Human Infection with Rickettsia sp. related to R. japonica, Thailand

To the Editor: Although rickettsioses caused by scrub typhus and typhus group rickettsiae are well recognized in Thailand, few spotted-fever group rickettsiae (SFGR), including Rickettsia honei TT118 and R. felis, have been documented to be associated with human illnesses (1,2). We report a case of human infection with an SFGR species closely related to R. japonica in Thailand. 
 
In January 2005, a 36-year-old man with prolonged fever, pneumonia, and septic shock was transferred from a private hospital to Phramongkutklao Army Hospital in Bangkok. Two weeks before the onset of fever, the patient had camped at Khao Yai National Park, ≈175 km northeast of Bangkok. The park is a popular location for tourists and the largest national park declared as a natural wildlife reserve area. The patient reported the presence of wild deer around the camping area but did not recall being bitten by an arthropod. Ten days before hospitalization, he developed flulike symptoms, fever, and sore throat. Six days later, he noted petechiae on his lower extremities, and his condition worsened. At the time of hospital admission, the patient had fever of 38.6°C, tachycardia, dyspnea, hypotension, nausea, vomiting, generalized maculopapular rash, and subconjunctival hemorrhage. Laboratory investigation showed thrombocytopenia (platelets 64,000/mm3), leukocytosis (14,000/mm3), and elevated levels of serum hepatic enzymes (aspartate aminotransferase 287 IU/L [reference 5–50 IU/L]; alanine aminotransferase 186 IU/L [reference 5–40 IU/L]). Chest radiograph showed interstitial pneumonitis. Serum antibody test results were negative for leptospira and dengue virus; blood smear was negative for malaria. 
 
Samples of the patient’s whole blood were collected in EDTA on days 10, 18, 20, and 25 after illness onset, and each sample was sent at the time of collection to the Armed Forces Research Institute of Medical Sciences, Bangkok, to be investigated for rickettsial infection. Plasma was separated and tested for scrub typhus, typhus group, and SFGR-specific immunoglobulin M (IgM) and IgG by immunofluorescence assay by using Orientia tsutsugamushi Karp-Kato-Gilliam strains and R. typhi Wilmington and R. honei TT118 whole cell antigens. No antibodies to rickettsiae were detected in the initial sample. On day 18, only antibodies against R. honei TT118 antigen were detected at a low titer, 50 for IgM and 200 for IgG, while antibodies to scrub typhus and typhus group rickettsiae remained negative (titers <50). Antibody level was unchanged on days 20 and 25. 
 
At the time of admission, the patient began receiving 2 g of intravenous ceftriaxone and 200 mg of oral doxycycline daily. Three days later, treatment with doxycycline was stopped because the initial serologic results for rickettsia were negative. However, doxycycline was resumed on day 21, after antibodies to Thai tick typhus agent were detected in a second specimen. Within 3 days, the patient was afebrile and asymptomatic. He was discharged from the hospital and continued oral doxycycline for an additional 7 days. At 2-week follow up, he had completely recovered. 
 
To identify which SFGR was responsible for the patient's illness, we used molecular approaches. We extracted DNA from the patient’s blood specimens by using QIAamp Mini blood kit (QIAGEN, Valencia, CA, USA) and subjected it to duplex nested–PCR assays targeting a 343-bp fragment of the rickettsial genus–specific 17-kDa antigen gene (3) and a 690 bp-portion of the Orientia 56-kDa antigen gene (4). An appropriate control panel included DNA from a reference sample of human blood, Coxiella sp., and Leptospira interrogans. Platinum Taq DNA Polymerase High Fidelity (Invitrogen, Carlsbad, CA, USA) enzyme mixture was used in PCR. By resolution on agarose gel, a PCR fragment of the expected size for the 17-kDa antigen gene was observed from the day-10 sample but not from the control samples. AluI restriction pattern of amplified 17-kDa fragment was similar to that of SFGR. Additional rickettsial gene fragments, 630 nt-ompA (nt 70–701) and 945 nt-gltA (RpCS.193F-5′-GTAGGGTATCTGCGGAAGCC-3′, RpCS.1143R-5′-GAGCGAGAGCTTCAAGTTCTATTGC-3′), were also amplified from the day-10 specimen. All amplicons were excised from agarose gels, purified by QIAEX II Gel Extraction Kit (QIAGEN), and then sequenced. BLAST analysis of 17-kDa antigen gene (GenBank accession no. {"type":"entrez-nucleotide","attrs":{"text":"DQ909071","term_id":"115184157","term_text":"DQ909071"}}DQ909071), gltA ({"type":"entrez-nucleotide","attrs":{"text":"DQ909073","term_id":"115184188","term_text":"DQ909073"}}DQ909073), and ompA ({"type":"entrez-nucleotide","attrs":{"text":"DQ909072","term_id":"115184175","term_text":"DQ909072"}}DQ909072) segments obtained from this patient showed 99% identity to corresponding genes of R. japonica. Phylogenetic analysis of these 3 genes indicates that the Rickettsia sp. from this patient is closely related and clustered within the same clade of R. japonica (Figure). Isolation of this rickettsial agent from the patient’s blood by animal inoculation and by cell culture methods is ongoing. 
 
 
 
Figure 
 
Phylogenetic relationships between Rickettsia sp. and rickettsial genes amplified from the patient (PMK 94) inferred from comparison with the rickettsial 17-kDa antigen gene, gltA, and ompA sequences by the neighbor-joining method. Bootstrap values of ... 
 
 
 
Persons visiting Khao Yai National Park are at risk for rickettsioses, particularly SFGR. Vectors for SFGR have been found in this area (5). The clinical and molecular findings in this case add to the accumulating data on the emerging rickettsial agents and their geographic distribution in Thailand.


Human Infection with Schineria larvae
To the Editor: Myiasis remains prevalent worldwide (1,2) and is infestation by larvae from fly species of live or dead tissues from vertebrate hosts (1,3,4). In humans, myiasis most frequently causes infection of exposed ulcers or traumatic wounds (1). In industrialized countries, most cases occur in tourists returning from tropical and subtropical areas (5,6), but autochthonous cases still exist. Several bacterial species have been associated with fly larvae, including species of the family Enterobacteriaceae and, more recently, Schineria larvae (7,8). S. larvae, a gram-negative bacterium, has been grown from larvae of Wohlfahrtia magnifica, a fly species responsible for myiasis (7,8). Its 16S rRNA gene has been amplified from a bacterial community of species involved in aerobic thermophilic bioprocesses (9). We report a case of S. larvae bacteremia in a man with wound myiasis.
On June 12, 2006, a 76-year-old man who had type 2 diabetes mellitus was examined at the emergency All material published in Emerging Infectious Diseases is in the public domain and may be used and reprinted without special permission; proper citation, however, is required. department of Drôme North Hospitals, Romans, France, for inflammation of chronic cutaneous ulcers of both legs and intermittent fever. The patient lived alone in a rural, crowded area and had received no medical care. He reported owning sheep and denied any recent travel outside France. At the time of admission, his body temperature was 37.8°C, he was malodorous, and he had swelling and painful wounds on both legs. Maggots were found in the leg wounds, scrotum ulcers, and at the anal margin. A radiographic examination of both legs did not show any osteolytic lesion. Laboratory data were as follows: C-reactive protein 71 mg/L (reference value, <5 mg/L), leukocyte count 18.2 × 10 9 /L (81% granulocytes), platelet and erythrocyte counts within normal limits, glucose 200 mg/dL, hemoglobin A1c level 13.8%. Serum protein electrophoresis showed hypoalbuminemia (20 g/L) and hypergammaglobulinemia (16.9 g/L) but no monoclonal gammopathy. Two blood samples and exudate from the leg wounds were collected for microbial cultures.
The patient was given a combination of amoxicillin-clavulanate and ofloxacin, and his cutaneous wounds were cleaned. After 24 h of incubation, leg wound cultures grew methicillin-susceptible Staphylococcus aureus, and the 2 blood cultures yielded the same S. aureus strain and an oxidative gram-negative bacterium (Romans strain). The Romans strain was found to be highly susceptible to β-lactams, aminoglycosides, chloramphenicol, cotrimoxazole, fluoroquinolones, and colistin. The antimicrobial drug therapy was changed to oxacillin and ofloxacin. The patient's condition improved rapidly, and his leg wounds healed progressively during hospitalization. He was discharged 27 days after initiation of antimicrobial therapy, which he continued for 7 more days. The patient was reexamined 1 month later and was considered cured.
The Romans strain was sent to the bacteriology laboratory at Grenoble University Hospital for identification. Using the API 20NE and Vitek II ID-GNB systems (bioMérieux, Marcy L'Etoile, France), we obtained, respectively, a "good" identification of Psychrobacter phenylpyruvicus and a "very good" identification of Oligella ureolytica. The nearly complete 16S rRNA gene of the Romans strain was amplified and sequenced; primers were Fd1 and rp2 (10) (GenBank accession no. EF120377). A BLAST search that used the network service of the National Center for Biotechnology Information (www. ncbi.nlm.nih.gov) showed 99.6% identity between the determined gene sequence of 16S rRNA and that of S. larvae type strain L1/68 T (accession no. AJ252143). The 16S rRNA gene sequences from several species belonging to the Gamma Proteobacteria order were aligned by using the ClustalW package (www.ncbi. nlm.nih.gov/). A consensus phylogenetic tree was constructed from Jukes-Cantor evolutionary distances based on the neighbor-joining method using Bacillus subtilis as the root. The Romans strain clustered with previously characterized S. larvae strains (Figure).
Our report demonstrates that S. larvae can induce bacteremia in humans. Because S. larvae has been associated with only fly larvae, we can speculate that bacteremia originated from maggots infesting the patient's wounds. We cannot affirm that W. magnifica was the fly species involved because maggots were not saved for identification. Phenotypic identification of S. larvae is tedious (7). Because it is an asaccharolytic species, erroneous identification may occur. We can speculate that difficulties in phenotypic identification of this species may explain why it has not been previously reported as a potential human or animal pathogen.
In conclusion, myiasis remains an unresolved problem in animals and humans worldwide. Physicians and microbiologists should be aware of the possibility of S. larvae bacteremia and should specifically search for S. larvae infection in myiasis patients. Also, animal myiasis is still responsible for severe economic losses to the livestock industry worldwide. The occurrence of S. larvae bacteremia in animals with myiasis may explain the evolution from disease to death, especially in chronically infected animals.

Isolation of Schineria sp. from a Man
To the Editor: Schineria larvae has been isolated from maggots of the fly Wohlfahrtia magnifica (1), which cause myiasis in animals and people in Eurasia and northern Africa. In industrialized nontropical countries, a range of species in the order Diptera cause facultative myiasis in patients with neglected wounds (2). Since the recent description of S. larvae, Schineria sp. isolates and clones have been detected in diverse environmental and animal sources, but in all cases a relation with flies could be established. We describe a case of bacteremia due to Schineria sp. in a human patient with myiasis.
In July 2005, a 39-year-old homeless man with medical history of polyneuropathy related to alcohol abuse was examined at Montpellier Hospital, Montpellier, France, and found to be in poor general health and to have an abnormal electrocardiogram, mild fever (38°C), metabolic disorders, increased C-reactive protein (254 mg/L) and fibrinogen (18.23 µmol/L), and a normal leukocyte count (7.8 × 10 9 /L). Removal of his shoes and socks, which he had worn continuously for 2 months, showed advanced maceration of his feet (trench foot) with wounds invaded by maggots. The following organisms were found in wound samples: Proteus mirabilis, Providentia stuartii, group G Streptococcus, Streptococcus sp., and Enterococcus sp. Aerobic blood culture, after 2 days of incubation, was positive for a gram-negative rod, strain ADV1107. 05. Subculture on MacConkey medium showed positive reactions for oxidase, catalase, and gamma-glutamyltransferase. Positive malate reaction with API 20NE system (bioMérieux, Marcy l'Etoile, France) identified the strain as Oligella urethralis, whereas VITEK2 (bioMérieux) with ID-GN card failed to identify the strain. Disk diffusion assay showed the strain to be susceptible to β-lactams, aminoglycosides, fluoroquinolones, tetracyclines, erythromycin, rifampin, and colistin but resistant to nalidixic acid and fosfomycin. Local therapy of debridement, bandaging, and sulfadiazin argentic, along with systemic antimicrobial therapy (ofloxacin 400 mg/day plus cefotaxime 6 g/day) for 2 weeks, led to clinical improvement and sterilization of the blood cultures. The local therapy was continued, and ofloxacin (400 mg/day) was prescribed for 15 days while the patient was in a rehabilitation center.
In October 2005, the patient was readmitted with the same symptoms. P. mirabilis, group A and group G streptococci, Morganella sp., Bacteroides fragilis, and Candida albicans were cultured from maggotinvaded wounds. Aerobic blood culture, after 1 day of incubation, was positive for strain ADV4155.05, which displayed the same phenotype as strain ADV1107.05 except for tetracycline resistance. Clinical improvement was observed after 2 weeks of the same local and systemic treatments as initially prescribed. The patient was transferred to an addiction care center and received oral antimicrobial therapy (ciprofloxacin 500 mg/day plus amoxicillin/clavulanic acid 3 g/day) for 20 days.
The 16S rDNA amplification and sequencing were performed with universal primers 27f and 1492r as described (3). The 1,414-bp sequences of the 2 isolates were identical and showed similarity level of 99.6% with the sequence of Schineria sp. 010793816 isolated from human urine (M. Vaneechoutte, pers. comm.) but only 98.3% with S. larvae L1/68 T 16S rDNA. This finding differed from the biochemical identification and underlined the usefulness of sequencing to precisely identify gram-negative bacilli that assimilate only a few