Serving high-risk foods in a high-risk setting: survey of hospital food service practices after an outbreak of listeriosis in a hospital.

BACKGROUND AND OBJECTIVES: Prepared ready-to-eat salads and ready-to-eat delicatessen-style meats present a high risk for Listeria contamination. Because no foodborne illness risk management guidelines exist specifically for US hospitals, a survey of New York City (NYC) hospitals was conducted to characterize policies and practices after a listeriosis outbreak occurred in a NYC hospital. METHODS: From August through October 2008, a listeriosis outbreak in a NYC hospital was investigated. From February through April 2009, NYC's 61 acute-care hospitals were asked to participate in a telephone survey regarding food safety practices and policies, specifically service of high-risk foods to patients at increased risk for listeriosis. RESULTS: Five patients with medical conditions that put them at high risk for listeriosis had laboratory-confirmed Listeria monocytogenes infection. The Listeria outbreak strain was isolated from tuna salad prepared in the hospital. Fifty-four (89%) of 61 hospitals responded to the survey. Overall, 81% of respondents reported serving ready-to-eat deli meats to patients, and 100% reported serving prepared ready-to-eat salads. Pregnant women, patients receiving immunosuppressive drugs, and patients undergoing chemotherapy were served ready-to-eat deli meats at 77%, 59%, and 49% of hospitals, respectively, and were served prepared ready-to-eat salads at 94%, 89%, and 73% of hospitals, respectively. Only 4 (25%) of 16 respondents reported having a policy that ready-to-eat deli meats must be heated until steaming hot before serving. CONCLUSIONS: Despite the potential for severe outcomes of Listeria infection among hospitalized patients, the majority of NYC hospitals had no food preparation policies to minimize risk. Hospitals should implement policies to avoid serving high-risk foods to patients at risk for listeriosis.

Antimicrobial resistance trends of Shigella serotypes in New York City, 2006-2009.

Shigellosis is the third most common enteric bacterial infection in the United States. Although infection is typically self-limiting, empiric treatment is often prescribed. Because of increasing antimicrobial resistance to Shigella, empiric treatment options are decreasing. Identifying resistance patterns can inform empiric treatment recommendations. The goals of our study were to examine risk factors associated with antimicrobial resistance of Shigella and examine issues related to empiric treatment and antimicrobial resistance of Shigella. During June 2006-February 2009, we attempted to interview all New York City patients reported to have shigellosis. Their Shigella isolates were tested for antimicrobial susceptibility to examine the level of resistance and identify risk factors for resistance. Analysis was conducted on two groups distinguished by a large outbreak that was documented during the data collection period. Of the 477 nonoutbreak patients, 333 (70%) patients reported taking an antibiotic for shigellosis and 36 (11%) were treated with an antibiotic to which their Shigella infection was resistant. Among this group, high levels of antimicrobial resistance were detected to amoxicillin-clavulanate (66%), ampicillin (68%), and trimethoprim-sulfamethoxazole (66%). Non-travel-associated ciprofloxacin-resistant Shigella (five patients) and ciprofloxacin-resistant Shigella sonnei (four patients) were reported for the first time to our knowledge. Antimicrobial resistance is significantly higher in New York City residents compared with national data. Some patients were treated with therapies that were not effective and to which the patient's Shigella infection was resistant. Shigella infections should not be treated with antibiotics unless the patient presents with severe or underlying illness and is at risk for systemic illness. When treatment is indicated, local monitoring of Shigella for antimicrobial resistance will provide local clinicians with the best guidance for effective empiric treatments.