We report the successful eradication of an environmental reservoir of NDM-KP in a low-incidence setting, over an eight-month period. During the investigation of this outbreak, WGS analysis allowed us to detect clonal relationships between clinical and environmental isolates, and to identify a probable transmission pathway contributing to the persistence of the epidemic strain in the unit.
Many reports have described outbreaks of MDRO linked to the hospital wastewater systems (10, 11, 15, 18–23). Highly discriminatory genomic methods, such as WGS, now allow for the confirmation of the key role that wastewater drains play as reservoir. (8, 9, 22, 24–26) Such transmissions have been mainly reported in ICU and hematology-oncology wards, predominantly affecting immunosuppressed patients exposed to several medical devices. (6) In Switzerland, carbapenem resistance remain rare, although numbers are increasing steadily, mirroring the situation in neighbouring countries. Klebsiella pneumoniae producing NDM, oxacillinase or Klebsiella pneumoniae carbapenemase and Escherichia coli producing oxacillinase or NDM are the most frequently observed CPE strains (2, 27, 28). Recently, Catho et al (29) published a report on a long-term outbreak of Pseudomonas aeruginosa producing Verona integron-encoded metallo-β-lactamase carbapenemase, which was genetically related to the hospital building’s wastewater in an intensive care unit in Geneva. This highlights the rising threat of carbapenemase-producing microorganisms in Swiss hospitals.
The biofilm present within the hospital wastewater system creates a particularly conducive environment for the development of a resistant microbiome. Indeed, the repeated exposure to various biological fluids from patients (30) or to antibiotic treatments poured into washbasins encourages the establishment and selection of antibiotic-resistant micro-organisms and therefore the potential horizontal transfer of resistance genes between species. Nevertheless, factors affecting MDRO establishment in the hospital wastewater environment are complex. In the case of carbapenemase-producing K. pneumoniae, Park et al (7) showed that positive patients can seed the wastewater environment in at least 6% of opportunities. Additionally, environmental sites that become colonised are more likely to remain positive, which is congruent with our experience.
Several routes of transmission between patients and aqueous reservoirs have been reported, including retro-contamination via splashes of the healthcare environment, (30, 31) healthcare equipment and medical devices (26). The routes can lead to patient contamination through direct or indirect contact. Sub-optimal design or misused of sinks may contribute to the dissemination of microorganisms colonizing the washbasin circuit. (32, 33) In our neurosurgical intermediate care unit, washbasins lack devices designed to minimise splashing, such as taps that do not flow directly into the drain and physical barriers that separate the adjacent area. Nevertheless, all waterpoints are located more than one meter far from patients’ beds, and patients’ clean items are stored far from the sink environment. However, on-site observations revealed several improper uses of washbasins during patient care and suboptimal hand hygiene adherence, which may have contributed to cross-transmission. The role of hand-carried transmission is difficult to evaluate and remains undefined, largely due to the likely contribution of multimodal factors. Nonetheless, several studies have reported a significant reduction in transmission following enhancements in hand hygiene compliance. (34)
Therefore, bundled approaches that aims to prevent the seeding of hospital drains with MDROs, by avoiding patient fluid exposition, eradicating colonized reservoirs, and interrupting cross-transmissions between patients and their water environment, have been proposed and led to a more or less satisfactory control of epidemic situations. (6) Modifying or renovating washbasin pipelines, including the revaluation of pipe material (e.g., copper) and the installation of self-cleaning sinks using vibration/heat/ultrasonic (27), can be challenging mostly due to cost and architectural constraints. Other feasible but difficult-to-implement measures include anti-splashing barriers (27, 35) and safe sink practices, such as hand hygiene-dedicated washbasins and disposal of human waste (including water in contact with patients or medical equipment) in a washer-disinfector. (33) Importantly, waterless patient care significantly contributes to the prevention of water environment-to-patient cross-transmission, but is not always accepted by healthcare providers (14, 29, 36)
Regarding the eradication of environmental reservoirs, various drain treatment protocols, primarily involving chemical agents (such as bleach and hydrogen peroxide) (37) or thermal disinfection, have been documented. (38) However, these methods have shown limited, if any, lasting effect on decolonizing MDROs from drain system. The presence of microbial biofilm in sink traps hinders eradication of microorganisms and may warrant the replacement of the whole washbasin circuit. Nevertheless, sink drain replacement alone is often insufficient, as biofilm-derived re-colonization can occur from distal (unchanged) components of the drain system. (11) In our case, partial replacement of the drainage circuit by changing the traps only was not successful, even when combined with weekly steam disinfection. Only the combination of a complete washbasin circuit replacement with bleach and heat disinfection allowed the sustained eradication of the environmental reservoir, after a 12-month follow-up. The persistent contamination of sink traps after initial replacement highlights the importance of replacing the connecting pieces, while working closely with the plumbing department. More data from prospective studies comparing different replacement and disinfection protocols are required.
Only a few studies reporting success stated the duration of follow-up after the intervention, which ranged, from 2 months to 3.5 years. (22, 35, 39, 40) In our situation, we continued microbiological screening of sink traps for 12 months after the identification of the last positive patient and 5 months after the completion of the sink’s disinfection protocol, providing proof for a successful and persistent control of the outbreak.
The unicentric and observational character of this report represents a key limitation. Moreover, we used a multimodal approach to control the NDM-KP outbreak and therefore, we were not able to assess each IPC measure separately. Nevertheless, we provide a structured and detailed description of an effective bundle of measures which can be reproduced in other settings.