ABSTRACT
Despite advances in medicine and technology, anastomotic healing problems and leaks are still the most important causes of postoperative morbidity and mortality in colorectal surgery. Growth factors are proven to have significant effects on wound healing. The aim of the present study was to determine the effects of different growth factors on experimental colonic anastomoses and the healing process.
The study included 48 Wistar rats, weighting 200-250 g, which were divided to 6 randomised groups (n=8): Group I-Sham group; Group II-Partial colostomy+colonic anastomosis; Group III-Partial colostomy+colonic anastomosis+125 μg/kg epidermal growth factor; IV-Partial colostomy+colonic anastomosis+6.5 μg/kg keratinocyte growth factor; V-Partial colostomy+colonic anastomosis+16 μg/kg fibroblast growth factor; Group VI-Partial colostomy+colonic anastomosis +150 μg/kg granulocyte-colony stimulating factor. All factors were injected subserosally to the perianastomotic area. On the postoperative 7th day, rats were sacrificed and perianastomotic tissue examples were obtained for colonic bursting pressure, tissue hydroxyproline levels and histopathologic examination.
Colonic bursting pressures were higher in Groups III and V compared to Groups II and VI (p<0.01). Tissue hydroxyproline levels also were higher in Groups III, IV and V compared to Groups II and VI (p<0.001). Histopathologic examination revealed that the healing parameters were higher in Groups III, IV and V, and lesion parameters were higher in Groups II and VI.
This study suggests that local application of epidermal, keratinocyte and fibroblast growth factors in the anastomotic area improves the healing process of colonic anastomoses.
Introduction
Despite the advances in technics, technology and suturing materials, comparing with surgical procedures applied to other organs, colorectal surgery is still one of the most complicating surgical procedure worldwide. Leakage that’s seen after colonic anastomoses increase the patient’s morbidity and mortality.1 It has been reported that when the anastomosis is closer to the anus, leakage rate is higher in elective colorectal surgery.2 Various factors, including preoperative using of steroids, elongation of the operation time and contamination of the surgical area; increase the anastomotic leakage rates.3
Recent studies notify that sufficient blood flow is essential for healing of colonic anastomoses and adequate oxygenization improves collagene synthesis and increases the safety of the anastomoses.4,5,6,7
Wound healing process is composed from inflammation, collagene accumulation and collagene maturation steps. Blood vessels are the most important elements of this early stage of trauma responce.8
The resistance of the anastomosis against streching and other factors is dependent on collagene.9,10 Bowel wall contains collagene type I, III and V.11 Collagene catabolism is higher than synthesis in first postanastomotic day, but in seventh day, thus collagene synthesis passes the catabolism.12 Wound healing is controlled by a variety of regulators.13 Experimental studies have shown that growth factors (GF) demonstrate considerable effects on wound healing.14,15 These factors regulate the cell functions, including growing, differantiation and metabolism.15 Each GF acts on different step of wound healing: e.g. inflammation, matrix synthesis and storage, angiogenesis, epithelisation and contraction.14,15
Epidermal GF (EGF) is a mithogenic polypeptide, containing 53 aminoacides, appears in many different tissues and body fluids of mammals.16 It has mitogenic properties on epithelial and mesothelial cells.17 EGF; exhibits strong action on cell division and differantiation both in vivo and in vitro by binding to own glycoprotein receptor, which is localised on the cell surface of fibroblasts, cornea, lence, small and large intestine, glia and epithelial cancer cells.18
Fibroblast GF (FGF) is a member of a wide polypeptide family and plays role in the regulation of the cell growing and differantiation.19 FGFs demonstrate mithogenic, chemotactic and angiogenetic effects on mesodermal cells, which are very important in tissue repair and regeneration. FGF-1, (acidic FGF) and FGF-2 (alkaline FGF) are the leading members of this family.20,21
Keratinocyte GF (KGF) is a member of the FGF family and is entitled as FGF-7.22 This factor is fairly specific for keratinocytes and induces dermal fibroblasts in case of skin wounds within 24 hours.23 KGF is a 28 kDa single chain polypeptide and is secreted by stromal cells of almost all epithelised organs. KGF effects the epithelial cells by inducing proliferation, migration and morphogenesis.22
Granulocyte-colony stimulating factor (G-CSF) is a 17 kDa glycoprotein. Recent studies showed that; the most of the G-CSF receptors are expressed on neutrophils and a little part on monocytes in vitro. G-CSF is used routinely in afebryl and febryl neutropenia due to chemotherapy, acute myeloid leukemia, myelodysplastic syndrome and bone marrow transplantations.24
Determination of the effects of perianastomotic injection of different GFs on colonic anastomoses in rats composes the goal of this study.
Materials and Methods
Anesthesia and Technique: The present study was performed in Medical and Surgical Research Centre Laboratory of the Eskişehir Osmangazi University Faculty of Medicine after approval of the Local Ethics Commitee.
A total of 48 male Sprague-Dawley rats, weighting 200-250 g, were housed at constant temperature with 14/10 h periods of light and dark exposure, respectively. In acclimation period of at least five days prior to experiments, animals were allowed to access standard rat chow and water ad libitum. Rats were starwated 12 hours before experiments and randomized into six groups (n=8). After anesthesia with subcutaneous injection of 50 mg/kg thiopental sodium (Pental Sodyum, İE Ulagay, Turkey) rats were fixed in supine position, shaved and disinfected with povidone-iodine 10% (Isosol, Merkez Laboratory&Medical Tic., İstanbul, Turkey). Dry operating field covered with sterile drape and median laparotomy was performed. Except for sham group, all contents of caecum and accending colon were pushed distaly and a partial colotomy was performed on accending colon. Thereafter all colotomies were closed with single layer separated 7/0 polypropylene sutures. Sham group rats were subjected only to colonic exploration. Experimental GFs were injected with 26 G syringes in subserosal area of the anastomotic region in groups III, IV, V and VI. Rats were divided into six randomised groups (n=8):
III. EGF + Colon anastomosis group (Grup EGFCA): 125 µg/kg EGF (SIGMA 2008, Germany).
After closure of the abdomen, animals were placed into separated cages and after 12 hours from operation allowed to oral feeding. On the 7th postoperative day relaparotomy was performed in the same fashion and after sampling of 2 cm proximal and distal perianastomotic area of the colon rats were sacrifised with decapitation.
Measurement of Colonic Bursting Pressure (CBP): The feces in sampled colonic lumen was cleaned. After placing of a catheter into proximal edge of the colon both egdes were closed with 2/0 silk in an effort to avoid air leakage. The uncovered edge of the catheter was connected to a standart sphygmomanometer and infussion pump through a fluid infussion set. Prepared colonic segment was placed into a glass container, filled with normal saline. By means of infussion pump, the intracolonic pressure was increased with air, infused with 6 mL/min velocity. The value of sphygmomanometer at the moment of detection of air bubles was recorded as CBP.
Measurement of Tissue Hydroxyproline (THP) Level: Following CBP measurement, the colonic segment was cutted 1 cm proximaly and distally to the anastomosis.
A 2 cm long colonic segment, containing the anastomosis at median point was acquired. This segment was cutted vertically to 2 equal segments, each segment have anastomosis at the median point. One of this segments was placed in 10% formalin solution for histopathologic examination and another was freezed immediately at-700 C for subsequent THP level measurement.
A ready-maid purchased hydroxyproline estimation kit was used for THP level measurement (HypronosticonR, Organon, HOLLAND). By inserting the obtained optic densities into below formulation, the THP level was calculated as mg/L: 2 x N tube’s optic density x 50/S tube’s optic density-N tube’s optic density.
By estimation and calculation of used tissue weight and added fluid amounts, this THP level was converted to THP consentration. The results were expressed as µg hydroxyproline/mg tissue.
Histopathologic Examination: From tissue samples in formalin solution paraffin blocks were prepared, then 4.5 micron in tickness sections were dyed with Haematoxylin-Eosin (H&E) and analysed blindly by a single pathologist. Tissues were assessed in terms of alterations leading to healing of the anastomoses (healing parameters) [i.e. vascular proliferation (VP), collagene tissue proliferation (CTP), fibrous tissue proliferation (FTP), and mononuclear leukocyte infiltration (MNLI)] and in terms of alterations leading to impairment of the anastomoses (lesional parameters) [i.e. mucosal ulseration (MU), perianastomotic oedema (PAO), polimorphonuclear leukocyte infiltration (PMNLI)].
Determined results were scored from 1 to 4 as follows: 0 to 25% changes as 1, 26 to 50% changes as 2, 51 to 75% changes as 3 and 76 to 100% changes as 4.
Statistical Analysis
CBP and THP levels were analysed using One-way ANOVA test, histopathological scores were analysed using Kruskal-Wallis test and multiple comparisons of both analyses were performed using Tukey’s method.
Results
Colonic Bursting Pressure Levels
Comparison of CBP levels of the groups revealed statistically significant difference (p<0.001). CBP levels of the Group S was significantly higher than all of other groups (p<0.001). Comparing the anastomosis groups; CBP values of the EGFCA and FGFCA groups were significantly higher in contrast to Group CA (p<0.01 and p<0.05, respectively). There was no diffference between Groups KGFCA, CSFCA and Group CA (p>0.05) (Figure 1).
Tissue Hydroxyproline Levels
There was significant difference between THP levels of the groups (F: 28.05; p<0.001). Accordingly, THP levels of Group S was significantly lower comparing with other groups, except Group CA (p<0.001). Comparing the anastomosis groups; the values of Groups EGFCA, KGFCA and FGFCA were higher than Group CA (p<0.001, p<0.001 and p<0.01, respectively). There was no difference between Group CSFCA and Group CA (p>0.05) (Figure 2).
Histopathologic Examination of the Anastomoses
Tissue samples of groups were analysed histopathologically by a single pathologist in a blinded fashion and scored in terms of healing parameters and lesional parameters (Figure 3, 4).
Comparison of the Healing Results
Sham Group Histopathological Examination (Picture 1): Normal colonic mucosa.
Vascular Proliferation (Picture 2): Comparison of groups from the point of VP revealed that only Group KGFCA scores were statistically higher than Group S (p<0.001). Comparison of colon anastomosis groups showed that only KGFCA groups scores were higher than Groups CA (p<0.001). There was no significant difference between Groups EGFCA, FGFCA, CSFCA and Group CA (p>0.05).
Collagene Tissue Proliferation (Picture 3): Although scores of all anastomosis groups were higher than Group S, there was no statistically significant difference between groups (p>0.05).
Fibrous Tissue Proliferation (Picture 4): Appraisal of groups in terms of FTP demonstrated that there was statistically significant difference between Groups FGFCA, KGFCA, CSFCA and Group S (p<0.001). Comparison of colon anastomosis groups indicated that there was significant difference between Groups FGFCA, KGFCA, CSFCA and Group CA (p<0.01, p<0.01 and p<0.001, respectively). There was no difference between Group EGFCA and CA (p>0.05).
Mononuclear Leukocyte Infiltration (Picture 5): The MNLI scores of all groups were significantly higher than Group S (p<0.001). Comparison of colon anastomosis groups revealed that there was significant difference only between Group EGFCA and Group CA (p<0.001). There was no difference between Group KGFCA, FGFCA, CSFKA and Group CA (p>0.05).
Comparison of the Lesional Results
Polimorphonuclear Leukocyte Infiltration (Picture 6): The PMNLI scores of all groups were higher than scores of Group S (p<0.001). Analysis of colon anastomosis groups demonstrated the decrease of scores in Group EGFCA and FGFCA in comparison with Group CA (p<0.001). There was no significant difference between Group FGFCA, CSFCA and Group CA (p>0.05).
Perianastomotic Oedema (Picture 7): The scores of Group CA and CSFCA were higher than Group S (p<0.001); however there was no distinction between Group EGFCA, FGFCA, KGFCA and Group S. The scores of Group EGFCA, FGFCA and KGFCA were better than scores of Group CA (p<0.001); there was no difference in comparison of group CSFCA and CA (p>0.05).
Mucosal Ulseration (Picture 8): The comparison of MU scores revealed that there was no difference between groups, although the scores of Groups EGFCA, FGFCA, KGFCA were better than Group S. There was no difference also between colon anastomosis groups (p>0.05).
Discussion
Despite advances in technology and colorectal surgery, there is still high morbidity and mortality rates particularly due to anastomotic leakages.1,2 The studies reported that when the anastomoses are closer to anus, leakage rates are higher.2,25 Emergency colorectal surgery and attenuation of perianastomotic blood flow due to trauma or technical problems are the other important factors leading to anastomotic leaks.1 Additionally, various factors like preoperative use of steroids, elongation of operation time and contamination of operation site increase the leaks rates.3 Platell et al.2 notify 2.4% overall leakage rate in a series of 1598 patients subjected to colorectal anastomoses while leakage rate is 1.5% in intraperitoneal anastomoses, the rate is 6.6% in extraperitoneal anastomoses.
Various physical and mechanical factors, such as different drug trials, examinations of angiogenetic mechanisms, a variety of colon cleaning methods, new suturing materials and surgical techniques were analysed in order to improve anastomotic safety.26,27 Experimental study performed by Lord et al.28 in order to compare colonic anastomoses performed with different sutures, such as chromic catgut, silk, polyglycolic acid, polypropylene, and teflon reported the eligibility of polypropylene and teflon sutures to cause less damage and inflammation. Chung29 compared the single layer and double layer anastomoses and informed that double layer anastomoses reduce the mucosal blood flow. In accordance with literature, the anastomoses in present study were performed with single layer polypropylene.
In published studies mechanical strenght of anastomoses was measured via two principal methods: CBP was the more prefered, and other was longitudinal breaking strenght.8,30 Biochemical method used for determination of the anastomotic healing in several studies was THP levels, and this measurement was accepted equal to collagene levels.8
Histopathology was also a beneficial assessment criterion in studies regarding to colonic anastomoses.31 These leading parameters were used in our study.
Each GF affects on one or more different steps in wound healing process. These steps are inflammation, matrix synthesis and storage, angiogenesis, epithelization and contraction.14,15 Recent studies research the effect of EGF on wound healing 18. EGF saturated sponges applied locally to colonic anastomoses, have shown to reverse the inhibitory effects of systemically applied methylprednisolone on wound healing.32
The appraisal of perianastomotic local EGF injection on colonic anastomoses revealed that EGF increases the CBP and THP levels and therefore the perianastomotic collagene levels. Additionally EGF has histopathologic benefits. MNLI has been significantly increased.
Although not singificantly, EGF increases other healing histopathologic parameters.
Furthermore the lesional parameters have been decreased. EGF significantly decreases the PMNLI and PAO. Although not significantly, the perianastomotic MU has been decreased.
The overall effects of EGF give chance to think that may be used in healing of colonic anastomoses. The insignificance of histopathologic parameters better than CA group, this may be explained with the insufficient number of experimental animals.
KGF is secreted by stromal cells of almost all epithelised organs and is mithogenic and chemotactic for epithelial cells.15,33 Egger et al.34 determined the beneficial effects of intraperitoneal EGF on colonic anastomoses and Cui et al. 35 on experimental esophagogastric anastomoses.
The appraisal of perianastomotic local injection of KGF in the present study revealed that; KGF increases the CBP, but not significantly and significantly increases the perianastomotic THP levels. Additionally histopathologic evaluation revealed the beneficial effects on VP and FTP. Although not significantly, other healing histopathologic parameters ameliorated. Moreover PAO decreased significantly. Other perianastomotic lesions diminished, but not significantly. These results showed that KGF is less effective than EGF and FGF on healing of colonic anastomoses. FGF is mithogenic for mesenchymal cells, stimulates angiogenesis and plays important role in wound healing.20,36 Nurata et al. 37 in an experimental study, designed to investigate the effects of local FGF on duramather injuries with serebrospinal fluid leaks, determined the favourable effects of FGF on healing of duramather. Ernst et al.38 found that local administration of FGF accelerates the healing of gastric ulcers in experimental conditions. The assessment of animals subjected to perianastomotic local FGF injection revealed the following results: 1) the CBP and perianastomotic THP levels and accordingly the collagene amount were increased by FGF; 2) FGF has positive effects on histopathological healing parameters. The FTP was significantly increased and although not significantly other healing parameters also have been increased; 3) additionally perianastomotic lesions have been reduced by FGF administration. There is significant decrease in PMNLI and PAO. Although not significantly the perianastomotic MU also have been decreased. All these detected effects occupies that FGF also may be used in healing of colonic anastomoses although less effectiveness to EGF. G-CSF stimulates the colony growth specific to granulocytes and provides the functional activation of PMNLs.39 Grzybowski et al.40 have determined the favourable effects of locally administered G-CSF, GM-CSF and EGF on healing of skin incisions, but these findings were not confirmed by other similarly desinged studies. Present study showed that there was not significant increase in CBP and perianastomotic THP levels after injection of G-CSF. On histopathological point of view, among healing parameters only the FTP was significantly increased by G-CSF administration. Additionally there was not decrease in perianastomotic lesions. These effects demonstrate that G-CSF has not significant contribution on healing of colonic anastomoses.
Evaluation of all results determined in our study demonstrated that; G-CSF has not beneficial effects on wound healing of colonic anastomoses, but KGF, FGF, and particularly EGF are agents which may increase the anastomotic safety and resistance against the intracolonic tension strength of colonic anastomoses. This experimental trial is suggestive in terms of demonstrated that the local use of EGF, KGF, and FGF in colorectal surgery may provide benefits to surgeons, however most efficient GF need to be tested. The effectiveness and dose dependent benefits and the exact way of action and administration way of these drugs need to be investigated via further and detailed studies.
Ethics
Ethics Committee Approval: DCR-D-10-00115.
Peer-review: Internally peer-reviewed.
Authorship Contributions:
Surgical and Medical Practices: Ergin Kopal, Bartu Badak, Necdet Fatih Yaşar, Ersin Ateş, Concept: Ergin Kopal, Bartu Badak, Necdet Fatih Yaşar, Ersin Ateş, Design: Ergin Kopal, Ersin Ateş, Data Collection or Processing: Bartu Badak, Necdet Fatih Yaşar, Analysis or Interpretation: Ergin Kopal, Bartu Badak, Necdet Fatih Yaşar, Ersin Ateş, Literature Search: Bartu Badak, Necdet Fatih Yaşar, Writing: Bartu Badak.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.
