Pseudo-outbreak ofMycobacterium gordonaeFollowing the Opening of a Newly Constructed Hospital at a Chicago Medical Center

Abstract

OBJECTIVETo identify the source of a pseudo-outbreak ofMycobacterium gordonaeDESIGNOutbreak investigation.SETTINGUniversity Hospital in Chicago, Ilinois.PATIENTSHospital patients withM. gordonae-positive clinical cultures.METHODSAn increase in isolation ofM. gordonaefrom clinical cultures was noted immediately following the opening of a newly constructed hospital in January 2012. We reviewed medical records of patients withM. gordonae-positive cultures collected between January and December 2012 and cultured potable water specimens in new and old hospitals quantitatively for mycobacteria.RESULTSOf 30 patients withM. gordonae-positive clinical cultures, 25 (83.3%) were housed in the new hospital; of 35 positive specimens (sputum, bronchoalveolar lavage, gastric aspirate), 32 (91.4%) had potential for water contamination.M. gordonaewas more common in water collected from the new vs. the old hospital [147 of 157 (93.6%) vs. 91 of 113 (80.5%),P=.001]. Median concentration ofM. gordonaewas higher in the samples from the new vs. the old hospital (208 vs. 48 colony-forming units (CFU)/mL;P<.001). Prevalence and concentration ofM. gordonaewere lower in water samples from ice and water dispensers [13 of 28 (46.4%) and 0 CFU/mL] compared with water samples from patient rooms and common areas [225 of 242 (93%) and 146 CFU/mL,P<.001].CONCLUSIONSM. gordonaewas common in potable water. The pseudo-outbreak ofM. gordonaewas likely due to increased concentrations ofM. gordonaein the potable water supply of the new hospital. A silver ion-impregnated 0.5-μm filter may have been responsible for lower concentrations ofM. gordonaeidentified in ice/water dispenser samples. Hospitals should anticipate that construction activities may amplify the presence of waterborne nontuberculous mycobacterial contaminants.Infect Control Hosp Epidemiol 2014;00(0): 1–6