Proactive infection control measures to prevent nosocomial transmission of carbapenem-resistant Enterobacteriaceae in a non-endemic area / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>Identification of hospitalized carbapenem-resistant Enterobacteriaceae (CRE)-positive patient is important in preventing nosocomial transmission. The objective of this study was to illustrate the implementation of proactive infection control measures in preventing nosocomial transmission of CRE in a healthcare region of over 3200 beds in Hong Kong between October 1, 2010 and December 31, 2011.</p><p><b>METHODS</b>The program included active surveillance culture in patients with history of medical tourism with hospitalization and surgical operation outside Hong Kong within 12 months before admission, and "added test" as an opportunistic CRE screening in all fecal specimens submitted to the laboratory. Outbreak investigation and contact tracing were conducted for CRE-positive patients. Serial quantitative culture was performed on CRE-positive patients and the duration of fecal carriage of CRE was analyzed.</p><p><b>RESULTS</b>During the study period, a total of 6533 patients were screened for CRE, of which 76 patients were positive (10 from active surveillance culture, 65 from "added test", and 1 secondary case from contact tracing of 223 patients with no nosocomial outbreak), resulting in an overall rate of CRE fecal carriage of 1.2%. The median time of fecal carriage of CRE was 43 days (range, 13-119 days). Beta-lactam-beta-lactamase-inhibitors, cephalosporins, and fluoroquinolones were associated significantly with high fecal bacterial load when used 90 days before CRE detection, while use of cephalosporins, carbapenems, and fluoroquinolones after CRE detection are significantly associated with longer duration of carriage. The duration of fecal carriage of CRE also correlates significantly with the initial fecal bacterial load (Pearson correlation: 0.53; P = 0.02).</p><p><b>CONCLUSION</b>Proactive infection control measures by enhanced surveillance program identify CRE-positive patients and data obtained are useful for the planning of and resource allocation for CRE control.</p>

An unprecedented outbreak investigation for nosocomial and community-acquired legionellosis in Hong Kong / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>The environmental sources associated with community-acquired or nosocomial legionellosis were not always detectable in the mainland of China and Hong Kong, China. The objective of this study was to illustrate the control measures implemented for nosocomial and community outbreaks of legionellosis, and to understand the environmental distribution of legionella in the water system in Hong Kong, China.</p><p><b>METHODS</b>We investigated the environmental sources of two cases of legionellosis acquired in the hospital and the community by extensive outbreak investigation and sampling of the potable water system using culture and genetic testing at the respective premises.</p><p><b>RESULTS</b>The diagnosis of nosocomial legionellosis was suspected in a patient presenting with nosocomial pneumonia not responsive to multiple beta-lactam antibiotics with subsequent confirmation by Legionella pneumophila serogroup 1 antigenuria. High counts of Legionella pneumophila were detected in the potable water supply of the 70-year-old hospital building. Another patient on continuous ambulatory peritoneal dialysis presenting with acute community-acquired pneumonia and severe diarrhoea was positive for Legionella pneumophila serogroup 1 by polymerase chain reaction (PCR) testing on both sputum and nasopharyngeal aspirate despite negative antigenuria. Paradoxically the source of the second case was traced to the water system of a newly commissioned office building complex. No further cases were detected after shock hyperchlorination with or without superheating of the water systems. Subsequent legionella counts were drastically reduced. Point-of-care infection control by off-boiled or sterile water for mouth care and installation of water filter for showers in the hospital wards for immunocompromised patients was instituted. Territory wide investigation of the community potable water supply showed that 22.1% of the household water supply was positive at a mean legionella count of 108.56 CFU/ml (range 0.10 to 639.30 CFU/ml).</p><p><b>CONCLUSIONS</b>Potable water systems are open systems which are inevitably colonized by bacterial biofilms containing Legionella species. High bacterial counts related to human cases may occur with stagnation of flow in both old or newly commissioned buildings. Vigilance against legionellosis is important in healthcare settings with dense population of highly susceptible hosts.</p>

Extensive contact tracing and screening to control the spread of vancomycin-resistant Enterococcus faecium ST414 in Hong Kong / 中华医学杂志(英文版)

<p><b>BACKGROUND</b>Proactive infection control management is crucial in preventing the introduction of multiple drug resistant organisms in the healthcare setting. In Hong Kong, where vancomycin-resistant enterococci (VRE) endemicity is not yet established, contact tracing and screening, together with other infection control measures are essential in limiting intra- and inter-hospital transmission. The objective of this study was to illustrate the control measures used to eradicate a VRE outbreak in a hospital network in Hong Kong.</p><p><b>METHODS</b>We described an outbreak of VRE in a healthcare region in Hong Kong, involving a University affiliated hospital and a convalescent hospital of 1600 and 550 beds respectively. Computer-assisted analysis was utilized to facilitate contact tracing, followed by VRE screening using chromogenic agar. Multi-locus sequence typing (MLST) was performed to assess the clonality of the VRE strains isolated. A case-control study was conducted to identify the risk factors for nosocomial acquisition of VRE.</p><p><b>RESULTS</b>Between November 26 and December 17, 2011, 11 patients (1 exogenous case and 10 secondary cases) in two hospitals with VRE colonization were detected during our outbreak investigation and screening for 361 contact patients, resulting in a clinical attack rate of 2.8% (10/361). There were 8 males and 3 females with a median age of 78 years (range, 40 - 87 years). MLST confirmed sequence type ST414 in all isolates. Case-control analysis demonstrated that VRE positive cases had a significantly longer cumulative length of stay (P < 0.001), a higher proportion with chronic cerebral and cardiopulmonary conditions (P = 0.001), underlying malignancies (P < 0.001), and presence of urinary catheter (P < 0.001), wound or ulcer (P < 0.001), and a greater proportion of these patients were receiving β-lactam/β-lactamase inhibitors (P = 0.009), carbapenem group (P < 0.001), fluoroquinolones (P = 0.003), or vancomycin (P = 0.001) when compared with the controls.</p><p><b>CONCLUSION</b>Extensive contact tracing and screening with a "search-and-confine" strategy was a successful tool for outbreak control in our healthcare region.</p>