Listeria monocytogenes-Associated Biliary Tract Infections

At present, little is known regarding Listeria monocytogenes-associated biliary tract infection, a rare form of listeriosis. In this article, we will study 12 culture-proven cases reported to the French National Reference Center for Listeria from 1996 to 2013 and review the 8 previously published cases. Twenty cases were studied: 17 cholecystitis, 2 cholangitis, and 1 biliary cyst infection. Half were men with a median age of 69 years (32–85). Comorbidities were present in 80%, including cirrhosis, rheumatoid arthritis, and diabetes. Five patients received immunosuppressive therapy, including corticosteroids and anti-tumor necrosis factor biotherapies. Half were afebrile. Blood cultures were positive in 60% (3/5). Gallbladder histological lesions were analyzed in 3 patients and evidenced acute, chronic, or necrotic exacerbation of chronic infection. Genoserogroup of the 12 available strains were IVb (n1⁄4 6), IIb (n1⁄4 5), and IIa (n1⁄4 1). Their survival in the bile was not enhanced when compared with isolates from other listeriosis cases. Adverse outcome was reported in 33% (5/15): 3 deaths, 1 recurrence; 75% of the patients with adverse outcome received inadequate antimicrobial therapy (P1⁄4 0.033). Biliary tract listeriosis is a severe infection associated with high mortality in patients not treated with appropriate therapy. This study provides medical relevance to in vitro and animal studies that had shown Listeria monocytogenes ability to survive in bile and induce overt biliary infections. (Medicine 93(18):e105) Abbreviations: BHI = brain heart infusion, CNS = central nervous system, Lm = Listeria monocytogenes, MIC = minimal inhibitory concentration, MLST = multilocus sequence typing, MN = maternal–neonatal, NRCL = National Reference Center for Listeria, S = septicemia. INTRODUCTION A severe foodborne infection that mostly occurs in immunecompromised patients is Listeria monocytogenes (Lm) that is a facultative intracellular Gram-positive bacterium responsible for listeriosis. Three main forms are described: septicemia (S), central nervous system (CNS), and maternal– neonatal (MN) infections. Aside from these typical presentations, localized infections are also reported, mostly as a consequence of a subclinical bacterial systemic dissemination. They include endocarditis, osteoarticular, and cutaneous infections as well as biliary tract infections, which have only been reported as isolated case reports, although Lm is well known to colonize the gut and survive in the bile. We undertook a comprehensive retrospective survey over a 17-year period to review all the cases referred to the national surveillance system of listeriosis in France since it has been established. Twelve cases were identified and analyzed. In addition, the 8 previously published case reports were reviewed. This study reveals that among biliary tract infections, those associated with Lm tend to exhibit specific features, with a higher frequency of comorbidities, of concomitant bacteremia and of adverse outcome, which are reported in 80%, 60%, and 33% of cases, respectively. Lmassociated biliary tract infections should be considered in the occurrence of biliary tract infection in immunocompromised patients. Their diagnosis requires a clinical and microbiological workup, and treatment is based on a specific amoxicillin-based antibiotic regimen to which Lm is sensitive, and which is, otherwise, not recommended as a first-line therapy for biliary tract infections. PATIENTS AND METHODS Data Collection Surveillance of human listeriosis in France is based on both mandatory reporting of cases to the Institut de Veille Sanitaire, France, since 1999 and voluntary submission of Lm Editor: Rana El Feghaly. Received: June 5, 2014; revised: August 7, 2014; accepted: August 10, 2014. From the Institut Pasteur, Biology of Infection Unit (CC, CF, LT, ML); Institut Pasteur French National Reference Center and WHO Collaborating Center for Listeria (CC, BC, HBD, AL, ML); Inserm U1117 (CC, CF, LT, ML); Universit e Paris Descartes, Sorbonne Paris Cit e, Centre d’Infectiologie Necker-Pasteur, Hôpital Necker-Enfants Malades, Institut Imagine (CC, ML); Service de Chirurgie visc erale, Centre Hospitalier Universitaire de Nantes (JP); Service de Chirurgie visc erale et digestive, Centre Hospitalier de Roubaix (DA, LG); Service d’Anatomopathologie et Cytologie, Centre Hospitalier Universitaire de Nantes (CB); Service d’Anatomopathologie et Cytologie, Centre Hospitalier de Roubaix (FC); and Service d’Anatomopathologie et Cytologie, Centre Hospitalier de Pau (VC). Correspondence: Caroline Charlier, Unit e de Biologie des Infections, Institut Pasteur, 28 rue du Dr Roux, Paris 75015, France (e-mail: [email protected]). CF and LT equally contributed in writing of this article. This study received financial support form Institut Pasteur, Inserm, Fondation BNP Paribas, Ville de Paris, Fondation pour la Recherche M edicale and Institut de Veille Sanitaire. The authors have no conflicts of interest to disclose. Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. This is an open access article distributed under the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ISSN: 0025-7974 DOI: 10.1097/MD.0000000000000105 Medicine • Volume 93, Number 18, October 2014 www.md-journal.com | 1 strains to the National Reference Center for Listeria (NRCL). The exhaustiveness of this reporting is estimated above 87%. We studied all listeriosis cases declared between January 1999 and March 2013 with mention of “cholecystitis,” “cholangitis,” “liver,” or “bile duct.” In addition, all patients with similar clinical data and for whom isolates were sent to the NRCL between 1996 and 1999, before the mandatory reporting era, were also included. Clinicians and microbiologists were contacted, and medical charts were directly analyzed according to a preestablished checklist. An appropriate local ethical committee (Comit e de Protection des Personnes Ile de France 8) considered the study as observational and hence exempted to the Institutional Review Board approval, according to the French legislation. Review of the Literature We searched the PubMed database for reports published between January 1966 and June 2013, using the terms “Listeria,” “listeriosis,” “cholecystitis,” “cholangitis,” “liver,” and “bile” without language restriction. Case Definition A case was defined as a person from whom Lm was isolated from the biliary tract. Infections were classified as cholecystitis, cholangitis, or biliary tract cyst infection. Liver abscesses without bile tract infection were excluded. Diagnosis of concurrent septicemia was based on a positive blood culture. L monocytogenes Typing Listeria isolates referred to the NRCL were identified with API Listeria (BioM erieux, Marcy l’Etoile, France), serotyped until January 2005, and then typed by multiplex polymerase chain reaction (PCR) genoserogrouping. PCR serogroups correspond to the 4 major serovars that cause human disease. Isolates were characterized by multilocus sequence typing (MLST) similar to the 745 other strains received in the NRCL, as previously described. Bile Resistance Assays Forty-two isolates were tested: 10 from biliary tract infection referred to the NRCL, the EGD bile susceptible, the LO28 bile-resistant reference strains, and 30 clinical isolates randomly selected among those received by the NRCL in 2012 (10 for each form: S, CNS, and MN). All were grown to log phase in brain heart infusion (BHI) broth (Becton Dickinson, Le Pont de Claix, France) at 37°C overnight, and then were inoculated by a A400 Multipoint Inoculator (Denley, West Sursex, UK) yielding 10 cells per spot on pork bile agar plates containing BHI base medium and 0%–10%, 15%, 20%, 25%, and 30% pork bile (Sigma Chemical Co, St Louis, MO) (pH1⁄4 7). The plates were incubated at 37°C in anaerobic condition for 24 hours and the minimal inhibitory concentration (MIC) of bile, defined as the lowest concentration totally inhibiting the growth of spots, was determined for each isolate. Mann–Whitney test was used to compare MICs. Indirect Immunofluorescence Assay of InlA Surface Expression Bacteria were grown overnight in liquid BHI at 37°C, rinsed, incubated with I4.4/L7.7 monoclonal antibodies against InlA (1:1000) for 1 hour, and then with a secondary goat anti-mouse Alexa Fluor 488 antibody (Life technologies, Carlfbad, CA) (1:500). Preparations were analyzed by epifluorescence microscopy (AxioObserver Z1 inverted microscope, Zeiss, Iena, Germany) and analyzed with the AxioVision software (Zeiss). Lm EGD that expresses surface InlA was used as positive control and Lm L028 that expresses a secreted truncated InlA was used as negative control. In case of absence of InlA surface expression, InlA truncation was confirmed by inlA sequencing. Biofilm Assays Forty-four isolates were tested: 12 bile tract infection isolates referred to the NRCL, the EGD, the LO28 Lm reference strains, and the 30 clinical isolates described above. Cultures were performed in BHI broth at 37°C upon shaking. Aliquot of BHI overnight liquid cultures (1:20) was added to fresh BHI medium. Exponential cultures were diluted in BHI medium, BHI with pork bile, at pH 5 and 7, to an optical density (OD600 nm) of 0.06 in 100 μL 96-well poly (vinyl chloride) microtiter plates (Falcon; Becton Dickinson Labware, Oxnard, CA). Biofilms were allowed to grow for 24 hours at 37°C. Unbound cells were removed by microplate inversion and tapping on absorbent paper. Microplates were washed in water and adherent cells were stained with crystal violet for 20 minutes. Excess stain was removed by 3 washes in water. Quantification of bound cells was performed by adding acetone–ethanol (20:80) and dissolved crystal violet was measured at OD595 nm. Each biomass was standardized relative to EGD reference strain, and Mann–Whitney test was used to compare each group. Histopathological Analyses Eight-micrometer-thick sections of paraffin-embedded tissue specimens were stained with hematoxylin eosin. Lm was labeled by immunohistochemistry using a polyclonal rabbit antiserum that detects Lm serotype 4b (Listeria O V/VI antiserum Seiken kit; Denka Seiken Co, Tokyo, Japan) and a goat anti-rabbit antibody coupled to peroxidase (EnVision+, Dako, Glosturp, Denmark), followed by hematoxylin counterstaining. Images were captured on a AxioImager A2 microscope (Zeiss) equipped with an AxioCam ICc 1 digital camera (Zeiss) and the AxioVision 4.8 software (Zeiss). RESULTS Clinical Cohort A retrospective analysis of all cases declared to the NRCL was performed as described in the “Patients and Methods” section. Among the 3231 human cases for which a clinical Lm strain was collected between January 1996 and March 2013, 12 involved patients with biliary tract infections (hereafter named the French cohort), representing 0.37% of the infections reported during the study period. They included 9 cholecystitis (75%), 2 cholangitis (17%), and 1 biliary cyst infection (8%) that is listed in Table 1. Eight additional cases were identified in the literature; all were cholecystitis and are also listed in Table 1. The patients from the French cohort and those previously reported were analyzed together to identify the main characteristics of Lmassociated biliary tract infections. 2 | www.md-journal.com ã 2014 Lippincott Williams & Wilkins Charlier et al Medicine • Volume 93, Number 18, October 2014