The trusted source for
healthcare information and
Synopsis: Studies in animals and humans suggest that atovaquone, administered in combination with a second agent, may be a safe and effective alternative to the standard clindamycin and with quinine for treatment of babesiosis.
A 33-year-old previously healthy phd molecu-lar biologist traveled to Madison, Wis, for a work-related function in late May 2000. Toward the end of her trip, she took a few days off to go hiking and camping about 1 hour north of the city. She returned to California on May 25 and within 8 days developed a high fever, headache, photophobia, neck stiffness, and swollen inguinal and axillary lymph nodes. She was seen in the emergency room, where she was found to be significantly orthostatic and was given parenteral fluids. Laboratory studies demonstrated mild leukopenia and elevations in hepatic transaminases. The presumed diagnosis was a "viral syndrome." During the next week, she continued to feel poorly with fevers to 103°F, and erythematous macules began to appear on her breasts, torso, and buttocks.
After surfing the internet, she approached her primary care physician (PCP) about the possibility of Lyme disease. Her PCP assured her that the rash did not resemble that of Lyme disease, but agreed to send an indirect immunofluorescence titer on June 8, which proved positive and was confirmed by Western blot. Before the results were available, however, the patient presented to an urgent care clinic on June 18 with an increasing number (~20) of pink and violaceous macules ranging in size from 1 to 6 cm, some with central clearing. She was treated with empiric doxycycline 100 mg twice daily for 3 weeks. Within 12 hours of the first dose, she experienced fever, rigors, and severe myalgias, which rapidly resolved.
After further surfing on the net, the patient approached her PCP regarding the possibility of co-infection with babesia. IFAs obtained on June 22 for B microti IgG and IgM were positive at 1:64 and 1:512. The patient was referred to an ID specialist for further evaluation. She was seen in consultation on July 18, at which time she was complaining of persistent fatigue and malaise, mild sweats, mild shortness of breath, a dry cough, and weight loss. The leukopenia had improved, the hepatic transaminases had normalized, a chest radiograph was unremarkable, and a direct smear for babesia was negative. But a PCR of whole blood for B microti obtained on August 8 was positive! The patient was reluctant to take quinine or clindamycin.
What Would You Do?
Comment by Carol A. Kemper, MD, FACP
Based on the patient’s desires for treatment, but also her valid concerns regarding drug toxicity (she was remarkably well informed), it was elected to treat her with a combination of quinine and azithromycin. Within 24 hours, she reported acute bilateral hearing loss, which quickly resolved with discontinuation of therapy. After further discussion, she decided to try atovaquone as a single agent 750 mg twice daily for 2 weeks. Two weeks after discontinuing treatment, a repeat PCR was negative, and the patient was feeling improved. Three months later, serologies for Lyme showed resolving infection, and babesia titers were less than 1:16.
Babesiosis is due to a malaria-like protozoan that parasitizes the erythrocytes of both wild and domesticated animals, birds, and humans.1,2 More than 100 species of Babesia have been identified, although only the rodent strain, B microti (in the United States), and the cattle strains B divergens and B bovis (in Europe), are known to infect humans. Cases of babesiosis in humans were first recognized in 1960-1970 along the northeastern coast of the United States, in places like Nantucket Island, Mass, New York, and Rhode Island, but infections have also been reported in recent years in Virginia, Georgia, Wisconsin, Minnesota, California, Washington, and Mexico.
Because the arthropod vector for Lyme disease, Ixodes dammini, also serves as the vector for B microti in the United States, co-infection with Borrelia burgdorferi (as well as Ehrlichia) may occur, as was the case here. (Serologies for Ehrlichia were not obtained.)
Most cases of babesiosis are generally mild, and subclinical infection in endemic areas is probably not uncommon. Approximately 3-7% of individuals living in endemic areas along the northeastern seaboard have serological evidence of prior infection. While most cases of babesiosis are self-limited, the infection can present as a fulminant, malaria-like illness with hypotension and circulatory collapse, and may be fatal in up to 5% of clinically apparent cases. Clinically apparent infections are more likely in older persons, asplenic individuals, and individuals with underlying disease. HIV-infected persons have been reported to be at greater risk for more frequent and more severe infection.3 However, 2 of the reported cases of babesiosis in HIV-infected individuals were also asplenic.4,5
Common laboratory findings include anemia, leukopenia, and abnormal hepatic transaminases. Hemolytic anemia has been reported, especially in splenectomized patients. With the exception of acute respiratory distress syndrome in patients with circulatory collapse,2 respiratory disease is not reported, although this patient complained of persistent dry cough without evidence of radiographic disease that resolved with treatment. The incubation period is typically 1-3 weeks, but the onset of symptoms can be delayed for up to 6 weeks after infection. Based on the evidence, it is more likely this patient was suffering from acute babesiosis than Lyme disease at initial presentation.
The diagnosis can be made by examination of thin blood smears, indirect immunofluorescent antibody tests, or PCR. Serologies may be positive for as long as 1-6 years following acute infection and are generally not useful for tracking response to therapy. PCR-based assays specific for B microtic and B divergens are now available. A positive PCR for babesial DNA is believed to be consistent with ongoing replication and active infection, even in the absence of a positive smear, although the significance of this finding, especially in asymptomatic persons, is not known. Wild and domestic animals have been known to harbor babesial parasites for long periods of time, and a similar "chronic carrier state" may also exist in humans. Transmission of the organism through contaminated blood products from unsuspecting hosts is not infrequent in endemic areas, and persistent low-grade parasitemia has been reported in asymptomatic persons for months to years. Patients with immunodeficiency or HIV infection may be more likely to develop persistent parasitemia.
Whether patients who are asymptomatic, but with evidence of persistent residual infection by PCR alone, should receive treatment is not known, as many of these patients should successfully clear their infection on their own. A combination of clindamycin plus quinine for 1 week has been recommended in patients with acute infection, especially those with severe disease, although this regimen is poorly tolerated. Exchange transfusions have been used in life-threatening cases.
Recent animal data suggest that atovaquone has significant activity against Babesia, although it may not be sufficiently active as a single agent to eradicate infection. Atovaquone successfully suppressed B divergens infection in gerbils with as little as 1.0 mg/kg, with increasing effectiveness in dosages up to 10 mg/kg.6 However, 10 days of treatment in a second study, even at higher dosages, failed to sterilize animals.7 A combination of atovaquone and clindamycin was more effective at suppressing parasitemia but also failed to eradicate infection. Clindamycin and quinine were less effective than atovaquone alone when administered to hamsters before the animals were inoculated.8
Studies in cats experimentally infected with Babesia felis suggest that primaquine, as well as a combination of rifampicin and sulfadiazine-trionethoprim, may also be effective in suppressing infection, although buparvaquone, enrofloxacin, and danofloxacin had no significant effect.9
The emergence of resistance during treatment with atovaquone may also occur. Recrudescent parasitemia occurred in hamsters following the administration of atovaquone as a single agent.10 When a blood specimen from hamsters with recrudescent parasitemia was inoculated into a subsequent series of animals, the second set of animals no longer responded to the administration of atovaquone. Better results were seen using a combination of atovaquone (100 mg/kg/d) plus azithromycin (150 mg/kg/d).
In humans, the combination of atovaquone and azithromycin appears to be similarly effective, but better tolerated, than the standard recommended regimen of quinine and clindamycin.11 A total of 58 patients were randomized to receive either orally administered atovaquone (750 mg bid) plus azithromycin (500 mg on day 1 followed by 250 mg on subsequent days; n = 40), or quinine (650 mg tid) plus clindamycin (600 mg tid; n = 18) for 1 week. Fever resolved within 7-8 days in both groups, and other symptoms had largely resolved within 3 months of treatment (65% of those receiving atovaquone plus azithromycin vs 73% of those receiving the alternate regimen). Three months post-therapy, smears and PCRs were negative in all subjects. Overall, the combination of atovaquone plus azithromycin was much better tolerated, resulting in side effects in only 15% of subjects vs. 72% of those receiving clindamycin plus quinine. The most common side effects to atovaquone plus azithromycin were diarrhea (8%), rash (8%) and vertigo (2%), compared with tinnitus (39%), diarrhea (33%), decreased hearing (28%), and vertigo (17%) in patients receiving clindamycin plus quinine.
Patients with life-threatening illness were excluded from this study, but successful treatment of a critically ill, splenectomized patient with azithromycin and atovaquone and exchange transfusion has been reported.12
Based on the improvement in the biologist’s symptoms and laboratory studies before the initiation of treatment, her infection would have undoubtedly resolved without treatment. But her improvement with therapy suggests that she benefitted from effective suppression of her infection. Based on animal data, the administration of atovaquone as a single agent may be sufficient for milder cases or cases similar to ours (eg, smear-negative/PCR-positive). However, a combination of atovaquone, plus either azithromycin or quinine, should probably be used in cases with higher grades of parasitemia and more severe symptomatology.
1. Homer MJ, et al. Clin Microbiol Rev. 2000;13:451-469.
2. Kemper CA. Pulmonary disease in selected protozoal infections. In: Sarosi GA, ed. Seminars in Respiratory Medicine. Philadelphia, Pa: WB Saunders Company; 1997:12(2):113-121.
3. Benezra D, et al. Ann Intern Med. 1987;107:944.
4. Machtinger L, et al. J Clin Apheresis. 1993;8:78-81.
5. Ong KR, et al. Lancet. 1990;336:112.
6. Pudney M, Gray JS. J Parasitol. 1997;83:307-310.
7. Pudney M, Gray JS. J Parasitol. 1999;85:723-728.
8. Hughes Wt, Oz HS. J Infect Dis. 1995;172:1042-1046.
9. Penzhorn BL, et al. J S Afr Vet Assoc. 2000;71:53-57.
10. Wittner M, et al. Am J Trop Med Hyg. 1996;55: 219-222.
11. Krause PJ, et al. N Engl J Med. 2000;343:1454-1458.
12. Bonoan JT, Johnson Dh, Cunha BA. Heart Lung. 1998; 27:424-428.