Disponible en ligne sur

ScienceDirect www.sciencedirect.com Médecine et maladies infectieuses 44 (2014) 229–231

Short communication

Management and control of a carbapenem-resistant Acinetobacter baumannii outbreak in an intensive care unit Prise en charge d’une épidémie d’Acinetobacter baumannii résistant à l’imipénème en réanimation S. Alfandari a,∗ , J. Gois a , P.-Y. Delannoy a , H. Georges a , N. Boussekey a , A. Chiche a , A. Meybeck a , P. Patoz b , N. Blondiaux b , E. Senneville c , H. Melliez c , O. Leroy a a

Service de réanimation et des maladies infectieuses, centre hospitalier de Tourcoing, 155, rue du Président-Coty, 59208 Tourcoing cedex, France b Laboratoire de biologie, centre hospitalier de Tourcoing, 59208 Tourcoing, France c Service universitaire des maladies infectieuses et du voyageur, centre hospitalier de Tourcoing, 59208 Tourcoing, France Received 1st March 2014; received in revised form 23 March 2014; accepted 24 March 2014 Available online 17 May 2014

Abstract Objective. – We had for aim to describe the identification and management of a 14-clonal carbapenem-resistant Acinetobacter baumannii (CRAB) outbreak, following admission of a known CRAB-infected patient in an ICU. Methods. – We reviewed the carriers’ files and outbreak management procedures. Results. – The index patient was admitted with strict isolation precautions. The outbreak started 2 months after his discharge. It persisted despite reinforcement of strict isolation precautions, staff and patient cohorting, and extensive environmental decontamination including 2 rounds of routine terminal cleaning and disinfection or 1 round of cleaning and disinfection followed by hydrogen peroxide treatment. A second epidemic peak, after 4 weeks without any case, led to another wide environmental sampling and decontamination rounds. The source of the CRAB outbreak was suspected to be the blood pressure cuffs Velcro. Switching to cuffs submersible in a disinfectant stopped the outbreak. Conclusions. – CRAB outbreaks are difficult to manage and sources of persistent colonization can be unexpected. © 2014 Published by Elsevier Masson SAS. Keywords: Acinetobacter baumannii; Outbreak; Multidrug resistance

Résumé Objectif. – Décrire la reconnaissance et la prise en charge d’une épidémie clonale de 14 cas d’Acinetobacter baumannii résistant à l’imipénème (ABRI) après admission d’un patient porteur d’une pneumonie à ABRI en réanimation. Méthode. – Revue des dossiers des porteurs et des mesures de contrôle. Résultat. – Le cas index avait été placé en précautions complémentaires dès l’entrée. L’épidémie a débuté 2 mois après sa sortie. Elle a persisté malgré le cohorting du personnel, dépistage des contacts, bionettoyage par détergent désinfectant, puis aérosols de peroxyde d’hydrogène, et audit extérieur. Un second pic épidémique après 4 semaines sans cas a fait renouveler les prélèvements environnementaux et un bionettoyage du service. De l’ABRI a été retrouvé uniquement sur le Velcro de 2 tensiomètres. Le service a changé de modèle pour utiliser des modèles entièrement immergeables dans une solution désinfectante, sans nouveau cas identifié depuis. Conclusion. – Les épidémies à ABRI sont difficiles à éradiquer. © 2014 Publié par Elsevier Masson SAS. Mots clés : Acinetobacter baumannii ; Épidémie ; Multirésistance

∗

Corresponding author. E-mail address: [email protected] (S. Alfandari).

http://dx.doi.org/10.1016/j.medmal.2014.03.005 0399-077X/© 2014 Published by Elsevier Masson SAS.

230

S. Alfandari et al. / Médecine et maladies infectieuses 44 (2014) 229–231

1. Introduction Acinetobacter baumannii is a Gram-negative non-fermenting rod colonizing hospital environment. It is responsible for opportunistic infections, mostly among intensive care patients [1]. Multidrug-resistant strains, particularly carbapenem-resistant A. baumannii (CRAB), can be responsible for severe outbreaks with few therapeutic options. A prompt identification and response is essential to prevent the extension of an outbreak. 2. Methods The Tourcoing General Hospital has 400 acute-care beds, with university affiliated 16 intensive care beds, and 41 infectious diseases unit beds. The intensive care unit (ICU) participates in a voluntary ongoing survey of nosocomial infections in French ICUs. Patients are screened for multidrug-resistant bacteria with nasal and anal swabs at admission, then once a week. Furthermore, tracheal aspirates from ventilated patient are cultured once a week. A 31-year-old French patient, presenting with CRAB pneumonia, was transferred to the Tourcoing hospital from Morocco, on September 30th, 2011. He was placed in a private room and managed with strict isolation precautions until his discharge in October 29th, 2011. A second patient, admitted on October 25th, 2011, was found to carry CRAB in a tracheal aspirate, on December 13th, 2011. A third patient presented with CRAB bacteremia on January 12th, 2012. Two new carriers were identified with CRAB in tracheal aspirates, in early February 2012. The infection control team launched an outbreak investigation at this point. Cases were defined as patients colonized or infected by CRAB. Contact patients were those managed by health care workers having cared for CRAB patients before identification of the outbreak and implementation of control measures. 3. Outbreak management On February 8, 2012, the ad hoc outbreak management team (infection control unit, ICU physicians and nurse supervisors, microbiologists, and administrative staff) held a crisis meeting and decided to implement the following actions: 24 h cohorting of the non-medical staff managing CRAB patients, reinforcement of hand hygiene and isolation precautions, twice weekly

screening of all contact patients, self-auditing of contact precautions, and limitation of carbapenem prescription. The outbreak was notified to the health authority, as required by the French law. Multiple environmental sampling and an extensive decontamination of the whole ICU, including technical and storage rooms, were performed as new cases were discovered. One round of cleaning and disinfection was performed in technical areas as well a daily cleaning of patients’ rooms. Initially 2 rounds were performed at CRAB patients’ discharge later replaced by 1 round of cleaning and disinfection followed by 1 round of hydrogen peroxide treatment when 2 CRAB positive samples were discovered after double disinfection of a patient’s room. We requested an external audit of the ICU from the regional infection control center, but it did not reveal any major error in infection control policies and compliance. The audit results led to recommending limiting storage of supplies in the rooms and preferring ready-to-use materials. No other case was observed in the ICU between March 13th and April 3rd, 2012. Two new cases were identified on April 3rd and 4th, 2012. The ICU stopped new admissions and repeated environmental sampling. A second cohorting sector for CRAB patients was created in an isolated 4-bed sector of the infectious diseases unit, for patients who were discharged from the ICU. Overall, 20 patient cases were identified between the index patient admission in September 2011, and the end of the outbreak (last case on June 22nd, 2012), 15 of which were infected in the ICU (Fig. 1). Infections occurred in 8 (40%) patients including the index case (4 cases of pneumonias, 3 of bacteremia, 1 of pyelonephritis). 6 patients died, 1 death being attributable to a CRAB bacteremia. 4. Epidemiological investigation All contact patients were screened with throat and anal swabs once a week, at discharge and at readmission, until 3 negative results were obtained. ICU patients were screened twice a week. Multiple (n = 222) environmental samples were collected in February 2012, in the patients’ rooms and the ICU common spaces (beds, supply carts, infusion pumps, ventilator pads, ultrasonographs, EKG, etc.), physicians’ and nurses’ phones, stethoscopes, before and after thorough room cleaning. Blood pressure cuffs Velcro were not sampled at that time. Five (2.3%) were positive for CRAB including 2 after room cleaning.

Fig. 1. New cases of carbapenem-resistant Acinetobacter baumannii per week, from December 2011 to April 2012 (not showing the index case in September 2011). Color indicates strain clone (black: clone 1; white: clone 2; dotted grey: undetermined). Nouveaux cas d’Acinetobacter baumannii résistants à l’imipénème, par semaine, décembre 2011 à avril 2012 (le cas index en septembre 2011 n’est pas indiqué). La couleur indique le clone en cause (noir : clone 1 ; blanc : clone : 2 ; pointillé gris : inconnu).

S. Alfandari et al. / Médecine et maladies infectieuses 44 (2014) 229–231

A second round of sampling was performed in April 2012 with 84 environmental and 13 health care worker hand samples. Two, from the blood pressure cuffs Velcro, yielded CRAB after disinfection and aerosolized hydrogen peroxide exposure. We changed all models for wholly submersible ones on the following days (Spengler, Aix-en-Provence, France). Antibiotic susceptibility testing was performed by disk diffusion and MICs were measured by the E-test. CRAB isolates were analyzed and genotyped by the French reference center for A. baumannii (Besancon University Hospital). All of the 19 evaluated infecting strains produced an OXA-23 carbapenemase. Pulsed-field gel electrophoresis revealed the presence of 2 different clones. Clone 1 was identified in 14 patients, 12 of whom were in the ICU and on the Velcro. This strain had been isolated from the index case and was only susceptible to colistin and tobramycin. Clone 2 was identified in 5 patients, including 2 in the ICU. It was susceptible to colistin, tobramycin, and amikacin and similar to a regional clone observed in various hospitals of the region.

231

transferred from health care institutions in other countries, to limit the spread of carbapenemase producing Enterobacteriaceae and glycopeptide-resistant enterococci. This strategy could be applied to CRAB [9]. We stress the ability of A. baumannii to persist on unexpected surfaces despite a stringent disinfection, as previously reported [10]. The characteristics of some materials might limit the effectiveness of simple cleaning and disinfection and, immersion in a disinfectant solution could be necessary. This might have been the case for the “hook and loop” Velcro structure. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Authors’ contribution: conception, design and drafting: S.A.; acquisition and analysis of data: S.A., J.G., P.-Y.D., H.G., N.B., A.C., A.M., E.S., H.M., O.L.; acquisition and analysis of microbiological data: P.P., N.B.; manuscript revision and final approval: all authors.

5. Discussion References We identified 2 clones during this outbreak. Clone 1 was responsible for most cases, most often in the ICU. The index patient’s strain was acquired in another ICU, abroad. The index patient was admitted with a strict isolation procedure; despite this, the strain was transmitted. We might have missed some cases, but given our comprehensive screening of drug-resistant organisms, this seems unlikely. Initial direct transmission is possible as all colonized/infected patients had a few days of hospitalization overlap. We also identified a possible source of contamination that could explain the outbreak dynamics. However, as there was no traceability of sphygmomanometers among patients, we can only suspect that colonized cuffs transferred from patient to patient. Nonetheless, the outbreak was controlled only after changing the sphygmomanometer model and no other clone 1 CRAB was isolated in the ICU after May 2012. In other reported outbreaks, A. baumannii was grown from sink and water tap [2], patient environment [3–6], and health care worker hand samples [6]. No source was identified for other outbreaks [7]. The second clone was isolated from patients with no hospitalization overlap, in different wards, which might reflect a low-level generalized spread of the strain in the region. The simultaneous discovery of 2 clones during an A. baumannii outbreak was previously reported [8]. The enhanced screening procedures during the outbreak might have helped identify this clone that would otherwise have gone undetected. Control of the outbreak required stringent measures, including a 3-month staff cohorting and a 1-month limitation of new admissions to the ICU, with a high financial cost for the hospital. The outbreak might have been shorter if it had been identified earlier but the administrative department might have been reluctant to engage in costly measures for, at the time, only a few cases. No current French guideline addresses CRAB management. French health authorities recommend a “search and isolate” strategy for patients

[1] Munoz-Price LS, Weinstein RA. Acinetobacter infection. N Engl J Med 2008;358:1271–81, http://dx.doi.org/10.1056/NEJMra070741. [2] Hong KB, Oh HS, Song JS, et al. Investigation and control of an outbreak of imipenem-resistant Acinetobacter baumannii infection in a pediatric intensive care unit. Pediatr Infect Dis J 2012;31:685–90, http://dx.doi.org/10.1097/INF.0b013e318256f3e6. [3] Thom KA, Johnson JK, Lee MS, Harris AD. Environmental contamination because of multidrug-resistant Acinetobacter baumannii surrounding colonized or infected patients. Am J Infect Control 2011;39:711–5, http://dx.doi.org/10.1016/j.ajic.2010.09.005. [4] Jamal W, Salama M, Dehrab N, Al Hashem G, Shahin M, Rotimi VO. Role of tigecycline in the control of a carbapenem-resistant Acinetobacter baumannii outbreak in an intensive care unit. J Hosp Infect 2009;72:234–42, http://dx.doi.org/10.1016/j.jhin.2009.03.023. [5] Choi WS, Kim SH, Jeon EGS, et al. Nosocomial outbreak of carbapenemresistant Acinetobacter baumannii in intensive care units and successful outbreak control program. J Korean Med Sci 2010;25:999–1004, http://dx.doi.org/10.3346/jkms.2010.25.7.999. [6] Enoch DA, Summers C, Brown NM, et al. Investigation and management of an outbreak of multidrug-carbapenem-resistant Acinetobacter baumannii in Cambridge, UK. J Hosp Infect 2008;70:109–18, http://dx.doi.org/10.1016/j.jhin.2008.05.015. [7] Garlantézec R, Bourigault C, Boles JM, et al. Investigation and management of an imipenem-resistant oxa-23 Acinetobacter baumannii outbreak in an intensive care unit. Med Mal Infect 2011;41:430–6, http://dx.doi.org/10.1016/j.medmal.2011.01.013. [8] Bourigault C, Corvec S, Bretonnière C, et al. Investigation and management of multidrug-resistant Acinetobacter baumannii spread in a French medical intensive care unit: one outbreak may hide another. Am J Infect Control 2013;41:652–3, http://dx.doi.org/10.1016/j.ajic.2012.08.003. [9] Lepelletier D, Andremont A, Grandbastien B, National Working Group. Risk of highly resistant bacteria importation from repatriates and travellers hospitalized in foreign countries: about the French recommendations to limit their spread. J Travel Med 2011;18:344–51, http://dx.doi.org/10.1111/j.1708-8305.2011.00547.x. [10] Manian FA, Griesenauer S, Senkel D, Setzer JM, Doll SA, Perry AM, et al. Isolation of Acinetobacter baumannii complex and methicillin-resistant Staphylococcus aureus from hospital rooms following terminal cleaning and disinfection: can we do better? Infect Control Hosp Epidemiol 2011;32:667–72, http://dx.doi.org/10.1086/660357.