ENHANCED RECOVERY OF LISTERIA FROM DAIRY-PLANT PROCESSING ENVIRONMENTS THROUGH COMBINED USE OF REPAIR ENRICHMENT AND SELECTIVE ENRICHMENT DETECTION PROCEDURES

The efficacy of using a repair step to increase sensitivity of recover y of injured Listeria from environmental sources in dairy processing p lants was investigated. The USDA-FSIS Listeria isolation protocol usin g UVM-modified Listeria enrichment broth medium University of Vermont (UVM) for primary enrichment was the standard method chosen for compar ison. UVM broth was used in conjunction with rapid methods (Organon Te knika and Gene-TrakTM), following manufacturer's guidelines. Listeria Repair Broth (LRB) was used as the repair enrichment medium in modifie d protocols of the standard and rapid procedures. LRB employs a nonsel ective period (2-5 hours) for repair of injured Listeria prior to sele ctive-agent addition. Of 80 environmental sites positive by any method , UVM and LRB showed similar recovery rates (87.5% and 88.8%, respecti vely). Thus LRB provided little advantage over current procedures for use in contaminated sites. UVM was superior when used in conjunction w ith either rapid method. The USDA and modified USDA (mUSDA) procedures gave identical recovery rates (93%), but 10 additional positive sites were attributed to the use of two enrichment broths. The culture meth od combined with either rapid method from each broth increased the sen sitivity to 97.5-98.8% when data from UVM and LRB was combined. False negative rates in the USDA method (7.1%) were attributed to the lack o f color change in Fraser secondary broth. Fraser broth also yielded ma ny false positive results (overall 66.2%) making this broth of limited value as a screening tool for highly contaminated samples. In order t o optimize methodology for detection of Listeria, suppression of backg round flora and the recovery of potentially injured Listeria in the pr ocessing environment must be addressed. Overgrowth occurring during th e nonselective enrichment period was suspected of causing suboptimal s ensitivity in LRB; however, the combination of UVM and LRB showed prom ising recovery rates. Ceftazidime was evaluated against 68 background isolates that survived throughout the various enrichment and detection methods. Inhibition of 57 of the contaminants indicates a potential r ole for ceftazidime in the LRB selective-agent regime for sites with h igh microbial background.