Questioning Anthrax Diagnosis

http://www.redflagsweekly.com/croweanthrax.html

QUESTIONING ANTHRAX DIAGNOSIS

David Crowe

David Crowe writes on medical and telecommunications topics. He has published numerous articles questioning medical dogmas in Canada’s “Alive” magazine. He has a degree in biology and mathematics and has peer-reviewed papers published in the areas of biosystematics and computer science.

The anthrax bioterrorism crisis began when a Florida journalist fell ill with flu-like symptoms and died within three days. Soon after, anthrax spores were detected on his computer keyboard at the American Media offices [Peltier 2001]. Bioterrorism was suspected, especially as the cases occurred in Florida, close to where some of the September 11th hijackers trained. Less than a month after the September 11th terrorist attacks that killed almost 4,000 people America was gripped by a new fear.

Those fears gathered strength when three anthrax-laden letters were mailed to media personalities and government officials from Trenton, New Jersey, and more people were diagnosed with anthrax.

As of December 5th, 2001, 59 people are believed to have been exposed to anthrax spores. Of these, eleven developed inhalational anthrax and another eleven the cutaneous form of the disease. Four of those infected have died.

There also have been several suspected cases of anthrax infection that were not confirmed.

Biological warfare has long been feared. Is it possible that those fears caused some ambiguous events to be interpreted with certainty? Will future testing of people believed to be at risk for anthrax result in false diagnoses?

And to what extent will the process of diagnosis involve short-cuts?

For example, according to New York Times medical reporter Lawrence Altman, the process of diagnosing an early case of anthrax was less than rigorous. Dr. Sherif R. Zaki, a pathologist, had claimed that a small sample of a skin lesion from a New York-based NBC employee was infected with anthrax bacteria (Bacillus anthracis). He based this on a test he had developed and not on a culture test, which is the gold standard of detection.

The director of the Centers for Disease Control (CDC), Dr. Jeffrey P. Kaplan, apparently had enough confidence in Zaki to report to New York Mayor Rudolph Giuliani that, with "a high degree of probability" the agency had detected a case of (cutaneous) anthrax. Giuliani was not satisfied with this indecision. When pressed for a yes or no, Kaplan said yes. [Altman 2001b].

DIAGNOSING ANTHRAX

The CDC continues to play a central role in the tracking of anthrax. It also provides information and educational materials to the public and health workers and performs much of the testing to confirm diagnoses of anthrax.The agency initially claimed it had a “robust” definition of a confirmed anthrax case, although this claim was later removed from its website.

Here is one CDC definition:
1. A clinically compatible case of cutaneous, inhalational, or gastrointestinal illness that is laboratory-confirmed by isolation of B. anthracis from an affected tissue or site, or
2. A clinically compatible case of cutaneous, inhalational, or gastrointestinal disease with other laboratory evidence of B. anthracis infection based on at least two supportive laboratory tests.

A slightly modified version now omits the reference to "clinically compatible" symptoms from part 2 above, although this may be an oversight [CDC 2001]. Both are based on the 1996 case definition [CDC 1996] which describes the clinical and laboratory evidence in somewhat more detail.

Here is the evidence required to meet the CDC 1996 case definition for inhalational anthrax [CDC 1996]:

Clinical
“a brief prodrome [early symptoms] resembling a viral respiratory illness, followed by development of hypoxia [lack of oxygen in the blood] and dyspnea [shortness of breath]”

X-Ray
radiographic evidence of mediastinal [space in the chest cavity between the lungs, containing the heart, trachea etc.] widening

Laboratory
Isolation of B. anthracis from a clinical specimen, or

Anthrax electrophoretic immunotransblot reaction to the protective antigen and/or lethal factor bands in one or more serum samples obtained after onset of symptoms, or

Demonstration of B. anthracis in a clinical specimen by immunofluorescence

Epidemiological Evidence

The circumstances surrounding a suspected case of anthrax play a surprisingly significant role in the diagnosis. The CDC states that the "epidemiologic profile" of cases can "guide the assessment of persons with ILI [influenza-like illnesses]" [MMWR 2001c]. Although the wording is vague, it implies that only people known to be in close proximity with anthrax spores should be seriously considered for an anthrax diagnosis. A flow-chart that provided for diagnosis lists "History of exposure, or occupational/environmental risk" as the first step [MMWR 2001b].

A recent overview of anthrax detection, diagnosis and treatment [Swartz 2001] also notes the importance of a “history of exposure” in the diagnosis of anthrax. New York Times reporter Altman noted that federal health officials are "urging doctors to look aggressively for the potentially lethal condition [anthrax] even when they see postal workers in the Washington area with mild, vague symptoms" [Altman 2001a].

In the case of two postal workers believed to have died from inhalational anthrax after being exposed to a letter addressed to a US senator, a preliminary diagnosis was made in both cases because health care workers (at two different hospitals) were "aware through media reports that two postal workers had been hospitalized in the local metropolitan area with inhalational anthrax" [Borio, 2001]. Early clinical symptoms were compatible with inhalational anthrax, but not diagnostic.

While the majority of people with compatible symptoms are not considered for an anthrax diagnosis because of lack of known exposure, perfectly healthy people believed to have been in the vicinity of anthrax spores will be considered for an anthrax diagnosis. For example, in the American Media case, although only two people had symptomatic disease, 1,075 who had been in the vicinity were tested with nasal swabs, resulting in one showing positive for anthrax [MMWR 2001a]. Later antibody tests on blood samples showed 5 positive tests, not including the person who had the only positive nasal swab [Canedy 2001a].

The detection of anthrax in a number of post offices has been considered as a possibility that perhaps more contaminated mail exists other than the three letters that have been the focus of attention. Another view holds that there has been cross-contamination of anthrax-laden mail. An alternative explanation worth considering is that anthrax is widely found in the form of spores in soil, surviving for decades after anthrax outbreaks [Dragon 2001, Swartz 2001, Altman 2001a]. It should therefore be seen as at least a theoretical possibility that some of these spores may have been blown around as dust and possibly ended up in some buildings. There is no evidence that postal and government buildings were free from anthrax spores prior to October 2001[Altman 2001a]. It is quite conceivable that many buildings around the world have long harbored anthrax spores, and that they rarely cause harm. Widespread testing of buildings well outside the area of exposure would help to determine whether anthrax spores are, in fact, quite commonly found. This does, however, have the potential to embarrass the many officials who have taken the detection of anthrax spores in buildings as concrete proof of bioterrorism.The emphasis on evidence for anthrax exposure as an important diagnostic factor could lead to a false association between exposure and disease.

Clinical Evidence

It is widely agreed that early clinical signs of anthrax, particularly inhalational anthrax, are difficult to distinguish from other diseases [Mayer 2001, Altman 2001a, Jernigan 2001]. Cough, fatigue, abdominal pain and fever are often experienced by people with influenza-like-illness (ILI, commonly known as the flu, but caused by a variety of viruses and some bacteria) [MMWR 2001c]. An investigation of the first 10 confirmed anthrax cases showed that only fever and fatigue were found in all patients [Jernigan 2001].

The CDC has attempted to clarify how the flu can be distinguished from anthrax. The agency notes, for example, that nasal congestion and rhinorrhea (runny nose) are common in ILI, but were only found in 1 out of 10 anthrax cases that they examined. This is not terribly useful, because obviously every case of flu-like symptoms without a runny nose is not anthrax, and every case with a runny nose cannot be eliminated from consideration as anthrax. The same is true of other physical symptoms.

X-Ray data is noted as a diagnostic criterion by the CDC, particularly mediastinal widening. This occurred in 7 of the 10 cases that the CDC reviewed. The agency did not note how frequently this occurs with ILI cases known not to be anthrax, but a recent examination of two surviving anthrax patients revealed that “there is nothing specific about this constellation of radiographic [X-Ray] findings that would, by itself, lead to a diagnosis of inhalational anthrax” [Mayer 2001]. [Jernigan 2001] All of the first 10 patients had at least one X-Ray abnormality, but in two cases the X-rays were first interpreted as normal, and abnormalities were only detected upon later review by a radiologist – a review that probably would not have occurred had anthrax not been suspected [Jernigan 2001].

Computed Tomography [CT] chest scans may be useful, but there are no baseline scans of previous anthrax cases to verify claims [Mayer 2001, Jernigan 2001].

In some cases of anthrax, the clinical evidence does not match those symptoms found in previous cases. “Profound, often drenching sweating" found in all 10 cases the CDC examined has not previously been identified as diagnostic [Jernigan 2001]. On the other hand, Mediastinal Widening, believed to be diagnostic on chest X-rays, was not found in all cases [Mayer 2001, Jernigan 2001]. White blood cell counts were also close to normal in some cases - not what would be expected during an intense bacterial infection [Altman 2001a, Jernigan 2001].

Active anthrax infections will obviously have clinical signs, but they do not appear to be specific enough to be diagnostic by themselves. Combining epidemiologic evidence with clinical evidence may result in the assumption that many people with flu, pneumonia or other lung conditions have anthrax, based on some evidence of a possible exposure to anthrax.

Laboratory Tests

The CDC defines laboratory tests for the confirmation of anthrax as at least two of:

(a) Evidence of B. anthracis DNA by polymerase chain reaction (PCR) from specimens collected from affected tissue or site,

(b) Demonstration of B. anthracis in a clinical specimen by immunohistochemical staining,

(c) Other laboratory tests (e.g., serology) that may become validated by laboratory confirmation [MMWR 2001a].

This is an update to the 1996 definition. In actual practice, the standards used by the CDC are reported as being somewhat different. The CDC is reported to also use a culture test and PCR [Mayer 2001] and to use a culture test and gamma phage lysis (indicating susceptibility to certain bacteria-destroying viruses) [Borio, 2001). Confirmation is also based on a culture test and either gamma phage lysis or direct fluorescent antibody staining [Swartz 2001]. Without identification of the tests being used for diagnostic purposes and without reference to scientific research that validated these tests, it is difficult to determine the frequency of false positives.

Overall, one major problem in testing for anthrax infection is the lack of baseline data. Without running the tests on thousands of healthy people, and, much more importantly, thousands of people with clinical symptoms that are similar to anthrax, the accuracy of the tests cannot be established with confidence. Tests cannot be validated by running them only on people suspected of having a disease.

A good example is the 1 out of 1,075 nasal swabs that tested positive in the Florida anthrax outbreak. If this testing methodology is 99.9% accurate (or less), at least 1 false positive in this number of tests would be expected. Since the antibody testing in this case was positive only in five cases that were negative by nasal swab, it is quite possible that some or all of these cases were false positives.

Until a wide ranging test program is initiated to obtain baseline data, it cannot be determined whether we are experiencing a rash of new anthrax cases or simply a rash of false anthrax diagnoses, based on the lack of opportunities to exercise these tests in the past.

Each type of test has its own specific limitations. PCR is highly sensitive, but this sensitivity makes it prone to false positive reactions. Immunological tests are prone to false positives due to reaction with compounds in the test other than the antibodies or antigens, due to high levels of non-specific reactivity or because of the presence of similar antibodies or antigens caused by things other than B. anthracis. Antibody tests in the American Media situation proved inconclusive (changes in antibody levels were supposed to be diagnostic, but were not) [Canedy 2001b).

Culture of the bacteria is usually held up as the gold standard. If the bacteria can be grown after inoculation into a sterile culture medium it is seen as proof of its presence. However, presence does not prove that the bacteria was the cause of the disease. It should also be shown that the amount of bacteria and its associated toxin are great enough to cause disease.

Furthermore, contamination is a possibility (particularly when multiple tests are run in parallel), and the laboratory tests necessary to identify the bacteria may be fallible. Sometimes culture is not successful even in people believed, based on other evidence, to be infected with anthrax [Swarz 2001].

Conclusions

The weaknesses in the theory that current anthrax cases are due to bioterrorism attacks are best illustrated by the last two reported deaths. One was a female hospital worker in New York; the other was a 94-year-old woman in Connecticut. It still remains unclear whether they had a greater exposure to anthrax than the average person who might encounter anthrax spores in the soil.

There is little question that three letters intentionally contaminated with anthrax spores were sent with the intent to cause widespread panic. In that, they succeeded. But whether the anthrax spores found in various buildings were due to these letters or had always been there is difficult to say.

The current concerns about anthrax are based on questionable evidence of exposure combined with non-diagnostic clinical symptoms and lab tests that have a risk of turning up false-positive results.

RECOMMENDED READING

[Altman 2001a]
Altman LK. Be alert to Anthrax clues, doctors are told. NY Times. 2001 Oct 25.

[Altman 2001b]
Altman LK. When everything changed at the C.D.C. NY Times. 2001 Nov 13.

[Canedy 2001a]
Canedy D, Yardley J. 5 More at Florida Office Test Positive for Anthrax. NY Times. 2001 Oct 14.

[Canedy 2001b]
Canedy D. Workers Being Kept on Cipro After Tests Prove Inconclusive. NY Times. 2001 Oct 27.

[CDC 1996]
Anthrax - 1996 case definition. CDC. 1996 Sep.

[CDC 2001]
CDC Update: CDC case definition of anthrax and summary of confirmed cases. CDC. 2001 Oct 20.

[Dragon 2001]
Dragon DC et al. Detection of anthrax spores in endemic regions of northern Canada. J Appl Microbiol. 2001 Sep; 91(3): 435-41.

[Jernigan 2001]
Jernigan JA et al. Bioterrorism-related inhalational Anthrax: The first 10 cases reported in the United States. EID. 2001; 7(5): 1-26.

[Mayer 2001]
Mayer TA et al. Clinical presentation of inhalational anthrax following bioterrorism exposure. JAMA. 2001 Nov 28; 286(20): 2549-53.

[MMWR 2001a]
Update: Investigation of Anthrax Associated with Intentional Exposure and Interim Public Health Guidelines. October 2001 . MMWR. 2001 Oct 19; 50(41): 889-93.

[MMWR 2001b]
Update: Investigation of bioterrorism-related Anthrax and interim guidelines for clinical evaluation of possible Anthrax. MMWR. 2001 Nov 2; 50(43): 941-8.

[MMWR 2001c]
Notice to Readers: Considerations for Distinguishing Influenza-Like Illness from Inhalational Anthrax. MMWR. 2001 Nov 9; 50(44): 984-6.

[Peltier 2001]
Second anthrax case detected in Florida. Reuters. 2001 Oct 8.

[Swartz 2001]
Swartz MN. Recognition and management of anthrax - an update. NEJM. 2001 Nov 29.

[Zielbauer 2001]
Zielbauer P. Inhalational Anthrax is Diagnosed in Connecticut Woman, 94. NY Times. 2001 Nov 21.

ALL INFORMATION, DATA, AND MATERIAL CONTAINED, PRESENTED, OR PROVIDED HERE IS FOR GENERAL INFORMATION PURPOSES ONLY AND IS NOT TO BE CONSTRUED AS REFLECTING THE KNOWLEDGE OR OPINIONS OF THE PUBLISHER, AND IS NOT TO BE CONSTRUED OR INTENDED AS PROVIDING MEDICAL OR LEGAL ADVICE. THE DECISION WHETHER OR NOT TO VACCINATE IS AN IMPORTANT AND COMPLEX ISSUE AND SHOULD BE MADE BY YOU, AND YOU ALONE, IN CONSULTATION WITH YOUR HEALTH CARE PROVIDER.