BACKGROUND: Salmonellais the leading cause of bacterial foodborne illness in the United States. Although salmonellosis is often self-limiting, persons with drug-resistant infections are at risk for severe clinical outcomes. Pennsylvania conducts integrated surveillance that compares isolates from clinical and retail meat sources.
METHODS: During 2009-2014, chicken breasts, ground turkey, ground beef, and pork chops (600 of each) were purchased from randomly selected retail outlets in southeastern Pennsylvania and cultured for enteric pathogens. Salmonellaisolates were analyzed by pulsed-field gel electrophoresis (PFGE). We compared the PFGE patterns from retail meat specimens with those of human clinical isolates in the Pennsylvania surveillance database. A subset of isolates was subtyped using a novel technique, CRISPR-MVLST, and compared to a database of clinical isolates. We performed antimicrobial susceptibility testing (AST) on retail meat isolates and a subset of matched clinical isolates.
RESULTS: A total of 177 Salmonella isolates were recovered from 2400 meat samples: 18.2% of chicken samples, 9.5% of turkey, 0.3% of beef, and 1.1% of pork were culture-positive. PFGE patterns of 86 (48.6%) retail meat isolates matched to patterns obtained from 1665 clinical isolates; 40 distinct PFGE patterns were represented among the shared patterns found in both meat and clinical isolates. Seventeen (43%) of the 40 shared PFGE patterns were associated with multiple drug resistance (MDR) (resistant to ≥ 3 drugs). Among the 175 retail meat, 74 (42%) were MDR; 10 (59%) of the 17 shared MDR-associated PFGE patterns had identical AST profiles in both meat and clinical isolates. One S. Dublin PFGE pattern (JDXX01.0004) found in 11 isolates (10 clinical and one beef) showed resistance to ≥6 antimicrobials, including resistance to ampicillin and ceftriaxone. One S. Typhimurium isolate from pork and 22 clinical S. Typhimurium isolates had an ACSSuT resistance profile. Four S. Kentucky meat isolates had same PFGE profile with six clinical isolates and all were resistant to streptomycin. CRISPR-MVLST analysis confirmed that Salmonellastrains overlap between retail meat and clinical isolates.
CONCLUSIONS: Salmonella isolates cultured from retail meat samples shared 40 unique PFGE patterns with isolates known to have caused human illness. Of these shared patterns, 43% were multidrug resistant. This comparative investigation suggests that some antimicrobial resistant Salmonellainfections in humans are likely related to exposure to contaminated meat.