Emerging and Re-emerging Tick-Transmitted Rickettsial and Ehrlichial Infections

Emerging andRe - emergingTick-TransmittedRickettsial andEhrlichial Infections

David H.Walker, MDa,*, Christopher D. Paddock, MDb,J. Stephen Dumler, MDc

KEYWORDS

� Rocky Mountain spotted fever� Human monocytotropic ehrlichiosis � Rickettsia parkeri� Ehrlichia ewingii � Human granulocytotropic anaplasmosis� Typhus

Recently in the field of rickettsiology, an explosion of new isolates of pathogens havereceived species designation1–5 and new disease names, all of which have beenrelatively neglected by primary care and infectious disease physicians.6,7

Rickettsial and ehrlichial diseases are remarkable for their uniform susceptibility todoxycycline but are clinically difficult to distinguish from many viral infections and eachanother, and therefore misdiagnosis and failure to treat have unfortunate and some-times tragic outcomes. Rocky Mountain spotted fever (RMSF) and human monocyto-tropic ehrlichiosis (HME) have substantial case-fatality rates. In North America, at leastfive well-established tick-borne, obligately intracellular bacterial pathogens (Rickettsiarickettsii, R parkeri, Ehrlichia chaffeensis, E ewingii, and Anaplasma phagocytophilum)and four other pathogens exist (R massiliae, R prowazekii, R felis, and E canis) thathave been identified in ticks elsewhere in the world, but remain to be definitively iden-tified as tick-transmitted infections in the United States. Finally, a broad group of othertick-associated rickettsial and ehrlichial agents of unknown pathogenicity exist (eg,R amblyommii) that may cause confusion in interpreting serologic surveys or a singleelevated antibody titer. Globally, many of these bacteria have been named (Table 1)

a Department of Pathology, University of Texas Medical Branch, 301 University Boulevard,Galveston, TX 77555-0609, USAb Centers for Disease Control and Prevention, Building CLFT Mailstop: G32, Atlanta, GA30329-4018, USAc DepartmentofPathology, Center for Biodefense andEmerging Infectious Diseases, TheJohnsHop-kins University, 720 Rutland Avenue, Ross Research Building, Room 624, Baltimore, MD 21205, USA* Corresponding author.E-mail address: [email protected] (D.H. Walker).

Med Clin N Am 92 (2008) 1345–1361doi:10.1016/j.mcna.2008.06.002 medical.theclinics.com0025-7125/08/$ – see front matter ª 2008 Elsevier Inc. All rights reserved.

mailto:[email protected]

http://medical.theclinics.com

Table 1Epidemology of tick-borne rickettsial infections

Agent Disease TickVectorGeographicDistribution

Rickettsiarickettsii

Rocky Mountainspotted fever

Dermacentorvariabilis

Eastern two thirdsof United Statesand Pacific Coast

D andersoni Rocky Mountainstates

Rhipicephalussanguineus

Arizona, northernMexico

Amblyommacajennense,A aureolatum

Central and SouthAmerica

Rickettsiaconorii

Boutonneuse fever Rh sanguineus Southern Europe,Africa, westernand southern Asia

Rh pumilio Southern Russia

Rickettsiaafricae

African tick bitefever

A hebraeum Southern Africa

A variegatum Central, east,and west Africa,West Indies

Rickettsiasibirica

North Asia ticktyphus andlymphangitis-associatedrickettsiosis

D nuttallii, Dsilvarum,Haemaphysalisconcinna,Hyalommaasiaticum,other species

Eurasia and Africa

Rickettsiaaustralis

Queensland ticktyphus

Ixodes holocyclus Eastern Australia

Rickettsia honei Flinders Islandspotted fever

Bothrocrotonhydrosauri,other species

Australia andsoutheastern Asia

Rickettsiajaponica

Japanese spottedfever

Vector status notestablished forticks that arehosts of theagent (H flava,H longicornis,I ovatus, Dtaiwanensis)

Japan and Korea

Rickettsia slovaca Tick-bornelymphadenopathy

D marginatus,D. reticularis

Europe

Rickettsia parkeri R parkeririckettsiosis

A maculatum United States,

A triste, Adubitatum

Brazil, Uruguay,Argentina

Rickettsiaaeschlimannii

Unnamed disease H marginatum Africa

Rickettsiaprowazekii

Not characterized A imitator, Htruncatum

North America,Africa

Walker et al1346

Tick-Transmitted Rickettsial and Ehrlichial Infections 1347

but the genetic differences among them are often small, and many of their clinicalmanifestations may not be distinguishable diagnostically (eg, R parkeri rickettsiosis,African tick bite fever).

CYCLIC FLUCTUATIONS IN INCIDENCEOF ROCKYMOUNTAIN SPOTTED FEVER

The United States incidence of RMSF cases reported to the Centers for DiseaseControl and Prevention8 has increased in recent years to 2288 cases in 2006 and2106 cases in 2007, the highest recorded levels in more than 80 years of national sur-veillance for this disease. RMSF also seems to be reemerging in several countries inLatin America (Fig. 1).9–11 The reasons for this trend are multifactorial, and patternsof dramatic cyclic waxing and waning have occurred previously over 30- to 40-yearintervals.12 An increase in the 1970s and early 1980s was followed by a decreaseduring the late 1980s and early 1990s. Proposed causes for these cyclic changes in-clude increased human contact with ticks because of recreational activities, extensionof homes into partially developed natural lands, and other variables that were subse-quently ruled out because of lack of change during periods of decreased incidence.13

Finally, increasing reliance on serologic assays that are not specific for R rickettsiiinfection may cause falsely elevated incidence rates of RMSF. In summary, the mech-anism of reemergence of tick-borne rickettsioses is not known.

ECOLOGYOF SPOTTED FEVER GROUP RICKETTSIAE

Rickettsia rickettsii, the most pathogenic of all known rickettsial species, resulted inthe death of 23% of infected persons in the preantibiotic era. It is known to be trans-mitted by three tick species in the United States, namely Dermacentor variabilis (Amer-ican dog tick), D andersoni (Rocky Mountain wood tick), and recently in an outbreakamong Native Americans in Arizona Rhipicephalus sanguineus (brown dog tick).14

In South America, R rickettsii is transmitted by two other tick species, Amblyommacajennense (Cayenne tick) and A aureolatum (golden dog tick). Many United Statestick species that frequently bite humans, including the highly prevalent Ixodes scapu-laris (blacklegged tick) and Amblyomma americanum (lone star tick), do not transmitR rickettsii.

Fig. 1. Annual incidence and number of reported cases of Rocky Mountain spotted fever(1920–1920). (Courtesy of J.S. Dumler, MD, Baltimore, MD.)

Walker et al1348

The lifecycle of hard ticks such as Dermacentor, Rhipicephalus, and Amblyommaspecies consists of three feeding stages: larvae that hatch from the eggs, nymphsthat develop from larvae, and adults that develop from nymphs and are differentiatedinto males and females. Each of these stages takes one blood meal and remains at-tached to the host for a few to several days, depending on the stage and other factors.The life span of a tick can be completed in 1 year or may require as many as 5 years,depending on success in obtaining the blood meals. Paradoxically, R rickettsii is path-ogenic for the ticks that maintain the rickettsiae through transovarial transmission inthe infected eggs that infected female ticks lay.15 Ticks may maintain R rickettsii bac-teria transovarially for several generations before these are killed by the rickettsial in-fection.16 Thus, to survive in nature, R rickettsii must be periodically transmittedhorizontally to vertebrate hosts, such as cotton rats, which develop rickettsemia ofsufficient magnitude to infect cofeeding ticks.17,18 The newly infected larval ornymphal ticks are the start of new lines of transovarially maintained rickettsiae. Thepathogenicity of R rickettsii for the tick hosts likely explains the low prevalence (gen-erally <0.1%) in ticks carrying this pathogen.15,16,19

Another factor that affects the distribution of R rickettsii in ticks is rickettsial interfer-ence, the phenomenon through which infection of a tick’s ovary with a nonpathogenicrickettsia such as R peacockii prevents the establishment of transovarian infection byR rickettsii in that tick.4,19,20 In most geographic locations, most ticks contain neitherpathogenic nor nonpathogenic rickettsiae.16 Currently, data are insufficient in theUnited States regarding the size of the tick population, the changes and mechanismsof the variation of R rickettsii prevalence in these ticks, and the incidence of human–tick encounters to develop a model that explains the reemergence and cyclic fluctu-ations of RMSF.

DIVERSITYOF SPOTTED FEVER GROUP RICKETTSIAE AND RICKETTSIOSES

The phylogeny of Rickettsia reveals four groups: the classic typhus and spotted fevergroups, an ancestral group (eg, R bellii, R canadensis), and a transitional group(R akari, R australis, R felis). All rickettsiae of the redefined spotted fever group aretick-associated. In North America, the named species of the spotted fever groupinclude the highly virulent R rickettsii, the moderately pathogenic R parkeri, many rick-ettsiae of undetermined pathogenicity (R amblyommii, R rhipicephali, and R monta-nensis), and nonpathogenic bacteria (eg, R peacockii).4,5 R massiliae, which hasbeen isolated recently from Rh sanguineus ticks collected in Arizona, is associatedwith one report of human disease in Sicily.21,22 R massiliae and Rh sanguineus weremost likely introduced into the New World from the Eastern Hemisphere on domesti-cated dogs. Many other nonspeciated isolates of spotted fever rickettsiae are alsoobtained from North American ticks, including strain 364D from D occidentalis andstrain Tillamook from Ixodes pacificus, which are mildly pathogenic in experimentallyinfected animals and require further genetic characterization to identify their unique-ness.23 The geographic distribution of R rickettsii, R parkeri, and R amblyommii extendthrough South America.24–26

No convincing evidence shows that R rickettsii infection can be asymptomatic.Thrombocytopenia, petechial rash, noncardiogenic pulmonary edema, hypotension,acute renal failure, meningoencephalitis, and death occur in many patients whohave RMSF who are not treated with doxycycline during the first 5 days of ill-ness.27–29 In contrast, R parkeri seems to be a less virulent organism. Although theagent was discovered in 1937, the first human infection was not documented until2004, and the incidence of the disease and its spectrum of severity remain to be


determined.30,31 However, no infections ascribed to R parkeri have been fatal or life-threatening. Other differences between A maculatum–transmitted R parkeri infectionand RMSF include the presence of an eschar at the site of Amblyomma tick bite,tender regional lymphadenopathy, and a vesicular or pustular rash. In Uruguay,many patients most likely infected with R parkeri do not have a rash.32 The eschar,an approximately 1-cm focus of epidermal and dermal necrosis associated withextensive vasculitis surrounded by an erythematous halo, is an important diagnosticfeature that should be sought during a complete skin examination, including the groin,waist, axillae, and scalp (Fig. 2).

R amblyommii is found in a high portion of A americanum (lone star ticks), one of themost prevalent and aggressive human-biting ticks in the southeastern and southcentral United States33–35 and is maintained transovarially in its arthropod host. Thepresence of R amblyommii and prevalence of A americanum tick bites are associatedwith a substantial occurrence of persons who have antibodies to spotted fever grouprickettsiae.36–39 The organisms of the spotted fever group are closely related geneti-cally and share antigens of the immunodominant rickettsial cell wall lipopolysaccha-rides and outer membrane proteins A and B. R rickettsii antigens have detected thepresence of antibodies in many healthy persons who have no history of RMSF orclinically compatible illness.40,41 These antibodies have sometimes been interpretederroneously as indicative of asymptomatic RMSF. In prospective studies of militarypersonnel who were heavily exposed to R amblyommii–infected A americanum ticks,numerous seroconversions to antigens of spotted fever group rickettsiae oc-curred.36,38,42 Although most persons who seroconverted did not report any illness,some symptoms, including headache, myalgia, rash, joint pain, fever, chills, dyspnea,and confusion, were reported more frequently (odds ratio, >2.0) in soldiers who hadantibodies to spotted fever group rickettsiae.38 Among 32 soldiers who had antibodiesto spotted fever group rickettsiae detected with enzyme immunoassay, 44% had anillness (odds ratio, 3.5; P < .0001). Evidence shows that some soldiers developedantibodies with specificity for an isolate of spotted fever group rickettsia from A amer-icanum.43 One could tentatively conclude that R amblyommii is an organism of mild-to-minimal virulence. R amblyommii has also been proposed to be a potential agent ofsouthern tick-associated rash illness, (STARI), which manifests as erythema migransand is confused clinically with Lyme disease.3 STARI had previously been associated

Fig. 2. Eschar associated with Rickettsia parkeri infection. (Courtesy of C.A. Ohl, MD,Winston-Salem, NC.)

Walker et al1350

with Borrelia lonestari, which is also found in A americanum ticks. However, subse-quent studies have not supported a borrelia origin.44,45

ATICK-BORNE TYPHUS GROUP RICKETTSIOSIS

In 1955 Reiss-Gutfreund46 reported the discovery of R prowazekii in Hyalomma ticks inEthiopia. The identity of the organism was confirmed and is indistinguishable fromR prowazekii Addis Ababa strain. However, other investigators were unable to confirmthe presence of R prowazekii in ticks in Ethiopia.47

The detection of a high prevalence of antibodies to typhus group rickettsiae amongpatients who had dengue-like illnesses in the state of Nuevo Leon in northern Mexicoled to field studies in which R prowazekii was detected in A imitator ticks and an isolatewas obtained48 (A. Medina-Sancheze, unpublished data, 2007).

This discovery may explain the occurrence of infection with R prowazekii in a 50-year-old man from New Mexico who had vacationed in south Texas, where the rangeof A imitator ticks extends.49 He developed headache, fever, photophobia, a stiff neck,abdominal pain, and cerebrospinal fluid pleocytosis. DNA of R prowazekii was ampli-fied from cerebrospinal fluid samples collected on two separate days, IgG indirectfluorescent antibodies to typhus group rickettsiae rose to a titer of 512, and treatmentwith doxycycline led to defervescence. The last recorded outbreak of louse-borne ty-phus in the United States occurred on the San Juan Navajo Indian Reservation in theFour Corners region of the southwestern United States during 1920 to 1921,50 makingthe possibilities of louse-borne typhus, or Brill-Zinsser disease (recrudescent typhus),seemingly remote because of the absence of human body lice on the patient or hiscontacts and lack of history of previous epidemic typhus.49 The few contemporaryUnited States cases or R prowazekii infection have been consistently associatedwith flying squirrels;51 however, the absence of flying squirrels in south Texas sug-gests another source of infection. The most likely possibility is Amblyomma tick–trans-mitted R prowazekii infection. The recognition of this epidemiologic situation mayprovide insights to the true origin of typhus fever and a previously unrecognized res-ervoir of the infection.

HUMAN EHRLICHIOSES ANDANAPLASMOSIS

Until 1987, ehrlichial infections were in the domain of veterinary medicine. Microcolo-nies of small coccobacilli in Romanovsky-stained peripheral white blood cells of a tick-exposed patient who had acute febrile illness were identified as Ehrlichia with electronmicroscopy and the presence of antibodies reactive with E canis in the patient’s se-rum.52 In 1991, the etiologic agent E chaffeensis was isolated and identified as a novelpathogen.1,53 In 1994 and 1999, molecular methods determined that A phagocytophi-lum and E ewingii, respectively, are also human pathogens.6,54,55 Investigators in Ven-ezuela have identified E canis as a cause of human infections that in some casesresemble human monocytotropic ehrlichiosis.56

E chaffeensis has been identified in several vertebrate animals, such as white-taileddeer, dogs, coyotes, and goats, with A americanum the most important vector(Table 2).57–60 These vertebrate hosts remain persistently infected and are bacteremicfor prolonged periods.60 Experimental studies have shown that A americanum tickscan acquire E chaffeensis while feeding as larvae or nymphs on white-tailed deer.61

They maintain the ehrlichial infection transstadially (ie, during molting to the nymphalor adult stage). These infected ticks can transmit E chaffeensis to other white-taileddeer during the next blood meal. Ehrlichiae do not pass from one generation of ticksto the next through transovarian transmission.62 E chaffeensis can overwinter in

Table 2Epidemology of tick-borne ehrlichioses and anaplasmosis

Agent DiseaseTickVector

GeographicDistribution

VertebrateHosts

Ehrlichiachaffeensis

Humanmonocytotropicehrlichiosis

Amblyommaamericanum,Dermacentorvariabilis,Ixodespacificus

Southeastern andsouth-centralUnited States,California

White-taileddeer, dogs,coyotes,goats

Ehrlichiaewingii

Ehrlichiosisewingii

A americanum Southeastern andsouth-centralUnited States,California

White-taileddeer, dogs

Ehrlichiacanis

Unnameddisease

Rhipicephalussanguineus

Worldwide Dogs, jackal,coyote, wildAfrican dog,red andgray fox

Anaplasmaphagocyto-philum

Humangranulocyto-tropicanaplasmosis

I scapularis NorthernUnitedStates

White-footeddeer mouse,white-taileddeer, dogs,horses,squirrels,chipmunks,red-backedvole

I pacificus PacificcoastalUnitedStates

Squirrels,wood rats,elk, horses,llama,black-taileddeer, deermice

I ricinus Europe red deer, roedeer, fallowdeer, horses,dogs, cattle,cats, sheep,bank voles,wood mice,yellow-neckedmouse,commonshrew


infected replete nymphs, unfed adult ticks, and deer. A high proportion (5%–15%) ofadult lone star ticks are infected with E chaffeensis.

The ecologic cycle of E ewingii is very similar, involving naturally infected white-taileddeer, dogs, and A americanum ticks.63 E ewingii can be acquired by A americanum ticksfeeding on infected dogs, is maintained when the ticks molt to the next life stage, and istransmitted to other dogs during the tick’s subsequent blood meal.64 E ewingii can alsobe abundant in lone star ticks collected in nature, although it is generally found less

Walker et al1352

frequently than E chaffeensis.63 A third Ehrlichia sp, designated provisionally as the Pan-ola Mountain Ehrlichia, has been detected in A americanum ticks and white-tailed deerfrom several southeastern states.65 This agent causes febrile disease in experimentallyinfected animals, although its role as a human pathogen remains undetermined.

Classic canine monocytic ehrlichiosis was first described in Tunis in 1935. It is trans-mitted by the brown dog tick, Rh sanguineus, which has been globally distributed bydomesticated dogs accompanying human migrations. E canis causes persistentinfection of dogs, from which brown dog ticks acquire the infection and maintainthe ehrlichiae transstadially but not transovarially.66 Rh sanguineus is not monophy-letic. Strains of Rh sanguineus in the United States feed predominantly on dogs andseldom bite humans. In contrast, European Rh sanguineus frequently feeds on rodentsand is the vector of R conorii (Mediterranean spotted fever or boutonneuse fever).Human infections resembling HME caused by E canis have been documented in SouthAmerica, presumably transmitted by brown dog ticks.56 Because E canis is not main-tained transovarially in ticks, the reservoir of ehrlichiae in nature is a cycle involvingpersistently infected vertebrate hosts and infected tick hosts.

Veterinary diseases caused by what is now designated as A phagocytophilum wereidentified in sheep in the United Kingdom in 1932, cattle in Scandinavia by the 1960s,and horses in the United States in the 1970s. During the early 1990s, Johan Bakkenand colleagues,6 a physician in Duluth, Minnesota, recognized the similarity of cytoplas-mic inclusions in peripheral blood neutrophils of hispatients to those in monocytes of pa-tients who had HME. Subsequently, Chen and colleagues55 used serology with antigensof peripheral blood leukocytes of infected horses, electron microscopy, universal 16SrRNA gene amplification and DNA sequencing, and immunohistochemistry to determinethat the etiologic agent was the organism currently classified as A phagocytophilum. In-fections are transmitted by I ricinus complex ticks and involve many vertebrate hosts inNorth America, Europe, and Asia (see Table 2). As forEhrlichia spp A phagocytophilum isnot vertically transmitted to the tick progeny, and depends on persistent bacteremia(deer, small rodents) and infection in high proportions of permissive mammalian andtick reservoir populations. Ecologic changes that affect these factors directly impact nat-ural transmission, and the proximity to and likelihood of human exposure permit periodicinfections in humans. Nonpermissive mammalian hosts, such as humans, horses, anddogs, develop significant inflammatory disease that interrupts persistent bacteremia,probably precluding a significant contribution to natural maintenance of anaplasmae.

APPROACH TO PATIENTSWHOHAVE UNDIFFERENTIATED FEBRILE ILLNESSAND POTENTIALTICK EXPOSURE

RMSF, R parkeri rickettsiosis, typhus fever, HME, E ewingii ehrlichiosis, human infec-tions with E canis, and human granulocytic anaplasmosis (HGA) manifest similarsymptoms at onset and during the first few days of illness. These symptoms includefever, headache, myalgia, and malaise.6,7,23,31,67–72 Nausea and vomiting also occurwith some infections. Thus, the diagnosis is difficult to distinguish from the syndromeassociated with many viral infections. Laboratory evaluation often shows similarabnormalities, including thrombocytopenia, elevated hepatic transaminases, andhyponatremia. Leukopenia (< 4000 white blood cells/mL) is observed more commonlyin HME and HGA but can occur in individual patients who have any of these infec-tions.7,71,73–75

The severity of each illness reflects the relative virulence of the causative agent andspecific host factors that confer a greater risk for disease severity.12,76–78 The approx-imate case fatality ratios range from 4% to 5% for RMSF and 3% for HME to 0.7% for


HGA. No deaths have been reported for human infections with R parkeri, E ewingii, andE canis. E ewingii has been diagnosed mostly as an opportunistic infection in personswho are immunocompromised.54,77

Presence of rash favors RMSF (90%), HME (31%), and R parkeri rickettsiosis (toofew cases have been described to assess frequency). Patients who have HGA seldommanifest a rash unless coinfected with B burgdorferi associated with erythemamigrans.67,74 Detection of an eschar would strongly favor diagnosis of R parkeriinfection. Meningoencephalitis with coma and seizures is a frequent manifestationof late-stage RMSF and typhus fever in untreated patients. Meningoencephalitis iswell documented in some patients who have HME but is rare in HGA.69 Respiratorydistress syndrome occurs more often in RMSF and HME than in HGA. Patients whoare immunocompromised and have HME often develop fatal overwhelming ehrlichialinfection.77 Severity of HGA and RMSF does not seem to be affected significantly bya state of immunocompromise.

The most important diagnostic key is obtaining a history of an attached tick or tickexposure by asking the patient the question, ‘‘Have you seen any ticks lately?’’ fol-lowed by inquiry about recreational and occupational activities. High grass alongroadsides, trails, forest edges, stream banks, and weedy yards harbor questing ticksthat attach to passing walkers, hikers, fishermen, persons doing gardening and yardwork, and outdoor workers. Domestic pets such as dogs are frequent vehicles thatbring ticks into the house. Depending on the particular disease, as many as three dif-ferent life stages of the same tick species are capable of transmitting the causativeagent to a human host when the tick obtains a blood meal, and larval and nymphalstage ticks can be extremely small, sometimes smaller than the head of a pin. Inthis context, as many as 40% of persons who have documented tick-transmitteddiseases are unaware of a tick bite, which is typically painless. Thus, probing theactivities of patients who have acute fever is worthwhile.

ROLE OF THE LABORATORY IN DIAGNOSIS OF RICKETTSIOSES, EHRLICHIOSES,AND HUMANGRANULOCYTIC ANAPLASMOSIS

The clinical laboratory offers little assistance in diagnosing rickettsial, ehrlichial, andanaplasmal infections during the early stage of illness when therapeutic decisionsare required.7,68,71,78–80 The presence of thrombocytopenia, leukopenia, or elevatedserum concentrations of aspartate aminotransferase and alanine aminotransferase in-crease the odds of one of these diagnoses. However, normal values do not excludethe diagnoses, and abnormal values are observed in other medical conditions. Furtherlaboratory evaluation of the illness, such as clinical chemistries, blood gases, andcerebrospinal fluid examination, helps define the pathophysiology and determinesupportive care, but will not define the etiologic agent.

Laboratory methods able to determine the diagnoses of these diseases are seldomavailable in the clinic or hospital. During the acute stage of RMSF and R parkeri rick-ettsiosis, immunohistochemical examination of biopsies of cutaneous lesions is themost sensitive (70% sensitivity) approach.31,80 Appropriate tissue samples for the im-munohistochemical diagnosis of HME are not readily accessible.81 This method isavailable from only a few reference and research laboratories, requires a qualifiedobserver, and cannot be applied until the appearance of an exanthem or eschar. Con-temporary improvements in nucleic acid amplification technology offer the opportunityto develop techniques, such as real-time polymerase chain reaction (PCR), and applythem to RMSF, HME, and HGA. Early reports of diagnostic PCR for RMSF weredisappointingly insensitive.82,83

Walker et al1354

Few patients who have RMSF have detectable antibodies to R rickettsii in their serumon day five of illness, after which the case fatality rate increases substantially. Mostrickettsial serologic tests are performed at reference laboratories, which use indirectimmunofluorescence and enzyme immunoassays. These tests do not distinguishwhich particular organism among the spotted fever group or typhus group rickettsiaestimulated the antibody response. Many patients’ antibodies reactive with R rickettsiidetected early in the illness may be actually stimulated at an earlier time by other spot-ted fever group rickettsiae, such as A americanum–inoculated R amblyommii. A four-fold rise in titer in a patient who has a clinically compatible illness is much strongerevidence for RMSF. However, convalescent sera are seldom submitted, and patientsinfected with R parkeri and clinical signs of rickettsiosis would develop a fourfold rise inantibody titer cross-reactive with R rickettsii. Methods such as cross-absorption usingselected rickettsial antigens are impractical except in research laboratories. Even thenone cannot be sure that the antigens selected to be examined represent all of thepotential etiologic agents, because patients infected with R parkeri, and likely otheras-yet-undiscovered Rickettsia species, will also generate antibodies that react withR rickettsii antigens in standard serologic assays.

Although rickettsiae can be cultivated in antibiotic-free cell culture, particularly withcentrifugation-enhanced shell vial techniques, this method does not yield rapid resultsand requires Biosafety Level 3 laboratory facilities and procedures, which are availablemainly in research laboratories.

HME and HGA were discovered through detecting circulating monocytes andneutrophils, respectively, with cytoplasmic vacuoles containing microcolonies ofbacteria.6,52 Although this diagnostic approach is available at clinical presentation, itis insensitive (<10%) for diagnosing HME except in severely immunocompromisedpatients who have overwhelming infection.69 In contrast, detecting inclusions in neu-trophils from patients who have HGA is more diagnostically sensitive (25%–75%)(Fig. 3). Because infected leukocytes are in the blood of patients who have HMEand HGA, diagnosis through PCR amplification of the organisms’ DNA is a relativelysensitive (60%–90%) method.69 This diagnostic test is available only in referenceand research laboratories. Currently, diagnosis of HME and HGA through isolatingthe etiologic agents is undertaken only in research laboratories.69,84

Fig. 3. Neutrophil in the peripheral blood of a patient who has HGA contains three morulae(cytoplasmic vacuoles containing microcolonies of A phagocytophilum; Wright-Giemsastain, original magnification �780). (Courtesy of J.S. Dumler, MD, Baltimore, MD.)


The serologic diagnosis of HME and HGA is useful in the portion of patients(22%–55%) who have developed antibodies when the sera are collected.69 Becausethe clinical manifestations of HME and HGA generally progress slower than in RMSF,patients tend to seek medical attention a few days later in the course, when they aremore likely to have already produced antibodies to ehrlichiae or anaplasma, thanthose who have RMSF would for rickettsiae. Serologic diagnosis is sometimes con-founded by cross-reactivity between E chaffeensis and A phagocytophilum and thehigh prevalence of antibodies against A. hagocytophilum in some geographic areas,underscoring the preference for testing acute and convalescent sera for a seroconver-sion or fourfold increase in antibody titer. Patients infected with E ewingii are diag-nosed specifically only with PCR because E ewingii antigens are not available andE ewingii and E chaffeensis are serologically cross-reactive.54 One could easily hy-pothesize that the presence of antibodies against E chaffeensis in healthy personsin the range of A americanum ticks is caused by mild or asymptomatic infectionwith mildly pathogenic E ewingii, or possibly other uncharacterized Ehrlichia spe-cies.84 Ehrlichial inclusions, if present, are observed in neutrophils and eosinophilsof patients who have E ewingii ehrlichiosis, but could be confused with A phagocyto-philum without other supportive laboratory data.

TREATMENT

Fortunately, the preferred drug for these tick-borne rickettsioses, ehrlichioses, and an-aplasmoses is the same—doxycycline—even in children younger than 8 years.85–87

Chloramphenicol has long been used as a less-effective secondary drug for patientswho have infections caused by Rickettsia. Some data suggest that chloramphenicolshould no longer be used for RMSF, because patients treated with that drug alonehave a higher risk for death than those treated with a tetracycline antibiotic, and thesame risk for death as patients treated with neither.88 However, chloramphenicolmay still have a role when tetracyclines are contraindicated, such as in patients whohave hypersensitivity or during pregnancy. Ehrlichia and Anaplasma are resistant tochloramphenicol. Rifampin inhibits the growth of ehrlichiae and anaplasma in vitroand has been used to treat HGA in children and during pregnancy.89,90 Fluoroquino-lones have not been proven effective in RMSF, HME, or HGA. Empiric oral treatmentwith doxycycline (100 mg twice daily) should be initiated in patients with clinical orepidemiologic evidence of rickettsial, ehrlichial, or anaplasmal infection, and contin-ued until patients feel substantially improved and have been afebrile for at least48 to 72 hours. Because some of these diseases are life-threatening and have littletime in which life-saving therapy can be administered effectively, treatment decisionsshould never be delayed while awaiting laboratory confirmation of the diagnosis.91

REASONS FOR THE EMERGENCEOF TICK-BORNEHUMAN RICKETTSIOSES,EHRLICHIOSES, ANDANAPLASMOSIS

Application of technologic advances led to the identification of novel agents (eg,E chaffeensis) and the association of previously known organisms with human infec-tions (eg, E ewingii, A phagocytophilum, R parkeri).1,2,31,54,55 Development of diagnos-tic assays enabled the epidemiology of the infections and ecology of the agents to bestudied. Thus, to some extent, emergence was actually recognition of organisms thathad been present for eons. Increasing populations of severely and moderately immu-nocompromised persons with conditions ranging from HIV-AIDS to advanced age ledto overwhelming infections in index patients with the visualization of organisms in theblood (eg, E. chaffeensis and A. phagocytophilum).6,52

Walker et al1356

However, ecologic changes were clearly responsible for dramatic increases in thewhite-tailed deer population. The remarkable reforestation of the Eastern United States,owing partly to growth of trees in expanding suburban areas, has been an important fac-tor in providing habitat for the large deer population and A americanum and I scapularisticks.34 The population of white-tailed deer is a critical factor in determining the popu-lation densities of A americanum and I scapularis ticks. Wild turkeys are also a determin-ing factor in the size of the lone star tick population.92 White-tailed deer and wild turkeyshave rebounded from near extinction to populations beyond historic records. Coyotepopulations and geographic range have also expanded remarkably. White-taileddeer, coyotes, and lone star ticks are the major reservoir of E chaffeensis.59 The largepopulation of lone star ticks infected with E chaffeensis resulted in increased transmis-sion to humans and the recognition of HME.92–95 White-tailed deer and lone star ticksprovide a significant reservoir of E ewingii, with similar opportunities for transmissionof E ewingii to humans. The emergence of HGA parallels the expansion of white-taileddeer and small mammal populations that support adult and immature stages of Ixodesspp ticks, as observed with B burgdorferi, which shares the same tick vector.96

The tremendous increase in reported cases of RMSF is confounded by the fact thatonly 15% of these cases are laboratory confirmed using Centers for Disease Controland Prevention criteria, and only approximately 5% are confirmed with diagnostic as-says that specifically identify the causative agent, R rickettsii.97 Even the serologic andimmunohistochemical criteria would not distinguish infections with R rickettsii, R par-keri, and R amblyommii.23 Many patients who have a single indirect immunofluorescentantibody titer of 64 or higher may have antibodies stimulated by the ubiquitous R am-blyommii. Failure to consider and investigate the potential diagnosis of ehrlichiosis byphysicians suggests that some, perhaps many, of these persons actually may have hadHME or HGA. The low case fatality rate in these patients implies that a large portion didnot have RMSF.97 Although something dramatic is occurring, inadequate longitudinaldiagnostic and epidemiologic investigation prevents knowledge of what the diseasesactually are. The high incidence of HME and HGA detected by active prospective stud-ies strongly suggests that these diseases are underreported, perhaps by as much astwo orders of magnitude.7,70,98–100 For example, before the discovery of HME, only12% of patients who had suspected RMSF could be confirmed with serologic tests. Af-ter HME serology became available, an additional 12% of these patients received thatdiagnosis, still leaving 76% of patients who had suspected rickettsial disease undiag-nosed.101 The hope is that many patients who never had a confirmed diagnosis or ac-curate public health reporting are being treated empirically early in the course withdoxycycline by primary care and emergency physicians. However, the recognitionthat 50% of all deaths caused by RMSF in the United States are misdiagnosed showsthat a large gap between theory and practice still exists.28 Accurate and rapid labora-tory confirmation of the diagnoses would assist public health strategic planning andsharpen medical knowledge of these diseases. Until then, the best defense againsteach of these tick-transmitted infections is increasing awareness among general prac-titioners, pediatricians, emergency room physicians, and other primary care doctors ofthe widespread occurrence, potentially severe manifestations, and rapid response todoxycycline that characterize these diseases in the United States.

REFERENCES

1. Anderson BE, Dawson JE, Jones DC, et al. Ehrlichia chaffeensis, a new speciesassociated with human ehrlichiosis. J Clin Microbiol 1991;29:2838–42.

2. Anderson BE, Greene CE, Jones DC, et al. Ehrlichia ewingii sp. nov., the etiologicagent of canine granulocytic ehrlichiosis. Int J Syst Bacteriol 1992;42(2):299–302.


3. Billeter SA, Blanton HL, Little SE, et al. Detection of ‘‘Rickettsia amblyommii’’in association with a tick bite rash. Vector Borne Zoonotic Dis 2007;7:607–10.

4. Niebylski ML, Schrumpf ME, Burgdorfer W, et al. Rickettsia peacockii sp. nov.,a new species infecting wood ticks, Dermacentor andersoni, in western Montana.Int J Syst Bacteriol 1997;47:446–52.

5. Yu XJ, Walker DH. Family I. Rickettsiaceae. In: Brenner DF, Kreig NR, Staley JT,editors, Bergey’s manual of systematic bacteriology. 2nd edition. vol. 2. NewYork: Springer Science1Business Media, Inc.; 2005. p. 96–116.

6. Bakken JS, Dumler JS, Chen SM, et al. Human granulocytic ehrlichiosis inthe upper midwest United States. A new species emerging? JAMA 1994;272:212–8.

7. Olano JP, Masters E, Hogrefe W, et al. Human monocytotropic ehrlichiosis, Mis-souri. Emerg Infect Dis 2003;9:1579–86.

8. CDC. Morbidity and Mortality Weekly Report MMWR 2008;56(51&52):1353.9. Hidalgo M, Orejuela L, Fuya P, et al. Rocky Mountain spotted fever, Colombia.

Emerg Infect Dis 2007;13:1058–60.10. Zavala-Castro JE, Zavala-Velazquez JE, Walker DH, et al. Fatal human infection

with Rickettsia rickettsii, Yucatan, Mexico. Emerg Infect Dis 2006;12:672–4.11. Paddock CD, Fernandez S, Echenique GA, et al. Rocky Mountain spotted fever

in Argentina. Am J Trop Med Hyg 2008;78:687–92.12. Hattwick MAW, O’Brien RJ, Hanson BF. Rocky Mountain spotted fever: epidemi-

ology of an increasing problem. Ann Intern Med 1976;84:732–9.13. Kaplan JE, Newhouse VF. Occurrence of Rocky Mountain spotted fever in rela-

tion to climatic, geophysical, and ecologic variables. Am J Trop Med Hyg 1984;33:1281–2.

14. Demma LJ, Traeger MS, Nicholson WL, et al. Rocky Mountain spotted fever froman unexpected tick vector in Arizona. N Engl J Med 2005;353:587–93.

15. Niebylski ML, Peacock MG, Schwan TG. Lethal effect of Rickettsia rickettsii onits tick vector (Dermacentor andersoni). Appl Environ Microbiol 1999;65:773–8.

16. Burgdorfer W. Ecological and epidemiological considerations of Rocky Moun-tain spotted fever and scrub typhus. In: Walker DH, editor. Biology of rickettsialdiseases. Boca Raton (FL): CRC Press; 1988. p. 33–50.

17. Gage KL, Burgdorfer W, Hopla CE. Hispid cotton rats (Sigmodon hispidus) asa source for infecting immature Dermacentor variabilis (Acari: Ixodidae) withRickettsia rickettsii. J Med Entomol 1990;27:615–9.

18. McDade JE, Newhouse VF. Natural history of Rickettsia rickettsii. Annu Rev Mi-crobiol 1986;40:287–309.

19. Walker DH, Fishbein DB. Epidemiology of rickettsial diseases. Eur J Epidemiol1991;7:237–45.

20. Burgdorfer W, Hayes SF, Mavros AJ. Nonpathogenic rickettsiae in Dermacentorandersoni: a limiting factor for the distribution of Rickettsia rickettsii. In: Burgdor-der W, Anacker RL, editors. Rickettsiae and rickettsial diseases. New York:Academic Press; 1981. p. 585–94.

21. Eremeeva ME, Bosserman EA, Demma LJ, et al. Isolation and identification ofRickettsia massiliae from Rhipicephalus sanguineus ticks collected in Arizona.Appl Environ Microbiol 2006;72:5569–77.

22. Vitale G, Mansueto S, Rolain J-M, et al. Rickettsia massiliae human isolation.Emerg Infect Dis 2006;12:174–5.

23. Paddock CD. Rickettsia parkeri as a paradigm for multiple causes of tick-borne rickettsioses in the Western Hemisphere. Ann N Y Acad Sci 2006;1063:315–26.

Walker et al1358

24. Labruna MB, Pinter A, Szabo MPJ. Rickettsia parkeri infectando um carrapatoAmblyomma triste em pauliceia, estrado de Sao Paulo. Rev Bras Parasitol Vet2004;13(1):359.

25. Pinter A, Labruna MB. Isolation of Rickettsia rickettsii and Rickettsia bellii in cellculture from the tick Amblyomma aureolatum in Brazil. Ann NY Acad Sci 2006;1078:523–9.

26. Labruna MB, Whitworth T, Bouyer DH, et al. Rickettsia bellii and Rickettsiaamblyommii in Amblyomma ticks from the state of Rondonia, Western Amazon,Brazil. J Med Entomol 2004;41:1073–81.

27. Kaplowitz LG, Fischer JJ, Sparling PF. Rocky Mountain spotted fever: a clinicaldilemma. Curr Clin Top Infect Dis 1981;2:89–108.

28. Paddock CD, Greer PW, Ferebee TL, et al. Hidden mortality attributable to RockyMountain spotted fever: immunohistochemical detection of fatal, serologicallyunconfirmed disease. J Infect Dis 1999;179:1469–76.

29. Kirkland KB, Wilkinson WE, Sexton DJ. Therapeutic delay and mortality in casesof Rocky Mountain spotted fever. Clin Infect Dis 1995;20:1118–21.

30. Parker RR, Kohls GM, Cox GW, et al. Observations on an infectious agent fromAmblyomma maculatum. Public Health Rep 1939;54:1482–4.

31. Paddock CD, Sumner JW, Comer JA, et al. Rickettsia parkeri: a newly recog-nized cause of spotted fever rickettsiosis in the United States. Clin Infect Dis2004;38:805–11.

32. Conti Diaz IA. Rickettsiosis por Rickettsii conorii (fiebre botonosa de Mediterra-neo o fiebre de Marsella). Estado actual en Uruguay. Revista Medica del Uru-guay 2001;17:119–24.

33. Childs JE, Paddock CD. The ascendancy of Amblyomma americanum as avector of pathogens affecting humans in the United States. Annu Rev Entomol2003;48:307–37.

34. Paddock CD, Yabsley MJ. Ecologic havoc, the rise of white-tailed deer, and theemergence of Amblyomma americanum-associated zoonoses in the UnitedStates. Curr Top Microbiol Immunol 2007;315:289–324.

35. Stromdahl EY, Vince MA, Billingsley PM, et al. Rickettsia amblyommii infectingAmblyomma americanum larvae. Vector Borne Zoonotic Dis 2007;8:15–24.

36. Kardatzke JT, Neidhardt K, Dzuban DP, et al. Cluster of tick-borne infections atFort Chaffee, Arkansas: rickettsiae and Borrelia burgdorferi in ixodid ticks.J Med Entomol 1992;29:669–72.

37. Marshall GS, Stout GG, Jacobs RF, et al. Antibodies reactive to Rickettsia rick-ettsii among children living in the southeast and south central regions of theUnited States. Arch Pediatr Adolesc Med 2003;157:443–8.

38. McCall CL, Curns AT, Rotz LD, et al. Fort Chaffee revisited: the epidemiology oftick-borne rickettsial and ehrlichial diseases at a natural focus. Vector BorneZoonotic Dis 2001;1:119–27.

39. Yevich SJ, Sanchez JL, DeFraites RF, et al. Seroepidemiology of infections dueto spotted fever group rickettsiae and Ehrlichia species in military personnel ex-posed to areas of the United States where such infections are endemic. J InfectDis 1995;171:1266–73.

40. Taylor JP, Tanner WB, Rawlings JA, et al. Serological evidence of subclinicalRocky Mountain spotted fever infections in Texas. J Infect Dis 1985;151:367–9.

41. Wilfert CM, MacCormack JN, Kleeman K, et al. Epidemiology of Rocky Mountainspotted fever as determined by active surveillance. J Infect Dis 1984;150:469–79.


42. Sanchez JL, Candler WH, Fishbein DB, et al. A cluster of tick-borne infections:association with military training and asymptomatic infections due to Rickettsiarickettsii. Trans R Soc Trop Med Hyg 1992;86:321–5.

43. Dasch GA, Kelly DJ, Richards Al, et al. Western blotting analysis of sera from militarypersonnel exhibiting serological reactivity to spotted fever group rickettsiae [ab-stract]. Journal of theAmericanSocietyofTropicalMedicine&Hygiene1993;49:220.

44. Barbour AG, Maupin GO, Teltow GJ, et al. Identification of an uncultivable Bor-relia species in the hard tick Amblyomma americanum: possible agent of a Lymedisease-like illness. J Infect Dis 1996;173:403–9.

45. James AM, Liveris D, Wormser GP, et al. Borrelia lonestari infection after a biteby an Amblyomma americanum tick. J Infect Dis 2001;183:1810–4.

46. Reiss-Gutfreund RJ. The isolation of Rickettsia prowazeki and mooseri from un-usual sources. Am J Trop Med Hyg 1966;15:943–9.

47. Burgdorfer W, Ormsbee RA, Schmidt ML, et al. A search for the epidemic typhusagent in Ethiopian ticks. Bull World Health Organ 1973;48:563–9.

48. Medina-Sanchez A, Bouyer DH, Alcantara-Rodriguez A, et al. Detection of a ty-phus group Rickettsia in Amblyomma ticks in the State of Nuevo Leon, Mexico.Ann N Y Acad Sci 2005;1063:1–6.

49. Massung RF, Davis LE, Slater K, et al. Epidemic typhus meningitis in the south-western United States. Clin Infect Dis 2001;32:979–82.

50. Armstrong C. Typhus fever on the San Juan Indian reservation. 1920 and 1921.Public Health Rep 1922;37:685–93.

51. McDade JE, Shepard CC, Redus MA, et al. Evidence of Rickettsia prowazekiiinfections in the United States. Am J Trop Med Hyg 1980;29:277–84.

52. Maeda K, Markowitz N, Hawley RC, et al. Human infection with Ehrlichia canis,a leukocytic rickettsia. N Engl J Med 1987;316:853–6.

53. Dawson JE, Anderson BE, Fishbein DB, et al. Isolation and characterization ofan Ehrlichia sp. from a patient diagnosed with human ehrlichiosis. J Clin Micro-biol 1991;29:2741–5.

54. Buller RS, Arens M, Hmiel SP, et al. Ehrlichia ewingii, a newly recognized agentof human ehrlichiosis. N Engl J Med 1999;341:148–55.

55. Chen S-M, Dumler JS, Bakken JS, et al. Identification of a granulocytotropicEhrlichia species as the etiologic agent of human disease. J Clin Microbiol1994;32:589–95.

56. Perez M, Bodor M, Zhang C, et al. Human infection with Ehrlichia canis accom-panied by clinical signs in Venezuela. Ann NY Acad Sci 2006;1078:110–7.

57. Dugan VG, Little SE, Stallknecht DE, et al. Natural infection of domestic goatswith Ehrlichia chaffeensis. J Clin Microbiol 2000;38:448–9.

58. Kocan A, Levesque GC, Whitworth LC, et al. Naturally occurring Ehrlichia chaf-feensis infection in coyotes from Oklahoma. Emerg Infect Dis 2000;6:477–80.

59. Paddock CD, Childs JE. Ehrlichia chaffeensis: a prototypical emerging patho-gen. Clin Microbiol Rev 2003;16:37–64.

60. Davidson WR, Lockhart JM, Stallknecht DE, et al. Persistent Ehrlichia chaffeen-sis infection in white-tailed deer. J Wildl Dis 2001;37(3):538–46.

61. Ewing SA, Dawson JE, Kocan AA, et al. Experimental transmission of Ehrlichiachaffeensis (Rickettsiales: Ehrlichieae) among white-tailed deer by Amblyommaamericanum (Acari: Ixodidae). J Med Entomol 1995;32:368–74.

62. Long SW, Zhang X, Zhang J, et al. Evaluation of transovarial transmissionand transmissibility of Ehrlichia chaffeensis (Rickettsiales: Anaplasmataceae)in Amblyomma americanum (Acari: Ixodidae). J Med Entomol 2003;40:1000–4.

Walker et al1360

63. Paddock CD, Liddell AM, Storch GA. Other causes of tick-borne ehrlichioses,including Ehrlichia ewingii. In: Goodman JL, Dennis DT, Sonenshine DE, editors.Tick-borne diseases of humans. Washington, DC: ASM Press; 2005. p. 258–67.

64. Anziani OS, Ewing SA, Barker RW. Experimental transmission of a granulocyticform of the tribe Ehrlichieae by Dermacentor variabilis and Amblyomma ameri-canum to dogs. Am J Vet Res 1990;51:929–31.

65. Loftis AD, Reeves WK, Spurlock JP, et al. Infection of a goat with a tick-transmittedEhrlichia from Georgia, U.S.A., that is closely related to Ehrlichia ruminantium.J Vector Ecol 2006;31:213–23.

66. Groves MG, Dennis GL, Amyx HL, et al. Transmission of Ehrlichia canis to dogsby ticks (Rhipicephalus sanguineus). Am J Vet Res 1975;36:937–40.

67. Aguero-Rosenfeld ME, Horowitz HW, Wormser GP, et al. Human granulocyticehrlichiosis: a case series from a medical center in New York state. Ann InternMed 1996;125:904–8.

68. Chapman AS, Bakken JS, Bloch KC, et al. Diagnosis and management of tick-borne rickettsial diseases: Rocky Mountain spotted fever, ehrlichioses, andanaplasmosis - United States. MMWR 2006;55:1–29.

69. Dumler JS, Madigan JE, Pusteria N, et al. Ehrlichioses in humans: epidemiology, clin-icalpresentation, diagnosis,and treatment. Clin InfectDis2007;45(Suppl1):S45–51.

70. Fishbein DB, Dawson JE, Robinson LE. Human ehrlichiosis in the United States,1985 to 1990. Ann Intern Med 1994;120:736–43.

71. Olano JP, Hogrefe W, Cullman L, et al. Clinical manifestations, epidemiology,and laboratory diagnosis of human monocytotropic ehrlichiosis in a commerciallaboratory setting. Clin Diagn Lab Immunol 2003;10:891–6.

72. Schutze GE, Buckingham SC, Marshall GS, et al. Human monocytic ehrlichiosisin children. Pediatr Infect Dis J 2007;26:475–9.

73. Bakken JS, Aguero-Rosenfeld ME, Tilden RL, et al. Serial measurements ofhematologic counts during the active phase of human granulocytic ehrlichiosis.Clin Infect Dis 2001;32:862–70.

74. Bakken US, Krueth J, Wilson-Nordskog C, et al. Clinical and laboratory charac-teristics of human granulocytic ehrlichiosis. JAMA 1996;275:199–205.

75. Hamilton KS, Standaert SM, Kinney MC. Characteristic peripheral blood findingsin human ehrlichiosis. Mod Pathol 2004;17:512–7.

76. Eremeeva ME, Dasch GA, Silverman DJ. Quantitative analysis of variations in theinjury of endothelial cells elicited by 11 isolates of Rickettsia rickettsii. Clin DiagnLab Immunol 2001;8:788–96.

77. Paddock CD, Folk SM, Shore GM, et al. Infections with Ehrlichia chaffeensis andEhrlichia ewingii in persons coinfected with human immunodeficiency virus. ClinInfect Dis 2001;33:1586–94.

78. Walker DH, Hawkins HK, Hudson P. Fulminant Rocky Mountain spotted fever: itspathologic characteristics associated with glucose-6-phosphate dehydroge-nase deficiency. Arch Pathol Lab Med 1983;107:121–5.

79. Aguero-Rosenfeld ME, Kalantarpour F, Baluch M, et al. Serology of culture-confirmed cases of human granulocytic ehrlichiosis. J Clin Microbiol 2000;38(2):635–8.

80. Walker DH, Burday MS, Folds JD. Laboratory diagnosis of Rocky Mountain spot-ted fever. South Med J 1980;73:1443–7.

81. Dumler JS, Dawson JE, Walker DH. Human ehrlichiosis: hematopathology andimmunohistologic detection of Ehrlichia chaffeensis. Hum Pathol 1993;24:391–6.

82. Sexton DJ, Kanj SS, Wilson K, et al. The use of a polymerase chain reaction as a di-agnostic test for Rocky Mountain spotted fever. Am J Trop Med Hyg 1994;50:59–63.


83. Tzianabos T, Anderson BE, McDade JE. Detection of Rickettsia rickettsii DNA inclinical specimens by enzymatic amplification using polymerase chain reactiontechnology. Ann NY Acad Sci 1990;590:553–6.

84. Standaert SM, Yu T, Scott MA, et al. Primary isolation of Ehrlichia chaffeensisfrom patients with febrile illnesses: clinical and molecular characteristics.J Infect Dis 2000;81:1082–8.

85. Marshal GS, Jacobs RF, Schutze GE, et al. Ehrlichia chaffeensis seroprevalenceamong children in the southeast and south-central regions of the United States.Arch Pediatr Adolesc Med 2002;156:166–70.

86. American Academy of Pediatrics. Ehrlichia infections (human ehrlichioses). In:Pickering LK, Baker CJ, Overturf GD, et al, editors. 2003 Red book: report ofthe Committee on Infectious Diseases. 26th edition. Elk Grove Village (IL): Amer-ican Academy of Pediatrics, Committee on Infectious Diseases; 2003. p. 266–9.

87. American Academy of Pediatrics. Rocky Mountain spotted fever. In:Pickering LK, Baker CJ, Overturf GD, et al, editors. 2003 Red book: report ofthe Committee on Infectious Diseases. 26th edition. Elk Grove Village (IL): Amer-ican Academy of Pediatrics; 2003. p. 532–4.

88. Holman RC, Paddock CD, Curns AT, et al. Analysis of risk factors for fatal RockyMountain spotted fever: evidence for superiority of tetracyclines for therapy.J Infect Dis 2001;184:1437–44.

89. Buitrago MI, Ijdo JW, Rinaudo P, et al. Human granulocytic ehrlichiosis duringpregnancy treated successfully with rifampin. Clin Infect Dis 1998;27:213–5.

90. Krause PJ, Corrow CL, Bakken JS. Successful treatment of human granulocyticehrlichiosis in children using rifampin. Pediatrics 2003;112:e252–3.

91. Hamburg BJ, Storch GA, Micek ST, et al. The importance of early treatment withdoxycycline in human ehrlichiosis. Medicine (Baltimore) 2008;87:53–60.

92. Lockhart JM, Davidson WR, Dawson JE, et al. Temporal association of Am-blyomma americanum with the presence of Ehrlichia chaffeensis reactiveantibodies in white-tailed deer. J Wildlife Dis 1995;31:119–24.

93. Lockhart JM, Davidson WR, Stallknecht DE, et al. Site-specific geographic asso-ciation between Amblyomma americanum (Acari: Ixodidae) infestations and Ehr-lichia chaffeensis-reactive (Rickettsiales: Ehrlichieae) antibodies in white-taileddeer. J Med Entomol 1996;33:153–8.

94. Lockhart JM, Davidson WR, Stallknecht DE, et al. Natural history of Ehrlichiachaffeensis (Rickettsiales: Ehrlicieae) in the piedmont physiographic provinceof Georgia. J Parasitol 1997;83:887–94.

95. Standaert SM, Dawson JE, Schaffner W, et al. Ehrlichiosis in a golf-orientedretirement community. N Engl J Med 1995;333:420–5.

96. Belongia EA. Epidemiology and impact of coinfections acquired from Ixodesticks. Vector Borne Zoonotic Dis 2002;2:265–73.

97. Chapman AS, Murphy SM, Demma LJ, et al. Rocky Mountain spotted fever in theUnited States, 1997–2002. Vector Borne Zoonotic Dis 2006;6:170–8.

98. Demma LJ, Holman RC, McQuiston JH, et al. Epidemiology of human ehrlichiosis andanaplasmosis in the United States, 2001–2002. Am J Trop Med Hyg 2005;73:400–9.

99. Dumler JS, Dotevall L, Gustafson R, et al. A population-based seroepidemio-logic study of human granulocytic ehrlichiosis and Lyme borreliosis on theWest coast of Sweden. J Infect Dis 1997;175:720–2.

100. Fishbein DB, Kemp A, Dawson JE, et al. Human ehrlichiosis: prospective activesurveillance in febrile hospitalized patients. J Infect Dis 1989;160:803–9.

101. Harkess JR, Ewing SA, Crutcher JM, et al. Human ehrlichiosis in Oklahoma.J Infect Dis 1989;159:576–9.

Documents

Emerging and Re-emerging Tick-Transmitted Rickettsial and Ehrlichial Infections