Search

Chronic administration of the ciliary neurotrophic factor (CNTF) or murine interleukin-6 (IL-6) to mice produced anorexia, lean tissue wasting and hepatic acute phase protein response. Mice showed a progressive reduction in carcass protein which caused increases in the rates of carcass protein breakdown. In contrast, IL-6 feeding had no effect on food intake but produced a similar hepatic acute response, suggesting that CNTF is distinct from IL-6.

Dendritic cells (DC) are unique in their ability to stimulate naive T cells to proliferate and to differentiate into effector T cells. DC, however, can also inhibit T cell activation and play a role in central and peripheral tolerance. IL-10 has been shown to render DC tolerogenic by unknown mechanisms. Using a combined monoclonal antibody/retroviral expression cloning approach, we show here that the inhibitory receptor LIR-2 (leukocyte immunoglobulin-like receptor-2, CD85d) is specifically up-regulated by IL-10 on maturing human DC. LPS-stimulated, LIR-2-transfected DC inhibited the proliferation of T cells in autologous, as well as allogeneic culture systems in vitro. In addition, overexpression of LIR-2 on resting T cells, which lack LIR-2 expression, inhibited T cell proliferation induced by TCR activation. A novel soluble form of LIR-2 was detected in culture supernatants of maturing DC. IL-10 treatment of DC potently inhibited the production of soluble LIR-2. Recombinant soluble LIR-2 was able to completely restore the proliferation of T cells activated with LPS-plus IL-10-treated DC. Thus, IL-10 renders DC hypostimulatory by up-regulating cell surface LIR-2 and by inhibiting soluble LIR-2 in vitro.

Inflammatory bowel disease (IBD) comprises the two disorders ulcerative colitis (UC) and Crohn's disease (CD). Although the etiology is still unclear, initiation and aggravation of the inflammatory processes seem to be due to a massive local mucosal immune response. An increased number of greatly activated macrophages seems to contribute to the onset of IBD by expressing upregulated costimulatory molecules (e.g., CD80/CD86) and a cytokine profile favouring a type I proinflammatory response. The release of interleukin 2 (IL-2) and Interferon-gamma (IFN-gamma) by naive T lymphocytes predominantly stimulates cytotoxic T lymphocytes, macrophages, and natural killer (NK) cells and increases the antigen-presenting potential of all these cell types. Opposite this proinflammatory immune reaction a compensatory type II antiinflammatory response has been suggested in the inflamed mucosa, involving mainly interleukin 4 and interleukin 10. Both cytokines are able to down-regulate inflammatory mediators including tumor necrosis factor-alpha (TNF-alpha) and interleukin 1 and favor a humoral immune response. The main goal of this clinical trial is the local liposome-mediated gene transfer of these two antiinflammatory cytokines, interleukin 4 and interleukin 10, in patients with severe IBD of the rectum. This local administration of antiinflammatory cytokines will avoid toxic systemic side effects, prevents blocking of the beneficial effects of proinflammatory cytokines, e.g., TNF-alpha in other tissue compartments and increases the local concentration of interleukin 4 and interleukin 10 over a prolonged period of time. The combined effects of IL-4 and IL-10 have been shown to shift the Th1/Th2 cell activation in favor of a Th2 immune response which seems to be essential for fighting against the inflammation and ultimative healing.

Background: Atherosclerotic vascular disease is the leading cause of death in patients with diabetes mellitus and end-stage renal disease. Advanced glycation end products (AGEs) are strongly suggested to be involved in the pathogenesis of atherosclerosis in these patients who also frequently experience infectious complications. We hypothesized that the interaction of AGEs and inflammatory mediators contributes to the up-regulation of endothelial cell activation., Methods: We investigated the effect of advanced glycated fibronectin in the presence or absence of inflammatory stimuli on the endothelial cell surface and mRNA expression of cell adhesion molecules. Furthermore, the influence of advanced glycated fibronectin on the transendothelial migration pattern of polymorphonuclear cells was analyzed., Results: Exposure to advanced glycated fibronectin together with inflammatory stimuli such as interleukin (IL)-1alpha, tumor necrosis factor-alpha (TNF-alpha) or lipopolysaccharide (LPS) led to a significant increase in the surface expression of the cell adhesion molecules E-selectin, ICAM-1, VCAM-1 and PECAM-1 on endothelial cells. Soluble AGEs in combination with advanced glycated fibronectin significantly enhanced the endothelial cell surface expression of ICAM-1, VCAM-1 and PECAM-1, whereas this was not the case for E-selectin. At the transcriptional level short-time exposure of endothelial cells to advanced glycated fibronectin and inflammatory mediators resulted in an increased expression of E-selectin, ICAM-1 and VCAM-1 mRNA levels, whereas PECAM-1 repeatedly showed a significant decrease of gene transcript levels. An increase of mRNA levels was also observed for E-selectin, ICAM-1, VCAM-1 and PECAM-1 following incubation with a combination of advanced glycated fibronectin and soluble advanced glycation end-products. Furthermore, polymorphonuclear cells responded with a sevenfold increase in transendothelial migration following exposure of endothelial cells to advanced glycated fibronectin and inflammatory mediators., Conclusions: These results suggest that the combination of matrix glycation and inflammation up-regulates the activation of the endothelial cell adhesion cascade, a mechanism that might contribute to the increased burden of atherosclerotic morbidity and mortality in patients suffering from diabetes mellitus or chronic renal failure.

Systemic anti-cytokine therapies have been unsuccessful in preventing mortality from gram-negative bacteremia in humans partly because of the failure to neutralize pro-inflammatory cytokines at sites of exaggerated production. In an attempt to deliver anti-inflammatory cytokines to organs directly, gene transfer was employed. Thirty-six BALB/c mice were injected intraperitoneally with cationic liposomes containing plasmids encoding the human interleukin-4 (hIL-4) or IL-13 gene. Both, hIL-4 and hIL-13 mRNA were detected by reverse transcription-polymerase chain reaction analysis in the liver and the spleen of the animals. Fourty-eight hours after the in vivo gene transfer, these 36 mice and 18 mock-transfected mice, were challenged with a lethal dose of E. coli lipopolysaccharide with D-galactosamine (D-GalN). Gene transfer with hIL-4 reduced the serum tumor necrosis factor (TNF)-alpha production in response to endotoxin/D-GalN by 80% from 113.1 pg/ml in mock-transfected animals to 22.2 pg/ml (p < 0.05); human IL-13 gene transfer reduced serum TNF-alpha levels by 90% (113.1 pg/ml to 11.6 pg/ml; p < 0.05). Survival was improved from 20% to over 83% in both treatment groups (p < 0.001). Our data demonstrate a potent in vivo anti-inflammatory action of both IL-4 and IL-13. In addition, the immune functions of peritoneal macrophages are significantly ameliorated in both treatment groups, with IL-13 demonstrating better macrophage immune modulation than IL-4 (p < 0.05).

Neutralizing monoclonal antibodies against the murine interleukin 1 (IL-1) type I (mAb 35F5) and type II receptor (mAb 4E2) were used to passively immunize mice prior to exogenous murine IL-1 alpha administration or a sterile-turpentine induced abscess. When mice were passively immunized with 35F5, the anorexia, weight loss and increased plasma acute phase protein levels in response to exogenous IL-1 alpha administration or a turpentine abscess were significantly attenuated. In contrast, passive immunization with 4E2 had only variable effects on food intake, body weight and the hepatic acute phase response in mice administered IL-1 alpha. In mice following a turpentine abscess, type II receptor blockade (4E2) either had no effect, or in some cases, actually increased the plasma IL-6 and acute phase protein responses. We conclude that in response to a turpentine abscess, the anorexia, weight loss and the induction of several hepatic acute phase reactants result in part from IL-1 binding to its type I receptor. Binding of IL-1 to the type II IL-1 receptor does not appear to be involved in the induction of these host nonspecific responses to inflammation.

Objective: To present our experience with laparostomy and necrosectomy in the treatment of acute necrotising pancreatitis, and to show how refinements in our treatment regimen improved mortality over the years despite no reduction in the severity of the disease., Design: Retrospective study., Setting: University hospital, Austria., Subjects: 125 patients treated by laparostomy/necrosectomy with repeated revisions during the period January 1983 to December 1991., Interventions: Laparostomy, blunt necrosectomy, operative lavage, and open drainage., Main Outcome Measures: Mortality and morbidity., Results: The severity of disease was assessed by the APACHE II score (median 15, range 4-30). In 106 of the 125 patients (85%) the necrotic pancreatic tissue was infected. Patients were operated on if they deteriorated clinically or if organ failure was suspected. A change in the protocol from revisions on demand (1983/4) to planned re-exploration at 48 hour intervals (1985/8) was associated with a reduction in mortality from 53% (16/30) to 28% (20/72). This was further reduced in 1989/91 to 17% (4/23) when a protocol of revisions planned for individual patients was introduced (p = 0.02). The incidence of gastrointestinal fistulas during the three periods was 6/30 (20%); 24/72 (33%); and 1/23 (4%); (p = 0.022), whereas that of intraabdominal bleeding remained much the same (7/23, 23%; 13/72, 18%; and 4/23, 17%; p = 0.56). The median (range) APACHE II scores for the three periods were 12 (4-27), 15 (5-30), and 14 (4-25)., Conclusion: By continual revision of our protocol, together with accompanying improvements in intensive care, our mortality decreased significantly during the nine year period.

Objective: The in vivo neutralizing activities of an anti-lipopolysaccharide (LPS) antibody HA-1A (Centoxin [Centocor, Malvern, PA]), a human immunoglobulin M monoclonal antibody, and of bactericidal/permeability-increasing protein (BPI), an endogenously produced human LPS-neutralizing protein, were studied in a primate model of lethal Escherichia coli bacteremia., Summary Background Data: HA-1A has been used with variable success against LPS activity in some animal models and in a recently reported clinical trial. However, no data assessing the efficacy of this agent in subhuman primates is available. Bactericidal/permeability-increasing protein is a product of polymorphomononuclear cells (PMNs) that is stored in azurophilic granules and exhibits LPS-neutralizing activity in vitro and in some in vivo models., Methods: Immediately after E. coli infusion and in a blinded fashion, three baboons were treated with BPI (5 mg/kg bolus infusion and 95 micrograms/kg/min infusion over 4 hr). Three animals received 3 mg/kg BW of HA-1A, whereas another three baboons received a placebo treatment., Results: The BPI-treated animals demonstrated significantly (p < 0.03) lower circulating LPS-limulus amoebocyte lysate (LAL) activity compared with the control animals, but this reduction in LPS-LAL activity was not associated with improved survival. HA-1A treatment did not reduce LPS-LAL activity. However, both BPI and HA-1A treatment did attenuate the pro-inflammatory cytokine response., Conclusion: The current data suggests that incomplete neutralization of endotoxin activity does not alter mortality from severe bacteremia. Given the diversity of mediator production under such circumstances, a strategy of combination therapy in the form of anti-lipopolysaccharide and anticytokine treatment may be necessary to achieve optimal survival.

Human neutrophil azurophilic granules contain an approximately 55-kDa protein, known as bactericidal/permeability-increasing protein (BPI), which possesses a high-affinity binding domain for the lipid A component of lipopolysaccharide (LPS). The in vivo LPS neutralizing activity of exogenous BPI was studied in a model of lethal Escherichia coli bacteremia. Five baboons were treated with BPI (5 mg/kg bolus injection followed by a 95 micrograms/kg/min BPI infusion over 4 hr), while four additional animals received a genetically engineered variant of BPI (NCY103). Five animals received a placebo treatment and served as controls. Both wild-type rhBPI and NCY103 significantly (P < 0.05) decreased blood levels of LPS throughout an 8-hr evaluation period following live bacterial challenge. Two hours following E. coli administration, LPS levels peaked in the controls, at 6.86 +/- 3.22 ng/ml, whereas LPS levels were 3.39 +/- 2.1 ng/ml in the BPI group and 2.04 +/- 1.18 ng/ml in the NCY103 group. Tumor necrosis factor-alpha (TNF-alpha) and interleukin-6 levels likewise were attenuated in the treatment groups, whereas circulating sTNFR I was significantly (P < 0.05) reduced only in the BPI group. Leukocytopenia and granulocytopenia were significantly (P < 0.02) lessened in the BPI group, by an average of 59% leukocytopenia and 65% granulocytopenia, respectively. This study supports the concept of E. coli LPS neutralization by BPI in vivo and demonstrates that a moderate (70%) reduction in peak LPS-LAL activity is sufficient to alter some hematologic and cytokine manifestations of bacteremia.

The appearance of endogenously produced inhibitors against tumor necrosis factor (TNF) (soluble TNF-receptor type I, sTNFR-I) and interleukin-1 (IL-1 receptor antagonist, IL-1ra) was evaluated acutely in five normal patients after experimental endotoxemia lipopolysaccharide (LPS) and prospectively during a one to 11 week period in 12 septic, critically ill patients. Increased levels of both factors remained detectable in the circulation for up to 24 hours after LPS (2 nanograms per kilogram body weight) administration in normal patients. Despite free TNF-a activity being detected only sporadically (3 percent of the samples) and that IL-1 beta was never detectable in the patients in the intensive care unit, IL-6 bioactivity was present in 90 percent of initial samples. Circulating sTNFR-I levels up to 62,000 picograms per milliliter and IL-1ra levels of 14,800 picograms per milliliter were noted in the critically ill patients and remained consistently detectable throughout the extended period of evaluation. While there was no difference in IL-1ra levels between patients who survived or ultimately died, sTNFR-I levels were significantly (p < 0.001) lower in survivors compared with nonsurvivors. A correlation between circulating sTNFR-I and concurrent cortisol levels (r = 0.64; p < 0.002) was also noted. Furthermore, a correlation between sTNFR-I and the severity of initial insult, as assessed by APACHE II scores (r = 0.54; p < 0.01) was demonstrable. These naturally occurring cytokine antagonists likely represent additional indicators of the presence of an infectious or other inflammatory process and seem to persist in the circulation even during conditions in which their respective proinflammatory cytokines are not demonstrable.

Objective: To determine if the inflammatory phospholipid platelet-activating factor (PAF) participated in the symptomatologic, metabolic, and counterregulatory hormonal responses of human endotoxemia., Design: In a double-blind, placebo-controlled study, five subjects received 10 mg of the PAF antagonist Ro 24-4736 orally, while five control subjects received a placebo. Eighteen hours later, all subjects were administered 4 ng/kg of endotoxin (lipopolysaccharide) intravenously., Setting: The Clinical Research Center of The New York Hospital-Cornell Medical Center., Participants: Healthy male volunteers., Main Outcome Measures: Repeated measurements of vital signs, symptoms, cytokine and hormone levels, resting energy expenditure, platelet aggregation, and bleeding times were performed during a 24-hour period., Results: Subjects who were pretreated with the PAF antagonist experienced fewer symptoms, including rigors at 1 hour (P < .05) and myalgias at 1 through 4 hours (P < .05) after administration of lipopolysaccharide. This was in concert with a diminished peak cortisol level (668 +/- 107 vs 959 +/- 159 nmol/L in controls; P < .05), epinephrine secretion (1057 +/- 165 vs 2029 +/- 431 nmol/L in controls; P < .05), and almost complete inhibition of PAF-induced platelet aggregation ex vivo., Conclusions: These findings in the face of unaltered circulating cytokines tumor necrosis factor alpha, interleukin 1 beta, and interleukin 6, as well as the tumor necrosis factor receptor-I s, suggest that PAF may influence some endotoxin-induced, counterregulatory hormonal responses and symptoms through cytokine-independent mechanisms. This study further supports the role of PAF antagonists as an adjunct to cytokine blockade in the treatment of gram-negative sepsis.

Objective: To find out if concentrations of tumour necrosis factor a (TNF alpha) and interleukin-6 (IL-6) play a part in the pathophysiology of intra-abdominal infection, and try to identify patients who would benefit from immunotherapy against TNF alpha., Design: Prospective open study., Setting: University hospital., Subjects: 19 consecutive patients (septic shock, n = 4; sepsis syndrome, n = 6; and no sepsis syndrome, n = 9, classified by the APACHE II score and the criteria of the Methyl-prednisolone Severe Sepsis Study Group) who were to undergo their first operation for intra-abdominal infection., Main Outcome Measures: Correlation between median (interquartile) concentrations of TNF alpha and IL-6 (pg/ml), and APACHE II score, plasma lactate concentration, and organ function., Results: Perioperative concentrations of both TNF alpha (p = 0.001) and IL-6 (p = 0.006) were significantly higher in patients with septic shock. Preoperative cardiovascular and respiratory failure were associated with significantly raised TNF alpha (p < 0.001 in both cases) and IL-6 concentrations (p = 0.02 and p < 0.001, respectively). The preoperative APACHE II score correlated with the increased TNF alpha concentration (r = 0.5, p < 0.001) and plasma lactate concentration with that of IL-6 (r = 0.7, p = 0.003)., Conclusion: Perioperative TNF alpha and IL-6 concentrations correlated with the severity of intra-abdominal infection, so it is possible that patients who present with either septic shock or the sepsis syndrome may benefit from immunotherapy against TNF alpha.

Cachexia and the acute-phase response are common manifestations of inflammation and are presumed to be the product of increased synthesis and release of cytokines, including tumor necrosis factor (TNF), interleukin-1 (IL-1) and interleukin-6 (IL-6). IL-1 receptor blockade has been previously shown to attenuate the weight loss, anorexia and acute-phase protein responses associated with a turpentine abscess. However, IL-1 receptor blockade was also associated with a reduced plasma IL-6 response, suggesting that the benefit achieved by IL-1 receptor blockade may be mediated by reduced systemic IL-6 production. To gain a better understanding of the role of IL-6 in this model of inflammation, C57BL/6 mice were passively immunized with either a monoclonal anti-IL-6 antibody (20F3), an anti-IL-1 type I receptor monoclonal antibody (35F5), a non-immune rat IgG, or a combined therapy of 35F5 and 20F3, before receiving a sterile turpentine abscess. IL-6 or IL-1 receptor blockade equally spared body weight and food intake. Compared to IL-1 receptor blockade, passive immunization against IL-6 further reduced the hepatic acute-phase protein response, as represented by serum amyloid P and complement 3. Combined blockade of IL-6 and IL-1 receptor did not result in a further sparing of body weights or improvement of food intake. These results confirm that IL-1 contributes to host cachexia and the acute-phase response following a turpentine abscess, but also show that these actions are dependent upon an IL-6 response. We conclude that the influence of IL-1 on cachexia and the acute-phase response is mediated, at least in part, through IL-6 and, thus, IL-6 may play a pivotal role in the cachexia and acute-phase response to inflammation.

There is, at present, considerable interest in the possible role for the proinflammatory cytokines, tumor necrosis factor-alpha, interleukin-1, interleukin-6, and interferon-gamma in the pathogenesis of cancer cachexia. Indirect evidence for such a role is based on the observation that chronic administration of many of these cytokines, either alone or in combination, can reproduce the myriad of host responses seen in experimental and human cancer cachexia. Elevated plasma levels of tumor necrosis factor-alpha, interleukin-2, and interferon-gamma have rarely been detected in patients or experimental animals with cancer, although interleukin-6 levels appear to correlate with tumor progression in animal models. The strongest evidence for a causal role for cytokines has come from rodent studies in which tumor-bearing animals have been passively immunized with antibodies directed against individual cytokines. Several groups have shown modest but significant improvements in food intake and lean tissue retention with antibodies directed against tumor necrosis factor-alpha, interleukin-1, interleukin-6, and interferon-gamma. However, there has been no consistent finding that one cytokine is universally involved in cancer cachexia in histologically distinct tumor models. One ominous finding in several tumor models has been that the endogenous production of cytokines appears to support tumor growth. Such findings raise the intriguing possibility that these cytokines, although contributors to tissue wasting and anorexia, may also serve the tumor as either direct or indirect cell growth factors.(ABSTRACT TRUNCATED AT 250 WORDS)

Objective: To identify prognostic factors in a consecutive series of patients with bleeding oesophageal varices and develop an optimum regimen of treatment., Design: Retrospective review., Setting: I Department of Surgery, University Hospital, Vienna, Austria., Patients: 301 consecutive patients with bleeding oesophageal varices., Outcome Measures: Median survival and survival at one year after sclerotherapy alone (n = 213), or sclerotherapy with portosystemic shunt (n = 54), Hassab's devascularisation (n = 29), or liver transplantation (n = 5)., Results: Prognosis was dependent on the severity of liver damage at the start of treatment. Median survival for Child's class A was 47 months, for Child's class B 54 months, and for Child's class C 2 months. The overall one year survival for patients in Child's class C was 33%, for sclerotherapy alone 28%, and for sclerotherapy and portosystemic shunt 42%, Hassab's devascularisation 50%, and liver transplantation 80%., Conclusion: Despite the small number of patients who underwent liver transplantation and their poor initial prognosis (Child's class C, n = 4; class B, n = 1) our results suggest that liver transplantation should be considered for the treatment of patients with end stage cirrhosis and bleeding varices.

One hundred and two patients with acute necrotizing pancreatitis were treated in accordance with a combined regimen of necrosectomy, open drainage by laparostomies, and repeated re-explorations. The severity of pancreatitis was assessed by the APACHE II score (median 15 on admission). Eighty-seven (85%) patients were classified as having infected pancreatic necrosis and only 15 (15%) as having pancreatic necrosis. Overall, 36 (35%) patients died, most of multiple organ failure. Survival was significantly impaired by bacterial contamination of pancreatic necrosis (p = 0.008), bacteremia (p = 0.0001) and infected bronchial secretions (p = 0.05). The mortality rate was reduced from 53% to 28% by changing the regimen of re-explorations from on demand to regular 48 hour intervals. Despite the fact that open packing was associated with a high frequency of gastrointestinal fistulas (30%), this concept seems to be a successful and recommendable approach in the therapy of pancreatic and infected pancreatic necrosis.

Tracheal problems in form of stenosis and malacia are a calculated risk of long-term tracheal intubation. Results with conservative treatment of such problems by bougienage, laser therapy, biopsy, cryotherapy, local steroids, tracheal stenting, and tracheostomy are not satisfactory in a higher percentage of cases. Resectional therapy of benign tracheal lesions has become an established technique, which combines excellent functional results with a low complication incidence. We have treated 40 patients of 17 to 76 years of age with postintubation tracheal lesions by cross resection of the affected segment. Of these patients 40% had received conservative therapeutical steps preoperatively. The mean resection length was 3.0 cm (1.5 to 6.5 cm). The perioperative morbidity was 7.8%, mortality was 2.5%. 85% of the patients operated between 1970 and 1989 were reached for a follow-up examination with x-ray, pulmonary function test and endoscopy. The patients subjective satisfaction with the operative result was good in 85%, minor in 12% and less in 3%. The objective investigations proved very good results in 90%. Our experience confirm the good results of other authors and recommend the resection treatment for cases of postintubation tracheal lesions.

Gastrointestinal lipomas are rare, but commonest in the colon and rectum, characteristically submucosal and seldom subserosal. An 18-year analysis revealed 17 cases of large-bowel lipoma, 13 presenting with colicky pain, abdominal discomfort, blood-stained feces or rectal bleeding and altered bowel habits and four asymptomatic. The 17 patients had totally 21 lipomas, all submucosal. No patients with multiple lipoma had evidence of lipoma at other sites. The ileocecal valve and cecum were most commonly affected, followed by the rectum, sigmoid colon and descending colon. Tumor size (largest diameter) was 0.5-10 cm, averaging 3.1 cm (3.5 cm in symptomatic, and 1.8 cm in asymptomatic patients). The primary diagnosis (with barium enema, colonoscopy and CT) was lipoma in only five cases, but CT gave the correct diagnosis in all three cases in which it was used. Two lipomas were found in surgical specimens from colorectal malignancy, while nine were misinterpreted as polyps and one as angiodysplasia. In symptomatic patients unnecessary colotomy or colonic resection may be avoidable by colonscopic removal of lipoma.

The incidence of postoperative wound infection ranges between 4.6% and 36% after gastrointestinal operations respectively. To evaluate the factors which influence the postoperative wound infection we prospectively analyzed our patients between 1/1989 and 1/1990. 444 patients from three general surgical units of our clinic entered this study. The overall wound infection rate was 6.3%. We classified the patients into 3 operative groups: Group I: subcutaneous operations; Group II: intraabdominal operations without opening the GI-tract; Group III: gastrointestinal operations. Wound infection rate in group I was 1.8%, in group II 7.3% and in group III 13.7%. The differences were highly significant. Both univariate (chi 2-test) as well as a multivariate (Cox-Model) analysis were done. We figured out that classification of patients (p = 0.000), operation time (p = 0.009), operating room (p = 0.000), intensive care unit (p = 0.026), long-term antibiotic prophylaxis (p = 0.001), subcutaneous haematoma (p = 0.000) and length of closed drainage time (p = 0.001) are of significant value. In the Cox model the classification of patients into 3 groups surpassed all the other factors. Postoperative hospital stay was lengthened in patients with wound infection significantly (p = 0.0025).