Foodborne botulism outbreaks

Human foodborne botulism outbreaks have typically been linked to the consumption of toxin-contaminated home-prepared or home-preserved foods (Maselli, 1998). The vast majority of foodborne botulism cases are attributed to toxin types A, B, or E. Maselli (1998) reports that type B is the most prevalent (52%) in the United States, followed by type A (34%) and type E (12%), while the CDC (1998) suggests 37.6% of all foodborne botulism outbreaks since 1950 were caused by type A, 13.7% by type B, 15.1% by type E, 0.7% by type F, and 32.9% were unidenhfied with respect to toxin type. Outbreaks of type F and G botulism are rare (Sonnabend et al., 1981 Maselli, 1998 Richardson et al., 2004), and only anecdotal reports of isolated type Cl and D botulism cases can be found in the published literature (e.g., Lamanna, 1959). 

The natural epidemiology of foodborne botulism provides additional insight into the similarities and discrepancies between the human disease and that represented in various animal models. In the United States, around 25% of reported human botulism cases are classified as foodborne and 72% are infant (Mackle et al., 2001). Human type A and B foodborne botulism cases occur worldwide and constitute the vast majority of reported human intoxications (Maselli, 1998). The majority of other botulism cases are attributed to serotype E and are typically associated with the consumption of contaminated seafood. Generalizations have been made regarding the geographic distribution of the most common C. botulinum strains within the United States. Most human foodborne botulism outbreaks occurring west of the Mississippi are due to type A toxin type B strains are more prevalent east of the Mississippi, while type E strains are typically isolated to Alaska and the Pacific Northwest (Amon et al., 2001 Richardson et al., 2004). 

The onset of symptoms in botulism depends upon the amount of toxin ingested or inhaled and the related kinetics of absorption. Time to onset can range from as short as 2 h to as long as 8 days, although symptoms typically appear between 12 and 72 h after consumption of toxin-contaminated food (Lecour et al., 1988 Amon et al., 2001). In a review of 13 foodborne botulism outbreaks involving 50 patients from 1970 to 1984, the incubation period ranged from lOh to 6 days (Lecour et al., 1988). 

Although the number of outbreaks has been relatively constant, the case to fatality ratio has improved markedly. From 1899 to 1950, foodborne botulism was associated with 60% mortahty from 1950 to 1996, the average annual mortahty fell to 15.5% (Shapiro et al., 1998), and decreased to 4% during the last decade (Sobel et al., 2004). These advances in survival have come primarily from improvements in critical care (Tacket et al., 1984 Sobel et al., 2004). Further reductions in morbidity and mortahty from botuhnum intoxication wiU require better methods for detection and diagnosis of BoNT outbreaks and availability of speciftc pharmacological treatments (Franz et al., 1997 Dickerson and Janda, 2006). 

Although infant botulism was not recognized until a large outbreak occurred in Califomia in 1976 (Pickett et ah, 1976), it is currently the most prevalent form of botulism in the United States, accounting for approximately 70% of all cases (Shapiro et al., 1998). Because infant botulism results from a continual production of BoNT, it appears to be more effectively treated by antitoxin than is foodborne botulism. In a recently concluded 5 year randomized clinical trial carried out with a human botulinum immune globulin (BIG-IV), it was found that administration of BIG-IV within 3 days of hospitalization resulted in a 3 week reduction in the mean hospital stay, as well as substantial reductions in the time needed for intensive care and mechanical ventilation (Amon et al., 2006). In a nationwide open label study, BlG-lV was found to be effective even when administered 4—7 days after hospital admission, although to a somewhat lesser extent than when infusion was initiated at 3 days (Arnon et al., 2006). 

Since inhalation bomlism does not occur in nature, aU outbreaks must be considered as suspicious. Prudence would dictate that each should be treated as a criminal or terrorist attack, unless other causes are found (Arnon et al., 2001). From the limited human and animal data currently available, inhalation bomlism does not have a unique presentation rather, the signs and symptoms resemble those of other forms of botulism. The latent period is comparable with that of foodborne bomlism without the early G1 signs (Adler, 2006). 

---