Liver cirrhosis on the colonic anastomotic healing in rats 1 Cirrose hepática na cicatrização de anastomose intestinal em ratos

PURPOSE: To investigate the effects of cirrhosis on colonic anastomosis healing in rats. METHODS: Fifty five Wistar male rats were used (23 in the control group and 32 in the cirrhosis group). On the first day of the procedure, the rats in the cirrhosis group underwent double ligation and folding of the common bile duct to induce liver cirrhosis, and the control rats underwent a laparotomy and intestinal manipulation. On the fourteenth and thirty-fifth days, all of the animals were biochemically assessed for serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase, bilirubin, total protein, and albumin and for liver histopathology. On the thirty-fifth day, cirrhosis was confirmed. On the twenty-eighth day, all of the animals were subjected to left colon transection and anastomosis. On the seventh day after the colonic anastomosis, the rats were sacrificed and macroscopically evaluated for dehiscence. The region of the colonic anastomosis was removed and subjected to hydroxyproline content measurement, conventional histology, and the immunohistochemical determination of vascular endothelial growth factor (VEGF) and matrix metalloproteinase type 1 (MMP 1). RESULTS: The biochemical and histopathological examinations confirmed cirrhosis in all of the animals in the cirrhosis group. More deaths occurred after anastomosis in the cirrhosis group (5/25) than in the control group (0/21), and anastomotic dehiscence was more frequent in the cirrhosis group (8/25) than in the control group (0/21). The average hydroxyproline concentration was lower in the cirrhosis group than in the control group. The immunohistochemical studies showed that the average VEGF expression in the cirrhosis group was lower than in the control group, and the average MMP1 expression was higher in the cirrhosis group. CONCLUSION: Hepatic cirrhosis leads to increased mortality and colonic anastomotic dehiscence, an increased distance between the mucosal edges of the anastomosis area, and a lower hydroxyproline concentration in the colonic anastomosis; therefore, these conditions are deleterious to healing.

Certain systemic conditions, such as hypotension, hypovolemia, intestinal ischemia, low oxygen tension, advanced malignant neoplasia, dehydration, uremia, age and malnutrition, can affect healing [8][9][10][11][12][13][14] .Some studies have shown a higher incidence of colonic anastomotic dehiscence in patients with intraoperative hypotension, anemia, and cirrhosis 1,6 .Some studies have shown that liver cirrhosis interferes with the healing of abdominal incisions and decreases collagen in these areas, which leads to dehiscence [15][16][17][18][19] .Other authors have published studies evaluating the healing of peptic ulcers in patients with hepatic cirrhosis 20 , but there are few studies on hepatic cirrhosis and anastomosis healing [21][22][23] .Patients with cirrhosis may require colonic anastomosis, which causes concern on the part of the surgeon 22 .This study evaluated the influence of hepatic cirrhosis on the healing of colonic anastomosis.

Methods
Healthy, male Wistar rats (55 total; 300 grams, 60 days old; Rattus norvegicus) were used in this study.The animals were divided into two groups: cirrhosis and control.
To induce liver cirrhosis, we used an experimental model that consisted of performing a double ligation and folding the common bile duct of the rat 24,25 .Clinical evaluations were performed by staining the sclera, performing choleric diuresis, and confirming cholestasis by determining the serum AST, ALT, bilirubin, alkaline phosphatase, total protein, and albumin concentrations 25 .We also performed a liver biopsy to confirm cirrhosis.
On the first day of the experiment, the rats in the cirrhosis group underwent biliary obstruction, and the rats in the control group underwent a laparotomy and bowel manipulation.On the fourteenth day, all of the animals were subjected to blood withdrawal (2 ml) from the tail vein for biochemical examination.
The rats in the cirrhosis group were injected with 0.2 ml of a solution containing 0.02 ml of vitamin K diluted in 0.2 ml of saline.
The control rats received 0.2 ml of saline subcutaneously.On the twenty-eighth day, all of the animals were subjected to transection and anastomosis of the left colon and received a subcutaneous injection of either vitamin K (cirrhosis group) or saline (control group), as previously described 8 .On the thirty-fifth day (seven days after the anastomosis), the animals were sacrificed, and their abdominal cavities were evaluated macroscopically for the presence of anastomotic dehiscence and peritonitis.Blood (2 ml) was also collected from the inferior vena cava for the biochemical study.A liver biopsy and a 2-cm resection from the area containing the anastomosis were collected to determine the colonic hydroxyproline content and for histopathological and immunohistochemical evaluations.A biochemical analysis was performed to determine the hydroxyproline concentrations 26,27 .
The hepatic biopsy tissues were stained with hematoxylin and eosin and Masson's trichrome, and the presence or absence of hepatic cirrhosis was evaluated histopathologically.The tissues containing areas of anastomosis were separated and evaluated by immunohistochemistry for VEGF) and MMP1.
Slides were prepared by immunohistochemistry for VEGF and MMP1 and were analyzed with an IBM-PC microcomputer and an Zeiss-Axiophot light microscope.The images were captured with a JVC-TK1270 camera and analyzed with KS 400 (version 2.0; Kontron Elektronik GmbH).We developed two macro applications for capturing the images and analyzing the immunohistochemical staining.We studied 20 fields in the center of each anastomosis, starting from the edge of the mucosa to the serosa.After capturing the images, the program calculated the percentage of the stained area, and the values were statistically analyzed.
Statistical analyses were performed using a nonparametric Mann-Whitney test 28 to analyze the bilirubin, AST, ALT, alkaline phosphatase, total protein, albumin, hydroxyproline, VGEF, and MMP1 results.We used Fisher's exact test to compare the macroscopic findings of the abdominal cavities (adhesions and anastomotic dehiscence) and the death rates.p<0.05 was considered statistically significant.

Results
On the fourteenth day (Table 1) and on the thirty-fifth day (Table 2) of the experiment, the animals in the cirrhosis group had significant increases in their average concentration of total bilirubin, direct bilirubin, AST, ALT, and alkaline phosphatase but a decreased concentration of albumin compared to control group.However, all of the concentrations were within normal ranges.
The biochemical values of the cirrhosis group on the fourteenth and thirty-fifth days were compared, and significant differences were observed for ALT, the total protein, and albumin (Table 3).

Clinical course of the animals
The cirrhosis group had a significantly greater number of deaths in the post-anastomotic period (p<0.05)compared to the control group.All of the animals in the cirrhosis group that survived until the day of sacrifice developed hepatic cirrhosis, which was confirmed by histological examinations.There was a greater frequency of anastomotic dehiscence in the cirrhosis group (8/20) than in the control group (0/22) (p<0.001).

Biochemical evaluation of the anastomosis
The mean hydroxyproline concentration was significantly lower in the cirrhosis group (430.46 ± 75.06 mg/g) than in the control group (662.27 ± 126.59 mg/g) (p<0.0001)(Figure 1).

Immunohistochemical analysis
VEGF expression was lower in the cirrhosis group (0.39 ± 0.14 of infiltrated area/mm 2 ) than in the control group (0.42 ± 0.12 of infiltrated area/mm 2 ) (Figure 2).MMP1 expression was greater in the cirrhosis group (0.39 ± 0.13 of infiltrated area/ mm 2 ) than in the control group (0.33 ± 0.06 of infiltrated area/ mm 2 ) (Figure 3).The animals in the cirrhosis group had a greater distance between the edges of the mucosal repair, which revealed worse healing (Figures 4 and 5).We used an experimental model to induce hepatic cirrhosis in rats, which consisted of double ligation and folding of the common bile duct.This model was used because it has the lowest incidence of duct recanalization 25 .
On the fourteenth day of the experiment, the biochemical analyses and liver functions were significantly different between the two groups for all of the serum parameters.The bilirubin, AST, ALT and alkaline phosphatase measurements were higher in the cirrhosis group than in the control group.These results were expected because common bile duct ligation is an excellent method for inducing extrahepatic cholestasis in rats 25 .
On the thirty-fifth day, the mean albumin level in the cirrhosis group (1.61 ± 0.34 mg/dl) was significantly reduced to below the normal level (2.93 ± 0.56 mg/dl).These findings show that cholestasis (but not liver cirrhosis) was established fourteen days after the bile obstruction.
More deaths occurred in the cirrhosis group during the period after the anastomosis; however, no rats died during this period in the control group.This difference was statically significant.
There was no anastomotic dehiscence in the control group, but dehiscence was observed in eight animals in the cirrhosis group (40.0%).This difference between the groups was statistically significant.In a study of fifty-eight cases of abdominal wall dehiscence, a significant relationship was found between dehiscence and hypoproteinemia 29 .Jaundice, when not accompanied by hypoproteinemia, does not increase the rate of dehiscence, probably because hepatic function is not significantly impaired before hypoproteinemia is established.Increased rates of abdominal wall dehiscence and incisional hernias have been found in jaundiced patients 20 .
We found a significantly lower hydroxyproline concentration in the cirrhosis group than in the control group.
Hydroxyproline is essential for the structural stability of collagen at the end of its synthesis and is very abundant.Proline and hydroxyproline constitute about 25% of the collagen molecule and provide stability to the molecule because they are made of fibrils.Hydroxyproline is measured to biochemically analyze the collagen concentration in the scar tissue, and its increased concentration is commonly associated with lower rates of healing.
A lower hydroxyproline concentration has been observed in the healing of the abdominal walls of cirrhotic rats 17,30 .
Healing worsens as the bile obstruction is prolonged 30 .In this study, we evaluated anastomosis and the lower hydroxyproline concentration, which correspond to less collagen and less efficient healing in cirrhotic rats.In the control group, the anastomosis area was virtually reconstructed and repaired on the seventh postoperative day.The distance between the mucosal edges was lower in the control group than in the cirrhosis group.This indicates less capacity for regeneration and epithelialization in the anastomosis areas of the rats in the cirrhosis group.
There was no statistically significant difference in VEGF expression between the two groups at the anastomosis level, but the cirrhosis group had less VEGF expression and angiogenesis.Thus, hepatic cirrhosis may have a possible anti-angiogenic effect and hinder healing of the anastomosis.This conclusion is based on the macroscopic and biochemical parameters that were analyzed.There was more MMP1 in the cirrhosis group than in the control group, but this difference was not significant.We know that metalloproteinases degrade collagen, and, therefore, have deleterious effects on healing.Our findings confirm the negative results regarding the hydroxyproline content and the increased number of dehiscences in the cirrhosis group.

Conclusion
Hepatic cirrhosis increases post-operative mortality and has deleterious effects on the healing of colonic anastomosis.

FIGURE 2 -
FIGURE 2 -Cirrhosis group.There is greater distance distance between the edges and mucosal repair area.

FIGURA 4 -FIGURA 5 -
FIGURA 4 -Control group.Smaller distance between the mucous membrane and the repair area.

TABLE 1 -
Mean values for serum determinations -14 th experimental day.

TABLE 2 -
Mean values for serum determinations -35 th experimental day.

TABLE 3 -
Mean serum levels on 14 th and 35 th experimental Day -Cirrhosis group.