The Coming Plague

by Laurie Garrett

  Vernon swore that every morning he was swallowing eleven pills, comprising three different antibiotics. And he insisted that he was always home after breakfast when the public health nurse came to inject amikacin into his shoulder.

  “Man, that hurts,” he said. “Stings, man. Burns going in, and takes its time getting there.”

  Brudney, for the first time since he arrived, fully agreed with something Vernon said.

  “It’s a four-cc injection, and it’s excruciating. And you wouldn’t have to be putting up with it if you had taken your pills in the first place,” she said.

  Vernon was now living at home in the South Bronx with his mother and older siblings. He had a girlfriend and a fifteen-month-old daughter, both of whom, so far, were free of TB. Until he got well, Vernon would live on welfare and social security funds, but, he said, “I’m gettin’ a job working on a movie that’s shooting in Harlem, just as soon as I lick this TB.”

  Brudney made a few notes on Vernon’s chart, handed the patient his prescriptions, and shook her head as he exited.

  “Everything that man says is a lie. It’s amazing. Every single word,” Brudney insisted. “For months he’s been checking in and out of homeless shelters, using false names so the Health Department couldn’t find him. And why? So he could deal drugs. I don’t know, he may even be selling his TB meds on the street. Some of the patients do.”

  Brudney noted that since 1989 Vernon had missed more than 75 percent of his appointments, was hospitalized four times, and was found hiding under an alias on two occasions.

  “That’s what we’re up against.”

  Two years earlier, Brudney and Columbia College of Physicians and Surgeons colleague Dr. Jay Dobkin had warned government officials that men like Vernon were breeding drug-resistant tuberculosis. The pair studied TB treatment records for Harlem Hospital, a public facility located in the middle of one of New York’s poorest neighborhoods, which, more than a decade earlier, Lee Reichman had identified as one of the communities with the highest incidence of TB in the United States.149 By 1985 it was also a neighborhood ranked in the top ten nationally for homelessness and narcotics use.

  Brudney and Dobkin examined the records of all patients hospitalized for tuberculosis between January 1, 1988, and September 30, 1988. Eight out of ten of the patients were men twenty-five to forty-five years of age, half of them were homeless, the remainder were listed as “unsteadily housed.” More than 80 percent of the patients were unemployed, 79 percent were alcoholic, and 40 percent were HIV-positive.

  More than a quarter of the patients—26 percent—were hospitalized for tuberculosis relapses, meaning that they had failed to properly take their medications. And a startling 89 percent of the patients disappeared sometime after hospital discharge, never returning for their mandated checkups and drug prescriptions. A subgroup of the patients—women who were addicted to crack cocaine—were 97 percent noncompliant with tuberculosis medication.

  “Within 12 months of discharge, 48 of 178 (27%) patients were readmitted with confirmed active tuberculosis at least once,” Brudney and Dobkin wrote.150 “Almost all of those discharged were again lost to follow-up, with 20 percent admitted a third time as of April 1989.”

  The two physicians noted that New York City spending for tuberculosis control stood at $40 million in 1968, more than 80 percent of which was spent on outpatient services, tracking patients, and ensuring their compliance with medication orders. In addition, the federal government added $1.4 million annually to New York’s TB effort during the 1970s.

  By 1988 that federal commitment had fallen below the $200,000 mark and New York City officials had dropped their fiscal expenditures for tuberculosis control to less than $2 million a year. In addition, at a time when the patient population was largely homeless and extremely difficult to follow, nearly all resources were directed to hospitalization costs rather than outpatient services and patient compliance issues.

  Meanwhile, the CDC had been monitoring laboratory tests on tuberculosis antibiotic resistance, finding a clear correlation between the number of times an individual had been treated for TB and the levels of resistance in the patient’s tuberculosis bacterial population. For example, based on lab data amassed between 1982 and 1986 on patients with resistant TB strains, the individuals were four times more likely to have isoniazid resistance if they had been previously treated for TB, more than three times more likely to be resistant to streptomycin if previously treated, and so on for all available drugs.

  In 1986, just as tuberculosis was making its reemergence in America, the federal government pulled the plug on the CDC’s drug-resistance tracking program. That explained, in part, why the new TB epidemic blindsided the watchdog agency.

  If significant numbers of TB patients in New York City were, as Brudney and Dobkin demonstrated, failing to adhere to proper medication schedules, the CDC’s findings indicated that widespread drug resistance was a virtually guaranteed outcome.

  When multiply drug-resistant strains of tuberculosis spread from the largely impoverished homeless population of New York City to their doctors, jail guards, and fellow patients inside hospitals, panic broke out. Though the first incidents occurred as early as 1989, word of the full extent of the problem and the number of health providers and patients so afflicted didn’t get out until early 1992.151 When the statistics were released by the CDC and the New York City Department of Health, nurses, physicians, people infected with HIV, and the general population were briefly shaken out of their complacency.

  During the first quarter of 1991, it turned out, 42.5 percent of all new tuberculosis cases diagnosed in New York City were caused by mutant strains that were resistant to the primary treatment drugs, isoniazid and rifampin. Worse yet, 60 percent of the relapse cases seen during the first twelve weeks of 1991 were multiply drug-resistant. Nowhere else in the nation were M. tuberculosis resistance levels that extreme. New Jersey and Florida ranked second and third nationally with 6.3 and 5.3 percent MDR (multiply drug-resistant) TB rates, respectively. Averaged nationally, 21.5 percent of all relapse TB cases were MDR, as were 8.2 percent of new cases.

  By 1989, New York had become the nation’s epicenter of four epidemics, each of which fed upon the other: HIV/AIDS, MDR tuberculosis, heroin addiction, and crack cocaine use.

  Three dreadful hospital tuberculosis outbreaks in New York and a fourth in Miami drew sharp attention to the interconnection between MDR-TB and HIV. In each instance a patient with active drug-resistant tuberculosis was in the same clinic or ward with HIV patients, and the immunodeficient individuals were terribly susceptible to both infection and death. Death rates among the newly infected HIV-positive patients ranged from 91 to 100 percent, most dying less than sixteen weeks after infection.152

  So grim were the prospects for the newly infected HIV-positive patients that officials referred to them as individuals who posed no direct public health threat: they didn’t survive to leave the hospital. They could, however, pose a risk for those who cared for them in the hospital.153

  When scientists with the CDC, various New York-based institutions, and research centers around the United States worked their way backward to understand why and how drug-resistant tuberculosis had emerged in the United States more than forty years after the invention of curative drugs, they were forced to conclude that the nation’s public health system had failed on every front.

  Twenty-six people caught TB in three Boston homeless shelters between February 1984 and February 1985; two died. Laboratory analysis revealed that fourteen of the individuals were newly infected with a strain of TB that was resistant to isoniazid and streptomycin. Searching for the source of the outbreak, researchers found two candidates, both of whom had MDR-TB. The first was a thirty-three-year-old alcoholic who had been in and out of
TB treatment for ten years. The other was a fifty-seven-year-old diagnosed schizophrenic who had suffered two bouts of TB since 1980.154 The outbreak demonstrated both that tuberculosis readily spread inside homeless shelters and that individuals who failed therapy could become carriers of chronically active MDR-TB.

  The most important points of vulnerability in the public health system were made apparent when a thirty-two-year-old man died of multidrug-resistant tuberculosis in Davidson County, North Carolina, on April 20, 1984. The cause of death was not confirmed as TB until over three months after his funeral; it took the North Carolina State Laboratory more than five months to determine the drug-resistance characteristics of the man’s TB strain. By then the individual had been six feet under for four months, his doctors having treated him with drugs rendered useless by a strain that proved resistant to isoniazid, rifampin, ethambutol, and streptomycin.155

  The system failures proved even more embarrassing when investigators from the CDC tested the North Carolina victim’s close friends, discovering that the dead man’s next-door neighbor had suffered chronic tuberculosis since 1978, passed it on to his live-in girlfriend, her brother living in Washington, D.C., and a drinking partner. All the cases had escaped the public health safety net, though they had been seen by physicians. And all were infected with powerfully drug-resistant mutant bacteria. All male members of the cluster died of the disease—only the female survived. The man who appeared to have been the first TB case was an alcoholic, and the group spent hours drinking together in a local bar. Because the antituberculosis drugs could not be tolerated with alcohol, the individuals failed to follow medication instructions.

  And nobody from the city, county, or state public health systems took steps at any time between 1978 and 1985 to track the recalcitrant patients or force medication compliance.

  The 1990s witnessed dangerous epidemics of MDR-TB first in Miami, San Juan (Puerto Rico), and New York City, later scattered all over the nation. Retrospective analysis of the New York City outbreak showed it began in September 1989 and continued well into 1994. In every case laboratory analysis of patient sputum and tissue samples was so slow that many victims were long dead by the time physicians knew which drugs might kill the particular TB strains in the victims’ bodies. Median time for laboratory diagnosis of tuberculosis was nine weeks, and median additional lab time for determining the bacteria’s drug-resistance patterns was six weeks. In other words, half of all New York City medical laboratories took nearly four months to reach a definitive diagnosis, and many required five to six months’ lab time. New York’s lab times were considered typical for the nation as a whole.156 Though HIV-positive people were the most vulnerable victims in the epidemic, health care workers, prison guards, homeless shelter employees, fellow HIV-negative patients, and relatives were also infected as the airborne mycobacteria spread.157

  To save money in the mid-1980s federal and state politicians had slashed TB control and surveillance budgets. By the time officials realized what had hit them, TB was draining financial resources at an astonishing rate. In 1991 direct tuberculosis treatment costs in the United States topped $700 million,158 and the costly cases kept coming well into 1994. In the state of New York, in 1991 direct hospital expenditures for TB ran to more than $50 million. 159 In response to the MDR-TB epidemic the city of New York had to build a special 140-bed tuberculosis unit in the Rikers Island jail, at a total cost over three years of $115 million. The city’s public hospitals spent $4 million to construct air-flow-controlled isolation rooms for TB patients that, for the first time, guaranteed that no other hospital employees or patients would be compelled to breathe air that was contaminated by an individual with tuberculosis.

  In addition, the federal government had to increase TB spending from $17 million in 1991 to $54.9 million in 1992, much of which went to New York City.160 When all the costs of the 1989–94 MDR-TB epidemic were totaled it was clear that more than $1 billion was spent to rein in the mutant mycobacteria. Saving perhaps $200 million in budget cuts during the 1980s eventually cost America an enormous sum, not only in direct funds but also in lost productivity and, of course, human lives.

  Amazingly, even as federal concern escalated, and TB reports from all over the country demonstrated a national upward trend in tuberculosis, cities and states, other than New York, continued to slash their TB budgets. A survey of 25 large-city health departments revealed that between 1988 and 1992 sixteen of them slashed their TB budgets.161 Though TB caseloads rose during that period in twenty-three of the cities, MDR-TB appeared in virtually all urban centers, expensive hospitalization was required in nearly twice as many cases, and the length of average treatment time increased by two months, cuts were the order of the day in most municipalities.

  By 1993 the MDR-TB epidemic had made its way to the suburbs, such as New York’s Long Island and Westchester County. 162 Jails and prisons all over the country reported MDR-TB outbreaks similar to that seen in Rikers in 1990–91. And Los Angeles, Chicago, Dallas, Detroit, and Miami all reported surges in the incidence of tuberculosis generally and MDR-TB in particular. Though New York City succeeded in bringing its TB incidence down that year, it remained fifty times greater than the national average—which itself was pretty bad. The CDC determined that 14.2 percent of the nation’s tuberculosis cases in 1993 involved MDR-TB.

  Further, studies showed that any diminution in the number of reported tuberculosis cases could only be considered a brief respite so long as the underlying conditions responsible for the emergence of MDR-TB remained unchanged. For example, Dr. Fred Gordin led a seventeen-center federal study in 1991–92 for the National Institute of Allergy and Infectious Diseases, looking at 4,314 indigent individuals around the country who were infected with the human immunodeficiency virus. About a quarter of the individuals came from poor communities of New York City, notably Harlem, the South Bronx, and eastern Brooklyn.

  Skin tests of New York individuals were 28 percent positive for TB infection, compared with a national infection rate among HIV-positive poor people of less than 8 percent. Because it had long been known that HIV-positive people failed to respond to the TB skin test due to the beleaguered state of their immune systems, Gordin went a step further. He conducted anergy tests on the individuals aimed at determining whether they could give skin-test responses to anything, and then used a mathematical model to estimate what percentage of the anergic patients were infected with tuberculosis.

  The result, Gordin said, was alarming: 51 percent of the New York area individuals were TB-infected.

  “It is really very scary in New York,” Gordin said. “We have found 10.2 percent of the New York cohort have actually had TB already, which is mind-boggling. It’s what you’d expect in a Third World country.”

  Another disturbing finding came out of the National Jewish Center for Immunology and Respiratory Medicine in Denver, Colorado. Established at the turn of the century when physicians believed that fresh mountain air held curative powers for people with tuberculosis, National Jewish was, by 1990, the last fully operational TB sanitarium and research center left in the United States. Its chief TB physician, Dr. Michael Iseman, was widely considered the preeminent expert in the United States on diagnosis and treatment of the disease, and doctors all over the country typically sent their most desperately ill tuberculosis patients to National Jewish.

  It came as grim news, indeed, when Iseman announced that even in his hands, in the best TB treatment center in the entire world, MDR-TB was extremely lethal. Of 171 patients (all HIV-negative) suffering from M. tuberculosis strains that were resistant to isoniazid and rifampin—as well as other drugs, in most cases—35 percent showed no response whatsoever to treatment with remaining, theoretically effective, drugs. And among those who did initially improve under Iseman’s care, many suffered relapses. Despite radical treatments, including surgical removal of TB-fill
ed lungs, more than half the patients never recovered from the disease; either they fell into the sort of lifelong tubercular state that Edgar Allan Poe and Charles Dickens had described eloquently more than a century earlier, or they died. Most of Iseman’s patients were not HIV-positive, and the Denver physician blamed the poor efficacy of the second- and third-string antituberculosis drugs for the dismal treatment outcome. 163

  When MDR-TB struck the United States in 1991, the CDC was swamped with requests for assistance from state agencies that were searching for second- and third-string drugs. The agency identified twenty-nine regions of the United States (of fifty-nine questioned) that were experiencing extreme shortages in anti-TB drug supplies. The U.S. government scrambled to persuade multinational drug companies to rapidly increase their production capacities. 164

  A confluence of events had played key roles in the emergence of New York’s MDR-TB epidemic. First, President Ronald Reagan’s declaration of a war on drugs and call for mandatory imprisonment for a range of drugassociated crimes coincided with an enormous surge in heroin and crack cocaine use in New York. Studies showed that some 80 percent of all MDR-TB index cases in 1989–90 (not including the secondary HIV-positive cases) were injecting drug and crack users, many of whom, as a result of federal and local crackdowns, drifted in and out of the jail and prison system. In 1991 some 295,000 arrests were made in New York City, 120,000 of which resulted in some period (days to years) of incarceration. Most city inmates were incarcerated for only short periods while they awaited arraignment or trial, so the situation from a microbial point of view, between the densely crowded jail ecology and the general community, was quite fluid. On any given day, 26 percent of the female inmates and 16 percent of the males were HIV-positive, providing the microbes with an enormous pool of unusually vulnerable Homo sapiens.

  Thus, what may have begun as isolated cases of MDR-TB among handfuls of scattered recalcitrant tuberculosis patients—men and women like Vernon—was amplified inside the city’s jails into a full-scale epidemic.165