Before the New York patient arrived at NIH, Palmore and her boss, Dr. David Henderson, deputy director for clinical care and associate director of epidemiology and quality improvement at the hospital, were brought in to consult on how to treat the patient and keep the germ from spreading. A mix of drugs, including colistin, was recommended. Palmore and Henderson also recommended that the patient be put into enhanced isolation.
The woman was placed in a room at the end of the ICU hall. Rooms near her were left empty. Staff and visitors had to wash their hands as they entered and exited the room and wear gowns and gloves. Equipment that touched the patient, such as an ultrasound or an X-ray machine, was disinfected; disposable pressure cuffs and stethoscopes were designated for her use only.
Within a day, the patient had improved. She was moved into a regular room, where she remained isolated, wearing gloves and a gown whenever she walked the hallway. She returned to the ICU one other time for 24 hours. But again, she recovered.
Meanwhile, ICU patients’ throats and groins were regularly swabbed (a practice known as “surveillance testing”) to ensure they hadn’t been contaminated by KPC. The results came back negative throughout that month.
By July 15, the woman was well enough to leave. (For privacy reasons, the NIH declined to share patients’ names or details about their conditions.) Palmore was relieved. Married to a lawyer and the mother of three, she returned to juggling her work and home life in D.C.
What neither Palmore nor the ICU staff knew was that the New York woman was “Patient One” in what would turn into an outbreak. The germ she carried had slipped past isolation controls in the ICU and infected a 27-year-old woman down the hall (later known as Patient Three). Because the second woman’s immune system was functioning, she didn’t get sick right away.
A healthy person isn’t at risk of getting sick from KPC, which is a mutated version of bacteria that live within our intestines. Other microbes in our gut normally render the bacteria harmless.
But patients in ICUs with impaired immune systems are highly susceptible to it. KPC can spread through a catheter, ventilator or surgical wound to the hands of a health care worker and then to another patient. A person with a healthy enough immune system can harbor KPC and show no symptoms, but he or she may be a silent carrier of the disease, like a Typhoid Mary.
The young woman, a transplant patient, became a silent carrier. Surveillance testing in July raised no alarms.
“We tried to assure ourselves that there wasn’t spread” by ordering the ICU staff to take repeated throat and groin cultures of ICU patients, Palmore says. But the clean cultures were misleading.
On Aug. 5, three weeks after Patient One had left, a 34-year-old male patient with a solid tumor in the ICU (later known as Patient Two) was found to have the microbe in his trachea. Palmore remembers her anxiety returning at the news of another KPC patient.
Immediately, she shifted into epidemiology mode. With a bug as small as bacteria, a lot of what Palmore does is guesswork. Epidemiologists study patterns, causes and effects of human disease and then try to connect the dots to discover a bacteria’s route of transmission. She created a spreadsheet on her computer. Where had Patient One and Patient Two overlapped? In the ICU? On the ward? In radiology? Which health care workers did they have in common?
Palmore grew up in the Virginia suburbs of D.C. and had been at the NIH hospital since 2001. Before that she interned at New York-Presbyterian Hospital and was a fellow at the National Institute of Allergy and Infectious Diseases, where she analyzed how infections spread in the ICU.
A 1992 Harvard graduate who studied history and literature, Palmore is adept at pulling together information, analyzing it, then explaining how and why an event occurred. She has enjoyed reading mysteries and solving puzzles since she was a teenager. And science, to which she also was drawn, is big on puzzles.
“I like the idea that you can take a bunch of data that looks like noise, and from that figure out what factors put people at risk for infection,” she says. “I like the idea of finding order in chaotic data of humans in the hospital.”
Palmore’s efforts to find a connection between Patient One and Patient Two yielded little, however. The two weren’t in the ICU or even on the same floor at the same time. She sent their KPC strains to the lab to see if they could discern any connection. But the existing technology wasn’t sensitive enough to determine if Patient Two had his own strain or got KPC from Patient One.
“We didn’t know how [the bug] could have hung around quietly for so long either in the patient or the environment,” Palmore says. “We thought it was either a transmission or a second introduction” unconnected to Patient One.
Palmore ordered the ICU staff to culture every patient and to test hall handrails, rooms, sinks and carts. Again, everything was negative. She reminded everyone about the importance of hand hygiene and infection control protocols.
Then, on Aug. 15, the woman who had been the silent carrier of the germ—Patient Three—tested positive in a surveillance test.
On Aug. 23, a 29-year-old male with lymphoma became Patient Four. Six days later, surveillance testing picked up KPC in a 54-year-old male with a tumor (now Patient Five). All had been in the ICU.
Eventually, Patients Two, Three, Four and Five died, though the deaths of Patients Four and Five were tied to their original illnesses, rather than to KPC.
“Every new case brought on a wave of unhappiness and stress for the whole hospital,” Palmore says. “We saw patients sick and dying from a hospital-acquired infection.”
Even as Palmore was getting news of the expanding outbreak, Segre was discussing an intriguing possibility with Evan Snitkin in mid-August. A soft-spoken resident of Silver Spring, Snitkin worked with Segre as a postdoctoral fellow in bioinformatics from Boston University.
Snitkin had recently finished a project with Segre and the Clinical Center’s microbiology staff, looking at antibiotic-resistant bacteria found in soldiers serving in Iraq and Afghanistan. His friends at the lab had told him about the hospital’s disturbing outbreak.
“There’ve been multiple patients with KPC at the hospital—and I think we should see if we can sequence it,” Snitkin recalls telling Segre.