A European perspective on nosocomial urinary tract infections II. Report on incidence, clinical characteristics and outcome (ESGNI−004 study)

Authors

  • E. Bouza,

  • R. San Juan,

  • P. Muñoz,

  • A. Voss,

  • J. Kluytmans,

  • the Co-operative Group of the European Study Group on Nosocomial Infections (ESGNI)


  • Corresponding author and reprint requests: E. Bouza, Servicio de Microbiología Clínica y Enfermedades Infecciosas-VIH, Hospital General Universitario ‘Gregorio Marañón’, Dr Esquerdo 46, 28007 Madrid, Spain

    Tel: +34 915868453

    Fax: +34 915044906

    E-mail: ebouza@microb.net

Abstract

Objectives  To estimate the incidence of nosocomially acquired urinary tract infections (NAUTI) in Europe and provide information on the clinical characteristics, underlying conditions, etiology, management and outcome of patients.

Materials and methods  We collected clinical information from NAUTI patients with a microbiology report on the named study day.

Results  A total of 141 hospitals from 25 European countries participated in the study. Written institutional bladder catheter guidelines were in place in 90.3% of EU hospitals and 55% of non-EU hospitals (P < 0.05). The total number of new NAUTI episodes on the day of the study was 298, representing an incidence of 3.55 episodes/1000 patient-days and an estimated prevalence of 10.65/1000. The five most commonly isolated micro-organisms were Escherichia coli, Enterococcus sp., Candida sp., Klebsiella sp. and Pseudomonas aeruginosa. Patients from non-EU countries were younger, with more severe underlying diseases with a higher incidence of obstructive uropathy/lithiasis. Overall, 22.8% of patients had no ‘classic’ UTI-predisposing factors. Catheter-associated UTI (CAUTI) was present in 187 patients (62.8%). A closed drainage system was used in only 78.5% of catheterised patients. The indication for bladder catheterisation was not considered adequate in 7.6% of cases and continuation of bladder catheterisation was considered unnecessary in 31.3%. Opening of the closed drainage system was the most frequent major error in catheter management (16.8%). Antimicrobial treatment was not considered adequate in 19.8% of all cases.

Conclusions  The incidence of NAUTI in a large European population is 3.55/1000 patient-days. There is clearly room for improvement in the area of bladder catheterisation, catheter care and medical management of NAUTI. We recommend that European authorities draw up and implement practical and specific guidelines to reduce the incidence of this infection.

Introduction

In a preceding study [European Study Group on Nosocomial Infections 003 (ESGNI-003)], we assessed the microbiology workload, diagnostic criteria, etiology and antimicrobial susceptibility caused by urinary tract infection (UTI) in patients hospitalised in 228 European institutions [1]. The aim of the present study (ESGNI-004) was to collect bedside information from patients with nosocomially acquired urinary tract infections (NAUTI), and to compare the situation between countries of the European Union (EU) and countries outside the EU (non-EU). We obtained information from 141 hospitals in 25 European countries. Our aim was to obtain baseline data on a broad basis and to establish opportunities for intervention and improvement.

Materials and methods

ESGNI-004 was a 1-day (29 February 2000) incidence study linked to ESGNI-003. Cases with microbiologically proven NAUTI on the study day had a bedside evaluation and follow-up for a maximum of 1 month. Data collected from each patient with significant bacteriuria or funguria included: age, sex, weighted index of co-morbidity, classification of the underlying disease according to the McCabe and Jackson groups, etiology of the episode, presence of fever, severity of illness according to the sepsis score, and predisposing conditions for infection. In patients with a urinary catheter (UC), we requested the following information: type of catheter, length of time the catheter had been in place on the study day, catheter indication, use of closed drainage systems, use of urinometer or silver-coated catheter, and indication, insertion and care adequacy according to the physician's opinion and institutional guidelines. Antimicrobial treatment was classified as adequate or inadequate, we registered the number of days of antimicrobial administration for the NAUTI episode and finally, patients were followed up for 1 month until discharge. Deaths were classified by the observer as attributable or not attributable to the UTI.

Definitions used

UTI episode. Episodes of significant bacteriuria [≥105 colony-forming units (CFU)/mL] or funguria (≥103 CFU/mL).

Nosocomially acquired UTI. A nosocomial episode is considered to be any UTI infection beginning at least 48 h after admission.

Polymicrobial UTI. Polymicrobial UTI is defined as isolation of two micro-organisms during a single UTI episode.

Associated bacteremia. Presence of positive blood cultures and micro-organism isolated not more than 3 days apart from the urinary isolate with the same micro-organism.

Severity of illness [2].Sepsis: Systemic response manifested by two or more of the following conditions as a result of infection: (a) temperature >38 °C or <36 °C; (b) heart rate >90 beats/min; (c) respiratory rate >20 breaths/min or PaCO2 <32 mmHg; (d) white blood cell count >12 000 cells/mm3, <4000 cells/mm3, or >10% immature (band) forms.

Severe sepsis: Sepsis associated with organ dysfunction, hypoperfusion, or hypotension. Hypoperfusion and perfusion abnormalities may include, but are not limited to, lactic acidosis, oliguria, or an acute alteration in mental status.

Septic shock: Sepsis with hypotension, despite adequate fluid resuscitation, along with the presence of perfusion abnormalities that may include, but are not limited to, lactic acidosis, oliguria, or an acute alteration in mental status. Patients who are on inotropic or vasopressor agents may not be hypotensive at the time perfusion abnormalities are measured.

Multiorgan failure: Failure of three or more organ systems during at least a 24-h period, as a consequence of NAUTI.

UTI-associated factors. (a) Urinary catheter, intravenous catheter, if present at the time of infection; (b) Presence of obstructive uropathy/lithiasis; (c) Urinary tract anatomic abnormalities; (d) Previous UTI; (e) Renal transplantation; (f) Fecal incontinence; (g) Pregnancy; (h) Uterine prolapse; (i) Surgery if it was during hospital admittance; (j) Corticosteroids if taken at a dose of 20 mg or more of prednisone daily (or equivalent) for at least 2 weeks, or 30 mg or more of prednisone daily for at least 1 week, before the urine culture; (k) Previous antimicrobials if the patient received any oral or parenteral antibiotic in the 15 days previous to the UTI episode; (l) Urological intervention, including prostatectomy if performed during the week previous to the UTI episode, and; (m) Other invasive procedures

Adequacy of treatment. Treatment was considered adequate if the patient received one or more antibiotics active in vitro against the micro-organisms isolated, when indicated.

Days of treatment. Only the number of days of adequate treatment have been considered.

Death attributable to UTI. Death is considered as attributable to NAUTI if it occurs during the phase of active infection or while the patient is undergoing antibiotic treatment.

McCabe and Jackson groups [3], and the Charlson weighted index [4] were used as comorbidity indexes.

Data analysis

We expressed continuous variables as the mean and standard deviation (SD) when normally distributed, or as the median and interquartile range (IQR) if their distribution was skewed, and discrete variables as percentages. We used Student's unpaired t-test to compare continuous variables, the Mann–Whitney U-test to compare continuous variables not normally distributed, and the χ2 or Fisher exact test to compare proportions. All statistical tests were two-tailed.

Results

A total of 141 hospitals from 25 countries (12 EU countries and 13 non-EU countries) participated in the ESGNI-004 study. (Table 1).

Table 1.  Participating hospitals
CountryNo. of hospitals
participating (%)
CountryNo. of hospitals
participating (%)
Austria4 (2.8)Lithuania1 (0.7)
Belgium10 (7.1)Netherlands2 (1.4)
Croatia3 (2.1)Poland6 (4.3)
Czech Republic6 (4.3)Portugal4 (2.8)
Denmark1 (0.7)Romania2 (1.4)
Finland1 (0.7)Russia2 (1.4)
France11 (7.8)Slovak Republic3 (2.1)
Germany10 (7.1)Slovenia4 (2.8)
Greece8 (5.7)Spain34 (24.1)
Hungary1 (0.7)Switzerland5 (3.5)
Israel1 (0.7)Turkey8 (5.7)
Italy11 (7.8)United Kingdom2 (1.4)
Latvia1 (0.7)  
TOTAL141  

Characteristics of participating institutions

We obtained this information from 141 hospitals of different sizes serving an estimated population of 99 759 000 (41.4% had <500 beds, 33.6% had 501–1000 beds, and 25% >1000 beds), 98 from EU countries and 43 from non-EU countries. Overall, the total number of estimated admissions in these institutions during 1999 was 4 410 500. With regard to administration and activity, 70.7% were teaching hospitals, 83.7% were public, 5.9% private and 10.4% both public and private. No statistically significant differences were noted between EU and non-EU institutions except for the percentage of hospitals with written institutional bladder catheter guidelines, 79.7% overall, 90.3% in EU hospitals and 55% in non-EU hospitals (P < 0.0001). On the study day, 83 962 beds were occupied (80% occupancy).

Incidence and estimated prevalence

The total number of NAUTI episodes confirmed by the microbiology laboratory on the study day was 298 (198 from EU countries, 100 from non-EU countries), giving an incidence of 3.55 episodes/1000 patient-days (3.43 in EU countries, 3.82 in non-EU countries; P > 0.05). Theoretically, a prevalence rate can be estimated from the incidence density value, which is proportional to the mean duration of the measured disease [5]. An accurate method for estimating prevalence rate from incidence rate has been revised recently [6], but depends on data that we cannot obtain from this study, such as the mean length of hospitalisation of patients who acquire one or more nosocomial infections (NI), the mean length of hospitalisation for all patients and the mean interval between admission and the onset of the first NI. Thus, if we estimate a duration of 3 days of UTI symptoms [7] we obtain a rough equivalent NAUTI prevalence value of 10.65 episodes/1000.

Etiology

The number of micro-organisms isolated from NAUTI episodes was 340. The etiology of the episodes is summarised in Table 2. Gram-positive bacteria represented 21.2% of all isolates, whereas Gram-negative bacteria were 65.9%. Yeasts were 12.9%. The five most commonly isolated micro-organisms were Escherichia coli, Enterococcus sp., Candida sp., Klebsiella sp. and Pseudomonas aeruginosa. Staphylococcus aureus represented 3.5% of all isolates, an even higher rate than coagulase-negative staphylococci. Overall, 14.1% of the episodes were polymicrobial (13.1% in EU countries versus 16% in non-EU countries; P > 0.05). The type of micro-organism was comparable between both groups of hospitals (EU and non-EU) with the single exception of P. aeruginosa, which was isolated more frequently in non-EU countries (P < 0.05), and E. coli, which represents only 21.6% in non-EU countries (35.3% in EU countries, P < 0.05).

Table 2.  Micro-organisms isolated in urine (>1%)
EU countries
(n = 224)
 Non-EU countries
(n = 116)
 Total
(n = 340)
 
  • *

    P < 0.05.

  • CNS, coagulase-negative staphylococci.

Escherichia coli*79 (35.3%)Escherichia coli*25 (21.6%)Escherichia coli104 (30.6%)
Enterococcus sp.34 (15.2%)Pseudomonas aeruginosa*16 (13.8%)Enterococcus sp.48 (14.1%)
Candida sp.29 (12.9%)Candida sp.15 (12.9%)Candida sp.44 (12.9%)
Klebsiella sp.22 (9.8%)Enterococcus sp.14 (12.1%)Klebsiella sp.34 (10%)
Proteus sp.15 (6.7%)Klebsiella sp.12 (10.3%)Pseudomonas aeruginosa28 (8.2%)
Pseudomonas aeruginosa*12 (5.4%)Proteus sp.10 (8.6%)Proteus sp.25 (7.4%)
Enterobacter sp.10 (4.5%)Staphylococcus aureus5 (4.3%)Enterobacter sp.14 (4.1%)
Staphylococcus aureus7 (3.1%)Enterobacter sp.4 (3.4%)Staphylococcus aureus12 (3.5%)
Citrobacter sp.6 (2.7%)CNS4 (3.4%)CNS7 (2.1%)
Morganella sp.3 (1.3%)Acinetobacter sp.3 (2.6%)Citrobacter sp.9 (2.6%)
CNS3 (1.3%)Citrobacter sp.3 (2.6%)  

Age, sex and underlying conditions

We obtained complete clinical information from the protocol on 298 individuals with nosocomially acquired UTI which was active on the study day (Table 3). There were 135 males (45.3%) and 163 females (54.7%), with more females in EU countries (60.2% versus 44%, P < 0.05). Mean age was 62.71 (SD 25), and patients from non-EU countries were significantly younger (mean 52.54 versus 67.95, P < 0.05).

Table 3.  Patient characteristics
 EU countries
(n = 198)
Non-EU countries
(n = 100)
Total
(n = 298)
  • *

    P < 0.05.

Age (SD)*67.95 (20.6)52.54 (29.4)62.71 (25)
Sex*
 Female118 (60.2%)44 (44%)162 (54.7%)
 Male78 (39.8%)56 (56%)134 (45.3%)
Charlson index (SD)3.21 (2.7)2.89 (2.9)3.1 (2.8)
McCabe and Jackson groups
 1 (Nonfatal)54.1%55.2%54.5%
 2 (Ultimately fatal)39.3%28.1%35.5%
 3 (Rapidly fatal)*6.6%16.7%10%

According to the McCabe and Jackson classification, 10% of the patients had rapidly fatal diseases (16.7% non-EU countries, 6.6% in EU countries, P < 0.05), 35.5% had ultimately fatal diseases and 54.5% had diseases considered as non-fatal. Co-morbidity was rated according to Charlson's criteria and the mean index was 3.1 (SD 2.8).

Table 4shows the distribution of underlying diseases and the potential predisposing factors in the nosocomial UTI population. Overall, 22.8% of patients had no ‘classic’ UTI-predisposing factors and 62.8% had a urinary catheter (UC). When EU and non-EU countries were compared, there were significant differences in the overall presence of obstructive uropathy/lithiasis (either associated or as a single condition); this finding was more frequent in non-EU patients (P < 0.05).

Table 4.  Global considered related factors
 TotalNon-EU countriesEU countries
n%n%n%
  • a

    Obstructive uropathy, anatomic abnormalities, urinary device, urinary manipulation.

  • b

    As only predisposing factor (without urinary catheter).

  • *

    P < 0.05.

No ‘classic’a UTI-predisposing factors6822.8%1717%5125.8
Urinary catheter18762.8%6868%11960.1%
Obstructive uropathy/lithiasis*5518.5%2828%2713.6%
Obstructive uropathy/lithiasis*b155%1111%42%
Urinary tract anatomic abnormalities196.4%99%105.1%
Urinary tract anatomic abnormalitiesb51.7%0052.5%
Recent urological intervention196.4%1010%94.5%
Recent urological interventionb20.7%00%21%
Previous urinary tract infections7525.2%2222%5326.8%
Previous urinary tract infectionsb165.4%44%126.1%
Renal transplantation20.7%11%10.5%
Previous antimicrobials11739.3%4242%7537.9%
Fecal incontinence4414.8%1414%3015.2%
Pregnancy62%00%63%
Uterine prolapse41.3%00%42%
IV catheter14448.3%5151%9347%
Corticosteroid treatment3511.7%77%2814.1%
Surgery during admission6021%1616%4422.2%
Other invasive procedures175.7%77%105.1%

Fever and sepsis

Of all patients with NAUTI, 51.5% were febrile on the study day, either due to UTI or to other causes (59% in non-EU countries, 44.7% in EU countries; P < 0.05). Of the whole population with NAUTI, and throughout the observation period, 2% went on to develop severe sepsis, 0.3% septic shock and 1.7% multiorgan failure, always according to the assignment to this category made by the observing physician (Table 5). In 2.7% of all patients with NAUTI the micro-organism recovered from the urine was also isolated from blood.

Table 5.  Clinical data
Clinical dataGlobalNon-EU countriesEU countries
n%n%n%
  • *

    P < 0.05.

Fever*15351.559599444.7
Same micro-organism isolated in blood and urine102.75452
Presence of sepsis10735.938386934.8
Severity of sepsis
 Plain sepsis9531.930306532.8
 Severe sepsis623331.5
 Septic shock10.31100
 Multiorgan failure51.74410.5

Catheter-associated UTI

Catheter-associated UTI (CAUTI) was present in 187 patients (62.8%) (Table 6). A bladder catheter was the most commonly used type (92.4%), and 90.8% were short-term catheterisations (<30 days): in 44.8% 1–7 days and in 46% 8–30 days. The main indication was incontinence (35.5%), followed by obstruction (20.1%), perioperative monitoring (19.5%) and non-surgical output measurement (17.8%). The indication for bladder catheterisation was not considered adequate in 7.6% of cases at the time of the first visit, and continuation of bladder catheterisation was considered unnecessary in 31.3% of patients from then on. A closed drainage system was used in only 78.5% of catheterised patients, a silver-coated catheter in 2.2% and a urinometer in 23.9%. Major errors in catheter management were observed in 24% of these patients. Opening of the drainage system was the most frequent (16.8%). An opened drainage system or a violated closed system was observed in 36.8% of catheterised patients. We estimated that in 53.1% of CAUTI patients an obvious preventable mistake had been made, either in the indication of catheter use or in its management. No significant differences were found between EU and non-EU patients in the UC data.

Table 6.  Urinary catheter data
 TotalNon-EU countriesEU countries
n%n%n%
  • Differences were statistically insignificant (i.e. P > 0.05).

  • *

    Either in indication or management.

Type of urinary catheter
 Bladder catheter17192.45988.111294.9
 Suprapubic catheter94.957.543.4
 Nephrostomy42.134.510.8
 Total18410067100117100
Days of catheterisation
 1–77844.83553.84339.4
 8–3080462436.95651.4
 More than 30169.269.2109.2
 Total17410065100109100
Global catheter indication
 Surgery3319.512202119.3
 Output measurement3017.89152119.3
 Obstruction3420.112202220.2
 Incontinence6035.52236.73834.9
 Other127.158.376.4
Catheter indication not adequate137.646.598.2
No catheter indication on study day5631.32029.43632.4
No initial or later indication6236.72336.53936.8
Closed drainage system14278.55582.18776.3
Silver-coated catheter used42.20043.5
Urinometer used4323.91319.73026.3
Errors in catheter management
 Catheter insertion31.70032.6
 Catheter care105.5710.432.6
 Drainage system opened3116.81420.61714.7
Open drain or violated closed drain6836.82739.14135.3
Global preventable errors*9353.13958.25450

Antimicrobial therapy and outcome

At the time of the first visit, 75.5% of patients with NAUTI were receiving one or more antimicrobial agents. Antimicrobial treatment was not considered adequate in 19.8% of all cases. The median number of days of planned treatment was 7 days (IQR 7 days; range 0–21). The mean number of days of planned treatment in non-EU patients was significantly higher than in EU patients (10.85 verss 6.83; P < 0.05). Urinary catheters were changed or withdrawn in 44.4% of cases during the observation period. The mortality rate of the study population was 11.7%, with 1.8% considered as attributable to UTI according the observer's own opinion (Table 8). In these areas there were no significant differences between EU and non-EU countries (Table 7).

Table 8.  Global related factors (urinary catheter excluded)
 TotalWithout UCWith UC
n%n%n%
  • *

    P < 0.05.

Obstructive uropathy/lithiasis*5518.5%1814.3%4220.8%
Urinary tract anatomic abnormalities196.4%54.5%147.5%
Urological intervention*196.4%21.8%179.1%
Previous urinary tract infections7525.2%2522.5%5026.7%
Renal transplantation20.7%10.9%10.5%
Previous antimicrobials*11739.3%3531.5%8243.9%
Fecal incontinence4414.8%65.4%3820.3%
Pregnancy62%43.6%21.1%
Uterine prolapse41.3%10.9%31.6%
IV catheter*14448.3%3733.3%10757.2%
Treatment with corticosteroids3511.7%1816.2%179.1%
Surgery on admission6021%1816.2%4222.5%
Other invasive procedures175.7%54.5%126.4%
Table 7.  Treatment and outcome
 TotalNon-EU countriesEU countries
n%n%n%
Antimicrobial treatment22275.5787814474.2
Antimicrobial treatment adequacy17880.26279.511680.6
Patients with urinary catheter
 Change of catheter7544.42746.64843.2
 Adequate antimicrobials + catheter change5232.71831.63433.3
Global mortality3511.713132211.1
Attributable mortality51.73321

Comparison of patients with or without catheter

Tables 8 and 9 compare patients with and without UC. Patients without UC were significantly younger with less severe underlying diseases (P < 0.05). The presence of obstructive uropathy, urological intervention, previous antimicrobial use and fecal incontinence were significantly more frequent in patients with UC (P < 0.05). Fever was more frequent in catheterised patients (P < 0.05), although the incidence of sepsis was similar in both groups.

Table 9.  Comparison with/without catheter
  Total
(n = 298)
 Without UC
(n = 111)
 With UC
(n = 187)
 
  • *

    P < 0.05.

  • a

    Exclusively according to the observer's own criteria.

Age* (SD) 62.71 (25) 56.62 (28.4) 66.24 (22.1) 
Charlson index mean (SD) 62.71 (25) 2.75 (3) 3.31 (2.6) 
n%n%n%
SexFemale16254.766609651.6
Male13445.344409048.4
 1* (Non-fatal)15254.56662.98649.4
2 (Ultimately fatal)9935.53230.56738.5
3 (Rapidly fatal)281076.72112.1
Fever* 15351.54742.310657
Presence of sepsis 10735.93329.77439.6
Severity of sepsisPlain sepsis9531.93228.86333.7
Severe sepsis620063.2
Septic shock10.310.900
Multiorgan failure51.70052.7
 Gram negatives22465.98368.614164.4
Gram positives7221.23024.84219.2
Candida*4412.986.63616.4
Polymicrobial UTI (two isolates) 4214.11093217.1
Same micro-organism isolated in blood and urine 102.721.583.4
Antimicrobial treatment given 22275.57669.714678.9
Antimicrobial treatment adequacy 19372.3737312071.9
Global mortality* 3511.732.73217.1
Attributable mortalitya 51.70052.7

Etiological differences between the two groups are described in Table 10. Polymicrobial infection was more frequent in the CAUTI group and slightly over the boundary of statistical significance (17.1% versus 9%, P = 0.06). Escherichia coli was clearly more frequent in patients without urinary devices (40.5% versus 25.1%; P < 0.05) whereas Candida and Pseudomonas were more frequently isolated in catheterised patients (16.4% versus 6.6% and 10.5% versus 4.1%, respectively; P < 0.05). The significant differences observed in Pseudomonas and E. coli were only seen in the group of patients without a urinary device (Table 11). No significant differences were found in the incidence of bacteremic UTI, which was more frequent in the UC group (3.4% versus 1.5%). Catheterised patients presented a higher rate of global mortality (P < 0.05).

Table 10.  Micro-organisms isolated in urine (>1%)
With UC
(n = 219)
 Without UC
(n = 121)
 Total
(n = 340)
 
  • *

    P < 0.05.

  • a

    CNS, coagulase-negative staphylococci.

Escherichia coli*55 (25.1%)Escherichia coli*49 (40.5%)Escherichia coli104 (30.6%)
Candida sp.*36 (16.4%)Enterococcus sp.19 (15.7%)Enterococcus sp.48 (14.1%)
Enterococcus sp.29 (13.2%)Klebsiella sp.12 (9.9%)Candida sp.44 (12.9%)
Pseudomonas aeruginosa*23 (10.5%)Proteus sp.9 (7.4%)Klebsiella sp.34 (10%)
Klebsiella sp.22 (10%)Candida sp.*8 (6.6%)Pseudomonas aeruginosa28 (8.2%)
Proteus sp.16 (7.3%)Pseudomonas aeruginosa*5 (4.1%)Proteus sp.25 (7.4%)
Enterobacter sp.11 (5%)Staphylococcus aureus4 (3.3%)Enterobacter sp.14 (4.1%)
Staphylococcus aureus8 (3.7%)CNS4 (3.4%)Staphylococcus aureus12 (3.5%)
Citrobacter sp.6 (2.7%)Enterobacter sp.3 (2.5%)Citrobacter sp.9 (2.6%)
CNSa4 (1.8%)Citrobacter sp.3 (2.5%)CNS7 (2.1%)
Acinetobacter sp.3 (1.4%)Streptococcus agalactiae3 (2.5%)Morganella sp.4 (1.2%)
Table 11.  Catheter-adjusted etiological differences between EU and non-EU countries
Micro-organismsWith UC (n = 219)Without UC (n = 121)
EU countriesNon-EU countriesEU countriesNon-EU countries
  • *

    P < 0.05.

Escherichia coli37 (27.2%)18 (21.7%)42 (47.7%)*7 (21.2%)*
Pseudomonas12 (8.8%)11 (13.3%)0 (0%)*5 (15.2%)*
Candida25 (18.4%)11 (13.3%)4 (4.5%)4 (12.1%)

Discussion

The incidence density of NAUTI in Europe obtained in our study was 3.55 per 1000 hospitalised patient-days. We made a rough estimate of an equivalent prevalence rate of 10.65 episodes per 1000, exclusively in order to compare this data with other European figures. Although the mean time of nosocomial infection duration used in this estimation should ideally have been derived from a more complex method recently revised by Gastmeier et al. [6], we did not have the necessary data to perform it. European information regarding this topic is generally fragmentary, covering different populations and areas and difficult to compare with our data [8,9]. The most recently reported prevalence figures that cover a more general population range from 0.5 per 1000 to 24 per 1000, with most around 20 per 1000 [10–15].

The clinical condition of patients with NAUTI showed significant differences between EU and non-EU patients. Non-EU patients were significantly younger, had more severe underlying diseases, and presented more frequently with obstructive uropathy as a predisposing factor for UTI. These facts may reflect different admission criteria and management of obstructive uropathy between EU and non-EU institutions.

With regard to etiology, this study is largely confirmatory of data obtained in ESGNI-003 [1]. It shows important shifts in the etiological agents of NAUTI, with an increase in yeasts and Gram-positive bacteria, such as Enterococcus, which has been described in other studies [16–18].

The majority of NAUTI studies and major references in infectious diseases identify NAUTI with CAUTI [19–25]. To our surprise, a high proportion of NAUTI in the present study (37%) were not related to an indwelling urinary catheter. In fact, 22.8% of patients were not associated with any ‘classic’ UTI-predisposing factor. Patients with NAUTI not associated with urinary catheter are younger, less compromised, have less frequent urological interventions and have a better outcome than those with CAUTI. They clearly deserve greater attention [25–27]. Most NAUTI (63%) are urinary catheter-related, although in a lower proportion than in other global NAUTI series [22,28]. In general, these patients are elderly, with severe underlying diseases and with high global mortality rates (over 17%). We found a higher incidence of fever than expected for CAUTI [29,30]. Otherwise, bacteremic NAUTI occurred in 2.7% of cases in our study, a figure similar to that previously reported [31,32].

The information revealed in our study gives cause for concern throughout Europe. Overall, 20% of European institutions surveyed do not have specific written guidelines for management and care of urinary catheters (45% of non-EU hospitals). The UC was not indicated in 37% of patients, either initially or at the moment of the NAUTI diagnosis. This is important in terms of primary prevention [28,33]. Although a closed system for urine drainage is now widely recommended [19,34,35], we found 21.5% of catheterised patients with open systems, confirming similar recent surprising data from other European studies [36]. Violation (opening) of closed systems is also a major error [37], and occurred in 17% of our cases.

It is also alarming that antimicrobial therapy was considered inadequate in almost one of every five patients treated for UTI, and that the catheter was changed in fewer than 50% of the patients with infection. The overall mortality of 1.8% (six patients) was considered attributable to NAUTI according to the observers' own criteria. All these patients had an indwelling urinary catheter and represented 3% of all catheterised patients. The contribution of CAUTI to mortality is unclear [29,38,39], so further directed studies should be performed in order to evaluate it.

The results of our study provide a panoramic view of nosocomial UTI in Europe, obtained with limited resources and with the efforts of many. Our figures must be confirmed by other studies, but the dimension of the problem is obvious. There is clearly room for improvement in the indication for bladder catheterisation, catheter care and medical management of NAUTI. Hence, specific guidelines must be implemented practically by European scientific societies.

Appendix

Appendix 1 esgni-004 authors

Dr Dieter, Adam (Chu Munich. Germany); Dr Agulla, Andres (C.H.A. Marcide. Spain); Dr Ros, Alaine (Hospital Nord. France); Dr Alvarez, Maria (Hc Asturias. Spain); Dr Aranaz, Carlos (Hospital Monte Naranco. Spain); Dr Arosio, Marco (Ospedali Riuniti. Italy); Dr Arribas, Jose Luis (Miguel Servet. Spain); Dr Arribi, Ana (Severo-Ochoa. Spain); Dr Arta, Balode (Latvia Republic Ch. Latvia); Dr Aspiroz, Carmen (Hospital Comarcal Alcañiz. Spain); Dr Aulami, Athina (Lainon Gh. Greece); Dr Bakir, Mehmet (Cumhuriyet University of. Hastanesi. Turkey); Dr Baraia-Etxaburu, Josu (H. Basurto. Spain); Dr Manau, Beatrice (Lariboisiere/Fernand Wida. France); Dr Bergerova, Tamara (University Hospital Plzen. Czech Republic); Dr Bernasconi, Enos (Or Lugano. Switzerland); Dr Boggian, Kattia (Kantonhospital. Switzerland); Dr Bogumica, Bober (Lctld Silesia. Poland); Dr Bonadio, Mario (S. Chiara H Pisa. Italy); Dr Bouza, Emilio (Hgu Gregorio Marañon. Spain); Dr Brands, Christiane (Middelheim G. Hospital. Belgium); Dr Hombrouck, Cecile (Jean Verdier. France); Dr Cetinkaya, Yesim (Hacettepe University Hospital Turkey); Dr Christos, Mathas (Gh Algia Olga. Greece); Dr Cisterna, Ramon (H. Basurto. Spain); Dr Clemenceau, Kahla (H Francois Quesnay. France); Dr Croix, Jean-Claude (C.H. Troyes (Hts Clos). France); Dr Crotti, Daniele (R. Silvestrini. Italy); Dr Govaerts, Danielle (Chu Andre Vesale. Belgium); Dr Delmee, Michel (Uh Saint Luc. Belgium); Dr Derrington, Petra (Southmead H. United Kindom); Dr Doganay, Mehmet (Erciyes University Hospital Turkey); Dr Dzupova, Olga (Bulovkauniversityhospital. Czech Republic); Dr Elisabeth, Nagy (Uh Szeged. Hungary); Dr Ena, Javier (H Villajoyosa. Spain); Dr Esen, Saban (Ondokuz Mayis. Turkey); Dr Espejo, Elena (Hospital De Terrassa. Spain); Dr Exposito, Ana (Hospital Monte Naranco. Spain); Dr Falagas, Matthew (Hygeia Hospital. Greece); Dr Fameree, Dominique (C.H.U. De Charleroi. Belgium); Dr Fariñas, Carmen (Hu Marques De Valdecilla. Spain); Dr Ferreira, Deolinia (Hu Coimbra. Portugal); Dr Ferrer, Isabel (Miguel Servet. Spain); Dr Ferretto, Roberto (Oc Maggiore Verona. Italy); Dr Freitas, Ema (Ch Furuchal. Portugal); Dr Fuchs, Karl (Lkh Voecklabruck. Austria); Dr Gabriels, Patrick (R.Z.Sint-Trudo. Belgium); Dr Gantenberg, Rolf (Frankfurt Gh. Germany); Dr Garau, Javier (H Mutua Terrassa. Spain); Dr Garrino, Maria-Grazia (Ucl Mont-Godinne. Belgium); Dr Gatnik, Vojko (H. Dr F. Derganc. Slovenia); Dr Gesu, Giovanni and Dr Ossi, Cristi (H San Rafaele Milano. Italy); Dr Giamarrellou, Helen (Sismanogliou Gh. Greece); Dr Gian-Luigi, Cartolano (Chg Poissy-St-Germain. France); Dr Gordts, Bart (Az Sint January. Belgium); Dr Grassi, Carlos (Xeral-Cies. Spain); Dr Grzesiowski, Pawel (P2-P10. Poland); Dr Gubler, Jacques (Stadtspital Triemli. Switzerland); Dr Gutierrez, Jose (San Cecilio. Spain); Dr Hauer, Thomas (Friburg Uh. Germany); Dr Hell, Markus (Gh Salzburg. Austria); Dr Hernãndez, Alberto (H Sant Jaume. Spain); Dr Hernández, Alberto (Hospital Sant Jaume. Spain); Dr Holcikova, Alena (University Childrens Hospital Czech Republic); Dr Honderlick, Patrick (Hospital Foch. France); Dr Hosoglu, Salih (Dicle University. Turkey); Dr Nahimana, Immaculee (Chuv. Switzerland); Dr Jansens, Hilde (University Hospital Antwe. Belgium); Dr Jezek, Petr (Mh Pribram. Czech Republic); Dr Juraj, Hanzen (HPL, s.r.o. Slovak Republic); Dr Kaltenis, Petras (Vilnius Uch. Lithuania); Dr Kniehl, Eberhard (Kalsruhe H. Germany); Dr Korten, Volkan (Marmara University. Turkey); Dr Kotulova, Daniela (Faculty Hospital. Slovak Republic); Dr Kouppari, Georgia (Amalia Fleming H. Greece); Dr Koutra, Grammati and Dr Saroglou, George (Gh Evangelismos. Greece); Dr Krajewska, Marta (Cch Warsaw. Poland); Dr Krcmery, Vladimir (St Elisabeth Ci. Slovak Republic); Dr Kucisec-Tepes, Nastja (Gh Sueti Duh. Croatia); Dr Lelekis, Moyssis (Athens General Hospital. Greece); Dr Liskova, Anna (Regional Hospital Nitra. Slovak Republic); Dr Luzzaro, Francesco (Ospedale Di Circolo. Italy); Dr Madaric, Vesna (Koprivn General Hospital. Croatia); Dr Malafiej, Eugeniusz (Polish Moth.Mem. Hospital. Poland); Dr Marroni, Massimo (Azienda Ospedaliera. Italy); Dr Martinez, Joaquin (H Barcelona. Spain); Dr Marty, Nicole (Chu Toulouse-Rangueil. France); Dr Mavridis, Anestis (Gh G. Hatzikosta. Greece); Dr Mewis, Alex (Virga Jesse Hospital. Belgium); Dr Michelle, Viot (Center A. Lacassagne. France); Dr Miftdae, Egidia (I. Diseases Hospital. Romania); Dr Montejo, Miguel (H. Baracaldo. Spain); Dr Moris Dela Tassa, Joaquin (H. Cabueñes. Spain); Dr Oteo Revuelta, Jose Antonio (Hospital De La Rioja. Spain); Dr Ozkan, Feriha (Ege University Medical Facult. Turkey); Dr Palomino, Julian (Hg Virgen Del Rocio. Spain); Dr Pastor, Vicente (H Princesa. Spain); Dr Paterson, David (Ismett. Italy); Dr Pavel, Babkin (Militar Medical Academy. Russia); Dr Paz Vidal, Isabel (Hospital Cristal-Piñor. Spain); Dr Pina, Elaine (H Capuchos Desterro. Portugal); Dr Pizzato, Enrico (Ospedale Boldrini. Italy); Dr Pombo, Vitor (Hu Coimbra. Portugal); Dr Prammer, Wolfgang (Kh Wels. Austria); Dr Presterl, Elisebeth (Gh Vienna. Austria); Dr Priv-Doz, Heinz-Michael (Nuernberg Clinic. Germany); Dr Radulescu, Amanda (I. Diseases Hospital. Romania); Dr Rafalsky, Vladimir (Smolensk Regional Hospital. Russia); Dr Raga, Xavier (St.Pau I Sta.Tecla. Spain); Dr Rene-Marc, Jolidon (Ch Yverdon. Switzerland); Dr Rodriguez, Javier (San Agustin De Aviles. Spain); Dr Rodriguez Brezmes, Juana Fe (Virgen De La Concha. Spain); Dr Roilides, Emmanuel (Hippokration. Greece); Dr Rubinstein, Ethan (Sheba Mc. Israel); Dr Ruyer, Olivier (Ch Belfort. France); Dr Ruzicka, Filip (St.Anna's Hospital, Brno. Czech Republic); Dr Salonen, Juha (Turku University Hospital. Finland); Dr Samet, Alfred (Public Hospital no. 1. Poland); Dr Sevillano, Joaquina (Povisa. Spain); Dr Shah, Pramod (J.W Goethe Uh. Germany); Dr Simeckova, Eva (Strakonice Hospital. Czech Republic); Dr Sion, Jean-Paul (Monica Campus Efka. Belgium); Dr Skerl, Marjeta (University Medical Center. Slovenia); Dr Soler, Marte and Dr Hernaez, Silvia (Cu Navarra. Spain); Dr Spencer, Robert (Bristol Royal Infirmary. United Kindom); Dr Tomic, Viktorija (University of Clinic Of Resp. Diseas. Slovenia); Dr Tripkovic, Vesna (Uhc Rebro. Croatia); Dr Ural, Onur (Medical Faculty Hospital. Turkey); Dr Van Griethuysen, Arjanne (Hospital Rijnstate Arnhem. Netherlands); Dr Vanhems, Philippe (Edouart Herriot Hospital. France); Dr Varela Otero, Jose (Hospital Xeral-Calde. Spain); Dr Vidal, Francesc (Hu Tarragona Joan Xxiii. Spain); Dr Von Wulffen, Hinrik (Allg Kh Barmbek. Germany); Dr Voss, Andreas (Umc St Radboud. Netherlands); Dr Wagenlehner, Florian (St Elisabeth H. Germany); Dr Yazdanfard, Younes (Hvidoure Hospital. Denmark); Dr Zupancic, Martina (Gh Jesenice. Slovenia).

Ancillary