David H. Walker is Professor and Chairman of the Department of Pathology at the University of Texas Medical Branch at Galveston. The members of the Task Force on Consensus Approach for Ehrlichiosis are Johan Bakken, Philippe Brouqui, James Childs, Louis Cullman, Jacqueline Dawson, J. Stephen Dumler, Volker Fingerle, Jesse Goodman, Patricia Hogan, Wayne Hogrefe, Jorge Liz, Louis A. Magnarelli, Jane Markley, Robert Massung, Edwin J. Masters, Cheryl Murphy, Juan Olano, Christopher Paddock, Miroslav Petrovec, Yasuko Rikihisa, Steven Standaert, Gary P. Wormser, Xue-Jie Yu, and Tatjana Avsic-Zupanc.

Article Links:

• Ehrlichiosis Agent Blocks Apoptosis, Preserving Host and Habitat
• Discovery of Second Human Granulocytotropic Ehrlichiosis

Links to Other ASM Pages:

• Table of Contents
• ASM News Issues

Despite shortcomings, immunofluorescence testing remains the best choice, with PCR and culture methods being valuable adjuncts

David H. Walker and the Task Force on Consensus Approach for Ehrlichiosis

Since the report of the first human case of ehrlichiosis in 1987, three infectious diseases caused by Ehrlichia species have been identified in the United States. Cultivation of Ehrlichia chaffeensis, the etiologic agent of human monocytotropic ehrlichiosis (HME) was reported in 1991. The agent of a human granulocytotropic ehrlichiosis (HGE), namely E. phagocytophila, was identified by molecular and morphologic methods as virtually identical to E. phagocytophila, and the disease itself was described in 1994, with cultivation of the agent two years later. Human infections with E. ewingii identified by PCR were reported in 1999.

HME frequently and HGE less often can cause severe, and sometimes fatal, illnesses. For instance, HME can cause aseptic meningitis, adult respiratory distress syndrome, multisystem disease resembling Rocky Mountain spotted fever or toxic shock syndrome, and an overwhelming fatal infection in patients with acquired immunodeficiency syndrome or other immunocompromising conditions. Meanwhile, HGE also can be fatal, particularly when it is complicated by opportunistic fungal and viral infections. Diagnosing any of these three diseases on the basis of clinical and epidemiologic findings is virtually impossible, making laboratory testing useful if not essential, particularly when therapeutic decisions are being made. Such information also would be very useful for analyzing the epidemiologic impact of these emerging infections, honing the diagnostic skills of individual physicians, and determining whether such infections stimulate protective immunity.

With the goal of avoiding the errors that beset the development of diagnostic laboratory criteria for another emerging infectious zoonosis, Lyme borreliosis, three workshops were held in 1999, the first in February, the second in June, and the third in December, to develop criteria and a consensus approach to diagnostic testing for HME and HGE. Although workshop participants concluded that a clinically based diagnosis of HME or HGE requires serologic confirmation, they also noted that measuring antibodies to E. chaffeensis or the agent of HGE, respectively, has not been standardized.

Nonetheless, they recommend use of an immunofluorescence assay (IFA) for detecting HME and HGE antibodies, with serologic testing for both IgM and IgG antibodies only for HGE. E. chaffeensis should be used as the antigen for HME serology and a human isolate of the agent of HGE, E. equi or E. phagocytophila as antigens for HGE serology. Patients with E. ewingii infection have developed antibodies crossreactive with E. chaffeensis. The sensitivity of this crossreaction for detecting antibodies stimulated by E. ewingii is not known, but their existence warns of the difficulty in distinguishing this infection and HME serologically.

When To Consider the Clinical Diagnosis of Ehrlichiosis

Figure 1

HGE, HME, or E. ewingii infection should be considered when a patient presents with an acute febrile illness in a geographic area and during the seasons when appropriate vector ticks are active. More than 90% of U. S. ehrlichiosis patients acquire their infections between mid-April and late October. The majority of cases of HGE occur in the northeast and upper midwestern United States, whereas HME occurs predominantly in the south-central and southeastern quadrant of the United States (Fig. 1). Although learning whether an individual traveled to known areas of disease endemicity may be useful in a presumptive diagnosis, these areas cannot be strictly delineated and are likely in flux.

In general, physicians should consider an otherwise unexplained acute febrile illness in a patient exposed to ticks in a region where Ixodes scapularis, I. pacificus, or Amblyomma americanum ticks are endemic within the preceding 14 days as a possible case of HGE or HME. Although other symptoms such as severe myalgias, headache, rigors, and gastrointestinal symptoms are common, upper respiratory symptoms are not. In HME, but not HGE, meningitis is also common. Questions regarding tick bite or exposure to tick-infested areas within two weeks preceding the onset of illness are particularly relevant.

Figure 2

If a patient's illness appears mild (e.g., ambulatory, taking fluids orally, and temperature under 38șC) and the patient is otherwise healthy (e.g., no known immune deficiency, including HIV infection, malignancy, or extreme age), delaying laboratory tests for 24-48 hours is reasonable. However, if the illness is severe (e.g., temperature over 38șC, unable to eat, drink, or ambulate) or the patient fails to improve during this period, a series of laboratory tests is recommended, including a complete blood count with differential and platelet counts, aspartate aminotransferase or alanine aminotransferase, and a Wright- or Giemsa-stained blood smear examination for clusters of intraleukocytic bacteria (morulae). If blood smears show characteristic morulae in granulocytes or monocytes (Fig. 2), therapy should be instituted immediately. Morulae may be sparsely distributed and extremely difficult to detect, even by experienced observers, and thus a negative blood smear cannot rule out ehrlichiosis. False-positive interpretations may also occur due to toxic granulations, Dohle bodies, or superimposed platelets or contaminant particles which may be mistaken for organisms. If a patient has leukopenia, thrombocytopenia, or elevated transaminases without another known cause, therapy should be instituted immediately.

Missing the diagnosis and failing to implement appropriate and timely therapy can be catastrophic. Thus, empiric treatment is usually appropriate when the disease is suspected. If clinical suspicion is high enough to warrant empiric therapy, appropriate clinical samples should be obtained for confirmatory laboratory testing, including serum or plasma for anti-ehrlichial antibody evaluation and EDTA-anticoagulated whole blood (and cerebrospinal fluid when appropriate) for PCR analysis. Although 60% of samples taken from patients when they first visit their physicians prove serologically nondiagnostic, more than 80% of such patients will eventually develop diagnostic levels of anti-ehrlichial antibodies. If available, PCR testing or culture for the agent of HGE and E. chaffeensis can assist in the diagnosis.

Patients, including children, who are not severely ill may be observed for 24 hours and, if unimproved, treated with doxycycline if HGE or HME is still suspected. For patients who are allergic to tetracyclines or are pregnant, an infectious diseases expert or officials at the Centers for Disease Control and Prevention should be consulted regarding alternative therapies. Doxycycline-treated patients should be reevaluated within 24-48 hours. Any failure to improve within this interval strongly suggests a diagnosis other than HGE or HME.

Laboratory Diagnosis of Ehrlichiosis

Although diagnosis of HME and HGE requires laboratory confirmation, detecting and measuring antibodies to E. chaffeensis and the HGE agent has not been standardized. Nonetheless, several approaches to such testing are available, with immunofluorescence assay (IFA) now providing the best overall option.

Typically, IFA screening of patient serum samples for HME is done at a dilution of 1:64 or whatever dilution a particular laboratory has optimized for its test system. The criteria for this assay remains somewhat subjective, dependent on reagents (e.g., anti-human immunoglobulin conjugate), the state and quality of the ultraviolet microscope, and observer experience. The criteria for the serum titers are based upon use of a conjugate that represents either total immunoglobulin or IgG antibodies. The assay has been evaluated using the Arkansas strain of E. chaffeensis that was cultivated in DH82 cells.

Based on use of a standard control serum with an established endpoint titer, satisfying any one of the following criteria represents a confirmed serologic diagnosis of HME in a patient with a clinically compatible illness: (i) seroconversion, (ii) fourfold or greater rise in titer, or (iii) single serum titer of >256. A patient with a single titer of 64 would be considered as a probable HME case.

IFA-based testing for HGE should target both IgM and IgG antibodies using HGE agents, such as human isolates or E. equi or E. phagocytophila strains, as a source of antigens. Cutoffs for positive serologic reactions should be determined in each laboratory based upon well-characterized sera from patients with proven HGE and an uninfected control population. A ``calibration'' panel of serum samples representing high, mid-range, low-range, and negative titers should be developed and made available to other labs.

Cross-Reactivity of Ehrlichia chaffeensis and E. ewingii

Because serum samples from between 3% and 30% of patients with HGE cross-reacts with E. chaffeensis, it makes sense to test simultaneously for antibodies to E. chaffeensis and the HGE agent, particularly in regions where vectors overlap and epidemiologic information indicates such infections are likely. States where both agents are found include Maryland, New Jersey, Virginia, New York, Connecticut, Florida, North Carolina, and California.

To be categorized as HME or HGE, the serum titer would need to be reactive at a fourfold or higher titer against E. chaffeensis or the agent of HGE, respectively. Those patients whose IFA serologic results fall within a two-fold difference are classified as having human ehrlichiosis of indeterminate etiology.

Protein immunoblotting can help to classify indeterminate cases. For instance, if a particular serum sample contains antibodies to p120 of E. chaffeensis, the diagnosis is likely HME if the serum also is reactive to p28 antigens of E. chaffeensis. Similarly reactivity with the MSP (42-49 kDa protein family) of the agent of HGE indicates that HGE is the likely diagnosis. However, lack of reproducibility among different antigen preparations remains a problem. Potential explanations include antigenic differences among isolates and, most importantly, technical variations in assay performance. Most data have been generated using polyvalent conjugates. The utility of IgM testing for HME remains in doubt until additional evaluation is completed.

Prospects for Ehrlichial Serology

Insufficient data are available to support the use of alternative testing procedures, such as ELISA, immunoblots, or recombinant protein serologies as the primary or screening assays for antibodies to the HGE agent or E. chaffeensis.

Recent advances in the development of recombinant p120 and the p28 family serologic tests suggest that there is a great promise in this approach for the diagnosis of HME. However, before these antigens can be used reliably in test kits, they will need to be critically evaluated for specificity and comparative sensitivity.

Molecular Diagnosis of Ehrlichiosis

With so many nucleic acid-based methods for identifying bacteria, each laboratory should establish its own assays for ehrlichia and validate them with appropriate samples. Although supplies are limited, test samples are available for validating PCR or similar assays. However, a standardized sample collection is needed.

When setting up such assays, EDTA-anticoagulated blood specimens are the most appropriate samples to use in PCR. Serum is not recommended because sensitivity is low even when testing samples obtained from acutely ill patients. Blood specimens should be obtained during the acute phase of illness before anti-ehrlichial therapy begins. PCR testing of samples from healthy individuals or those with chronic illness is probably inappropriate. For PCR testing, samples are best maintained at 4oC, not frozen, and should be processed within 48 hours after collection. All controls should be subjected to the same manipulations as samples, including extraction, PCR amplification, and detection.

Primer design and selection are important. Various assay procedures have been published and the amount of validation varies among these reports. Some are well tested in various laboratories and are known to function well as diagnostic tools, including some assays of 16S rRNA gene targets. Other targets may function equally well or better, but require additional validation. Reduction of the potential for amplicon contamination is critical. The group advocates that standard methods for containment including use of different rooms or facilities for critical steps of the assay.

Several PCR-based molecular assays are available, including those based upon the 16S rRNA, VLPT, groEL, 120-kDa, and nadA genes for E. chaffeensis, and 16S rRNA genes, the msp gene, groEL, and epank1 targets for the agent of HGE-E. phagocytophila group. Others will be developed, and some of them will likely be evaluated as targets for diagnostic PCR testing.

Interpretation of PCR Results, Detection of Morulae in Smears, and Isolation of Ehrlichiae

Culture of ehrlichia is a useful adjunct to molecular testing. Because false-positive PCR results are possible and recognition of the presence of morulae can readily be misinterpreted, in the absence of serologic evidence for ehrlichiosis or of cell culture confirmation, a confirmed case entails both a positive PCR result and detecting morulae in a blood smear. This double-positive result is particularly important for HGE in which ehrlichial morulae are relatively frequently identified. Moreover, an immunologic or genetic method should be used to identify ehrlichiae isolates as to species. For now, IFA should be used as the principal laboratory diagnostic method.