BACKGROUND: Vibrio parahaemolyticus is the most commonly reported species causing vibriosis in the United States, an estimated 45,000 infections annually. We describe the epidemiology of V. parahaemolyticus infections in the United States.
METHODS: State Health Departments report V. parahaemolyticus infections to the Centers for Disease Control and Prevention (CDC) using the Cholera and Other Vibrio Illness Surveillance (COVIS) report form. Reporting began in 1988, and became officially notifiable in 2007; most states were reporting by 2000. Cases reported during 1999 through 2012 were analyzed. Transmission route was categorized as foodborne, non-foodborne, or unknown based on site of specimen collection and reported exposures. Crude incidence rates (cases per million) for the years 1999-2011 were calculated using US Census Bureau estimates. Since 2000, states have submitted many V. parahaemolyticus isolates to CDC for serotyping.
RESULTS: Between 1999 and 2012, 3,726 V. parahaemolyticus infections were reported. Most (78%) patients had V. parahaemolyticus isolated from a gastrointestinal site, 14% from a skin or soft tissue site, and 2% from a blood or other normally sterile site. The median age of patients was 46 years (range 1-94) and 65% were male. Twenty-one percent of patients were hospitalized; 1% died. Forty percent of cases occurred in the Pacific region, followed by the Atlantic (30%). The states of California, Washington, New York, Florida, and Texas reported 56% of all infections. Overall, 81% of cases were categorized as foodborne (range 72%-90%, annually), 16% non-foodborne (range 8%-24%, annually), and 3% unknown (range 0%-9%, annually). All transmission routes peaked during the summer months. Among foodborne cases, 57% reported oyster consumption (92% of oysters consumed raw) and 17% clam consumption (51% of clams consumed raw). Exposure to a body of water was reported for 85% of non-foodborne cases. In 1999, the overall incidence of infections was 0.45 per million; annual increases in occurred in 8 of the 12 subsequent years, reaching 0.99 in 2011 following a peak of 1.3 in 2006. Trends in foodborne infections mirrored this pattern (1999 [0.37], 2006 [1.2], 2011 [0.75]). In 1999, the incidence of non-foodborne infections was 0.07; it declined to a low of 0.04 in 2001, and then increased steadily to 0.22 in 2011.The most common serotype reported was O4:K12 (14% of 1,378 isolates serotyped).
CONCLUSIONS: The incidence of both foodborne and non-foodborne V. parahaemolyticus infections is increasing in the United States. More stringent prevention measures need to target both foodborne and non-foodborne transmission routes.