1 Department of Gynecology and Obstetrics, 2 Department of General Surgery and 3 Department of Physiology, School of Medicine, Kocaeli University, Kocaeli 41900, 4 Department of Histology and Embryology, School of Medicine, Istanbul University, Ciapa, Istanbul 34891, 5 Department of Internal Medicine, School of Medicine, Cumhuriyet University, Sivas 58040, 6 Department of Biochemistry, School of Medicine, Dokuz Eylül University, Izmir 35048 and 7 Department of Pathology, School of Medicine, Hacettepe University, Ankara 06912, Turkey
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: cyclophosphamide/granulocytemacrophage colonystimulating factor/neutrophils/peritoneum/postoperative complications
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The mechanism of adhesion formation represents a variation in the physiological healing process. Adhesions seem to be a result of an inflammatory mechanism (Pados and Devroey, 1992). There are several indications that polymorphonuclear leukocytes (PMN) are involved in the pathophysiology of intraperitoneal adhesion formation. Peritoneal injury during surgery, in the absence of infection, causes a rapid and transient influx of PMN into the peritoneal cavity followed by the accumulation of mononuclear cells, mainly macrophages (Ellis et al., 1965
). However, the role of neutrophils in the pathophysiological cascade leading to intraperitoneal adhesion formation has not been well studied.
The present study evaluated the role of neutrophils in postoperative adhesion formation. Cyclophosphamide and granulocytemacrophage colony-stimulating factor (GM-CSF) were preferred because of their contrary effects on leukocytes and a number of activities such as chemotaxis, migration, cell adhesion, synthesis and release of cytokines, and superoxide release. We tested the hypothesis to ascertain whether or not the development of neutrophil function with GM-CSF would affect postoperative adhesion formation in the rat. Since adhesion formation is a kind of fibrosis, hydroxyproline concentrations as a marker of collagen metabolism have been measured in adhesive tissue.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Experiments
The animals were randomly divided into three groups, group I (n = 12) receiving saline (saline solution every 12 h for 2 days prior to the experiment, injections were continued for 5 days after induction of adhesion formation), group II (n = 14) receiving a single dose of cyclophosphamide which caused neutropenia (Dokmetas et al., 1991) (Sigma Chemical Co., St Louis, MO, USA) (250 mg/kg i.p. 2 days prior to the experiment), and group III (n = 14) receiving human GM-CSF (Sandoz Chem Co., Basel, Switzerland) (1 mg in phosphate-buffered saline solution every 12 h for 2 days prior to the experiment, injections were continued for 5 days after induction of adhesion formation). The neutrophil count and neutrophil phagocytosis and adherence index were determined in each group 1 day after the induction of adhesion.
Midline laparotomy was performed under ketamineHCl (60 mg/kg) and xyalasin (4 mg/kg) anaesthesia using a clean surgical technique. The serosa was stripped from the uterine horn at its midportion for a distance of 1.5 cm and the caecum was grasped and denuded of serosa over a 2 cm length until punctuate haemorrhage occurred. The abdomen was closed at the end of the surgery in two layers with a continuous 2/0 silk (Ethicon) suture (Liebman et al., 1993; Heidrick et al., 1994
).
All animals in each group underwent a peritoneal lavage 24 h after adhesion induction. Peritoneal lavage was performed using a silastic tube introduced into the peritoneal cavity through a small midline incision. 200 ml of warm sterile saline containing heparin was then infused into the peritoneum and gentle external manipulation of the abdomen was applied in order to ensure proper distribution. Neutrophil counts were determined from the aspirated fluid.
Adhesion assessment
A grading system (Linsky et al., 1987) was used to determine the adhesion score at the end of the second week after adhesion induction; this was practical, easy and detailed: 0, no adhesion; 1, adhesion on 25% of the traumatized area; 2, adhesion on 50% of the traumatized area; and 3, total area involved. The severity of the adhesions was evaluated according to the scale: 0.0, no resistance to separation; 0.5, moderate force required for separation; 1.0, sharp dissection needed for separation.
A total score was obtained by adding the two grading scales with a range from 0.0 to 4.0 assigned for each region evaluated. The total score represented both extent and severity.
Histological assesment
Preparation of tissues for light microscopy
The tissue from adhesions of each animal was fixed with formaldehyde for 24 h, dehydrated and embedded in paraffin wax. The thickness of the sections was 5 µm and they were stained with haematoxylin and eosin.
Preparation of tissues for electron microscopy
The abdominal cavity was opened and five tissue samples from adhesions of each group were placed in 2.5% glutaraldehyde in 0.1 M cacodylate buffer at 4°C. Fixation was continued for 24 h. Primary fixation was carried out using 1% osmium tetroxide in 0.1 M phosphate buffer. Tissue fixed in glutaraldehyde was washed in several changes of buffered sucrose and then post-fixed for 2 h in 1% osmium tetroxide in 0.1 M phosphate buffer. The specimens were subsequently dehydrated in graded solutions of ethanol, passed through epoxypropane and embedded in Epon 812. Sections 0.1 µm thick were cut on a Reichert microtome, and stained with uranyl acetate and lead citrate prior to examination through a Jeol 100-C electron microscope.
Studies on neutrophil function
White cell count, neutrophil count
Peripheral blood for blood smears and quantitation of the absolute numbers of circulating leukocytes were obtained from tail bleeding. The total circulating white blood cell count/mm3 was determined using an automatic cell counter (Cell Dyne 300). The percentage and absolute number of circulating white blood cell subsets was determined after performing a differential count of 100 cells/smear.
Neutrophil adhesiveness
Siliconized glass beads 0.1 mm in diameter were packed to a height of 3 cm over a small piece of glass wool in a siliconized glass syringe 1 cm in diameter, the whole being enclosed in a water jacket at 37°C. Two millilitres of heparinized whole blood were added to the column, and, after 10 min, allowed to flow out from the column. The flow rate was adjusted by controlling the pressure on the plunger of the syringe to 0.1 ml/min. An absolute neutrophil count (total white cell count/mm3x% neutrophil) was performed in both influent and effluent blood and the ratio of the former to the latter termed as the adhesiveness index (AI) (Brandt, 1965).
Neutrophil phagocytosis
Heat-killed yeast cells were prepared by stirring desiccated baker's yeast in normal saline, placing the suspension in a water bath at 100°C for 30 min and filtering the suspension through layers of gauze. The suspension of yeast cells was adjusted to a particle concentration of 5x104/mm3 with phosphate-buffered normal saline (pH 7.4). Vigorous shaking was important to disperse small cell clumps.
The leukocyte suspension was prepared from a buffy coat, obtained by simple sedimentation (without addition of dextran) of heparinized whole blood (50 units/ml), and the concentration of mature neutrophils was adjusted to 5000/mm3 with cell-free plasma. If the absolute neutrophil count of the buffy coat was <5000/mm3, the concentration of yeast cells was reduced so that a constant yeast-cell to neutrophil ratio of 10:1 was maintained. The yeast cell and leukocyte suspensions was mixed together in amount of 0.4 ml, and placed in a water bath at 37°C with gentle inversion of the tube every 5 min. After 30 min, 0.08 ml of 5% dipotassium ethylenediamine tetra-acetate was added, the contents of the tube were well mixed, and films prepared rapidly on slides from the mixture, and stained with MayGrunwaldGiemsa. One hundred neutrophils were counted and classified according to their yeast-cell content; those containing more than six yeast cells were placed in one group because of the frequent difficulty of containing the exact number when more than six were packed in one neutrophil. The mean yeast-cell number per neutrophil was termed the phagocytosis index (PI) (Penny et al., 1966).
Biochemical analysis
Adhesive tissue samples were removed surgically and immediately frozen. These tissue samples were homogenized. Purified acid-soluble collagen (dissolved in 50 mM acetate buffer, pH 3.5) was included as a test sample for the estimation of hydroxyproline. Aliquots of standard hydroxyproline prepared from stock solution and test samples were mixed gently with sodium hydroxide in a total volume of 50 µl. Samples were hydrolysed. 450 µl of chloramine-T was added to the hydrolysate, mixed gently, and oxidation was allowed to proceed for 25 min at room temperature. 500 µl of Erlich's aldehyde reagent was added to each sample, mixed gently, and the chromophore was developed by incubating the samples at 65°C for 20 min. The absorbency of each sample was read at 550 nm using a spectrophotometer. The collagen content of adhesive tissue was calculated (Reddy and Enwemeka, 1996).
Statistical analysis
The statistical analysis of adhesion scores and blood and peritoneal neutrophil counts among the groups was carried out using non-parametric KruskalWallis variant analysis and the Tukey test. Our data were normally distributed. P < 0.05 was considered to be statistically significant.
This study was approved by the ethical committee of Kocaeli University and performed following standard guidelines for the care and use of laboratory animals.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Blood cell counts, neutrophil functional parameters
The peripheral white blood cell counts were determined in each group and were found to be decreased in group II and increased in group III as expected. The same results were also recorded for the neutrophils (P < 0.05). Thus, neutropenic and neutrophilic effects of cyclophosphamide and GM-CSF respectively, were observed (Table I). Additionally the total numbers of white blood cells recovered from the peritoneum significantly decreased in group II and increased in group III. These neutropenic and neutrophilic effects of the agents were also reflected in the peritoneal fluid and the number of the cells in group III was significantly higher than in both group I and group II (P < 0.05) (Table I
).
|
|
|
|
Histological results
Light microscopic findings
Group I:
Collagen formation was pronounced and fibroblasts were prominent. Small blood vessels were also present within the adhesive tissue. Occasional foreign body granulomas and PMN were seen (Figure 3a).
|
Group III:
Several macrophages and PMN were seen within the adhesive tissue. Inflammatory granulation tissue with fibroblasts and new blood vessels were observed.
Electron microscopical findings
Micrographs were usually taken from the middle of the adhesive area. The results obtained using the electron microscope largely supported those of the light microscopical study but also revealed some interesting features not shown by light microscopy.
Group I:
Collagen formation was well advanced and fibroblasts were rarely beginning to align themselves parallel with one another (Figure 4).
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Although many multifactorial processes are involved in the adhesion formation, one of these has been shown to be the inflammatory process (Pados and Devroey, 1992). Surgical injury of the intact peritoneum produces a serosangineous exudate within 3 h (Pados and Devroey, 1992
). Following an immediate inflammatory reaction, vascular permeability is increased and a fibrin-rich exudate is released (Ellis et al., 1965
; Liebman et al., 1993
). If the fibrin is not completely lysed through the plasminogenplasmin cascade, adhesions may form through collagen deposition on a framework of proteoglycan and fibronectin. This exudate includes various inflammatory cells such as neutrophils, and mononuclear cells (Rodgers and DiZerega, 1992
). However, under normal conditions the peritoneal cavity contains mainly macrophages and lymphocytes in a small amount of fluid (Pados and Devroey, 1992
). During the early period of injury (up to 3 days), the most pronounced elements in the healing of peritoneal lesions are PMN and fibrin (Raftery, 1973
; Rodgers and DiZerega, 1992
). The PMN cells, recruited in large numbers into the peritoneal cavity after the injury, play a role in the formation of peritoneal adhesions.
It has also been shown that enhancement of macrophages markedly decreases postoperative adhesion formation (Rodgers and DiZerega, 1992). Additionally, it has been reported that inhibition of PMN adherence results in increased peritoneal adhesion formation (Arfors et al., 1987
; ArRajab et al., 1996
). The opposite mechanisms were also suggested. It was shown that neutrophils migrate into the peritoneum 1 day after adhesion-producing intraperitoneal injuries. PMN first adhere to the vascular endothelial surface and highly reactive metabolites derived from both the endothelium and PMN as a result of the adherence, produce endothelial injury and, consequently, microvascular permeability, oedema and thrombosis occur (Harlan, 1987a
,b
; Harlan et al., 1987
; Price et al., 1987
; Barton et al., 1989
; Argenbright et al., 1991
; ArRajab et al., 1996
). Neutrophilneutrophil aggregation further compromises the microvascular circulation by occluding capillaries and postcapillary venules, extending the zone of ischaemia and subsequent necrosis (ArRajab et al., 1996
). Additionally, most animal studies have shown that corticosteroids and non-steroidal anti-inflammatory drugs, which prevent leukocyte migration besides their other effects, have reduced adhesion formation (Replogle et al., 1966
; Siegler et al., 1980
). In the present study cyclophosphamide-induced neutropenia resulted in decreased peritoneal adhesion formation. The decreased number of neutrophils was accompanied by a decreased phagocytosis index but an unchanged adherence index. Although GM-CSF-induced neutrophilia was accompanied by an increased phagocytosis index, the adherence of PMN did not change in this group. Adhesion formation was increased in this group but the difference between this and the control group was statistically insignificant. So we conclude that changes in phagocytosis index could be more important in the formation of adhesion than changes in the adherence index.
As shown previously, adhesion formation can form through collagen deposition (Thompson et al., 1989). Collagen principally forms the matrix of the developed adhesion. Neutrophils were claimed to have the potential to degrade the extracellular matrix (Sopata et al., 1989
). Serine protease and metalloproteinases released from the neutrophils have been shown to play a role in neutrophil-mediated degradation of basement membrane collagen (Vissers and Winterbourn, 1988
). In some studies, the release of free oxygen radicals and neutral protease from neutrophils has been claimed to cause collagen degradation (Hogstrom and Haglund, 1986
; Jonsson and Hogstrom, 1991
). Although collagen synthesis is known to reach significant amounts on the fifth day, controversial results have been reported (Baykal et al., 1997
). Collagen solubilization by neutrophils was measured by hydroxyproline release (Miligan and Raftery, 1974
; Palmgren et al., 1992; Baykal et al., 1997
). In this study we also used a similar method to determine the amount of connective tissue present on adhesive tissue as a chemical marker. Interstitial connective tissue fragments were known to be chemotactic for neutrophils. Native and synthetic collagen peptides were shown to stimulate free oxygen radical and elastase release as synthetic peptides containing hydroxyproline from neutrophils (Laskin et al., 1994
). Although the collagen concentrations did not show a significant difference between the groups in this study, the degradation product, hydroxyproline, was significantly increased in the neutrophilia-induced group. But the adhesion formation was not found to have decreased in this group. This may be explained by extensively increased numbers of neutrophils in the peritoneum, since we believe that neutrophils cause adhesion formation but, at the same time, they start the degradation of collagen through the substances released by themselves and through other undetermined factors. Thus in the GM-CSF-induced neutrophilia group this balance can be changed in favour of adhesion formation. Increased phagocytosis, which was present in accordance with the increased hydroxyproline concentrations, may not be sufficient to prevent the adhesion formation. In the neutropenia induced group, hydroxyproline concentrations were not different since phagocytosis was very low. At the same time adhesion formation was significantly low which might be explained by a decreased number of neutrophils. Light and electron microscopy findings also supported this suggestion. It has been reported that inhibition of neutrophil adherence and consequently neutrophil activation and migration into the peritoneum reduce neutrophil-dependent collagen degradation and result in increased adhesion formation (ArRajab et al., 1996
). On the contrary, our results have shown that in the neutrophilia-induced group adhesion formation has been increased. Moreover, as stated in the literature (Bigatti et al., 1995
), vascularization played a role in the organization of adhesions during the early stages of this study.
Finally, it is concluded that inhibition and activation of neutrophil counts and function affect the cascade of postoperative adhesion formation. These results might suggest that neutrophils have a role in modulating postoperative adhesion formation. Although we showed the beneficial effects of cyclophosphamide on adhesion formation, by significant reduction of neutrophil counts and neutrophil phagocytosis index and insignificant inhibition of neutrophil adhesivity index, the bioacceptance and usefulness of these factors in the formation of postoperative abdominal and pelvic adhesion require further investigation prior to possible clinical use.
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Argenbright, L..W., Letts, L.G. and Rothlein, R. (1991) Monoclonal antibodies to the leukocyte membrane CD18 glycoprotein complex and to intracellular adhesion molecule-1 inhibit leukocyte-endothelial adhesion in rabbits. J. Leukocyte Biol., 49, 25357.[Abstract]
ArRajab, A., Mileski, W., Sentementes, J.T. et al. (1996) The role of neutrophils in peritoneal adhesion formation. J. Surg. Res., 61, 14346.[ISI][Medline]
Barton, R.W., Rothlein, R., KsÉazek, R. and Kennedy, C. (1989) The effect of anti-intracellular adhesion molecule-1 on phorbol-ester-induced rabbit lung inflammation. J. Immunol., 141, 127882..
Baykal, A., Onat, D., Rasa, K. et al. (1997) Effects of polyglycolic acid and polypropylene meshes on postoperative adhesion formation in mice. World J. Surg., 21, 579583.[ISI][Medline]
Bigatti, G., Boeckx, W., Gruft, L. et al. (1995) Experimental model for neoangiogenesis in adhesion formation. Hum. Reprod., 10, 22902294[Abstract]
Brandt, L. (1965) Adhesiveness to glass and phagocytic activity of neutrophilic leukocytes in myeloproliferative diseases. Scand. J. Haematol., 2, 126129.[ISI]
Dokmetas, I., Doganay, M. and Bakir, M. (1990) IgG therapy of experimental E.coli septicemia in neutropenic mice. Mikrobiyol. Bult., 24, 205213.
Ellis, H., Harrison, W. and Hugh, T.B. (1965) The healing of peritoneum under normal and pathological conditions. Br. J. Surg., 56, 471475.
Harlan, J.M. (1987a) Consequences of leukocyte-vessel wall interaction in inflammatory and immune reactions. Semin. Thromb. Hemostasis. 13, 43444.[ISI][Medline]
Harlan, J.M. (1987b) Neutrophil-mediated vascular injury. Acta Chim. Scand., 715 (Suppl.), 123129.
Harlan, J.M., Schwartz, B.R., Wallts, W.J. and Pohlmann, T.H. (1987) The role of neutrophil membrane proteins in neutrophil emigration. In Movat, H.Z. (ed.), Leukocyte Emigration and its Sequel. Krager, Basel, p. 94.
Heidrick, G.W., Pippitt, C.H., Morgan, M.A. and Thurnau, G.R. (1994) Efficacy of intraperitoneal sodium carboxymethylcellulose in preventing postoperative adhesion formation. J. Reprod. Med., 39, 575578.[ISI][Medline]
Hogstrom, H. and Haglund, U. (1986) Neutropenia prevents decrease in strength of rat intestinal anastomosis: Partial effect of oxygen free radical scavengers and allopurinol. Surgery, 99, 716723.[ISI][Medline]
Jonsson, T. and Hogstrom, H. (1991) Neutrophil-dependent decrease in early wound margin strength. Arch. Surg., 126, 14231431.[Abstract]
Laskin, D.L., Soltys, R.A., Berg, R.A. and Riley, D.J, (1994) Activation of alveolar macrophages by native and synthetic collagen like polypeptides. Am. J. Respir. Cell. Mol. Biol., 10, 5864.[Abstract]
Liebman, S.M., Langer, J.C., Marshall, J.S. and Collins, S.M. (1993) Role of mast cells in peritoneal adhesion formation. Am. J. Surg., 165, 127130.[ISI][Medline]
Linsky, G.B., Diamond, M.P. and Cunningham, T. (1987) Adhesion reduction in the rabbit uterine horn model using an absorbable barrier. TC & J. Reprod. Med., 32, 1720.[ISI][Medline]
Miligan, D.W. and Raftery, A.T. (1974) Observations on the pathogenesis of peritoneal adhesions: a light and electron microscopical study. Br. J. Surg., 61, 274281.[ISI][Medline]
Pados, G.A. and Devroey, P. (1992) Adhesions. Curr. Opin. Obstet. Gynecol., 4, 412418.[ISI][Medline]
Palmgren, M.S., deShazo, R.D. and Carter, R.M. (1992) Mechanisms of neutrophil damage to human alveolar extracellular matrices: The role of serine and metalloproteases. J. Allergy Clin. Immunol., 89, 905915.[ISI][Medline]
Penny, R., Galton, D.A., Scott, J.T. and Eisen, V. (1966) Studies on neutrophil function I: Physiological and pharmacological aspects. Br. J. Haemat., 12, 623632.[ISI][Medline]
Pijlman, B.M., Dorr, P.J., Brommer, E.J.P. and Vemer, H.M. (1994) Prevention of adhesions. Eur. J. Obstet. Gynecol. Reprod. Biol., 53, 155163.[ISI][Medline]
Price, T., Beatty, P.G. and Corpu, S.R. (1987) In vivo inhibition of neutrophil function in the rabbit using monoclonal antibody to CD18. J. Immunol., 139, 41744177.
Raftery, A.T. (1973) Regeneration of parietal and visceral peritoneum in the immature animals: a light and electron microscopical study. Br. J. Surg., 60, 969975.[ISI][Medline]
Reddy, G.K. and Enwemeka, C.S. (1996) A simplified method for the analysis of hydroxyproline in biological tissues. Clin. Biochem., 29, 225229.[ISI][Medline]
Replogle, R.L., Johnson, R. and Gross, R. (1966) Prevention of postoperative intestinal adhesions with combined promethazine and dexomethasone therapy. Experimental and clinical studies. Ann. Surg., 163, 580591.[ISI][Medline]
Rock, J.A. (1991) Infertility. Surgical aspects. In Yen, S.S.C. and Jaffe, R.B. (eds), Reproductive Endocrinology. W.B.Saunders, Philadelphia, p. 710.
Rodgers, K.E. and DiZerega, G.S. (1992) Modulation of peritoneal re-epithelization by postsurgical macrophages. J. Surg. Res., 53, 542549.[ISI][Medline]
Siegler, A.M., Kontopoulos, V. and Wanz, C.F. (1980) Prevention of postoperative adhesions in rabbits with ibuprofen, a nonsteroidal anti-inflammatory agent. Fertil. Steril., 34, 4654.[ISI][Medline]
Stangel, C.L. and Gomel, V. (1980) Postoperative adhesions: Etiology, prevention and therapy. Clin. Obstet. Gynecol., 23, 12131220.[Medline]
Sopata, I., Wojtecka-Lukasik, E., Wize, J. and Maslinski, S. (1989) Neutrophil enzyme activities in carrageenan-induced inflammation in rats. Agents Actions, 28, 8992.[ISI][Medline]
Thompson, J.N., Paterson-Brown, S., Harbourne, T. et al. (1989) Reduced human peritoneal plasminogen activating activity: possible mechanism of adhesion formation. Br. J. Surg., 76, 382384.[ISI][Medline]
Vissers, M.C. and Winterbourn, C.C. (1988) Gelatinase contributes to the degradation of glomerular basement membrane collagen by human neutrophils. Colloq. Relat. Res., 8, 113122.
Yaacobi, Y., Israel, A.A. and Goldberg, E.P. (1993) Prevention of postoperative abdominal adhesions by tissue precoating with polymer solutions. J. Surg. Res., 55, 422426.[ISI][Medline]
Submitted on May 27, 1998; accepted on October 5, 1998.