CID 1999;29 (October)
Brief Reports
antibiotic exposure, however, the potential for the development of
resistance cannot be ignored. In contrast, therapy with a 1-g dose of
azithromycin would be expected to provide intracellular levels of drug
that exceed the MIC for Shigella for as long as 14 days [7]. The data
presented here expand the experience reported from Bangladesh,
where 5 days of azithromycin (total dose, 1.5 g) was as effective
against shigellosis as 5 days of ciprofloxacin [5]. Although the use of
azithromycin in the treatment of other diarrheal diseases is still under
evaluation, a single 1-g dose of azithromycin is effective and welltolerated treatment of epidemic dysentery.
References
1. Malakooti MA, Alaii J, Shanks GD, Phillips-Howard PA. Epidemic
dysentery in western Kenya. Trans R Soc Trop Med Hyg 1997;91:
541–3.
2. Shanks GD, Ragama OB, Aleman GM, Andersen SL, Gordon DM.
Azithromycin prophylaxis prevents epidemic dysentery. Trans R Soc
Trop Med Hyg 1996;90:316.
3. Inversen ER, Colding H, Petersen L, Ngetich R, Shanks GD. Epidemic
Shigella dysenteriae in Mumias, western Kenya. Trans R Soc Trop Med
Hyg 1998;92:30 –1.
4. Islam MS, Siddique AKM, Salam K, et al. Microbiological investigation
of diarrhoea epidemics among Rwandan refugees in Zaire. Trans R Soc
Trop Med Hyg 1995;89:506.
5. Khan WA, Seas C, Dhar U, Salam MA, Bennish ML. Treatment of
shigellosis: V. Comparison of azithromycin and ciprofloxacin. A
double blind, randomized, controlled trial. Ann Intern Med 1997;
126:697–703.
6. Gendrel D, Moreno JL, Nduwimana M, Baribwira C, Raymond J.
One-dose treatment with pefloxacin for infection due to multidrugresistant Shigella dysenteriae type 1 in Burundi. Clin Infect Dis
1997;24:83.
7. Dunne MW, Foulds G, Retsema J. Rationale for the use of azithromycin
as Mycobacterium avium prophylaxis. Am J Med 1997;102:37– 49.
Acute Dysentery Study Group Members
A. Caacas, A. Honnas, M. Schmit, M. Tiemessen, E. Van Eijik,
and O. Van Doom.
G. D. Shanks, B. L. Smoak, G. M. Aleman, J. Oundo,
P. G. Waiyaki, M. W. Dunne, L. Petersen, and
the Acute Dysentery Study Group*
U.S. Army Medical Research Unit–Kenya and Kenya Medical Research
Institute, Nairobi, and St. Mary’s Hospital, Mumias, Kenya; and Pfizer
Central Research, Groton, Connecticut, USA
Isolation of Legionella longbeachae Serogroup 1 from
Potting Soils in Japan
Legionella longbeachae is the major causative agent of legionnaires’ disease in Australia, which is considered by many to be the
result of use of potting soil containing the bacterium. Australian
potting soil is not sterilized during manufacture and contains a
complex mixture of bacteria and free-living amebae [1]. Elevated
storage temperatures appear to foster the intraamebic growth of
L. longbeachae in potting soil [2]. We report the first cultureproven case of L. longbeachae pneumonia in Japan and the results
of a survey of Japanese potting soil for the bacterium.
A 52-year-old gardener who had recently traveled to Guam was
admitted to a Japanese clinic because of fever, cough, and dyspnea
in 1996. L. longbeachae serogroup 1 was isolated from his sputum.
He died of pneumonia despite treatment with multiple antibiotics
including erythromycin. One-half year later, potting soil was sampled from the company where the patient had worked. Legionella
bozemanii and Legionella spiritensis, but not L. longbeachae, were
isolated from this sample.
In 1998, we surveyed 17 other Japanese potting soils for the
presence of Legionella species by using a variety of selective
media and amebic enrichment techniques. The soils were purchased from stores in Hyogo, Nagasaki, and Okinawa between
January and September 1998. The potting soils were ready-to-use
products, of which nine contained composted wood products and
Reprints or correspondence: Dr. Michio Koide, First Department of Internal
Medicine, Faculty of Medicine, University of the Ryukyus, 207 Uehara
Nishihara-cho, Okinawa 903-0125, Japan ([email protected]).
Clinical Infectious Diseases 1999;29:943– 4
© 1999 by the Infectious Diseases Society of America. All rights reserved.
1058 – 4838/99/2904 – 0042$03.00
943
Table 1.
soils.
Isolation of Legionella species from 17 Japanese potting
Legionella species isolated by
Sample
no.
Type of
sample
Direct method*
1
2
Wood
Mix
2
2
3
4
5
Wood
Wood
Mix
2
2
2
6
7
8
Mix
Wood
Wood
9
10
11
12
Wood
Mix
Mix
Wood
13
14
15
Wood
Mix
Mix
16
17
Mix
Wood
2
L. bozemanii
L. bozemanii,
Legionella species
2
L. bozemanii
2
L. micdadei,
Legionella species,
L. pneumophila
serogroup 12
L. bozemanii
L. longbeachae
L. longbeachae,
L. birminghamensis
L. bozemanii
Legionella species
Enrichment method†
L. micdadei, L. gormanii
L. cincinnatiensis, Legionella
species
2
L. longbeachae, L. bozemanii
L. bozemanii, L. micdadei
L. pneumophila serogroup 4
L. bozemanii, L. longbeachae
L. bozemanii, L. longbeachae
L. micdadei, Legionella
species
L. micdadei
L. bozemanii, L. longbeachae
L. bozemanii
L. micdadei, Legionella
species, L. pneumophila
serogroup 12
L. bozemanii
L. longbeachae
L. longbeachae
L. bozemanii, L. longbeachae
L. longbeachae, Legionella
species
NOTE. Mix 5 complex soil mixes; wood 5 composted wood products;
2 5 negative for Legionella species.
* By direct inoculation onto agar media.
†
By an amebic enrichment technique.
944
Brief Reports
eight contained a mixture of composted wood products, sand, mineral
fertilizer, and manure. Thirty-one strains of Legionella were recovered (table 1). L. longbeachae serogroup 1 was isolated from 8
samples; L. bozemanii serogroup 1, 9; Legionella micdadei, 5; Legionella pneumophila serogroup 4, 1; and L. pneumophila serogroup 12,
1. An unidentifiable Legionella species was isolated from four soil
samples; Legionella gormanii, Legionella birminghamensis, and Legionella cincinnatiensis were isolated from one soil sample each.
L. longbeachae was isolated from three composted wood soils and
five complex soil mixes.
Japanese potting soils contain multiple types of Legionella
species, most of which have been implicated in human disease. In
contrast to the Australian products, the Japanese products are
composed of different wood types. Unlike the situation in Australia, L. longbeachae pneumonia and other forms of legionnaires’
disease are rarely diagnosed in Japan, although routine diagnostic
testing for the disease is uncommonly undertaken. It is unknown if
the Japanese potting soils present any risk for legionnaires’ disease. Further investigation is required to determine the clinical and
CID 1999;29 (October)
epidemiological relevance of the presence of Legionella species in
potting soils.
Michio Koide, Atsushi Saito, Miki Okazaki, Bunichi Umeda,
and Robert F. Benson
First Department of Internal Medicine, Faculty of Medicine, University
of the Ryukyus, Okinawa, and Department of Respiratory Diseases,
Kobe City General Hospital, Hyogo, Japan; and Respiratory Diseases
Branch, Centers for Disease Control and Prevention,
Atlanta, Georgia, USA
References
1. Steele TW. Interaction between soil amoebae and soil legionellae. In:
Barbaree JM, Breiman RF, Dufour AP, eds. Legionella: current status
and emerging perspectives. Washington, DC: American Society for
Microbiology, 1993:140 –2.
2. Ross IS, Mee BJ, Riley TV. Legionella longbeachae in Western Australian
potting mix. Med J Aust 1997;166:387.
Disseminated Acanthamoeba Infection in a Patient with
AIDS: Response to 5-Fluorocytosine Therapy
Free-living amebas belonging to the genera Naegleria, Acanthamoeba, and Balamuthia are pathogenic protozoa for humans
and animals. Acanthamoeba species and Balamuthia mandrillaris
are opportunistic pathogens that cause infections involving preferentially the skin, sinuses, and CNS (granulomatous amebic encephalitis) [1]. These infections occur mainly in immunocompromised patients, particularly those with advanced HIV infection.
This report describes a patient with AIDS who had cutaneous and
sinus infections due to Acanthamoeba. The original aspect was a
protracted clinical course related to medical treatment.
A 29-year-old woman was found to be seropositive for HIV type 1
(via heterosexual transmission) in 1992. Between 1992 and 1995, her
medical history included oral candidiasis, pulmonary tuberculosis,
herpes zoster, B cell non-Hodgkin’s lymphoma of a tonsil, tonsillar
actinomycosis, and chronic sinusitis with recurrent epistaxis. In June
1995, 3 months after a trip to Thailand, she developed three cutaneous
nodules on the left thigh without fever or weight loss (her weight, 50
kg). Therapy at that time included zidovudine (400 mg/d) and
trimethoprim-sulfamethoxazole (80/400 mg/d). One month later, new
skin lesions (painful, nontender, erythematous, purplish papules and
nodules ranging from a few millimeters to several centimeters in
diameter) appeared on all four extremities. Some of the lesions became ulcerated and crusted (figure 1). Differential diagnosis included
cryptococcal dermatitis and Penicillium marneffei infection.
Laboratory studies showed a WBC count of 2.9 3 109/L and a
CD4 cell count of 2 3 106/L. No microorganisms were observed in
cultures of blood, urine, and smear of a skin ulcer. A biopsy specimen
from the cutaneous nodule was stained with periodic acid–Schiff and
hematoxylin-eosin stains. Examination revealed a polymorphous in-
Reprints or correspondence: Dr. T. Casper, Service des Maladies Infectieuses
et Tropicales-CISIH, Hôpital Gui de Chauliac, 2 Avenue Bertin Sans, 34 295
Montpellier Cedex 5, France ([email protected]).
Clinical Infectious Diseases 1999;29:944 –5
© 1999 by the Infectious Diseases Society of America. All rights reserved.
1058 – 4838/99/2904 – 0043$03.00
Figure 1. Papules and crusted lesions on the arm of a patient with
AIDS who had disseminated acanthamoeba infection.
flammatory infiltrate and necrotizing vasculitis. A multinucleated
giant cell was noted (unpublished data, A. J. Martinez). Some round
structures were observed with a dense central nucleolus, a less dense
nucleus, and a thick, well-stained wall. The presence of trophozoites
and cysts of free-living amebas was suggested.
Cultures of skin biopsy samples inoculated on nonnutriment
agar plates coated with Escherichia coli yielded marked growth of
free-living amebas, both trophozoite and cyst forms. Immunofluorescence testing of the skin biopsy samples confirmed the presence of free-living amebas of the genus Acanthamoeba (unpublished data, G. S. Visvesvara). MRI of the head showed
pansinusitis. CSF examination revealed a protein level of 0.83 g/L.
Treatment was started in August 1995 with the combination of
5-fluorocytosine (7.5 g/d intravenously) and itraconazole (400
mg/d orally). Marked improvement in the skin lesions was noted
from the first week of treatment. On the 15th day of treatment,
5-fluorocytosine was discontinued because of hematologic toxicity. After 13 days of monotherapy with itraconazole, new skin
nodules and ulcers appeared. Administration of 5-fluorocytosine