Clinical Consultation: Could your patient's flu-like illness really be inhalation anthrax?

In 2001, 11 US residents contracted inhalation anthrax as a result of the deliberate mailing of 4 letters containing anthrax spores. Five died as a result. Several of these patients presented to physicians before becoming fatally ill, but their illness was misdiagnosed as influenza.¹

To prevent similar diagnostic errors, physicians have gained greater knowledge about the clinical presentation of inhalation anthrax. Unfortunately, this might not be enough. While clinical differences between inhalation anthrax and influenza-like illnesses exist, there also is sizable overlap.² Many of the differences are not absolute, but are differences in the frequency of occurrence of a particular clinical characteristic; for example, rhinorrhea occurs in 10% to 15% of anthrax cases compared with 80% to 90% of influenza cases.^3,4

In addition, there is no diagnostic test for inhalation anthrax that is rapid, sensitive, specific, and relatively inexpensive. Pending the development of such a diagnostic test, the initial differentiation between the 2 illnesses is based on clinical assessment.

Clinical presentation

Inhalation anthrax classically presents as a biphasic illness, initially mimicking a mild viral-like illness, with fever, cough, and myalgias lasting 2 to 4 days. This process is invariably followed by a rapid progression to severe fulminant pneumonitis, dyspnea, mediastinal lymphadenopathy, pleural effusions, sepsis, and often death.

The initial presentations of influenza-like illnesses are also characterized by an acute onset of fever, cough, myalgias, and headache, with a fever lasting 3 to 4 days. In addition, rhinorrhea and sore throat are prominent features. However, progression to severe illness is less likely, except in patients at the extremes of age and those who are immunosuppressed.

Since many of the initial clinical manifestations of inhalation anthrax and influenza-like illness are similar, it is often difficult to distinguish between the 2 diseases at an early stage when intervention could significantly improve outcome. However, 2 recent articles highlight some of the differences that may be useful in adjusting the clinical suspicion for anthrax.^3,5

First, Hupert and associates³ attempted to create screening criteria for detecting inhalation anthrax. They studied patients with inhalation anthrax, influenza, and non-influenza viral respiratory illness and compared the likelihood ratios for the presenting symptoms. They found that the presence of non-headache neurologic symptoms (such as dizziness), dyspnea, and nausea or vomiting is more suggestive of inhalation anthrax than of influenza-like illnesses. In contrast, rhinorrhea or sore throat is more suggestive of a viral-like illness.

However, the likelihood ratios for the various clinical conditions were approximately 5, meaning a particular trait was 5 times more likely to be seen in one entity than in the other. This number is well below the number that typically indicates a significant difference, which is closer to 10. In addition, the reliability of these calculations is somewhat tenuous, since the study depended on a total of only 28 cases of anthrax.

A second study by Kuehnert and associates⁵ approached the same question from a slightly different perspective. These authors created a scoring system to distinguish inhalation anthrax from influenza. They discovered that patients with inhalation anthrax were more likely to have tachycardia, a high hematocrit, and low albumin and sodium levels. In addition, these patients were less likely to complain of myalgias, headache, and nasal symptoms.

Using a combination of clinical and laboratory data, a total score of 4 points or more had a sensitivity of 100% and a specificity of 96%. As exciting as these results appear, one must interpret them with caution, because the study was based on just 11 cases of anthrax.

In addition, this 11-patient group was used to derive the scoring system--the derivation set. To ensure the scoring system is effective, it must be tested on a second set of anthrax patients--the validation set. For obvious reasons, this has not occurred. In the absence of a validation set, the conclusions from this study are tentative.

Combining the results of these 2 studies, certain patterns emerge that may help differentiate the 2 diseases (Table).

Clinical implications

We are still left with the question, "Is there a way to accurately estimate the probability that a patient has anthrax?" Given the limits of our current data, the answer to this question is probably "not yet."⁶ So how does the clinician approach the patient with the viral-like illness when he or she is concerned about anthrax?

A reasonable approach is as follows: In the absence of any confirmed cases of anthrax or direct evidence of a biologic attack, clinicians must assume that the risk of a patient presenting with anthrax is near zero. They should proceed as though the patient has an influenza-like illness unless compelling evidence exists to the contrary. This implies that the first few cases of inhalation anthrax will be missed. If, however, an index case has occurred, or strong evidence exists that spores have been released in the community, then inhalation anthrax should be included in the differential diagnosis.

If patients present with fever, rhinorrhea, headache, sore throat, normal vital signs, and normal physical findings, they are less likely to have inhalation anthrax than an influenza-like illness. In this situation, a positive influenza test result is helpful. Assuming the chest radiograph is normal, the patient can be discharged with close follow-up.

In contrast, patients complaining of fever and several symptoms more suggestive of inhalation anthrax, such as vomiting, dizziness, dyspnea, abnormal breath sounds, and tachycardia, should have a more thorough evaluation, including a chest radiograph, laboratory tests, and possibly a chest CT scan to look for hilar lymphadenopathy and pleural effusions. Even if the symptoms are more consistent with an influenza-like illness, suspicion should remain high if the patient belongs to a group identified as being at high risk for potential exposure, such as postal workers were during the 2001 release.

Until more research data become available, the initial differentiation between the lethal inhalation anthrax and the influenza-like illness will remain a clinical decision. Epidemiologic data, reports from health and government authorities, and specific clinical features and laboratory results should help in the decision-making process.

References:

REFERENCE

1. Jernigan JA, Stephens DS, Ashford DA, et al. Bioterrorism-related inhalational anthrax: the first 10 cases reported in the United States.

Emerg Infect Dis.

2001; 7:933-944.
2. Fine AM, Wong JB, Fraser HSF, et al. Is it influenza or anthrax? A decision analytic approach to the treatment of patients with influenza-like illnesses.

Ann Emerg Med.

2004;43:318-328.
3. Hupert N, Bearman GM, Mushlin AI, Callahan MA. Accuracy of screening for inhalational anthrax after a bioterrorist attack.

Ann Intern Med.

2003;139:337-345.
4. Centers for Disease Control and Prevention. Notice to readers: considerations for distinguishing influenza-like illness from inhalational anthrax.

MMWR.

2001; 50:984-986.
5. Kuehnert MJ, Doyle TJ, Hill HA, et al. Clinical features that discriminate inhalational anthrax from other acute respiratory illnesses.

Clin Infect Dis.

2003;36:328-336.
6. Schultz CH. Chinese curses, anthrax, and the risk of bioterrorism.

Ann Emerg Med

. 2004;43:329-332.