Hggm.es

A European perspective on intravascular catheter-related infections: reporton the microbiology workload, aetiology and antimicrobial susceptibility(ESGNI-005 Study)E. Bouza1, R. San Juan1, P. Mun˜oz1, J. Pascau2, A. Voss3 and M. Desco2 on behalf of the CooperativeGroup of the European Study Group on Nosocomial Infections (ESGNI) 1Clinical Microbiology and Infectious Diseases and 2Experimental Medicine, Hospital GeneralGregorio Maran˜o´n, Madrid, Spain and 3Medical Microbiology, University Hospital Nijmegen,Nijmegen, The Netherlands The laboratory workload, microbiological techniques and aetiology of catheter-related infections inEuropean hospitals are mostly unknown. The present study (ESGNI-005) comprised a 1-day (22 October2001), laboratory-based, point-prevalence survey based on a questionnaire completed by microbiologylaboratories in European (European Union (EU) and non-EU) hospitals. Also included were questionsrequesting retrospective information for the year 2000. In total, 151 hospitals from 26 European countriesparticipated, of which 78.1% were teaching institutions. Overall, the estimated population served bythese institutions was 121 363 800, and the estimated number of admissions during 2000 was 6 712 050.
The total number of catheter tips processed during 2000 was 142 727, or 21 ⁄ 1000 admissions, of which23.7% were considered to be positive in the institutions using semiquantitative or quantitativetechniques. Overall, EU centres received significantly more catheter tip samples ⁄ 1000 admissions andhad a significantly higher rate of ‘positivity’ (p < 0.0001) than non-EU centres. Of the institutionssurveyed, 11.4% (7.2% in EU countries and 23.7% in non-EU countries; p 0.04) used only qualitativetechniques for catheter tip sample processing. On the day of the study, 167 microorganisms wererecovered from significant catheter tip cultures (122 patients), of which Gram-positive bacteriarepresented 70.7%, Gram-negative bacteria 22.2%, and yeasts 7.2%. The five most common microor-ganisms were coagulase-negative staphylococci, Staphylococcus aureus, Candida spp., Enterococcus spp.
and Pseudomonas spp. Overall, 19% of catheter tip cultures were polymicrobial. In the case of S. aureus,40% of isolates were resistant to oxacillin, as were 63.4% of coagulase-negative staphylococcus isolates.
Of 37 Gram-negative isolates, 35% were resistant to cefotaxime, 31% to ceftazidime, and 27% tociprofloxacin. Imipenem and cefepime had the lowest reported rates of resistance (11%).
Catheter tip, epidemiology, intravascular catheter-related infections, laboratory workload, microbio- Original Submission: 1 October 2003; Revised Submission: 18 December 2003; Accepted: 16 January 2004 Clin Microbiol Infect 2004; 10: 838–842 Culture of catheter tips with semiquantitative or quantitative methods is currently the standard Intravenous catheter-related infections (IV-CRIs) microbiological test for the diagnosis of catheter are common nosocomial infections, which are colonisation and is an essential part of the currently responsible for > 60% of nosocomial diagnosis of catheter-related bloodstream infec- bacteraemic episodes in European hospitals [1].
tions [2]. The techniques used to diagnose cath-eter laboratories have not been clearly defined and Corresponding author and reprint requests: E. Bouza, Servicio the corresponding workload is not known.
de Microbiologı´a Clı´nica y Enfermedades Infecciosas-VIH, The aim of this study by the ESCMID Study Hospital General Universitario ‘Gregorio Maran˜o´n’, Dr Esqu- Group on Nosocomial Infections (ESGNI) was to erdo 46, 28007 Madrid, SpainE-mail: [email protected] obtain general information on the microbiology Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases workload and techniques used for IV-CRI diag- variables, the Mann–Whitney U-test to compare continuous nosis in Europe, as well as the aetiology and variables not normally distributed, and the chi-square orFisher exact test to compare proportions. All statistical tests antimicrobial susceptibility patterns of pathogens (IV-CC). The clinical aspects of intravascularcatheter-related bloodstream infection will be In total, 151 hospitals from 26 countries partici-pated in the study (Table 1), comprising 111 hospitals from 12 European Union (EU) countriesand 40 hospitals from 14 non-EU countries.
The present study (ESGNI-005) was a 1-day (22 October 2001),laboratory-based, point-prevalence study based on a question-naire sent to the microbiology laboratories of every European hospital with one or more ESGNI members (c. 400 invitationsto participate). The questionnaire also included a request for Of the 151 participating institutions, 78.1% and information on the results for the year 2000 in the participating 21.9%, respectively, were teaching and non-teach- ing institutions. Hospital sizes varied from < 500 The questionnaire requested information on the population served by the hospital, total number of beds, total number of beds (34%), to 500–1000 beds (40.4%) and > 1000 admissions, total number of catheter tips sent to microbiology beds (25.2%). Overall, the estimated population laboratories, and whether a qualitative, semiquantitative or served by these institutions was 121 363 800, and quantitative technique was used for processing. Also reques- the number of estimated admissions during 2000 ted was information on the total number of sterile samples andtotal number of significant positive samples (‡ 15 CFU ⁄ mL of any semiquantitative culture, or any quantitative culture Results regarding the microbiology workload yielding ‡ 103 CFU) processed during the year 2000. Further- in 2000 are shown in Table 2. The total number of more, information on the use of indirect methods of assessing catheter tips processed during 2000 was 142 727, catheter tip colonisation was requested (e.g., comparativequantitative blood cultures, skin and hub semiquantitative or 21 ⁄ 1000 admissions, of which 23.7% were cultures, differential time to growth between blood cultures considered to be positive in the institutions using taken from peripheral veins and catheters).
For each episode of IV-CC on the day of study, information was requested regarding the microorganisms present and theirantibiotic susceptibility profile, regardless of the laboratory Table 1. Distribution of participating hospitals (n = 151) method used for detection. The list of antimicrobial agents studied included: penicillin, ampicillin, ticarcillin, amoxycil- lin–clavulanate, oxacillin, cefazolin, cefuroxime, cefotaxime,ceftazidime, cefepime, imipenem, aztreonam, co-trimoxazole, ciprofloxacin, levofloxacin, gentamicin, tobramycin, amika- cin, streptomycin, rifampicin, erythromycin, clindamycin, vancomycin, tetracycline and chloramphenicol.
IV-CC was defined as any semiquantitative culture yielding ‡ 15 CFU, or any quantitative culture yielding ‡ 103 CFU of bacteria or fungi. Any colony counts below these levels were considered to be negative. Data from institutions usingqualitative techniques for catheter tip processing were notincluded in the study.
Table 2. Microbiology workload associated with cathetertip samples in the year 2000 Discrete variables were expressed as percentages, and con- tinuous variables as the mean and standard deviation when normally distributed, or as the median and interquartile rangeif their distribution was not normal. For discrete variables with missing values, percentages were calculated from the total of valid cases whenever missing values did not exceed 20% (in these instances, the variable was excluded from the analysis).
Student’s unpaired t-test was used to compare continuous Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 838–842 840 Clinical Microbiology and Infection, Volume 10 Number 9, September 2004 Overall, EU centres received significantly more Table 4. Data for catheter tip samples obtained on the had a significantly higher rate of ‘positivity’ (p < 0.0001) than non-EU centres.
Diagnostic techniques used in differentinstitutions sive care units (41%), surgical services (25.4%) Of all the institutions surveyed, 11.4% (7.2% in and general medical services (23.8%).
EU countries and 23.7% in non-EU countries; The number of microorganisms isolated on the p 0.04) used only qualitative techniques to process day of the study from significant catheter tip catheter tip samples. Most (63.8%) institutions cultures was 168 (Table 5). Overall, 19% of processed these samples using semiquantitative cultures were polymicrobial. Gram-positive bac- teria comprised 70.7% of all isolates, and Gram- between EU and non-EU countries. Quantitative negative bacteria 22.2%. Yeasts were isolated techniques were used in 24.8% of microbiology from 7.2% of catheter tip samples. The five laboratories (27.9% in EU countries vs. 15.8% in organisms isolated most commonly were, in non-EU countries; p 0.07) (Table 3). In 68.5% of decreasing order: coagulase-negative staphylo- the participating institutions, an indirect method cocci, Staphylococcus aureus, Candida spp., Entero- of assessing catheter colonisation, e.g., comparat- coccus spp. and Pseudomonas spp. The organisms ive quantitative blood culture, infusate culture, isolated were similar in both groups of hospitals internal brushing, and skin or hub culture, was (EU and non-EU), with the single exception of used. These methods were used most frequently S. aureus, which was isolated more frequently in in EU countries (75.7% vs. 47.4%; p 0.02).
non-EU countries (34.3% vs. 6.1%; p < 0.0001).
While Candida was the second most frequentlyisolated microorganism in catheter tips in EU countries (9.1%), laboratories from non-EU coun- Data regarding microbiology workload obtained tries did not report any Candida isolates; however, on the day of the study differed only slightly from this difference did not reach statistical signifi- the questionnaire data relating to the year 2000.
On the day of the study, 676 catheter tip sampleswere cultured in 151 institutions. Assuming a Resistance patterns of the most frequent isolates similar number of admissions in the year 2001, theestimated number of catheter tip samples during Table 6 lists the antibiotic resistance data for the day-prevalence study was 36.8 ⁄ 1000 admis- organisms isolated from catheter tips in the sions. According to the criteria of each individual different centres. The sensitivity assays were not laboratory, 168 (25%) samples were reported as performed in a central laboratory, and local susceptibility testing results were taken at face It was possible to collect information for 122 value. In the case of S. aureus, 40% of isolates were positive catheter tip samples, of which 94 were resistant to oxacillin, as were 63.4% of coagulase- from EU countries and 28 from non-EU countries.
negative staphylococcus isolates. Of the 118 These samples were taken from patients in inten- Gram-positive isolates from catheter tips in Eur-ope, 51.5% were oxacillin-resistant, but vancomy-cin was reported to be active against all but two Table 3. Techniques used for the diagnosis of catheter tipinfection single isolates (coagulase-negative staphylococcusand E. faecalis). Of the 37 Gram-negative organ- isms isolated, only 65% were susceptible to cefotaxime, and only 69% to ceftazidime. Ciprofl- oxacin resistance was reported for 27% of these Semiquantitative or quantitative techniques isolates, with imipenem and cefepime having the Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 838–842 CONS, coagulase-negative staphylococci.
aSpeciation of Candida isolates was not requested in the questionnaire.
bp < 0.0001.
Table 6. Overall resistance rates in organisms isolated were in non-EU countries. Fortunately, most laboratories use the recommended techniques, and in > 23% of EU microbiology laboratories, the most recent quantitative methods have replaced Maki’s semiquantitative catheter tip culture.
In parallel, more conservative diagnostic tests, which do not depend on catheter withdrawal, have been developed. These include comparative quantitative blood cultures taken from a catheter CONS, coagulase-negative staphylococci; NT, not tested.
and peripheral veins [6], skin and hub cultures[7], and differential time of growth between blood comprised three Acinetobacter isolates and one obtained from a catheter and that obtained from peripheral veins [8]. The wide use of thesetechniques reported in the present study (nearly70%) seems to be proof of the acceptance in European institutions of conservative diagnostic Most literature on clinical microbiology, infec- techniques for IV-CRI. Nevertheless, it should not tious diseases and laboratory medicine gives no be inferred that these institutions are using such insight into the workload generated by catheter techniques exclusively on a day-to-day basis.
tip sampling in microbiology laboratories. The The aetiology of IV-CRI in Europe, as shown in present data, obtained from 151 hospitals in the present study, is associated mostly with Europe, show that 21 catheter tip cultures were Gram-positive microorganisms (coagulase-negat- processed ⁄ 1000 admissions in 2000, with a signi- ive staphylococci and S. aureus), as reported in ficant difference between the workloads in EU other European studies [9, 10]. Candida spp. is the and non-EU hospitals. EU hospitals processed third most common pathogen isolated in cases of more samples each year, with a higher rate of IV-CRI, reflecting the reported rising trend of this positivity, which could partly reflect more inva- microorganism as a major nosocomial pathogen sive procedures in EU countries, with a parallel in developed countries [11–13]. The significance higher use of central intravenous lines. It should of the higher incidence of S. aureus in non-EU be noted that new guidelines discourage routine countries cannot be explained clearly, although culturing of all removed catheters, and only it may be associated with differences in the recommend culturing catheters from patients procedures employed in the insertion of intravas- when bloodstream infection is suspected.
cular catheters and in operating theatres, as well Although qualitative culture has no role in the diagnosis of IV-CC [3–5], and should be clearly The resistance patterns of the IV-CRI isolates discouraged, it was the only method used in reported in this study should be considered with > 11% of European laboratories, most of which caution for several reasons: the study was not Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 838–842 842 Clinical Microbiology and Infection, Volume 10 Number 9, September 2004 designed to evaluate the resistance rates directly, 3. Mermel LA, Farr BM, Sherertz RJ et al. Guidelines for the the isolates were not tested in a central laboratory, management of intravascular catheter-related infections.
Clin Infect Dis 2001; 32: 1249–1272.
and only a few microorganisms were studied.
4. Nahass RG, Weinstein MP. Qualitative intravascular catheter tip cultures do not predict catheter-related bac- drawn from the data in this study. There were teremia. Diagn Microbiol Infect Dis 1990; 13: 223–226.
high rates of resistance in Gram-negative isolates 5. SEIMCySEMICIUC. Conclusiones de la Conferencia de from catheter tips (> 30% to cefotaxime and Consenso sobre ‘Infecciones relacionadas con cate´teresintravasculares de corta permanencia en adultos SEIMC- ceftazidime, and nearly 30% to ciprofloxacin). A SEMICIUC’. Toledo: Sociedad Espan˜ola de Microbiologı´a similar observation was made in a European study of nosocomial bacteraemia isolates by 6. Capdevila JA, Planes AM, Palomar M et al. Value of dif- Dornbusch et al. [14], in which it was also noted ferential quantitative blood cultures in the diagnosis ofcatheter-related sepsis. Eur J Clin Microbiol Infect Dis 1992; that countries previously belonging to the ‘eastern block’ had higher rates of resistance. Another 7. Cercenado E, Ena J, Rodrı´guez-Cre´ixems M, Romero I, concern was that 40% of S. aureus isolates from Bouza E. A conservative procedure for the diagnosis of catheter tips were oxacillin-resistant in the pre- catheter-related infections. Arch Intern Med 1990; 150: sent study, which supports the use of glycopep- 8. Blot F, Nitenberg G, Chachaty E et al. Diagnosis of cath- tides as first-line empirical antimicrobial agents eter-related bacteraemia: a prospective comparison of the for catheter-related bloodstream infections.
time to positivity of hub-blood versus peripheral-blood In conclusion, the microbiological diagnosis of cultures. Lancet 1999; 354: 1071–1077.
IV-CRI is increasingly more effective in European 9. Richet H, Hubert B, Nitemberg G et al. Prospective multicenter study of vascular-catheter-related complica- institutions, but there are still some deficiencies in tions and risk factors for positive central-catheter cultures the techniques used, especially in non-EU coun- in intensive care unit patients. J Clin Microbiol 1990; 28: tries. The recognition of new pathogens such as Candida spp., and the increasing rates of antibiotic 10. Donelli G, De Paoli P, Fadda G, Marone P, Nicoletti G, resistance in Gram-positive and Gram-negative Varaldo PE. A multicenter study on central venous cath-eter-associated infections in Italy. J Chemother 2001; bacteria, stress the need for a general European 11. Chen YC, Chang SC, Sun CC, Yang LS, Hsieh WC, Luh KT. Secular trends in the epidemiology of nosocomialfungal infections at a teaching hospital in Taiwan, 1981 to 1993. Infect Control Hosp Epidemiol 1997; 18: 369–375.
12. Gastmeier P, Schumacher M, Daschner F, Ruden H. An The authors are indedted to the 139 colleagues from 29 analysis of two prevalence surveys of nosocomial infection countries who participated in this study.
in German intensive care units. J Hosp Infect 1997; 35:97–105.
13. Pittet D, Harbarth S, Ruef C et al. Prevalence and risk factors for nosocomial infections in four university hospi-tals in Switzerland. Infect Control Hosp Epidemiol 1999; 20: 1. Lyytikainen O, Lumio J, Sarkkinen H, Kolho E, Kostiala A, Ruutu P. Nosocomial bloodstream infections in Finnish 14. Dornbusch K, King A, Legakis N. Incidence of antibiotic hospitals during 1999–2000. Clin Infect Dis 2002:; 35: e14– resistance in blood and urine isolates from hospitalized patients. Report from a European collaborative study.
2. Rijnders BJ, Van Wijngaerden E, Peetermans WE. Cath- European Study Group on Antibiotic Resistance (ESGAR).
eter-tip colonization as a surrogate end point in clinical Scand J Infect Dis 1998; 30: 281–288.
studies on catheter-related bloodstream infection: howstrong is the evidence? Clin Infect Dis 2002; 35: 1053–1058.
Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10, 838–842

Source: http://www.hggm.es/umce/imagen_old/pdfs/ClinMicrobiolInfect(Study-005).pdf