Jean-François TIMSIT CHU Grenoble UJF/Inserm U 823 Conduite à tenir devant une suspicion d’infection liée aux cathéters en réanimation Jean-François TIMSIT CHU Grenoble UJF/Inserm U 823 Nice - Juin 2007

ILC : Le traitement depend de sévérité du sepsis maladies sous-jacente (immunodépression, prothèses). Micro-organismes identifiés ou suspectés HC positives ou négatives Utilité et facilité de l’abord veineux central Faible niveau de preuve 1

En cas de sepsis grave le cathéter DOIT être enlevé Deux constraintes : Eviter l’ablation inutile des CVCs (75% cases) et le risque associé de complications mécaniques Sauver les malades et éviter que l’infection se complique En cas de sepsis grave le cathéter DOIT être enlevé 8

2 situations Fièvre sans signes de sepsis sévère en réanimation Sepsis sévère de cause inconnu Ablation du CVC (or ou échange sur guide?) Quels antibiotiques? Comment dépister les complications et les traiter? Fièvre sans signes de sepsis sévère en réanimation Hémoculture positive Est il possible de conserver le cathéter sans risques?

Le cathéter? Ablation du cathéter Est associée à un plus grand nombre de guérison et une amélioration du pronostic 2. Diagnostic cathéter en place 3. Echange sur guide (GWX)

Schneegurt, MA. Wichita St. University, Microbiology 103. Biofilm formation Schneegurt, MA. Wichita St. University, Microbiology 103.

Why form a bioflim? Jefferson KK. FEMS. 2004;236:163-73.

Susceptibility of biofilm organisms Antibiotic MIC or MBC (mcg/mL) Effective [ ] vs. biofilm (mcg/mL) S. aureus (NCTC 8325-4) Vancomycin 2 (MBC) 20 P. aeruginosa (ATCC 27853) Imipenem 1 (MIC) >1,024ª E. coli (ATCC 25922) Ampicillin 2 (MIC) 512ª P. pseudomallei Ceftazidime 8 (MBC) 800 S. sanguis Doxycycline 0.063 (MIC) 3.15 ª Minimal biofilm eradication Adapted from Donlan RM, et al. Clin Microbiol Rev. 2002;15:167-93.

CVC maintenance is always risky… 1- Bacterias with slime production have an increased MICs and MBCs to ABx 2- The Biofilm increases the resistance of bacteria to ABt SCN culture Here is the MIC s of a slime producer Sraf epidermidis. For example the MIC of gentamycin was increased by more than a hundered fold in presence of slime. The MBCs of attached bacteria are also increased by a more than 128 fold. Finaly CVC maintenance is always associated with an increased risk of failure in case of catheter-related sepsis . However in three forth of the cases, catheters are removed but innocent!!! CVC maintenance is always risky…

Catheter removal and duration of candidemia Rex et al -Decrease of the duration of the candidemia New site 5.6 days vs Other 2.6 days - Bias: APACHE II 14.5 vs 16.9 p=0.03 Other catheter: 1.2 vs 1.8,p<0.001 - GWX: 6.3 + 1.8 j Post hoc analysis of a randomized study comparing fluconazole and amphotericine B Catheter removal should be prefered

Candidemia: CVC removal and mortality: meta-analysis Analyses are biased because CVCs removal is associated with severity… 4 studies with severity scores adjustment Anaissie 1998 (n=491) Retro adj OR: 2 (1.4-2.9, p=0.06) Nucci (CID) 1998 (n=54) Pro adj OR: NS Nucci (2) 1998 (n=145) Pro adj OR: 4.22 (2-11.6) Luzzatti 2000 (n=189) retro adj OR: 1.61 (1.01-2.63, p=0.047) The removal of all central venous catheters (CVCs) from all patients with candidemia is considered to be standard care. However, this practice is not always possible, and it is associated with significant cost and potential complications. To evaluate the effect of CVC removal on the outcome of patients with candidemia, a literature review was conducted that selected studies that evaluated CVC removal as a prognostic factor (of mortality) in candidemia, performed a multivariate analysis with odds ratios and 95% confidence intervals, and included a validated severity of illness score. Of 203 studies of candidemia, only 4 fulfilled these criteria. One study showed a benefit from CVC removal in a subset of 21 neutropenic patients; another study showed no benefit; and the remaining 2 studies showed a marginal benefit from this strategy. Although it is possible that removal of CVCs may reduce the rate of complications due to candidemia, including death, the findings of this literature review do not substantiate this consensus recommendation. Nucci – Clin Infect dis 2002; 34:591

Management of CVCs in patients with cancer and candidemia Management of CVCs in patients with cancer and candidemia Raad I et al – Clin Infect Dis 2004; 38:1119 1993-1998: 404 episodes of candidemia (50% ICU) with 1 CVCs for more than 1 days 3 categories Primary candidemia : 241 (60%) Secondary candidemia: 52 (13%) CVC related candidemia : 111 (27%) + tip cult (66) or quantitative BC > 5:1 (45) % Management of Central Venous Catheters in Patients with Cancer and Candidemia Issam Raad,1 Hend Hanna,1 Maha Boktour,1 Essam Girgawy,1 Hadi Danawi,3 Masoud Mardani,1 Dimitrios Kontoyiannis,1 Rabih Darouiche,2 Ray Hachem,1 and Gerald P. Bodey1 1Department of Infectious Diseases, Infection Control and Employee Health, University of Texas M.D. Anderson Cancer Center, 2Baylor College of Medicine, and 3University of Texas School of Public Health, Houston To determine the need and appropriate timing of catheter removal in patients with candidemia, the records for 404 patients with cancer and central venous catheters (CVCs) who developed candidemia during the period of 1993 1998 were retrospectively reviewed. Of the total 404 cases of candidemia, 241 (60%) were due to a primary source, 111 (27%) were catheter related, and 52 (13%) were secondary cases of candidemia caused by a source other than the catheter. Multivariate analysis showed that catheter removal 72 h after onset improved response to antifungal therapy exclusively in patients with catheter-related candidemia (P = .04). Clinical characteristics that suggested a noncatheter source for the candidemia were disseminated infection (P < .01), previous chemotherapy (P < .01), previous corticosteroid therapy (P = .02), and poor response to antifungal therapy (P < .03). CVC removal 72 h after onset should be considered in patients with suspected catheter-related candidemia who have no evidence of dissemination, recent corticosteroid therapy, or chemotherapy.

Is candidemia catheter-related Is candidemia catheter-related? Raad I et al – Clin Infect Dis 2004; 38:1119 111 catheter-related candidemia and 52 secondary candidemia No corticosteroids within 1 month: OR 3.5 (1.3-9.4), p=0.02 No chemotherapy within 1 month: OR 4.3 (1.5-13.3), p<0.01 Non disseminated infection * OR 9.7 (3.5-26.3), p<0.01 Good response to antifungal therapy** OR 2.9 (2.2-7.2), p=0.03 Management of Central Venous Catheters in Patients with Cancer and Candidemia Issam Raad,1 Hend Hanna,1 Maha Boktour,1 Essam Girgawy,1 Hadi Danawi,3 Masoud Mardani,1 Dimitrios Kontoyiannis,1 Rabih Darouiche,2 Ray Hachem,1 and Gerald P. Bodey1 1Department of Infectious Diseases, Infection Control and Employee Health, University of Texas M.D. Anderson Cancer Center, 2Baylor College of Medicine, and 3University of Texas School of Public Health, Houston To determine the need and appropriate timing of catheter removal in patients with candidemia, the records for 404 patients with cancer and central venous catheters (CVCs) who developed candidemia during the period of 1993 1998 were retrospectively reviewed. Of the total 404 cases of candidemia, 241 (60%) were due to a primary source, 111 (27%) were catheter related, and 52 (13%) were secondary cases of candidemia caused by a source other than the catheter. Multivariate analysis showed that catheter removal 72 h after onset improved response to antifungal therapy exclusively in patients with catheter-related candidemia (P = .04). Clinical characteristics that suggested a noncatheter source for the candidemia were disseminated infection (P < .01), previous chemotherapy (P < .01), previous corticosteroid therapy (P = .02), and poor response to antifungal therapy (P < .03). CVC removal 72 h after onset should be considered in patients with suspected catheter-related candidemia who have no evidence of dissemination, recent corticosteroid therapy, or chemotherapy. (*) Dissemination to non contiguous sites (**)Resolution of fever and chills, BC neg.

Outcome of candidemia: time of catheter removal after the first positive culture Raad I et al – Clin Infect Dis 2004; 38:1119 Management of Central Venous Catheters in Patients with Cancer and Candidemia Issam Raad,1 Hend Hanna,1 Maha Boktour,1 Essam Girgawy,1 Hadi Danawi,3 Masoud Mardani,1 Dimitrios Kontoyiannis,1 Rabih Darouiche,2 Ray Hachem,1 and Gerald P. Bodey1 1Department of Infectious Diseases, Infection Control and Employee Health, University of Texas M.D. Anderson Cancer Center, 2Baylor College of Medicine, and 3University of Texas School of Public Health, Houston To determine the need and appropriate timing of catheter removal in patients with candidemia, the records for 404 patients with cancer and central venous catheters (CVCs) who developed candidemia during the period of 1993 1998 were retrospectively reviewed. Of the total 404 cases of candidemia, 241 (60%) were due to a primary source, 111 (27%) were catheter related, and 52 (13%) were secondary cases of candidemia caused by a source other than the catheter. Multivariate analysis showed that catheter removal 72 h after onset improved response to antifungal therapy exclusively in patients with catheter-related candidemia (P = .04). Clinical characteristics that suggested a noncatheter source for the candidemia were disseminated infection (P < .01), previous chemotherapy (P < .01), previous corticosteroid therapy (P = .02), and poor response to antifungal therapy (P < .03). CVC removal 72 h after onset should be considered in patients with suspected catheter-related candidemia who have no evidence of dissemination, recent corticosteroid therapy, or chemotherapy.

Predictors of failure to respond to antifungal therapy Predictors of failure to respond to antifungal therapy Raad I et al – Clin Infect Dis 2004; 38:1119 Management of Central Venous Catheters in Patients with Cancer and Candidemia Issam Raad,1 Hend Hanna,1 Maha Boktour,1 Essam Girgawy,1 Hadi Danawi,3 Masoud Mardani,1 Dimitrios Kontoyiannis,1 Rabih Darouiche,2 Ray Hachem,1 and Gerald P. Bodey1 1Department of Infectious Diseases, Infection Control and Employee Health, University of Texas M.D. Anderson Cancer Center, 2Baylor College of Medicine, and 3University of Texas School of Public Health, Houston To determine the need and appropriate timing of catheter removal in patients with candidemia, the records for 404 patients with cancer and central venous catheters (CVCs) who developed candidemia during the period of 1993 1998 were retrospectively reviewed. Of the total 404 cases of candidemia, 241 (60%) were due to a primary source, 111 (27%) were catheter related, and 52 (13%) were secondary cases of candidemia caused by a source other than the catheter. Multivariate analysis showed that catheter removal 72 h after onset improved response to antifungal therapy exclusively in patients with catheter-related candidemia (P = .04). Clinical characteristics that suggested a noncatheter source for the candidemia were disseminated infection (P < .01), previous chemotherapy (P < .01), previous corticosteroid therapy (P = .02), and poor response to antifungal therapy (P < .03). CVC removal 72 h after onset should be considered in patients with suspected catheter-related candidemia who have no evidence of dissemination, recent corticosteroid therapy, or chemotherapy.

Proposed algorythm for candidemia Proposed algorythm for candidemia Raad I et al – Clin Infect Dis 2004; 38:1119 Management of Central Venous Catheters in Patients with Cancer and Candidemia Issam Raad,1 Hend Hanna,1 Maha Boktour,1 Essam Girgawy,1 Hadi Danawi,3 Masoud Mardani,1 Dimitrios Kontoyiannis,1 Rabih Darouiche,2 Ray Hachem,1 and Gerald P. Bodey1 1Department of Infectious Diseases, Infection Control and Employee Health, University of Texas M.D. Anderson Cancer Center, 2Baylor College of Medicine, and 3University of Texas School of Public Health, Houston To determine the need and appropriate timing of catheter removal in patients with candidemia, the records for 404 patients with cancer and central venous catheters (CVCs) who developed candidemia during the period of 1993 1998 were retrospectively reviewed. Of the total 404 cases of candidemia, 241 (60%) were due to a primary source, 111 (27%) were catheter related, and 52 (13%) were secondary cases of candidemia caused by a source other than the catheter. Multivariate analysis showed that catheter removal 72 h after onset improved response to antifungal therapy exclusively in patients with catheter-related candidemia (P = .04). Clinical characteristics that suggested a noncatheter source for the candidemia were disseminated infection (P < .01), previous chemotherapy (P < .01), previous corticosteroid therapy (P = .02), and poor response to antifungal therapy (P < .03). CVC removal 72 h after onset should be considered in patients with suspected catheter-related candidemia who have no evidence of dissemination, recent corticosteroid therapy, or chemotherapy.

Jensen et coll Arch Intern Med 2002;162:25-32

Absence of CVC removal is always risky… Candida sp: 427 CRB Death 41% vs 21 % N’guyen et al - Arch Intern Med 1995;155:2429 S. aureus: 50 CRB (retrospective) Persistent BC:11 vs 56% (p=0.01), Deaths: 5 vs 20% Malanovski GJ - Arch Intern Med 1995;155:1161 X. maltophilia: % cured: 49/49 vs 32/62 (p<0.0001) Elting et al - Medicine 1990;69:296 Boktour et al – Cancer2006; 106:1967 Gram negative bacili % relapse: 1/67 vs 5/5 (p<0.001) Hanna et al – ICHE 2004; 25:646 Enterococci (n=61) % cured: 5/13 vs 40/47 (p<0.01) especially if aminoglycosides are not associated with cell-wall agent Sandoe JA –JAC 2002; 50:577 CNS: Deaths: 4/36 vs 4/34, recurrence after 3 months: 1/36 vs 6/34 Raad et al ICHE 1992 Malanovski: Long-term (16) or short-term (34) CVCs

In case of tunnelitis Antimicrobials alone is not sufficient… Microorganisms Cured Failures (n=5) (n=15) S. aureus 1 1 P. aeruginosa 0 7 polymicrobial 1 5* Negative culture 3 2 * 4 with P. aeruginosa et 1 with P. maltophilia Benezra et al, Am. J. Med., 1988, 85, 495

Culture après ablation des cathéters Culture qualitative (trop peu spécifique à abandonner) Culture semi-quantitative + si > 15 cfu/ml Maki et al. N Engl J Med 1977; 296: 1305-1309 Culture quantitative: Portion endo et extra-luminale préférable ultrasonication Sherertz et al – J Clin Microbiol 1990 Vortexage dans 1 ml de RL stérile Brun-Buisson - Arch Int Med 1987; 147:873 9

Influence de la culture des KT sous antibiotiques actifs KT intrapéritonéaux/souris Infectés à S. epi puis traité par TEICO ou RMP A J1 culture neg ou micro-colonies Culture vs détection du mRNA (bactéries viables)+PCR quanti J2 Sensibilité >1000 cfu/ml >100 cfu/ml Contrôle 94% (30/32) TEICO 72% (49/68) 81% (55/68) RMP 86% (62/72) 94% (68/72) À J1 il n’y a que des microcolonies La technique mRNA permet de detecter les bactéries encore viable et de les quantifier Qualitative broth culture and quantitative culture on agar were compared with quantitative polymerase chain reaction (PCR) for the diagnosis of foreign body infections (FBI) in a rat model with and without exposure to systemic antibiotics (teicoplanin and rifampin). The 3 methods had a similar and high yield without antibiotics. Antibiotics decreased the number of CFU/foreign body and increased the number of culture-negative foreign bodies and the variability of the results in quantitative culture. The yield of broth culture remained high under antibiotics although prolonged incubation (2-5 days) was required. The yield of the PCR was equivalent or even superior (for teicoplanin) to the yield of broth culture. Quantitative PCR had a higher yield and lower variability than quantitative culture and was not affected by antibiotics. The simultaneous isolation of RNA from all samples indicated viability of the bacteria. Quantitative PCR seems a promising method for the diagnosis of FBI. Vandecasteele et al – Diagnostic Microbiology and Infectious Disease 48 (2004) 89–95

The CVC ? CVC removal Diagnosis catheter in place Direct examination Other methods based on culture results 3. Guidewire exchange (GWX)

Modes de colonisations Extraluminale Endoluminale

Endo ou extra-luminale? Nb KT/durée 139/8.6 156/ 15 109/18.2 113/23.9 22/20 400/23 42/ 114 Nb inf sys 53 11 6 28 20 24 Hub 12 1 3 21 14 9 8 Peau 30 4 3 7 2 5 Mixte 8 2 Cercenado 1990 Fan 1988 Cicco 1989 Salzman 1993 Linares 1985 Segura 1993 Weightman 1988

Diagnostic catheter in place A negative cutaneous swab culture of skin entry  100% Negative predictive value Paired (Peripheral/central) quantitative BC > 5/1 or Differential time to positivity of BC > 120 mn In the absence of severe sepsis, for exemple in case of fever without no evident cause which is a frequnet circumtance in the ICU, it shoud be interresting to perform diagnostic procedures without removing catheter It should be known that a negative cutaneous swab culture is a strong argument for the absence of cathetr-related sepsis. In this circumstance, paired peripheral and central blood culture with differential quantitative cultures of differential time to positivity is able to acurately make the diagnosis of CRS at least for long-term CVCs and oncologic patients. Se/Sp > 90% 10

Test diagnostic rapide Kite et al Lancet 1999; 354:1504 100 µL de sang par le KTC Traitement par l'acide édétique lyse et centrifugation puis pastilles de cytocentrifugation puis coloration acridine orange et Gram 100 champs, 2 colorations Gram + AOLC test Positif Négatif ILC+ 48 2 ILC- 5 57

Diagnosis of Catheter - Related Infections Endoluminal brush and Acridine Orange stain Se: 83% 50 CVC 50 CVC Se: 18% 17 cult + 18 cult + 15 AOLC + Sensibilisation de la technique grace à la brosse endoluminale 2 AOLC + Group 1: Acridine orange stain Group 2: Acridine orange stain and endoluminal brush Tighe et al. J Parent Enter Nutr 1996; 20: 215-218

Culture cutanée: valeur prédictive 134 CVC de réanimation, 70% S.clav. Durée d'insertion:10 + 6 jours écouvillonnage de 25 cm2 site d'insertion 75 cultures peau positives / 26 CVC > 103cfu/ml concordance bactérienne avec la culture du KT dans 23/24 cas de colonisation de CVC Se 92.3% Sp 52.7% VPP: 32% VPN 96.7% VPP moins bonne pour les G+ (24% vs 47%) Mahé I et al. Reanim Urg 1998;7:17 2

Prélèvements cutanés orientés 132 Kt, hématologie, culture (Maki +Sheretz) Cultures systématiques tous les mois vs Culture en cas de suspicion d'infection N 87 15 Se 18 75 Sp 83 100 VPP 13 100 VPN 88 92 Systématiques Orientés (*) écouvillon de 24 cm 2, culture quantitative en milieu liquide Raad Clin Infect Dis 1995; 20:593 16

Hémoculture quantitative comparative en réanimation 14/283 infectés, 19 ont au moins une HC sur CVC + Seuil KT/P=2 Se 98 % Sp 98% Seuil KT/P=8 Se 92.8 % Sp 98.8% Seuil KT/P=100 Se 79% Sp 99% Que faire des HC centrales positives isolées? Quilici - CID 1997; 25:1066

Délai de positivité des hémocultures (DTP) Turbidité du sang fonction de l’inoculum bactérien HC sur cathéter 0 4 8 heures HC périph. 0 4 8 heures DPT = 4 h.

Délai de positivité Validation in-vitro Blot F et al - J Clin Microbiol. 1998;105-109 Validation in-vivo (réanimation cancérologique) Seuil DTP= 120 mn Blot F - Lancet 2000; 354: 1071 MAIS Que faire de hémocultures dissociées? Explore essentiellement le mode de contamination endoluminale…utilité en réanimation? Rijnders BJ et al - Crit Care Med. 2001 Jul;29(7):1399-403 Cependant valeur diagnostique aussi bonne pour les CVCs de moins ou de plus de 30 jours Raad et al – Ann Intern Med 2004; 140:18-25

Blot F - Lancet ; 354: 1071-77 14 mois, 93 suspicions d ’ILC CVC courte et longue durée, dispositifs implantables Paires d ’hémocultures et ablation du KT dans les 48 heures Sp: 91 (95% CI 59 -100%) Se: 94 (95% CI 71 - 100%)

Paired blood cultures Total CRI Absence of CRI Positive (H+/P+) 28 17 11 DTP >120 min 17 16 1 DTP <120 min 11 1 10 Dissociated 19 3 16 (H+/P­) 17 3 14 (H­/P+) 2 0 2 Negative (H­/P­) 46 4 42 Total 93 24 69 Classification basée sur la clinique et la culture du KT et la régression à l ’ablation du KT 34 paires sont prélevé pour les 18 ayant une CR BSI et un couple d ’HC positive 23/34 seulement ont un couple d ’HC positif 9 ont H+/P-, et 2 H-/P- donc, les H+/P- sont difficiles à classer chez les patients avec plusieurs H+ P+ toutes donnent des delai de positivité sup à 120 mn Blot F - Lancet 1999; 354: 1071-77

Endoluminal colonization: in which lumen Endoluminal colonization: in which lumen? Dobbins et al – CCM 2003; 31: 1688 CVCs not suspected No CRBSI (n=50) CVCs suspected (n=25) CRBSI N lumens colonized* 1 2 3 6 4 10 5 N CVCs Maki roll + 28 14 20 (*) endoluminal brushes> 100 CFUs

positivity, and endoluminal brushing In situ diagnosis of intravascular catheter-related bloodstream infection: A comparison of quantitative culture, differential time to positivity, and endoluminal brushing Catton et al – Crit Care Med 2005; 33:787 Objective: To compare the accuracy of three techniques that do not require central venous catheter removal to diagnose catheter- related bloodstream infection. Design: Prospective cohort study of central venous catheters from suspected cases of catheter-related bloodstream infection. Setting: University teaching hospital. Patients: One hundred and twenty-five central venous catheters from patients with suspected catheter-related bloodstream infection (a raised peripheral white blood cell count, temperature >37°C, and/or local signs of infection at the catheter skin entry site) in intensive care and surgical patients in a large teaching hospital were assessed. Interventions: None. Measurements: Three techniques were compared: the differential time to positivity of central venous catheter vs. peripheralblood cultures, quantitative culture of central venous catheter vs. peripheral blood, and the endoluminal brush with peripheral blood culture. Main Results: Central venous catheters with a median dwell time of 11 days were examined. There were 36 episodes of catheter-related bloodstream infection, defined as a positive result from at least two of the three tests in the presence of a peripheral blood culture growing the same microorganism and without an identifiable alternative source of sepsis. The sensitivities of the endoluminal brush, quantitative culture, and differential time to positivity techniques were 100%, 89%, and 72%, respectively, with corresponding specificities of 89%, 97%, and 95%. Blood could be directly aspirated from only 231 of 312 (74%) lumens. In the 20 cases of catheter-related bloodstream infection associated with multiple-lumen central venous catheters, endoluminal brushing was positive for one, two, and three lumens in nine (45%), six (30%), and five (25%) cases, respectively. Conclusions: All three techniques had relatively high sensitivity. However, inability to obtain samples via central venous catheters is a major drawback of the differential time to positivity and quantitative blood culture approaches. Differential time to positivity is simple to perform and has high specificity and therefore could be used as a first line approach, with the endoluminal brush reserved for cases where blood cannot be obtained. All lumens of multiple-lumen central venous catheters must be sampled to ensure maximal sensitivity. (Crit Care Med 2005; 33:787–791) KEY WORDS: central venous catheter; catheter-related bloodstream infection; differential time to positivity; quantitative culture; endoluminal brush Population: One hundred and twenty-five CVCs were examined from 119 patients. No patients had catheters impregnated with antibiotics or antiseptic, and none received immunosuppressive dosages of systemic steroids. The following types of catheter were assessed: 25 Hickman catheters (21 single lumen and four double) and 18 single-, 63 triple-, and 19 quadruple- lumen MLCVCs. The median catheter dwell time was 11 days (range, 2–712).

Watchful waiting vs immediate CVC removal in the ICU - Watchful waiting vs immediate CVC removal in the ICU - Rijnders BJ et al – Intens Care Med 2004; 30: 1073-80 Exclusion: Neutropenia, foreign body, transplantation BSI (positive BC) Erythema, induration or purulence HD instability Previous DNR Abstract Objective: To find a subset of patients with suspected central venous catheter (CVC)-related infection (CRI) in whom CVC removal is not needed. Design: Randomized controlled trial. Setting: Thirty-threebed ICU. Patients and participants: One hundred and forty four patients with suspected CRI in which a change of CVCs was planned were evaluated for inclusion. Interventions: Hemodynamically stable patients without proven bacteremia, no insertion site infection, and no intravascular foreign body were randomized to a standard-of-care group (SOC, all CVCs were changed as planned) or a watchful waiting group (WW, CVCs changed when bacteremia was subsequently confirmed or hemodynamic instability occurred). Measurement and results: Study groups were compared for incidence of CVC-related bloodstream infection (CR-BSI), resolution of fever, C-reactive protein, SOFA score, duration of ICU stay, and mortality. Of 144 patients with suspected CRI, 80 patients met exclusion criteria. Sixty-four were randomized. Forty-seven of 80 excluded patients were shown to be bacteremic, 20 (25%) of whom had a CR-BSI. Five of 64 (8%) included patients had a CR-BSI during their subsequent ICU stay (two in SOC and three in WW group). All 38 CVCs were changed in the SOC group versus 16 of 42 in the WW group (62% reduction, P<0.01). Resolution of fever, C-reactive protein, SOFA score, duration of ICU stay, and ICU mortality did not differ between SOC and WW group (P>0.1 for all). Conclusions: The use of a simple clinical algorithm permits a substantial decrease in the number of unnecessarily removed CVCs without increased morbidity.

Watchful waiting vs immediate CVC removal in the ICU - Watchful waiting vs immediate CVC removal in the ICU - Rijnders BJ et al – Intens Care Med 2004; 30: 1073-80 (2) New Abx after inclusion: 13 of 32 patients in the WW 22 of 32 in the SOC-(P=0.04).

limitations Weak and subjective exclusion criterias Low power Rate of non bacteremic sepsis not reported Decrease in the rate of suspicion of CR-BSI during the study: First half 85/704 vs 2nd half 59 / 790 p=0.003 Rijnders BJ et al – Intens Care Med 2004; 30: 1073-80

The CVC ? 1. CVC removal 2. Diagnosis catheter in place 3. Guidewire exchange (GWX) Associated with fewer mechanical complications OR: 0.48 [[0.89-3.33] But a trend toward a higher risk of infection of the 2nd CVCs OR: 1.72 [0.89-3.33] Cook DJ Crit Care Med 1997;25:1417 12

Changement sur guide Palmer S – ICHE 2005; 26:506 158 changements sur guide / 13 cultures de guide positives (8.2%) Même germes sur les 2 CVCs et le guide dans 6 / 7 cas Colonisation du guide prédictif de la colonisation du CVC posé (p=0.05)

Guidewire exchange (GWX) 1. When to start antimicrobials?  Before the guidewire exchange 2. Attitude with the second CVC  Keep it if culture neg.  Remove it if culture pos.  It might be possible to keep the 2nd CVC in case of CNS or Enterobacteriaceae???? 18

Critères diagnostiques Infection bactériemique CVC + ou HC différentielles + ou culture du site d’insertion + et HC au même germe Absence d’autre site + expliquant les HC ILC non bactériémique C.V.C.+ Et Une régression totale ou partielle dans les 48 h ou Orifice purulent ou tunnelite. Réactualisation du consensus – Réanimation 2003;12: 258-265 21

Catheter tip colonization: a surrogate? Meta-analysis 1990- 2002 randomized study 29 studies selected Quantitative or semiquantitative cult and CR-BSI Correlation: R squared= 0.48, p< 0.001 BSI=0.77 + 0.73(CTC) Rijnders et al – Clin Infect Dis 2002; 9:1053

Should we always prescribe systemic antimicrobials ? Always if severe sepsis or septic shock Positive blood cultures - Yes, always For CNS (2 positive BC) In case of negative BC ???? 19

Which micro-organisms are associated with severe complications Which micro-organisms are associated with severe complications? ?(n = 102) Shock Sepsis Thrmb. Sept. Other Total (%)* CNS 3 1 1 1 6/33 (18) S. aureus 3 3 4 8 12/32 (38) Enterococci 0 0 0 0 0/3 GNB 2 0 0 0 2/10 (20) P.aeruginosa 1 0 1 0 2/4 (50) Candida spp. 0 7 0 0 7/11 (64) Polymicrob. 2 1 1 0 4/9 (44) * Nb Complications/Nb of events Arnow PM et al. 1993 Clin Infect Dis

Antimicrobials (BC neg) Situation Antimicrobials Candida spp, S. aureus or P. aeruginosa Sepsis after CVC removal Yes No fever after CVC removal No ? Other micro-organisms  Fever after CVC removal No* If GWX or CVC in place Yes __________________________________________________ * Except immunosuppression 21

Quelles molécules doit t on utiliser? C-CLIN Sud Est 2000 C-CLIN Paris-Nord Réa Cat 2000 Colonisation CNS 40 % 44 % S. aureus 10 % 6 % Entérocoques 3 % - BGN dont pyocyanique 12 % 37 % 15 % Candida 2 % Infection 28 % 19 % - 47 % 22 % 1 %

Grandes variations selon les centres

L’épidémiologie varie en fonction des années et des épidémies… from U.H.L.I.N Bichat: I Lolom, JC Lucet

Infection sytémique de KT Microorganismes voie fémorale Timsit et al – Ann intern Med 1999 Infection sytémique de KT Culture >103 cfu/ml Tunnelisés (15/14) 1 4 3 2 Controles (21/19) 1 4 2 7 Groupes (N /N events) S. aureus SCN Enterococcus P. aeruginosa A. baumannii E. coli Autres Gram neg. Champignons Tunnelisés (6/5) 2 1 Controles (17/15) 1 2 4 6 9 4 21 17 2 2

Choice of the molecules Situations active on CNS If severe, consider immediately GNB and yeast Molecules Glycopeptide + gentamicin If GNB suspected: activity against P. aeruginosa Candida: fluconazole (800 mg laoding dose) or Ampho B (unstable patients) Rex et al N Engl J Med 1994 ;331:1325 Antimicrobials should be adapted to blood and catheter cultures LNZ? Lipopeptides? Eichinocandins? AmpB - L 28

Biofilm production and antifungal effects In the biofilm (C. albicans and C. glabrata): AMPHO B > Voriconazole > fluconazole Regrowth was noted in the biofilm Lewis et al – Antimicrob Agent Chemother 2002; 3499 Killing of the biofilm cells better with eichinocandins (caspofungin) (activity against fungal cell wall +++) Kuhn DM - Antimicrob Agent Chemother 2002; 1773 Ramage R - Antimicrob Agent Chemother 2002; 3634 Bachmann SP- Antimicrob Agent Chemother 2002;3591 Caspo kills > 99¨of sessile cells within the biofilm at attainable concentrations Probably because the fungal cell wall is one of the component of the candidal biofilm and the target of this new agent (blocking of the 1,3 B D glucan). Figure: biofilms is less hyphal and some of the cells within the biofilms presented abberrant morphologies Red fluorescence characterize metabolic activities ….treated biofilm with 0.5 µg/ml of caspofungin showed a diffuse green fluorecence characteristics of dead cells

Favorable outcome : per-protocol % 80,7 79,5 64,9 64,9 Mora-Duarte J et al. NEJM 2002.

Background Invasive candidosis is increasingly prevalent in seriously ill patients. Our aim was to compare micafungin with liposomal amphotericin B for the treatment of adult patients with candidaemia or invasive candidosis. Methods We did a double-blind, randomised, multinational non-inferiority study to compare micafungin (100 mg/day) with liposomal amphotericin B (3 mg/kg per day) as fi rst-line treatment of candidaemia and invasive candidosis. The primary endpoint was treatment success, defi ned as both a clinical and a mycological response at the end of treatment. Primary analyses were done on a per-protocol basis. This trial is registered with ClinicalTrials.gov, number NCT00106288. Findings 264 individuals were randomly assigned to treatment with micafungin; 267 were randomly assigned to receive liposomal amphotericin B. 202 individuals in the micafungin group and 190 in the liposomal amphotericin B group were included in the per-protocol analyses. Treatment success was observed for 181 (89·6%) patients treated with micafungin and 170 (89·5%) patients treated with liposomal amphotericin B. The diff erence in proportions, after stratifi cation by neutropenic status at baseline, was 0·7% (95% CI –5·3 to 6·7). Effi cacy was independent of the Candida spp and primary site of infection, as well as neutropenic status, APACHE II score, and whether a catheter was removed or replaced during the study. There were fewer treatment-related adverse events—including those that were serious or led to treatment discontinuation—with micafungin than there were with liposomal amphotericin B. Interpretation Micafungin was as eff ective as—and caused fewer adverse events than—liposomal amphotericin B as fi rst-line treatment of candidaemia and invasive candidosis. Kuse et al - Lancet 2007; 369: 1519–27

What should be done in case of failure What should be done in case of failure ? (sepsis and/or + BC > 3 days)  Pharmacologic failure MRSA/glycopeptides  Thrombophlebitis  New CVC colonization  Other septic foci (endocarditis+++) 2

Vancomycin Pharmacocinetic variable and unpredictable: Dosage+++ Low level associated with failure Maintain trough > 15-20 µg/ml especially if MIC > 1 µg/ml Consider association: Gentamicin if possible, rifampin, linezolid?, dalfopristin-quinupristin? SUBSEQUENT DE-ESCALATION IF Methicillin sensitive+++ 25

Septic thrombophlebitis Clinically silent Ultrasound Doppler. Ligation of the vein: very invasive, rarely indicated Optimizing the antimicrobial : Antibiotic dosing, 2 antimicrobials Longer duration: 4-6 weeks Heparin and fibrinolytic ? 23

Trans-oesophagal echography and S.aureus 26 * 7 * P < 0,0005 Adapted from Fowler et al. JACC 1997 20

Relapse increases if treatment is less than 10 days Duration of treatment and complications: P=0.01 S. aureus: Relapse increases if treatment is less than 10 days Malanovski GJ - Arch Intern Med 1995;155:1161

S. aureus CRB : Short treatment Jernigan et al - Ann Intern Med 1993;119:304 S. aureus CRB : Short treatment Meta-analysis 11 studies/ 132 Pts Late complications after treatment < 14 days 6.1% [95% CI, 2.0% - 10.2%] Rare but severe: 3 Endocarditis (1 surgery) 2 epidural abscesses (1 surgery) 2 bacteremias (1 death)

Duration of treatment proposals (Positive BC) Microorganism Duration (d) S. aureus 14 (4-6 weeks*) P. aeruginosa 14 Candida spp. 14 (28*) CNS 7 (14/ 21**) Enterobacteriaceae 7 (14/ 21**) ______________________________________________________ * complications ** If CVC left in place or immunosupression 26

Duration of treatment proposals (Negative BC) Nothing!! Probably not justified if afebrile after CVC removal? S. aureus et P. aeruginosa or immunosupression  ?? (7d?) 27

Antibiotic lock in ICU? Antimicrobial concentration high (X 50 to 100) Volume 2 ml (+ héparine if vanco, cipro, teico) Anticrobials stable: (even with heparine) vanco, cefazolin, ticar-clavu,cipro (Anthony et al, AAC 1999;2074) New locks:Minocyclin-EDTA, Ethanol, Taurolidine CVC use is impossible during the lock… Injection 2 fold a day, for 2 to 3 weeks Associated IV antimicrobials Contra-indications: fungal infections, neutropenia, thrombophlebitis, tunnelitis, septic shock

Chambres implantables Verrou (VLA) ou AB IV   AB IV seuls VLA +/- AB IV   14 essais (1982  1995) Succès: 342/514 (66,5%) CVC tunnélisés 7 essais (1990  1995) Succès: 138/167 (82,6%) Chambres implantables 5 essais (1988  2001) Succès : 90/120 (75%) Problèmes de définition des infections Sites d’infection inconstamment cités Paramètres d’évaluation de l’efficacité différents

AAC 2007; 78-83 Antimicrobial lock solutions may be needed to salvage indwelling catheters in patients requiring continuous intravenous therapy. We determined the activity of minocycline, EDTA, and 25% ethanol, alone or in combination, against methicillin-resistant Staphylococcus aureus and Candida parapsilosis catheter-related bloodstream infection strains in two established models of biofilm colonization. Biofilm-colonized catheter segments from a modified Robbins device and a silicone disk biofilm colonization model were exposed to these antimicrobial agents for 15 or 60 min, respectively. After exposure, segments were sonicated and cultured. To determine regrowth after incubation at 37°C, following the brief exposure to the antimicrobial agents, an equal number of segments were washed, reincubated for 24 h, and then sonicated and cultured. The triple combination of minocycline-EDTA (M-EDTA) in 25% ethanol was the only antimicrobial lock solution that completely eradicated S. aureus and C. parapsilosis in biofilm of all segments tested in the two models, and it completely prevented regrowth. In addition, M-EDTA in 25% ethanol was significantly more effective in rapidly eradicating the growth or regrowth of methicillin-resistant S. aureus and C. parapsilosis biofilm colonization in the two models than the other solutions—minocycline, EDTA, M-EDTA, 25% ethanol, and EDTA in ethanol. We conclude that M-EDTA in 25% ethanol is highly effective at rapidly eradicating S. aureus and C. parapsilosis embedded in biofilm adhering to catheter segments. (*)I.R. is the inventor of catheter lock technology that involves alcohol. This patent is the property of The University of Texas M. D. Anderson Cancer Center.

70% Ethanol lock 45/51 success Arch Pediatr Adolesc Med. 2006;160:1049-1053 70% Ethanol lock 45/51 success Objectives: To use the ethanol-lock technique (in conjunction with systemic antibiotics) to salvage central lines from removal and to prevent persistence of catheterrelated infections among pediatric patients with longterm intravascular devices. Design: Medical records of patients treated with ethanol locks were retrospectively reviewed from June 1, 2004, through June 22, 2005. Setting: Childrens Hospital Los Angeles, Los Angeles, Calif, a tertiary care pediatric hospital. Patients: Forty children with diverse underlying disorders were treated for 51 catheter-related infections using the Childrens HospitalLosAngeles ethanol-lock technique. Interventions: Eligible infected central lines were instilled with a dose volume of 0.8 to 1.4 mL of 70% ethanol into the catheter lumen during 12 to 24 hours and then withdrawn. The volume of ethanol used was based on the type of intravascular device. Main Outcome Measures: Clearance of infection and incidence of recurrence. Results: Of the 51 ethanol-lock treatments in 40 children, no catheters were removed because of persistent infection. Eighty-eight percent (45/51) of the treated episodes cleared without recurrence (defined as a relapse within 30 days with the same pathogen). Twelve (75%) of 16 polymicrobial isolates and 33 (94%) of 35 monomicrobial isolates were successfully treated. There were no adverse reactions or adverse effects reported. Conclusion: This retrospective study supports the use of the ethanol-lock technique in conjunction with systemic antibiotics as an effective and safe method to retain the use of a previously infected central venous catheter, decrease the need for line removal, and eradicate persistent pathogens in catheter-related infections. Treatment success was defined as resolution of fever within 24 hours, no recurrence of positive blood cultures with the same organism, and retention of the IVD. Treatment failure was defined as recurrence within 30 days with the same pathogen or removal of the IVD because of a persistent infection.

Comité d'organisation : Responsables pour la commission des référentiels: B Guidet, R Robert, M Wolff, S Leteurtre Chargé de projet : adulte : JF Timsit, pédiatrie : Ph Durant Experts : adulte : G Nitenberg, pédiatrie : Dageville Membres de l'ancien jury : G Bleichner, Y Letulzo, M Pinsard. Experts extérieurs : JC Lucet, B Souweine, L Soufir, P Longuet, J Merrer , A Lepape, F Blot, C Martin, G Nitenberg, O Mimoz, Ph Eggiman, G Colas, C Brun-Buisson Reanimation 2003