Infection (2014) 42:1013–1022 DOI 10.1007/s15010-014-0678-1
ORIGINAL PAPER
Risk factors for enterococcal infection and colonization by vancomycin-resistant enterococci in critically ill patients M. Papadimitriou-Olivgeris • E. Drougka • F. Fligou • F. Kolonitsiou • A. Liakopoulos • V. Dodou • E. D. Anastassiou • E. Petinaki • M. Marangos K. S. Filos • I. Spiliopoulou
•
Received: 30 June 2014 / Accepted: 13 August 2014 / Published online: 21 August 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Present Address: A. Liakopoulos Central Veterinary Institute (CVI) of Wageningen UR, Lelystad, The Netherlands
Methods A prospective study regarding patients admitted in ICU (n = 497) was conducted during a 24-month period. Rectal swabs were collected upon admission and during hospitalization and inoculated onto selective medium. Enterococci were phenotypically characterized. van genes were investigated by PCR and clones were identified by Pulsed-Field Gel Electrophoresis and Multilocus Sequence Typing. Epidemiologic data were collected from the ICU database. Results Risk factors for VRE carriage upon ICU admission (71/497) were: duration of previous hospitalization, glycopeptide administration, chronic heart failure, malignancy, insulin-dependent diabetes mellitus, and previous enterococcal infection (VRE and/or VSE). Risk factors for VRE colonization during ICU stay (36/250) were: quinolone administration, chronic obstructive pulmonary disease, chronic renal failure, and number of VRE-positive patients in nearby beds. Risk factors for enterococcal infection during ICU stay (15/284), including VRE and VSE cases, were: administration of third- or fourth-generation cephalosporins, cortisone use before ICU admission and VRE colonization, whereas, enteral nutrition was a protective factor. Conclusions Previous VRE colonization and antibiotic usage are essential parameters for enterococcal infection (by VRE or VSE) during ICU stay. Previous enterococcal infection, co-morbidities and antibiotic usage are associated with VRE colonization upon ICU admission, whereas, patient to patient transmission, co-morbidities and antibiotic usage constitute risk factors for VRE colonization during ICU hospitalization.
V. Dodou Intensive Care Unit, Saint Andrew’s General Hospital, 26335 Patras, Greece
Keywords Enterococci VRE Colonization Infection ICU Risk factors Clones
Abstract Purpose Vancomycin-Resistant Enterococci (VRE) are important causes of Intensive Care Unit (ICU) infections. Our goal was to identify the prevalence and risk factors for VRE colonization upon ICU admission and during ICU stay, as well as, their impact in enterococcal infection including vancomycin-susceptible cases (VSE).
A part of this work was presented as a poster presentation at the 22nd European Congress of Clinical Microbiology and Infectious Diseases, 31 March–3 April 2012, London, United Kingdom. M. Papadimitriou-Olivgeris M. Marangos Division of Infectious Diseases, School of Medicine, University of Patras, Rion, 26504 Patras, Greece E. Drougka F. Kolonitsiou E. D. Anastassiou I. Spiliopoulou (&) Department of Microbiology, School of Medicine, University of Patras, Rion, 26504 Patras, Greece e-mail:
[email protected] F. Fligou K. S. Filos Department of Anaesthesiology and Critical Care Medicine, University of Patras, Rion, 26504 Patras, Greece A. Liakopoulos E. Petinaki Department of Microbiology, School of Medicine, University of Thessalia, Biopolis, 41222 Larissa, Greece
123
1014
Introduction Since their emergence in 1986, Vancomycin-Resistant Enterococci (VRE) became an important nosocomial pathogen in European countries, especially in Greece [1–3]. According to data submitted to the European Antimicrobial Resistance Surveillance System (EARSS) [http://ecdc. europa.eu], a significant increase of VRE rates was detected in Greece from 2000 to 2007, followed by decline [2, 3]. Even though these data rely on different methods from participating laboratories, they are in accordance with surveillance data from our Institution. More specifically, VRE were 19 % in 2003, reached 24 % in 2007, while during the study period, 2010 and 2011, were 10 and 8.5 %, respectively, among the total enterococcal infections in our setting [4]. In Greece, like in other countries, the majority of VRE are Enterococcus faecium carrying vanA gene, while E. faecalis is rare and vanB is identified to a lesser extent [2–5]. Asymptomatic VRE faecal colonization precedes infection, while, colonization rates exceed those of infections [6]. As for many multidrug resistant pathogens, VRE colonization is associated with administration of different classes of antibiotics (especially vancomycin), invasive procedures, medical devices and underlying disease [6–8]. Epidemiological studies have revealed that VRE dissemination in a hospital setting may be a combination of strains’ clonal expansion with horizontal gene transfer among different strains [4–7]. These data suggest that infection control measures such as screening for VRE upon admission of patients at risk, as well as, isolation of identified cases should be applied in specific hospital wards. The purpose of the present study was to implement active surveillance for VRE faecal colonization upon ICU admission and during patients’ stay, to identify the risk factors for colonization and infection during ICU hospitalization, and to assess these factors for infection control practices.
Materials and methods Hospital and bacterial isolates This is a prospective study performed in two general ICUs, the ICU of the University General Hospital of Patras (UGHP, ICU-A) and the ICU of ‘‘Saint Andrew’s General Hospital’’ of Patras (ICU-B) that was transferred and hosted from June 2009 to February 2011 to the UGHP because of an earthquake. The study was carried out during a 24-month period (January 2010 to March 2011 and June 2011 to February 2012) at the ICU-A, whereas, for ICU-B during a 14-month period (January 2010 to February 2011). ICUs A and B have 13 and 6 beds, and 366 and 80
123
M. Papadimitriou-Olivgeris et al.
admissions per year, respectively. During April–May 2011, the largest hospitalization room of the ICU-A with 10 beds, was closed and disinfected. The study was carried out under the ‘‘Hospital Surveillance Programme for multidrug resistant bacterial colonization of patients at risk and HCWs’’, and was approved by the University Hospital Ethics Committee that waived the need for informed consent (HEC No: 571). Rectal swabs were obtained within 24 h after admission and follow-up cultures were collected every week thereafter, till final outcome. Patients’ enterococcal infections in total, including vancomycin-resistant and -susceptible cases 1 month before and those acquired during ICU hospitalization were recorded. Enterococcal urinary tract infections were defined on the basis of clinical findings, pyuria and a positive urine culture of C103 and \105 CFU/ mL with no more than two species of microorganisms [9]. Colonizing and infecting isolates were further analyzed. Demographic characteristics (age, sex), severity of illness, scores of severity upon admission, chronic illnesses, length of hospitalization, surgery during the last month, antibiotic usage for at least 2 days, and use of intravenous catheters were collected from patients’ chart reviews. Briefly, rectal swabs were inoculated into Brain-HeartInfusion-Broth with colistin and caspofungin for 24 h in order to inhibit the growth of multi-drug resistant Gramnegative bacteria and fungi. Subcultures were performed onto chromogenic agar (ChromID VRE, bioMerieux, La Palme, Les Grottes, France). Thirty-seven enteroccoci recovered from patients fulfilling infection criteria (22 before ICU admission and 15 during ICU hospitalization) were also analyzed. Enterococci were phenotypically identified by Gram stain, catalase test, and Vitek 2 Advanced Expert System (bioMerieux). Antibiotic susceptibility testing It was performed by the disk diffusion method against: penicillin, ampicillin, erythromycin, chloramphenicol, vancomycin and teicoplanin. Minimal Inhibitory Concentrations (MICs) of vancomycin, teicoplanin, linezolid, daptomycin, tigecycline and gentamicin (high-level) were determined by the Etest (bioMerieux) [10]. Genotypes of isolates The presence of vanA and vanB genes (conferring resistance to vancomycin) was investigated by PCR in all isolates [11]. Clones were identified by Pulsed-Field Gel Electrophoresis (PFGE) of chromosomal SmaI DNA digests and interpreted according to already established criteria [12]. PFGE types (pulsotypes) of E. faecalis were named by capital letters and those of E. faecium by lower
Risk factors for enterococcal infection and colonization
1015
case letters. Furthermore, colonizing VRE strains were characterized by Multilocus Sequence Typing (MLST) [www.mlst.net]. Statistical analysis SPSS version 19.0 (SPSS, Chicago, IL) software was used for data analysis. Categorical variables were analyzed using the Fisher exact test or chi2 and continuous variables with the Mann–Whitney U test, as appropriate. Odds Ratios (ORs) and 95 % Confidence Intervals (CIs) were calculated to evaluate the strength of any association. Univariate analysis was used to determine those risk factors associated with development of VRE colonization and enterococcal Table 1 Enterococcal isolates recovered from rectal swabs of colonized Intensive Care Unit (ICU) patients during the studied period
infection. Backward stepwise multiple logistic regression analysis used all those significant variables from the univariate analysis with a P \ 0.08 in order to determine which ones were independently associated with development of VRE colonization and enterococcal infection. Factors contributing to multicollinearity were excluded from the multivariate analysis. All statistical tests were two-tailed and P \ 0.05 was considered statistically significant.
Results Among 497 patients admitted in both ICUs, 71 (14.3 %) were colonized by VRE upon admission (66 by E. faecium
Species
CC
STs (number)
PFGE types (number)
E. faecium (100)
17
ST117 (58)
a (55); f (2); i (1)
ST17 (18)
CC Clonal Complex, ST Sequence Type identified by MLST, PFGE pulsed field electrophoresis types a
van genes: A: vanA gene, B: vanB gene
E. faecalis (7)
d (14); c (4)
van genesa (number)
Upon admission
During stay
A (42)
31
11
B (16)
8
8
A (17)
11
6
B (1)
0
1
ST203 (11)
b (11)
A (11)
9
2
ST125 (3)
g (3)
A (3)
1
2
94
ST226 (5)
h (5)
A (5)
3
2
Singleton
ST786 (5)
e (5)
A (5)
3
2
ST6 (3)
A (3)
A (3)
1
2
ST41 (2)
F (2)
B (2)
2
0
ST19 (1) ST28 (1)
D (1) E (1)
B (1) B (1)
1 1
0 0
Fig. 1 Newly discovered VRE-colonized patients per month in both ICUs (study period: January 2010–February 2012). VRE: vancomycin-resistant enterococci; ICU: Intensive Care Unit. Samples were
collected from ICU-B as long as it was hosted in UGHP, until February 2011 (month 14). During April and May 2011 (months 16 and 17) no rectal samples were collected
123
123
E. faecalis
Susceptible
A
259
vanA
b
E. faecium
Bacteraemia
629
E. faecium
E. faecium
Bacteraemia
535
b
E. faecium E. faecium
E. faecium
518 vanA
a b
529
E. faecium
Susceptible vanA
E. faecium
E. faecium
Intra-abdominal
SSTI
499
E. faecium
510
E. faecium
a b
E. faecium
vanA
497
E. faecium
Susceptible
Bacteraemia
443
E. faecium
Bacteraemia
429
a
E. faecium vanA
E. faecium
E. faecium
SSTI
a
E. faecium
416
Susceptible
a
417
E. faecium
Susceptible
A
E. faecium
E. faecium
E. faecium
E. faecium
E. faecium
E. faecium
Bacteraemia
E. faecium
Susceptible
A
A
A
A
A
A
400
Intra-abdominal
393
E. faecalis
Susceptible
Susceptible
Susceptible
Susceptible
Susceptible
Susceptible
E. faecium
Bacteraemia
383
E. faecalis
E. faecalis
E. faecalis
E. faecalis
E. faecalis
E. faecalis
395
SSTI
Intra-abdominal
Intra-abdominal
355
373
Intra-abdominal
329
375
Bacteraemia
Bacteraemia
291
264
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanA
vanB
vanB
vanB
vanA
vanA
vanA
b
b
b
b
b
a
a
b
a
a
a
a
a
a
a
a
b
a
a
203
203
203
203
203
117
117
203
117
117
117
117
117
117
117
117
203
117
117
CRBSI
SSTI
Intra-abdominal
SSTI
Bacteraemia
SSTI
Bacteraemia
Urinary tract
E. faecium
Bacteraemia
28
251
252
E
Urinary tract Bacteraemia
vanB
15
Intra-abdominal
Urinary tract Bacteraemia
E. faecalis
d
19
6
117 117
786
232
vanB
D
A
a a
e
CRBSI
E. faecium
vanB
vanA
vanA vanA
vanA
Type of infection
E. faecium
E. faecium
E. faecium
E. faecium
E. faecium
E. faecalis
E. faecium
E. faecalis
E. faecalis
E. faecalis
E. faecium
E. faecium
E. faecalis E. faecalis
Species
vanA
vanA
Susceptible
vanA
vanA
Susceptible
Susceptible
Susceptible
Susceptible
Susceptible
Susceptible
Susceptible
Susceptible Susceptible
Vancomycin resistance
Enterococcal infection during ICU stay (N: 15)
226
211
E. faecalis E. faecalis
D
186
209
Susceptible
E. faecium E. faecium
E. faecalis
e A
171 178
Urinary tract
vanA Susceptible
E. faecium
E. faecium
E. faecalis
Bacteraemia
Urinary tract
ST
92
PFGE type
121
Vancomycin resistance
Species
PFGE type
Species
Type of infection
Vancomycin resistance
Colonization by VRE upon ICU admission (N: 28)
Enterococcal infection before ICU admission (N: 22)
Patient
Table 2 Characterization of enterococci isolated from colonized and infected patients one month before and during ICU hospitalization
b
b
a
a
a
A
d
A
E
A
d
a
A A
PFGE type
1016 M. Papadimitriou-Olivgeris et al.
Bacteraemia 117
17 d
a vanA E. faecium
E. faecium 636
Pneumonia
Bacteraemia
E. faecium
E. faecium
Susceptible
Susceptible
a
d 635
622
vanA
Vancomycin resistance PFGE type Vancomycin resistance Species
PFGE Pulsed Field Gel Electrophoresis, ST Sequence Types identified by MLST, SSTI Skin and Soft Tissue Infection, CRBSI Catheter-Related Bloodstream Infection
a E. faecium
Susceptible
Vancomycin resistance Species Type of infection ST Species Type of infection
PFGE type
Colonization by VRE upon ICU admission (N: 28) Enterococcal infection before ICU admission (N: 22) Patient
Table 2 continued
Enterococcal infection during ICU stay (N: 15)
PFGE type
Risk factors for enterococcal infection and colonization
1017
and 5 by E. faecalis). Among the 71 VRE-colonized patients, 12 (16.9 %) were admitted directly to the ICU, without previous hospitalization. Out of 250 patients that had a longer than seven days stay in ICU (24.0 ± 23.1 days), 36 (14.4 %) became colonized during ICU hospitalization (34 by E. faecium and 2 by E. faecalis). No difference was observed between the two ICUs. Overall, 100 isolates were E. faecium and seven E. faecalis (Table 1). Eighty-six VRE (83 E. faecium and three E. faecalis) were vanA-positive, while, the remaining 21 (17 E. faecium and four E. faecalis) carried the vanB gene. The interventions introduced to prevent VRE transmission upon detection were cohort nursing of colonized or infected patients when possible, contact precautions, reinforcement of hand hygiene policies, and refreshing courses of the staff for compliance with infection control practices. The susceptibility rates among VRE isolates were: penicillin (0 %), erythromycin (0 %), ampicillin (6 %), high-level gentamicin (20 %), vancomycin (8 %), teicoplanin (20 %), chloramphenicol (85 %), linezolid (93 %), daptomycin (100 %) and tigecycline (100 %). Antibiotic resistance phenotypes revealed that vanA-positive isolates expressed resistance to most tested compounds. Nine out of the 21 vanB-positive strains (five E. faecium and four E. faecalis) showed MICs to vancomycin ranging 1.5–4 mg/ L, below the resistance limit of the antibiotic (B4 mg/L), whereas teicoplanin MICs ranged 0.125–1.5 mg/L [10]. E. faecium were classified into nine pulsotypes and six sequence types (STs), (Table 1), predominantly ST117 (58 strains). Four of these STs (ST117, ST17, ST203, ST125) including 90 strains, belonged to a common Clonal Complex (CC17). E. faecalis were classified into four pulsotypes and four STs (ST6, ST41, ST19, ST28). The number of patients admitted in the ICU with VRE carriage and those colonized during ICU hospitalization per month are shown in Fig. 1. No seasonal variation was observed concerning VRE colonization or enterococcal infections. When comparing the first 15 months of the study period (period A: January 2010–March 2011) and the last nine (period B: June 2011–February 2012) a statistically significant decrease in the colonization during ICU was observed during period B (18.0 vs. 5.6 %, P: 0.016), even though no difference was observed in the incidence of VRE-colonized patients admitted in the ICU (13.5 vs. 16.1 %, P: 0.484). Altogether, 37 strains recovered from 36 patients (21 E. faecium and 16 E. faecalis) were recognized to cause infection, either the previous month before ICU entrance (22 strains), or during their stay (15 strains). Twenty-eight patients were VRE colonized upon ICU admission. PFGE type comparison was performed among colonizing and infecting enterococci (Table 2). Seventeen out of 22 patients being infected during the last month before ICU
123
1018
M. Papadimitriou-Olivgeris et al.
Table 3 Univariate analysis of risk factors for Vancomycin-Resistant Enterococcus (VRE) enteric colonization upon ICU admission and during hospitalization Characteristics
Upon admission Colonized patients (n = 71)
During hospitalization Non colonized patients (n = 426)
P
Colonized patients (n = 36)
Non colonized patients (n = 214)
P
Demographics Age (years)
61.5 ± 17.0
55.4 ± 19.3
0.015
60.6 ± 17.6
53.4 ± 19.9
0.037
Male gender
46 (64.8 %)
286 (67.1 %)
0.685
20 (55.6 %)
149 (69.6 %)
0.123
Co-morbidities (number)
\0.001
1.6 ± 1.3
0.8 ± 1.0
Insulin-dependent diabetes mellitus
9 (12.7 %)
12 (2.8 %)
Chronic obstructive pulmonary disease
21 (29.6 %)
46 (10.8 %)
Chronic heart failure
11 (15.5 %)
34 (8.0 %)
0.070
6 (16.7 %)
11 (5.1 %)
0.022
Chronic renal failure
7 (9.9 %)
17 (4.0 %)
0.121
4 (11.1 %)
2 (0.9 %)
0.004
Malignancy
19 (26.8 %)
49 (11.5 %)
0.001
3 (8.3 %)
15 (7.0 %)
0.730
Chemotherapy Obesity
5 (7.0 %) 19 (26.8 %)
8 (1.9 %) 93 (21.8 %)
0.026 0.360
1 (2.8 %) 13 (36.1 %)
1 (0.5 %) 52 (24.3 %)
0.286 0.152
Cortisone use
14 (19.7 %)
25 (5.9 %)
3 (8.3 %)
12 (5.6 %)
0.460
0.001 \0.001
\0.001
1.3 ± 1.2
0.6 ± 0.8
0.002
5 (13.9 %)
4 (1.9 %)
0.004
11 (30.6 %)
20 (9.3 %)
0.001
Admission data APACHE II score upon admission
16.4 ± 7.18
14.7 ± 7.1
0.080
17.6 ± 9.2
15.3 ± 6.3
0.178
SAPS II score upon admission
40.1 ± 12.4
35.5 ± 12.8
0.002
39.4 ± 12.9
36.2 ± 12.1
0.141
Prior hospitalization
59 (83.1 %)
144 (33.8 %)
–
–
–
\0.001
Duration (days)
28.9 ± 46.1
3.4 ± 8.7
\0.001
–
–
–
Prior ICU stay
23 (32.4 %)
22 (5.2 %)
\0.001
–
–
–
Admission at ICU Aa
62 (87.3 %)
354 (83.1 %)
0.487
27 (75.0 %)
172 (80.4 %)
0.503
Emergency surgery
22 (31.0 %)
128 (30.0 %)
0.889
9 (25.0 %)
77 (36.0 %)
0.256
Abdominal surgery
31 (43.7 %)
83 (19.5 %)
\0.001
10 (27.8 %)
37 (17.3 %)
0.165
Previous enterococcal infection
17 (23.9 %)
5 (11.7)
\0.001
–
–
–
3.8 ± 3.6
0.8 ± 1.5
\0.001
4.5 ± 2.2
3.6 ± 1.7
0.026
Number of antibiotics administered Carbapenems
42 (59.2 %)
49 (11.5 %)
\0.001
29 (80.6 %)
178 (83.2 %)
Quinolones third- or fourth-generation cephalosporins
34 (47.9 %) 17 (23.9 %)
43 (10.1 %) 42 (9.9 %)
\0.001 0.002
14 (38.9 %) 11 (30.6 %)
28 (13.1 %) 38 (17.8 %)
\0.001 0.109 0.568
1.000
Beta-lactam/lactamase inhibitors
27 (38.0 %)
39 (9.2 %)
\0.001
14 (38.9 %)
70 (32.7 %)
Colistinb
12 (16.9 %)
6 (1.4 %)
\0.001
14 (38.9 %)
82 (32.7 %)
1.000
Aminoglycosides
13 (18.3 %)
11 (2.6 %)
\0.001
12 (33.3 %)
74 (38.3 %)
1.000
Glycopeptides
47 (66.2 %)
70 (16.4 %)
\0.001
33 (91.7 %)
173 (80.0 %)
0.259
Metronidazole
23 (32.4.2 %)
24 (5.6 %)
\0.001
6 (16.7 %)
30 (14.0 %)
0.616
Tigecycline
6 (8.5 %)
3 (0.7 %)
\0.001
5 (13.9 %)
23 (10.7 %)
0.570
Linezolid
16 (22.5 %)
12 (2.8 %)
\0.001
8 (22.2 %)
51 (23.8 %)
1.000
–
–
28 (77.8 %)
129 (60.3 %)
0.061
ICU procedures Tracheotomy Number of catheters
c
–
–
–
–
1.1 ± 1.0
1.0 ± 1.2
0.005
Abdominal catheter and/or colostomy
–
–
–
8 (22.2 %)
35 (16.4 %)
0.473
Cortisone administration
–
–
–
21 (58.3 %)
110 (51.4 %)
0.477
Parenteral nutrition
–
–
–
20 (55.6 %)
90 (42.1 %)
0.277
Enteral nutrition
–
–
–
27 (75.0 %)
160 (74.8 %)
1.000
–
–
–
25.7 ± 34.8
17.2 ± 13.3
VRE colonization pressure Days at risk
123
\0.001
Risk factors for enterococcal infection and colonization
1019
Table 3 continued Characteristics
Upon admission
During hospitalization
Colonized patients (n = 71)
Non colonized patients (n = 426)
Number of VRE-positive patients in nearby beds per day
–
–
Number of VRE-positive patients in ICU per day
–
–
P
Colonized patients (n = 36)
Non colonized patients (n = 214)
P
–
0.5 ± 0.4
0.1 ± 0.2
\0.001
–
1.1 ± 0.8
0.8 ± 0.8
0.069
Data are number (%) of patients or mean ± SD APACHE II Acute Physiology and Chronic Health Evaluation II, SAPS II Simplified Acute Physiology Score II a
ICU A: ICU of the University General Hospital of Patras
b
Only intravenous administration of colistin is included
c
All patients after ICU admission were intubated, mechanically ventilated and were continuously monitored with a central venous catheter, an arterial catheter and a urinary catheter. Number of catheters does not include the aforementioned catheters
hospitalization (VRE and/or VSE) were found to be colonized upon ICU admission with VRE, six of them with the same strain (patients numbered 92, 417, 443, 510, 535 and 629). Five more patients were colonized with vanA/or vanB-positive strains of the same PFGE type as the susceptible infecting ones (patients numbered 209, 393, 400, 499 and 636) (Table 2). Among the 15 infected patients during their ICU stay, 12 were previously VRE colonized. Four of them were infected by the colonizing VRE strains (patients numbered 400, 416, 518 and 529), while four more by VSE of the same PFGE type (patients numbered 211, 232, 497 and 635, Table 2). Risk factors for VRE-colonization upon ICU admission identified by logistic regression analysis were: duration of previous hospitalization (P: 0.001; OR: 1.0; CI 1.0–1.1), usage of glycopeptides (P: 0.007; OR: 2.9; CI 1.2–6.4), chronic heart failure (P: 0.012; OR: 3.0; CI 1.3–7.0), malignancy (P: 0.001; OR: 3.9; CI 1.8–8.3), insulindependent diabetes mellitus (P: 0.001; OR: 6.6; CI 2.2–20.0) and previous enterococcal infection including VSE (P \ 0.001; OR: 10.2; CI 3.2–32.7). To determine the risk factors for VRE colonization during ICU stay, 250 patients who were negative for VRE faecal colonization upon ICU admission and were hospitalized for at least 7 days were included. The average time for colonization was 25 days. Risk factors identified by multivariate analysis were: quinolone administration (P: 0.042; OR: 3.1; CI 1.0–9.1), chronic obstructive pulmonary disease (P: 0.013; OR: 4.2; CI 1.4–13.1), chronic renal failure requiring dialysis (P: 0.047; OR: 10.2; CI 1.0–100.3), and number of VRE-positive patients in nearby beds per day (P \ 0.001; OR: 87.9; CI 21.6–357.3). Table 3 depicts the univariate analyses of risk factors for VRE-colonization upon ICU admission and during stay. In order to evaluate the risk factors for enterococcal infection (VRE and VSE cases), all patients with ICU stay of at least 7 days and more were included (Table 4).
Multivariate analysis showed that administration of thirdor fourth- generation cephalosporins (P: 0.020; OR: 4.8; CI 1.3–17.8), cortisone use before ICU admission (P: 0.014; OR: 7.4; CI 1.5–36.7) and VRE colonization (P: \ 0.001; OR: 14.3; CI 3.4–60.3) were statistically significant risk factors, whereas, enteral nutrition was identified as a protective factor (P: 0.001; OR: 0.10; CI 0.03–0.42).
Discussion In our setting, enterococci constitute the sixth of the most common pathogens accounting for 7.5 % of total infections during the study period. VRE remain important nosocomial pathogens and pose a great threat due to limited therapeutic options [1, 2]. In the present study, a high percentage (14.3 %) of patients were VRE carriers upon ICU admission, while, an additional 14.4 % became colonized during their ICU stay. This phenomenon reflects a high patient to patient transmission, despite the cohorting of colonized patients and implementation of infection control practices. These rates are higher than those reported from studies in other countries, but lower from those of previous Greek publications, reflecting differences in the epidemiology of VRE that may be related to patients’ characteristics prior to ICU admission (surgical wards, transplantation units, etc.) [2, 5, 7, 8, 13]. The fact that among 203 patients admitted directly to the ICU, without previous hospitalization, 12 (5.9 %) were already colonized with VRE, suggests the dissemination of VRE in the community, as published in a study from Czech Republic where 0.4 % of people living in the community were colonized. [14] E. faecium predominate among colonizing isolates, as reported elsewhere [13]. Clonal spread was also observed among colonized patients, a fact reinforcing the risk of patient to patient transmission. Even though ST17 and ST203 were previously identified among E. faecium in our Hospital, new clones have
123
1020 Table 4 Univariate and multivariate analyses of risk factors for enterococcal infection, including vancomycin-resistant and susceptible cases, during ICU stay
M. Papadimitriou-Olivgeris et al.
Characteristics
Univariate analysis
Multivariate analysis
Enterococcal infection (n = 15)
No infection (n = 269)
18.9 ± 23.2
16.7 ± 15.7
0.095
65.7 ± 12.5 8 (53.3 %)
55.1 ± 19.95 182 (67.7 %)
0.036 0.268
1.3 ± 1.2
0.8 ± 1.1
0.016
Insulin-dependent diabetes mellitus
2 (13.3 %)
10 (3.7 %)
0.126
Chronic obstructive pulmonary disease
6 (40.0 %)
39 (14.5 %)
0.019
Chronic heart failure
1 (6.7 %)
20 (7.4 %)
1.000
Chronic renal failure
0 (0.0 %)
9 (3.3 %)
1.000
Malignancy Chemotherapy
2 (13.3 %) 0 (0.0 %)
20 (7.4 %) 2 (0.7 %)
0.327 1.000
Days at risk
P
P
OR (95 % CI)
Demographics Age (years) Male gender Co-morbidities (number)
Obesity
6 (40.0 %)
69 (25.7 %)
0.235
Cortisone use
4 (26.6 %)
18 (6.7 %)
0.021
11 (73.3 %)
221 (82.2 %)
0.489
0.014
7.4 (1.5–36.7)
0.020
4.8 (1.3–17.8)
Admission data Admission at ICU Aa APACHE II score upon admission
21.6 ± 8.6
15.5 ± 6.8
0.011
SAPS II score upon admission
44.7 ± 9.5
36.4 ± 12.2
0.018
Prior emergency surgery
6 (40.0 %)
89 (33.1 %)
0.583
Prior abdominal surgery
7 (46.7 %)
55 (20.4 %)
0.025
3.8 ± 2.7
3.6 ± 1.7
0.984
10 (66.7 %)
227 (84.4 %)
0.072
Number of antibiotics administered Data are number (%) of patients or mean ± SD
Carbapenems Quinolones
2 (13.3 %)
43 (13.1 %)
1.000
VRE Vancomycin-Resistant Enterococci, OR adjusted odds ratio, CI confidence interval, APACHE II Acute Physiology and Chronic Health Evaluation II, SAPS II Simplified Acute Physiology Score II
third- or fourth-generation cephalosporins
7 (46.7 %)
49 (16.0 %)
0.014
Beta-lactam/lactamase inhibitors
4 (26.7 %)
86 (32.0 %)
0.782
Colistinb
6 (40.0 %)
93 (34.6 %)
0.782
Aminoglycosides
3(20.0 %)
85 (31.6 %)
0.407
Glycopeptides
9 (60.0 %)
218 (81.0 %)
0.088
Metronidazole
2 (13.3 %)
37 (13.8 %)
1.000
Tigecycline
3 (20.0 %)
19 (7.1 %)
0.089
a
ICU A: ICU of the University General Hospital of Patras
b
Only intravenous administration of colistin is included
c
All patients after ICU admission were intubated, mechanically ventilated and were continuously monitored with a central venous catheter, an arterial catheter and a urinary catheter. Number of catheters does not include the aforementioned catheters
Linezolid
5 (33.3 %)
58 (21.6 %)
0.336
Mean antibiotic use per day
2.1 ± 1.2
2.5 ± 0.8
0.174
ICU procedures Tracheotomy
8 (53.3 %)
180 (67.0 %)
0.274
Number of cathetersc
0.9 ± 2.2
1.0 ± 1.2
0.120
Cortisone administration
6 (40.0 %)
140 (52.0 %)
0.432
Parenteral nutrition
5 (33.3 %)
117 (43.5 %)
0.594
6 (40.0 %) 12 (80.0 %)
198 (73.6 %) 0.014 0.001 58 (21.6 %) \0.001 \0.001
Enteral nutrition VRE colonization
evolved, such as ST117, ST125, ST226 and the newly identified ST786 [4]. vanA-positive strains express a more resistant phenotype, while, among 21 vanB-positive strains nine showed MIC to vancomycin within the susceptible range (MIC B4 lg/ml). Results indicate that vanB-positive
123
0.1 (0.03–0.42) 14.3 (3.4–60.3)
enterococci may be undetected and disseminate if only phenotypic methods are used, as previously suggested [15]. Statistical analysis showed that insulin-dependent diabetes mellitus, chronic heart failure, malignancy, chronic obstructive pulmonary disease, chronic renal failure,
Risk factors for enterococcal infection and colonization
previous enterococcal infection (VRE and/or VSE cases) and antibiotic administration (glycopeptides and carbapenems) constitute predisposing factors for VRE colonization upon ICU admission or extended stay. Cortisone usage was identified as a risk factor for enterococcal infection during ICU stay. These findings underline the importance of comorbidities in promoting colonization or infections by VRE and multidrug-resistant pathogens in general, as depicted by previous studies [6, 16–18]. It is already reported that previous vancomycin use, third generation cephalosporins and quinolone therapy, are predisposing risk factors for VRE colonization or infection [6, 7, 13, 16, 17, 19–22]. Empirical therapy in our setting consisted of piperacillin/tazobactam or a carbapenem for broad-spectrum combined with a glycopeptide for antiGram positive coverage. Length of hospitalization prior to ICU admission has also been found to be an independent risk factor for VRE carriage [7, 23]. Longer hospitalization means high probability of receiving antibiotics such as cephalosporins, quinolones and glycopeptides and also increased risk of being hospitalized near a patient who was already colonized by VRE. In the present study, it is shown that previous enterococcal infection, even with VSE strains, predisposes to VRE colonization. The average number of VRE-positive patients in nearby beds per day was identified as independent risk factors for VRE colonization during ICU stay, as previously reported [19, 20, 22, 23]. This suggests that colonization pressure is the most important factor in promoting VRE dissemination in the ICU setting, whereas, clonal relationship of isolated strains in our study verifies this parameter. Since cohort nursing was not always applicable, reinforcement of hand hygiene policies, education of the staff by infection control personnel combined sometimes with more stringent measures, are actions that should be implemented. This was verified with a decrease of VRE colonization during ICU stay comparing period B with period A in our setting (Fig. 1). Although 107 patients were VRE-colonized in total, only four (3.7 %) developed VRE infection during ICU hospitalization (with strains of the same clone), which is within the range (0–45 %) already published [23, 24]. It is reported that VRE enteric colonization predicts VRE bloodstream infection [25]. Eleven patients in the present study developed infection by vancomycin-susceptible strains. It is important to note that VRE colonization was proved as an important factor for developing infection also by vancomycin-susceptible strains, in four cases by strains of the same PFGE type. This finding suggests that horizontal gene transfer combined with clonal spread occurred during hospitalization. Polyclonality among VRE colonizing strains verifies the presence and evolution of these strains in our setting for a
1021
long time. Identification of vancomycin-susceptible strains of the same PFGE type with VRE, suggests a continuing horizontal spread of van genes. Besides multiple risk factors for VRE colonization and infection during ICU hospitalization, to the best of our knowledge, this is the first study showing that previous enterococcal infection, even with VSE strains, constitutes a risk factor for VRE colonization and vice versa. This fact indicates that all patients who develop enterococcal infection during prior hospitalization must be isolated and screened for VRE faecal colonization. Acknowledgments This research was supported by funding of the Department of Microbiology, School of Medicine, University of Patras, Greece. Conflict of interest vant to this article.
All authors report no conflicts of interest rele-
Ethical approval The study was approved by the University Hospital Ethics Committee that waived the need for informed consent (HEC No: 571).
References 1. Cattoir V, Leclercq R. Twenty-five years of shared life with vancomycin-resistant enterococci: is it time to divorce? J Antimicrob Chemother. 2013;68:731–42. 2. European Centre for Disease Prevention and Control (ECDC). Annual Epidemiological Report, Reporting on 2011 surveillance data and 2012 epidemic intelligence data. Stockholm: ECDC; 2013. p. 2013. 3. Werner G, Coque TM, Hammerum AM, et al. Emergence and spread of vancomycin resistance among enterococci in Europe. Euro Surveill. 2008;13. 4. Damani A, Klapsa D, Panopoulou M, et al. A newly described vancomycin-resistant ST412 Enterococcus faecium predominant in Greek hospitals. Eur J Clin Microbiol Infect Dis. 2010;29:329–31. 5. Sofianou D, Pournaras S, Giosi M, et al. Substantially increased faecal carriage of vancomycin-resistant enterococci in a tertiary Greek hospital after a 4 year time interval. J Antimicrob Chemother. 2004;54:251–4. 6. Kim YJ, Kim SI, Kim YR, et al. Risk factors for vancomycinresistant enterococci infection and mortality in colonized patients on intensive care unit admission. Am J Infect Control. 2012;40:1018–9. 7. Ostrowsky BE, Venkataraman L, D’Agata EM, et al. Vancomycin-resistant enterococci in intensive care units: high frequency of stool carriage during a non-outbreak period. Arch Intern Med. 1999;159:1467–72. 8. Song JY, Cheong HJ, Jo YM, et al. Vancomycin-resistant Enterococcus colonization before admission to the intensive care unit: a clinico-epidemiologic analysis. Am J Infect Control. 2009;37:734–40. 9. Centers for Disease Control and Prevention. Catheter associated urinary tract infection (CAUTI) event. January 2012. http://www. cdc.gov/nhsn/pdfs/pscManual/7pscCAUTIcurrent.pdf. Accessed August, 2014. 10. Clinical and Laboratory Standards Institute. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, Approved standard M7-A7. 12th ed. Wayne: CLSI; 2011.
123
1022 11. Bell J, Turnidge J, Coombs G, O’Brien F. Emergence and epidemiology of vancomycin-resistant enterococci in Australia. Commun Dis Intell. 1998;22:249–52. 12. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol. 1995;33:2233–9. 13. Christidou A, Gikas A, Scoulica E, et al. Emergence of vancomycin-resistant enterococci in a tertiary hospital in Crete, Greece: a cluster of cases and prevalence study on intestinal colonisation. Clin Microbiol Infect. 2004;10:999–1005. 14. Kolar M, Pantucek R, Vagnerova I, et al. Prevalence of vancomycin-resistant enterococci in hospitalized patients and those living in the community in the Czech Republic. New Microbiol. 2006;29:121–5. 15. Endtz HP, Van Den Braak N, Van Belkum A, et al. Comparison of eight methods to detect vancomycin resistance in enterococci. J Clin Microbiol. 1998;36:592–4. 16. Beltrami EM, Singer DA, Fish L, et al. Risk factors for acquisition of vancomycin-resistant enterococci among patients on a renal ward during a community hospital outbreak. Am J Infect Control. 2000;28:282–5. 17. Batistao DW, Gontijo-Filho PP, Conceicao N, et al. Risk factors for vancomycin-resistant enterococci colonisation in critically ill patients. Mem Inst Oswaldo Cruz. 2012;107:57–63. 18. Goetz AM, Rihs JD, Wagener MM, et al. Infection and colonization with vancomycin-resistant Enterococcus faecium in an acute care Veterans Affairs Medical Center: a 2-year survey. Am J Infect Control. 1998;26:558–62.
123
M. Papadimitriou-Olivgeris et al. 19. MacIntyre CR, Empson M, Boardman C, et al. Risk factors for colonization with vancomycin-resistant enterococci in a Melbourne hospital. Infect Control Hosp Epidemiol. 2001;22:624–9. 20. Bonten MJ, Slaughter S, Ambergen AW, et al. The role of ‘‘colonization pressure’’ in the spread of vancomycin-resistant enterococci: an important infection control variable. Arch Intern Med. 1998;158:1127–32. 21. Sakka V, Tsiodras S, Galani L, et al. Risk-factors and predictors of mortality in patients colonised with vancomycin-resistant enterococci. Clin Microbiol Infect. 2008;14:14–21. 22. Ajao AO, Harris AD, Roghmann MC, et al. Systematic review of measurement and adjustment for colonization pressure in studies of methicillin-resistant Staphylococcus aureus, vancomycinresistant enterococci, and Clostridium difficile acquisition. Infect Control Hosp Epidemiol. 2011;32:481–9. 23. Ziakas PD, Thapa R, Rice LB, et al. Trends and significance of VRE colonization in the ICU: a meta-analysis of published studies. PLoS One. 2013;8:e75658. doi: 10.1371/journal.pone. 0075658. eCollection 2013 24. Pan SC, Wang JT, Chen YC, et al. Incidence of and risk factors for infection or colonization of vancomycin-resistant enterococci in patients in the intensive care unit. PLoS One. 2012;7:e47297. doi:10.1371/journal.pone.0047297. 25. Liss BJ, Vehreschild JJ, Cornely OA, et al. Intestinal colonisation and blood stream infections due to vancomycin-resistant enterococci (VRE) and extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBLE) in patients with haematological and oncological malignancies. Infection. 2012;40:613–9.
