Surveillance Trends in the epidemiology of catheter-related bloodstream infections; towards a paradigm shift, Spain, 2007 to 2019 Laia Badia-Cebada1,2 , Judit Peñafiel³ , Patrick Saliba⁴ , Marta Andrés⁵ , Jordi Càmara6,7,8 , Dolors Domenech9 , Emili Jiménez- Martínez10 , Anna Marrón11 , Encarna Moreno12 , Virginia Pomar13 , Montserrat Vaqué14 , Enric Limón⁴ , Úrsula Masats15 , Miquel Pujol16,17 , Oriol Gasch2,11,18 , on behalf of the VINCat programme (Infection Control Catalan Programme)19 1. Internal Medicine Department, Hospital Universitari Parc Taulí, Sabadell, Spain 2. School of Medicine, Universitat Autònoma de Barcelona, Sabadell, Spain 3. Unit of Statistics, Hospital Universitari de Bellvitge/Institut d’Investigació Biomèdica de Bellvitge-IDIBELL, L’Hospitalet de Llobregat, Spain 4. VINCat programme: Infection Control Catalan Programme, Barcelona, Spain 5. Infectious Diseases Unit, Department of Internal Medicine, Hospital Consorci de Terrassa, Spain 6. Department of Microbiology, Hospital Universitari de Bellvitge, L’Hospitalet de Llobregat, Spain 7. Institut d’Investigació Biomèdica de Bellvitge-IDIBELL, L’Hospitalet de Llobregat, Spain 8. CIBER de Enfermedades Respiratorias (CIBERes), ISCIII, Madrid, Spain 9. Infection Control Nurse, Hospital Josep Trueta, Girona, Spain 10. Infection Control Nurse, Hospital Universitari de Bellvitge, Institut d’Investigació Biomèdica de Bellvitge-IDIBELL, L’Hospitalet de Llobregat, Spain 11. Infectious Diseases Department, Hospital Universitari Parc Taulí, Sabadell, Spain 12. Infection Control Nurse, Parc Sanitari Sant Joan de Déu, Sant Boi de Llobregat, Spain 13. Infectious Diseases Unit, Department of Internal Medicine, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain 14. Infection Control Nurse, Hospital de Barcelona, Barcelona, Spain 15. Infection Control Nurse, Hospital Universitari Mútua Terrassa, Terrassa, Spain 16. Department of Infectious diseases, Hospital Universitari de Bellvitge, L’Hospitalet de Llobregat, Spain 17. Institut d’Investigació Biomèdica de Bellvitge-IDIBELL, L’Hospitalet de Llobregat, Spain 18. Institut d’Investigació i Innovació Parc Taulí, Sabadell, Spain 19. The members of the network are acknowledged at the end of the article Correspondence: Oriol Gasch (ogasch@tauli.cat) Citation style for this article: Badia-Cebada Laia, Peñafiel Judit, Saliba Patrick, Andrés Marta, Càmara Jordi, Domenech Dolors, Jiménez-Martínez Emili, Marrón Anna, Moreno Encarna, Pomar Virginia, Vaqué Montserrat, Limón Enric, Masats Úrsula, Pujol Miquel, Gasch Oriol, on behalf of the VINCat programme (Infection Control Catalan Programme). Trends in the epidemiology of catheter-related bloodstream infections; towards a paradigm shift, Spain, 2007 to 2019. Euro Surveill. 2022;27(19):pii=2100610. https://doi.org/10.2807/1560-7917.ES.2022.27.19.2100610 Article submitted on 05 Jun 2021 / accepted on 03 Dec 2021 / published on 12 May 2022 Background: Catheter-related bloodstream infections non-intensive care unit (ICU) wards, 62.7% (n = 5,822) (CRBSI) are frequent healthcare-associated infec- were related to central venous catheter (CVC), 24.1% tions and an important cause of death. Aim: To ana- (n = 2,236) to peripheral venous catheters (PVC) and lyse changes in CRBSI epidemiology observed by 13.3% (n = 1,232) to peripherally-inserted central the Infection Control Catalan Programme (VINCat). venous catheters (PICVC). Incidence rate fell over Methods: A cohort study including all hospital- the study period (IRR: 0.94; 95%CI: 0.93–0.96), espe- acquired CRBSI episodes diagnosed at 55 hospitals cially in the ICU (IRR: 0.88; 95%CI: 0.87–0.89). As a (2007–2019) in Catalonia, Spain, was prospectively whole, while episodes of CVC CRBSI fell significantly conducted. CRBSI incidence rates were adjusted per (IRR: 0.88; 95%CI: 0.87–0.91), peripherally-inserted 1,000 patient days. To assess the CRBSI rate trend catheter CRBSI (PVC and PICVC) rose, especially in per year, negative binomial models were used, with medical wards (IRR PICVC: 1.08; 95%CI: 1.05–1.11; the number of events as the dependent variable, IRR PVC: 1.03; 95% 1.00-1.05). Conclusions: Over the and the year as the main independent variable. From study, CRBSIs associated with CVC and diagnosed in each model, the annual rate of CRBSI diagnosed per ICUs decreased while episodes in conventional wards 1,000 patient days and the incidence rate ratio (IRR) involving peripherally-inserted catheters increased. with its 95% confidence intervals (CI) were reported. Hospitals should implement preventive measures in Results: During the study, 9,290 CRBSI episodes were conventional wards. diagnosed (mean annual incidence rate: 0.20 epi- sodes/1,000 patient days). Patients’ median age Introduction was 64.1 years; 36.6% (3,403/9,290) were female. The use of vascular devices in hospitalised patients In total, 73.7% (n = 6,845) of CRBSI occurred in is essential for their treatment, which frequently www.eurosurveillance.org 1 Table 1 2 Clinical and demographic characteristics of annual catheter-related bloodstream infections diagnosed at VINCat hospitals, Catalonia, Spain, 2007–2019 (n = 9,290) ALL 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Characteristics (n = 9,290) (n = 741) (n = 784) (n = 834) (n =  775) (n =  896) (n = 752) (n = 588) (n =  694) (n = 703) (n = 678) (n = 619) (n = 688) (n = 538) na %b na %b na %b na %b na %b na %b na %b na %b na %b na %b na %b na % na %b na %b Patient Age, mean (SD) 64.1 (15.7) 62.2 (16.2) 61.9 (15.8) 64.0 (16.2) 63.7 (15.5) 63.1 (16.0) 64.4 (15.4) 64.2 (14.8) 64.5 (15.8) 64.6 (15) 65.9 (14.9) 65.2 (15.6) 65.0 (15.4) 65.4 (16.4) Female sexc 3,403 36.6 286 38.6 315 40.2 320 38.4 311 40.1 320 35.7 264 35.1 206 35.0 237 34.1 240 34.1 244 36.0 208 33.6 245 35.6 207 38.5 Catheter Days since 3.0 4.0 5.0 5.0 3.0 3.0 3.0 3.0 3.0 4.0 2.0 3.0 3.0 3.0 catheter insertion, (0.0–14.0) (0.0–17.5) (1.0–19.0) (1.0–18.0) (0.0–12.0) (0.0–14.0) (0.0–13.0) (0.0–13.0) (0.0–14.0) (0.0–13.0) (0.0–12.8) (0.0–12.0) (0.0–13.0) (0.0–10.0) Md (IQR) 10.0 10.0 11 10.0 10.0 10.0 9.0 10.0 10.0 10.0 10.0 9.0 9.0 8.0 Days since admission, Md (IQR) (6.0–17.0) (6.0–17.0) (7.0–17.0) (6.0–18.0) (6.0–17.0) (5.3–18.0) (6.0–16.0) (6.0–18.0) (5.0–17.0) (6.0–18.0) (6.0–18.0) (6.0–18.0) (5.0–17.0) (5.0–16.0) Catheter type CVC 5,822 62.7 541 73.0 616 78.6 608 72.9 558 72.0 605 67.5 447 59.4 386 65.6 417 60.1 423 60.2 362 53.4 321 51.9 319 46.4 219 40.7 PVC 2,236 24.1 137 18.5 117 14.9 151 18.1 148 19.1 205 22.9 207 27.5 127 21.6 160 23.1 188 26.7 179 26.4 179 28.9 237 34.4 201 37.4 PICVC 1,232 13.3 63 8.5 51 6.5 75 8.9 69 8.9 86 9.6 98 13.0 75 12.8 117 16.9 92 13.1 137 20.2 119 19.2 132 19.2 118 21.9 Site of acquisition Medical ward 3,911 42.1 262 35.4 304 38.8 318 38.1 324 41.8 397 44.3 318 42.3 242 41.2 268 38.6 297 42.2 279 41.2 284 45.9 345 50.1 273 50.7 Surgical ward 2,934 31.6 242 32.7 217 27.7 264 31.7 231 29.8 247 27.6 238 31.6 184 31.3 245 35.3 240 34.1 238 35.1 208 33.6 213 31.0 167 31.0 ICU 2,445 26.3 237 32 263 33.5 252 30.2 220 28.4 252 28.1 196 26.1 162 27.6 181 26.1 166 23.6 161 23.7 127 20.5 130 18.9 98 18.2 Catheter use Haemodialysis 444 4.8 37 4.9 52 6.6 46 5.5 64 8.3 60 6.7 34 4.5 30 5.1 25 3.6 25 3.6 23 3.4 16 2.6 21 3.1 11 2.0 PN 2,467 26.6 202 27.3 216 27.6 253 30.3 214 27.6 219 24.4 195 25.9 147 25 180 25.9 204 29 181 26.7 165 26.7 177 25.7 114 21.2 Other 6,379 68.7 502 67.7 516 65.8 535 64.1 497 64.1 617 68.9 523 69.5 411 69.9 489 70.5 474 67.4 474 69.9 438 70.8 490 71.2 413 76.8 Aetiology d CoNS 3,652 39.5 345 46.6 343 43.8 368 44.3 316 40.8 328 36.8 286 38.0 233 39.6 248 35.7 250 35.6 268 39.6 259 42.1 234 34.9 174 33.6 S. aureus 2,268 24.6 153 20.7 165 21.0 176 21.2 146 18.9 201 22.5 166 22.1 137 23.3 186 26.8 190 27.1 165 24.4 156 25.4 229 34.2 198 38.2 Enterobacteriaceae 1,700 18.4 110 14.9 134 17.1 153 18.4 157 20.3 183 20.5 157 20.9 104 17.7 131 18.9 137 19.5 131 19.4 109 17.7 111 16.6 83 16 Enterococcus spp. 459 5.0 46 6.2 29 3.7 34 4.1 49 6.3 58 6.5 49 6.5 39 6.6 41 5.9 31 4.4 22 3.3 18 2.9 27 4.0 16 3.1 Candida spp. 540 5.9 41 5.5 45 5.7 51 6.1 45 5.8 56 6.3 43 5.7 34 5.8 42 6.1 42 5.9 47 6.9 36 5.8 31 4.6 27 5.2 www.eurosurveillance.org P. aeruginosa 483 5.2 36 4.9 46 5.9 30 3.6 50 6.5 55 6.2 44 5.8 39 6.6 38 5.5 48 6.8 36 5.3 27 4.4 20 2.9 14 2.7 Other 135 1.5 9 1.2 22 2.8 19 2.9 11 1.4 11 1.2 7 0.9 2 0.3 8 1.2 4 0.6 8 1.2 10 1.6 18 2.7 6 1.2 CoNS: coagulase-negative staphylococci; CVC: central venous catheter; ICU: intensive care unit; IQR: interquartile range; Md: median; P. aeruginosa: Pseudomonas aeruginosa; PICVC: peripherally-inserted central venous catheter; PN: parenteral nutrition; PVC: peripheral venous catheter; S. aureus: Staphylococcus aureus; SD: standard deviation; VINCat: Infection Control Catalan Programme. a Numbers of counts are presented in this column, unless otherwise specified by the row header. b Percentages are presented in this column, unless otherwise specified by the row header. c Information on sex was collected as a binary variable. d Data on aetiology were missing for 53 catheter-related bloodstream infections, including one in 2007, three in 2009, one in 2010, four in 2011, one in 2015, one in 2016, four in 2017, 18 in 2018 and 20 in 2019. involves the administration of drugs and fluids, par- The 55 Catalan hospitals participating in the VINCat enteral nutrition, or haemodialysis. The prevalence of programme are classified into three categories accord- peripheral (PVC) and central (CVC) venous catheter ing to complexity and to the number of beds available use among hospitalised patients estimated in differ- for hospitalisation: 500 beds or more (Group I), 200 to ent European surveys in the last decade is around 70% 499 beds (Group II), and fewer than 200 beds (Group and 10% respectively [1-3]. In in a prospective cohort III). A table with the number of hospitals participating study published in 2010, catheter-related bloodstream in each year of the study, stratified by group, is pro- infections (CRBSI) were the most important compli- vided in Supplement S1, with a footnote providing fur- cations reported from 15 Spanish hospitals, with 821 ther details on the function of these hospitals. bloodstream infections (BSI) episodes, representing almost 25% of all nosocomial BSI [4]. According to a Data from each hospital are continuously monitored. paper from 2006 reporting a systematic review of 200 The annual incidence rates are compared with the published prospective studies, the incidence rate of hospitals’ records from previous years, and with the CRBSI per 1,000 catheter days generally ranges from aggregate data compiled in the VINCat programme. 0.1 episodes for PVC to 2.7 episodes for CVC [5]. Results are presented at general clinical sessions and a public annual report is published within the VINCat CRBSI are an important cause of morbidity and mortal- website [9]. This study presents data from all episodes ity. Patients with these infections usually have more recorded between January 2007 and December 2019. severe underlying illness and are more likely to have other healthcare-associated infections (HAI) during Definitions their admission, with a mortality ranging from 12% to Terms used in this study are described as follows [10]. 25%, according to a prospective nationwide surveil- lance study in the United States (US) from March 1995 Catheter-related bloodstream infection through September 2002 [6]. CRBSI are also associated A bacterial infection in a patient using a venous catheter with longer hospital admissions and higher economic is defined with the following criteria. It has to be detected costs [7]. with at least one set of blood cultures obtained from a peripheral vein and two sets in the case of habitual skin- The application of prevention programmes in intensive colonising microorganisms (coagulase-negative staph- care units (ICU) in recent decades has resulted in signif- ylococci (CoNS), Micrococcus spp., Propionibacterium icant reductions of CRBSI incidence rates [8]. Bundles acnes, Bacillus spp. and Corynebacterium spp.). These of preventive measures have been applied including cultures must be associated with clinical manifestations hand hygiene, use of chlorhexidine alcohol solution for of infection (fever > 37.5 °C, chills and/or hypotension) skin antisepsis, full barrier precautions, daily review of and the absence of any apparent alternative source of need for catheterisation and femoral site avoidance [8]. BSI. The Infection Control Catalan Programme (VINCat) was These conditions must be accompanied by one or more launched in 2006, with the main objective of reducing of the following: the incidence of HAI through continuous active moni- toring and implementation of preventive programmes. (i) semiquantitative culture of catheter tip (> 15 col- Surveillance of CRBSI at the hospitals in our region is ony forming units (CFU) per catheter segment) or quan- a priority [9]. The aim of this study is to describe the titative culture (> 103 CFU per catheter segment), with changes in the incidence and epidemiology of CRBSI detection of the same microorganism as in blood cul- in the hospitals participating in the VINCat programme tures obtained from the peripheral blood; over a 13-year period. (ii) quantitative blood cultures with detection of the Methods same microorganism, with a difference of 5:1 or greater between the blood obtained from any of the lumens of Setting a venous catheter and that obtained from a peripheral BSI associated with the use of venous catheters is vein by puncture; continuously monitored under the VINCat programme. Participation of hospitals in VINCat is voluntary. All (iii) difference in time to positivity of the blood cul- detected nosocomial episodes of CRBSI, which are tures of above 2 hours between cultures obtained diagnosed in adult patients at each of the participating from a peripheral vein and from the lumen of a venous hospitals are prospectively followed and reported to catheter; the VINCat programme by the infection control teams. The detection of cases is based on the daily evaluation (iv) presence of inflammatory signs or purulent secre- of all patients with positive blood cultures. This infor- tions in the insertion point or the subcutaneous tunnel mation is provided to the infection control team by the of a venous catheter. A culture of the secretion showing microbiology laboratory at each hospital. The applica- growth of the same microorganism as the one detected tion of precise definitions allows the identification of in the blood cultures is also recommended (but not CRBSI. obligatory); www.eurosurveillance.org 3 Table 2 4 Annual incidence rate per 1,000 patient days of catheter-related bloodstream infections diagnosed at VINCat hospitals stratifying by hospital ward and catheter type, Catalonia, Spain, 2007–2019 (n = 9,290) Incidencea Variables 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Number of CRBSI IRR (95%CI)c 692 763 695 658 652 750 578 688 703 678 619 688 533 episodesb Number of 2,347.947 2,991,053 2,970,611 3,067,156 3,161,235 3,416,998 3,497,772 3,700,237 3,818,378 3,817,357 3,855,069 3,936,649 3,953,391 patient days CVC 2.01 1.96 1.48 1.3 1.15 1.06 0.78 0.77 0.62 0.63 0.52 0.46 0.34 0.87 (0.86–0.88) PVC 0.09 0.14 0.13 0.06 0.06 0.09 0.09 0.03 0.07 0.07 0.03 0.07 0.04 0.92 (0.88–0.96) ICU PICVC 0.23 0.12 0.20 0.18 0.18 0.14 0.13 0.18 0.15 0.17 0.13 0.13 0.13 0.97 (0.94–1) Subtotal 2.33 2.22 1.81 1.55 1.39 1.29 1.0 0.98 0.83 0.88 0.68 0.66 0.50 0.88 (0.87–0.89) CVC 0.14 0.13 0.1 0.1 0.08 0.08 0.07 0.06 0.06 0.05 0.05 0.05 0.03 0.90 (0.89–0.92) Medical PVC 0.06 0.04 0.06 0.06 0.07 0.08 0.05 0.06 0.07 0.06 0.06 0.09 0.07 1.03 (1.00 –1.05) wards PICVC 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.02 0.02 0.03 0.03 0.02 1.08 (1.05–1.11) Subtotal 0.21 0.18 0.17 0.17 0.16 0.18 0.13 0.14 0.15 0.14 0.14 0.16 0.13 0.97 (0.96–0.98) CVC 0.14 0.13 0.12 0.10 0.10 0.10 0.09 0.10 0.11 0.09 0.07 0.07 0.06 0.94 (0.93–0.96) Surgical PVC 0.05 0.02 0.02 0.03 0.02 0.03 0.01 0.03 0.03 0.03 0.03 0.03 0.02 0.98 (0.94–1.02) wards PICVC 0.02 0.01 0.03 0.01 0.01 0.03 0.02 0.03 0.02 0.04 0.03 0.03 0.03 1.05 (1.01–1.09) Subtotal 0.21 0.17 0.18 0.14 0.13 0.16 0.12 0.16 0.15 0.15 0.13 0.13 0.10 0.97 (0.95–0.98) Total 0.29 0.26 0.23 0.21 0.21 0.22 0.17 0.19 0.18 0.18 0.16 0.17 0.13 0.94 (0.93–0.96) CI: confidence interval; CRBSI: catheter-related bloodstream infection; CVC: central venous catheter; ICU: intensive care unit; IRR: incidence rate ratio; PVC: peripheral venous catheter; PICVC: peripherally- inserted central venous catheter; VINCat: Infection Control Catalan Programme. a Incidence rates per 1,000 patient days for each year of the study are presented in the columns, unless otherwise specified by the row header. b Episodes diagnosed in hospitals, where the number of patient days were available. c Incidence risk ratio by year (with the inclusion of CRBSI episodes diagnosed in hospitals, for which patient days were available that year). www.eurosurveillance.org Figure 1 Annual incidence rate of catheter-related bloodstream infection adjusted per 1,000 patient days stratified by (A) catheter type and (B) hospital unit type, Catalonia, Spain, 2007–2019 (n = 9,290) A. Incidence rate per catheter type B. Incidence rate per hospital unit type 0.25 1,000 3 1,000 Incidence rate of CRBSI per 1,000 patient days Incidence rate of CRBSI per 1,000 patient days 0.20 750 750 2 0.15 Number of CRBSI Number of CRBSI 500 500 0.10 1 250 250 0.05 0.00 0 0 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Year Year CVC PICVC PVC Intensive care unit Medical unit Surgical unit CRBSI: catheter-related bloodstream infection; CVC: central venous catheter; PIVC: peripherally-inserted central venous catheters; PVC: peripheral venous catheters. (v) resolution of clinical signs and symptoms after cath- at an outpatient service; CRBSI associated with arterial eter withdrawal with or without appropriate antibiotic catheters. treatment. For the clinical diagnosis of PVC-BSI, the presence of signs of phlebitis is required (induration, Statistical analysis pain or signs of inflammation at the insertion point or Categorical variables were presented as the number the catheter route). This last criterion is the only one of cases and percentages. Continuous variables were that is not considered in the point prevalence survey presented as means and standard deviation (SD) or protocol for microbiology confirmed catheter-related medians and interquartile range (IQR), depending on infection (CRI3-CVC) [11]. whether the distribution was normal or non-normal. Normality of variables was assessed graphically (quan- Type of catheter tile-quantile-plot and density plots). A CVC is defined as a catheter inserted in a subclavian, jugular or femoral vein, percutaneously (with or with- The annual incidence rate of CRBSI was obtained out tunneling). Fully implanted catheters (type Port-a- by dividing the total number of episodes of CRBSI Cath) are not included in the surveillance programme. with the total number of patient days in 1 year A peripherally-inserted central venous catheter (PICVC) and this was then adjusted for 1,000 patient days is a catheter inserted percutaneously through a vein to give the annual incidence rate of CRBSI diag- in the forearm (usually a basilica vein). Its distal end nosed per 1,000 patient days (annual incidence of reaches the right heart cavities. These catheters are CRBSI diagnosed per 1,000 patient days  = total num- generally used in the same way as conventional CVCs. ber of CRBSIs detected in 1 year x 1,000 /number of A PVC is a short- or medium-length catheter inserted patient days). percutaneously in a peripheral location (usually an arm or forearm). A negative binomial model was used to assess the trend over the study period of the rate CRBSIs diag- Hospital wards nosed at VINCat hospitals per year. The number of Hospital wards where CRBSI are identified are classi- admissions per year was used as offset, the num- fied as medical, surgical, or ICUs. ber of events (i.e. CRBSI) as the dependent variable, and the year as the main independent variable. The Exclusion criteria effect of hospital ward, catheter type and the inter- Episodes in the following patients were not included in action between year and catheter type, catheter use the study: patients up to 18 years of age; outpatients and aetiology were also assessed. Stratified analysis with a hospital stay of less than 48 hours at time of according to hospital ward and catheter type was also BSI detection; patients in whom CRBSI was detected performed. From each scenario, the annual incidence www.eurosurveillance.org 5 Figure 2 Number of expected catheter-related bloodstream infection per year, stratified by (A) catheter type, (B) catheter use and (C) aetiology, Catalonia, Spain, 2007–2019 (n = 9,290) A.Catheter type B. Catheter use Expected catheter−related bloodstream infection Expected catheter−related bloodstream infection 600 600 400 400 200 200 0 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Year Year CVC PICVC PVC Haemodyalisis/parenteral nutrition Other C. Aetiology Expected catheter−related bloodstream infection 600 400 200 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Year CoNs Staphylococcus aureus Gram−negative bacilli Other CoNS: coagulase-negative staphylococci; CVC: central venous catheter; PIVC: peripherally-inserted central venous catheters; PVC: peripheral venous catheters. Gram-negative bacilli include Enterobacteriaceae and Pseudomonas aeruginosa. rate of CRBSIs diagnosed per 1,000 patient days was accordance with standard operating procedures. Every reported, as was the incidence rate ratio (IRR) with its microorganism is identified using standard microbio- 95% confidence interval (CI). The interpretation of IRR logical techniques at each centre. was focused on the annual rate increase or decrease. The expected annual numbers of CRBSIs were plotted. Results To estimate catheter days, we obtained the total During the study period, a total of 9,290 CRBSI epi- adult patient days from all the centres during the study sodes were reported (Table 1). The incidence rate was period and then multiplied this total by the mean prev- 0.20 episodes/1,000 patient days. Patients’ median alence rate of catheter daily use over this period [12]. age was 64.1 years, and 36.6% of patients were female All analyses were performed with a two-sided signifi- (information on sex was collected as a binary variable). cance level of 0.05 and conducted with the R software BSI was diagnosed a median of 10 days (IQR: 6–17) after version 4.0.2 [13]. admission and a median of 3 days (IQR: 0–14) after cath- eter insertion. In total, 26.3% of episodes occurred in Microbiology the ICU, while 42.1% and 31.6% were acquired in medi- Two sets of two blood samples from a peripheral vein cal and surgical wards, respectively. Among the whole are usually obtained from all patients with a suspected cohort, 62.7% episodes were related to CVC, 24.1% to BSI. An additional blood sample is also collected PVC and 13.3% to PICVC, while catheter use was dis- through the catheter. When possible, the catheter tip tributed as haemodialysis (4.8%), parenteral nutri- is cultured after removal. Blood samples are processed tion (26.6%) and other uses (68.7%). Meanwhile, the at the microbiology laboratories of each centre in most frequent responsible microorganisms were CoNS 6 www.eurosurveillance.org Table 3 Incidence rates of episodes acquired in the ICU, Negative binomial model with interaction between years regardless of catheter type (CVC, PVC or PICVC), fol- and episode characteristics, Catalonia, Spain, 2007–2019 (n = 9,290) lowed a downward trend. Meanwhile, in the medical wards annual incidence rates of CVC BSI fell signifi- cantly (IRR: 0.90; 95%CI: 0.89–0.92) but those of PVC Predictors IRR CI p and PICVC increased (IRR: 1.03; 95%CI: 1.00–1.05 Interaction between years and catheter types and IRR: 1.08; 95%CI: 1.05–1.11, respectively). In Intercept 0.00 0.00–0.00 < 0.001 the surgical wards, CVC episodes fell significantly Year 0.91 0.90–0.92 < 0.001 (IRR: 0.94; 95%CI: 0.93–0.96) while CRBSI associated PICVC vs CVC 0.20 0.16–0.23 < 0.001 with PICVC increased (IRR: 1.05; 95%CI: 1.01–1.09) PVC vs CVC 0.09 0.07–0.11 < 0.001 (Figure 1B, Table 2). Interaction between PICVC and 1.11 1.08–1.14 < 0.001 year An estimation of incidence rate of CRBSI Interaction between PVC and year 1.14 1.12–1.17 < 0.001 adjusted per 1,000 catheter days was carried out to Interaction between years and catheter use overcome possible bias associated with catheter use, Intercept 0.00 0.00–0.00 < 0.001 and a significant downward trend in the incidence Year 0.93 0.92–0.94 < 0.001 rates of CRBSI associated to CVC was also observed, Other uses vs haemodialysis/PN 1.71 1.47–1.98 < 0.001 both in ICUs and conventional wards (Supplement S1). Interaction between other uses 1.04 1.02–1.06 < 0.001 and year Interaction between years and episode Interaction between years and aetiology characteristics Intercept 0.00 0.00–0.00 < 0.001 The interactions between year and catheter type or Year 0.93 0.92–0.95 < 0.001 catheter use were respectively statistically significant Staphylococcus aureus vs CoNS 0.41 0.34–0.49 < 0.001 (Figure 2A, Figure 2B, Table3). For catheter use, signifi- Gram-negative bacilli vs CoNS 0.51 0.43–0.61 < 0.001 cance was observed despite the absolute number of Other vs CoNS 0.17 0.14–0.22 < 0.001 episodes of all three categories (parenteral nutrition, Interaction haemodialysis and other uses) falling during the study between Staphylococcus 1.07 1.04–1.09 < 0.001 period. On the other hand, the category of ‘other uses’ aureus and year increased compared with the other two (Table 3). The Interaction between Gram- 1.02 0.99–1.04 0.138 interaction between year and aetiology was statisti- negative bacteria and year cally significant, with a significant downward trend in Interaction between other the rate of episodes caused by CoNS. Simultaneously, 0.99 0.96–1.02 0.405 pathogens and year incidence rates of S. aureus rose significantly (Figure CI: confidence interval; CoNS: coagulase-negative staphylococci; 2C, Table3). Figure 3 shows the annual incidence rate CVC: central venous catheter; IRR: incidence rates ratios; PICVC: peripherally-inserted central venous catheter; PN: parenteral of CRBSI adjusted per 1,000 patient days stratified by nutrition; PVC: peripheral venous catheter. catheter type, catheter use and microorganism. Gram-negative bacilli include Enterobacteriaceae and Pseudomonas aeruginosa. Discussion This study is a comprehensive description of the changes in the epidemiology of CRBSI at hospitals (39.5%), followed by Staphylococcus aureus (24.6%) in our healthcare region in Spain. Our large series of and Enterobacteriaceae (18.4%). Candida species CRBSI reveals a significant increase in the incidence accounted for 5.9% episodes, Pseudomonas aerugi- rate of PVC and PICVC CRBSI in conventional wards, in nosa 5.2% and Enterococcus spp. 5.0%. parallel to a notable reduction in all CRBSI in ICUs. Annual incidence trends During the study period, three interventional pro- The annual incidence rate of CRBSI fell from 0.29 epi- grammes were implemented in hospitals belonging to sodes per 1,000 patient days in 2007 to 0.13 in 2019 the VINCat. First, since 2007 the comparative results (IRR: 0.94; 95%CI: 0.93–0.96) (Table 2). This down- (benchmarking) of CRBSI surveillance are shared with ward trend was mostly associated with the progres- professionals involved in prevention of complications sive decrease in the annual incidence rate of CVC associated with vascular catheters [14]. As was already CRBSI, which ranged from 0.22 per 1,000 patient days demonstrated in 1985 by a study in the US, implemen- in 2007 (507 episodes) to 0.05 (217 episodes) in tation of intensive infection surveillance and control 2019 (Figure 1A). A downward trend in CRBSI epi- programmes can reduce the rate of nosocomial infec- sodes acquired in the ICU was also observed tions by up to 30% [15]. during the study period, from 2.33 to 0.5 epi- sodes/1,000 patient days (IRR: 0.88; 95%CI: 0.87– The second intervention implemented at our hospi- 0.89), while the annual incidence rate of episodes tals was the so-called ‘Bacteraemia Zero Programme’. acquired in medical and surgical wards presented It started in 2009 at 192 Spanish ICUs, including all in significantly lower decreases (IRR: 0.97; 95%CI: 0.96– hospitals in Group I and resulted in a reduction in the 0.98 and IRR: 0.97; 95%CI: 0.95–0.98, respectively). risk of CRBSI of 50% in 18 months [16]. Consistently, www.eurosurveillance.org 7 Figure 3 Annual incidence rate of catheter-related bloodstream infection adjusted per 1,000 patient days stratified by (A) catheter type and aetiology and (B) catheter type and use, Catalonia, Spain, 2007–2019 (n = 9,290) A. Catheter type and aetiology B. Catheter type and catheter use 800 800 0.15 0.15 Incidence rate of CRBSI per 1,000 patient days Incidence rate of CRBSI per 1,000 patient days 600 600 0.10 0.10 Number of CRBSI Number of CRBSI 400 400 0.05 0.05 200 200 0.00 0 0.00 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Year Year CVC−CoNs CVC−haemodialysis/parenteral nutrition CVC−other CVC−other PICVC−CoNs PICVC−haemodialysis/parenteral nutrition PICVC−other PICVC−other PVC−CoNs PVC− haemodialysis/parenteral nutrition PVC−other PVC−other CRBSI: catheter-related bloodstream infection; CoNS: coagulase-negative staphylococci; CVC: central venous catheter; PICVC: peripherally- inserted central venous catheter; PVC: peripheral venous catheter. in recent years, prevention programmes in ICUs have trend in the PVC CRBSI incidence rate in medical wards, enabled to bring down CVC CRBSI incidence rates. One in agreement with recent prospective studies [22-25]. of the most influential of these programmes was imple- mented in 2004 in 103 ICUs from 67 Michigan hospitals, Our results suggest the need for programmes in con- as described by Pronovost et al. [8]. These authors ventional wards to prevent PVC CRBSI, similar to the reported a 66% reduction in CRBSI rates, which fell to ones conducted in ICUs in recent years. Indeed, a zero infections per 1,000 catheter days 18 months after multimodal intervention performed in conventional the programme start [8]. A meta-analysis including wards of a selection of Spanish hospitals was associ- 43 studies, published in 2014, also showed a decrease ated with a reduction of CVC CRBSI incidence in this in CVC CRBSI incidence rates after the implementation setting. However, no impact on PVC CRBSI incidence of prevention programmes in ICUs [17]. Therefore, the was observed. It was concluded that compliance with impact of these programmes in ICUs is beyond doubt. the bundles related to catheter insertion and mainte- nance was lower for PVCs, probably due to the lower The third intervention implemented in VINCat hospitals perception of risk of complications with their use [18]. was conducted in 2010 and included the monitoring of However, we stress that other similar experiences CRBSI related to all PVC and CVC catheters inserted at achieved significant reductions in PVC CRBSI incidence conventional wards of 11 hospitals of the VINCat pro- rates in conventional settings [26-28]. gramme [18]. Studies assessing the incidence rate of CRBSI in conventional wards compared to ICU present In this study, the incidence rate of CRBSI caused by a wide range of results [19-21]. In our study, the inci- CoNS followed a downward trend, in agreement with dence rate of CVC CRBSI in the ICU was higher than a previous report [29]. In contrast, it was observed that observed in conventional settings, but the differ- that, relative to the CoNS, the Gram-negative bacilli ence between the rates fell over the course of the study incidence increased, as recently described elsewhere period. In the present study we also found an upward [25,30]; the risk was associated with solid organ transplantation, prior use of antibiotics, previous 8 www.eurosurveillance.org Ethical statement neurological or gastrointestinal conditions, and longer hospital stay [25,29,31]. Although femoral catheters Participation in the VINCat programme is voluntary at a hos- have also been associated with higher infection rates pital level. Data confidentiality is guaranteed. This study was evaluated and approved by the Parc Taulí Hospital Research due to Gram-negative bacilli [8,32], we did not observe Ethics Committee. The study did not require ethical approval this relationship (data not shown), perhaps because or informed consent statement. prevention programmes in recent years have argued against their use and their insertion is less frequent today than in the past [33]. Other studies have under- lined the importance of hand hygiene to prevent CRBSI caused by Gram-negative bacteria [28]. Acknowledgements This study was carried out as part of LBC’s doctoral thesis in Similarly, S. aureus CRBSI episodes also increased Medicine of Universitat Autònoma de Barcelona (Department during the study period. These episodes are frequently of Medicine). OG received a personal research grant from the “Pla estratègic de recerca i innovació en salut (PERIS) 2019- associated with catheters inserted in emergency 2021” (Departament de Salut. Generalitat de Catalunya). rooms, where contamination during catheter insertion This work was supported by Spanish Ministry of Economics is frequent [17,34]. Interestingly, most cases of CRBSI and Competitiveness. Instituto de Salud Carlos III Expedient: in our study were diagnosed within the first days after PI20/01563, the Red Española de Investigación en Patología hospital admission and catheter insertion. The rapid- Infecciosa (REIPI) and Centro de Investigación Biomédica en Red (CIBER) de Enfermedades Respiratorias (CIBERES ity of the occurrence of these episodes was probably CB06/06/0037), an initiative of the Instituto de Salud Carlos due to the lack of aseptic conditions at the time of III, Madrid, Spain. insertion, while the episodes that occurred later on were associated with catheter maintenance [22]. This Additional members of the VINCat Programme: We thank means that special attention should be paid to these members of the VinCat program for the surveillance and col- lection of CRBSI data at each hospital. These members include catheters. Notably, interventions applied to reduce the Alfredo Jover, Dolors Castellana, Montserrat Olona, Antonia incidence of CRBSI have a greater impact on episodes García Pino, Josep Rebull Fatsini, Mª France Domènech, caused by S. aureus [26,27] and Gram-negative bacilli Miquel Pujol, Ana Hornero, Carmen Ardanuy Tisaire, Dàmaris [28] than those caused by other pathogens. Berbel Palau, Joaquín López-Contreras, Engracia Fernández, Xavier Salgado, Dolors Domènech, Ana Lérida, Lydia Martin, Nieves Sopena, Irma Casas Garcia, Rafel Pérez, Encarna The main limitation of the study is the lack of clinical Maraver, Eva Palau, Pepi Serrats, José Antonio Martínez, information regarding the presence of chronic diseases Gemina Santana, Ana Martínez, Lourdes Ferrer, Mª José or other health disorders that might have influenced Moreno, Esther Calbo, Carolina Porta, Alex Smithson, Maria the risk of CRBSI. In addition, CRBSI incidence rates de la Roca Toda, Teresa Aliu, Susanna Camps, Montserrat were adjusted by patient days and not by catheter days, Ortega, Vicens Diaz-Brito, Encarna Moreno, Carme Agustí, Miquel Perea Garcia, Marta Andrés, Laura Grau Palafox, as this strategy would not be achievable for surveil- Raquel Carrera, Anna Besolí, Juan Pablo Horcajada, Cristina lance of all types of catheters inserted at all hospital Gonzalez, Jordi Cuquet, Demelsa Maria Maldonado López, wards, and not only in ICUs. To overcome a possible Rosa Benítez, Mireia Duch, David Blancas, Esther Moreno, bias associated to catheter use, we estimated the Naiara Villalba, Sara Martínez, Àngels García Flores, Roser incidence of CRBSI per 1,000 catheter days, which gave Ferrer, Josep Bisbe, Montse Blascó, Antoni Castro Salom, Ana Felisa López, Joan Espinach Alvarós, Àngels Perez, similar results as the adjustment by patient days. Also, David Castander, Elisabet Calaf, Mercè Clarós, Núria Bosch due to the multicentre nature of the study, the inter- Ros, Irene Montardit, Roser Porta, Pilar De la Cruz Sol, Mª ventions and control programmes may not have been Rosa Coll Colell, Rosa García Penche Sanches, Josep Maria homogeneous across the different hospitals. To over- Tricas, Eva Redon, Montse Brugués, Laura Linares, Maria come this limitation, the VINCat programme attempts Cusco, Mª Pilar Barrufet‚ Elena Vidal, Sandra Barbadillo, Mariló Marimón, Yolanda Meije, Montserrat Vaqué, M. Rosa to standardise definitions and preventive actions. In Laplace Enguinados, Blanca Vila, Ana Guadalupe Coloma, addition, the reported data are audited annually, and Lucrecia López, Magda Campins, Benito Almirante, Carme deviations are analysed together with the person in Ferrer, Natalia Juan Serra, Josep Farguell Carrera, Àngels charge at each centre. Garc¡a Flores, Roser Ferrer, Marta Milian Sanz, Alexandra Moise, Ana Mª Jiménez Zarate, M. Carmen Eito Navasal, Maria Gracia Garcia Ramirez, Ana Mª Jiménez Zarate, Mar Conclusion Armario Fernández. This surveillance programme enabled us to trace the changes in the epidemiology of CRBSI, which remains an important HAI. The present study highlights the Funding need for interventional programmes focusing on PVC, especially in non-ICU wards. Our group is currently Spanish Ministry of Economics and Competitiveness. leading a prospective preventive programme at hospi- Instituto de Salud Carlos III Expedient: PI20/01563. tals in Catalonia that aims to reduce the rate of CRBSI in conventional wards. Conflict of interest None declared. www.eurosurveillance.org 9 Authors’ contributions 14. Almirante B, Limón E, Freixas N, Gudiol F, VINCat Program. Laboratory-based surveillance of hospital-acquired catheter- Laia Badia-Cebada analysed data and wrote the article, related bloodstream infections in Catalonia. Results of the VINCat Program (2007-2010). Enferm Infecc Microbiol Clin. 2012;30(Suppl 3):13-9. https://doi.org/10.1016/S0213- Judit Peñafiel did the statistical analysis 005X(12)70091-5 PMID: 22776149 15. Haley RW, Culver DH, White JW, Morgan WM, Emori TG, Munn Patrick Saliba, Enric Limón and Miquel Pujol coordinate the VP, et al. The efficacy of infection surveillance and control VinCat Program and participated with the data analyses programs in preventing nosocomial infections in US hospitals. Am J Epidemiol. 1985;121(2):182-205. https://doi.org/10.1093/ Marta Andrés, Jordi Càmara, Dolors Domenech, Emili oxfordjournals.aje.a113990 PMID: 4014115 Jiménez-Martínez, Anna Marrón, Encarna Moreno, Virginia 16. Palomar M, Álvarez-Lerma F, Riera A, Díaz MT, Torres F, Agra Y, et al. , Bacteremia Zero Working Group. Impact of a Pomar, Montserrat Vaqué, Úrsula Masats and Oriol Gasch national multimodal intervention to prevent catheter-related are members of the Catheter-related infections study group. bloodstream infection in the ICU: the Spanish experience. Crit Care Med. 2013;41(10):2364-72. https://doi.org/10.1097/ Oriol Gasch leaded the study and the analyses and wrote the CCM.0b013e3182923622 PMID: 23939352 article. 17. Blot K, Bergs J, Vogelaers D, Blot S, Vandijck D. Prevention of central line-associated bloodstream infections through quality improvement interventions: a systematic review and meta-analysis. Clin Infect Dis. 2014;59(1):96-105. https://doi. org/10.1093/cid/ciu239 PMID: 24723276 References 18. Freixas N, Bella F, Limón E, Pujol M, Almirante B, Gudiol F. 1. Zarb P, Coignard B, Griskeviciene J, Muller A, Vankerckhoven Impact of a multimodal intervention to reduce bloodstream V, Weist K, et al. . The European Centre for Disease Prevention infections related to vascular catheters in non-ICU wards: a and Control (ECDC) pilot point prevalence survey of healthcare- multicentre study. Clin Microbiol Infect. 2013;19(9):838-44. associated infections and antimicrobial use. Euro Surveill. https://doi.org/10.1111/1469-0691.12049 PMID: 23130638 2012;17(46):20316. https://doi.org/10.2807/ese.17.46.20316- 19. Marschall J, Leone C, Jones M, Nihill D, Fraser VJ, Warren DK. en PMID: 23171822 Catheter-associated bloodstream infections in general medical 2. Sociedad Española de Medicina Preventiva. Estudio EPINE- patients outside the intensive care unit: a surveillance study. EPPS no 30: 2019. Estud EPiNE. 2019; Infect Control Hosp Epidemiol. 2007;28(8):905-9. https://doi. 3. Zingg W, Metsini A, Balmelli C, Neofytos D, Behnke M, org/10.1086/519206 PMID: 17620235 Gardiol C, et al. . National point prevalence survey on 20. Zingg W, Sax H, Inan C, Cartier V, Diby M, Clergue F, et al. healthcare-associated infections in acute care hospitals, Hospital-wide surveillance of catheter-related bloodstream Switzerland, 2017. Euro Surveill. 2019;24(32). https:// infection: from the expected to the unexpected. J Hosp Infect. doi.org/10.2807/1560-7917.ES.2019.24.32.1800603 PMID: 2009;73(1):41-6. https://doi.org/10.1016/j.jhin.2009.05.015 31411135 PMID: 19646788 4. Rodríguez-Baño J, López-Prieto MD, Portillo MM, Retamar P, 21. Tedja R, Gordon SM, Fatica C, Fraser TG. The descriptive Natera C, Nuño E, et al. . Epidemiology and clinical features of epidemiology of central line-associated bloodstream infection community-acquired, healthcare-associated and nosocomial among patients in non-intensive care unit settings. Infect bloodstream infections in tertiary-care and community Control Hosp Epidemiol. 2014;35(2):164-8. https://doi. hospitals. Clin Microbiol Infect. 2010;16(9):1408-13. https:// org/10.1086/674856 PMID: 24442079 doi.org/10.1111/j.1469-0691.2010.03089.x PMID: 19845694 22. Mermel LA. Short-term Peripheral Venous Catheter-Related 5. Maki DG, Kluger DM, Crnich CJ. The risk of bloodstream Bloodstream Infections: A Systematic Review. Clin Infect Dis. infection in adults with different intravascular devices: 2017;65(10):1757-62. https://doi.org/10.1093/cid/cix562 a systematic review of 200 published prospective PMID: 29020252 studies. Mayo Clin Proc. 2006;81(9):1159-71. https://doi. 23. Pujol M, Hornero A, Saballs M, Argerich MJ, Verdaguer org/10.4065/81.9.1159 PMID: 16970212 R, Cisnal M, et al. Clinical epidemiology and outcomes of 6. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel peripheral venous catheter-related bloodstream infections at RP, Edmond MB. Nosocomial bloodstream infections in a university-affiliated hospital. J Hosp Infect. 2007;67(1):22-9. US hospitals: analysis of 24,179 cases from a prospective https://doi.org/10.1016/j.jhin.2007.06.017 PMID: 17719678 nationwide surveillance study. Clin Infect Dis. 2004;39(3):309- 24. Guembe M, Pérez-Granda MJ, Capdevila JA, Barberán J, Pinilla 17. https://doi.org/10.1086/421946 PMID: 15306996 B, Martín-Rabadán P, et al. , NUVE Study Group. Nationwide 7. Stevens V, Geiger K, Concannon C, Nelson RE, Brown J, study on peripheral-venous-catheter-associated-bloodstream Dumyati G. Inpatient costs, mortality and 30-day re-admission infections in internal medicine departments. J Hosp Infect. in patients with central-line-associated bloodstream 2017;97(3):260-6. https://doi.org/10.1016/j.jhin.2017.07.008 infections. Clin Microbiol Infect. 2014;20(5):O318-24. https:// PMID: 28716670 doi.org/10.1111/1469-0691.12407 PMID: 24112305 25. Ripa M, Morata L, Rodríguez-Núñez O, Cardozo C, Puerta- 8. Pronovost P, Needham D, Berenholtz S, Sinopoli D, Chu H, Alcalde P, Hernández-Meneses M, et al. Short-term peripheral Cosgrove S, et al. An intervention to decrease catheter- venous catheter-related bloodstream infections: Evidence for related bloodstream infections in the ICU. N Engl J Med. increasing prevalence of Gram-negative microorganisms from 2006;355(26):2725-32. https://doi.org/10.1056/NEJMoa061115 a 25-year prospective observational study. Antimicrob Agents PMID: 17192537 Chemother. 2018;62(11):e00892-18. https://doi.org/10.1128/ 9. VinCat. Barcelona: Catalan Health Service. [Accessed: AAC.00892-18 PMID: 30126952 19 Apr 2022]. Available from. https://catsalut.gencat. 26. Saliba P, Hornero A, Cuervo G, Grau I, Jimenez E, Berbel D, et cat/ca/proveidors-professionals/vincat/index. al. Interventions to decrease short-term peripheral venous html#googtrans(ca%7Cen) catheter-related bloodstream infections: impact on incidence 10. Nosocomial Infection Surveillance Programme at Catalan and mortality. J Hosp Infect. 2018;100(3):e178-86. https://doi. Hospitals (VINCat). 2015 Manual. 2015 Available from: https:// org/10.1016/j.jhin.2018.06.010 PMID: 29928942 catsalut.gencat.cat/web/.content/minisite/vincat/documents/ 27. Rhodes D, Cheng AC, McLellan S, Guerra P, Karanfilovska manuals/Manual-VINCat-2015-english.pdf D, Aitchison S, et al. Reducing Staphylococcus aureus 11. European Centre for Disease Prevention and Control (ECDC). bloodstream infections associated with peripheral intravenous Point prevalence survey of healthcare-associated infections cannulae: successful implementation of a care bundle at a and antimicrobial use in European acute care hospitals. large Australian health service. J Hosp Infect. 2016;94(1):86-91. Protocol version 5.3. Stockholm: ECDC; 2016. Available from: https://doi.org/10.1016/j.jhin.2016.05.020 PMID: 27346623 https://www.ecdc.europa.eu/sites/default/files/media/en/ 28. Garcia-Gasalla M, Arrizabalaga-Asenjo M, Collado-Giner C, publications/Publications/PPS-HAI-antimicrobial-use-EU- Ventayol-Aguiló L, Socias-Mir A, Rodríguez-Rodríguez A, et al. acute-care-hospitals-V5-3.pdf Results of a multi-faceted educational intervention to prevent 12. Trinh TT, Chan PA, Edwards O, Hollenbeck B, Huang B, Burdick peripheral venous catheter-associated bloodstream infections. N, et al. Peripheral venous catheter-related Staphylococcus J Hosp Infect. 2019;102(4):449-53. https://doi.org/10.1016/j. aureus bacteremia. Infect Control Hosp Epidemiol. jhin.2019.02.004 PMID: 30771370 2011;32(6):579-83. https://doi.org/10.1086/660099 PMID: 29. Braun E, Hussein K, Geffen Y, Rabino G, Bar-Lavie Y, Paul M. 21558770 Predominance of Gram-negative bacilli among patients with 13. R Core Team. (2020). R: A language and environment for catheter-related bloodstream infections. Clin Microbiol Infect. statistical computing. R Foundation for Statistical Computing, 2014;20(10):O627-9. https://doi.org/10.1111/1469-0691.12565 Vienna, Austria PMID: 24461043 10 www.eurosurveillance.org 30. Marcos M, Soriano A, Iñurrieta A, Martínez JA, Romero A, Cobos N, et al. Changing epidemiology of central venous catheter-related bloodstream infections: increasing prevalence of Gram-negative pathogens. J Antimicrob Chemother. 2011;66(9):2119-25. https://doi.org/10.1093/jac/dkr231 PMID: 21665905 31. Bouza E, Eworo A, Fernández Cruz A, Reigadas E, Rodríguez- Créixems M, Muñoz P. Catheter-related bloodstream infections caused by Gram-negative bacteria. J Hosp Infect. 2013;85(4):316-20. https://doi.org/10.1016/j.jhin.2013.08.008 PMID: 24183319 32. Lorente L, Jiménez A, Santana M, Iribarren JL, Jiménez JJ, Martín MM, et al. Microorganisms responsible for intravascular catheter-related bloodstream infection according to the catheter site. Crit Care Med. 2007;35(10):2424-7. https://doi. org/10.1097/01.CCM.0000284589.63641.B8 PMID: 17717493 33. O’Grady NP, Alexander M, Burns LA, Dellinger EP, Garland J, Heard SO, et al. , Healthcare Infection Control Practices Advisory Committee. Guidelines for the prevention of intravascular catheter-related infections. Am J Infect Control. 2011;39(4) Suppl 1;S1-34. https://doi.org/10.1016/j. ajic.2011.01.003 PMID: 21511081 34. Stuart RL, Cameron DR, Scott C, Kotsanas D, Grayson ML, Korman TM, et al. Peripheral intravenous catheter-associated Staphylococcus aureus bacteraemia: more than 5 years of prospective data from two tertiary health services. Med J Aust. 2013;198(10):551-3. https://doi.org/10.5694/mja12.11699 PMID: 23725270 License, supplementary material and copyright This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0) Licence. You may share and adapt the material, but must give appropriate credit to the source, provide a link to the licence and indicate if changes were made. Any supplementary material referenced in the article can be found in the online version. This article is copyright of the authors or their affiliated in- stitutions, 2022. www.eurosurveillance.org 11