A total of 10 persons were counted as outbreak associated cases; five were laboratory confirmed with an indistinguishable genetic strain of E. coli O157:H7 as determined by pulsed field gel electrophoresis (PFGE). Four patients were hospitalized. Three patients developed Hemolytic Uremic Syndrome (HUS). Outbreak associated cases ate at Coco Loco on April 15, April 16 or April 17. Two patients became ill on April 17. Five patients experienced symptom onset on April 18. The last patient became ill on April 22. The outbreak was assigned Cluster Identification Number 1305TXEXH-1.
PFGE testing showed that Coco Loco outbreak associated cases were infected with a common genetic strain of E. coli O157:H7, designed by PulseNet as pattern combination identification numbers EXHX01.0047/EXHA26.0015. Additional testing by a procedure known as Multiple Locus Variable-number Tandem Repeat Analysis (MLVA) conducted at the Centers for Disease Control and Prevention (CDC) revealed that the genetic strain of Coco Loco cases was new to the PulseNet database based on MLVA results. The additional testing results of MLVA led investigators to conclude that source of E. coli O157:H7 in Coco Loco patients was distinct from sources other patients with strain EXHX01.0047/EXHA26.0015 had. Id.
Public health investigators conducted a case control study to identify a specific Coco Loco food or menu item that was the source of the outbreak. For the case control study, a confirmed case was defined as a person laboratory confirmed with E. coli O157 who ate at Coco Loco between April 15-17 with illness onset between April 17-22. A probable case was defined as a person with symptoms consistent with an E. coli O157 infection who was epidemiologically linked to a confirmed or probable case. There were five confirmed cases and five probable cases included in the case control study. Controls were selected from credit card charges at Coco Loco between April 15 -17 and through interviews with cases. One health department employee who ate at Coco Loco during the dining dates of interest was included as a control. In total there were 9 controls in the case control study. Id.
Results of the case control study showed that menu items containing ground beef were statistically associated with illness. Cases were more than twenty times likely to consume ground beef prepared at Coco Loco than controls. The regular odds ratio (28.0, 95% Confidence Limits 1.5-1400, p=0.015) was calculated as well as Fishers Exact Odds Ratio (21.4, 95% Confidence Limits 1.53-1396.0) due to the small sample size. No other food item or menu item was statistically associated with illness in this outbreak. Id.
Findings that Coco Loco outbreak associated cases were infected with a unique strain of E. coli O157:H7 along with epidemiologic data implicating ground beef as the source of the outbreak allowed public health investigators to focus on ground beef supplied to the restaurant.