She was bleeding from everywhere. There was blood in her urine, blood in her stool. She had blood-streaked sputum and bloody vomit. As we doctors huddled around her bed, blood was trickling from her nose.We were not gathered around a patient in some hushed room in an intensive care unit. Outside, I could hear children screaming, laughing, and the sounds of diesel trucks laboring down the nearby road. I was working for the medical relief group World Vision in St. Mary's Hospital Lacor, a 460-bed missionary hospital in the small farming town of Gulu in northern Uganda. Everything was different here. And I had no idea what was wrong with this woman. Doctors base their diagnosis on what's common in the community. In the United States, a doctor finding an enlarged spleen in a reasonably healthy young woman would suspect mononucleosis. In South America, a doctor might consider Chagas' disease; in Ethiopia, Ewing's tumor. Most likely, each would be right. But put an American doctor in the sub-Sahara or a Saudi doctor in Minneapolis, and suddenly each doctor is clueless. On the other hand, the post-9/11 threat of unfamiliar pathogens used as bioterrorist weapons has only highlighted this weakness. Now physicians in American emergency rooms are quietly studying up on symptoms of rare, deadly diseases such as anthrax and smallpox. Before I came to work in Uganda for a summer, I pictured myself teaching the ins and outs of cutting-edge Western medicine. As it turned out, I was the one who needed educating.
Day one: A resident arrived a little early for morning rounds. He had come to request a day off. "I'm aching all over, I've got a fever, I'm nauseated," he said. "Sounds like the flu," I said. "No," he said with a sigh. "It's just my malaria acting up again." After a month or so on the job, I got better at dealing with common complaints. I had learned that you do a rectal biopsy for belly pain. That's because the likely cause is schistosomiasis, a parasitic worm found in the drinking water; the best way to diagnose it is with a "rectal snip." Still, every day I would stumble over differences in medical practice. One day a resident diagnosed kidney failure in a 29-year-old man with two wives and four children. When I said the patient needed dialysis, the resident laughed. "There are maybe 10, maybe 12 dialysis patients in all of Uganda," he said. "That's how many people can actually afford it." He turned to the patient and patted his shoulder. "Go home and make your will," he said in English. The nurse translated. "And go make your peace with God. You are going to die." My boss was Dr. Matthew Lukwiya, the director of the hospital. He was Acholi—born and raised in the Gulu district, dirt-poor like everyone else here. Dr. Matt was a brilliant student, though, and somehow someone found enough money to send him to Kampala to study medicine. He graduated at the top of his class in 1983—the only Acholi tribesman to become a medical doctor. Not only that, he was later appointed medical director of St. Mary's Hospital Lacor, the best hospital in northern Uganda. Some said it was the best hospital in all of Uganda. He was much like all the other physicians I had worked with in the States, and like all doctor-administrators I have known, he wore an air of terminal exasperation. Day after day, he dealt with the nuns who ran the hospital, the residents who rotated through their clinical clerkships, the patients and their unremitting parade of AIDS-related diseases and other ailments, and on occasion, even expatriate doctors like me. Dr. Matt endured us all. Normally, after seeing a patient so mysteriously ill, I would have gone to find him, but he was in Kampala working on a research project. I was on my own.
"What happened?" I asked the girl sitting at the head of the bed. She couldn't have been more than 13 or 14—probably a daughter taking care of her dying mother. "She got sick on Friday night," the girl told me through the translator. Today was Sunday. I bent over to examine the multiple blue blotches on the woman's hands. "What are those?" I asked the residents. "Purpura," someone said. Purpura is a subcutaneous hemorrhage—bleeding under the skin. It is a sign that the clotting mechanism has gone awry. I checked her wrist for a pulse. Nothing. When I felt her neck, even her carotid pulse was weak. Context didn't matter here. Africa or America, the woman was in shock. If nothing was done, only death could follow. Back in the States, this woman would be in some intensive care unit with tubes and catheters, monitors and ventilators, and a whole swarm of health-care professionals. Here there was just a cot with a plastic sheet. We could do so little. "Okay," I said. "What are the common causes of shock associated with this kind of bleeding?" By this I meant "bleeding from every pore." There weren't many. "End-stage liver disease," one of the residents said. I nodded. But that kind of liver failure takes time to develop. This patient had gotten sick overnight. "What else?" Silence. "How about DIC?" I asked. Disseminated intravascular coagulation is a paradoxical disease that causes a patient to bleed and form clots simultaneously. The residents responded as they do in the States. You could see the little light go on: "DIC. Oh, yeah." Certain diseases, for reasons that aren't well understood, can set off the first few steps in the blood-clotting cascade. The reactions then spread to involve blood everywhere in the body. Platelets clump and intravascular micro-clots form. Blood begins to sludge, and no-blood-flow regions spring up all over the body. As the process continues, clotting factors get used up until the patient can no longer make clots when or where they are really needed. End result: chaos and death. "Now, what is the most common cause of DIC?" I asked. The correct answer is "bacterial sepsis," which means an infection, usually by a class of pathogen known as gram-negative bacteria. The infection overwhelms the body's defense mechanisms and trips the wire on the clotting cascade. "Snakebite," one of the residents said. This brought me up short. Bacterial sepsis was the most common cause of DIC in the United States. In Africa, snakebite could cause 90 percent of all cases for all I knew. "Snakebite," I said nodding. "Not the first thing I would think of. What else?" "Abruptio placentae," another resident said. This occurs when, late in pregnancy, the placenta pulls away from the intrauterine wall. The amniotic fluid activates the blood-clotting mechanism, resulting in the downward spiral of DIC. "How do you make the diagnosis?" I asked, and then regretted it. In the United States you could run a dozen different tests. Here all we could measure was a platelet count. Except, of course, today—Sunday—when the hospital laboratory was closed. I moved on quickly. "And how do you treat it?" Again, silence. In Africa there is no treatment for this disease. I could see each resident wondering what miracle Western medicine had dreamed up. Unfortunately, the West has no magic potions for this condition either. Some experts recommend heparin, which may prevent clotting, but it doesn't seem to be particularly effective. One resident spoke up. "You treat the underlying disease." "Correct," I said. Then it occurred to me that there was a difference between Africa and the United States. We in America didn't have a cure, but we did have a "promising new therapy." The latest experimental drug prevents the first step in the chain reaction of DIC—but so far, only in the laboratory. I knelt down next to the patient and took her feverish hand in mine. Her eyes had rolled back so that not much more than the whites showed through narrow slits. For a moment, looking at her prunelike fingers, I thought, This woman has cholera. But no, it was just that she had entered the final common pathway of all diseases—the collapse of all the smoothly integrated organ systems. I squatted there helplessly for a moment, patting her hand clumsily as her daughter looked on. "Well," I finally asked, in order to say something, "what did her malarial preps show?" In Africa, when in doubt, think malaria. We could do little more than start fluids and administer a blast of antibiotics in hopes of arresting any infection. We moved on to other patients and to other diseases we couldn't cure. The next afternoon, her bed was empty, and the rubber mat that served as a mattress had been pulled outside and was drying in the waning sun. A little spooky, though, was the woman's neighbor, there with pneumonia, who had developed a nosebleed that wouldn't stop. By the time Dr. Matt returned from Kampala the following day, she had died as well. I told him about the two cases: "Classic gram-negative sepsis." "How about something viral?" he said. "What about one of the viral hemorrhagic fevers?" "Oh," I said. "Yeah, of course." I mentally kicked myself. I had completely forgotten the little I knew about hemorrhagic fevers. These are a distinct class of viral infection that features massive bleeding, whether by DIC or other mechanisms. I'd never seen a case because the hemorrhagic fevers, such as Lassa fever and Crimean-Congo fever, are found primarily in the tropics. The only hemorrhagic fever someone in America might ever see would be hantavirus pulmonary syndrome, an illness that was identified in outbreaks in Arizona and New Mexico. "Yeah," Matt said. "Lassa fever . . . yellow fever . . ." Yellow fever, the other hemorrhagic viral fever. I had forgotten about it too. "Well," I said lamely, "there's not much more we could have done for her even if we knew she had yellow fever." Actually, there is an antiviral agent, ribavirin, for treating yellow fever. But it's expensive, and needless to say, no one at St. Mary's Hospital could afford that drug. "No, no." Matt shook his head over me. "You've got to make the diagnosis. That's the only way you know if you've got a public-health problem. You need to know what's out there." Why was I always looking in the wrong direction? Matt slapped my back, laughing. "I'll make you an tropical disease expert. Just give me another 10 years." The following Saturday, though, I had my chance to show my stuff. A patient came in with new-onset diabetes and diabetic ketoacidosis—a life-threatening imbalance in the blood that results when the body lacks enough insulin to metabolize sugar. He was a young soldier, diagnosed as malarial the Friday before and only dragged into the hospital now because his buddies had noticed the antimalarial pills didn't seem to be working. His honey-sweet smell was enough for me to make a diagnosis. The bedside glucometer (the only one, I think, in all of northern Uganda) confirmed it. At last a disease I knew. We started him on an insulin drip, gave him saline to rehydrate him, and sent off for multiple blood tests—modern medicine in action. And then he abruptly and inexplicably died. He just stopped breathing. No CPR here. We all just stood around looking at his inert body—his corpse. He was 32. Not long after that, my time was up, and I returned to the United States and to an emergency room full of sore throats and runny noses. The tropical diseases, the dying patients, the dusty town of Gulu all became so distant that when I saw the word Gulu on the front page of The New York Times, I didn't really recognize it as something I knew about firsthand. Once it registered, though, I was so startled I couldn't read the article. I just scanned it until I saw another word: Lacor. I finally picked out the point of the story: Ebola. Ebola, the classic hemorrhagic fever. An outbreak of Ebola virus had occurred; the epicenter seemed to be St. Mary's Hospital Lacor. Personnel from the World Health Organization (WHO) were camping out all over Gulu. I scanned the article to find out when the WHO personnel thought the epidemic started. I was thinking of that woman bleeding from everywhere. Had I touched her? Had I treated a woman with Ebola? The only reliable way to get messages in and out of Gulu was through the World Food Bank satellite mail system. I tried to get through but couldn't reach anyone. I ended up following the epidemic through the newspapers, helpless. WHO identified the index case, the first human case confirmed, as a woman who lived in a village about a mile from Gulu. She had died suddenly. No one knew—or even really wondered—why. As was customary in Gulu, her daughters washed her body prior to burial. Then the daughters got sick. One died immediately, but the other made it to the hospital. Apparently Dr. Matt had been suspicious enough to send some blood samples off to the reference lab in Kampala. The result came back positive for Ebola. There had never been an outbreak of Ebola in Uganda. The organisms, or vectors, that carry other hemorrhagic-fever viruses are known, but no one knows the vector for Ebola. Researchers suspect that its natural reservoir is some creature of the forest, and that outbreaks occur when the virus somehow spreads into another host, such as a monkey or a chimp. Contact with infected, sickened animals could begin the chain of infection in humans. A case appears and an epidemic ignites, blazes through a population, and eventually burns itself out. The virus then returns to its hiding place, its secrets intact. St. Mary's Hospital remained the center of the outbreak. Within two weeks there were 71 suspected cases, including 35 deaths. WHO had recognized immediately that the epidemic had international implications, and they sent out the troops. Their field personnel began gathering data about the outbreak and helped guide hospital care for victims. They instituted barrier nursing methods at the hospital and provided the necessary supplies. (When I was there we washed and reused gloves many times, not a satisfactory arrangement in an epidemic.) The disease spread through Gulu Town, then on to the Masindi and Mbarara districts and north toward Kitgum. All are meaningless names unless you've been there. Every time I read them in a newspaper, they stung like a whip. WHO officially proclaimed the epidemic over on February 28, 2001, some 150 days after it began. Agencies disagreed over how many had died, but the Centers for Disease Control and Prevention in Atlanta gives the total as exactly 425. The case fatality rate was calculated at 53 percent.
Click on the image to enlarge (38k)
The Ebola cases documented since the virus was identified in 1976 form a scattershot pattern across central and western Africa. The most recent outbreak began in January 2003 in the Republic of the Congo. By the end of March, it seemed to be over. A total of 128 cases had been reported, and only 14 people had survived. The outbreak appears to have begun after people handled and ate the meat of dead gorillas.Map by Matt Zang
In a follow-up report, WHO noted three important risk factors: having contact with victims of Ebola in one's family, attending funerals of Ebola victims, and "providing medical care to Ebola patients without using adequate personal protective measures and practices." But even with the best protective measures WHO could employ, the nurses and doctors treating Ebola were at high risk. About a third of those caring for Ebola patients in the isolation ward at St. Mary's Hospital developed Ebola themselves, despite all the barrier methods. Students, doctors, nurses all lay side by side with their patients, dying. One night a patient broke loose and escaped from the isolation ward. He ran through the hospital, raining blood and body fluids everywhere. Dr. Matt was called. He gowned up, chased the patient down, and wrestled him back to the ward. Two days later Dr. Matt developed fever and muscle pains, but he continued to work until he finally collapsed and took his place in the isolation ward. On December 5, 2000, he died of Ebola, one of the last victims of the outbreak.
Dr. Matthew Lukwiya, medical director for St. Mary's Hospital Lacor in Gulu, Uganda, began to suspect in October 2000 that some of his patients were dying of Ebola. Less than two months later, the virus claimed his life. On his deathbed he asked to be buried next to his mentor, Lucille Teasdale, the Canadian surgeon who established the hospital with her husband, Piero Corti. Teasdale died in 1996 after battling HIV, which she had contracted while operating on a patient. Photograph courtesy of Mrs. Lukwiya and World Vision Uganda.
How Ebola Works
Stage 1: People become infected by handling material from infected animals or by contact with the fluids— blood, saliva, or semen—of infected people. The virus slips into cells, hijacks the cellular machinery to make more viral RNA and protein, then packages that material into new viruses. Once inside a cell, the virus takes about eight hours to reproduce.
Stage 2: The Ebola virus targets cells lining the blood vessels and the immune cells called macrophages. Once enough cells are infected, the amount of the virus increases rapidly, outstripping immune defenses. A gene that codes for a surface protein may be involved in both interrupting immune-cell activation and killing host cells.
Stage 3: The virus spreads to the lungs, lymph nodes, kidneys, and liver. About five to 12 days after infection, the patient feels feverish, dizzy, and nauseated. Within a few days, blood circulation begins to fail. Patients often start bleeding from the gastrointestinal tract, lungs, and gums.
Stage 4: Typically, the virus kills 50 percent to 90 percent of its victims within about two weeks of infection. Ebola victims do not bleed to death, nor does the virus itself kill them. Powerful signals from infected immune cells flood the bloodstream, prompting blood vessels to dilate and release blood into tissue. The final cause of death is usually shock—organ failure caused by a collapsing circulatory system.All a physician can do for an Ebola victim is give fluids and supportive care. An Ebola vaccine, however, is in the works. In a preliminary test, vaccinated guinea pigs and monkeys did not become infected after exposure to Ebola. Some people may fend off the virus naturally. One small study found Ebola antibodies among individuals who cared for Ebola victims and never got sick.
Tracking Down a Stealthy Killer
The earliest known Ebola outbreaks occurred in central Africa in 1976; another smaller outbreak followed in 1979. Then the virus vanished, only to reappear in western Africa in 1994. Nobody knows where Ebola hid between those outbreaks. To find the Ebola reservoir, virologists look where other viruses that cause Ebola-like infections hide. For example, the virus that causes hantavirus pulmonary syndrome, which broke out in 1993 in the southwestern United States, maintains itself within rodents in the region; humans become infected through contact with rodent droppings. No rodent in Africa has definitively been shown to harbor Ebola, however. Nonhuman primates are another potential reservoir. Ebola infections have occurred among monkeys, chimpanzees, and gorillas, yet these animals are unlikely to be primary hosts because there is no evidence that they can carry Ebola infections without becoming ill. Epidemics among monkeys, for example, seem to burn out as quickly as those in humans. Bats are another suspected carrier because they can be infected with Ebola in the lab. But the Ebola virus itself has never been detected in wild bats. One way epidemiologists try to pinpoint a mysterious vector is by studying the index case, the patient whose illness seems to have precipitated the epidemic. In a 1995 outbreak in the Democratic Republic of the Congo, the index case was a charcoal maker whose daily rounds took him from his home in the savanna to a forested area, where he kept his kiln, and to a site deeper in the forest, where he had a small vegetable garden. Fieldworkers in protective gear retraced the man's usual rounds, collecting insects and mammals they found along the way. Despite screens of 3,000 vertebrates (including 500 bats) and 30,000 invertebrates, researchers have come away empty-handed. Some researchers have postulated that Ebola arises when a benign human virus mutates into a virulent form. Yet Ebola and its kin, the Marburg virus, appear to be genetically stable with a mutation rate 1/100 that of HIV and influenza. A recent study turned up a new suspect: birds. Researchers had known that the genetic sequence of Ebola resembled that of some viruses found in birds, and a 2002 study by David Sanders of Purdue University strengthened the suspicion. He identified a biochemical feature in the protein coat of the Ebola virus that is common among avian viruses. Since his study was published, Sanders has received several calls from researchers who intend to begin screening birds for Ebola. — P.G.
For a brief scientific overview of Ebola outbreaks in Africa, visit the Eurosurveillance Project Web site: www.eurosurveillance.org/em/v07n03/0703-222.asp.
The Centers for Disease Control and Prevention has information about Ebola, its symptoms, and its transmission: www.cdc.gov/ncidod/dvrd/spb/mnpages/dispages/ebola.htm.
World Vision International organizes medical relief efforts around the world. To learn about their projects or make a donation, visit www.wvi.org.
For more information about St. Mary's Hospital Lacor and how to support their work, visit www.lhospital.org.
