A Primer on E. coli O145: A Little Contaminated Lettuce and a Lot of Different Os

Here is a quick summary of the recent Escherichia coli O145 outbreak associated with Romaine lettuce, and it highlights the amazingly quick FDA actions to prevent additional disease:

• On April 16th, public health authorities recognized an outbreak of bloody diarrhea at a university in Michigan that was later confirmed as being due to E. coli O145. The outbreak was subsequently linked to other similar school-based clusters in Ohio and New York.
• On April 27th, preliminary information linked these illnesses to a common supplier of Romaine lettuce.
• By the next day, April 28th, FDA had determined that the implicated production lots of Romaine lettuce were produced in late March from a single farm and accounted for all of the illnesses. These implicated lots were no longer in commerce, and no recall was necessary.
• Subsequent laboratory investigations of Romaine lettuce showed at least intermittent contamination on later production days from the processor and triggered preemptive recalls of first a single contaminated lot and then all production from the implicated farm. No illness has been associated with these later lots of recalled lettuce.

Recent reports of one case from Tennessee and another from Pennsylvania do not reflect expansion of the outbreak but, rather, the fact that information about those cases was recently reported. These cases are actually part of the original cluster, which was linked to the original implicated lot of lettuce from March. This long lag time between somebody getting sick and their illness finally being reported to a national database remains a challenge for our national foodborne illness detection and response system.

I was not very familiar with the serotypes of E. coli other than E. coli O157, so I turned to our resident expert, Dr. Patricia Griffin for basic information that health and lab professionals need to know.  I thought it might be helpful to share her expertise with you. As laboratory testing expands, I expect that we will recognize more such outbreaks involving groups of E. coli other than O157.

E. coli Basics for Health and Lab Professionals

 Shiga toxin-producing Escherichia coli (STEC) can cause diarrhea, bloody diarrhea (hemorrhagic colitis), and hemolytic uremic syndrome (HUS) in persons of any age. The usual reservoirs are cattle and other ruminants (mammals that rechew their cud [food that returns from the stomach], including goats and sheep). Features of HUS include low platelet count, anemia due to broken blood cells, and kidney failure. Among those with HUS due to STEC O157, the mortality in children <5 years old is 3% and that in persons >60 years old is even higher.

Nomenclature and toxins

All STEC produce Shiga toxin. Some investigators, e.g., many in Canada, use the term verocytotoxin-producing E. coli (VTEC); this term is completely interchangeable with STEC. STEC strains that make people sick only rarely cause illness in animals. All STEC produce one or more of two similar toxins, Shiga toxin 1 and Shiga toxin 2. The toxins are named for Dr. Shiga, who first described the bacterial origin of dysentery caused by a rare type of Shigella (Shigella dysenteriae type 1) that produces Shiga toxin 1.

Serogroups

About 70% of non-O157 STEC isolated from humans fall into six serogroups, which are (in order of frequency): O26, O103, O111, O121, O45, and O145. Among the >50 other STEC serogroups isolated from ill persons, most are probably human pathogens. Some STEC serogroups isolated from animals are almost never isolated from ill persons. Many of these may not be human pathogens. STEC that are known to cause human illness are sometimes called enterohemorrhagic E. coli (EHEC).

Comparison of O157 and non-O157 STEC

We estimate that in the United States about half the diarrheal illnesses caused by STEC are due to O157 and about half to STEC with subtypes of E. coli other than O157. However, we estimate that about 90% of diarrhea-associated HUS is caused by STEC O157. Overall, for STEC O157 illnesses, about 6% (15% in children <5 years old) result in HUS and 42% of affected individuals are hospitalized. Among all STEC illnesses with subtypes other than O157, about 2% result in HUS and 12% are hospitalized. It is important to note that some of these STEC illnesses due to serogroups other than O157 can cause illness just as severe as that caused by STEC O157.

Current and previous STEC O145 outbreaks

The current outbreak is the first U.S. outbreak due to STEC O145 in which food was implicated.  STEC O145 was isolated in two other U.S. outbreaks. In Oregon in 2005, several pathogens, including STEC O145, caused illness in 60 persons; transmission was probably through drinking water. In Minnesota in 1999, STEC O145 caused illness in 2 persons; transmission was person-to-person in a child care facility. In a report of a foodborne outbreak of STEC O145 and O26 infections from Belgium, ice cream was the vehicle. CDC has compiled a list of recognized non-O157 STEC outbreaks in the United States through 2008.

Lab issues

STEC O157 has a biochemical marker that permits screening by culture on special laboratory media. We estimate that about 70% of U.S. labs use this media on all submitted stool specimens. Most non-O157 STEC, including O145, lack a biochemical marker that distinguishes them from non-pathogenic E. coli. In 1995, an enzyme-linked immunoassay (EIA) that can detect Shiga toxin came on the market. Since then, the number of labs using it has gradually increased. However, we estimate that only about 5% use it on all submitted stool specimens.  Once the presence of Shiga toxin is determined, suspect E. coli must be forwarded to a state laboratory or to CDC for the serogroup to be determined. Partly because this takes time, CDC advises that all labs test all stool specimens both for STEC O157 using selective media and for Shiga toxin. Because so few labs now test routinely for STEC subtypes other than O157, most of non-O157 STEC illnesses are not diagnosed.