FIRST CHALLENGE IN ANTHRAX CASE: NOT MISSING IT  



about Epidemiology & the department

Epidemiology academic information

Epidemiology faculty

Epidemilogy resources

sites of interest to Epidemiology professionals



Last Updated

12 Nov 2002

Source: New York Times, December 4, 2001.

THE DOCTOR'S WORLD

First Challenge in Anthrax Case: Not Missing It

By LAWRENCE K. ALTMAN, M.D.

Is anthrax really as rare in this country as experts believe? Given the resemblance of its early stages to other ailments -- skin anthrax can look like a spider bite or acne, and the inhaled form like a viral respiratory infection -- many people have wondered whether other cases may have occurred in the past, unrelated to terrorism, and been misdiagnosed.

Until the intentional spread of anthrax in recent months, only 18 cases of inhalation anthrax had been reported since 1950, and 227 of the skin form from 1955 to the beginning of the intentional spread this fall. Additional anthrax cases may have gone undetected over the years, but the number would be small, infectious disease experts said in interviews.

"No matter how I think about this, I doubt that we would miss more than the rare, rare, rare case," said Dr. David Gilbert, the president of the Infectious Diseases Society of America.

"I am sure we can come up with some extenuating circumstances where patients have slipped through the system, but it is very unlikely that we've missed lots of cases" because the illness is so striking, said Dr. Gilbert, who practices at Providence Portland Medical Center in Oregon.

No one knows the precise number. Since the death of Ottilie Lundgren, 94, on Nov. 21, federal and state epidemiologists have scoured medical records in hospitals and offices in Connecticut for clues to missed cases and turned up none.

Still, widespread publicity about the recent outbreak unquestionably led doctors to diagnose anthrax in some cases that they otherwise would not have detected because they have had so little direct experience with the infection.

One such case is that of a 7-month-old boy in Manhattan, who is thought to have contracted the infection when his mother, an ABC news producer, took him to work on Sept. 28. The child nearly died from skin anthrax in October.

The boy's physician, Dr. William Borkowsky, a pediatric infectious disease specialist at New York University, said he made a point of studying anthrax when bioterrorism became a threat in recent years. Even so, Dr. Borkowsky said he would have diagnosed the infant's lesion as a spider bite if he had not known that anthrax was present in New York City.

The reason is that when Dr. Borkowsky first examined the infant's skin lesion on Oct. 2, tests showed no evidence of anthrax. The infant also suffered kidney damage, bleeding and a type of anemia that rapidly destroyed red blood cells and required four blood transfusions.

"The syndrome this child had had never been described in anthrax" but had been linked to spider bites, Dr. Borkowsky said. "The skin lesions were compatible with anthrax, but the rest of the picture was incompatible because we reviewed all published cases of anthrax dating to the 1950's and never found the particular blood problem, microangiopathic hemolytic anemia, linked to anthrax in an adult or a child."

But as soon as Dr. Borkowsky learned about the first cutaneous anthrax case in New York City on Oct. 12, he said he called an official of the New York City Health Department and also sent an e-mail message describing the case. But he received no reply.

Frustrated, Dr. Borkowsky then called another branch of the Health Department, the Poison Control Center, which was handling calls for suspected anthrax. But he had trouble finding an official who knew about anthrax.

Eventually, Dr. Borkowsky spoke to Dr. Farzad Mostashari, an epidemiologist assigned by the Centers for Disease Control and Prevention to the New York City Health Department, who examined the child that night. Dr. Mostashari said the infant's lesion looked exactly like the one that he had seen four hours earlier in Erin O'Connor, 38, the first recognized anthrax victim in New York City.

Dr. Borkowsky had frozen some of the child's blood with the intent of testing it later for evidence of a spider bite. Instead, the doctors used the blood to confirm anthrax.

It took a month for the child's kidneys to regain normal function, Dr. Borkowsky said, and now he is fine.

In reflecting on the case, Dr. Borkowsky said that if he had not learned about the presence of anthrax in New York City, he would have diagnosed a spider bite because the baby's symptoms were "so atypical of what I was expecting for anthrax."

Milder cases of skin anthrax may have gone undetected because they can be cured by antibiotics, even if the drugs are prescribed for some other reason. Also, skin anthrax can heal on its own and even resemble bad acne.

In comparison, inhalation anthrax progresses much more swiftly and strikingly than skin anthrax. Yet there are disquieting signs that some cases may have been missed because of recent changes in medical practice and education that have made doctors less vigorous in pursuing a diagnosis at the bedside or in a microbiology laboratory. These changes began even before managed care affected the economics of medical practice and teaching, and have accelerated since.

"It is possible that a fluke case or two of inhalation anthrax over the years, not every year, could have gone undetected," said Dr. Anthony S. Fauci, the director of the National Institute of Allergy and Infectious Diseases. "But we have not missed six or seven cases a year."

"When these people get overwhelming sepsis associated with inhalational anthrax, it is tough to miss, even when you are not thinking about it," Dr. Fauci said.

Referring to a laboratory test known as the Gram stain that can help detect Bacillus anthracis, Dr. Fauci said: "It is not easy to miss these horrible looking Gram positive rods that look like boxcars. They are very striking" and can be found in blood cultures even when patients seem healthy.

Dr. Fauci cited the case of a man who walked into a hospital with just mild intestinal symptoms and no fever; blood cultures yielded enormous numbers of anthrax bacilli.

But doctors must order such tests before the microbes can be detected, and whether they do depends on how hard they and the microbiology laboratory are willing to look. The initial culture may not distinguish B. anthracis from bacteriological cousins that are common skin contaminants, and laboratories often discard such cultures without precisely identifying the bacillus.

Because immune function tends to wane with age, older people seem to be more susceptible than younger people to infection from the same microbes. Yet doctors tend not to investigate pneumonias in the very old as thoroughly as they do in younger people. So some cases of inhalation anthrax may be missed in old people.

"If it weren't for the recent number of cases, and I had a 90-year-old lady who came in with fulminant pneumonia, and she had a Gram positive rod in her sputum, I would assume it was a bacillus contaminant species and would not have pursued it," and neither would many other doctors, said Dr. Marvin Turck, the editor of The Journal of Infectious Diseases.

"We have a lot of similar cases," said Dr. Turck, who is a professor of medicine at the University of Washington. He also said: "It would be an inappropriate way of using resources if every laboratory tried to identify every bacillus species on every sputum. But having said that, I still don't know how prevalent anthrax is in the community, though I know it is not a large number."

The traditional, standard practice was to obtain sputum and other secretions from patients with suspected pneumonia and other respiratory infections so cultures could be started before antibiotics were given.

But doctors now often do not obtain cultures from patients with suspected pneumonia or other conditions before prescribing antibiotics. In selecting the antibiotics, doctors often use a shotgun approach and use drugs that will treat a broad array of microbes, including the classical ones, like pneumococci, and atypical ones, like B. anthracis, Dr. Turck said.

Dr. Joseph S. Pagano, emeritus director of the Lineberger Comprehensive Cancer Center at the University of North Carolina, said recent changes in practice might lead to undetected anthrax cases. Dr. Pagano said that earlier in his career, when he was an infectious disease expert, he would have been reprimanded "if I ever dared to put somebody on antibiotics without first obtaining a culture and making a diagnosis because that approach makes you less likely to miss something real important."

Dr. Pagano also said he "deplored the idea of giving a patient antibiotics for a fever without even having an idea about the possibilities that might be causing it."

"But that type of sharp diagnostic thinking hardly goes on anymore," in part because doctors tend to be impatient about going through the process, Dr. Pagano said.

Dr. Turck, of the University of Washington, said that even when doctors in teaching hospitals obtained appropriate cultures, the particular cause of many pneumonias was not identified. Most pneumonias are assumed to be the result of pneumococcal bacteria, Dr. Turck said, "but about half the time you do not prove any specific microbe as the cause."

One reason is that even in the best hospitals the sputum that is collected for testing often is mostly saliva from the mouth and not secretions from the lungs, making it difficult to be sure the offending microbe is present in the sample, Dr. Turck said.

Correctly detecting the cause of an infection often requires diligence, shoe leather epidemiology and some luck.

For example, an outbreak of skin diphtheria affected more than 1,100 people in the skid row population of Seattle from 1972 to 1982. The first cases were detected "only by chance because the skin lesions did not respond to the antibiotics initially given," said Dr. Turck, who was an author of a scientific paper reporting the outbreak.

For other infections, outbreaks were caused by previously unknown microbes. Two examples are Legionnaire's disease in Philadelphia in 1976 and hantavirus pulmonary infection in New Mexico in 1993. Subsequent research showed that both microbes had caused illness that was attributed to other conditions in the past.

Inhalation anthrax causes bleeding and swollen lymph nodes in an area behind and between the lungs known as the mediastinum. The damage is so distinctive that pathologists would quickly detect them in an autopsy. But with a striking decline in the number of autopsies performed, such damage could easily escape detection.

Dr. Gilbert, the infectious diseases society president, said that although "it is worrisome that we don't do as many autopsies as we used to do in people that die with unidentified illnesses, the reason we don't do autopsies is that our pre-mortem diagnostic capabilities are so, so much better."

Nevertheless, studies have shown that such diagnostic tests often fail to detect serious conditions during life.

Reflecting on the changes in practice, Dr. Pagano said, "Maybe one of the good things that could come out of this is that infectious disease physicians will sharpen their thinking, realize the need to go back to basics and devote more time to avoid missing a rare case."