Recent phylogenetic studies have prompted an extensive taxonomic reorganization of bacteria from the Ehrlichieae tribe, a group of parasitic microbes that invade and occupy specific compartments in host cells, called vacuoles, that are responsible for nutrient uptake and the release of cellular waste. The updated classification scheme, implemented over the last several years, now places these bacteria in the family Anaplasmataceae, which currently consists of six genera, including Anaplasma and Ehrlichia. These two genera comprise all tick-transmitted pathogens that infect cellular elements of peripheral blood, such as erythrocytes (red blood cells), leukocytes (white blood cells) and platelets.

The term “ehrlichiosis” is still sometimes used generally to describe infection by any organism from the former the Ehrlichieae tribe, and at other times to refer specifically to infection by species within the newly reorganized Ehrlichia genus. This inconsistency can cause confusion; for the purpose of this review, “ehrlichiosis” and “anaplasmosis” will be used separately to describe infection with species belonging to the Ehrlichia and Anaplasma genera, respectively. Although the clinical manifestations of the two diseases are similar, there are notable zoonotic and epidemiological differences between them.



Bacteria from the genus Ehrlichia have long been recognized as veterinary pathogens, but the first human case of ehrlichiosis was not identified until 1986. Since that year, the number of case reports has grown fairly steadily and currently stands at around 500 per year. Although ehrlichiosis is a nationally reportable disease, reporting is passive, and the true incidence of the Ehrlichia infection is thus assumed to be significantly higher. This suspicion is bolstered by the high rates of background seroprevalence (~12-15%) in endemic areas, a finding that also indicates that many infections are mild and self-limiting or asymptomatic.

Ehrlichia are small, gram-negative bacteria, round or ellipsoidal in shape. They preferentially invade mononuclear phagocytes, such as monocytes and macrophages, and in some cases neutrophils. In all of these cell types they occupy cytoplasmic vacuoles, usually in bacterial microcolonies known as morulae. Ehrlichia cycle in nature between ticks and mammals, and can cause disease in many mammalian species.


E. chaffeensis
morulae in the cytoplasm of
a human monocytePhotograph courtesy

of the Centers for Disease Control.

Image showing E. ewingii morulae in the

cytoplasm of a neutrophil.  Photograph

courtesy of the Centers for Disease


The two known primary agents of human ehrlichiosis are E. chaffeensis and E. ewingii. (A third species, E. canis, has recently been found to infect humans, but its significance as a human pathogen is not well understood at this point.) E. chaffeensis targets monocytes and is therefore referred to as the agent of “human monocytic ehrlichiosis” (HME). In contrast, E. ewingii preferentially invades neutrophil granulocytes. In this regard, it resembles the Anaplasma pathogen A. phagocytophilum (see section on Anaplasmosis, below), the agent of human granulocytic anaplasmosis (HGA), although it is genetically and serologically much closer to E. chaffeensis. To avoid confusion with HGA, most researchers prefer to call this disease entity “human ewingii ehrlichiosis.”

E. chaffeensis is known to be transmitted by the Lone star tick, Amblyomma americanum, and white tail deer appear to be its most important natural mammalian reservoir. In the United States, cases of the disease generally track the known distribution of the Lone star tick, occurring throughout the south central, southeastern and mid-Atlantic states, although there have also been scattered case reports in states with no known population of these ticks, such as Montana and Utah. E. ewingii is also thought to be transmitted by Lone star ticks, but less is known about its natural history and enzootic cycle.

Patients are most likely to be infected with Ehrlichia in spring and summer months, though cases occur into autumn as well. Unlike Lyme disease and Rocky Mountain spotted fever, ehrlichiosis strikes older people preferentially, probably due to immunological host factors. However, severe and even fatal cases have also been reported in children and young adults.

Signs and Symptoms

Although E. chaffeensis and E. ewingii invade different host cells, they seem to produce a similar clinical course in humans. Most patients develop symptoms 1-2 weeks after the tick bite, and over 70% will have fever, chills, severe headache and myalgias. Less common symptoms include nausea and vomiting, as well as confusion. A maculopapular rash (easily distinguishable from Rocky Mountain spotted fever) can also occur. As with many other tick-borne diseases, the symptoms are largely non-specific, thus confounding diagnosis.

Although most cases of HME are uncomplicated, it is a potentially serious illness. Hospitalization rates in symptomatic patients are estimated to be 40-50%, and fatalities run in the 2-3% range. At greatest risk are patients with underlying immunosuppression, such as organ transplant recipients or HIV or cancer patients. These patients are also at increased risk for complications in human ewingii ehrlichiosis, but no fatalities have been reported for this infection.

The main complications of ehrlichiosis are prolonged fever, a toxic or septic shock-like syndrome, coagulopathy, adult respiratory distress syndrome, and central nervous system manifestations such as meningoencephalitis, seizures and coma. Peripheral neuropathies, primarily cranial neuritis, are rarer, but have also been reported.


The potential severity of ehrlichial infections makes early diagnosis critical. Common findings on conventional blood tests include leukopenia, thrombocytopenia and elevated serum transaminases, and this triad, which is also found in Rocky Mountain spotted fever, should prompt physicians to seriously consider empiric antibiotic treatment, especially if the patient is from an endemic area and has had recent tick exposure.

From the standpoint of timeliness, the most useful diagnostic test for ehrlichiosis is probably polymerase chain reaction (PCR). Sensitivity has been reported to range between 60-85% for E. chaffeensis; the sensitivity for E. ewingii infections is not known, but PCR is the only definitive diagnostic test for E. ewingii, which has so far never been cultured in vitro. PCR sensitivity is negatively affected by prior antibiotic treatment, so blood samples for PCR testing should be drawn before treatment has been initiated.

Examination of Wright-stained blood smears for the classic ehrlichial morulae colonies is clearly diagnostic for E. chaffeensis if positive in monocytes. This test can be performed rapidly, but is of limited real-world utility due to its lack of sensitivity, which rarely exceeds 25%. As with PCR, prior antibiotic therapy reduces sensitivity.

Culture of E. chaffeensis is possible from either blood or cerebrospinal fluid, but usually takes at least two weeks. Thus, this method is useful only for retrospective confirmation of the diagnosis. Similarly, changes in antibody titers detected by indirect immunofluorescent assay (IFA) during the convalescent phase can buttress the diagnosis, but this testing method is not useful during acute illness, when treatment decisions need to be formulated. Physicians should also be aware that IgG antibodies can remain high for years after the infection, and false positive results have been associated with many other conditions, including several tick-borne diseases (Lyme disease, Rocky Mountain spotted fever, and Q fever).


No treatment studies have been performed for ehrlichiosis, but empiric evidence indicates that tetracyclines are highly effective against both E. chaffeensis and E. ewingii. The most commonly employed regimen is oral doxycycline at a dose of 100 mg every 12 hours, for 5-14 days. (Doxycycline is also recommended for pediatric patients.) In severe cases, intravenous therapy is used or antibiotic treatment is extended. Consensus exists that in all cases, treatment should be continued in all patients for at least 3-5 days after the fever subsides.

In cases where doxycycline is contraindicated, such as pregnancy or allergy, rifampin is usually the alternative choice. Little data exists to support the use of any other antibiotic, as cephalosporins, macrolides, beta lactams and aminoglycosides are all inactive against Ehrlichia organisms in vitro.



The first case of human anaplasmosis was described in 1990, when a Wisconsin patient developed a severe febrile illness following a tick bite and died two weeks later. Blood smears revealed clusters of bacteria within the patient’s neutrophils, similar to the morulae seen in monocytes with E. chaffeensis infection. However, cultures and serologic tests for E. chaffeensis were negative. Nevertheless, the patient’s clinical course suggested ehrlichiosis of some kind, and when several additional cases of the disease were reported in the northern Midwest in ensuing years, it was posited that a new species of Ehrlichia might be emerging. The new disease was tentatively given the name “human granulocytic ehrlichiosis,” or HGE.


Ultrastructure of A.phagocytophilum by transmission

electron microscopy. Photo by V.Popov, reprinted

from Dumler JS et al. Human granulocytic

anaplasmosis and Anaplasma phagocytophilum.

Emerg Infect Dis;11:1828-34.

In 1994, DNA sequencing studies revealed that the HGE agent was clearly distinct from E. chaffeensis but essentially identical to two previously known ehrlichial veterinary pathogens, E. equi and E. (Cytoecetes) phagocytophila. Under a new taxonomic scheme since implemented (see Introduction), these three organisms have been united as a single species within a new genus, Anaplasma. The new species is referred to as Anaplasma phagocytophilum, and the disease it causes is now known as human granulocytic anaplasmosis, or HGA.

Like Ehrlichia species, Anaplasma organisms are small, gram-negative and intracellular. A. phagocytophilum targets neutrophils, alters their function in the host, and forms morulae within vacuoles. Most of the damage it causes appears to be related to host inflammatory processes, as there is little evidence of a correlation between the number of organisms and host disease severity.

Anaplasmosis is a global infection, occurring in North America, most of Europe and eastern Asia. Ticks from the Ixodes persulcatus-complex are the vectors: I. scapularis in the northeastern and upper Midwestern regions of the United States; I. pacificus in the Pacific Northwest; I. ricinus in Europe and I. persulcatus in Asia. A. phagocytophilum is maintained in nature by cycling between these ticks and various small mammals, primarily mice and other small rodents. Because Ixodes ticks are also the vectors for Lyme disease, babesiosis and tick-borne encephalitis, Anaplasma coinfection with these other diseases can and does occur in humans.

In the last decade, cases of HGA have outnumbered those of HME in the United States. Similar to HME and human ewingii ehrlichiosis, the median age of patients with HGA is around 50 years old. Over 4000 total cases have been reported in the CDC’s Morbidity and Mortality Weekly since the disease became nationally reportable; as with most tick-borne diseases, the true incidence is suspected to be considerably higher.

Signs and Symptoms

The clinical course of HGA is very wide, ranging from asymptomatic infection to fatal disease. When initial symptoms appear, usually 5-10 days after tick bite, they are largely non-specific and similar to those of HME: fever, chills, severe headache and myalgias. Nausea, cough and arthralgias also occur. Rash is uncommon but has been reported.

Compared with HME, HGA appears less likely to involve the central nervous system, but peripheral neuropathies are more common and can last weeks to months. Among the neurologic findings reported in the medical literature are facial palsy, demyelinating polyneuropathy and brachial plexopathy. Respiratory distress syndrome and a septic or toxic shock-like syndrome have been reported, but appear to be less common than in HME. The overall fatality rate from HGA also seems to be slightly lower than that of HME, with most of the deaths resulting from opportunistic infections (for example, herpes simplex esophagitis, Candida pneumonitis, and pulmonary aspergillosis) in immunocompromised patients.


Standard blood tests in HGA usually reveal findings similar to those seen in HME: leukopenia, thrombocytopenia and liver function abnormalities (elevated transaminases). However, the hematological abnormalities frequently resolve by the second week of symptoms, so their absence should be interpreted in that context if patients are presenting later in the course of their illness. In general, empiric antibiotic treatment should be considered for patients in endemic areas who present with an acute febrile illness suggestive of HGA.

For specific diagnosis, Wright or Giemsa-stained blood smears have a slightly higher yield than with HME, but are still not optimal for general clinical utility, given that there appears to be a wide variation (25-75%) in the sensitivity of these tests in visualizing morulae in host neutrophils. More helpful, but not always available, are polymerase chain reaction (PCR) tests, which are estimated to have a sensitivity of 67-90%. Prior antibiotic therapy dramatically reduces the sensitivity of both  of these diagnostic methods.

Serologic testing is useful to confirm the diagnosis of anaplasmosis. The most commonly used method is indirect immunofluorescence (IFA) of IgM and IgG anti-A. phagocytophilum antibodies. Seroconversion is perhaps the most sensitive laboratory evidence of A. phagocytophilum infection, but is not always obtained in a timely enough manner to provide useful input on clinical (i.e., treatment) decisions.


The optimal dose and duration of antibiotic treatment for anaplasmosis has not been definitively established, but it is clear that A. phagocytophilum is highly sensitive to tetracyclines. Thus, oral doxycycline is the recommended treatment, at the same dose used for Ehrlichia infections: 200 mg/day in two divided doses. The usual treatment duration is 5-10 days, which is extended if there is suspected coinfection with B. burgdorferi, the agent of Lyme disease. In any case, treatment should continue for at least three days after the patient’s fever resolves. Response to treatment is usually rapid; if the patient remains febrile more than two or three days after initiation of doxycycline therapy, the diagnosis should be revisited.

As with Ehrlichia infections, rifampin is used in cases where doxycycline is contraindicated, such as pregnancy or allergy. Rifampin has also been used successfully in pediatric cases, and thus is sometimes employed in mild cases of pediatric A. phagocytophilum infection. If coinfection with B. burgdorferi is suspected in a pediatric case, doxycycline is sometimes used as an initial treatment for 3-5 days, with another antibiotic employed thereafter to complete the somewhat longer recommended treatment period for early Lyme disease.

| Top |