MAJOR ARTICLE

Severe Babesiosis in Long Island: Review of 34 Cases and Their Complications Jeffrey C. Hatcher,a Pietra D. Greenberg, Julie Antique, and Victor E. Jimenez-Lucho Division of Infectious Diseases, Department of Medicine, State University of New York at Stony Brook, Stony Brook, and Veterans Administration Medical Center, Northport, New York

Thirty-four consecutive patients were hospitalized with diagnosis of severe Babesia infection over the course of 13 years. The average time from onset of symptoms to diagnosis was 15 days. When compared with uninfected febrile control patients, affected patients complained significantly more often of malaise, arthralgias and myalgias, and shortness of breath (P ! .05 ), and they more often had thrombocytopenia and abnormal liver function (P ! .05). Forty-one percent of patients with Babesia developed complications such as acute respiratory failure, disseminated intravascular coagulation, congestive heart failure, and renal failure. Analysis of data revealed that complicated babesiosis was more commonly associated with the presence of severe anemia (hemoglobin level !10 g/dL; P p .01 ) and higher parasitemia levels (110%; P p .08 ). Patients were treated with a combination of drugs that included clindamycin, quinine, atovaquone, or azithromycin. Despite treatment, parasitemia persisted for an average of 8.5 days (range, 3–21 days). Exchange transfusion was performed for 7 patients, and it effectively reduced the high levels of parasitemia. Three patients died. Improved outcomes may result with prompt recognition and treatment of babesiosis. Babesiosis is a tickborne illness caused by the RBC parasite Babesia microti. Infection with this parasite can cause a mild syndrome with symptoms similar to those of the flu; in some cases, it can be as severe, causing sepsis and death [1]. Although most infections are relatively asymptomatic in the immunocompetent patient, severe infection can occur in patients whose spleens have been removed [2]. However, a systematic analysis of the patterns of severe babesiosis that occur in endemic areas is lacking. In the United States, the disease is centered around the East Coast areas of Massachusetts, New York, and Connecticut. More recently, babesiosis outbreaks in the

United States have also been reported in Minnesota, Wisconsin [3], California [4], and Washington State [5]. The University Hospital at Stony Brook and the Veterans Administration Medical Center at Northport are both tertiary-care centers located in Suffolk County on the eastern end of Long Island, New York, an area in which babesiosis is endemic. We performed a chart review of all patients with babesiosis admitted to our institutions over the previous 13 years to examine the clinical factors associated with severe infection and to determine the presence of complications such as renal failure, respiratory failure, or disseminated intravascular coagulation.

Received 1 March 2000; revised 7 August 2000; electronically published 26 March 2001.

PATIENTS AND METHODS

a

Present affiliation: Division of Infectious Diseases, Vanderbilt University Medical Center, Nashville. Reprints or correspondence: Dr. Victor Jimenez, HSC T15-080, State University of New York at Stony Brook, Stony Brook, NY 11794-8153 (jimenez.victor @northport.va.gov). Clinical Infectious Diseases 2001; 32:1117–25  2001 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2001/3208-0001$03.00

We reviewed the hospital records of patients with babesiosis who required hospitalization at the University Hospital at Stony Brook and the Northport Veterans Affairs Medical Center during the previous 13 years. All patients were diagnosed with Babesia infecSevere Babesiosis in Long Island

• CID 2001:32 (15 April) • 1117

1118

Table 1.

Demographics and clinical characteristics of 34 patients with severe babesiosis.

Patient

Age (y), sex

1

43, F

10

Fever, chills, arthralgias, myalgias, chest pain

Diabetes

2

35, M

7

Fever, weakness, chills

Alcoholism, b-thalassemia minor, splenectomy

3

55, M

21

Fever, fatigue, anorexia

Alcoholism

22.0

20.0

4

10,a M

3

Fever, diarrhea, anorexia

Hemophagolymphocytosis

10.0

16.0

5

66, M

7

Weakness, lethargy, fatigue, anorexia, nausea

CAD, CHF, COPD

12.0

33.2

6

50, F

21

Fever, chills, myalgias, headaches, fatigue, hematuria

Lyme disease, peptic ulcer disease

8.0

7

72, F

28

Fever, weakness, lethargy

COPD, diverticulitis splenectomy

8

27, M

3

Fever, headaches, diaphoresis

Splenectomy

9

82, M

1

Fever, chills

10

74, M

1

11

68, F

14

12

41, M

13

Time to diagnosis, d

Complaints at presentation

Comorbidity

Parasitemia, Hematocrit Persistence, Tick % level, % d bite

Lyme disease

Complications

10.0

22.0

8

N

N

DIC, ARDS

2.0

28.6

4

Y

N

ARDS

6

N

N

ARDS

21

N

?

9

N

N

18.0

6

N

Y

—

12.0

28.3

10

Y

?

—

4.2

37.6

5

Y

Y

—

COPD, CAD, polio, hypertension, colitis

5.9

27.4

4

N

Y

—

Fever, weakness, chills, myalgias

Diabetes, hypertension, CAD

4.5

30.2

3

N

Y

ARDS

Fever, lethargy, diarrhea

hypertension

1.0

27.9

9

N

N

CHF

21

Fever, chills, night sweats, myalgias, arthralgias

HIV, hepatitis, peptic ulcer, splenectomy

0.3

28.0

7

Y

N

DIC

74, M

14

Fever, weakness, fatigue, confusion

COPD, hypertension, CHF

11.0

23.0

9

N

?

—

14

77, M

10

Fever, weakness, SB, fatigue

Atrial fibrolation, CAD, aortic stenosis, hyperlipidemia

11.0

30.0

9

Y

Y

CHF; died

15

48, M

7

Fever, chills, myalgias, arthralgias

Alcoholism

20.0

23.0

15

Y

Y

DIC

16

62, M

14

Weakness, chills, anorexia

Felty syndrome, alcoholism, cirrhosis, splenectomy

30

27.0

20

N

Y

DIC, ARDS, renal failure; died

— Renal failure, CHF

17

58, M

14

Fever, fatigue, anorexia, malaise

Esophageal cancer, ethyl alcohol consumption, hypertension, splenectomy

18.0

34.0

18

N

?

18

73, M

42

SB, headache, malaise, myalgias

19

69, M

18

Fever, chills, nausea, vomiting, abdominal pain

20

85, F

10

21

50, M

21

22

77, M

23

—

CAD, COPD, CHF, G6PD deficiency

8.0

30.0

4

Y

Y

BPH, arthritis

1.0

33.6

6

Y

N

—

Mental status change, fall

Hypothyroidism

0.5

41.0

6

N

Y

—

Fever, chills, sweats, abdominal pain

Anemia

1.0

30.8

?

N

Y

21

Mental status change, fall, arthralgias

Atrial fibrolation, arterial myxoma

15.0

24.0

9

N

Y

35, M

21

Fever, chills, malaise

ITP, splenectomy

7.0

34.5

9

N

N

—

24

52, M

3

Fever, SB, fatigue, malaise, anorexia

Hereditary spherocytosis, arthritis, splenectomy

7.6

41.3

10

N

Y

—

25

3,a M

44

Fever, irritability

None

4.0

28.8

18

N

Y

—

ARDS, CHF; died

— DIC

1119

26

23, M

7

Fever, chills, myalgias, headache

Splenectomy

5.8

44.2

113

N

N

—

27

76, M

3

Fever, chills, headache, vomiting

Lyme disease, CAD anemia, splenectomy

0.7

30.7

5

N

Y

—

28

6, M

5

Fever, anorexia, nausea, vomiting

Cerebral palsy, seizure disorder

4.0

29.9

4

N

N

29

73, M

14

SB, weakness, chills, anorexia, malaise

BPH

5.6

28.5

6

N

N

30

53, F

14

Fever, chills, fatigue, myalgias, night sweats

None

3.0

19.0

4

N

N

—

31

53, M

30

Fever, chills, headache, photophobia, night sweats, myalgias

Hypothyroidism

0.1

39.5

?

N

N

—

32

17, M

4

Fever, hematuria

Hereditary, spherocytosis, splenectomy

2.0

37.4

4

N

N

—

33

52, M

35

Fever, chills, headache

Splenectomy

0.1

33.8

8

Y

Y

—

34

82, M

3

Fever, chills, CAD

COPD, hypertension

5.9

27.4

11

Y

Y

—

— DIC, ARDS

NOTE. ARDS, acute respiratory distress syndrome; BPH, benign prostatic hyperplasia; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; DIC, disseminated intravascular coagulation; HIV, human immunodeficiency virus; ITP, idiopathic thrombocytopenic purpura; N, no; SB, shortness of breath; Y, yes. a

Months.

Table 2. Characteristics and symptoms of patients with severe babesiosis and control patients.

Characteristic

Patients with babesiosis (n p 34)

Control patients (n p 44)

P

Age, y, mean  SE

53.1  4.0

45.5  2.7

.2

Sex, male/female

26/8

28/16

.2

Duration of illness before diagnosis, d, mean  SE

15.4  2.0

14.1  5.0

.8

Temperature, C, mean  SE

39.4  0.4

39.2  0.2

1.0

Chills, n, %

23 (67.6)

19 (43.2)

Malaise, n, %

18 (53.0)

5 (11.4)

!.003

Arthralgias or myalgias, n, %

14 (41.2)

2 (4.5)

!.003

8 (23.5)

7 (15.9)

Anorexia, n, % SB, n, %

.6

10 (29.4)

5 (11.4)

.04

8 (23.5)

7 (15.9)

.6

7 (20.6)

9 (20.5)

1.0

6.2  0.7

!.003

Headache, n, % Dry cough, n, % Length of stay, d, mean  SE

.1

12.7  1.2

NOTE. Values listed as mean  SE were evaluated using Student’s t test to determine P values. Values listed as percentages were evaluated using the x2 test to determine P values. SB, shortness of breath; SE, standard error.

tion, as documented by parasitemia on peripheral blood smear. Patients were examined with respect to age, sex, comorbid conditions, history of splenectomy, recollection of a tick bite, days of illness before diagnosis, length of hospital stay, complete blood count, atypical lymphocytosis, percentage of parasitemia, liver and renal functions, Lyme disease serology, treatment, complications, and outcome. A case-control study was undertaken to examine the clinical features of babesiosis that distinguish it on presentation from other febrile illnesses. Control patients were selected from a sample of the records from our hospitals for patients admitted and discharged with a diagnosis, International Classification of Diseases, 9th Revision (ICD-9) of fever of unknown origin. Case and control patients were matched for age and sex. Con-

trol patients were used if other infectious causes had been excluded, which include a peripheral blood smear examination that was negative for parasites. Retrospective review of control patients was continued until a sufficient number (1 or 2 per case patient) were selected. Data from both affected patients and control patients were compared with respect to age; sex; comorbid conditions; body temperature at admission; presence of malaise, chills, anorexia, arthralgias, or myalgias; cough; and laboratory values, including complete blood count, atypical lymphocytosis, renal function tests, and liver function tests. Patients who developed severe complications such as respiratory failure or disseminated intravascular coagulation were compared with patients without complications with regard to the following risk factors: age 160 years, hemoglobin level ⭐10 g/dL, parasitemia level 110%, delay in diagnosis of ⭓2 weeks, history of splenectomy, concomitant Lyme disease, and duration of parasitemia for ⭓10 days. Patients who underwent exchange transfusion were also analyzed with regard to the same risk factors. Statistical analysis of the data was performed by use of Student’s t, the x2, or the r 2 test, as appropriate, with use of the statistics computer program Quattro (Corel Corporation) or by manual calculation with standard formulas.

RESULTS Figure 1. Monthly admissions of patients with babesiosis (light bars) and control patients (dark bars) to any of the participating medical facilities. 1120 • CID 2001:32 (15 April) • Hatcher et al.

Of 34 patients with severe babesiosis, 30 (88.2%) had a diagnosis made before admission or transfer from another facility, and 4 (11.8%) were diagnosed after hospitalization at our fa-

Table 3. Comorbid conditions in patients with severe babesiosis and control patients.

Condition Splenectomy

Patients with babesiosis (n p 34)

Control patients (n p 44)

P

a

11 (32.4)

1 (2.3)

!.003

CAD

8 (23.5)

None

!.003

Hypertension

6 (17.6)

9 (20.5)

COPD

6 (17.6)

None

.04

Alcoholism

5 (14.7)

None

.09

Diabetes mellitus

3 (8.8)

2 (4.5)

.5

.6

Congestive heart failure

3 (8.8)

None

.04

Previous Lyme disease

2 (5.9)

None

.1

Peptic ulcer disease

2 (5.9)

None

.6

Diverticulitis

2 (5.9)

None

.6

Atrial fibrillation

2 (5.9)

None

.6

Prostate hypertrophy

2 (5.9)

None

.6

Hypothyroidism

2 (5.9)

3 (6.8)

.6

Anemia

2 (5.9)

None

.6

Hereditary spherocytosis

2 (5.9)

None

.6

b-thalassemia minor

1 (2.9)

None

.3

Polio

1 (2.9)

None

.3

HID

1 (2.9)

10 (22.7)

.01

Hepatitis B

1 (2.9)

None

.3

Gout

1 (2.9)

None

.3

Hyperlipidemia

1 (2.9)

None

.3

Aortic stenosis

1 (2.9)

None

.3

Felty’s syndrome

1 (2.9)

None

.3

Cirrhosis

1 (2.9)

None

Cancer

1 (2.9)

13 (29.5)

Cerebral palsy

1 (2.9)

None

.3

ITP

1 (2.9)

None

.3

Atrial myxoma

1 (2.9)

None

.3

G6PD deficiency

1 (2.9)

None

.3

End-stage renal disease

None

4 (9.1)

.07

.3 !.003

NOTE. CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; HID, human immunodeficiency disease; ITP, idiopathic thrombocytopenic purpura. a

P values were calculated using the x2 test.

cilities (table 1). The clinical and demographic characteristics of these patients were compared with those of patients with cases of undiagnosed febrile illness (tables 2 and 3). Eight (23.5%) infected patients were female and 26 (76.4%) were male; mean age was 53.1 years (range, 3 months to 85 years; median, 42.6 years). These parameters were not statistically different when compared with those of the control patients. The average time from onset of symptoms to final diagnosis of babesiosis was 15.4 days (range, 3–44 days; median, 23.5 days). Thirty-two percent of the patients recalled receiving a tick bite. Of the patient characteristics and symptoms, the presence of malaise, arthralgias and myalgias, and shortness of

breath were all significantly associated with Babesia infection (P ! .05), whereas other symptoms were not. Patients were evaluated by which month they were admitted to any medical facility (figure 1). Sixteen (47.1%) patients were admitted during the month of July; 10 (29.4%) in August; 4 (11.8%) in September; 2 (5.9%) in December; and 1 (2.9%) each in June and January. Only during the month of July was an admission more likely to be statistically associated with a patient with babesiosis than with a control patient (P ! .05). Thirty-two (94.1%) of the patients had an associated comorbidity. The list of concurrent medical conditions is shown in table 3. Comorbid conditions more common in the babesiosis group were as follows: splenectomy and coronary artery disease (32.3% and 23.5%, respectively; P ! .003 ); and chronic obstructive pulmonary disease and congestive heart failure (CHF; 17.6% and 8.8%; P p .04). Comorbid conditions more common in the control group were cancer (29.5%), HIV infection (22.7%), and end-stage renal disease (9.1%). Lyme borreliosis coinfection occurred in 17 (53.1%) of the 32 patients for whom serologic tests were performed, and Ehrlichia coinfection was present in 1 (patient 29, table 1). The laboratory findings are summarized in table 4. The mean WBC count was 9.0 cells ⫻109/L (median, 9.5 ⫻ 109/L; range, 1.6–17.4 ⫻ 109/L), and the mean hemoglobin level was 10.0 g/ dL (median, 10.1 g/dL; range, 5.1–15.0 g/dL), with a mean hematocrit level of 29.7% (median, 30.1%; range, 16.0%– 44.2%). These values were not significantly different between the affected patients and the control patients. The mean platelet count was 91.8 ⫻ 109/L (median, 127 ⫻ 109/L; range, 10–244 ⫻ 109/L) in the infected group, which was significantly lower than that of the control patients (P ! .003). Both the mean number of atypical lymphocytes and the proportion of patients with atypical lymphocytes were similar in the affected patients and in the control patients. Liver enzymes were significantly higher in the affected patients when compared with the control patients, with a mean serum aspartate transaminase of 120.7 U/L (median, 277.5 U/L; range, 22–533 U/L) and a mean serum alanine transaminase of 99.2 U/L (median, 272.5 U/L; range, 23–522 U/L; P ! .05). Complications of severe Babesia infection were as follows: acute respiratory failure in 7 (20.6%) patients, disseminated intravascular coagulation in 6 (17.6%), CHF in 4 (11.8%), and renal failure in 2 (5.9%). In the acute respiratory failure subgroup, the mean Po2 was 46.9 mm Hg (median, 55.5 mm Hg; range, 39–72 mm Hg), and the mean respiratory rate was 27.3 breaths/min (median, 26.5 breaths/min; range, 21–32 breaths/ min). The disseminated intravascular coagulation subgroup was characterized by a mean prothrombin time of 15.5 s (median, 17.1 s; range, 11.6–22.9 s; normal range, 10.1–12.7 s); mean partial tissue thromboplastin time, 40.4 s (median, 47.1 s; range, 29.7–64.4 s; normal range, 23–38 s); mean fibrin split products Severe Babesiosis in Long Island • CID 2001:32 (15 April) • 1121

Table 4. Laboratory findings for patients with severe babesiosis and for control patients.

Finding

Patients with a babesiosis (n p 34)

WBC count ⫻ 109cells/L

a

Control patients (n p 44)

P

b

9.0  1.0

7.6  0.8

.3

Hemoglobin level, g/dL

10.0  0.4

11.4  0.3

.06

Hematocrit level, %

29.7  1.1

33.2  1.0

Platelet count, ⫻ 109 cells/L

91.8  9.8

187.7  14.9

c

24.8  3.5

Urea level, mg/dL

.06 .9

1.3  0.2

1.5  0.3 50.2  13.6

99.2  17.1

51.8  7.3

Alanine aminotransferase, U/L

.7

18.8  1.9

120.7  22.7

Creatinine level, mg/dL Aspartate aminotransferase, U/L

2.5  0.6b

2.5  0.5

Atypical lymphocytes, %

.2 !.003

.003 !.003

a

Mean  SD. Means were evaluated using Student’s t test to determine P values. Percentages were evaluated using the x2 test to determine P values. c Twenty-one (75%) of 28 patients with babesiosis, and 16 (48.5%) of 33 control patients; P p .06. b

of lower limit 152 mg/mL and upper limit !108 mg/mL (normal range, !10 mg/mL); and mean fibrinogen, 473 mg/dL (median, 481 mg/dL; range, 342–620 mg/dL; normal range, 200–400 mg/ dL). The patients who developed complications were analyzed against those who did not develop complications with respect to the following risk factors (table 5): age 160 years, hemoglobin concentration of ⭐10 g/dL, parasitemia level of ⭓10%, delay in diagnosis of ⭓2 weeks, history of splenectomy, concurrent Lyme infection, and duration of parasitemia for ⭓10 days. When examining patients with medical complications as a whole, only severe anemia (defined as a hemoglobin level ⭐10 g/dL) was associated with the presence of complicated babesiosis (P p .01). Although a parasitemia level 110% was also associated with complicated babesiosis, this association did not reach statistical significance (P p .08; table 5). When the individual complications were analyzed separately with the above criteria, a parasitemia level of 110% was associated with an increased risk of disseminated intravascular coagulation, although this association did not reach statistical significance (P p .08; table 6). None of the examined criteria were associated with an increased risk of development of acute respiratory failure. Because of the small number of cases of CHF and renal failure, they were not analyzed separately. However, 2 of the 3 patients with a history of CHF developed worsening of this condition during their acute illness. Renal failure occurred in 2 of the infected patients, 1 of whom required dialysis. The mean peak parasitemia was 7.6% (median, 15.1%; range, 0.1%–30%; figure 2). Parasitemia persisted for an average of 8.5 d (median, 12 d; range, 3–21 d) after initiation of treatment (complete data were available for 30 patients). Correlation sta1122 • CID 2001:32 (15 April) • Hatcher et al.

tistics was performed to determine whether the level of parasitemia correlated with the degree of anemia. The r 2 value was 0.14, which shows a poor correlation between the level of parasitemia and the degree of anemia. Treatment of babesiosis was with clindamycin in 33 (97.1%) patients; quinine, 28 (82.3%); atovaquone, 15 (44.1%); and azithromycin, 2 (5.9%). Eight (23.5%) patients had an adverse reaction to quinine. Exchange transfusion was performed in 7 (20.6%) patients. The effect of exchange transfusion in rapid reduction of the parasitemia in 1 patient is shown in figure 3. The mean parasitemia in the exchange transfusion group was 16% (median, 20%; range, 10%–30%). The exchange transfusion group was analyzed with the complication risk factors listed previously (table 7). Of the risk factors, only parasitemia ⭓10% was significantly associated with exchange transfusion (P ! .003). Table 5.

Factors associated with complicated babesiosis. Complicated babesiosis, no. (%) of patients

Factor

Yes

No

Pa

Age 160 y

8/13 (62)

7/21 (33)

.1

Hemoglobin level !10 g/L

8/12 (67)

6/20 (30)

.01

Parasitemia level ⭓10%

7/13 (54)

5/21 (43)

.08

Parasitemia for ⭓10 d

2/13 (15)

9/21 (43)

.1

⭓2 weeks for diagnosis

7/13 (54)

11/21 (52)

.9

Previous splenectomy

3/13 (23)

8/21 (38)

.2

Lyme disease

8/13 (62)

9/19 (47)

.9

NOTE. One patient record did not contain the hematocrit level, and not all patients had been tested for Lyme disease. a

P value was calculated using the x2 test.

Table 6. Factors associated with disseminated intravascular coagulation (DIC). DIC, no. (%) of patients Yes

No

Pa

Age 160 y

3/6 (50)

12/28 (43)

.8

Hemoglobin level !10 g/L

4/6 (67)

11/27 (41)

.3

Parasitemia level ⭓10%

4/6 (67)

8/28 (29)

.08

Parasitemia for ⭓10 d

2/6 (33)

7/28 (25)

.7

⭓2 weeks for diagnosis

4/6 (67)

14/28 (50)

0.5

Previous splenectomy

2/6 (33)

9/28 (32)

1.0

Lyme disease

4/6 (67)

13/26 (50)

.4

Factor

NOTE. One patient record did not contain the hematocrit level, and not all patients had been tested for Lyme disease. a

P value was calculated using the x2 test.

The mean length of hospital stay in the infected patients was 12.7 days (median, 18 days; range, 3–33 days) and in the control patients was 6.2 (P 1 .05). Despite severe disease, outcome in the infected patients was recovery in 31 (91.2%) patients and death in 3 (8.8%). The death of 1 patient was attributed to acute myocardial infarction; and for the other 2, the cause of death was not determined (no postmortem examinations were performed). It is of note that parasitemia in these 3 patients before their deaths was 0%, 1%, and 0.1%, respectively. DISCUSSION Babesiosis can be a severe and life-threatening infection, and the diagnosis may be difficult, even in an area in which babesiosis is endemic. This study was performed in 2 referral centers located in eastern Long Island, an area of endemicity. In this series, the average number of days of illness before the diagnosis was 15.4; and the mean length of hospital stay was long (13 days). The diagnosis was delayed because of the nonspecific clinical findings of babesiosis. Analysis of our data suggests that subtle abnormalities may help the clinician differentiate Babesia infection from other febrile illnesses. Constitutional symptoms such as malaise, arthralgias and myalgias, and shortness of breath were significantly more common in patients with babesiosis compared with other febrile control patients. Among the laboratory findings, a normal WBC count and mild to moderate anemia were observed, which reinforced the findings of previous research [1]. Thrombocytopenia and abnormal liver enzymes were also significantly more common in patients with Babesia infection. In contrast to a previous study [6], the presence of atypical lymphocytes was not significantly more common in infected patients. Babesiosis has traditionally been considered to be an infection acquired during the summer. However, in our study, only during the month of July were the admitted patients more likely

to have babesiosis. Recent studies have shown that Babesia can persist asymptomatically for long periods [7] and can be acquired from sources other than ticks, such as blood transfusions [8]. Therefore, in endemic areas, babesiosis should be considered and investigated appropriately even during months not typical for increased tick activity. As previously reported, severe babesiosis can occur in patients after splenectomy [2]. In our series, splenectomy was present in about a third of the patients. However, other comorbid conditions seemed to contribute to increased severity of illness. Concomitant coronary artery disease, chronic obstructive pulmonary disease, alcoholism, or CHF were commonly present in patients with severe babesiosis. Also 53% of the patients had concurrent Lyme infection. As previously noted [9], coinfection with Lyme disease may lead to increased symptomatology and prolonged duration of illness. Our series is unique in that 41% of the patients developed complications such as acute respiratory failure, CHF, disseminated intravascular coagulation, and renal failure. We attempted to determine which clinical factors were associated with the presence of complicated babesiosis. We found that severe anemia, defined as a hemoglobin level of ⭐10 g/dL, was significantly associated with the presence of a complication (P p .01). In addition, a high parasitemia level (110%) was associated with the presence of a complication, although this association did not reach statistical significance (P p .08), probably because of the small sample size. A poor correlation between the degree of anemia and the level of peripheral blood parasitemia was noted. In the subgroup analysis, a high level of parasitemia (110%) was also associated with the presence of disseminated intravascular coagulation (P p .08 ), but it was not associated with acute respiratory failure. Although the pathogenesis of complicated babesiosis in humans is unknown at present, these findings support the view that factors other than parasitemia

Figure 2. Time course of peripheral blood parasitemia during treatment in patients with babesiosis. Severe Babesiosis in Long Island • CID 2001:32 (15 April) • 1123

Figure 3. Effect of exchange blood transfusion in the peripheral blood parasitemia of a patient infected with Babesia.

itself may play a role. For instance, malarial disease pathogenesis is now presumed to be associated with an excessive production of proinflammatory cytokines such as TNF and IL-1. Interestingly, a similar range of complications has been noted in cases of babesiosis in rodents, cattle, and dogs—species that share an innate susceptibility to bacterial endotoxin [10]. In addition, macrophages obtained from Babesia-infected mice have been shown to be unusually primed for cytokine release when exposed to endotoxin [11]. Whether a similar process occurs in human babesiosis remains to be determined. Interestingly, other factors such as older age, delay in diagnosis of ⭓2 weeks, history of splenectomy, concurrent Lyme disease, and long duration of parasitemia (110 days) were not significantly associated with the presence of disseminated intravascular coagulation or acute respiratory failure The role of the comorbid conditions in this population is also suggested by the high frequency of worsening CHF in patients with a history of CHF. It is likely that this reflected the use of high-volume infusions in these patients during treatment of babesiosis.

Treatment with clindamycin and quinine was administered to 82.4% of the patients. However, nearly one-quarter of the patients had an adverse reaction to quinine. Atovaquone was usually added because of intolerance to other drugs, and 2 affected children (patients 4 and 25, table 1) were treated with azithromycin. Both of these drugs have been previously shown to have significant antibabesial activity [12, 13]. However, persistence of parasitemia for ⭓10 days occurred in 23.5% of patients after treatment was begun. In accord with our data, babesial DNA has been noted to persist in infected patients for up to 16 days after treatment [9]. Thus, there is need for more efficacious and less toxic therapies for babesiosis to be developed. Exchange transfusion has been used effectively in the past as an adjunct for treatment. In our study, only the presence of high parasitemia ⭓10%) was significantly associated with the likelihood of exchange transfusion, and the other risk factors examined were not. This probably reflects our clinicians’ preference for the treatment of high levels of parasitemia at our institutions. Although the frequency of complications was not significantly different in patients who underwent exchange transfusion, it is possible that more complications could have been prevented if exchange transfusion had been performed earlier in the clinical course. Further clinical research will need to be done to determine the efficacy of this modality of treatment. In addition, it should be noted that several patients received blood transfusions (patients 22, 29, and 30; table 1) along with antimicrobial therapy, and this adjunct treatment remains an option in areas where exchange transfusion is not available. Caution must be used when interpreting this analysis because of its retrospective nature and the relatively small sample size. However, this study reinforces the severity of complicated babesiosis and the clinical factors associated with it.

References Table 7. (ET).

Factors associated with exchange transfusion

ET, no. (%) of patients Factor

Yes

No

Pa

Age 160 y

3/7 (43)

11/27 (41)

.4

Hemoglobin level !10 g/L

5/7 (71)

10/26 (38)

.5

Parasitemia level ⭓10%

7/7 (100)

5/27 (19)

Parasitemia for ⭓10 d

3/7 (43)

6/27 (22)

.3

⭓2 weeks for diagnosis

4/7 (57)

14/27 (52)

.8

Previous splenectomy

3/7 (43)

8/27 (30)

.5

Lyme disease

3/5 (60)

14/27 (52)

.7

!.003

NOTE. One patient record did not contain the hematocrit level, and not all patients had been tested for Lyme disease. a

P value was calculated using the x2 test.

1124 • CID 2001:32 (15 April) • Hatcher et al.

1. Gelfand JA. Babesia. In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas and Bennett’s principles and practice of infectious diseases. 4th ed., Vol 2. New York: Churchill Livingstone, 1995:2497–500. 2. Rosner F, Zarrabi MH, Benach JL, Habichi GS. Babesiosis in splenectomized adults: a review of 22 reported cases. Am J Med 1984; 76: 696–701. 3. Steketee RW, Eckman MR, Burgess EC, et al. Babesiosis in Wisconsin: a new focus of disease transmission. JAMA 1985; 253:2675–8. 4. Persing DH, Herwaldt BL, Glaser C, et al. Infection with a Babesialike organism in northern California. N Engl J Med 1995; 332:298–303. 5. Quick RE, Herwaldt BL, Thomford JL, et al. Babesiosis in Washington State: a new species of Babesia? Ann Intern Med 1993; 119:284–90. 6. Rosenbaum GS, Johnson DH, Cunha BA. Atypical lymphocytosis in babesiosis. Clin Infect Dis 1995; 20:203–4. 7. Krause PJ, Speilman A, Telford SR, et al. Persistent parasitemia after acute babesiosis. N Engl J Med 1998; 339:160–5. 8. Dobroszycki J, Herwaldt BL, Boctor F, et al. A cluster of transfusionassociated babesiosis cases traced to a single asymptomatic donor. JAMA 1999; 281:927–30.

9. Krause PJ, Telford SR, Speilman A, et al. Concurrent Lyme disease and babesiosis: evidence for increased severity and duration of illness. JAMA 1996; 275:1657–61. 10. Clark IA, Jacobson LS. Do babesiosis and malaria share a common disease process? Ann Trop Med Parasitol 1998; 92:483–8. 11. Wood PR, Clark IA. Macrophages from Babesia and malaria infected mice are primed for monokine release. Parasite Immunol 1984;

6:309–17. 12. Wittner M, Lederman J, Tanowitz HB, Rosenbaum GS, Weiss LM. Atovaquone in the treatment of Babesia microti infections in hamsters. Am J Trop Med Hyg 1996; 55:219–22. 13. Weiss LM, Wittner M, Wasserman S, Oz HS, Retsema J, Tanowitz HB. Efficacy of azithromycin for treating Babesia microti infection in the hamster model. J Infect Dis 1993; 168:1289–92.

Severe Babesiosis in Long Island • CID 2001:32 (15 April) • 1125