The Effect of Vacuum-Assisted Closure in Bacterial Clearance of the Infected Abdomen

Abstract

Background:

Laparostomy with vacuum-assisted closure (VAC) plays an important role in improving survival in the presence of abdominal infection. We conducted a study of the qualitative changes in the bacterial flora of the peritoneal cavity in patients with severe abdominal infection treated with laparostomy and a VAC device.

Methods:

Thirty-nine patients with severe abdominal infection treated with abdominal opening and VAC were registered in a clinical study. When an incidence of 53.8% of hospital-acquired peritoneal infection (HAPI) was found in the study patient population, it was decided to divide the patients in two groups according to whether or not they developed a HAPI. The patients' outcomes were then analyzed.

Results:

The durations of abdominal opening (p=0.04), length of stay in the intensive care unit (ICU) (p=0.01), and of hospitalization (p=0.04) were significantly greater in patients with HAPI than in those without it, whereas mortality did not differ on the basis of these three variables.

Conclusions:

Superinfection is common in laparostomy done with a VAC device for managing severe abdominal infection. The data in the present study show that VAC does not alter the quality of the bacterial burden in primary abdominal contamination, nor does it seem to prevent a high incidence of HAPI. However, VAC is as effective in reducing mortality among patients with HAPI as among those without it.

Severe sepsis of abdominal origin is defined as sepsis plus organ dysfunction, caused by a condition arising in the peritoneal cavity [1]. The essentials of the management of patients with severe sepsis remain unchanged and can be divided broadly into source control and hemodynamic stabilization. Control of sepsis mainly involves surgical exploration and temporary or permanent elimination of the infectious agent [2]. Currently, open abdominal treatment remains among the valid approaches to severe abdominal sepsis [3 –5]. Topical negative pressure with an abdominal vacuum-assisted closure (VAC) device appears to be the optimal means for managing an infected open abdomen [5 –10]. This method has several advantages, chiefly related to: (1) The opportunity to close hermetically the abdominal compartment; (2) an easily performed “second-look” laparotomy; and (3) multiple revisions of the abdominal cavity if needed to improve infection control [5,8,9,11].

Although evidence-based data are missing, topical negative pressure in patients with severe abdominal infection appears to make a positive contribution to the treatment of these critically ill patients [12,13]. An important issue to be addressed is whether negative abdominal pressure alters the composition of the bacterial population in the peritoneal cavity. We hypothesized that in patients with severe abdominal infection treated with the open-abdomen technique, use of a VAC device would reduce the bacterial burden of the abdomen.

Patients and Methods

To determine whether use of a VAC device would reduce the bacterial burden of the abdomen in patients with infection treated with the open-abdomen technique, we performed a retrospective analysis of a prospectively formed database that was approved by the AHEPA University Hospital Research Ethics Board. The analysis was performed from January 2009 to December 2011. Written informed consent was obtained from the closest relatives of the patients in our study, in accordance with the Good Clinical Practice Guidelines of the International Conference on Harmonisation [14]. The severe abdominal infection in these patients was not of traumatic origin. Laparostomy was performed when the Sequential Organ Failure Assessment (SOFA) score was >7 or the Mannheim peritonitis score was >29 [15].

The criteria for inclusion in the study were: (1) Age >18 y, (2) consent to participate in the study, and (3) laparostomy managed with a VAC system [V.A.C.; KCI International, San Antonio, TX]. The only exclusion criterion was death within 24 h after the patient's admission. With use of the criteria named above, 39 patients were included in the study. Because of the severity of their infections, all patients in the study were intubated or tracheostomized. It is our department's policy to perform ostomies in all cases of advanced peritonitis, to avoid the risks inherent in anastomosis. No intra-abdominal drains were placed in the study patients.

Bacteriologic sampling and culture

At initial exploratory laparotomy, 5 mL of peritoneal fluid was aspirated with a syringe. The sample was emptied into a rubber membrane-sealed, sterile glass tube and transported immediately to the laboratory, where it was processed directly. The same process was repeated at each change of dressings. For bacterial culture, a 0.1 mL sample of the aspirated fluid was inoculated into both aerobic and anaerobic agar plates, incubated at 37°C for 7 d, and inspected daily for bacterial growth. Microorganisms were quantified as the number of colony-forming units [CFU]/mL and identified through routine microbiologic techniques [16]. Bacterial identification to the species level and initial antibiotic susceptibility testing were performed with the VITEK2–automated system (bioMérieux, Marcy l'Etoile, France). Additionally, for multi-drug-resistant strains of organisms, testing of susceptibility to a range of antimicrobial agents was assessed with the disc diffusion method according to the guidelines of the Clinical and Laboratory Standards Institute (CLSI, formerly the National Committee on Clinical Laboratory Standards [NCCLS]) and with the Etest (AB, Biodisk, Solna, Sweden), according to the manufacturer's instructions [16]. Cultures of peritoneal fluid that grew more than one species of pathogen were considered polymicrobial. For the purpose of the study, all subspecies of Enterococcus and Escherichia coli were considered collectively under the name of intestinal bacteria. Data on other healthcare-associated infections were not collected.

Antimicrobial prophylaxis

All patients were given piperacillin/tazobactam 4.5 g i.v. empirically 30 min before the induction of anesthesia. The antibiotic administration regimen was later adjusted according to the results of peritoneal fluid culture.

Sponge changes and abdominal closure

Patients returned to the operating room every 2 d for sequential fascial closure and replacement of sponges and sutures, with a resulting decrease in the fascial defect from laparotomy of approximately 3–4 cm per session until final abdominal closure, according to the protocol proposed by Pliakos et al. [17]. After sampling was done, the various abdominal spaces were thoroughly lavaged with a small volume (1,500 mL) of saline solution. In our opinion, there is no need during the period when the abdomen is open to irrigate with a large volume of normal saline (NS) at every dressing change. Less than 1500 mL of NS seems to be more than adequate for this purpose. The major portion of pus and the bacterial burden on the abdomen was certainly removed during the primary operation done on each patient. When sequential fascial closure was not feasible because of a lack of immediate normalization of the patient's physiology, treatment through a planned ventral hernia repair was required.

Statistical analyses

The operative and clinical variables recorded and analyzed in the study were age, gender, pathology, initial Sequential Organ Failure Assessment (SOFA) score, Acute Physiology and Chronic Health Evaluation II (APACHE II) score [18], intra-abdominal pressure (IAP), results of peritoneal culture, hospital-acquired peritoneal infection (HAPI), duration of an open abdomen, number of dressing changes, hospitalization, rate of successful abdominal closure, and mortality. The data were analyzed with standard statistical methods. Results are expressed as mean±standard deviation (SD) or as median and range. The Mann–Whitney U test, the Student t-test, the χ² test, analysis of variance (ANOVA), and the Kruskal–Wallis test were used as appropriate to determine statistical significance, which was set at p<0.05. A marginal homogeneity test was used to determine whether dependent categorical data differed from one another. Statistical analysis was done with the SPSS 17 data analysis program (SPSS, Inc., Chicago, IL).

Results

The mortality rate was 35.9% (14 patients). All 39 patients in the study were admitted on an emergency basis. Of the 39 patients in the study, 32 patients (82.1%) had a diagnosis of purulent peritonitis, five patients (12.8%) had a diagnosis of fecal peritonitis, and two patients (5.1%) had a diagnosis of necrotizing pancreatitis. Sepsis originated in the colon and rectum in 27 patients, in the stomach in eight patients, in the pancreas in two patients, and in the small intestine in two patients. None of the patients had fistulas, and second-look operations revealed no abscesses.

Thirteen patients died before closure of the fascia. In the remaining 26 patients, the outcome in one patient (3.8%) was partial closure of the fascia and complete closure of the skin (planned ventral hernia); in two other patients (7.7%) no closure of the fascial defect was possible; and in 23 patients (88.5%), a full closure was feasible, although the outcome in one of these patient was fatal. Following the development of a layer of granulation tissue, the two patients without fascial closure underwent skin grafting. The clinical characteristics and outcomes of the patient groups with and without HAPI are shown in Table 1.

Table 1.

Epidemiologic-Clinical Characteristics and Outcome of the Patients

Patient characteristics	HAPI group (n=21)	Non HAPI group (n=18)	Total (n=39)	p^
Age (y), mean [SD]	57.38 [17]	59.94 [19.73]	58.5 [18.1]	0.66
SOFA Score (points), median (range)	9 (8–12)	9 (8–14)	9 (8–14)	0.47
APACHE II Score [SD]	19.67 [5.06]	20.33 [6.28]	20 [5.6]	0.71
Initial IAP (mmHg), median (range)	11 (5.1–21.3)	10.3 (6.6–16.6)	10.3 (5.1–21.3)	0.79
SOFA Score after 1^st dressing change, mean [SD]	4.71 [3.38]	4.89 [3.53]	4.79 [3.45]	0.87
Dressing changes, median (range)	3 (1–12)	2 (1–6)	3 (1–12)	0.06
Open abdomen duration (d), median (range)	13 (3–58)	8.5 (3–19)	11 (3–58)	0.04^*
ICU stay (d), median (range)	23 (2–79)	8.5 (1–30)	13 (1–79)	0.01^*
Hospitalization (d), median (range)	43.86 (42.44)	22.11 (11.2)	26 (3–172)	0.04^*
Primary abdominal closure (%)	8 (38.1)	14 (77.8)	22 (56.4)	0.06
• Death prior to abdominal closure (%)	9 (42.9)	4 (22.2)	13 (33.3)
• Primary abdominal closure & death (%)	1 (4.8)	-	1 (2.6)
• Skin grafting (%)	2 (9.6)	-	2 (5.1)
• Planned hernia (%)	1 (4.8)	-	1 (2.6)
Mortality (%)	10 (47.6)	4 (22.2)	14 (35.9)	0.09

HAPI=Hospital Acquired Peritoneal Infection; SOFA=Sequential Organ Failure Assessment; APACHE II=Acute Physiology and Chronic Health Evaluation II; IAP=Intrabdominal Pressure; ICU=Intensive Care Unit.

Statistically significant.

^ concerns comparison between HAPI and non-HAPI groups.

Initial diagnosis had no influence on the morbidity in the study (dressing changes, p=0.6; duration of open abdomen, p=0.7; ICU stay, p=0.3; duration of hospitalization, p=0.4; and mortality, p=0.8. The results of initial peritoneal culture and cumulative results of successive peritoneal cultures are shown in Table 2. In 21 patients (53.8%), different kinds of bacteria than those found on initial peritoneal culture were isolated and were considered to represent HAPI. In this group we distinguished 11 patients with negative initial peritoneal cultures, seven patients with initial peritoneal cultures positive for intestinal bacteria, and three patients with initial polymicrobial peritoneal contamination. Analytically, the successive peritoneal cultures of the 11 patients who had negative initial peritoneal cultures showed polymicrobial peritoneal contamination (three patients) and contamination by Pseudomonas aeruginosa (four patients), Acinetobacter baumanni (one patient), Proteus mirabilis (one patient), Klebsiella pneumoniae (one patient) and Staphylococcus epidermidis (one patient). In the remaining 10 patients in the group with pathogens other than those initially isolated, HAPI was caused by additional species of bacterial pathogens. Two patients in this group had multi-drug-resistant bacteria, consisting of Acinetobacter baumanni in one patient and Pseudomonas aeruginosa in the other.

Table 2.

Initial and Successive Peritoneal Culture Findings

Initial peritoneal culture	Patients (%)	Successive cultures	Patients (%)
Negative	20 (51.2)	Negative	9 (23.1)
Intestinal bacteria	14 (35.9)	Intestinal bacteria	7 (17.9)
Polymicrobial	4 (10.3)	Polymicrobial	14 (35.9)
Fungal	1 (2.6)	Fungal	1 (2.6)
		Acinetobacter baumanni	1 (2.6)
		Proteus mirabilis	1 (2.6)
		Pseudomonas aeruginosa	4 (10.3)
		Klebsiella pneumoniae	1 (2.6)
		Staphylococcus epidermidis	1 (2.6)
Total	39 (100)		39 (100)

Of the 18 (46.2%) patients in the study who did not have an HAPI, peritoneal culture remained negative in nine patients. The initial peritoneal contamination with intestinal bacteria persisted in seven other patients, the initial polymicrobial peritoneal infection in one additional patient remained, and an initial peritoneal infection by Candida albicans in one other, immunosuppressed patient, also remained. One patient in the group with HAPI developed an infection with multi-drug-resistant bacteria (Enterococcus spp.). Three parameters showed statistically significant differences between the groups with and without HAPI. These were: (1) Duration of ICU stay; (2) duration of hospitalization; and (3) duration of an open abdomen (Table 1).

Discussion

Treatment through an open abdomen should be considered after celiotomy in patients with sepsis who would benefit from planned, early re-exploration of the peritoneal cavity for additional surgical procedures [4]. Massive contamination of the peritoneal cavity in severe abdominal infection, with the need for repeated abdominal lavage, and edema of the bowel from high-volume fluid therapy, are the two most common reasons for the use of laparostomy [4,5].

The use of various vacuum-assisted techniques for temporary abdominal closure was a turning point in the management of laparostomies, minimizing bowel injury, fistulas, loss of peritoneal fluid, and loss of body heat [5,8,10,11,19]. The decision to continue management with an open abdomen is made at the time of each re-operation after an initial laparostomy and is based on the need for a new, planned re-exploration of the abdomen. The technique of VAC appears to be a reliable method for temporary closure of the open abdomen in patients with severe abdominal sepsis [5,9,13]. Although VAC has proven its efficacy in improving survival in several studies [9,13], it is not clear how the negative pressure it creates contributes to the treatment of these patients. Many authors have implied a role of negative pressure in removing the microbial burden on the infected abdomen [8,12,13]. However, there is no evidence that VAC has such an effect. In the present study, in which a high incidence of HAPI (53.8%) was found in the patient population, we observed qualitative changes in the microbial environment of the open abdomen treated with VAC, but no difference in the incidence of HAPI with and without VAC.

Abdominal sepsis usually resolves after surgical intervention and appropriate antimicrobial therapy, with the complete recovery of affected patients. However, in several types of patients this secondary infectious process is overwhelming and cannot be contained through customary procedures [20,21]. Hospital-acquired peritoneal infection is probably related to factors in a patient's initial host response, to impaired host-defense mechanisms, to the ICU environment, and to the VAC application technique.

The likelihood of HAPI is probably accentuated when laparostomy is performed. In this procedure the skin, which is the natural barrier of the abdomen, is left open and a polyurethane sponge with an air-tight plastic sheet takes its place. When several instances of sponge replacement are necessary, the peritoneal cavity becomes more vulnerable to infection from an external source. In thw clinical trial reported here we studied the influence of HAPI on the outcomes of patients with severe abdominal infection in terms of number of dressing changes, duration of an open abdomen, ICU stay, duration of hospitalization, rate of abdominal closure, and mortality. Our data indicate that in treatment through an open abdomen, the primary infection does not influence the final outcome in terms of number of dressing changes (p=0.6), duration of an open abdomen (p=0.7), duration of ICU stay (p=0.3), duration of hospitalization (p=0.4), or mortality (p=0.8). However, HAPI played a decisive role in the outcomes of the patients in our study cohort. In patients with HAPI, the duration of an open abdomen (p=0.04), duration of ICU stay (p=0.01), and duration of hospitalization (p=0.04) were significantly longer than in patients without HAPI. Although a greater number of dressing changes (p=0.06), primary closure rate (p=0.06), and mortality (p=0.09) was observed in the group of patients with HAPI than in the group without HAPI, the difference did not reach statistical significance. Thirteen patients had primary abdominal closure within 7 d after laparostomy. In this group, five patients (38.4%) developed HAPI. In 26 patients primary abdominal closure occurred in more than 7 d after laparostomy. In this group of patients, 16 patients (61.5%) developed HAPI. In our study, the overall rate of primary closure, which is a very significant component of morbidity in open abdominal therapy, was 88.5%. This rate is not much lower than that in our previous study, of 93.3% [17]. However, the low rate of primary abdominal closure (42.9%) in patients with HAPI is alarming.

Nosocomial infections are a common, severe problem worldwide, and are associated with significant morbidity and mortality as well as with rapidly increasing multi-drug resistance to antibiotics of the pathogens responsible for these infections [23 –27]. Patients in the ICU are at greater risk of nosocomial infection than patients in the community or in other hospital units, and the rates of resistance to antimicrobial agents of pathogens isolated in ICU infections are substantially higher than those in community-acquired or other in-hospital infections [23 –26,28,29]. The high incidence of infection and multi-drug resistance in ICU patients is attributable to severe underlying medical conditions, the frequent and sometimes unnecessary use of broad-spectrum antibiotics, mechanical ventilation, poor compliance with regimens for hygiene, inadequate disinfection or sterilization of devices and equipment, prolonged use of invasive devices, overcrowding and inefficient isolation of infected patients, prolonged hospital lengths of stay, and greater risk of cross-transmission of resistant microorganisms [30 –34].

Regarding the microbiologic profile of the HAPIs in our study population, polymicrobial infection was the most common finding and P. aeruginosa was the second most frequently identified pathogen. A. baumanni, P. mirabilis, K. pneumoniae, and S. epidermidis, respectively, were identified in four patients, and the same pathogens (although in different proportions) were identified in polymicrobial infections. Confirmation of the microbial nature of a disease relies either on the microscopic visualization of pathogens in tissue samples or, more commonly, on culture of the responsible pathogen from tissue samples. However, results of culture usually take at least 24 h to become available, and even then culture permits positive identification in only about 50% of cases [35,36]. We believe that the negative initial cultures in our series were the result either of previous antimicrobial treatment or of the presence of fastidious or slow-growing pathogens.

The exact origin of HAPIs in our patients remains unknown. Some of the implied pathogens were probably translocated via the blood (e.g., urinary tract infection, pulmonary infection) because positive cultures of the same pathogens were found in bronchoalveolar lavage specimens, arterial catheters, central and peripheral venous catheters, and urinary catheters. Furthermore, we cannot exclude microbial translocation as the result of poor hygienic practices on the part of ICU personnel or from the inefficient isolation of patients, especially during dressing changes.

These findings indicate the importance of effective prevention and treatment of infections in ICU patients. Because each unit in a hospital has a distinct bacteriologic profile and antibiotic-resistance pattern, knowledge of these differences is critical for planning effective prophylaxis and treatment for infections and reducing infection-related cost, morbidity, and mortality [29,36]. The eradication of pathogens requires implementation of rigorous infection-control measures, prudent antibiotic use, and effective antimicrobial therapy. The recognition of ICU patients' clinical and microbiologic characteristics is essential for preventing and treating infections.

Conclusions

The present study is the first prospective clinical study directed at evaluating the qualitative changes in the bacterial flora of the peritoneal cavity in patients with severe abdominal sepsis undergoing open abdominal therapy with a VAC device. Despite the proven efficacy of VAC in the management of severe abdominal infection, our data show that it cannot alter the bacterial burden resulting from primary abdominal contamination. Furthermore, the use of VAC does not seem to prevent a high incidence of HAPI. This prolongs the ICU stay and duration of hospitalization of patients and multiplies the morbidity of patients with HAPI, although without adversely influencing their mortality. The study reported here is undoubtedly of qualitative value, but the limited number of patients involved limits its potential for generalization. More extensive, stratified studies are needed to reach more definitive conclusions about the effect of VAC in laparostomy.

Author Disclosure Statement

No competing financial interests exist.

References

Calandra

, Cohne

. The international sepsis forum consensus conference on definitions of infection in the intensive care unit. Crit Care Med, 2005; 33:1538–1548.

Ordoñez

, Puyana

. Management of peritonitis in the critically ill patient. Surg Clin North Am, 2006; 86:1323–1349.

Duff

, Moffat

. Abdominal sepsis managed by leaving abdomen open. Surgery, 1981; 90:774–778.

Stawicki

, Brooks

, Bilski

, et al. The concept of damage control: Extending paradigm to emergency general surgery. Injury, 2008; 39:93–101.

Pliakos

, Papavramidis

, Michalopoulos

, et al. The value of vacuum-assisted closure in septic patients treated with laparostomy. Am Surg, 2012; 78:957–961.

Papavramidis

, Marinis

, Pliakos

, et al. Abdominal compartment syndrome—Intra-abdominal hypertension: Defining, diagnosing, and managing. J Emerg Trauma Shock, 2011; 4:279–291.

Papavramidis

, Michalopoulos

, Mistriotis

, et al. Abdominal compliance, linearity between abdominal pressure and ascitic fluid volume. J Emerg Trauma Shock, 2011; 4:194–197.

Barker

, Green

, Maxwell

, et al. Experience with vacuum-pack temporary abdominal wound closure in 258 trauma and general and vascular surgical patients. J Am Coll Surg, 2007; 204:784–792.

Perez

, Wildi

, Demartines

, et al. Prospective evaluation of vacuum-assisted closure in abdominal compartment syndrome and severe abdominal sepsis. J Am Coll Surg, 2007; 205:586–592

10.

Campbell

, Chang

, Fabian

, et al. Management of the open abdomen: From initial operation to definitive closure. Am Surg, 2009; 75(11 Suppl):S1–22.

11.

Stevens

. Vacuum-assisted closure of laparostomy wounds: A critical review of the literature. Int Wound J, 2009; 6:259–266.

12.

Amin

, Shaikh

. Topical negative pressure in managing severe peritonitis: A positive contribution?. World J Gastroenterol, 2009; 15:3394–3397.

13.

Horwood

, Akbar

, Maw

. Initial experience of laparostomy with immediate vacuum therapy in patients with severe peritonitis. Ann R Coll Surg Engl, 2009; 8:681–687.

14.

International Conference on Harmonisation (ICH). Guidelines good clinical practice (E6) as adopted by the FDA, (1997). Available from: http://www.ich.org/LOB/media/MEDIA482.pdf.

15.

Vincent

, Moreno

, Takala

J. S

, et al. The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med, 1996; 22:707–710.

16.

Clinical and Laboratory Standards Institute. Performance Standards for Antimicrobial Susceptibility Testing; Twentieth Informational Supplement, CLSI document M100-S20-U. Wayne, PA; Clinical and Laboratory Standards Institute, 2010.

17.

Pliakos

, Papavramidis

, Mihalopoulos , et al. Vacuum assisted closure in severe abdominal sepsis with or without retention sutured sequential fascial closure: A clinical trial. Surgery, 2010; 148:947–953.

18.

Knaus

, Draper

, Wagner

, et al. APACHE II: A severity of disease classification system. Crit Care Med, 1985; 13:818–829.

19.

Subramonia

, Pankhurst

, Rowlands

, et al. Vacuum- assisted closure of postoperative abdominal wounds: A prospective study. World J Surg, 2009; 33:931–937.

20.

Bosscha

, van Vroonhoven

, van der Werken

. Surgical management of severe secondary peritonitis. Br J Surg, 1999; 86:1371–1377.

21.

Dellinger

, Levy

, Carlet

et al; International Surviving Sepsis Campaign Guidelines Committee; American Association of Critical-Care Nurses; American College of Chest Physicians; American College of Emergency Physicians; Canadian Critical Care Society; European Society of Clinical Microbiology and Infectious Diseases; European Society of Intensive Care Medicine; European Respiratory Society; International Sepsis Forum; Japanese Association for Acute Medicine; Japanese Society of Intensive Care Medicine; Society of Critical Care Medicine; Society of Hospital Medicine; Surgical Infection Society; World Federation of Societies of Intensive and Critical Care Medicine. Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med, 2008; 36:296–327.

22.

Vincent

, Bihari

, Suter

, et al. The prevalence of nosocomial infection in intensive care units in Europe. Results of the European Prevalence of Infection in Intensive Care (EPIC) Study. EPIC International Advisory Committee. JAMA, 1995; 274:639–644.

23.

Sligl

, Taylor

, Brindley

. Five years of nosocomial gram-negative bacteremia in a general intensive care unit: Epidemiology, antimicrobial susceptibility patterns, and outcomes. Int J Infect Dis, 2006; 10:320–325.

24.

Streit

, Jones

, Sader

, et al. Assessment of pathogen occurrences and resistance profiles among infected patients in the intensive care unit: Report from the SENTRY Antimicrobial Surveillance Program (North America, 2001). Int J Antimicrob Agents, 2004; 24:111–118.

25.

Safdar

, Maki

. The commonality of risk factors for nosocomial colonization and infection with antimicrobial-resistant Staphylococcus aureus, Enterococcus, gram-negative bacilli, Clostridium difficile, and Candida. Ann Intern Med, 2002; 136:834–844.

26.

Fridkin

, Steward

, Edwards

, et al. Surveillance of antimicrobial use and antimicrobial resistance in United States hospitals: Project ICARE phase 2. Project Intensive Care Antimicrobial Resistance Epidemiology (ICARE) hospitals. Clin Infect Dis, 1999; 29:245–252.

27.

Kourkoumpetis

, Velmahos

, Ziakas

, et al. The effect of cumulative length of hospital stay on the antifungal resistance of Candida strains isolated from critically ill surgical patients. Mycopathologia, 2011; 171:85–91.

28.

Zhanel

, DeCorby

, Laing

, et al; Canadian Antimicrobial Resistance Alliance (CARA). Antimicrobial-resistant pathogens in intensive care units in Canada: Results of the Canadian National Intensive Care Unit (CAN-ICU) study, 2005–2006. Antimicrob Agents Chemother, 2008; 52:1430–1437.

29.

Edgeworth

, Treacher

, Eykyn

. A 25-year study of nosocomial bacteremia in an adult intensive care unit. Crit Care Med, 1999; 27:1421–1428.

30.

Richards

, Edwards

, Culver

, et al. Nosocomial infections in combined medical-surgical intensive care units in the United States. Infect Control Hosp Epidemiol, 2000; 21:510–515.

31.

Hanberger

, Garcia-Rodriguez

, Gobernado

, et al. Antibiotic susceptibility among aerobic gram-negative bacilli in intensive care units in 5 European countries. French and Portuguese ICU Study Groups. JAMA, 1999; 281:67–71.

32.

Rello

. Acinetobacter baumannii infections in the ICU: Customization is the key. Chest, 1999; 115:1226–1229.

33.

Markogiannakis

, Pachylaki

, Samara

, et al. Infections in a surgical intensive care unit of a university hospital in Greece. Int J Infect Dis, 2009; 13:145–153.

34.

Martin

, Mannino

, Eaton

, et al. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med, 2003; 348:1546–1554.

35.

Lever

, Mackenzie

. Sepsis: Definition, epidemiology, and diagnosis. BMJ, 2007; 335:879–883.

36.

Cookson

, Macrae

, Barrett

, et al.; Combined Working Party of the Hospital Infection Society and Infection Control Nurses Association. Guidelines for the control of glycopeptide-resistant enterococci in hospitals. J Hosp Infect, 2006; 62:6–21.