The Washington Post - September 14, 2010
Egg-loving salmonella bacteria have been sickening people for decades
By David Brown
The unfolding story of how salmonella bacteria infected two giant egg
operations in Iowa this summer is the latest chapter of a mysterious narrative
about how a minor bacterial annoyance took off 35 years ago to become the second
most common cause of food-borne illness in the United States.
Like the things that cause AIDS, Lyme disease, Legionnaire's disease and West
Nile fever, the egg-loving germ (whose formal name is Salmonella enterica
serovar Enteritidis) is a classic "emerging infectious agent." Sometimes called
SE, it's a microbe that has been around a long time and has found a new or
better way to reach its human victims.
For more than three decades, the strain of salmonella bacteria with a fondness
for eggs has taken advantage of changes in this country's animal husbandry, food
distribution and eating habits.
Along the way, scientists and public health officials have paid increasing
attention to it, culminating recently in the Food and Drug Administration's
71-page "egg safety rule," which took effect in July. The federal government
hopes those regulations will prevent 80,000 of the 142,000 cases of egg-related
salmonella infection that occur in the United States each year (out of an
estimated 1.54 million cases of food-borne salmonella illness). They hope the
rule will cut the number of infected "table eggs," currently estimated at 2.3
million of the 47 billion produced each year. The new standards might even
reduce health-care costs by $1.4 billion.
But they are unlikely to eradicate the problem.
The problems in Iowa hark back to the early days of the SE pandemic. (The
bacterium's worldwide, near simultaneous resurgence starting about 1980 has
earned that designation.) Since April, health departments in 10 states have
investigated 29 outbreaks of salmonella illness, all of them traceable to
restaurants. Fifteen appear to be linked to eggs from a group of Iowa farms
known as Wright County Egg, and at least one to a nearby operation, Hillandale
Farms. In all, about 1,500 people have gotten sick. Such large, geographically
dispersed outbreaks traceable to a single source are rare these days.
"This current outbreak is like deja vu," said Christopher R. Braden, the
epidemiologist at the Centers for Disease Control and Prevention who is leading
the Iowa investigation.
There are many species of salmonella bacteria, the most dangerous being
Salmonella typhi, which causes typhoid fever, a frequently fatal bloodstream
infection. One of the species that causes intestinal infection is Salmonella
enterica, which has 2,500 distinct types, or serovars, all causing pretty much
the same symptoms: vomiting, stomach pain, diarrhea and fever.
One of those serovars, SE, behaves differently from the others in crucial ways.
In hindsight it's clear that something happened in the 1970s to break out it out
of the pack.
From the mid-1970s to the mid-1980s, SE infections increased sixfold in the
northeastern United States. In 1986, a large outbreak in New England that was
linked to stuffed pasta shells gave epidemiologists two crucial insights: All
the illness was egg-related and traceable to a single source, a farm in
Connecticut that provided the eggs used in the stuffing.
For a while the SE problem was confined to the Northeast, where human cases
peaked in 1989. Then it surfaced dramatically in California. It has slowly
spread to the rest of the country, although it has always been rare in the
Southeast for reasons unknown.
Not so perfect
As it happens, this wasn't SE's first appearance in eggs.
In the 1920s and 1930s, there was so much intestinal illness associated with
duck eggs that that food -- once more popular than chicken eggs -- disappeared
from the American table. There is good evidence that SE was the germ, one that
would remain a low-grade threat for decades to come.
The discovery in the 1960s that the strains found in egg-associated outbreaks
were the same ones found in the feces of the hens that laid the eggs led to
rules requiring most eggs to be washed and inspected for cracks before they were
sold. That caused a noticeable decline in egg-associated outbreaks, and by 1976
SE was responsible for only 5 percent of illness caused by all salmonella
The reemergence of SE in the 1980s led to the search for other causes of
illness, because eggs with dirty shells clearly weren't the explanation. It
turns out the problem was "vertical transmission": passage of the microbe from
mother to offspring, or, in this case, mother to egg.
In experiments done in the late 1990s and early 2000s at the U.S. Department of
Agriculture, Richard Gast proved that SE-infected chickens could lay infected
When a hen eats something contaminated with SE, the bacterium can move from the
intestine to the bloodstream and from there to the reproductive tract. If an
ovary becomes infected, bacteria are often deposited in the yolk. If the
oviduct, downstream from the ovary, is infected, the microbe is more likely to
end up in the white. If the eggs were fertile, the resulting chicks may be
infected. The hens often recover, and even when they are infected many of their
eggs are entirely normal.
Nevertheless, news that nature's perfect food packages were sometimes
disease-carrying grenades did not go over well.
"I think it would be fair to say that a lot of people on the industry side said
that was not a viable theory," said Braden.
But the discovery should not have been a surprise.
The early-20th-century outbreaks of salmonella that ravaged chicken flocks,
killing nearly 80 percent of newly hatched chicks, led to the creation of the
National Poultry Improvement Plan in 1935. It advised farmers how to eradicate
the bacteria and certify their flocks as disease-free. The NPIP was so
successful that the serovar responsible for "fowl typhoid," Salmonella
gallinarum, essentially disappeared.
Biologists later discovered that that bacterium's cell wall has a molecular
feature, called the O antigen, identical to one in SE cells. They theorize that
birds infected with S. gallinarum were immune to SE, which was more commonly
carried by rodents. (A single mouse dropping can contain 100,000 salmonella
cells.) When S. gallinarum disappeared, SE moved into its territory -- chickens.
From 1985 through 2003, 997 outbreaks of food-borne illness caused by SE were
reported to health authorities. Three-quarters of the outbreaks involved eggs.
The number of foods involved was extremely varied: ice cream, salad dressing,
eggnog, omelets, crab cakes, chiles rellenos, pasta dishes, cream pies.
In all 33,687 people fell ill, and 82 died. However, that was just the tip of
the iceberg; the CDC estimates that for every reported illness there are 38
Those alarming numbers led some states to create voluntary "egg quality
assurance programs" (EQAP) in which egg farmers instituted "biosecurity"
practices, rodent-proofed their barns, routinely cultured henhouses for
salmonella, tested eggs if the samples were positive, and followed specific
disinfection procedures. Pennsylvania's program, started in 1994, is the model
for the FDA rules that became mandatory nationwide this summer.
The programs made a big difference. A study published in 2004 showed that states
with EQAPs experienced an average 72 percent fall in SE incidence in the 1990s.
(Iowa, which produces more eggs than any other state and is the site of the
recent problems, notably did not have an EQAP.) But since 2000 that decline has
Changes in bacterial or viral genes sometimes play a role in emerging
infections. But, as often happens, factors beyond the microbe have contributed
to the rise of the bug that is making consumers queasy.
"There are usually changes in the environment that permit rapid spread of an
organism, where previously there wasn't an open door," said J. Glenn Morris, a
physician who heads the Emerging Pathogens Institute at the University of
Florida in Gainesville.
Industrial farming put huge numbers of birds in confined spaces heavily
contaminated with their manure, which can harbor salmonella for as long as two
years. Many farmers bought chicks from large suppliers, spreading the bacteria
between regions. Some chicken farmers forced their hens to molt by starving them
for a brief period. That dramatically increases the length of a hen's egg-laying
life, but the stress can increase a bird's susceptibility to SE infection and
increase the likelihood it will produce infected eggs.
At the same time, restaurants and food processors often pooled hundreds or
thousands of eggs in cooking, allowing a few infected eggs to contaminate many
The "infectious dose" of SE isn't known, although some outbreaks suggest that
fewer than 50 cells of the bacterium can make a person ill. The best way to
avoid infection is to cook eggs until both white and yolk are hard, and to cook
dishes containing eggs until the internal temperature reaches 160 degrees.
But what caused the U.S. pandemic to break out in New England in the late 1970s?
And why did SE's prevalence increase simultaneously in South America and Europe?
Was it the cumulative effect of these variables, or did something else happen?
Did the bacterium become more virulent or American chickens more susceptible?
"That's the $64,000 question," said Robert V. Tauxe, a CDC scientist who has
spent more than two decades studying food-borne infections.
The USDA's Gast agrees. "It's hard for me to believe that this just suddenly
happened," he said recently. "It's very hard to explain under any of the