plasminogen activator activity «5%), the urokinase plasminogen activator activity was omitted from further analysis. Adhesion Formation
For all survival times the mean adhesion percentage per peritoneal defect showed little variation, although the type changed markedly from filmy, easily separable to organized and vascular within 1 month postoperatively. Adhesion formation was confined to the peritoneal defects and no other noticeable complications were encountered. Adhesion percentages ranged from 25% to 100%, with a mean of 75%. 1020
Bakkum et at Fibrinolysis and adhesions
Failure of the peritoneal fibrinolytic system in case of postoperative ischemia is described to be an important factor in the formation of postsurgical adhesions (7-10, 17). When peritoneum is damaged, the plasminogen activator activity is reduced (6, 8, 10), through which deposited fibrin is not removed. This fibrin is organized subsequently into vascular, permanent adhesions (2, 18). Fibrinolytic activity in rats has been measured using either a single layer of cells (Raftery [9, 10], fibrin-slide technique) or by applying full-thickness biopsies of the peritoneal layer (Buckman et al. [7, 8], fibrin-plate technique). With both techniques, the fibrinolytic activity is measured as the resultant zone of lysis. Raftery (9, 10) observed that the fibrinolytic activity disappeared from the area of a peritoneal defect after 2 days of inflicting various traumas. At day 3, an irregular band was observed, indicating a partial recovery, and, at day 8, a wide band was detected that was even higher than normal peritoneum, indicating a full recovery. However, their study is limited by the use of only a single layer of cells, taken from the surface of the peritoneal wound area. In addition, inflammatory cells present at the surface, which do not have intrinsic fibrino-
250
200
!! S
150
'"
«
Il.
.:.
100
50
3d
ad
3m
1m
6m
ly
time alter surgery
Figure 2 Mean tissue plasminogen activator activity (U/g) and mean adhesion percentage (%) at different postoperative time intervals after a standard peritoneal trauma. (- -) Mean tissue plasminogen activator activity; (- - -) mean adhesion percentage. t-PA, tissue plasminogen activator activity. **p < 0.01 compared with control peritoneum (sham-operated animals).
Fertility and Sterility®
r
lytic capacity, might have affected the results. Buckman et al. (7, 8) observed a similar reduction in fibrinolysis. Abrasion trauma produced an immediate, significant decrease of total plasminogen activator activity, which was sustained at 48 and 72 hours, with some recovery by 96 hours. Crushing induced an immediate depression, with a maximum at 24 hours, and a return to normal by 72 hours. Deperitonealized surfaces did not show such depression and were found to be associated with a reduction in adhesions. Buckman et aL (7, 8) did not follow the parameters beyond 96 hours, although in both groups the fibrinolysis did not seem to have recovered completely. In our study, no depression of the fibrinolytic activity was found. Instead, an increase was found, both in total plasminogen and in tissue plasminogen activator activity. This increase differed significantly from control peritoneum from day 3 to 1 month postoperatively (Figs. 1 and 2). Mter 1 month, a decline was observed with a return to normal at 6 months. When the extent of adhesion formation was related to the total plasminogen and tissue plasminogen activator activity, no correlation was discerned (Pearson's correlation test). The slight, insignificant decrease in adhesion percentage at days 3, 8, and 1 month, might be concomitant with the increase in fibrinolysis after day 1. However, in earlier studies, this decrease was attributed to scar formation and retraction when the adhesions become progressively more organized postoperatively (19). Recently, other authors also have suggested that intraperitoneal (fluid phase) fibrinolysis is unaltered or even enhanced with infectious processes or endometriosis (20, 21). In peritoneal fluid of patients with or without endometriosis, or with or without adhesive disease, Batzofin et aL (20) found no difference. In a study on patients with pelvic inflammatory disease, Dorr (21) states that fibrinolysis occurs at a high rate, as measured in the peritoneal fluid and plasma. We were not able to measure fibrinolysis in the peritoneal fluid because of the minimal amount that can be harvested in the rat. The controversial finding that high levels of fibrinolysis were found up to 1 month postoperatively, concomitant with the persistence of adhesions, needs further evaluation. Our assay method might not measure the relevant plasminogen activator activity, in contrast to the assay of Buckman et aL (7, 8). This could indicate that the mesothelium, in fact, is decisive in measuring fibrinolysis. It also might indicate that the fibrinolysis theory, as proposed, is not correct, although this is not confirmed by the observation that tissue plasminogen activator activity applied after a peritoVol. 66, No.6, December 1996
neal trauma does reduce the formation of adhesions (22). Another explanation might be that, in our model, fibrin production is compensated for inadequately, even though fibrinolysis is enhanced significantly. In animal studies that are comparable to our study, application of tissue plasminogen activator activity in the peritoneal cavity prevented adhesion formation in a dose-related manner (22, 23), which seems to be in agreement with this. Alternatively, the role of blood in our model may be of importance. Raftery (9, 10) achieved hemostasis by applying pressure with a gauze swab. Buckman et al. (7, 8) did not create hemostasis, but left the defects unrepaired. In previous studies using our model, the amount of blood loss did not influence the extent or type of postsurgical adhesion formation (13, 19). Therefore, hemostasis was not considered necessary. Differences in tissue reactions between peritoneal traumas also might explain the discrepancy found in plasminogen activator activity. Orita et al. (24) showed that the total plasminogen activator activity of macrophages in the rabbit was higher at days 3 and 4 postoperatively and increased sixfold at day 10. In our study, the total plasminogen activator activity increased at day 3, reaching a maximum at day 8. These intervals are concomitant with the influx of fibroblasts and mesothelial cells, both of which are known to possess intrinsic fibrinolytic capacity. Final discussion is required concerning the different techniques for determining plasminogen activator activity. Buckman et al. (7, 8) measured the zone oflysis that occurred on the fibrin plate, beneath the mesothelium. With this technique, all fibrinolytic activity that reached the plate from the surface of the mesothelium and from deeper layers was measured. These diffusing components also may be synthesized during the assay procedure itself. For instance, plas- , minogen activator inhibitor-1 is released immedi-' ately after synthesis and is not contained within cells, in contrast to tissue plasminogen activator activity, which is strongly retained in endothelial cells. Therefore, Buckman et aL (7, 8) (synthesis) predominantly measured plasminogen activator inhibitor-1 and, thus, inhibition, whereas, in the present study (extraction technique), the fibrinolytic capacity that is present immediately after removal of the biopsy is measured. Hence, tissue plasminogen activator activity is predominant and, thus, activation. Buckman et al. (7, 8) might have measured a reduced fibrinolysis due to diffusing inhibiting activity. The plasminogen activator inhibitor should be measured using the extraction method to differentiate between this. In the animal experimental situation, no antisera are available, although one might be able to Bakkum et a1. Fibrinolysis and adhesions
1021
measure plasminogen activator inhibitor with plasminogen activator inhibitor-l messenger RNA, or otherwise, as performed by Vipond et al. (25) in humans with the fibrin plate technique. Clinical implications of our study could be that avoidance of massive fibrin production by minimalization of peritoneal trauma might be effective to prevent adhesion formation. Microsurgical techniques and gentle tissue handling then should be emphasized again. Furthermore, postsurgical adhesion formation might not be caused by a reduction in peritoneal fibrinolytic capacity but instead by an increase. REFERENCES 1. Ellis H. The cause and prevention of postoperative intraperitoneal adhesions. Surg Gynecol Obstet 1971; 133:497 -511. 2. Milligan DW, Raftery AT. Observations on the pathogenesis of peritoneal adhesions: a light and electron microscopical study. Br J Surg 1974;61:274-80. 3. Holtz G. Prevention and management of peritoneal adhesions. Fertil Steril 1984;41:497 -507. 4. Hartwell SW. The mechanics of healing in human wounds. Springfield (IL): Thomas, 1955. 5. Myhre-Jensen 0, Larsen SB, Astrup T. Fibrinolytic activity in serosal and synovial membranes. Arch Pathol 1969;88: 623-30. 6. Gervin AS, Puckett CL, Silver D. Serosal hypofibrinolysis, a cause of postoperative adhesions. Am J Surg 1973; 125:808. 7. Buckman RF, Buckman PD, Hufnagel HV, Gervin AS. A physiologic basis for the adhesion-free healing of deperitonealized surfaces. J Surg Res 1976; 21:67 -76. 8. Buckman RF, Woods M, Sargent L, Gervin AS. A unifying pathogenetic mechanism in the etiology of intraperitoneal adhesions. J Surg Res 1976;20:1-5. 9. Raftery AT. Regeneration of peritoneum: a fibrinolytic study. J Anat 1979; 129:659-64. 10. Raftery AT. Effect of peritoneal trauma on peritoneal fibrinolytic activity and intraperitoneal adhesion formation. Eur Surg Res 1981;13:397-401. 11. Whawell SA, Vipond MN, Scott-Coombes DM, Thompson IN.
1022
Bakkum et al. Fibrinolysis and adhesions
12.
13.
14.
15.
16.
17. 18. 19. 20.
21.
22.
23.
24.
25.
Plasminogen activator inhibitor 2 reduces peritoneal fibrinolytic activity in inflammation. Br J Surg 1993;80:107 -9. diZerega GS. The peritoneum and its response to surgical injury. In: diZerega GS, Malinak LR, Diamond MP, Linsky CB, editors. Treatment of postsurgical adhesions. New York: Wiley-Liss, 1990:1-11. Bakkum EA, van Blitterswijk CA, Dalmeijer RAJ, Trimbos JB. A semi-quantitative for intraperitoneal postoperative adhesion formation. Gynecol Obstet Invest 1994;37:99-105. Padro T, van den Hoogen CM, Emeis JJ. Distribution of tissue-type plasminogen activator (activity and antigen) in rat tissues. Blood Coag Fibrinolysis 1990; 1:601-8. Padro T, van den Hoogen CM, Emeis JJ. Plasmin inhibitory effect of rat tissues: comparison with anti plasmin activity of plasma. Scand J Clin Lab Invest 1991;51:599-603. Verheijen JH, Mullaart E, Chang GTG, Kluft C, Wijngaards G. A simple, sensitive spectrophotometric assay for extrinsic (tissue-type) plasminogen activator applicable to measurements in plasma. Thromb Haemost 1982;48:266-9. Ellis H. The aetiology of post-operative abdominal adhesions: an experimental study. Br J Surg 1962;50:10-6. Jackson BB. Observations on intraperitoneal adhesion. Surg 1958; 44:507 -14. Bakkum EA, Trimbos-Kemper TCM. The natural course of postsurgical adhesions. Microsurgery 1995; 16:650-4. Batzofin JH, Holmes SD, Gibbons WE, Buttram VC Jr. Peritoneal fluid plasminogen activator activity in endometriosis and pelvic adhesive disease. Fertil Steril 1985;44:277-9. Dorr PJ, Brommer EPJ, Dooijewaard G, Verner HM. Peritoneal fluid and plasma fibrinolytic activity in women with pelvic inflammatory disease. Thromb Haemost 1992;68:102-5. Doody KJ, Dunn RC, Buttram VC Jr. Recombinant tissue plasminogen activator reduces adhesion formation in a rabbit uterine horn model. Fertil Steril 1989;51:509-12. Menzies D, Ellis H. Intra-abdominal adhesions and their prevention by topical tissue plasminogen activator. J R Soc Med 1989;82:534-5. Orita H, Campeau JD, Gale JA, Nakamura RM, DiZerega GS. Differential secretion of plasminogen activator activity by postsurgical activated macrophages. J Surg Res 1986; 41:569-73. Vipond MN, Whawell SA, Thompson IN, Dudley HAF. Peritoneal fibrinolytic capacity and intra-abdominal adhesions. Lancet 1990;335:1120-2.
Fertility and Sterility®
Vol. 66, No.6, December 1996 Printed on acid-free paper in U. S. A.
1996 American Society for Reproductive Medicine
long-term analysis of peritoneal plasminogen activator activity and adhesion formation after surgical trauma in the rat model
Erica A. Bakkum, M.D., Ph.D. *t+ Jef J. Emeis, Ph.D.§ Remco A. J. Dalmeijer, B.Sc.*
Clemens A. van Blitterswijk, Ph.D. * J. Baptist Trimbos, M.D., Ph.D.t Trudy C. M. Trimbos-Kemper, M.D., Ph.D.t
Leiden University Medical Center, Leiden, The Netherlands
Objective: Recent literature has shown that a common pathway in postsurgical adhesion formation is a transient reduction in local plasminogen activator activity, shortly after peritoneal trauma. This deficit in fibrinolysis permits deposited fibrin to become organized into fibrous, permanent adhesions. Although adhesion formation is a process that continues beyond the first postoperative days, long-term analysis of this theory has not been performed. Design: A standardized peritoneal adhesion model in the rat. Main Outcome Measure(s): Long-term analysis of the peritoneal fibrinolytic activity (extraction technique) was related to the extent of postsurgical adhesion formation, up to 1 year postoperatively. Result(s): Total and tissue plasminogen activator activity were significantly increased at days 3 and 8, and 1 month postoperatively. A mean adhesion percentage of 75% per peritoneal defect was found to persist throughout all evaluation times, which was directly related to the increase of fibrinolysis. Conclusion(s): In contrast to the classical concept that adhesion formation is related to a reduction in fibrinolysis, an impressive increase of the fibrinolysis was found to be associated with the persistence of adhesions. Fertil Steril® 1996;66:1018-22 Key Words: Postsurgical adhesions, fibrinolytic activity, peritoneal side wall model, rat
Injury to the peritoneum leads to the formation of a fibrinous exudate. This fibrinous exudate is part of the hemostatic process and aids in tissue repair by providing a matrix for invading fibroblasts and new vessels. Fibrin deposition is an essential component of normal tissue repair while resolution of this fibrin deposit is required to restore the preoperative conditions. An imbalance between these two processes is thought to lead to adhesion formation when the deposited fibrin is not removed effectively and fibrinous adhesions become organized into fibrous, permanent adhesions (1-3). Normal peritoneum has
Received October 16, 1996; revised and accepted July 23, 1996. * Biomaterials Research Group, Laboratory for Biocompatibility. t Biomaterials Research Group, Department of Gynecology. :j: Reprint requests: Erica A. Bakkum, M.D., Ph.D., Department of Gynecology, Groene Hart Hospital, GraafFlorisweg 77-79, 2805 HH Gouda, The Netherlands (FAX: 31-182-566449). § Gaubius Laboratory TNO-PG, Dutch organization for applied scientific research: prevention and health. 1018
Bakkum et al. Fibrinolysis and adhesions
an inherent fibrinolytic activity, as first described by Hartwell (4) and confirmed by other authors (5, 6). Damaged peritoneum possesses a reduced fibrinolytic capacity. As a result, plasminogen is not converted sufficiently into plasmin and lysis of fibrin will not take place; this process is described as being related to postsurgical adhesion formation (7 -11). The time course and return of this fibrinolytic capacity after peritoneal trauma was studied first by Raftery (9), who found a reduction in fibrinolytic capacity immediately after the initial trauma. In that study, fibrinolysis was measured in the layer of mesothelial cells located at the surface of the peritoneum, which means that fibrinolysis was detected only at the surface ofthe wound area. Therefore, this assay method measures only part of the fibrinolytic capacity, which is known to reside not only in the mesothelium but also in the submesothelial blood vessels (3, 6). In addition, it has been reported that regeneration of the mesothelium starts at the base of the wound (12). Evaluation of the fibrinolytic parameters therefore should be performed in the total Fertility and Sterility®
peritoneal layer. This was performed in a later study by Buckman et al. (8), who found a reduction in fibrinolysis with full-thickness peritoneal biopsies, although they only measured fibrinolysis during the first 96 hours postoperatively. The present study was undertaken to determine whether a standard peritoneal sidewall trauma, known to produce uniform and consistent adhesion formation (13), was associated with a depression of fibrinolysis in the peritoneal layer, especially at longer postoperative intervals. Full-thickness biopsies were taken to ensure measurement of the fibrinolytic capacity of the entire depth of the peritoneal layer. Fibrinolytic capacity was evaluated at different postoperative time intervals, from directly postoperative to :=;1 year. Total plasminogen activator activity, tissue plasminogen activator activity, and urokinase plasminogen activator activity were chosen as parameters to detect fibrinolytic capacity. In the presence of fibrin, tissue plasminogen activator activity and plasminogen will bind to fibrin. Subsequently, plasminogen is converted into plasmin, which can disintegrate fibrin. Urokinase plasminogen activator activity has a similar working mechanism but does not require fibrin as a cofactor and is probably more important in extracellular proteolytic processes. Tissue plasminogen activator activity has a significant role in the modulation of intravascular thrombi. The observed changes in the fibrinolytic capacity were related to the extent of postsurgical adhesion formation. MATERIALS AND METHODS
Twenty-four female Wistar rats of reproductive age (Harlan CPB, Zeist, The Netherlands), weighing 180 to 200 g, were anesthetized with ether. A standard adhesion model was used, as described previously, which comprised bilateral excision and suturing of an oval-shaped parietal peritoneal defect. The defect measured 1.5 x 1.2 cm, with a depth of 3 mm, and encompassed the mesothelial and both underlying muscular layers (13). After 1, 3, and 8 days and 4, 12, 24, and 52 weeks, the rats were killed. By means ofthe three Vicryl sutures (Johnson and Johnson, Amersfoort, The Netherlands) with which the defect was closed, the defect was subdivided into eight areas of 12.5%. Adhesions were scored according to extent, as a percentage of the peritoneal defect covered with adhesions (13). At each interval, three rats were killed, thereby allowing six peritoneal defects per survival time to be evaluated. Three rats received an abdominal incision without additional peritoneal trauma and thus served as controls (n = 6). After scoring of adhesions, full-thickness biopsies Vol. 66, No.6, December 1996
from each defect (6 mm in diameter), including the mesothelial and the underlying muscular layers, were dissected with a dermal punch instrument, at a standard localization between the first and second suture. In the control animals without peritoneal defects, biopsies were taken at the same locations and are referred to as normal peritoneal fibrinolytic activity. After excision, the biopsies were frozen immediately and stored at -70°C. Subsequently, they were triturated under liquid nitrogen, resuspended in Camiolo's buffer (40 mg of wet tissue weight/mL) and extracted according to Padro et al. (14, 15). In the extracts, three parameters were determined spectrophotometrically, as described by Verheijen et al. (16); these parameters were the total plasminogen activator activity, tissue plasminogen activator activity, and urokinase plasminogen activator activity. Tissue plasminogen activator activity is the plasminogen activator activity quenched by anti-rat tissue plasminogen activator activity immunoglobulin (Ig)G as described by Padro et al. (15). Urokinase plasminogen activator activity is the activity quenched by anti-mouse urokinase plasminogen activator activity IgG (American Diagnostics, Greenwich, CT), which cross-reacts with rat urokinase plasminogen activator activity. All total plasminogen activator activities were corrected for the plasmin-inhibitory activity present in the extract (15) and will be expressed as units (U) of activity of 1 ng recombinant human tissue plasminogen activator activity) per g wet tissue weight. Total plasminogen activator activity represents tissue plasminogen activator activity and urokinase plasminogen activator activity together. In each individual peritoneal defect, the total plasminogen activator activity and tissue plasminogen activator activity subsequently were related to the extent of postsurgical adhesion formation. Statistical analysis was performed with the oneway analysis of variance and the Newman-Keuls multiple-range-test. Statistical significance was defined as P < 0.01. Data were expressed per peritoneal defect as mean adhesion percentage, mean total plasminogen activator activity, mean tissue plasminogen activator activity, and mean urokinase plasminogen activator activity (±SD). RESULTS Fibrinolytic Parameters
Normal peritoneal plasminogen activator activity as measured in the control animals, averaged 11.8 ± 4.5 U/g. The effect of peritoneal trauma on the plasminogen activator activity is depicted in Figure 1. At days 1 and 3, an increase was found (33.3 ± 9.2 Bakkum et a1. Fibrinolysis and adhesions
1019
250 200
~ .s
150
~
100
«
---f --- -
- ---
--- {
A mean adhesion percentage of 89.6% ± 5.1% was found at day 1 postoperatively, which increased slightly to 93.8% ± 6.8% on day 3. After day 3, a minimal decline was seen (Figs. 1 and 2). At 1 year, a mean adhesion percentage of 75% ± 13.7% was found per peritoneal defect. No statistical significant difference was found between any of the postoperative intervals.
50
Od
DISCUSSION
0"---...1----'-----'---"------'-----'--1d 3d ad 1m 3m 8m 1y Od time after surgery
Figure 1 Mean total plasminogen activator activity (U/g) and adhesion percentage (%) at different postoperative intervals after a standard peritoneal trauma. (- -) Mean total plasminogen activator activity; (- - -) mean adhesion percentage. P AA, total plasminogen activator activity. **p < 0.01 compared with control peritoneum (sham-operated animals).
and 99.9 ± 31.8 U/g; P < 0.01), with a further increase to 173.2 ± 69.7 U/g (P < 0.01) at day 8 postoperatively. From this interval onward, the activity decreased until the control values were reached at 3 months postoperatively (Fig. 1). The tissue plasminogen activator activity followed a similar time course (Fig. 2). Normal control peritoneal tissue plasminogen activator activity was 6.2 ± 3 U/g. Peritoneal trauma caused the tissue plasminogen activator activity to increase, with the highest level being 165.8 ± 69.4 U/g found at day 8 (P < 0.01). After day 8, a decline was observed, similar to the total plasminogen activator activity (Fig. 2). The urokinase plasminogen activator activity showed a different course. Normal peritoneal urokinase plasminogen activator activity was 5.8 ± 2.4 U/g. This activity fell to 3.6 ± 3.7 U/g on day 1 and increased to 7.9 ± 6.7 U/g on day 3, after which a decline was seen. Because urokinase plasminogen activator activity represents only an insignificant amount of the total plasminogen activator activity «5%), the urokinase plasminogen activator activity was omitted from further analysis. Adhesion Formation
For all survival times the mean adhesion percentage per peritoneal defect showed little variation, although the type changed markedly from filmy, easily separable to organized and vascular within 1 month postoperatively. Adhesion formation was confined to the peritoneal defects and no other noticeable complications were encountered. Adhesion percentages ranged from 25% to 100%, with a mean of 75%. 1020
Bakkum et at Fibrinolysis and adhesions
Failure of the peritoneal fibrinolytic system in case of postoperative ischemia is described to be an important factor in the formation of postsurgical adhesions (7-10, 17). When peritoneum is damaged, the plasminogen activator activity is reduced (6, 8, 10), through which deposited fibrin is not removed. This fibrin is organized subsequently into vascular, permanent adhesions (2, 18). Fibrinolytic activity in rats has been measured using either a single layer of cells (Raftery [9, 10], fibrin-slide technique) or by applying full-thickness biopsies of the peritoneal layer (Buckman et al. [7, 8], fibrin-plate technique). With both techniques, the fibrinolytic activity is measured as the resultant zone of lysis. Raftery (9, 10) observed that the fibrinolytic activity disappeared from the area of a peritoneal defect after 2 days of inflicting various traumas. At day 3, an irregular band was observed, indicating a partial recovery, and, at day 8, a wide band was detected that was even higher than normal peritoneum, indicating a full recovery. However, their study is limited by the use of only a single layer of cells, taken from the surface of the peritoneal wound area. In addition, inflammatory cells present at the surface, which do not have intrinsic fibrino-
250
200
!! S
150
'"
«
Il.
.:.
100
50
3d
ad
3m
1m
6m
ly
time alter surgery
Figure 2 Mean tissue plasminogen activator activity (U/g) and mean adhesion percentage (%) at different postoperative time intervals after a standard peritoneal trauma. (- -) Mean tissue plasminogen activator activity; (- - -) mean adhesion percentage. t-PA, tissue plasminogen activator activity. **p < 0.01 compared with control peritoneum (sham-operated animals).
Fertility and Sterility®
r
lytic capacity, might have affected the results. Buckman et al. (7, 8) observed a similar reduction in fibrinolysis. Abrasion trauma produced an immediate, significant decrease of total plasminogen activator activity, which was sustained at 48 and 72 hours, with some recovery by 96 hours. Crushing induced an immediate depression, with a maximum at 24 hours, and a return to normal by 72 hours. Deperitonealized surfaces did not show such depression and were found to be associated with a reduction in adhesions. Buckman et aL (7, 8) did not follow the parameters beyond 96 hours, although in both groups the fibrinolysis did not seem to have recovered completely. In our study, no depression of the fibrinolytic activity was found. Instead, an increase was found, both in total plasminogen and in tissue plasminogen activator activity. This increase differed significantly from control peritoneum from day 3 to 1 month postoperatively (Figs. 1 and 2). Mter 1 month, a decline was observed with a return to normal at 6 months. When the extent of adhesion formation was related to the total plasminogen and tissue plasminogen activator activity, no correlation was discerned (Pearson's correlation test). The slight, insignificant decrease in adhesion percentage at days 3, 8, and 1 month, might be concomitant with the increase in fibrinolysis after day 1. However, in earlier studies, this decrease was attributed to scar formation and retraction when the adhesions become progressively more organized postoperatively (19). Recently, other authors also have suggested that intraperitoneal (fluid phase) fibrinolysis is unaltered or even enhanced with infectious processes or endometriosis (20, 21). In peritoneal fluid of patients with or without endometriosis, or with or without adhesive disease, Batzofin et aL (20) found no difference. In a study on patients with pelvic inflammatory disease, Dorr (21) states that fibrinolysis occurs at a high rate, as measured in the peritoneal fluid and plasma. We were not able to measure fibrinolysis in the peritoneal fluid because of the minimal amount that can be harvested in the rat. The controversial finding that high levels of fibrinolysis were found up to 1 month postoperatively, concomitant with the persistence of adhesions, needs further evaluation. Our assay method might not measure the relevant plasminogen activator activity, in contrast to the assay of Buckman et aL (7, 8). This could indicate that the mesothelium, in fact, is decisive in measuring fibrinolysis. It also might indicate that the fibrinolysis theory, as proposed, is not correct, although this is not confirmed by the observation that tissue plasminogen activator activity applied after a peritoVol. 66, No.6, December 1996
neal trauma does reduce the formation of adhesions (22). Another explanation might be that, in our model, fibrin production is compensated for inadequately, even though fibrinolysis is enhanced significantly. In animal studies that are comparable to our study, application of tissue plasminogen activator activity in the peritoneal cavity prevented adhesion formation in a dose-related manner (22, 23), which seems to be in agreement with this. Alternatively, the role of blood in our model may be of importance. Raftery (9, 10) achieved hemostasis by applying pressure with a gauze swab. Buckman et al. (7, 8) did not create hemostasis, but left the defects unrepaired. In previous studies using our model, the amount of blood loss did not influence the extent or type of postsurgical adhesion formation (13, 19). Therefore, hemostasis was not considered necessary. Differences in tissue reactions between peritoneal traumas also might explain the discrepancy found in plasminogen activator activity. Orita et al. (24) showed that the total plasminogen activator activity of macrophages in the rabbit was higher at days 3 and 4 postoperatively and increased sixfold at day 10. In our study, the total plasminogen activator activity increased at day 3, reaching a maximum at day 8. These intervals are concomitant with the influx of fibroblasts and mesothelial cells, both of which are known to possess intrinsic fibrinolytic capacity. Final discussion is required concerning the different techniques for determining plasminogen activator activity. Buckman et al. (7, 8) measured the zone oflysis that occurred on the fibrin plate, beneath the mesothelium. With this technique, all fibrinolytic activity that reached the plate from the surface of the mesothelium and from deeper layers was measured. These diffusing components also may be synthesized during the assay procedure itself. For instance, plas- , minogen activator inhibitor-1 is released immedi-' ately after synthesis and is not contained within cells, in contrast to tissue plasminogen activator activity, which is strongly retained in endothelial cells. Therefore, Buckman et aL (7, 8) (synthesis) predominantly measured plasminogen activator inhibitor-1 and, thus, inhibition, whereas, in the present study (extraction technique), the fibrinolytic capacity that is present immediately after removal of the biopsy is measured. Hence, tissue plasminogen activator activity is predominant and, thus, activation. Buckman et al. (7, 8) might have measured a reduced fibrinolysis due to diffusing inhibiting activity. The plasminogen activator inhibitor should be measured using the extraction method to differentiate between this. In the animal experimental situation, no antisera are available, although one might be able to Bakkum et a1. Fibrinolysis and adhesions
1021
measure plasminogen activator inhibitor with plasminogen activator inhibitor-l messenger RNA, or otherwise, as performed by Vipond et al. (25) in humans with the fibrin plate technique. Clinical implications of our study could be that avoidance of massive fibrin production by minimalization of peritoneal trauma might be effective to prevent adhesion formation. Microsurgical techniques and gentle tissue handling then should be emphasized again. Furthermore, postsurgical adhesion formation might not be caused by a reduction in peritoneal fibrinolytic capacity but instead by an increase. REFERENCES 1. Ellis H. The cause and prevention of postoperative intraperitoneal adhesions. Surg Gynecol Obstet 1971; 133:497 -511. 2. Milligan DW, Raftery AT. Observations on the pathogenesis of peritoneal adhesions: a light and electron microscopical study. Br J Surg 1974;61:274-80. 3. Holtz G. Prevention and management of peritoneal adhesions. Fertil Steril 1984;41:497 -507. 4. Hartwell SW. The mechanics of healing in human wounds. Springfield (IL): Thomas, 1955. 5. Myhre-Jensen 0, Larsen SB, Astrup T. Fibrinolytic activity in serosal and synovial membranes. Arch Pathol 1969;88: 623-30. 6. Gervin AS, Puckett CL, Silver D. Serosal hypofibrinolysis, a cause of postoperative adhesions. Am J Surg 1973; 125:808. 7. Buckman RF, Buckman PD, Hufnagel HV, Gervin AS. A physiologic basis for the adhesion-free healing of deperitonealized surfaces. J Surg Res 1976; 21:67 -76. 8. Buckman RF, Woods M, Sargent L, Gervin AS. A unifying pathogenetic mechanism in the etiology of intraperitoneal adhesions. J Surg Res 1976;20:1-5. 9. Raftery AT. Regeneration of peritoneum: a fibrinolytic study. J Anat 1979; 129:659-64. 10. Raftery AT. Effect of peritoneal trauma on peritoneal fibrinolytic activity and intraperitoneal adhesion formation. Eur Surg Res 1981;13:397-401. 11. Whawell SA, Vipond MN, Scott-Coombes DM, Thompson IN.
1022
Bakkum et al. Fibrinolysis and adhesions
12.
13.
14.
15.
16.
17. 18. 19. 20.
21.
22.
23.
24.
25.
Plasminogen activator inhibitor 2 reduces peritoneal fibrinolytic activity in inflammation. Br J Surg 1993;80:107 -9. diZerega GS. The peritoneum and its response to surgical injury. In: diZerega GS, Malinak LR, Diamond MP, Linsky CB, editors. Treatment of postsurgical adhesions. New York: Wiley-Liss, 1990:1-11. Bakkum EA, van Blitterswijk CA, Dalmeijer RAJ, Trimbos JB. A semi-quantitative for intraperitoneal postoperative adhesion formation. Gynecol Obstet Invest 1994;37:99-105. Padro T, van den Hoogen CM, Emeis JJ. Distribution of tissue-type plasminogen activator (activity and antigen) in rat tissues. Blood Coag Fibrinolysis 1990; 1:601-8. Padro T, van den Hoogen CM, Emeis JJ. Plasmin inhibitory effect of rat tissues: comparison with anti plasmin activity of plasma. Scand J Clin Lab Invest 1991;51:599-603. Verheijen JH, Mullaart E, Chang GTG, Kluft C, Wijngaards G. A simple, sensitive spectrophotometric assay for extrinsic (tissue-type) plasminogen activator applicable to measurements in plasma. Thromb Haemost 1982;48:266-9. Ellis H. The aetiology of post-operative abdominal adhesions: an experimental study. Br J Surg 1962;50:10-6. Jackson BB. Observations on intraperitoneal adhesion. Surg 1958; 44:507 -14. Bakkum EA, Trimbos-Kemper TCM. The natural course of postsurgical adhesions. Microsurgery 1995; 16:650-4. Batzofin JH, Holmes SD, Gibbons WE, Buttram VC Jr. Peritoneal fluid plasminogen activator activity in endometriosis and pelvic adhesive disease. Fertil Steril 1985;44:277-9. Dorr PJ, Brommer EPJ, Dooijewaard G, Verner HM. Peritoneal fluid and plasma fibrinolytic activity in women with pelvic inflammatory disease. Thromb Haemost 1992;68:102-5. Doody KJ, Dunn RC, Buttram VC Jr. Recombinant tissue plasminogen activator reduces adhesion formation in a rabbit uterine horn model. Fertil Steril 1989;51:509-12. Menzies D, Ellis H. Intra-abdominal adhesions and their prevention by topical tissue plasminogen activator. J R Soc Med 1989;82:534-5. Orita H, Campeau JD, Gale JA, Nakamura RM, DiZerega GS. Differential secretion of plasminogen activator activity by postsurgical activated macrophages. J Surg Res 1986; 41:569-73. Vipond MN, Whawell SA, Thompson IN, Dudley HAF. Peritoneal fibrinolytic capacity and intra-abdominal adhesions. Lancet 1990;335:1120-2.
Fertility and Sterility®