This is another it what will be a long – too long – series of outbreak investigations where we have represented consumers in what I hope will be a cautionary tale, and a learning experience, for manufacturers of food.

On July 7, 2007, the Centers for Disease Control and Prevention (“CDC”)[1] learned that two siblings in Texas were critically ill with botulism and that their illnesses were likely acquired by eating contaminated food.  The two children were admitted to pediatric intensive care, and there required mechanical ventilation.  The CDC released doses of botulinum antitoxin,[2] which was administered to the children the next morning.

Four days later on July 11, public health officials in Indiana reported to the CDC that a married couple in Indiana was suspected of having foodborne botulism.  Serum samples were collected from each of them on July 10 and then sent to the Botulism Reference Laboratory at the CDC.  On July 16, one day after the lab received the serum samples, botulinum toxin type A was detected by mouse bioassay in the man’s serum sample.  Botulinum toxin was also detected by mouse bioassay in serum submitted by the wife, but the sample volume was insufficient to determine the toxin type.  Investigations conducted by state and local health departments in both Texas and Indiana revealed that all four patients had eaten types of Castleberry’s hot dog chili before symptom onset.

Texas investigators found an unopened can of Castleberry’s Austex Hot dog Chili Sauce Original date stamped with a manufacture date and time of May 7 at 9:41 p.m. at the children’s home and tested it for botulism.  The Texas Department of Health Services laboratory tested an aliquot from this can using an enzyme-linked immunosorbent assay (ELISA) for botulinum toxin and did not detect the toxin.

The Indiana couple had an unlabeled, sealed plastic bag of leftover chili mixture in their refrigerator that local health officials collected and sent to the CDC for C. botulinum toxin testing.  On July 16 the CDC detected botulinum toxin type A by mouse bioassay in the chili mixture.  Empty, well-rinsed cans of Castleberry’s Hot Dog Chili Sauce Original and chili made by another company were found in the couple’s recycling bin. CDC re-rinsed the two cans and tested the rinse water for botulinum toxin by mouse bioassay; both were negative.  The label on the can of Castleberry’s Hot Dog Chili Sauce Original indicated a production-date of May 8, and a time of 2:23 AM—less than five hours after the production-time indicated on the can collected from the Texas home.

On July 17, CDC staff provided information regarding the production-dates and times to the FDA.  The evidence strongly suggested that brands of Castleberry’s hot dog chili sauce were the common source of the four ill persons with botulism.  On July 18, FDA issued a consumer advisory.  On that same day, after being informed about the outbreak, and findings from the FDA investigation of the canning facility, Castleberry’s Food Company issued a voluntary recall that included a limited number of production dates of Castleberry’s Hot Dog Chili Sauce Original, Castleberry’s Austex Hot Dog Chili Sauce Original, and Kroger Hot Dog Chili Sauce. The recall was expanded on July 21 to include all production dates for 91 types of canned chili sauce, chili, other meat products, chicken products, and dog food that were manufactured in the same set of cookers, or “retorts” as the hot dog chili sauce at the Castleberry’s facility in Augusta, Georgia.

By August 24, eight cases of botulism had been reported to the CDC.  In addition to the Indiana couple, the mother of the children in Texas had developed symptoms of botulism, which brought the total number of Castleberry-associated cases in Texas to three.  There were also three unrelated residents of Ohio who had developed botulism consuming Castleberry’s hot dog chili sauce in the week before symptom onsets.  Botulinum toxin was identified in leftover chili sauce collected from the refrigerator belonging to one of the Ohio cases.

The Castleberry’s manufacturing facility in Georgia produces products regulated both by the FDA and USDA-FSIS.  Initial reports of illnesses were linked to meatless hot dog chili sauce and thus, fell under the jurisdiction of the FDA.  The agency’s Atlanta District Office took the lead in the investigation of facilities.

The inspection started on the evening of July 17.  FDA investigators requested company maintenance records, which were not immediately available because they were stored on a laptop of a vacationing employee.  Finally, three days later, under threat of severe penalty, the company produced some of the requested records.  Included in records provided to federal investigators was a 42-page report written by a consultant hired by Castleberry’s to investigate swollen cans of stew, chili, and hash produced in April and May 2007.  The consultant had attributed spoilage to post-process handling operations in one of the plant’s cooking equipment.  Reports by two other company-hired consultants would also implicate post processing as the reason for swollen cans.  Unfortunately, Castleberry’s had not investigated the issues further.

On July 18 and 19, a team of federal investigators was sent to the firm’s warehouse.  Samples of Castleberry’s Austex and Castleberry’s brand Hot Dog Chili Sauce with the “best by May 7, 2009” and “best by May 8, 2009” lot codes were collected and sent to FDA laboratories for testing.[3]  FDA testing of sample 428113, consisting of 17 swollen cans, found C. Botulinum toxin in 16 of the cans.  This sample included the same time-stamp and lot code from the May 8, 2007 production as the can found in the Indiana home. FDA testing of sample 420352, consisting of six swollen cans, found C. Botulinum in four cans.  FDA sample 420353 included one swollen can, and its contents tested positive for C. Botulinum toxin.

Federal investigators conducted extensive tests on Castleberry equipment.  The findings are presented in an FDA report issued on August 10, 2007, FDA Inspectional Observations dated 08/10/2007, (Summary pages only).  Noted observations include:

  • The system, equipment, and procedures used for thermal processing of foods in hermetically sealed containers were not operated and administered in a manner that ensures commercial sterility is achieved.
  • Each retort did not have an accurate temperature records device.
  • Failure to supply a suitable water valve used for water cooling to prevent leakage of water into the retort during processing.
  • The condensate bleeder was not checked with sufficient frequency to ensure removal of condensate or equipped with an automatic alarm system for the continuous monitoring of condensate bleeder functioning.
  • Required information was not entered on designated forms at the time the observation was made by the retort or processing system operator or designated person.
  • Failure to maintain fixtures in repair sufficient to prevent food from becoming adulterated.
  • Failure to properly adjust the temperature-recording device.  The temperature recorded on the temperature-recording device chart was higher than the mercury-in-glass thermometer during processing.

The report ultimately placed blame on Castleberry management saying there was no commitment from employees in making the products and there was not adequate management oversight.  As one Castleberry employee noted:  “Two years ago the [implicated reports] were maintained very well, but they are maintained poorly now.”  The FDA plainly agreed, citing Castleberry’s for the “failure to maintain fixtures in repair sufficient to prevent food from becoming adulterated.”

Castleberry made substantial fixes at its plant and then reopened in the fall of 2007.  The company re-branded its line to American Originals, and redesigned product labels. But in March, 2008, the plant was forced to close again after a February 27 joint-inspection by the FDA and USDA revealed deviations in some equipment operations on the processing line.  The line was not related to deficiencies noted in the summer of 2007 but because under-processing caused the botulism outbreak, the plant’s operating permit was suspended.


[1]           The information about the outbreak comes primarily from the CDC-published report issued July 30, 2007. See MMWR; supra note 1, at 1-2.  The information specific to the Castleberry’s manufacturing facility is taken from the FDA Establishment Inspection Report, FEI No. 1010894.

[2]           The CDC is the only source of the therapeutic antitoxin, which is stocked in locations around the country for rapid release. See Sobel; supra note 2, at 1607.

[3]           This was despite Castleberry’s questionable efforts to apparently destroy evidence.  As noted in the FDA Inspection Report, “Castleberry’s personnel had sorted through lots and destroyed most of the swollen cans prior to our visit.”  Although the FDA has not imposed any specific penalty for this conduct yet, it is safe to assume that the conduct would be scrutinized by a court as potential spoliation.