Acanthamoeba spp. in Urine of Critically Ill Patients

To the Editor: Free-living amebae are ubiquitous protists able to endure extreme temperature and pH in diverse environments (1). In past decades, interest in them increased as causes of infections, such as keratitis (2) and granulomatous encephalitis (3). Acanthamoeba spp. can harbor pathogenic microorganisms as endosymbionts, such as bacteria (e.g., Legionella spp., Pseudomonas aeruginosa, and Vibrio cholera), fungi, and mycobacteria (4). 
 
The occasional observation of amebae in urine specimens in a clinical microbiology laboratory (by L.C.S.) motivated our study. To estimate the prevalence of Acanthamoeba spp. in urine, on March 3 and 4, 2008, we collected urine samples from indwelling urinary catheters of all critically ill patients in the intensive care units of a tertiary care 2,000-bed university hospital (Hospital das Clinicas, University of Sao Paulo). Medical records were evaluated for patient age, sex, underlying diseases, length of hospital stay, use of central venous catheter, mechanical ventilation, antimicrobial drug use, and duration of urinary catheterization. 
 
Chemical urinalysis was performed by using a urine dipstick (Uriquest; Labtest Diagnostica, Lagoa Santa, Minas Gerais, Brazil), and urine leukocyte and erythrocyte counts were performed. Samples were examined microscopically and cultured for amebae. Amebae were characterized on the basis of morphologic criteria (cyst morphology and trophozoite shape and motility) (5). For ameba culture, 10 mL of urine was centrifuged at 2,500 rpm for 5 min. The supernatant was discarded, and the sediment (1 mL) was added to 5 mL of brain-heart infusion (Oxoid, Cambridge, UK). Cultures were incubated at 25oC for 48 hr and microscopically examined. For this study, finding trophozoites on direct examination or growing Acanthamoeba on culture were considered a positive result for the patient. 
 
Pipet tips (10), vacuum containers (10), plastic 15-mL tubes (10), syringes (10), glass slides (20), and tubes containing medium (10) were submitted for direct examination and culture for ameba to ensure that they were not contaminated. Urine samples were cultured for bacteria and fungi. Urine samples were submitted for bacterial and fungal direct examinations, and cultures were performed (6). If found, organisms were identified by morphologic, biochemical, and Gram stain characteristics. 
 
Data from patients with and without Acanthamoeba spp. were compared. For dichotomous variables, we used χ2 to calculate odds ratios and 95% confidence intervals. For continuous variables, the Mann-Whitney test was used. Results were significant at p<0.05. 
 
A total of 63 urine samples were evaluated; 17 (26%) were positive for Acanthamoeba spp. (Table). All samples of the control materials and medium tested showed negative results. 
 
The high prevalence of Acanthamoeba spp. in the urine of critically ill patients is difficult to explain. Although Acanthamoeba spp. can cause severe infections, amebae also carry pathogenic microorganisms (4). Bacteria may serve as food for amebae, but other interactions exist; for example, bacteria take advantage of the protection offered by amebae, especially in the cystic form (4). P. aeruginosa, Escherichia coli, and Proteus mirabilis can infect free-living amebae (7). The presence of Acanthamoeba spp. in critically ill patients may be advantageous to potentially pathogenic bacteria in the urine, protecting them against antimicrobial drugs, disinfectants, and host immunity. In this sense Acanthamoeba spp. may be a reservoir for pathogenic bacterial agents in severely ill patients or, as Khan described, a “Trojan horse for bacteria” (1). However, we found no association between the presence of Acanthamoeba spp. of bacteria and fungi in the urine. 
 
Biofilms are attractive niches for Acanthamoeba spp. that may provide nutritional requirements and protection against disinfectants and antimicrobial drugs (1). After 7 days in a patient, most urinary catheters contain a biofilm (8). However, we found no association between duration of catheterization and presence of Acanthamoeba spp. Also, P. aeruginosa isolates from clinical infections have shown more virulence toward Acanthamoeba spp. than environmental samples (9). 
 
Another possible explanation is the direct pathogenic activity of Acanthamoeba spp. In a study to determine whether ameba-associated microorganisms are a cause of nosocomial pneumonia, in 5 of 210 cases, Acanthamoeba sp. was considered the only cause of infection (10). In our study, patients positive for Acanthamoeba spp. had a higher mean and median of urine leukocytes and erythrocytes, suggesting aggression by the amebae. On the other hand, even these higher counts among positive patients are considered relatively low. Cardiovascular disease, cancer, and diabetes were associated with carriage of Acanthamoeba spp., which may occur in more severely ill patients. As a final possibility, Acanthamoeba spp. in the urine could have no role at all and may even reflect contamination during catheterization. 
 
This study has limitations. It was a small, preliminary investigation designed only to evaluate the presence of Acanthamoeba spp. in urine. However, our findings should lead to further studies to increase knowledge about the role of free-living amebae in nosocomial infections. 
 
 
 
Table 
 
Univariate analysis of variables potentially associated with Acanthamoeba spp. in urine samples from critically ill patients, Hospital das Clinicas, University of Sao Paulo, Brazil, March 2008*

48 hr and microscopically examined. For this study, finding trophozoites on direct examination or growing Acanthamoeba on culture were considered a positive result for the patient.
Pipet tips (10), vacuum containers (10), plastic 15-mL tubes (10), syringes (10), glass slides (20), and tubes containing medium (10) were submitted for direct examination and culture for ameba to ensure that they were not contaminated. Urine samples were cultured for bacteria and fungi. Urine samples were submitted for bacterial and fungal direct examinations, and cultures were performed (6). If found, organisms were identified by morphologic, biochemical, and Gram stain characteristics.
Data from patients with and without Acanthamoeba spp. were compared. For dichotomous variables, we used χ 2 to calculate odds ratios and 95% confidence intervals. For continuous variables, the Mann-Whitney test was used. Results were significant at p<0.05.
A total of 63 urine samples were evaluated; 17 (26%) were positive for Acanthamoeba spp. (Table). All samples of the control materials and medium tested showed negative results.
The high prevalence of Acanthamoeba spp. in the urine of critically ill patients is difficult to explain. Although Acanthamoeba spp. can cause severe infections, amebae also carry pathogenic microorganisms (4). Bacteria may serve as food for amebae, but other interactions exist; for example, bacteria take advantage of the protection offered by amebae, especially in the cystic form (4). P. aeruginosa, Escherichia coli, and Proteus mirabilis can infect free-living amebae (7). The presence of Acanthamoeba spp. in critically ill patients may be advantageous to potentially pathogenic bacteria in the urine, protecting them against antimicrobial drugs, disinfectants, and host immunity. In this sense Acanthamoeba spp. may be a reser-voir for pathogenic bacterial agents in severely ill patients or, as Khan described, a "Trojan horse for bacteria" (1). However, we found no association between the presence of Acanthamoeba spp. of bacteria and fungi in the urine.
Biofilms are attractive niches for Acanthamoeba spp. that may provide nutritional requirements and protection against disinfectants and antimicrobial drugs (1). After 7 days in a patient, most urinary catheters contain a biofilm (8). However, we found no association between duration of catheterization and presence of Acanthamoeba spp. Also, P. aeruginosa isolates from clinical infections have shown more virulence toward Acanthamoeba spp. than environmental samples (9).
Another possible explanation is the direct pathogenic activity of Acanthamoeba spp. In a study to determine whether ameba-associated microorganisms are a cause of nosocomial pneumonia, in 5 of 210 cases, Acanthamoeba sp. was considered the only cause of infection (10). In our study, patients positive for Acanthamoeba spp. had a higher mean and median of urine leukocytes and erythrocytes, suggesting aggression by the amebae. On the other hand, even these higher counts among positive patients are considered relatively low. Cardiovascular disease, cancer, and diabetes were associated with carriage of Acanthamoeba spp., which may occur in more severely ill patients. As a final possibility, Acanthamoeba spp. in the urine could have no role at all and may even reflect contamination during catheterization.
This study has limitations. It was a small, preliminary investigation designed only to evaluate the presence of Acanthamoeba spp. in urine. However, our findings should lead to further studies to increase knowledge about the role of free-living amebae in nosocomial infections.  To the Editor: Ranaviruses (family Iridoviridae) are emerging pathogens of farmed and wild amphibians and cause high mortality rates in these animals (1). These viruses are associated with massive population decreases of some species (2,3); outbreaks have been reported in the United States, Asia, Micronesia, and Europe. At the general meeting held by the International Epizootic Office in May 2008, iridoviruses of amphibians were added to the list of pathogens of wildlife that should be monitored (www.oie.int/aac/eng/Publicat/Cardsenglish/Ranavirus%20card_final.pdf, www.oie.int/eng/normes/fcode/en_ chapitre_2.4.2.htm, and www.jcu.edu. au/school/phtm/PHTM/frogs/other diseases-viruses.htm). We report an outbreak of ranavirus disease in amphibians in Japan.
A mass die-off of wild North American bullfrog (Rana catesbeiana) larvae was discovered in a 1,000-m 2 pond in western Japan. The die-off lasted from September 10 through October 20, 2008, with an epidemic peak on September 20, during which several thousand carcasses were collected daily. No dead adults of R. catesbeiana or other amphibian species were found. Fish (families Cyprinidae and Gobiidae) in the pond were unaffected.
Clinical signs in frogs were depression; lethargy; palpebral hyperemia; abdominal edema, petechiae, and erythema on the ventral surface; skin ulcers; limb and tail necrosis; and emaciation. Pathologic changes were similar in all larvae. At necropsy, subcutaneous edema, body cavity effusions, and swollen and friable livers were observed. Histologic examination showed extensive glomerular necrosis with renal tubular hyaline droplet degeneration (online Appendix Figure, available from www.cdc. gov/EID/content/15/7/1146-appF. htm) and various degrees of hepatic cell degeneration and necrosis. Myxosporidia were not observed within any renal tubules. Electron microscopy showed cytoplasmic ranaviruslike particles within glomerular endothelial cells. These particles were icosahedral with a diameter of ≈120 nm. Bacterial colonies were observed on the skin and within multiple organs in some larvae examined. These colonies were interpreted to be opportunistic organisms and microbial cultures were not performed.
We determined MCP DNA sequences of 1,472 nt that included the complete coding region (nt positions 17-1408, 1,392 nt) and proximal flanking regions. Sequences were deposited in the DNA Data Bank of Japan/Gen-Bank/European Molecular Biology Laboratory DNA databases under the accession no. AB474588. Phylogenetic analysis showed that virus detected in this study, designated RCV-JP, showed greater similarity to TW07-440 virus than to other ranaviruses, including tadpole edema virus (5), frog virus 3 (6), and R. catesbeiana virus Z (7). Liver tissues of fish (Gnathopogon