By Rose Hoban
When David Sweat received a phone call late one Monday afternoon that a Gaston County child was hospitalized with an E. coli infection, he knew he’d miss dinner that night, and for many nights to come.
It was Oct. 8. Sweat, an epidemiologist with the Communicable Disease Branch of the state Division of Public Health, knew a county fair had just wrapped up in the Gaston County vicinity. Sweat also knew that only a small percentage of people exposed to E. coli end up with kidney failure like the child in the hospital.
“We knew immediately this was likely to be a bigger outbreak,” Sweat said.
Sweat stayed late that night, making phone calls to learn more of what was happening and to consult with doctors and lab technicians where the child was hospitalized.
The next morning, two more calls came in. Sweat’s colleagues dropped what they were doing and started looking for more cases. They found four that day, all hospitalized in Charlotte.
“We knew if there was severe illness, they’d go to a Charlotte hospital,” said branch chief Evelyn Foust. “We were checking first with hospitals in that area to see if there were similar illnesses that were being cared for with similar diagnosis and not reported yet.”
Eventually, Foust’s experts identified an outbreak of more than a hundred people. Most of those are probable cases, identified by symptoms only; 17 are confirmed cases, identified by bacteria culled from fecal samples.
Those people got sick from exposure to the bacteria Escherichia coli, a bug found commonly in the guts of ruminants such as goats, sheep and cows. One bacteria subtype, called 0157, produces a toxin that has the potential to kill by first causing kidney failure and then multi-organ failure.
And kill it did. One 2-year-old Gaston County boy is dead, his brother is sick from their exposure and at least 13 people overall have been hospitalized.
Finding a bacteria in a haystack
“We started getting cases reported every day and we suspected the connection was the county fair, but we had to confirm,” said Foust. She explained that people in her office are hesitant to narrow their search too quickly for fear that they’ll miss something significant that could come out of left field.
But they’re also racing against time.
“If there’s something happening acutely, we need to take measures to stop people from getting any sicker,” she said. That includes telling parents of children with probable cases that they need to stay home from daycare or school to prevent secondary infections – infections of others who didn’t get sick from the initial cause.
By the end of the day Wednesday, Oct 10, the folks in the Communicable Disease Branch had ruled out other possibilities: The source was the Cleveland County Fair.
One of the epidemiologists, Zack Moore, sent out a blast communication to doctors, clinics and hospitals in surrounding counties – including neighboring South Carolina counties – alerting them to look for patients with symptoms characteristic of E. coli infection and who had some connection to the fair.
“We’ve learned only 30 percent of people with E coli will have bloody diarrhea,” Moore said. Symptoms can include diarrhea, bloody diarrhea, fever and other flu-like symptoms.
“So we figure many other people got sick, but [we] wouldn’t have found them because their symptoms are not extreme enough to go to the doctor or hospital,” Moore said. Their strategy is to actively seek out cases rather than waiting for reports of illness to roll into state epidemiology offices.
The decision also meant telling officials at the Cleveland County fairgrounds in Shelby that they had to cancel upcoming events.
Two people were dispatched to the fairgrounds, where the exhibits and animals had been packed up days earlier. They collected swabs to confirm the relationship between the E. coli in stool samples and the bacteria on the ground.
“We were crawling around, collecting dirt and poop with disposable spoons,” said Carl Williams, a public-health veterinarian who collected more than 40 samples into sterile cups that came back to the state lab. There technicians compared bacteria from the soil samples with that from patients.
The bacteria persists. After the state fair E. coli outbreak in 2004, Foust’s team found the bacteria in the soil five months later. Aedin’s Law, passed in the wake of that outbreak, insures that the soil under animal exhibits where an E. coli outbreak takes place remains unused for six months.
“E. coli can be recovered from surfaces for more than five months,” Sweat said. “It can still be cultured, can still make you sick.”
But the bulk of the cases come in an initial torrent. Because E. coli can make a person sick up to 10 days after exposure, Sweat reckoned they’d find new cases until at least Oct. 17, 10 days after the fair ended.
Meanwhile, the team was looking to create a comparison group of people who attended the fair who didn’t get sick. Their call for “case controls” resulted in dozens of phone call from people in the area who volunteered to share their experiences. All of those people answered an eight-page questionnaire, similar to the one given to every patient who got sick. In all, the team has surveyed 150 people about the Cleveland County Fair. Still, that’s only a fraction of the more than 175,000 people who went to Shelby.
CSI it isn’t
Foust’s office always employs more than 20 doctors, nurses, epidemiologists, veterinarians and three people placed in the state by the Centers for Disease Control and Prevention, all people with experience tracing outbreaks.
The group has been meeting seven days a week, holding conference calls with health-department officials from at least eight counties and several in South Carolina, hunting out cases, working 10 and 12-hour days since the outbreak began.
Before calls about E. coli came in, the group was already working on finding patients who might have been exposed to steroid injections contaminated with fungus that has made people in 18 states sick. Several people in North Carolina have gotten sick and one woman has died.
“We eat a lot of lunches standing up,” Foust said with a laugh. “One day, we were working so much, I was so hungry, I just stood there and ate the cheese from my Lean Cuisine.”
She said the collective brains of the group are more powerful than any computer used by disease detectives on television.
“CSI this isn’t,” she said. “There’s no commercial break and then you come back and we have the answers. This takes time.”
State lab supervisor Shermalyn Greene explained that her lab doesn’t have a computer to compare the genes of different bacteria; instead, that’s done by the experienced eye of a lab technician.
“We take the patterns [created by the molecular analysis] and run them against a national database maintained by the CDC called PULSENET,” she said.
“When we enter patterns into PULSENET, it could show that it’s a local cluster, or it could tell us that the outbreak is part of something bigger,” like the Trader Joe’s peanut butter Salmonella outbreak a few months ago, Greene said.
In this instance, comparisons allowed her lab to figure out that two different genetic subtypes were present in the samples from people who attended the Cleveland County Fair.
In the end, the team counted 106 cases. Not all of them can be confirmed because they don’t have samples from all the patients. But based on history and symptoms, those patients will be part of the official tally.
And no one will ever know the real toll – a number that’s much higher.