As anybody who had ever taken antibiotics knew, many microbial residents of the gastrointestinal tract performed beneficial functions. As they digested food for their own purposes, these bacteria assisted in the task of breaking down the fats, sugars, proteins, and unwanted chemicals that regularly flowed through the human body. In their absence, human digestion was a difficult, often painful process, as signaled by the constipation, cramps, and gas that often resulted from antibiotic treatment. Antibiotics disrupted the balance of both commensal and pathogenic microbes.
Not long after the advent of the antibiotic medical revolution, an equally radical change in veterinary and livestock practices took place. Expensive livestock lived longer when their ailments were treated with antibiotics. Prophylactic treatment seemed even wiser, as animal husbandry soon included routine antibiotic dosing of chickens, cattle, and dairy cows. The shelf life of meat, poultry, eggs, and dairy products was extended through antibiotic treatment of the animals.
On the face of it this made sense. Why risk Salmonella poisoning due to consumption of undercooked meat if it was possible to sterilize the steer’s body before slaughter? Why not extend the unrefrigerated shipping distance of eggs by injecting them, or the hens, with antibiotics?
Of course, the microbes were every bit as likely to share genes and mutate around the antibiotics while inside the gut of a cow as they were in a Homo sapiens intestinal tract. So in addition to the five billion humans on the planet that might take antibiotics, there were billions of cows, chickens, pigs, cattle, sheep, ducks, and other livestock undergoing prophylactic or treatment exposure to the chemicals, more than doubling the global selective pressure upon bacterial populations.
In the 1970s, Dr. Stuart Levy of Tufts School of Medicine in Boston showed that giving high doses of antibiotics to chickens resulted in the emergence of resistant Salmonella strains that could be found in both the meat and the eggs of the animals. 70 Only thorough high-heat cooking could safely destroy the mutant organisms before humans consumed the poultry products.71
A Dutch study in 1990 showed that the use of expensive fluoroquinolone-type antibiotics on chickens and eggs led to the emergence of strains of the enteric bacteria Campylobactr jejeuni and C. coli that were resistant to the drugs in people.72 Two years later Spanish physicians reported that half of all uncooked chicken in the country contained strains of the bacteria that were resistant to fluoroquinolones. In 1989 the Spanish group had seen virtually no evidence of such resistant strains in randomly tested human stools; by 1993 half of all samples contained the resistant C. jejeuni/coli.73 The same Spanish group had discovered in 1988 that chloramphenicol use in pigs led to the emergence of a strain of resistant Yersinia enterocolitica that made its way into humans who ate pork.74
In February 1983 an unfortunate landmark was reached when Michael Osterholm in Minnesota discovered that low-level livestock use of antibiotics led to mutant bacterial emergence and subsequent human disease.75 Having recently learned the wily ways of the microbes during their controversial investigations of Toxic Shock Syndrome, the Minnesota State Health Department’s Osterholm and the CDC’s Mitch Cohen and Scott Holmberg were already on the alert for emergent bacteria when hospitals statewide began reporting an increase in Salmonella newport food poisonings.
“This is definitely out of the norm,†Holmberg said. “Newport is a southern bacterium. You never see it up there.â€
The S. newport bacterium was almost exclusively found in animals and foods common to the near-tropical states along the Gulf of Mexico, yet dozens of people, some acutely ill, were turning up in the ice-cold climes of Minnesota. And the bacterial strain found in their bodies was resistant to penicillin, ampicillin, carbenicillin, and tetracycline.
The patients who took ill in the 1983 Minnesota outbreak were far sicker than was usual: six of them had to be hospitalized for more than a week, several had passed bloody stools, and all of them had suffered at least one of the following symptoms: high fever, diarrhea, stomach cramps, chills, and vomiting.
The investigators first searched victims’ medicine cabinets looking for contaminated or aberrant antibiotics, but that proved to be a blind alley. So they turned to the patients’ kitchens, hoping to find something unusual in the individuals’ eating habits. But what they found were absolutely typical Minnesota diets: lots of meat, potatoes, dairy products, eggs, frozen foods, and prepackaged snack foods.
After several months of frustrated investigation, Osterholm sent out an official memo to his counterparts in neighboring states, asking for clues and suggestions. South Dakota officials called immediately—they had five cases. North Dakota had one. And the common thread between the tristate cases suddenly was obvious: all the sick individuals had consumed hamburger meat shortly before taking ill.
Osterholm, Holmberg, and Cohen traced the hamburger shipments to a herd of cattle that had been grazing in the area where the South Dakota cases were clustered. From there, the 105 cattle were shipped to southern Minnesota and slaughtered on January 8, 1983. Fifty-nine beef carcasses were further processed for retail sale in a Nebraska packaging center on January 10, and shipped to beef brokers in Minnesota and Iowa. From there they were sent to supermarkets across the region. The investigators estimated that 40,000 pounds of contaminated beef, mostly in the form of hamburger, eventually reached Minneapolis, and an untold quantity reached markets in North and South Dakota and Iowa.
It turned out that the herd had been fed antibiotic-laced feed, dosed at levels well below legal standards, thought to be utterly safe. Over time, resistance developed in the Salmonella that were infecting the cattle. Most of the humans who fell violently ill were taking antibiotics for other problems, such as sore throats, when they consumed the contaminated hamburger meat. The first antibiotics had cleared their bodies of many other microbes, creating a wide-open, noncompetitive field for S. newport colonization.
In a few cases people passed their hamburger-acquired Salmonella on to others, via hospital instruments or household exposure. 76
Levy felt that the Minnesota case proved the folly of continuing to regulate human use of antibiotics in the United States and much of the industrialized world while allowing virtually unregulated sales of the drugs to the agricultural industry, to veterinary medicine, and to most of the world’s human population, since few countries required prescriptions for purchase of antibiotics.77
Despite these and a host of other examples78 of the transmission of antibiotic-resistant bacteria from meat, dairy, and poultry products to human consumers, the U.S. Food and Drug Administration, its counterparts in Europe, and the European Community (under the Maastricht Treaty) all failed to take actions that might have limited the use of antibiotics on animals. Government agencies were reluctant to take steps that might impinge upon their country’s competitive status in world agricultural markets.
One of the clearest and most troubling examples of animal/human cross-species transmission of mutant bacteria was Escherichia coli. The bacteria were ubiquitous, rod-shaped microscopic creatures found in the intestines of all humans and many other mammalian species. Most of the time, in most people, they were harmless. And there was no microscopic organism that was better understood than E. coli, as it had been the focus of the majority of the world’s molecular and cellular biology research since the 1940s. Scientists liked to work with E. coli because all the complex machinery of life was there to study, packaged inside a predictable tubular structure which, almost like clockwork, stretched itself out every 120 minutes, duplicated its DNA, divided down its middle, and—voilà —there were two E. coli. The hearty bacteria would readily perform this feat of reproduction in the laboratory, always doubling their total population every two hours.
Of course, the bacteria were capable of similar feats of reproduction inside human intestines. If unchecked by the host’s immune system, or if of a particularly virulent strain prone to prod
ucing tough toxins, the bacteria would cause diarrhea and vomiting. Typically, this occurred in small children whose immune systems weren’t yet fully developed, and the ailment was particularly dangerous in malnourished or otherwise seriously ill infants.
In 1982 something new showed up: E. coli 0157:H7. It was an apparently novel organism that was capable of causing dangerous hemorrhages of the colon, bowel, and kidneys of human beings of all ages. And it hit suddenly in several U.S. states, as if out of nowhere. 79
Ten years later the details of 0157:H7 emergence would remain obscure, but its source would not: most cases came from contaminated meat. Like most E. coli strains in the 1980s, it was moderately resistant to ampicillin and tetracycline. More important, the mutant bacteria appeared to have acquired the ability to produce Shigella-like toxins. Studies of dozens of emergent bacterial species showed that genes for antibiotic resistance and virulence often resided in the same regions of the microbes’ DNA, and could move together from one organism to another. Thus, the same selection pressures that led to the emergence of resistance—in this case, use of antibiotics on livestock—also promoted greater virulence.
Because of both agricultural and medical misuse of antibiotics, E. coli strains of all kinds were rapidly acquiring broad ranges of resistance during the 1970s and 1980s.80
Stuart Levy showed in 1989 that E. coli readily spread from pigs and cows to people living and working on a farm. And the resistance factors themselves could move from E. coli that were inhabiting a pig, for example, to bacteria that were infecting other higher animals, including humans.81
In 1991 in the apple-growing region of Massachusetts there was a small outbreak of E. coli 0157:H7 infection, producing serious illness in twenty-seven people, ten of whom required hospitalization. All the cases occurred during the fall apple harvest months. It turned out that the bacteria were in local apple cider. And the cider was made from apples plucked from trees that were fertilized with livestock manure. Presumably, then, the manure was the excreta of 0157:H7-infected animals.82
The stage was set for public health disaster.
In January 1993 more than 500 people in Washington State became seriously ill after eating hamburgers prepared in ninety-three Jack-in-the-Box fast-food restaurants. Fifty of the hamburger consumers developed the E. coli hemorrhagic syndrome, and four of them—all small children—died. The culprit was E. coli 0157:H7, which had arisen in the cattle and was in the hamburger.83
Three months later a smaller outbreak occurred in a Sizzlers restaurant in Grants Pass, Oregon. Five diners were hospitalized in that E. coli 0157:H7 incident.
Politics immediately entered the picture, as consumer and legal groups demanded that the U.S. government take steps to ensure public safety. They claimed that upward of 25 million Americans suffered food poisoning each year, 6,000 of whom were victims of E. coli 0157:H7. The Clinton administration responded by ordering increased meat inspections. But the administration took no steps to get to the source of the problem: the unregulated use of antibiotics on livestock.
II
Many bacteria were capable of using sporulation to their advantage in the face of antibiotics and other threats. Like plant seeds, they would go dormant, toughen their cell walls to a nearly impermeable state, and wait. When conditions were favorable, the bacteria would reactivate, their cell walls once again becoming permeable. Some forms of resistance involved the bacteria’s use of genes that triggered sporulation when the microbes were threatened, or created an even less vulnerable cell wall at the time of sporulation.
Under such conditions, microbes could drift about unharmed in solutions designed specifically to kill them. Disinfectants, such as chlorine- and ammonia-based cleansers, soaps, fungicides (yes, fungi could also sporulate), extremely salty or acid solutions, even high heat could all be withstood by hearty sporulation mutants. 84
By 1992 a number of organisms, including strains of cholera, E. coli, and the Legionnaires’ Disease bacteria, had developed some resistance, via such sporulation mechanisms, and other means, to chlorine. “Resistance†might have been a misnomer—“partial tolerance†comes closer, because the microbes were able to survive in doses of chlorine that usually killed their species. To ensure safe drinking water in the presence of such bugs, higher doses of chlorine were needed.
Dumping more chlorine in Lima, Peru’s water system at the height of its 1991 cholera outbreak provoked little local objection. But in the wealthy United States, where public fear of cancer far outweighed concerns about infectious diseases, thousands of municipalities were lowering their chlorine levels during the late 1980s and the 1990s. Though most evidence indicated that the real chemical carcinogens were chlorinated pollutants, such as polychlorinated biphenyls (PCBs) or dioxins, much public anger was aroused toward all uses of chlorine, including sanitation.85 Greenpeace, the Environmental Defense Fund, and other leading environmental organizations argued that chlorinated compounds accumulated in human body fat and the cancer risk rose over time with each additional chlorine exposure.86
That put government—from the municipal to the federal level—in a tight spot, forced to balance the need to limit environmental carcinogens against the threat of infectious diseases. At a time when chlorine-resistant strains were emerging, governments were being pressured to lower sanitation uses of disinfectants.
The first warning shot from the microbes came in January 1987 at West Georgia College in Carrollton, Georgia. Students in record numbers fell ill with acute gastroenteritis due to Cryptosporidium, a one-celled tiny parasite. Not much larger than most bacteria,87 Cryptosporidium caused painful intestinal infections and severe diarrhea.
Rural Carroll County, with a population of only 64,900 people, suffered 13,000 cases of cryptosporidiosis in less than a month. Every household that received water from the central municipal system was struck. The Carrollton water supply met federal standards for water purification, and researchers at the time were uncertain why the outbreak had occurred.88
Two U.S. federal agencies whose charters occasionally conflicted on matters of community exposure to chlorine—the Environmental Protection Agency and the Centers for Disease Control—had passive surveillance systems in place. Neither agency conducted active surveillance, aggressively searching for cases of contamination or the emergence of newly resistant strains of microbes. The problem was left to the states and municipalities. And the quality of local surveillance activity varied radically from state to state; some states had no ongoing system in place.
Even this admittedly weak federal data base showed, however, that there was trouble afoot. Between 1991 and 1992, thirty-four outbreaks of disease associated with drinking water were reported to the federal agencies. In 27 percent of the cases the microbial contaminant was identified; 68 percent were unsolved mysteries. Half the cases involved an identified malfunction or deficiency in local water treatment and purification procedures; but in 6 percent of the cases investigators were unable to identify how the water got contaminated.89
The key microbes involved were Giardia, Cryptosporidium, hepatitis A, and Shigella. Ten of the outbreaks occurred in communities that used proper chlorine purification. In some of the cases treatment failure occurred when microbially contaminated agricultural wastes got into the water supply. And though Giardia had, since the early 1970s, been the dominant microbial contaminant in U.S. drinking water, by 1992 cryptosporidiosis cases equaled those of giardiasis. Cryptosporidium were commonly found in cows and their excreta.
In at least three outbreaks the local water treatment facilities were, by all standards, top of the line. They used proper amounts of chlorine, kept the water moving at a rate that could make it impossible for sporulated bacteria to form protective colony clusters adhering to solid surfaces, and passed the water through efficient filtration systems.
Yet the systems failed.
“Evidence suggests that a su
bstantial proportion of non-outbreak-related diarrheal illness may be associated with consumption of water that meets all current water quality standards,†the CDC concluded. The agency was also forced to conclude that “Cryptosporidium oocysts are resistant to disinfection by chlorine.â€
The clearest evidence of chlorine failure could be seen in the sudden surge of Legionnaires’ Disease, cryptosporidiosis, and giardia among people who used chlorinated hot tubs, swimming pools, and public spas.90
In April 1993 some 400,000 residents of Milwaukee, Wisconsin, fell ill with cryptosporidiosis, and the city’s AIDS population faced a mortal threat in their drinking water, as their immune systems couldn’t control the microbe. The problem was blamed on a combination of chlorine-resistant Cryptosporidium and a decrease in filtration efficiency due to a drop in water levels that left the liquid unusually high in particulate levels. U.S. Environmental Protection Agency laboratory studies later showed that the Milwaukee strain could actually live on Clorox.
In July 1993 some 35,000 residents of New York City had to switch to boiled water when it was discovered that E. coli 0157:H7 had made its way into the water supply. The bacteria survived chlorination and a faulty filtration system. And residents of the nation’s capital and outlying Virginia suburbs were forced to boil their water for a week in December 1993 because Cryptosporidium had made its way into the Washington, D.C., water supply, due to the same set of factors. Similarly, in 1993 in Cabool, Missouri, the water supply, despite chlorination, was contaminated with E. coli 0157:H7; three elderly residents of the town died as a result.91
In a review of the U.S. water systems the Natural Resources Defense Council, a citizens’ action group, concluded that nearly one million Americans were falling ill annually due to water contamination and 900 were dying as a result. The organization named 250,000 violations of federal drinking water laws nationwide.92 Some 83 percent of the nation’s water systems—those that serviced small towns and rural areas—went virtually unmonitored by state and federal agencies, the organization charged.
The Coming Plague Page 66