Anatomy Of An Outbreak

By Jeffrey Kluger

Tammy Lowery couldn't see the blood vessels rupturing in her gut, but the way she was feeling, she didn't have to. Lowery had been sick for five days, growing steadily worse as the week wore on. First had come the stomach pains. Then the bloody diarrhea. Then the paralyzing cramps. She had laid off food for a while, figuring the problem would pass. It didn't. Finally, as July 4 approached--when Lowery should have been at the Alpine, Wyo., gift shop where she works, preparing for the crush of campers and tourists who make the Independence Day weekend such a busy one--she noticed that her son Sean, 5, had come down with the same symptoms.

That did it. Struggling to get to the car, Lowery drove Sean to the office of Dr. Donald Kirk, a physician who serves many of Alpine's 470 year-round residents. She got there just in time; shortly after she walked into Kirk's waiting room, Lowery passed out on the floor.

On Friday of the same weekend in late June that Lowery started feeling wobbly, a young motorcyclist who lives in a nearby town was passing through Alpine. Dropping in on some friends, he gulped two glasses of water drawn straight from an Alpine tap. Before too many more days had passed, he realized that he was falling seriously ill.

Just as the motorcyclist was riding out of Alpine, members of the Scott family were driving in from all over the northwest. The Scotts had been planning a reunion in Alpine for months, and on June 26, a Friday, 43 of them showed up. They spent the weekend picnicking, playing touch football and rafting on the Snake River. Like most families planning a weekend outdoors, they brought their own food along. The Scotts didn't like Wyoming water, so they brought their own water too. But the kids filled their squirt guns from a faucet and took a few sips from the barrel ends of their water pistols, and the adults began to drink from the tap when they ran out of bottled water. If there was something seriously wrong with Alpine water, they couldn't tell by the taste.

Regardless of how Alpine's water tasted, there was in fact something grievously--perhaps lethally--wrong with it. That something was a particularly dangerous strain of the E. coli bacterium called E. coli O157:H7, or O157 for short. Ordinarily a benign organism found in the intestines of human beings and animals, E. coli has a nasty ability to mutate and proliferate. Lately it has been proliferating with a vengeance. Five years ago, the fast-food industry was rocked when four children died and 500 other people fell ill after eating E. coli O157-contaminated hamburgers at Jack in the Box restaurants across the Pacific northwest. Massive as that outbreak seemed at the time, it was, for the bacterium, merely a shot across the bow.

In the half-decade since, there have been as many as 30 E. coli outbreaks each year in the U.S., resulting in thousands of very ill people and hundreds of lost lives. In the past seven months, new outbreaks have been reported in Indianapolis, Ind.; Milwaukee, Wis.; Washington; and Chicago. Americans have grown sick after eating ground beef, coleslaw, cheese and various fruits. Just last week California-based Odwalla Inc. pleaded guilty to violating federal food-safety laws and agreed to pay a $1.5 million fine for selling a batch of contaminated apple juice that sickened 70 people and killed a 16-month-old girl in 1996. With the country sweating through one of the hottest summers on record, such problems are likely to grow, as more people cook outside, eating too rare hamburger, unwashed fruit and other foods that have been mishandled or badly prepared.

Making matters worse, some people are getting E. coli infections without eating any food at all. In Spokane, Wash., last winter, seven children in a day-care center became sick; prime suspects: contaminated sand, soil and modeling clay. In June, 26 kids, including the three-year-old son of Atlanta Braves shortstop Walt Weiss, contracted the bug in a pool at a Georgia water park. Brody Weiss recovered in time to watch his dad play in the All-Star game. One of the other children in the water park that day, two-year-old McCall Akin, developed kidney complications and died last week. In all, some 20,000 people in the U.S. suffer E. coli infections each year, 250 of them fatally.

If there is any seawall against the spread of E. coli, it's the epidemiologists from the Centers for Disease Control and Prevention and other health organizations who make it their business to monitor all serious disease outbreaks, from swine flu to salmonella to HIV. It's the job of those scientists to track the pathogen back to its source, where it can be eradicated or--at the very least--studied, and to spot weaknesses in the public health system that will help them prevent future outbreaks. The medicine these doctors practice is less academic than it is forensic, relying on both high-tech hardware and old-fashioned shoe leather. When the size of the epidemic grew beyond the money and staff available to the Wyoming experts, they called the CDC for reinforcements.

Dr. Kirk knew as soon as Tammy Lowery collapsed in his waiting room that he had a problem on his hands. A cursory examination of the Lowerys told him two disturbing things: he had never seen symptoms like these before; and, unfamiliar as he was with the symptoms, he was pretty sure E. coli was to blame. Kirk dispatched Tammy to the hospital in nearby Afton and then began to work the phone.

Canvassing the other doctors who serve the area, he learned that in the past few days, six people in Alpine had come down with the same symptoms. In Jackson Hole, also close by, two other Alpine residents had been hospitalized. Only moments later, Kirk learned that lab tests on the Jackson Hole patients confirmed that the victims were carrying not only E. coli but also the deadly--if prosaically named--O157:H7 strain. With that, Kirk made one more call, to Donn Wooden, the mayor of Alpine. "We have a real potential health problem in the community," Wooden recalls Kirk telling him.

A problem indeed. Of all the bacteria that bloom in the body, E. coli is usually one of the most beneficial, helping to metabolize food in the intestine. In 1982, however, scientists discovered that E. coli wasn't always so benign. That year 26 people in Oregon were felled by a violent infection and intestinal disorder, and when doctors analyzed the bug behind the illness, they found that it was all but indistinguishable from ordinary E. coli, with but a small exception: this breed of the bacterium contained a few strands of genetic reweaving that cause it to produce a powerful toxin its less potent kin can't.

When this fortified E. coli, which researchers dubbed the O157:H7 strain, takes hold in the body, it behaves savagely. Doctors believe the bacterial toxin first destroys blood vessels in the intestines, which accounts for the bloody diarrhea that is the signature symptom of the infection. The toxin then passes into the bloodstream, where it probably damages vessels throughout the body. This produces gummy clots that clog organs like the kidneys. Up to 5% of all people with O157 infection develop a kidney condition known as hemolytic uremic syndrome; up to 5% of all HUS cases are fatal. The clotting caused by E. coli can also damage the heart, the lungs and even the central nervous system.

Certainly, O157 E. coli is not an easy bug to pick up. It's not an airborne pathogen like a flu virus, and it can have an ill effect only if it's ingested. The vast majority of people who do come down with the infection survive if they are kept hydrated and, in some cases, hospitalized. But up to 1% do die--mostly children, the elderly and people with compromised immune systems. In all cases, antibiotics are not only useless but may actually make things worse, causing the bacteria to rupture and spill their toxin even more widely throughout the gut. Says Nancy Donley, a safe-food activist whose son died of E. coli infection: "We're not talking about minor gastrointestinal distress. It is a brutally ugly death."

When Dr. Kirk got Mayor Wooden on the phone, he told him that they had a serious public health threat on their hands. Wooden didn't need to have his arm twisted. "If it's a real problem," he said, "let's deal with it."

For Kirk, dealing with the problem next meant reporting the outbreak to the county infectious-disease nurses, who in turn reported to Gayle Miller, Wyoming's chief epidemiologist. Miller and her nurses knew immediately that even six cases meant an epidemic. They began canvassing the region to locate others who had been infected, and each time they found someone sick, they began interviewing that person, looking for a common source of infection. After a week they had 26 confirmed E. coli cases in four states, and the numbers seemed likely to grow.

A contagion this big was beyond the state's capabilities, and at 3:30 p.m. on July 9, Miller sent e-mail to CDC headquarters in Atlanta, asking that a group of scientists from the centers' Epidemic Intelligence Service--the SWAT team of infectious medicine--be sent to Wyoming. For epidemiologists, a billet in the EIS is a much coveted post. Researchers compete vigorously for the few available spots, and when a position is granted, the tenure lasts just two years. After that, EIS investigators are encouraged to seek work in the same field at the local level. Miller had been an EIS member, and when she called in the CDC, she knew she would get the best. "We speak the same language," she says.

Early the next morning, a three-member EIS team arrived in Alpine and set up shop in the town's city hall, an unlovely, one-room structure used for basketball games, karate classes, the occasional play and, when absolutely necessary, running the local government. The leader of the federal team was Thomas Breuer, a 37-year-old German physician in his second year as an EIS researcher. Working with him were Sonja Olsen and Malinda Kennedy.

Breuer, a veteran of four previous infectious-disease outbreaks, appreciated the enormousness of the job ahead of him. Miller's nurses had turned up no food or other contaminant all the victims had shared. This left only two possibilities: "Water and air," Breuer says, "two things epidemiologists hate." A contamination of this kind was a hit-and-run affair; the bacterial colonies could stream into a community, do their damage and flow out of the ecosystem in a matter of days, before the epidemiologists could even get their equipment unpacked.

Once Breuer, Olsen and Kennedy got themselves set up, they found a community in a state of high alert. Mayor Wooden had already switched the Alpine water supply from the natural springs to a chlorinated well system and instructed townspeople to boil water before drinking it. Residents brought the CDC researchers ice that might be needed to keep stool samples cool during the eight-hour drive from Alpine to the state laboratory in Cheyenne. The phone company provided extra telephone lines for the duration of the crisis.

Breuer, Olsen and Kennedy quickly made use of those lines, spending the next several days phoning people who had already got sick, who might be getting sick or who had remained uninfected, looking for anything they did--or didn't--have in common. The calls turned up some promising leads. One woman worked in an Alpine day-care center and routinely drank eight glasses of tap water a day and even gave some to her infant daughter. Yet both of them were healthy. That seemed to exonerate the water supply, until the woman added one final detail. When she is home on the weekends, she told Kennedy, she drinks only from her own well. It was during the last weekend in June that many victims said they began feeling sick. By avoiding town water on those days, the woman and her daughter might have dodged an infectious bullet. If the water supply was in fact contaminated, this one clue might help pinpoint just when the bug was present.

Expanding the scope of their calls, the EIS team turned up the motorcyclist who had passed through town long enough for his two-glass dose of local water on Friday, June 26. In the days since, he had developed a confirmed O157 infection. Because E. coli can be passed by touch from one person to another before it's unknowingly ingested, it was possible that he had picked up the bug from one of his friends in Alpine. But the water-bacteria link was too promising to ignore. Breuer also contacted LaFonda Scott, the woman who had organized the family reunion in Alpine. Scott reported that she and several of her relatives had tested positive for O157. Over the next few days, Breuer interviewed 41 of the 43 Scotts who had visited Alpine and came up with some sobering numbers. None of those who had avoided the local water were ill. Of those who had drunk even a few sips, half had become infected.

By now the Alpine outbreak had spread to 13 states--or at least it seemed to have. Thirteen states could also mean 13 separate outbreaks. Earlier this year, the CDC took a step to eliminate such uncertainty, employing an innovative network of biotech machines called PulseNet. The hardware allows scientists to scan a bacterium and come up with a sort of genetic fingerprint unique to that cell line. Studying samples of the Wyoming E. coli as well as bugs from the surrounding states, the EIS researchers discovered that their profiles matched perfectly. The Alpine infection, it appeared, was indeed widespread.

The EIS team had been in Wyoming for a week, and the database they had compiled was enormous. On the evening of Thursday, July 16, they gathered in the city hall to enter all the findings into their laptop computers and see what conclusions they could draw. Their early observations and calculations all pointed to Alpine's water, but they had not yet run any final statistical analyses, and they knew not to draw conclusions before all the tallies were in. Says Breuer: "I'm always worried before we look at the numbers."

The three investigators worked those numbers late into the night, bending over their computers as a neon light buzzed overhead and their forgotten dinner--a pineapple-and-bacon pizza--grew rubbery and congealed. Over the course of hours, graphs and tables flashed on the screen and streamed from the printer in a growing pile. Finally, Breuer was ready to ask the computer his final questions.

From the data that had been compiled, he inquired, was there any statistical link between eating hamburgers and the current O157 infection? No, the computer answered. What about venison? No. Beef jerky? No. Any relationship between contact with cattle and the appearance of the disease? Again, no. At last, Kennedy tap-tapped the commands asking the computer about a link between Alpine tap water and E. coli. The researchers leaned in as the question was processed and the numbers were tallied. After a long moment, the screen began to fill with numbers.

"Wow," someone said. "What a relief," someone else echoed.

A person who drank Alpine tap water, the computer reported, was eight times as likely to become infected with E. coli O157 as someone who didn't. Someone who was in town during that weekend in June was 14 times as likely. Testing the reliability of the numbers, the computer concluded that if the same study were conducted 10,000 times, those results would appear by chance only nine times. "Which is nice," Breuer said. "Which is very nice."

How the bug got into the Alpine water supply was anybody's guess, but the EIS team had an idea. The winter storms around the town had been fierce enough in the early part of the year to topple fences erected to keep animals away from the springs. If even a single animal did wander in, any feces it left behind could have been washed into the water supply by spring rains. Bacteria in the feces would have moved through the Alpine pipes in a single foul rush and then drained away. "Once the E. coli hit town it was at once everywhere," Breuer says. And then, just as quickly, it was gone.

In the end, 61 people tested positive for the Alpine bug--19 of whom were hospitalized--and an additional 159 were suspected of being infected, making it the largest waterborne outbreak of O157 in the U.S. So far none of the Alpine victims have died; given the bacteria's low but consistent mortality rate, however, that is as much a stroke of luck as anything else.

Hoping for similar good fortune in the future is not the best way to fight the microbe. Until a drug treatment is available, the best answer is prevention, and several groups are taking action. Alfalfa seeds, which have been implicated in past E. coli outbreaks, are now being irradiated to kill bacteria, and last year the Food and Drug Administration gave the cattle industry the green light to treat meat the same way. The procedure is generally effective, but critics are uneasy, questioning whether it's ever a good idea to irradiate something consumers are intended to ingest.

More effective than cleaning food after it's been contaminated is preventing it from getting dirty in the first place. Already public health advocates are calling for stricter standards in slaughterhouses and tighter inspection procedures to keep E. coli-contaminated meat out of the food chain. While this has been a common refrain in the past after outbreaks of salmonella and other food-borne pathogens, the severity of the O157 outbreaks may at last prod the government into action. President Clinton is proposing a modest $100 million in new funding to improve food safety, but he is still meeting resistance in Congress.

Until Washington does bestir itself, consumers must look after themselves. All fresh vegetables--including packaged, so-called triple-washed vegetables--should be thoroughly cleaned. Steaks can be safely eaten relatively rare, since E. coli is typically found only on the surface of food, and cooking will generally kill it. Hamburger is another matter. Grinding beef distributes bacteria throughout the patty, so the only safe burger is a well-done burger.

For human-to-human infection, common sense is the rule. Hand washing after bathroom trips was a good idea when the fiercest E. coli was ordinary E. coli; with O157 at large, it's even more important. Parents should take similar precautions when changing diapers, and diaper-age babies should not go into swimming pools. Even in wading pools, toddlers should wear a waterproof bathing-suit liner.

Apart from these measures, there's not much even the most germ-wary person can do. Eating, drinking and human contact are not optional, and as long as O157 is out there, some people are going to get sick. As long as some people get sick, at least a handful are going to die.

A few weeks after the Alpine outbreak subsided, LaFonda Scott and one of her daughters attended a church luncheon. Still shaky after their battles with the bug, mother and daughter made their way to a nearby table, where seven-year-old Janessa spotted a pitcher of water. Eying it warily, she asked, "Mom, is that safe water?"

Her mother could only reply, "I don't know."

--Reported by Dick Thompson/Alpine

With reporting by Dick Thompson/Alpine