Coinfecting Deer-Associated Zoonoses: Lyme Disease, Babesiosis, and Ehrlichiosis
- Ellie J. C. Goldstein, Section Editor
- 1Department of Pediatrics, University of Connecticut School of Medicine, Farmington, and the Connecticut Children's Medical Center, Hartford, Connecticut
- 2Department of Infectious Diseases and Immunology, Harvard School of Public Health, Boston, Massachusetts
- Reprints or correspondence: Dr. Peter J. Krause, Dept. of Pediatrics, Connecticut Children's Medical Center, 282 Washington St., Hartford, CT 06106 (PKrause{at}CCMCKIDS.org), or Dr. Andrew Spielman, Dept. of Infectious Diseases, Harvard School of Public Health, 677 Huntington Ave., Boston, MA 02115 (ASpielma{at}HSPH.Harvard.edu).
Abstract
The heightened worldwide recognition of the health burden of tickborne infection derives largely from the increasing incidence of Lyme disease, human babesiosis, and human granulocytic ehrlichiosis, both individually and in concert. Because these infections share the same rodent reservoir and tick vector hosts, they can be cotransmitted to human hosts. Indeed, human coinfections involving various combinations of these pathogens are common, and some tend to be particularly severe. Diagnostic procedures and clinical management of the resulting disease syndrome is rendered complex by the diversity of pathogens involved and by the unusual diversity and duration of symptoms.
Lyme disease, babesiosis, and human granulocytic ehrlichiosis are caused by a guild of cotransmitted pathogens that began to affect residents of southern New England, New York, and the northern Midwest in the 1960s [1–5]. Their etiologic agents, Borrelia burgdorferi, Babesia microti, and Ehrlichia phagocytophila, are maintained by the same vector tick (the deer tick or Ixodes dammini, frequently designated as Ixodes scapularis) and reservoir rodent (the white-footed mouse or Peromyscus leucopus). White-tailed deer (Odocoileus virginianus) serve as the definitive hosts of these vector ticks, and their proliferation in the mid-20th century has been the basis for the emergence of these infections [6]. Infection by any one of these agents frequently accompanies infection by ⩾1 of the others [3, 7–13]. The clinical manifestations of simultaneous Lyme disease and babesiosis tend to be more severe than would be expected if these infections occurred separately [3, 9, 12]. Because their tick vector thrives solely where deer are abundant, the risk of human infection reflects deer abundance and the presence of any of these pathogens [14]. Our objective is to analyze the growing health burden of these coinfecting deer-associated zoonoses.
History
The burden that the deer-associated zoonoses impose on human health began to be recognized after 1969, when a case of human babesiosis was first diagnosed on Nantucket Island [15]. Previously, babesiosis had been described only in a few asplenic people in Europe reportedly infected by Babesia divergens, a piroplasm of cattle. Following a cluster of similar diagnoses that were registered in 1973 and 1974, the name “Nantucket Fever” was used to designate this American disease [15, 16]. The etiologic agent of these disease episodes was B. microti, a parasite of white-footed mice that is one of numerous species in the genus that infect mammals. Such organisms were first identified in cattle in 1888 by the Hungarian pathologist Babes [17]. A few years later, Smith and Kilbourne discovered that Babesia bovis is transmitted between cattle by a particular tick, which was the first demonstration that a hematophagous arthropod transmits a disease-producing pathogen [18]. One of us (A.S.) established that human babesiosis is transmitted by deer ticks [19], lives in white-footed mice [20], and requires the presence of deer [6]. Zoonotic babesiosis has since been recognized in many parts of the world. The vector and reservoir combination maintains the agents of Lyme disease, B. burgdorferi, and human granulocytic ehrlichiosis (HGE), E. phagocytophila.
Although the characteristic erythematous rash of Lyme disease was first noted in Europe at the turn of the century, the full spectrum of this disease and its etiologic agent were not described until recently. The first reported case of “Lyme arthritis” in the Americas dates to 1969 [21], and the first cluster was reported in the mid-1970s by Steere and colleagues [22]. Clusters of people living in 3 Connecticut communities were afflicted by an illness characterized by recurrent arthritis and preceded by an annular rash. This “Lyme arthritis” syndrome was subsequently identified as a facet of a multisystem disease and renamed Lyme disease, or Lyme borreliosis [2]. In 1982, Burgdorfer suggested that a spirochete, soon named B. burgdorferi, is the etiologic agent of this disease [23]. Lyme disease has since been recognized as the most common arthropod-borne illness in parts of the United States and in much of Eurasia [2].
As with babesiosis and Lyme disease, the epidemiological and clinical significance of ehrlichiosis was recognized only recently. The first human case was described in 1954 by Japanese investigators, who detected a rickettsial organism, Ehrlichia sennetsu, in a man experiencing a mononucleosis-like illness [24]. The first American case was identified in Arkansas in 1986, and was followed by numerous case reports of a similar rickettsial-like illness [25]. Ehrlichia chaffeensis was recognized as its etiologic agent in 1991 [26]. This rickettsia is transmitted by the Lone Star tick, Amblyomma americanum, which, like E. sennetsu, infects monocytes. Tick fever of sheep is caused by E. phagocytophila, which is borne by Ixodes ricinus [27]. In the Americas, this organism was first found in voles (Microtis pennsylvanicus) on the island of Martha's Vineyard, Massachusetts, and was described as Cytoecetes microti [28]. This third human ehrlichial infection was later designated as HGE [4]; a similar condition occurs in Europe [29]. More recently, another ehrlichial species, Ehrlichia ewingii, was discovered in human granulocytes [30].
The I. ricinus complex of ticks also transmits a group of flaviviruses that includes tickborne encephalitis virus in Eastern Europe and Asia [31] and deer tick or Powassan virus in the northeastern United States [32]. The health burden imposed by tickborne encephalitis virus is heavy, and deer tick virus presents a potential threat to human health; however, these agents will not be discussed in this article, because little information is available regarding their relationship to the other pathogens in this guild.
Shortly after the discovery of the public health significance of these tickborne, deer-associated zoonoses, episodes of human coinfection by ⩾2 of these pathogens became evident [8–10]. We (P.J.K., A.S., and colleagues) carried out the first epidemiological and clinical study of coinfection and found that ∼10% of patients with Lyme disease in southern Connecticut and Rhode Island are coinfected by the agent of babesiosis [3]. The mean number of symptoms and duration of illness in patients with simultaneous Lyme disease and babesiosis is greater than the sum of such symptoms in patients with either infection alone [3, 9, 12]. We do not know whether HGE similarly affects the clinical expression of Lyme disease [12, 13].
Ecology
Lyme disease, babesiosis, and HGE are transmitted by members of the I. ricinus complex of ticks (including I. dammini in the northeastern and north-central United States, Ixodies pacificus in the western United States, I. ricinus in the British Isles and much of Europe, and Ixodes persulcatus in Eastern Europe and Asia [1, 33–35]. Larval ticks ingest the pathogens while feeding on mice and subsequently transmit ⩾1 of these microbes to other mice or to human hosts when they feed again during their nymph stage [1, 14]. Such ticks feed only once in each developmental stage, and blood is their only food (figure 1). In the spring, adult females deposit clutches of several thousand eggs. Larvae hatch in August and soon begin to quest for a host. In the course of feeding, a larva may ingest⩾1 pathogens that subsequently develop in the tick during its molt to the nymph stage. Nymphs transmit the organisms to rodents or human beings in the late spring or summer [1]. Although both nymphal and adult ticks can transmit infection to people, nymphs do so most frequently. Most people who develop symptoms of infection with one or another of these pathogens do not recall tick contact, because tick attachment is painless and the maximum size of a replete nymph is less than that of a lowercase “o” in this text. Nymphs feed for ∼4 days.
Figure 1
Life cycle of Ixodes dammini (also known as Ixodes scapularis) [1] (Reprinted, with permission, from the Annual Review of Entomology, volume 30, copyright 1985 by Annual Reviews [http://www.AnnualReviews.org]).
The white-footed mouse is the main reservoir host of the causative agents of Lyme disease, babesiosis, and HGE and may be infected by ⩾1 of these pathogens while remaining asymptomatic [1, 36–38]. White-tailed deer serve as the main definitive hosts for vector ticks and provide the blood meal that adult deer ticks require for reproduction during the winter season [1]. Domestic animals, such as dogs, may be infested by adult deer ticks but appear to contribute little to their abundance. Dogs can be infected by B. burgdorferi, the agent of human granulocytic ehrlichiosis, and E. ewingii but not B. microti [14, 30, 39, 40]. The unprecedented increase in deer density that occurred during the 20th century appears to be the central factor in the proliferation and increased geographic range of deer ticks and the resulting increase in deer-associated zoonotic disease in the United States [1, 41, 42] (figure 2).
Figure 2
Relationship between the number of deer killed by hunters each year in Massachusetts, as recorded by the Massachusetts Division of Fisheries and Wildlife, and the number of cases of Lyme disease in the United States, as recorded by the Centers for Disease Control and Prevention.
Although the agent of Lyme disease generally infects some 20%–50% of vector tick nymphs, the prevalence of the other cotransmitted pathogens varies greatly from place to place [37, 43–49]. In western New Jersey, ∼55% of deer ticks are infected by B. burgdorferi, B. microti, or the agent of HGE, and 10% are infected by ⩾2 of these pathogens [44]. On 2 Massachusetts islands, 8%–18% of deer tick nymphs carry both the agent of Lyme disease and the agent of human babesiosis [45]. On Nantucket Island, 10% are infected by the agent of HGE, and 20% of these ticks are coinfected by the agent of Lyme disease [38]. Similar frequencies characterize other sites in the United States and Europe [45–49]. In Germany, Switzerland, and the United Kingdom, for example, the frequencies of various infections in ticks are as follows: B. burgdorferi, 36% in Germany, 49% in Switzerland, and 37% in the United Kingdom; the agent of HGE, 1.6% in Germany, 2% in Switzerland, and 7% in the United Kingdom; and both B. burgdorferi and the agent of HGE, <1% in Germany, 2% in Switzerland, and 1.7% in the United Kingdom [35, 48, 49].
The prevalence of Lyme disease infection in reservoir mammals exceeds that in vector ticks in a given area of endemicity. More than 80% of white-footed mice are infected by the agents of Lyme disease, babesiosis, or HGE, and as many as 40% are coinfected by ⩾2 of these pathogens [1, 50, 51]. In southeastern Connecticut, white-footed mice were captured in or near the homes of 8 patients with babesiosis (5 of whom concurrently had significant levels of IgG or IgM antibody to B. burgdorferi) [51]. Twenty-seven (46%) of 59 white-footed mice had antibody against B. microti, and 25 of these mice (93%) also had antibody against B. burgdorferi. The prevalence of B. burgdorferi in white-footed mice on Prudence Island, Rhode Island, is 86%, while for both B. burgdorferi and B. microti together, the prevalence is 43% [52]. In Connecticut, antibody against the agents of B. burgdorferi, B. microti, and HGE occurs in 83%, 77%, and 53% of mice, respectively [50]. In the upper Midwest, both B. burgdorferi and B. microti occur in 20%–54% of white-footed mice [53]. Dogs have antibody against both Ehrlichia and Babesia species, as well as Bartonella and Rickettsia species [54–56]. About 50% of white-tailed deer in Connecticut similarly have antibodies against both E. phagocytophila genogroup and B. burgdorferi [57].
Human Disease
Epidemiology
Deer-associated zoonoses have become a major public health concern in the United States because human contact with deer ticks has increased as a result of the proliferation of deer, abandonment of farmland that reverts to thick secondary vegetation, and increased use of coastal sites for human recreation or habitation. Recognition of deer-associated infections by physicians and the public also help to explain the increasing frequency of reported human cases of Lyme disease, babesiosis, and HGE (figure 3).
The number of reported cases of Lyme disease far exceeds that of babesiosis or HGE; currently, ∼15,000 cases per year are reported in the United States. Some authorities estimate that the true incidence exceeds the reported incidence 10-fold. Although reporting is mandatory in many states, health departments generally depend on passive modes of reporting that markedly underestimate the true rate of infection. Unlike babesiosis and HGE, Lyme disease appears to be uniformly dispersed in areas of endemicity in the Northeast and northern Midwest. The disparity between the incidence of Lyme disease and babesiosis and HGE also can be attributed to the readily identified erythema migrans rash that is the hallmark of Lyme disease; symptomatic patients with babesiosis or HGE generally experience a flulike illness that is difficult to distinguish from numerous other conditions, including viral infections.
Four Babesia species cause disease in human beings, B. microti, WA1, and MO1 in North America, and Babesia divergens in Europe [59–63] (table 1). Human babesial infection generally is transmitted by ticks, although >30 cases due to blood transfusions have been described [64–66]. Because babesial infection generally is not diagnosed, the number of reported cases probably represents only a small fraction of the actual number of cases. This appears to be especially true for children. Only a few pediatric cases of babesiosis have been reported, even though babesial seroprevalence in children is similar to that in adults [60]. The scarcity of reported pediatric cases of babesiosis may derive, in part, from the commonly held concept that babesiosis is a geriatric rather than a pediatric disease. Many more febrile illnesses affect children than adults, and any febrile illness in an adult tends to be evaluated more aggressively, including evaluation for possible babesiosis, than is febrile illness in a child [60].
Table 1
Deer-associated zoonoses transmitted by ticks in the Ixodes ricinus complex.
Human ehrlichiosis in the United States includes 2 distinct diseases, HGE and human monocytic ehrlichiosis (HME) [4, 5, 67, 68]. HGE (which is due to E. phagocytophila) occurs mainly in the Northeast and northern Midwest, whereas HME (which is due to E. chaffeensis) generally is reported in the southeastern and south-central United States.
Marked differences may exist in the incidence of coinfection and in the combination of pathogens that cause coinfection, because the geographic distribution of these pathogens tends to vary locally. Serosurveys of human hosts have helped define the geographic range and prevalence of coinfection with deer-associated zoonoses, despite the difficulty in distinguishing between simultaneous and sequential infection. About 66% of subjects in a Long Island study had antibody against B. burgdorferi as well as B. microti [7]. In southern New England, between 9.5% and 14% of serum samples obtained from subjects enrolled in studies of deer-associated infection reacted against both B. burgdorferi antigen and B. microti antigen [3, 60, 69]. Multiple reactivity to B. burgdorferi, B. microti, and E. phagocytophila is present in 6.6% of serum samples from areas of endemicity in Connecticut and Minnesota [70]. Among volunteers at high risk for infection on Long Island, New York, 13% had antibodies against >1 tickborne organism, but only 5 subjects had evidence of dual infection [71]. Serum samples from residents of Bulgaria, Denmark, Switzerland, and Norway frequently react against the agents of Lyme disease and ehrlichiosis [72–75].
Simultaneous infection by the agents of Lyme disease, babesiosis, and ehrlichiosis is common [3, 8–13, 59, 76–80]. Diagnosis generally is based on clinical findings (such as erythema migrans rash) and identification of the causative pathogen by microscopic and/or PCR analysis, in addition to serological studies. Among residents of southern New England, 10% of patients with Lyme disease also had babesial infection, in areas where both diseases were zoonotic [3]. Coinfection rates in the northern Midwest seem to range from 9% to 16%, with various combinations of B. burgdorferi, B. microti, and the agent of HGE [13, 81]. Deer-associated zoonoses vary locally, in both the kind and frequency of coinfection.
Pathogenesis
Coinfection by deer-associated zoonotic infections can affect various aspects of the chain of transmission, including the following: (1) it can cause cooperative or competitive interactions of the pathogens in rodent reservoir hosts and tick vector hosts, (2) it can affect transmission from rodent to tick and from tick to rodent or human, and (3) it can increase the severity of disease in human hosts. Where these infections are enzootic, reservoir mice generally are infected by both the pathogens that cause Lyme disease and babesiosis, yet the agent of Lyme disease is twice as prevalent in ticks as is the agent of babesiosis [36]. Indeed, laboratory experiments suggest that this disparity may be due to differences in the efficiency of transmission of B. burgdorferi and B. microti from mice to ticks. Deer ticks fed on mice that are infected concurrently with B. burgdorferi and B. microti become infected twice as frequently with B. burgdorferi as with B. microti, which suggests that the efficiency of acquisition and transstadial passage of these pathogens may differ by a factor of 2 [37]. Such disparity in acquisition of infection by ticks may not occur during transmission to mammalian hosts. One study indicated that B. burgdorferi is transmitted just as frequently as is the agent of human granulocytic ehrlichiosis [82]. The mechanism that causes increased disease severity in people coinfected by the agents of Lyme disease and babesiosis remains uncertain. Babesial infection appears to enhance Lyme disease myocarditis in mice [83], and it may impair host defense mechanisms in cattle, mice, and human beings [84, 85]. Much remains to be learned about the pathogenesis of deer-associated zoonotic coinfections.
Clinical Presentation
The clinical manifestations of babesiosis range from subclinical illness to a fatal fulminating disease [3, 59, 86–92]. Babesial infections are often not apparent [60, 70, 87, 89]. In most clinically apparent cases, overt signs and symptoms generally commence after an incubation period of 1–6 weeks, measured from the beginning of tick feeding. Although patients may present with a mild flulike illness, the presentation generally is more severe and begins with malaise, anorexia, and fatigue that is followed by intermittent fever with a temperature as high as 40°C and ⩾1 of the following symptoms: headache, chills, sweats, myalgia, arthralgia, nausea, and vomiting [86–91]. Less commonly noted are emotional lability and depression, hyperesthesia, sore throat, abdominal pain, conjunctival infection, photophobia, weight loss, and nonproductive cough [86–91]. Physical examination findings often consist solely of fever [86]. Splenomegaly, hepatomegaly, or both are noted occasionally. Abnormal laboratory findings reflect lysis of erythrocytes by the parasite; they include mild to moderately severe hemolytic anemia and an elevated reticulocyte count. Findings of liver function tests are elevated for about one-half of patients. Thrombocytopenia is common. The leukocyte count is either normal or slightly low [92]. Complications of babesial infection are limited. The illness usually lasts for a few weeks to as long as several months; recovery sometimes requires as long as 18 months [86–89]. Patients may remain parasitemic after all symptoms have resolved. As noted with malaria, parasitemia and relapse of illness can occur up to 27 months after the initial episode [89]. Patients at risk for severe babesiosis include those who lack a spleen, who are immunocompromised because of infection with HIV or because of corticosteroid therapy, or who are >50 years of age. These patients may have serious illness, characterized by high fever and severe hemolytic anemia, that results in death or a prolonged convalescence [10, 16, 91, 93, 94]. The case fatality rate for human babesiosis approaches 5% [59].
The Lyme disease spirochete resides mainly in fixed tissues and causes a rash, flulike illness, arthritis, and, less frequently, carditis or neuropathy [2, 95]. The clinical manifestations of Lyme disease have been classified into early, disseminated, and late phases [2]. The symptoms of early Lyme disease begin 2–30 days after the tick bite and feature a characteristic erythema migrans rash and/or a flulike illness. Erythema migrans is defined as an expanding, erythematous rash measuring ⩾5 cm in diameter. The typical lesion usually is not associated with symptoms, although warmth, burning, pruritis, tenderness, hyperesthesia, or dysesthesia may occur. Untreated erythema migrans generally fades in 2–4 weeks but clears within days after appropriate antibiotic therapy is begun [2, 96, 97]. The disseminated stage of Lyme disease is characterized by ⩾1 secondary erythema migrans lesions (which occur in 6%–48% of patients with Lyme disease), arthritis, neurological illness, or carditis. In untreated patients, Lyme disease results in arthritis in ∼60% of patients, neurologic disease in ∼15%, and carditis in ∼5%; these symptoms occur weeks to months after the onset of illness [2, 98]. Characteristics of late Lyme disease include (1) acrodermatitis chronica atrophicans lesions of the skin, (2) chronic arthritis consisting of persistent arthritis of 1 joint, and (3) chronic neurological symptoms [98–100]. The wide range of clinical symptoms of Lyme disease may be explained in part by differences in Borrelia genospecies or substrains in different geographic areas [101, 102]. This is particularly evident in Europe; although only 1 genospecies is associated with human disease in the United States, several such genospecies are clinically relevant in Europe [2] (table 1).
HGE is an acute, systemic, febrile illness often accompanied by headache, chills, malaise, myalgias, arthralgias, nausea, vomiting, anorexia, and acute weight loss [4, 5, 67, 103]. Rash occurs in ∼10% of cases of HGE. Complications include disseminated intravascular coagulation, meningitis, pulmonary infiltrates, respiratory failure, bone marrow hypoplasia, and renal failure. Laboratory abnormalities include anemia, thrombocytopenia, hyponatremia, elevated levels of liver enzymes, and CSF fluid pleocytosis with a predominance of lymphocytes and an elevated total protein concentration. The disease typically lasts 1–2 weeks without sequelae, but neurologic sequelae may occur and ∼5% of cases are fatal [67, 103]. Fatalities have been associated with delays in diagnosis and failure to give appropriate antibiotic therapy early in the course of illness.
Coinfection by the agents of Lyme disease and babesiosis appears to increase disease severity [3, 9, 10, 12]. In previous case-finding studies in Rhode Island and Connecticut, patients coinfected by the agents of Lyme disease and babesiosis had more symptoms that lasted longer than did patients infected with either alone [3, 12]. It is less clear whether patients with both Lyme disease and HGE have a different clinical outcome than do patients with Lyme disease alone [12, 13]. The long-term outcome of coinfection, however, is reported not to differ from that of monoinfection [104]. Because this negative finding was based on a retrospective serological study, exposure to these parasites would most frequently have been sequential, rather than synchronous, a circumstance that limits its validity.
Diagnosis
Patients infected by the agents of babesiosis or HGE (and occasionally Lyme disease) experience a flulike illness that is similar to that of other summertime illnesses, including those due to viral infections. A specific diagnosis generally requires that a thorough history be obtained and that a physical examination and appropriate laboratory tests be performed. The history should include a description of tick exposure, the kinds of tickborne infections present around the place of residence, and any sites where the patient may have traveled. Physicians should obtain a complete blood count as an aid in evaluating patients with suspected deer-associated illness. In patients with babesiosis, anemia is common; in patients with HGE, leukopenia, granulocytopenia, or lymphopenia is common; and in patients with babesiosis or HGE, thrombocytopenia is commonly seen. A specific diagnosis can be made on the basis of microscopic study of this blood smears, in the case of Babesia or Ehrlichia infection; the results of culture, in the case of Borrelia or Ehrlichia infection; PCR amplification of borrelial, babesial, or ehrlichial DNA; or identification of specific antibody in blood samples (table 2).
Table 2
Laboratory methods that allow specific diagnosis of deer-associated diseases.
Babesiosis or HGE may be diagnosed on the basis of microscopic examination of Giemsa-stained films of blood samples treated with the anticoagulant ethylenediamine tetraacetic acid [3, 67]. At least 100 fields (at magnification ×400) should be examined before declaring the sample free of piroplasms or ehrlichia. Intracellular objects suggestive of piroplasm merozoites in erythrocytes or HGE morulae in leukocytes should further be scrutinized at magnification ×1000.
Serological evidence of exposure to the Lyme disease spirochete may be detected by use of ELISA or immunoblot [105, 106]. The Centers for Disease Control and Prevention suggest a 2-step diagnostic approach: first use a sensitive ELISA, followed by a more specific immunoblot test for all serum samples with a borderline or reactive ELISA result. Serum samples tested by immunoblot generally are considered to have a positive response if the IgG immunoblot contains ⩾5 of the 10 most common B. burgdorferi-specific bands and/or if the IgM immunoblot contains ⩾2 of the 8 most common IgM bands [105]. Babesial infection may be diagnosed serologically by an indirect immunofluorescence assay [107, 108]. Serum samples are diluted 1 : 32 in PBS. The secondary antibody is goat antihuman immunoglobulin labeled with fluorescein isothiocyanate and diluted in PBS with 0.001% Evans Blue. Slides are examined by epifluorescence microscopy at magnification ×630. For comparison, each series of tests should include testing of serum samples from a subject who has had babesiosis (a “positive control”), serum from a noninfected adult (a “negative control”), and PBS. A specimen that reacts at a dilution of ⩾1 : 64 is considered to be positive for Babesia antibody. HGE infection may be diagnosed serologically on the basis of results of an immunofluorescence assay or an ELISA, as described elsewhere [109, 110].
PCR assays are available for detecting the DNA of the agents of Lyme disease, human babesiosis, and HGE [35, 111, 112]. The latter 2 agents are readily detected in whole blood from an acutely infected patient. B. burgdorferi may be detected in joint fluid or CSF but infrequently in whole blood.
Treatment
The choice of treatment for patients infected by >1 of these Ixodes-borne pathogens must be guided by the severity of the clinical course as well as the organisms implicated in the disease process.
Lyme disease. Doxycycline (100 mg twice daily) or amoxicillin (500 mg 3 times daily) for 14–21 days is the regimen of choice for early localized or early disseminated Lyme disease associated with erythema migrans, in the absence of neurological involvement or third-degree atrioventricular heart block [113]. For children <8 years of age, amoxicillin (50 mg/kg/day [maximum of 500 mg/dose], divided into 3 doses per day) is recommended. Cefuroxime axetil is recommended in the event of penicillin allergy. Erythromycin is an alternative choice if a patient cannot take doxycycline, amoxicillin, or cefuroxime. A similar regimen is recommended for uncomplicated Lyme arthritis, but the course should be for 28 days. For patients with more severe disseminated disease, including persistent or recurrent arthritis, meningitis or radiculopathy, or third-degree atrioventricular heart block, the drug of choice is ceftriaxone (for adults, 2 g once daily iv; for children, 75–100 mg/kg/d in a single iv dose [maximum of 2 g/day]) for 14–28 days. Intravenous penicillin G, cefotaxime, or doxycycline for 14–28 days may be satisfactory alternatives [113]. Tetracyclines are relatively contraindicated for women who are pregnant or breast-feeding and for children <8 years of age.
Babesiosis. The combination of clindamycin (20 mg/kg/day, ⩽600 mg every 6 h) for 7–10 days and quinine (25 mg/kg/day, ⩽650 mg every 6–8 h) for 7–10 days was first used for treating babesiosis in 1982 and has subsequently become the treatment of choice [64, 114]. This combination frequently produces untoward reactions, however, such as tinnitus, vertigo, and gastrointestinal upset. Treatment failures have been reported with clindamycin and quinine in patients who are asplenic, infected by HIV, or receiving concurrent corticosteroid therapy.
In a recently completed prospective, randomized trial, atovaquone and azithromycin was compared with clindamycin and quinine for a 7-day treatment of adults with B. microti infection [114]. The combination of atovaquone (750 mg every 12 h) and azithromycin (500 mg on day 1, then 250 mg/day thereafter) cleared parasitemia and resolved symptoms as rapidly as did the combination of clindamycin (600 mg every 6 h) and quinine (650 mg every 8 h). After 3 months, no evidence of piroplasms in thin blood smears was detected for either group, and no B. microti DNA could be amplified by PCR. Significantly fewer adverse effects are associated with the atovaquone and azithromycin combination than with the clindamycin and quinine regimen. Three-fourths of patients who received clindamycin and quinine experienced adverse drug reactions, and for one-third, the dosage had to be decreased or the medication discontinued. Adverse effects of therapy included hearing loss, tinnitus, syncope, hypotension, and gastrointestinal symptoms (anorexia, vomiting and diarrhea). By contrast, only 18% of patients in the azithromycin and atovaquone group experienced symptoms consistent with an adverse drug reaction, and 1 (2%) had to have medication diminished on day 6 of therapy. This suggested that (1) antibabesial therapy with atovaquone and azithromycin generally is superior to therapy with clindamycin and quinine, mainly because the atovaquone and azithromycin regimen is better tolerated, and that (2) physicians should consider the use of atovaquone and azithromycin in adult patients experiencing mild or moderate babesial symptoms and in others who cannot tolerate clindamycin and quinine [114]. Clindamycin (iv) and quinine should be administered to patients who are more severely ill and have intense parasitemia (>5% of erythrocytes infected), significant hemolysis, or renal or pulmonary compromise. Exchange transfusion also should be considered for such patients [115].
Ehrlichiosis. For the treatment of HGE, the drug of choice is doxycycline (for adults, 100 mg twice daily; for children, 3 mg/kg/day in 2 divided doses). Tetracycline (25 mg/kg/day in 4 equally divided doses) also has been used successfully. The treatment course should continue for a minimum of 5–7 days [67, 116]. Longer courses may be necessary for more severe or complicated disease. Untreated ehrlichiosis may be severe or fatal, and early treatment can help to reduce complications. Administration of doxycycline in children <8 years of age is controversial, because of the risk of dental staining, but doxycycline should be administered in severe cases [117, 118].
Footnotes
-
Financial support: National Institutes of Health grants AI-42402 (to P.J.K. and A.S.) and AI-19693, AI-37993, and AI-39002 (to A.S.)
- Received January 5, 2001.
- Revision received March 8, 2001.
- © 2001 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
-
- Clinical Practice
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
