ABSTRACTS Twenty-fourth European Congress on Surgical Infections Leon,Spain May 25–28,2011

Dine Alan 1 Singh Jay 2 Cohen Stephen 3 Lieberman Harry 3

1 I-Flow Corporation 2 Piedmont Hospital 3 Atlanta Colon and Rectal Surgery, USA 1. Postoperative Pain Management Strategies as Risk Factors for Surgical Site Infection

Introduction and objectives: Surgical site infections (SSIs) occur in up to 2–5% of clean extra abdominal procedures and up to 20% of intra abdominal procedures. These infections are a major contributor to patient injury, mortality and health care costs. The consequences of these infections may result in increased pain, increased hospital length of stay (LOS), increased costs to patient and facility, and potentially sepsis and mortality. Narcotics are the current standard of care for postoperative pain management after surgery. Both unrelieved pain and narcotic usage have been implicated as additional risk factors for SSIs. The use of a continuous infusion of local anesthetic to the surgical site has been shown to significantly reduce both pain and narcotic usage and may impact the risk of SSIs.

Material and methods: We undertook a prospective, multicenter, randomized, open, controlled, surveillance study of 289 patients undergoing elective colorectal surgery. Following IRB approval, patients were randomized to receive either continuous, surgical site, local anesthetic infusion with the ON-Q® SilverSoaker™ (ON-Q group) or traditional narcotic pain management (control group) in order to compare the SSI rate associated with each modality. Outcome measures included the presence or absence of SSI at 30 days postoperative, hospital length of stay (LOS) and readmission rates due to SSI.

Results: A total of 289 patients were recruited for the study: 152 in the ON-Q SilverSoaker study group and 137 in the control group. The incidence of SSI was found to be significantly less in the ON-Q group compared to the control group (6.6% versus 14.6%, p = 0.033). The LOS was lower in the ON-Q group compared to the control group (6.1 days versus 8.4 days, p = 0.0003).

Conclusions: The goal of achieving a 0% SSI rate is elusive, as there are many risk factors that present themselves when patients have surgery. These known risks are diabetes, obesity, length of procedure, and clean vs clean contaminated vs contaminated cases. The effect that pain and pain management therapies have on the risk of SSI is well documented yet less widely understood. Narcotic and pain reduction protocols may provide another link in the reduction of SSI. This study reveals that a continuous, surgical site, local anesthetic infusion with ON-Q SilverSoaker may be an effective approach to further reduce SSIs and LOS in patients undergoing elective colorectal surgery.

Sancho-Insenser JJ 3 Membrilla-Fernández E 3 Argudo-Aguirre N 3 Luna-Aufroy A 1 Girvent-Montllor M 2 Sitges-Serra A 3

1 Hospital Parc Taulí 2 Universitat Pompeu Fabra 3 Hospital Universitari del Mar, Catalonia, Spain 2. Significance of Enterococcus Spp. in Secondary Peritonitis. Results of a Prospective Multicentric Study

Introduction: The evolution of emergent pathogens in both community-acquired (CAP) and postoperative peritonitis (POP) drives the evolution of empiric antibiotic therapy. One of the more prominent emergent pathogens has been Enterococcus spp. with a sustained increasing prevalence. The benefits and suitability of its coverage by empiric treatment (ET) has been the object of intense debate.

Patients and methods: During 28 months, an observational prospective multicentric clinical study was carried out in 24 General Surgery Departments. Inclusion criteria were patients operated on for peritonitis encompassing at least two quadrants and with positive peritoneal cultures. A comprehensive general and peritonitis-specific work-up was carried out in every patient. Antibiotics administration regimes, resistance profiles and re-infection/super-infection patterns centred the analysis, focused on Enterococcus spp. The end-points were success/failure of ET and complications.

Results: There were 362 peritonitis, 164 women and 198 men. Mean age 62 yr. IC_95% = [58.7–64.3], including 262 CAP(72.4%) and 100 POP(27.6%). Enterococcus spp. was isolated in 113 samples from 101 patients (31.2% of all patients), and was significantly more prevalent in POP (53% vs 23%;P < 0.0001). Isolated species were E.faecalis 51%, E.faecium 22%, E.avium 16%, E.durans 4%, and undetermined 7%. Resistance patterns included Imipenem/cilastatin (22%), Piperacillin/tazobactam (30%), and two Vancomycin-Resistant strains both Linezolid sensible. Enterococcus was the leading cause of inadequate empiric treatment: Out of 101 patients with Enterococcus, it remained untreated by ET in 55 (54%), whereas only 30% of patients with any other germ (including C.albicans) were left untreated by ET (P < 0.00001). Enterococcus was associated to higher rates of treatment failure (35% vs 21%;P = 0.001), both as super-infection (P = 0.0015) and re-infection (P = 0.026). Enterococcus was also overrepresented in the group of patients with treatment failure, appropriate source control and appropriate ET (P = 0.018). Among comorbidities, prevalence of Enterococcus was associated with malignancy (44% vs. 22%;P = 0.045). Patients harbouring Enterococcus were older (65 ± 16 vs. 60 ± 19 yr; P = 0.027) and had higher ASA (P = 0.021) and McCabe&Jackson (P = 0.004) scores. Systemic inflammatory response was stronger in patients with Enterococcus, as signalled by higher prevalence of tachycardia (P = 0.009), tachypnoea (P = 0.009) and azotemia (62 ± 49 vs 48 ± 33; P = 0.002). Patients with Enterococcus suffered a higher proportion of septic complications (P = 0.004), surgical site infection (P = 0.030), septic shock (P = 0.001), ICU admissions (P = 0.009), overall mortality (58% vs 25%;P = 0.0001) and even higher mortality after CAP (13.5% vs.3.6%;P = 0.004).

Conclusions: Prevalence of Enterococcus spp. is growing, especially in older patients and severe peritonitis, and it is associated to more septic complications and mortality. Enterococcus is also the germ leading the inadequacy of ET, and its coverage in secondary peritonitis is recommended.

Badia-Perez JM 2 Jimeno J 3 Aldeano A 2 Navinés J 2 Oms LL 1 Guirao X 2 Sitges-Serra A 3

1 Hospital General de Terrassa 2 Hospital General de Granollers 3 Hospital Universitari del Mar, Catalonia, Spain 3. Randomised Trial of a Short Course of Postoperative Antibiotic Therapy in Low-Risk Acute Cholecystitis

Introduction: Short courses of antibiotics have been suggested for resectable surgical infections. Early surgical operation is the treatment of choice for acute cholecystitis. Adjuvant antibiotic treatment is always given, but the optimal choice, timing and duration of antibiotics in operated cholecystitis remain unclear. Aim: To compare the early postoperative results in terms of surgical site infection (SSI), other morbidity and hospital stay between two groups of patients operated for acute cholecystitis with different lenght of antibiotic therapy.

Methods: Multicenter prospective randomised clinical trial comparing a short course of antibiotic treatment (SC: preoperative treatment with Piperacillin-tazobactam until 24 hours of operation) with a long course (LC: same treatment prolonged for 5 days after surgery) for acute cholecystitis treated by early cholecystectomy. Inclusion criteria were: acute cholecystitis confirmed by ultrasonography, admitted <2 days after the onset of symptoms, and operated <4 days of admission. Exclusion criteria were: subhepatic abscess, biliary peritonitis, suspicion of cholangitis, and confirmed bacteremia. Patients were followed for 30 days after operation. An intention-to-treat analisis was used. The chi-square and Student-Fischer tests were used for statistical analysis. Level of significance was stated at p < 0.05.

Results: One-hundred forty-five patients entered the study. Six were excluded, 66 were allocated to SC group and 73 to LC group. Both groups were comparable in sex, age, weight, level of inflammation encountered at surgery, surgical technique (open or laparoscopic), and duration of operation. ASA level were higher in SC group (ASA I-II: 54 SC vs 49 LC, p = 0.01). No statistical differences were found in non-infectious postoperative complications (4.5% SC vs 6.8% LC, p = 0,7) or SSI: overall (SC 13.5% vs LC 8.2%, p = 0.4), superficial wound infection (SC 10.6% vs LC 8.2%, p = 0.8), and organ-space infection (SC 3% vs LC 1.4%, p = 0,6).

Conclusions: In a preliminary analysis of results, no significant differences were found between a short course of antibiotic treatment -covering the preoperative period and the day of operation- and a long course after early cholecystectomy in uncomplicated cases of acute cholecystitis.

Nespoli L Tamini N Bolzonaro E Dinelli M Ardito A Frego R Redaelli A Nespoli A Gianotti L Milano-Bicocca University, Italy

4. Comparison of Endoscopic Stenting as a Bridge to Elective Operation Versus Emergency Surgery in the Treatment of Acute Colonic Obstruction. Impact on Surgery-Related Infectious Complications

Introduction and objective: Colonic obstruction often requires an emergency treatment. Surgery in such setting may be demanding for the surgeon, at high risk for the patient and often is a temporary or palliative therapeutic option. Self-expandable metallic stent (SEMS) might be an alternative as a bridge to elective operation or definitive palliation avoiding surgery, but its role and safety is still debated. The aim of the present study was to evaluate the outcome of patients with colon obstruction and treated with different approaches.

Materials and methods: We admitted 117 patients with large bowel obstruction over a period of 6 years. Sixty-seven of 117 (57.2 %) were treated by endoscopic stenting. Forty-two of 67 (62.7%) patients with the purpose of bridging to elective surgery (SEMS group) and 25 (37.3%) of definitive palliation. Fifty of 117 (42.8%) patients underwent emergency surgery (ES) without stenting.

Results: Obstruction was successfully treated by SEMS in 97% of the cases (65/67) within a median of 24 hrs from admission. We observed 3 (4.5%) self-limiting bleeding and 3 stent migration with necessity of further endoscopic replacement. No perforation was recorded. In 2 patients (3%) the endoscopic procedure failed for technical reasons. We compared the SEMS group to the ES group. The median time from SEMS positioning to elective operation was 5 days. The two groups were well matched for baseline characteristics. The overall rate of postoperative complications was respectively 45.2% (19/42) vs.72.0% (36/50) (p = 0.009).

OPEN IN VIEWER

Complication	SEMS	ES	P value
Wound infection	26.2% (11\42)	54.0% (27\50)	0.007
Pneumonia	9.5% (4\42)	38.0% (19\50)	0.002
Urinary tract infect	9.5% (4\42)	12.0% (6\50)	0.492
Abdominal abscess	11.9% (5\42)	38.0% (19\50)	0.005
Septic shock	4.8% (2\42)	10.0% (5\50)	0.345
Peritonitis	4.8% (2\42)	8.0% (4\50)	0.437
Re-operation	7.1% (3\42)	16.0% (8\50)	0.198
ICU for infections	9.5% (4\42)	28.0% (14\50)	0.032

The mean length of hospitalization was 13.9 ± 6.3 in the SEMS group vs. 22.5 ± 20.2 days in the ES group (p = 0.01).

Conclusions: Endoscopic colon stenting is safe and feasible to manage patients with acute obstruction. Preoperative SEMS as a bridge to elective surgery is associated with a significant reduction of infectious complications and a shorter hospitalization when compared with emergency surgery.

Carlisle E 3 Poroyko V 3 Caplan M 1 Alverdy J 3 Morowitz M 2 Liu D 3

1 Northshore University Healthsystems 2 University of Pittsburgh Medical Center 3 University of Chicago Medical Center, USA 5. Gut Microbiota and Host Genome are Significantly Altered After 72 Hours of Formula Feeding

Introduction: Formula fed infants are at increased risk for bacterial mediated intestinal diseases such as necrotizing enterocolitis. Yet the impact of diet on infant gut bacterial colonization and host gene expression remains relatively unexplored. Here, we quantitatively compare the colonic microbiota and transcriptional profile of formula fed mice (FF) and maternal fed mice (MF) to test the hypothesis that diet impacts gut colonization and host gene expression. Methods: 7 day old C3HeB/FeJ pups were randomly assigned to 72h of FF or MF. After 72h of differential feeding, colonic tissue was collected for DNA and RNA extraction. 16S ribosomal subunit sequences were amplified and sequenced on the Roche GS-FLX pyrosequencing platform. Taxonomy was assigned to sequences using the Ribosomal Database Project classifier tool. Additionally, cDNA libraries were constructed and sequenced using the Solexa sequencing platform. cDNA reads were annotated by BLAST search against the mouse RNA database(NCBI build 37). Reads were functionally classified using KOG database (NCBI). The differentially abundant features were identified via Metastats methodology.

Results: After only 72h of differential feeding, 16S pyrosequencing demonstrated that Firmicutes (p < 0.001) was the dominant phylum in the MF pups while Proteobacteria (p < 0.001) was dominant in FF mice. At the genus level, colonic microbiota of FF mice had significantly more Streptococcus (p < 0.001) than that of MF mice. However, the colonic microbiota of MF mice had a greater abundance of Lactobacillus (p < 0.001) and Enterococcus (p < 0.001) than that of FF mice. After quality filtering, Solexa sequencing of 8 cDNA libraries resulted in 29,315,823 reads with average read length of 100bp. Blast search against mouse RNA database yielded 18,014,657 hits to 19,703 mouse mRNAs. Statistical analysis revealed relative similarity of transcriptional patterns between MF and FF mice. However, a statistically significant (p < 0.05) difference in mRNA transcript number was detected for 149 genes. Most notable was a 25 fold increase in hemeoxygenase1 (Hmox1), a gene involved in protection against oxidative mediated stress injury, in FF mice. This suggests that formula feeding alone may induce oxidative stress in the gut. Additionally, a 47% decrease in Vinculin (Vcl) (p < 0.05), a cytoskeletal protein associated with adherens junctions, and a 49% decrease in serine peptidase inhibitor, clade H, member 1 (Serpinh1) (p < 0.05), a protein that stabilizes collagen cross-linkages, in FF pups suggests that gut structural integrity may be impaired with formula feeding.

Conclusions: Formula feeding shifted both the intestinal microbiota and host genes involved in the structural integrity and oxidative response of the epithelial barrier-key features of necrotiing enterocolitis. Further in depth interrogation of microbial and host gene expression patterns in formula fed infants has the potential to shed new light on the pathogenesis of NEC and improve current feeding formulations.

ten Broek R van Goor H Radboud University Nijmegen Medical Centre, The Netherlands

6. Adhesiolysis Time is a Risk Factor for Surgical Site Infection

Objectives: To establish predictive factors for surgical site infection following elective abdominal surgery with special emphasis on the impact of adhesiolysis on surgical site infection.

Background: Despite the use of antibiotic prophylaxis, surgical site infection remains a common problem following elective abdominal surgery. Dissecting adhesions at repeat laparotomy is an important risk factor for a large number of per- and postoperative complications, however its impact on surgical site infection has never been investigated. Adhesiolysis might increase the risk for surgical site infection by increasing operative time and blood loss. Further, adhesiolysis is correlated with a high risk of seromuscular injury and enterotomies (with gross spillage) that might further increase the risk of surgical site infection.

Methods: Between June 2008 and June 2010 demographics, disease characteristics and perioperative data of all patients undergoing elective laparotomy or laparoscopy were included in a prospective cohort study. Detailed description of adhesiolysis was obtained by a trained researcher present during every operation. Predictive factors for surgical site infection were assessed through univariate and multivariate logistic regression.

Results: 755 abdominal operations were included during the study. Surgical site infections developed after 56 operations. 14 patients had superficial incisional infection, 24 had deep incisional infection and 21 had organ/space infection. In univariate analysis, adhesiolysis time (20 min or longer OR 3.02; 95% CI 1.74 – 5.254), seromuscular injury (OR 2.63; 95% CI 1.47 – 4.70), contaminated or dirty wound class (OR 4.86; 95% CI 2.11 – 11.22 and OR 53.1; 95% CI 5.22 – 539, respectively), blood loss (per 100 ml increase OR 1.02; 95% CI 1.01 - 1.04), ex-smoker or smoker (OR 2.17; 95% CI 1.06 – 4.43 and OR 2.84; 95% CI 1.28 – 6.30, respectively), age (per year increase OR 1.00; 95% CI 1.00 - 1.04), and male gender (OR 2.18; 95% CI 1.18 – 4.01) were significant predictive factors for surgical site infection. In multivariate analysis, wound class, adhesiolysis time, blood loss and male gender were independent risk factors for surgical site infection. The AUC of ROC from the multivariate model was 0.720 (0.644 – 0.795).

Conclusions: An adhesiolysis time longer than 20 minutes is correlated with a higher risk for surgical site infection. Traditional risk factors such as wound class, blood loss and male gender were also significant and independent risk factors for surgical site infection.

Olivas A Zaborina O Zaborin A Poroyko V Valuckaite V Musch M Alverdy J University of Chicago, USA

7. Opportunistic Pathogens Cause Anastomotic Leak in the Radiated Bowel VIA Switching to a Wound Disrupting Phenotype

Introduction: Colorectal anastomotic leak rates continue to be high, especially in patients who receive preoperative chemoradiation. The aim of this study was to define potential mechanisms by which pathogenic microbes, present in the intestinal tract of complex patients, affect anastomotic leakage. Specifically, we sought to determine whether the local microenvironment of an anastomosis could induce the virulence properties of intestinal Pseudomonas aeruginosa to attenuate wound healing and cause anastomotic leakage.

Methods: Male rats (350g) received fractionated pelvic radiation with a total dose of 25 Gy followed one week later by a low colon resection and anastomosis. At laparotomy, the cecum was directly injected with P. aeruginosa. Additional groups underwent anastomosis formation only, radiation and anastomosis formation without P. aeruginosa injection, and anastomosis formation with P. aeruginosa injection without radiation. Six days after surgery, rats were euthanized and assessed for anastomotic leaks. Localization of P. aeruginosa in the bowel was determined with DNA quantification using QRT-PCR. To determine if in vivo activation of P. aeruginosa influenced the development of anastomotic leak, P. aeruginosa was isolated from anastomotic sites and analyzed for swarming motility (epithelial-disrupting phenotype), lethality (C. elegans killing assay), and ability to impair healing of wounded rat intestinal epithelial cells (IECs). To determine if the local anastomotic microenvironment itself could transform the phenotype of P. aeruginosa, anastomotic tissue segments from non-contaminated rats were homogenized and co-incubated ex vivo with the P. aeruginosa parental strain, and recovered strains were analyzed for phenotype switching.

Results: Rats exposed to radiation followed by a colorectal anastomosis and P. aeruginosa bowel contamination had a significantly higher anastomotic leak rate compared to non-radiated animals (73% vs 6%, respectively; p < 0.01). Radiated and non-radiated rats without P. aeruginosa injection had no leaks, implicating a causal role for P. aeruginosa. P. aeruginosa was found to preferentially segregate to the anastomotic sites of all rats, and strains recovered from these sites were significantly enhanced in swarming motility (p < 0.01) and killing ability of C. elegans (p < 0.05) compared to the parental strain. Experiments performed on wounded IECs showed impaired healing in the presence of P. aeruginosa recovered from the anastomotic sites, suggesting that in vivo virulence activation of P. aeruginosa plays a direct role in epithelial healing. In ex vivo experiments, robust swarming motility was demonstrated in 95% of colonies recovered after incubation with radiated anastomoses, 33% from non-radiated anastomoses, and 0% after incubation with normal colon tissue.

Conclusion: The presence of intestinal pathogenic flora as a result of patient complexity may play an underappreciated role in anastomotic leaks. The specific microenvironment of an anastomosis after radiation exposure can transform P. aeruginosa to a tissue-destroying phenotype, which may enhance its ability to disrupt the healing of high risk anastomoses.

ten Broek R van Goor H Bleichrodt RP Radboud University Nijmegen Medical, The Netherlands

8. Morbidity of Adhesiolysis in Abdominal Wall Surgery

Objectives: To establish predictive factors for enterotomy made during adhesiolysis in abdominal wall repair and to establish the impact of enterotomies or longduring adhesiolysis on postoperative morbidity (such as sepsis, wound infections, abdominal complications and pneumonia) and socioeconomic parameters.

Background: The surgical repair of hernias is frequently complicated by wound and mesh infections, seroma formation and inadvertent enterotomy. Enterotomies made during adhesiolysis might specifically have a large impact on morbidity of patients, especially surgical site infections. Little is known on the incidence and burden of enterotomies and longduring adhesiolysis in ventral hernia repair.

Methods: Between June 2008 and June 2010 demographics, disease characteristics and perioperative data of all patients undergoing elective laparotomy or laparoscopy were included in a prospective cohort study. Detailed description of adhesions and enterotomies was obtained by a trained researcher present during every operation. A cohort of abdominal wall repairs was selected by operation codes from the database. Predictive factors for enterotomy were assessed through univariate and multivariate analyses.

Results: 33 enterotomies were made in 17 patients (12.7%). Two patients had a delayed diagnosed bowel perforation. Adhesiolysis time, hernia size >10cm and fistula were significant predictive factors in univariate analysis. In multivariate only adhesiolysis time was a significant and indepent predictive factor for enterotomy (P = 0.004). Trend towards an increased risk were seen for patients with mesh in situ and hernia size >10 cm. Patients with enterotomy had significant more urgent reoperations (P = 0.029) and more often required parenteral feeding (P = 0.037). Moreover, patient with adhesiolysis time >30 min. more often suffered from wound infection (9/63 vs. 2/70; P = 0.025), abdominal complications (5/63 vs. 0/70; P = 0.022) and sepsis (4/63 vs. 0/70; P = 0.048).

Conclusions: The incidence and morbidity of enterotomy at abdominal wall repair is high and associated with the length of adhesiolysis. Long adhesiolysis is associated with both local and systemic postoperative infectious complications.

Membrilla-Fernández E 3 Iñigo-Noain JJ 1 Fraccalvieri D 5 Adrio G 6 Girvent-Montllor M 3 Sancho-Insenser JJ 3 Monzón A 4 Luna-Aufroy A 2 Sitges-Serra A 3

1 Hospital de Navarra 2 Hospital Parc Taulí 3 Hospital Universitario del Mar 4 Hospital Royo de Vilanova 5 Hospital Universitario de Bellvitge 6 Hospital Arquitecto Marcide, Spain 9. Causes of Therapheutic Failure in Secondary Peritonitis in Patients with Adequate Empiric Antibiotic Treatment and Correct Focus Control

Introduction: The source control and adequate empiric therapy are essentials in the management of secondary peritonitis. Otherwise, not a low percentage of patients with source control and adequate empiric therapy had therapeutic failure. This study was made to analyze the causes of therapeutic failure in this kind of patients.

Materials and methods: An open, observational, prospective and multicenter study was done at 24 Spanish hospitals (04/2005-08/2007). There were included 362 patients with secondary peritonitis; 262 of them (72,4%) with community peritonitis (PC) and the other 100(27,6%) with postoperative peritonitis (PPO). There were only included cases with two quadrants abdominal affectation and positive peritoneal culture. The studied variables were: The index of gravity (ASA, APACHE II, Mannheim Peritonitis Index, Charlson, McCabe&Jackson), origin of peritonitis, kind of surgery, gender, specie and microbial sensitivity to antibiotics, septic complications and mortality. The dependent studied variable was therapeutic failure (SSI or exitus).

Results: This studied group had 204 patients (56%), 98 women and 106 men, with mean age of 60 years IC_95% = [57,8–63,1]. The postoperative peritonitis was related to more percentage of failure than community peritonitis (58% vs. 37%; P = 0,017). The origin of peritonitis wasn't associated with a great percentage of therapeutic failure. The group with therapeutic failure had more percentage of patients with ASA III/IV (45% vs. 28%; P = 0,0039), a greater prevalence of cardiac insufficiency (13% vs. 5%; P = 0,03) and a Mannheim Peritonitis Index greater than healed group (27 ± 7 vs. 22 ± 7; P = 0,001). It hadn't significative differences at the APACHE II index, neither at Charlson index. The group with therapeutic failure had more percentage of patients with preoperative peritonitis over 24 hours (77% vs. 63%; P = 0,043) and also with organic failure (53% vs. 15%; P = 0,001). The group with failure therapeutic had a higher cardiac frequency than the group with no complications (99 ± 19 vs. 93 ± 17; P = 0,019) and higher respiratory rate (22 ± 7 vs. 19 ± 7; P = 0,038) as well as a bigger value of PCR (18 ± 25 vs. 39 ± 28; P = 0,003) and severe hypoalbuminemia (2,68 ± 0,9 vs. 3,10 ± 0,8; P = 0,021). Otherwise, the group with therapeutic failure had more percentage of anaerobe bacteria (44% vs. 32%; P = 0,049) and Enterococcus spp. (24% vs. 12%; P = 0,018) at initial cultures.

Discussion: In patients with secondary peritonitis and with an adequate empiric antibiotic treatment the index that appeared as main responsibles of postoperative septic complications were cardiac insufficiency and hypoalbuminemia. This patients has an elevated Mannheim Peritonitis Index, especially, it depends of the existence of organ failure.

Sawyer R Popovsky K Hranjec T Rosenberger L Politano A Riccio L Turza Campbell K University of Virginia, Richmond, USA

10. Excessive Duration of Antibiotics in the Treatment of Intra-Abdominal Infections is Associated with Subsequent Extra-Abdominal Infections and Death: A Study of 2552 Consecutive Infections

Introduction and objectives: An excessively long duration of antibiotic use for intraabdominal infections (IAI) may be associated with an increased risk of antimicrobial resistance. We hypothesized that due to changes in normal host flora that an over long duration of antibiotic use is also associated with an increased risk of extraabdominal infections (EAI).

Material and methods: All IAI in a single institution occurring between 1997 and 2010 were reviewed. Demographic data on these infections and all subsequent EAI occurring in these patients were collected. Variables between patients with and without EAI were compared. Due to a difference in the severity of illness between these groups, a 1:2 matching between patients with and without EAI was performed, based on APACHE II score ± 1 point. Comparisons utilized Student's t-test, chi-square analysis, or Wilcoxon rank sum test.

Results: 2552 IAI were identified, of which 549 (21.5%) were complicated by subsequent EAI. IAI complicated by subsequent EAI were associated with longer initial duration of antimicrobial therapy for IAI (median 14 days [interquartile range 10–22] versus 10 [6–15] days for patients without subsequent EAI, p < 0.01), a higher APACHE II score (16.6 ± 0.3 vs. 11.2 ± 0.2, p < 0.01), and a higher in-hospital mortality (17.1% vs. 5.4%, p < 0.01). Patients initially treated with 0–7 days of antimicrobials for IAI had a rate of subsequent EAI of 13.3% versus 25.1% for those treated for >7 days (p < 0.01). The most common sites of EAI were blood (324), lung (287), and urine (287), and the most common pathogens isolated from EAI were Candida species (242), Enterococcus species (230), and P. aeruginosa (164). 469 IAI patients with subsequent EAI were successfully matched with 938 patients without subsequent EAI, resulting in a mean APACHE II score of 15.2 for each group. Even after matching, IAI complicated by subsequent EAI were associated with longer duration of initial antimicrobial therapy for IAI (median 14 [9–22] days versus 11 [7–16] days for patients without subsequent EAI, p < 0.01) and a higher in-hospital mortality (14.9% vs. 9.0%, p < 0.01).

Conclusions: Even after controlling for severity of illness, the overuse of antimicrobials in the treatment of intraabdominal infections is associated with an increased risk of subsequent extraabdominal infections, and these infections are caused by healthcare associated-pathogens. Further, the development of extraabdominal infection is strongly associated with death. Increased efforts to limit the duration of antimicrobial therapy for intraabdominal infections will likely lead to fewer complications and better outcomes.

Culebras-Fernández J 3 Belda-Albero A 4 Sanz-Valero J 1 Wanden-Berghe C 2

1 University of Alicante & University Miguel Hernández 2 University General Hospital & University CEU Cardenal Herrera 3 Institute of Biomedicine (IBIOMED) & University Hospital of Leon 4 University of Alicante, Alicante, Spain 11. Perioperative Prebiotics and Probiotic use to Prevent Postoperative Surgical Infection

Introduction: Cross Infection and wound complications after surgery increase mortality, morbidity, hospital stay and the health care costs.

Objective: To know with the highest level of evidence the benefits of pre and probiotics to prevent the surgical wound infection.

Material and methods: Systematic review of articles from peer-reviewed journals. The following databases were consulted: MEDLINE (via PubMed), EMBASE, ISI Web of Knowledge, The Cochrane Library Plus, CINAHL, International Pharmaceutical Abstracts, Latin American and Caribbean Health Sciences Literature (LILACS) and Spanish Health Sciences Bibliographic Index (IBECS). In addition, a secondary search was carried out in order to reduce potential publication bias. The Descriptors (MeSH) selected were: “Probiotics”[Mesh]; “Prebiotics”[Mesh] and “Surgical Wound Infection”"[MeSH]. The search was limited to human's adults. The resulting search equation was defined with the Boolean connectors. The final equation was adapted to each of the bibliographical databases consulted. Additionally, as a secondary search, the bibliographies given in the selected articles were reviewed in order to identify studies not found by the primary search. Selection criteria: We included only randomized clinical trials (RCT) irrespective of language or publication status.

Results: A total of 13 RCT were found. 7 (53.853%) were repeated articles and 1 (7.69%) was not a RCT, leaving a total of 4 (30,77%) studies suitable for this research; (the table of the review will be displayed in the poster). In four trials, the incidence of infectious complications in the prebiotics or probiotics group was significantly lower than in controls. In another, there were no measurable effects in elective surgical patients (bacterial translocation 12.1% vs. 10.7%, p = 0.808; gastric colonization 41% vs. 44%, p = 0.719; septic complications 32% vs. 31%, p = 0.882). In addition, the duration of antibiotic therapy was significantly shorter in the probiotic-group.

Conclusions: The use of prebiotics and probiotics offers promise, costs can be reduced by shortening the duration of antibiotic therapy. There is no clear evidence that any of the interventions may benefit the patients. Further randomized clinical trials are necessary.

De Waele J Imaeva K De Vos M Ghent University Hospital, Belgium

12. Diagnosis of Pancreatic Infection in Acute Pancreatitis: Systematic Review and Meta-Analysis

Introduction and objective: Pancreatic infection (PI), most frequently infected pancreatic necrosis (IPN), is a major determinant of outcome in acute pancreatitis (AP). Timely diagnosis of PI determines therapeutic choices and improves patient outcomes. We performed a systematic review and meta-analysis to determine the accuracy of clinical, laboratory and radiological markers of PI in patients with severe AP.

Materials and methods: We identified studies examining or reporting test performance within Medline and Web of Science databases. Pooled estimates of diagnostic accuracy were calculated using random effects model. Diagnostic odds ratio (DOR) and area under the summary ROC curve (AUC) represented global test performance. We evaluated the study quality and heterogeneity and conducted sensitivity analyses.

Results: Twenty-seven of 228 reports identified were included (N = 1159 patients with AP, n = 581(50%) with PI). The respective sensitivity, specificity and DOR for markers evaluated in 2 or more studies were as follows: computed tomography (CT) or ultrasound (US) guided fine-needle aspiration (FNA): 91%, 92%, and 88.6, retroperitoneal gas on CT: 48%, 99%, and 27, fever: 95%, 44%, and 8.39, C-reactive protein (CRP): 87%, 89%, and 36.2, and procalcitonin: 77%, 89%, and 24. Between-study heterogeneity was not significant. Combined clinical parameters were very specific, but individual clinical parameters, and white blood cell count performed only poor.

Conclusions: FNA is most accurate to diagnose PI. CRP and procalcitonin are helpful markers which need investigation in larger diagnostic studies. Presence of extraintestinal gas on CT is highly suggestive of PI and should limit further invasive testing.

Függer R 1 Gangl O Fröschl U Dutta-Függer B Sautner T Krankenhaus der Elisabethinen, Linz, Austria

13. Infection as a Significant Cause of Unplanned Reoperation following Pancreatic Resection

Background: The incidence of unplanned reoperation and reintervention is an accepted benchmark to assess surgical quality. The aim of this study was to determine to which proportion infectious complications following pancreatic resection contribute to unplanned reoperation and reintervention.

Methods: 251 pancreatic resections, performed between 2001 – 2010, were prospectively documented in a pancreatic surgery data base. Data were searched for unplanned reoperations and percutaneous or angiographic reinterventions. The underlying complications were classified as infectious or other origin, and groups compared regarding their impact on reoperation and reintervention rates.

Results: 30-day mortality was 2.4% (6/251). Unplanned reoperation and reintervention rates were 6.4% (16/251) and 9.2% (23/251). Reinterventions performed were 11 CT-guided and 6 ultrasound guided drainages and 6 angiographic interventions (embolization, stenting). Three (3/6, 50%) deaths, 10 (10/16, 62.5%) unplanned reoperations, 10 (10/11, 90.9%) percutaneous drainages and three (3/6, 50%) angiographies were caused by infectious complications. In a logistic regression model, haemorrhage associated with pancreatic fistula was the predominant cause of unplanned reoperation (RR 116.667, p < 0.0001). Pancreatic fistula alone (RR 41.795, p < 0.001) and fistula associated with haemorrhage (RR 245.998, p < 0.001) were independent risk factors for unplanned reintervention.

Conclusion: Infectious complications are significant risk factors for unplanned reoperation and reintervention. 50% of postoperative mortality and the majority of unplanned revisions and interventions was related to infection.

Nve Obiang E CasallRosell M Juvany M Amador S Guirao X Tàrrech JM Badia JM Hospital General de Granollers, Catalonia, Spain

14. A Survey of the Timing and Approach to the Surgical Management of Acute Cholecystitis Among Spanish Surgeons

Introduction: Despite strong evidence advocating early laparoscopic cholecystectomy for acute cholecystitis, this approach is not widely used. The type of empirical antibiotic and the duration of antibiotic therapy are also controversial. This study was promoted by the Infection Section and the Hepatobiliary Section of the Spanish Association of Surgeons (AEC) to assess the current management of acute calculous cholecystitis (ACC) by Spanish surgeons.

Methods: An e-mail was sent to all the members of the AEC, containing a link to an internet questionnaire in order to ascertain their current preferred management of patients with acute cholecystitis and factors limiting their preferred practice. They were asked to provide details related to subspecialist interests, timing of surgery, open or laparoscopic approach, empirical antibiotic therapy and duration of antibiotic treatment.

Results: 752 surgeons answered the questionnaire. 93% of the responding surgeons routinely treat patients with ACC. 63.2% of the surgeons perform fewer than 50 cholecystectomies per year, while 36.8% perform more than 50. Routine intraoperative cholangiography is performed only in 3.6% of cases. A 42% of surgeons perform it selectively, while 54.4% never perform it in ACC. The preferred timing for surgery in ACC is cholecystectomy during the first 24 h of admission for 58.3% of responding surgeons, early operation (1–4 days of admission) for 34.3% and delayed elective cholecystectomy for 7.4%. However, in their real practice immediate, early and delayed cholecystectomy was adopted by 42.2%, 29. 9% and 27.8% of surgeons, respectively. Reasons for delayed cholecystectomy are comorbidity, ASA score III-IV, lack of emergency operating room, and a different policy in their surgical unit. The majority of members of the Spanish Association of Surgeons adopt a laparoscopic approach for ACC (89.3%). Most used empirical antibiotics were piperacillin-tazobactam (49.8%) and amoxicillin-clavulanate (47.9%).

Ertapenem was used in 24.3% of instances and different cephalosporins in 17.6%. When he patient is operated in the first admission, the duration of antibiotic therapy is one postoperative day in 22.1% of cases and less than 3 days in 44.4% of patients. Half of surgeons used a prefixed duration of antibiotic treatment while the other half discontinued the treatment when there was an improvement in the patient condition. When a policy of delayed cholecystectomy is chosen, antibiotics are continued at first admission during 7 or more days in 60.6% of cases.

Conclusions: A policy of early cholecystectomy (during the first 4 days of admission) is preferred for ACC by 92.6% of the responding surgeons, but this policy can be actually followed in only 72.1% of cases, mostly due to organizative problems. Laparoscopic surgery is the surgical technique of choice for 89.3% of surgeons. Beta-lactamic antibiotics are most used for empirical treatment, and a short course (1–3 days) of postoperative antibiotic therapy is preferred in ACC (66.5% of cases).

Cuevas MJ 1 Tiempo J 3 Culebras-Fernández JM 1 2 Possa-Marroni N 4 González-Gallego J 1 Tuñón –González MJ 1

1 Institute of Biomedicine (IBIOMED), University of Leon 2 Hospital of León 3 Porto Alegre Clinical Hospital 4 Porto Alegre Clinical Hospital and Federal University of Rio Grande do Sul, Spain and Brazil 15. Quercetin Suppresses Amphiregulin/EGFR Signals Involved in the Progression to Liver Cancer in Cirrhotic Rats

Introduction: Elevated expression of mitogenic factors for the hepatocytes is one of the major molecular mechanisms that drive the hyperproliferative condition in chronically injured liver, and its perpetuation is thought to participate in the multistep process of hepatocarcinogenesis. In this study we investigated the potential of quercetin to delay or prevent tumor progression in cirrhotic rats by assessing its effects on the amphiregulin/epidermal growth factor receptor (EGRF) signal and on the activation of downstream pathways leading to cell growth.

Methods: Rats were divided into four groups (8 rats/group): rats subjected to common bile duct ligation (CBDL), Sham (rats subjected to simulated CBDL), quercetin-treated sham (Sham-Q), and quercetin-treated CBDL (CBDL-Q). Quercetin (50 mg/kg i.p. injection) was administered daily for 2 wk starting on d 14 after surgery. Over-expression of amphiregulin, EGFR, tumor necrosis factor (TNF)alpha, interleukin (IL)-6, transforming growth factor (TGF)beta, platelet derived growth factor (PDGF), extracellular regulated kinase (ERK), Akt, cycloxygenase (COX)-2, and glioma-associated oncogenes (GLI)-1 and-2 were observed in liver of cirrhotic rats after 4 wk of bile duct ligation.

Results: Cirrhotic rats treated with quercetin showed a significant diminished expression of amphiregulin and EGFR when compared to untreated. CBDL rats. Furthermore, mRNA levels of TNFalpha, IL-6, TGFbeta and PDGF, and protein content of COX-2, GLI-1 and GLI-2 were significantly lower in CBDL rats treated with quercetin than untreated cirrhotic rats.

Conclusion: Findings indicate that quercetin ameliorated the activation of survival pathways and down-regulated the expression of genes related to precancerous conditions. Suppression of amphiregulin/EGFR signals may contribute to this effect.

Supported by grants from CAPES/FIPE-HCPA. CIBERehd is funded by the Instituto de Salud Carlos III.

Troyano D Balibrea-del-Castillo JM Molinos-Abos S Troya J Ausina V Oller-Sales B Fernández-Llamazares J Hospital Germans Trias i Pujol, Catalonia, Spainl

16. Bactibilia in Elective Cholecystectomy: Study of Microbial Populations, Multidrug Resistance and Surgical Outcomes

Background: Frequently, bactibilia is a factor associated with surgical complications, mainly surgical site infection (SSI). Commonly isolated specimens include E.coli y Enterococcus sp. But important changes in microbial populations and increasing rates of multidrug resistenace have also been reported. Thus, bactibilia in an elective cholecystectomy context remains an interesting topic because its presence might change some therapeutic aptitudes. Our aim was to study the presence of bacteria in bile and tissue samples from elective cholecystectomies, identify possible risk factors for its presence and to determine whether the its presence was associated with a higher rates of surgical complications (specially SSI) or not.

Patients and methods: We present a prospective, observational and multidisciplinary study performed between june 2007 and june 2010. Bactibilia was determinated by the presence of bacterial growing in culture of both bile and gallbladder mucosa. All samples were obtained during a elective cholecystectomy, excluding those with inflammatory signs. Data was collected prospectively including previous patient history, microbial and surgical outcomes. Antibiogram was performed on all samples and post-hoc anaylisis between clinical variables and appearance of bactibilia was done by univariate analysis. P values under 0,05 were considered as statistically significant.

Results: Samples from 368 patients were collected (10 were excluded for not fulfilling the inclusion criteria) and 103 positive cultures were obtained. The most frequent species isolated were E. coli (29 cases, 24,2% amoxicillin-clavulanic acid resistant), Enterococcus spp(20 cases, 60% amoxicillin-clavulanic acid resistant) and Enterobacter spp. (20 cases, 100% amoxicillin-clavulanic acid resistant). After univariant analysis, bactibilia predisposing factors were male, >65 years old, admision from emergency department or from another center, diabetes mellitus, preoperative administration of antibiotics, history of previous biliar disease, history of previous interventionism in the bile duct, high ASA risk (III, IV), preoperative pain and preoperative fever. No association was found between the presence of bactibilia and a higher rate of surgical site infection in study sample.

Conclusions: When compared to previous studies, our results show a greater diversification of bactibilia species and higher antibiotic resistance. The overall rate of resistance to amoxicillin-clavulanic acid in our sample is 53.7%. Conversely, presence of bactibilia was not correlated with a higher rate of surgical complications such as SSI. Our microbiological results suggest that administration of amoxicillin-clavulanic acid could be ineffective in preventing SSI in elective cholecystectomies. On the other hand, the weak correlation between bactibilia and SSI does not support systematic administration profilactic antibiotics.

Nve Obiang E Casal M Juvany M Amillo M Ciscar A Guirao X Tàrrech JM Badia JM Hospital General de Granollers, Catalonia, Spain

17. Timing of Surgery in Acute Cholecystitis. Comparison of Two Surveys of the Spanish Association of Surgeons (2003 and 2010)

Introduction: There is strong evidence either in the open surgery or in the laparoscopic era favoring early cholecystectomy in patients with acute cholecystitis. However, this approach is not universally adopted. This study assesses the evolution in the management of acute calculous cholecystitis (ACC) by Spanish surgeons, comparing two different audits in 2003 and 2010 promoted by the Spanish Association of Surgeons (AEC).

Methods: In 2003, a postal survey was sent to each Head of Department of Surgery of Spanish hospitals, addressing the timing in the surgical management of ACC. In 2010, an internet questionnaire was addressed to the members of the AEC in order to ascertain their current preferred management of patients with acute cholecystitis and factors limiting their preferred practice. Results of both surveys are compared.

Results: In 2003, 89 surgeons responded to the postal survey. In 18.8% of Departments of Surgery patients with ACC were operated during the first 24 h of admission, 51.2% during the first 3 days, while in 30% of them a medical treatment and a delayed cholecystectomy in a second admission was preferred. In 2010, 752 surgeons answered the questionnaire. Cholecystectomy within the first 24 h of admission was the preferred timing for 58.3% of responding surgeons, early operation (1–4 days of admission) for 34.3% and delayed elective cholecystectomy for 7.4%. However, in actual practice immediate, early and delayed cholecystectomy was adopted by 42.2%, 29.9% and 27.8% of surgeons, respectively. Reasons for delayed cholecystectomy were comorbidity, ASA score III-IV, lack of emergency operating room, and a different policy in their surgical unit.

Conclusions: A marked improvement in the surgical management of ACC has been observed during the seven years elapsed from the first to the second survey. Currently, a policy of early cholecystectomy (during the first 4 days of admission) is advocated for ACC by 92.6% of the responding surgeons. However, this policy can be actually followed in only 72.1% of cases, mostly due to lack of availability of experienced surgeons, radiological investigations and the limited availability of theatre space. Organizative changes must be implemented in Spanish hospitals to assure the availability of emergency theatre time and ultrasonography investigations.

Kiewiet J 2 Andeweg C 1 Daniels L 2 Hendriks J 1 Reitsma J 2 van Goor H 1 Boermeester M 2

1 UMC St Radboud 2 Academic Medical Center, The Netherlands 18. External Validation of Two Tools for the Diagnosis of Acute Diverticulitis Without Imaging

Objective: The objective of this study was to compare the diagnostic accuracy of two tools aiding the clinical diagnosis ‘acute diverticulitis' and perform an external validation.

Methods: The derivation datasets for a scoring system of seven variables derived from a cohort of patients admitted for suspected diverticulitis (‘admitted cohort/tool’) and a decision triad derived from patients with acute abdominal pain in the emergency department (‘ED cohort/triad') were used crosswise to externally validate each other. Multivariable regression analysis in the validation cohorts reassessed the predictive values of the variables included in the two tools. A cut off analysis was performed where the positive predictive value (PPV) of the ‘admitted tool’ was set at a clinically relevant 90% in its derivation cohort.

Results: Patients in the admitted cohort had more complicated diverticulitis compared to the ED cohort (Hinchey stage ≥2: 33% versus 8%, p < 0.001). The PPV of the ‘ED triad’ dropped from 97% in the ED cohort to 82% in the admitted cohort while the triad occurred in 24% and 20% of the cohorts, respectively. When the PPV of the ‘admitted tool’ was set at 90%, scores above this cut-off occurred in only 6% of patients in the admitted cohort compared to 19% in the ED cohort with slight improvement of the PPV to 92%.

Conclusion: The more complicated diverticulitis rate needing admission normally lies below 10% of diverticulitis patients. Therefore, the ED cohort resembles better the unselected patient population in which a clinical decision tool might be of use. Since the ‘ED triad’ performed best in the ED cohort, it can be used to aid the clinical diagnosis in unselected patients with suspected diverticulitis and possibly omit additional imaging in about a quarter of them.

Noguera J Cuadrado A García J Olea J Morales R Vicens J Hospital Son Llàtzer, Palma de Mallorca, Spain

19. Surgical Site Infection in N.O.T.E.S. Surgery

Introduction: NOTES surgery in a new modality of minimally invasive surgery. Transvaginal approach is the most used one to perform intraabdominal human procedures. We must know if this new approach is better of laparoscopic surgery in some questions, like the surgical site infection.

Material and methods: Prospective non-randomized clinical series of 50 patients with elective cholecystectomy for gallstones. 25 patients with transvaginal hybrid NOTES surgery and 25 with a conventional laparoscopic approach. Surgical infection in both groups was evaluated: general infections, surgical site infections and intraabdominal infections.

Results: We found an urinary infection in one case with transvaginal approach (the eigth case: we performed rutinary urethral catheterization in the first ten cases) and one umbilical site infection in laparoscopic group.

No intraabdominal infections were found in the study.

Conclusions: Transvaginal hybrid NOTES for cholecystectomy is a safe approach in relation to the surgical infections. According to the minor abdominal trauma in transvaginal NOTES we can expect less abdominal complications (infections, ventral hernia). The elective urethral catheterization must be avoided and only used if urinary bladder difficult the transvaginal approach. Transvaginal NOTES approach is as safe as laparoscopy in terms of surgical infection.

Hviid C Samulin-Erdem J Kunke D Wang YY Ahmed Shakil Attramadal H Aasen AO Institute for Surgical Research, Oslo University Hospital, University of Oslo, Norway

20. Tissue Specific Regulation of Matricellular CCN Proteins in Experimentally Induced Sepsis

Introduction: A major predictor of sepsis-related death is the presence of consecutive organ failure, and the clinical outcome is directly correlated to the number of organs failing. Much is known about the immune-pathophysiological mechanisms in sepsis, but the molecular mechanisms behind sepsis-induced multiple organ failure (MOF) are a matter of debate. To address this critical aspect of sepsis, a novel project using a rodent cecal ligation and puncture (CLP) model to investigate organ specific molecular mechanisms of MOF was established. The CCN (Cyr61, CTGF, NOV)-proteins has been identified as players in diseases such as cardiac and renal failure, and acute pulmonary damage. These proteins (denoted CCN1-CCN6) belong to the matricellular, dynamically expressed factors in the extra cellular matrix (ECM) which serve no obvious structural role. Rather they modulate cellular function in response to environmental stimulation and are involved in regulation of parenchyma function. A protective role for the proteins in isolated organ failure has been described. Moreover, the CCN expression is highly sensitive to environmental perturbations like; release of cytokines and growth factors, oxidative stress, and hypoxia. This indicates a potential role for the CCN proteins in the development of sepsis-induced MOF. The present study is the first to provide evidence for such a role of the CCN proteins.

Objectives: The present study aimed to investigate the regulation of CCN1-CCN3 during the development of sepsis-induced MOF using a short-term rat model.

Materials and methods: A rat CLP model (18 hours) was established. The animals were randomized to Sham or CLP operation. Blood samples were collected and analyzed for markers of organ function and inflammation. The lungs, liver, and heart were harvested and analyzed for CCN mRNA and protein expression, using real-time PCR (qPCR) and western-blot. Furthermore, cytokine expressions in the tissues were determined by qPCR.

Results and conclusions: The CLP animals displayed reduced platelet and leukocyte count as well as tissue induction of pro-inflammatory cytokines. Serum markers of hepatic damage were significantly elevated, while markers of cardiac damage, and arterial blood gasses were unaffected. Q-PCR analysis revealed tissue specific CCN gene expression in the CLP animals. Hepatic and pulmonary CCN1 and CCN3 mRNA levels were up-regulated, whereas CCN2 expression was induced in liver but vice-versa, suppressed in the lungs. Interestingly, cardiac CCN2 mRNA levels were significantly down-regulated but due to small differences in expression levels, it was difficult to confirm protein regulation. Taken together, the study provides novel evidence for gene regulation of several matricellular CCN proteins in sepsis-induced MOF. Moreover, tissue specific profiles of the expressions were found. This indicates involvement of CCN1-CCN3 already in the early stages of sepsis-induced organ dysfunctions.

Piva M Carbajo-Pescador S Martin-Renedo J 1 Ordoñez R 1 Culebras JM 2 Mauriz JL 1 González-Gallego J 1

1 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd) Instituto de Biomedicina, University of León 2 Complejo Asistencial Universitario de León, Spain 21. Antiproliferative Effects of PTK 787 on Human Hepatocarcinoma Cells

Introduction and Objetives: Hepatocellular carcinoma (HCC) is responsible for over 600,000 deaths every year, being the fifth most common cancer among men and eighth among women. PTK787 is a very promising antiangiogenic anticancer therapy, especially for HCC, a highly vascularized tumor characterized by a reduced sensitivity to conventional treatments, with frequent occurrence of chemoresistance. PTK787 has been shown to be a potent inhibitor of VEGF receptor tyrosine kinases, inhibiting both VEGFR1 and VEGFR2. In addition, PTK787 seems to be able to act directly as a regulator of apoptosis.

The aim of this study was to determine the PTK 787's effect over the proliferation of the human's HCC cell line HuH7; and to verify whether their antiproliferative effects are related to the Mitogen-activated protein kinases pathway – MAPK/ERK and the cell cycle progression.

Material and Methods: Huh7 cell line was treated with PTK787 at a concentration of 40 μM for 12 or 24 hours and untreated cells were used as control. The cell viability was determined by MTT and the protein expression of p-ERK, ERK, CDK 4 and Cyclin D1 were measured by Western Blot.

Results: PTK787 was capable to induce a reduction in Huh7 cell line proliferation with the administered dose of 40 μM in a time dependent manner, with a maximum at 24 hours of treatment. Statistically significant decrease in the expression of the phosphorylated form of ERK was observed, while there was no change in its unphosphorylated form. Cyclin D1 and CDK 4 expressions were also clearly diminished after treatment with PTK 787.

Conclusion: VEGF receptors inhibition by PTK 787 causes a decrease in ERK phosphorylation impairing cell proliferation in a time-dependent manner. In addition to interrupting the MAPK/ERK pathway, the treatment induces cell cycle arrest in G1 due to the decreased expression of Cyclin D1 and CDK 4, which contributes to the proliferation's decline observed in the MTT. Although further studies are needed, our experiments indicate that the drug PTK787 seems to be very promising in the treatment of HCC due to its angiogenic effects associated with its ability to prevent cell proliferation.

Chang TP Malbon S Javaid A Lisy P Tameside General Hospital NHS Foundation Trust, United Kingdom

22. Pre-Operative CT Scan Measurement of Thickness of Subcutaneous Fat is Predictive Marker of Surgical Site Infection in Elective Colorectal Cancer Resection

Objectives: The aim of this study is to identify predictive markers of surgical site infection (SSI) in patients undergoing elective colorectal cancer resection.

Methods: We conducted a case-controlled study on 178 patients who underwent elective colorectal cancer resection between January 2008 and November 2010. We included patient's blood parameters (haemoglobin, albumin, C-reactive protein), presence of cardiovascular co-morbidities (ischaemic heart disease, renal dysfunction, diabetes mellitus and obesity) and intraoperative factors (antibiotic prophylaxis) in this study. The infraumbilical thickness of subcutaneous fat of the anterior abdominal wall was measured using pre-operative staging CT scan of the abdomen and pelvis. The end-point in this study is the presence of SSI.

Results: The incidence of SSI in this study was 33 (19%) and this was significantly higher in patients with increased subcutaneous fat thickness (median thickness 34 mm vs 22 mm; p = 0.003). In particular, a significantly higher percentage of patients with thickness of more than 30 mm had SSI compared to those less than 30 mm (48% vs 17%, P = 0.002). Although obesity was associated with a higher incidence of SSI, it has lost its significance on multivariate analysis that included CT measurement of subcutaneous fat thickness. None of the remaining blood parameters, co-morbidities or intraoperative factors included was associated with SSI.

Conclusion: The results of this study suggest that the risk of SSI is higher in patients with extraperitoneal central obesity. The use of pre-operative CT scan in the staging work-up of colorectal cancer patients could also help to identify patients who are higher risk of developing SSI.

Gianotti L 2 Nespoli L 2 Panelli M 1 Nespoli A

1 San Gerardo Hospital 2 Milano-Bicocca University, Italy 23. Early Oral Feeding after Colorectal Resection is Safe and Feasible also Outside a Fast Track Programme

Introduction and objective: Postoperative ileus after colorectal surgery remains a relevant clinical problem. In the last years, a multimodal enhanced recovery programme or “fast track surgery” (FTS) has been hypothesized as an efficacious way to preserve gut function. Nevertheless, this programme is not yet largely applied and it finds difficulties to be implemented worldwide. Part of the reason may be due to the profound, radical and numerous changes that surgeons and anaesthetists should perform altogether to apply entirely the suggested programme. Recently, it has been reported that similar beneficial results on surgical outcome can be obtained also if only a part of this program is carried out. This was study was designed to verify safety, feasibility and tolerance of early oral postoperative feeding (EOF) outside an FTS program.

Materials and methods: One hundred patients candidate to elective colorectal resection were prospectively enrolled in a EOF program. Feeding was started on postoperative day (POD) 1 with oral nutritional supplement (OSN). On POD 2, patients had normal food plus ONS to reach 1000–1200 Kcal/day with progressive increase until 1800–2000 kcal/day to be reached within POD 5. Results were compared with historical controls (n = 100) in whom oral feeding was allowed only after full recovery of bowel function (median 5 days after operation). FTS program was not applied in both groups except for epidural analgesia used in 58% in the EOF group and in 31% in the control group.

Results: The two groups were well-matched for baseline and surgical characteristics. The EOF group had a better recovery of short half-life protein synthesis compared to the control group (P < 0.001). Stool canalization occurred after a median of 3 days (range 1–6) in the EOF group versus 5 days (range 2–8) in the control group (P = 0.001). The EOF protocol was completed in 89 patients within POD 5. Tolerance to resumption of feeding was similar in the two groups. Postoperative nausea/vomiting was observed in 19 patients in the EOF group and in 14 patients in controls (P = 0.63). Naso-gastric tube had to be replaced in 4 cases in the EOF group and in 1 case in controls (P = 0.37). The overall rate of post-operative complications was 22% in the EOF group vs. 27% in the control group (P = 0.51). The incidence of infectious complications was 15% vs. 21% respectively (P = 0.36). The median length of hospitalization was 9 days (range 6–25) in the EOF group vs. 12 days (range 6–31) in controls (P = 0.01).

Conclusions: The present data suggest that EOF after surgery is safe and feasible also outside a FTS program. This might convince surgeons devoted to a more traditional and conservative surgical care that limited changes in daily practice are possible without jeopardising patient outcome

Nagai L Yaguchi A Suzuki H Takeda M Harada T Moroi R Namiki M Tokyo Women's Medical University, Japan

24. The Utility of EAA (Endotoxin Activity Assay) for Diagnosis of Sepsis Due to Gram Negative Infection

Introduction: Endotoxin Activity Assay (EAA^TM, SPECTRAL DIAGNOSTIC Inc., Toronto, Canada) is a useful test to suspect of severe sepsis and rule-out of Gram negative infection. However, the evaluation of intermediate levels of EAA (0.4–0.59) has not been elucidated. The purpose of this study is to clarify the interpretation of the intermediate level of EAA in clinical practice.

Methods: Adult patients, who admitted to the intensive care unit (ICU) in our university hospital and were suspected developing sepsis in their clinical course, and whose EAA showed intermediate levels were enrolled in this study. The measurements of EAA, WBC counts and differentiations, CRP (C-reactive protein), PCT (procalcitonin) and blood culture were performed with same whole blood samples. Other bacteriological cultures were also performed. Patients were divided into three groups, 1) With Gram negative bacteria in blood culture, 2) With Gram negative bacteria in other cultures, and 3) No Gram negative bacteria in any cultures by bacteriological test. Data were analyzed by Kruskal-Wallis test, Mann-Whitney U-test, a χ² test, Fisher's exact probability test. P < 0.05 was considered to be statistically significant.

Results: Of 33 patients who showed the intermediate levels of EAA, there were 1) 6 patients with Gram negative bacteria in blood, 2) 11 patients with Gram negative bacteria in others, and 3) 16 patients with no Gram negative bacteria. There were no significant differences in age, gender, WBC counts (10135 ± 9273 vs. 12141 ± 8706 vs. 11953 ± 7637/mm³, respectively, p = 0.79), the proportion of neutrophlies (67 vs. 73 vs. 88%, respectively, p = 0.47), the appearance of Döle body (33 vs. 18 vs. 13%, respectively, p = 0.53) or CRP values (18.5 ± 6.2 vs. 11.3 ± 5.4 vs. 15.5 ± 14.0 mg/dL, respectively, p = 0.33) in three groups. However, procalcitonin values were statistically significantly higher in group 1) patients (30.5 ± 53.3 ng/mL, p = 0.02) and in group 2) patients (9.4 ± 11.4 ng/mL, p = 0.08) than in group 3) patients (4.4 ± 8.3 ng/mL).

Conclusions: EAA is a rapid and simple test to evaluate for endotoxemia in clinical practice at the bed side. The present study shows PCT values could be the most useful additional data with the intermediate levels of EAA to diagnose sepsis due to Gram negative infection.

De Waele J 3 Tellado JM 1 Alder J 2 Reimnitz P 5 Jensen M 4 Hampel B 4 Arvis P 4

1 H.G.U. Gregorio Marañon 2 Bayer HealthCare 3 Ghent University Hospital 4 Bayer Schering Pharma 5 Bayer Schering Pharma AG, Spain and Belgium 25. Efficacy and Safety of Moxifloxacin Vs Ertapenem in Peritonitis

Introduction: Peritonitis is an important cause of morbidity and may be associated with a poor prognosis. Alongside source control, antibiotic therapy has an important role in managing patients with peritonitis. Moxifloxacin (MXF) has proven clinical efficacy in complicated intra-abdominal infections (cIAIs), and activity against the vast majority of causative organisms. The PROMISE study was carried out to compare the efficacy and safety of MXF and ertapenem (ERTA) in the treatment of patients with cIAIs. The current analysis focuses on the subgroup of patients with peritonitis.

Methods: PROMISE was a prospective, randomised, double-dummy, double-blind, multinational trial in patients with cIAIs. Patients were treated for 5–14 days with MXF, 400 mg IV qd, or ERTA, 1g IV qd. The primary efficacy variable was clinical response 21–28 days after the end of therapy, with bacteriological response being a secondary efficacy variable. Descriptive efficacy analyses were carried out for patients with peritonitis.

Results: A total of 642 patients enrolled in the PROMISE study had a primary diagnosis of peritonitis, and 562 were valid for the per protocol analyses. Of these, 67.3% were men; the mean age (SD) was 45.8 (17.7) years. Most frequent indications for surgery were perforations from complicated appendicitis (48.4%), stomach/duodenum (22.1%), colon (13.3%), gall bladder (7.8%) and small bowel (7.3%). The mean (SD) Mannheim Peritonitis Index score was 19.8 (7.1) (MXF) and 19.6 (7.4) (ERTA). Localised peritonitis occurred in 103 MXF- and 97 ERTA-treated patients, while diffuse peritonitis was seen in 177 MXF- and 184 ERTA-treated patients. Clinical efficacy rates were high and similar in MXF and ERTA as shown in all analysis populations (Table) Good bacteriological efficacy was also seen in microbiologically valid patients (bacteriological success rates: MXF 203/232, 87.5%, ERTA 196/215, 91.2%; 95% CI: –9.5, 2.0; P = 0.20). Bacteriological success rates against the key pathogens Escherichia coli and Bacteroides fragilis were 88.7% (189/213) and 80.3% (57/71) for MXF, and 90.5% (172/190) and 93.4% (71/76) for ERTA, respectively. Similar numbers of patients in both treatment arms experienced treatment-emergent drug-related adverse events (intent-to-treat/safety population: MXF 64/327, 19.6%; ERTA 63/315, 20.0%; P = 0.991). Table: clinical efficacy in peritoniotis patients.

OPEN IN VIEWER

Table:

clinical efficacy overall and by extent of peritonitis

	MXF	ERTA	P-value (95% CI)
Overall clinical cure n/N (%)
PP	255/281 (90.7)	264/281 (94.0)	0.174 ( −7.4, 1.3)
ITT	270/327 (82.6)	277/315 (87.9)	0.056 (−10.7, 0.1)
MBV	208/232 (89.7)	199/215 (92.6)	0.286 (−8.1, 2.4)
ITT with organisms	220/268 (82.1)	207/239 (86.6)	0.137 (−10.9, 1.5)
Clinical cure by extent of peritonitis n/N (%)
LocalisedPPMBV	93/100 (93.0)77/84 (91.7)	90/96 (93.8)71/76 (93.4)	0.721 (−8.4, 5.8)0.414 (−11.8, 4.9)
DiffusePPMBV	162/181 (89.5)131/148 (88.5)	174/185 (94.1)128/139 (92.1)	0.192 (−9.2, 1.9)0.431 (−9.4, 3.9)

Conclusions: MXF monotherapy had similar efficacy to ERTA and was well-tolerated in patients with peritonitis. This included good efficacy in patients with diffuse peritonitis.

Lundeland B 1 Østerholt H 2 Gundersen Y 1 Opstad PK 1 Thrane I 1 Vaagenes P 1 1 Norwegian Defence Research Establishment 2 University of Oslo. Norway 26. Toll-Like Receptor 4 Expression on CD14 + Monocytes is Inversely Related to Temperature

Introduction: The body temperature is normally regulated within a limited range, but disturbances are frequently seen in clinical practice. Hypothermia is a common complication to surgery and multiple trauma, but can also be exploited therapeutically, e.g. after cardiac arrest. Fever is often seen in connection with infections and inflammation, and even after intensive physical activity core body temperature can increase. An adequate response by the innate immune system is important for resistance to infections and repair of tissue damage. Toll-like receptors (TLRs) play a crucial role in recognizing pathogen associated molecular patterns, e.g. lipopolysaccharide (LPS) from Gram-negative bacteria. This molecule activates TLR4, which in turn initiates a cascade of enzymatic reactions that eventually trigger the production of cytokines, i.a. TNF-α. In the present study we have investigated how temperature affects TLR4 expression on monocytes, a main producer of pro-inflammatory cytokines, and looked for a possible correlation with LPS-stimulated TNF-α production.

Methods: Blood from nine male volunteers were incubated with LPS 10 ng/ml for 6h in an ex vivo whole blood model at 33, 37 and 40 ΰC. After incubation the surface expression of TLR4 on CD14 + monocytes was measured by flow cytometry, and the concentration of TNF-α was measured in the supernatant.

Results: The monocyte expression of TLR4 declined with increasing temperature (Fig.). There was no difference in TNF-α concentration between incubation temperatures 33 and 37 ΰC, whereas incubation at 40 ΰC induced a significantly higher value. No correlation was found between monocyte TLR4 expression and plasma TNF-α concentration.

OPEN IN VIEWER

Conclusions: Temperature and monocyte TLR4 expression are inversely related. Hyperthermia induced significantly higher LPS-induced ex vivo production of TNF-α compared with incubation at 37 ΰC, in contrast to hypothermia that caused no change. The findings suggest that TLR4 expression on monocytes is not a main regulator of LPS-induced TNF-α production.

28. Prevention of Postoperative Infections in Abdominal Surgery using Reabsorbable Suture with Antibacterial Activity (Vicryl Plus) Versus Reabsorbable Standard Sutures

Background: Postoperative complications in abdominal surgery may not only retard normal wound healing but also induce life-threatening clinical situations, particularly in patients with gastrointestinal sutures insufficiency and anastomosis leakage.

Objectives: The aim of study was to compare the incidence of postoperative inflammatory complications in elective surgery on the stomach and colon using Vicryl Plus versus reabsorbable sutures without antibacterial activity to close the abdominal wall and digestive anastomosis.

Material and methods: This was a prospective, randomized, open-label, comparative study. The study group consisted of 65 patients with clean-contaminated operations from 18 years old, operated upon benign and malignant diseases of stomach and colon using the reabsorbable sutures coated by Triclosan with antibacterial activity (Vicryl Plus). The control group consisted of 68 patients with clean-contaminated operations from 18 years old, operated upon for benign and malignant diseases of stomach and colon using different reabsorbable sutures without antibacterial activity. Groups are homogenous in a number of basic characteristics, making it possible to compare them according to certain criteria. The groups are similar in age and sex: the patient's age in both groups range from 20 to 85 years, average age is about 60 years, median age is equal to 63 years. Moreover these groups are similar in rate of malignant tumors – it is about 70% of all the patients operated on. All patients in study and in control groups received intravenous antibiotic prophylaxis 30–45 minutes before skin incision. Most of patients received amoxycillin-clavulanate in a single dose 1, 2 g or cephalosporins 2 generation.

Statistical analysis was done. P ≥ 0, 05 were considered to be statistically significant.

Results: The incidence of postoperative complications statistically significant decreased from 23% in the control group to 12, 3% in the study group. The anastomosis leakage was observed in 2 patients in the study group and in 4 patients in the control group. 2 patients in the control group who had been operated on for stomach cancer and pancreas cancer there occurred anastomosis leakage that lead to development of diffuse peritonitis and fatal outcome. The incidence of surgical wound infections decreased from 20,5% in the control group to 9,2% in the study group. At the same time in the control group of the patients deep wound infections were observed more. The mortality rate was 3, 0% in the control group. In the study group there were no fatal outcomes. The mean hospital stay duration depends on the different factors and there was no statistical difference in two groups. A more exact (accurate) characteristic – duration of stay in Intensive care unit, was lower in the study group in comparison with the control group. The mean duration in intensive care unit was 2, 8 ± 0, 1 in study group and 3, 1 ± 0, 15 in the control group.

Conclusion: Use of triclosan-coated sutures for abdominal wall closure and digestive anastomosis formation can reduce the number of intraabdominal postoperative complications and the number of wound infections in patients with clean-contaminated operations on stomach and colon.

29. Lack of Early Effects of Fluid Resuscitation with NACL 7.5 Per Cent on Selected Parameters of Innate Immunity in a Porcine Model of Serious Gunshot Trauma

Introduction: Major trauma and blood loss rapidly lead to dramatic and complex alterations of host immune responses. The extensive destruction of tissues together with introduction of foreign microbes and debris expose the organism to an overwhelming load of damage-associated molecular patterns. Exaggerated autodestructive inflammatory responses followed by immunosuppression may later result in sequential dysfunction or failure of remote organs. Hypertonic saline (HTS) is an excellent volume expander. Several studies also suggest independent and favourable effects on the post-traumatic immune function. In the present study we wanted to investigate the impact of HTS infusion on post-traumatic alterations of selected markers of innate immune function.

Material and methods: The study was conducted as part of a course in Traumatology and War Surgery, arranged by the Norwegian Armed Forces and the University of Oslo. In general anaesthesia 21 Norwegian landrace pigs pigs weighing 58.0 ± 2.6 kg (HTS group, n = 10) and 54.0 ± 3.5 kg (controls, n = 11) suffered two standardised gunshots towards abdomen and thigh. First aid treatment and surgical intervention were started without delay. The animals were randomised to receive either NaCl 7.5% 4 ml/kg iv in 30 minutes, or an equivalent volume of normal saline, starting 10 min after trauma. Additional fluid was infused as needed. The anaesthesiologist in charge was blinded as to treatment given. Haemodynamic parameters were continuously monitored. Blood samples were collected at baseline and 90 min after injury. Effects on markers of organ injury were measured in plasma. To search for effects of HTS on circulating immune cells, blood drawn at baseline and 90 min was stimulated ex vivo with LPS 10 ng/ml or an equivalent amount of normal saline for 6 h at 38 ΰC. Selected cytokines (TNF-α and IL-1 β) and MMP 9 were measured in the supernatant.

Results: Calculated serum osmolality after 90 min was 298 mOm/kg in the controls vs. 315 mOsm/kg in the HTS group (p < 0.05). Haemodynamic measurements were equal in both groups. No difference was seen in organ injury. As measured in ex vivo whole blood, trauma and haemorrhage induced a strong tolerance towards LPS stimulation, likewise without difference between the groups. Treatment with HTS did not affect the concentration of TNF-α, IL-1 β or MMP 9.

Conclusions: Infusion of NaCl 7.5 % 4 ml/kg as part of the immediate resuscitation fluid regimen did not induce any measurable effects on the early post-traumatic alterations of selected parameters of innate immune function.

32. Evaluation of the Effect of Intrabdominal Irrigation with Normal Saline and a with a Gentamicin-Clindamicin Solutionon the Appearance of Postoperative Intrabdominal Abscess and Wound Infection After Elective Surgery for Colorectal Cancer

Introduction: Colorectal surgery, even the elective one, is very prone to develop infections, thus the extravasation of microrganisms from the colonic lumen is unavoidable, despite meticulous aseptic measures. Colonic surgery is considered a contaminated procedure, conditioning the possible apearance of postoperative intrabdominal abscess and favouring a wound infection by contamination from the abdominal cavity.

Matherial and methods: A prospective randomized study of all the patients undergoing colorectal suregery for neoplasms at Hospital General Universitario de Elche during 2010 was performed. Patients were divided in 2 groups: Group 1 (intrabdominal irrigation with normal saline) and Group 2 (intraperitoneal irrigation with a gentamicin 240mg y clindamicin 600mg solution). Wound infection and intrabdominal abscess were investigated.

Results: 128 patients were analyzed, 64 in each group. There were no significant differences in age, gender, comorbilities and colorectal surgery performed among the patients of both groups. Wound infection rate was 41,9% in Group 1 vs 9,5 in Group 2 (p = 0,009). Intrabdominal infection rate, excluding those cases diagnosed of anastomotic leak (5%), was 16,3% in Group 1 vs 0% in Group 2 (p = 0,014).

Conclusion: Intraoperative irrigation with gentamicin-clindamicin solution reduces intrabdominal and wound infections

33. Percutaneous Drainage of Intrabdominal Abscess After Surgery: Role of Interventional Radiology

Objetive: This study evaluates the role and accuracy of interventional radiology (IR) procedures to manage intrabdominal abscess after abdominal surgery.

Methods: During the year 2010, we underwent 50 percutaneous drainages of intrabdominal abscesses in 50 patients (22 female and 28 male patients, age range 20–89 y-o) after abdominal surgery. A retrospective review is performed of the records of patients.

Results: All the lesions were developed after abdominal surgery (onset range 3 - 15 days). Thirty-seven of fifty abscesses (74%) were located in the peritoneal cavity, where the main related cause included cholecystectomy (16/50), colorectal surgery dehiscence (8/50), apendicetomy (6/50) and splenectomy (3/50). Ten of fifty abscesses (6%) were located in the retroperitoneal cavity and 10 abscesses (20%) in the pelvis.

Forty-seven cases (94%) were performed under ultrasonographic (US) guide. In two cases (4%) CT scan guiding added to US was necessary. Conventional fluoroscopy added to US was used in one case (2%).

A definitive treatment was obtained in more that 90% of cases. Failure most commonly occurred with multiloculated lesions or lesions associated with fistulous communication. There were no complications.

Conclusions: Abscess complications occurring after abdominal surgery can be managed effectively using IR, thus minimizing morbidity and the need for reoperation, most of the time only under US guide. Even if a total cure is not achieved, a beneficial temporary effect may be obtained. This procedure should be indicated for the initial treatment of postsurgical abscesses.

Abstract Author Indexby abstract number