Aortic Endograft Infection with Mycobacterium chimaera and Granulicatella adiacens, Switzerland, 2014

We describe an aortic endograft infection caused by Mycobacterium chimaera and Granulicatella adiacens, successfully treated with prolonged antimicrobial drug therapy after complete explantation of the infected endoprosthesis and extra-anatomical reconstruction. Whole-genome sequencing analysis did not indicate a close relationship to bacterial strains known to cause infections after cardiac surgery.

The diagnostic workup in May 2015 revealed an incidental 5-mm small pulmonary nodulus in the right upper lobe, which was observed to be metabolically active in PET-CT. After recovery from the abdominal intervention, the patient underwent wedge resection, and a localized squamous-cell carcinoma of the lung was confirmed. In April 2016, a relapse of his neoplasia occurred. Despite intensified chemotherapy, the patient died in August 2017 because of progressive pulmonary cancer; no autopsy was performed.
We cultured mycobacteriologic samples in BD MGIT tubes (BD, Franklin Lakes, NJ, USA) on Middlebrook 7H11 agar plates (BD) according to previously published methods (3). Air and water mycobacterial cultures were performed as suggested by the European Centre for Disease Prevention and Control (5).
The HCU-related outbreak of disseminated M. chimaera infections led us to investigate the hybrid operating  room where the patient had undergone his initial surgery (online Technical Appendix Figure). The referring hospital did not use HCUs or extracorporeal membrane oxygenation devices. In summer 2015, we obtained water and air samples from the operating room (Table); results were negative for M. chimaera. According to a signature single-nucleotide polymorphism-based classification, the patient isolate was similar to the group 1 strains of M. chimaera (2). We therefore included all group 1 strains with sufficient WGS data from published studies together with the patient isolate in a combined analysis of a total of 437 strains (Figure 2). The patient isolate did not cluster with subgroup 1.1, which represented all but 1 of the reported cases of disseminated M. chimaera infections associated with contaminated HCUs.
Instead, the patient strain clustered with strains from subgroup 1.11 and branch 1 of group 1 (2); however, the patient strain had no close relationship to any other strain included in the comparison.
The endoprosthetic graft (Excluder RMT261214/ PXC121200) of our patient was produced by Gore Medical (Newark, DE, USA). The Swiss Agency for Therapeutic Products submitted a medical device report for the implicated graft to the manufacturer.

Conclusions
We report an endovascular AGI caused by M. chimaera and G. adiacens, which was successfully treated with extra-anatomic bypass and prolonged antimicrobial therapy. Because   granulomatous necrotizing inflammation and detection of sparse acid-fast rods in Ziehl Neelsen stain, we outweighed the importance of M. chimaera compared with G. adiacens. Patients at risk for NTM infections are elderly patients with preexisting pulmonary conditions or immunocompromised patients. At AGI diagnosis, the localized pulmonic cancer in this patient was in an early stage, and the patient was not known to be immunocompromised. Blood cultures and repeated sputum specimens were negative for mycobacteria, and PET-CT did not reveal any distant foci. Therefore, we considered a hematogenous spread of a localized and naturally acquired infection to be unlikely. Water and air samples from the operating room were negative for M. chimaera; thus, local contamination in the operating room was unlikely. When we compared the patient's isolate with other available M. chimaera strains with available WGS data (2,6-10), we observed no association with the cardiac surgery cluster or any other closely related strain in the collection. Because the cardiac surgery cluster originated from M. chimaera-contaminated water in medical devices, a contamination of the medical prosthesis at the production site was considered, especially because the poorly soluble polytetrafluoroethylene polymerization is conducted as an emulsion in purified water. However, according to the graft manufacturer, its grafts are produced in a controlled environment, and ethylene oxide gas (EOG) is used for sterilization as recommended by the International Organization for Standardization (standard no. 11135-2007). EOG is widely used because of its good bactericidal activity on many bacterial species and even bacillus spores (11). However, studies showing the effect of EOG on mycobacteria are lacking, and cases of NTM infections caused by inadequate implant sterilization have been reported (12). As the logical next step in the investigation, testing environmental water samples from the production site or from fresh implants for NTM contamination was proposed. However, because of a paperwork assessment, the company decided not to pursue the case further.
Because our investigation involved a single case of an abdominal AGI caused by M. chimaera and G. adiacens, it is too early to draw any conclusions. If further infections emerge, investigations into the adequacy of EOG sterilization for arterial implants should be conducted. In this case, the combination of prolonged antimicrobial therapy, graft explantation, and extra-anatomic reconstruction resulted in sustained healing. etymologia Sources