Search

International audience; Bacteria genetically related to Coxiella burnetii have been found in ticks. Using molecular techniques, we detected Coxiella-like bacteria, here named Candidatus Coxiella massiliensis, in skin biopsy samples and ticks removed from patients with an eschar. This organism may be a common agent of scalp eschar and neck lymphadenopathy after tick bite.

To the Editor: Pulmonary involvement is a well-documented complication of scrub typhus caused by Orientia tsutsugamushi (1). Lung involvement manifests as bronchitis and interstitial pneumonitis of various grades that progress to acute respiratory distress syndrome (ARDS), a serious complication that occurs in ≈11% of scrub typhus patients (2). The death rate among scrub typhus patients with ARDS can reach 25% (3). Older age, thrombocytopenia, and the presence of early pneumonitis have been proposed as risk factors for the development of ARDS in scrub typhus patients (3). We report the detection and culture of O. tsutsugamushi in a bronchoalveolar lavage specimen from a patient with scrub typhus–associated ARDS. A 50-year-old woman from Lyon, France, was admitted to the hospital in November 2013 with fever (39°C), dizziness, diarrhea, dyspnea, and nonproductive cough. The woman, who had just returned from travel to a jungle in Laos, reported that the fever and diarrhea had begun immediately before her return home. Examination revealed that she had an oval eschar on her back and a faint maculopapular rash. Laboratory values showed elevated C-reactive protein and liver enzyme levels, lymphocytopenia, and thrombocytopenia. Extensive microbiological testing was done, including tests to rule out malaria, dengue, viral hepatitis, and leptospirosis; all results were negative. Salmonella sp. infection was suspected, and treatment with ofloxacin was started. On hospitalization day 5, the patient showed development of septic shock, renal failure, and ARDS. She was transferred to an intensive care unit, and treatment with ceftriaxone was started. On hospitalization day 6, a skin biopsy of the eschar (2 mm × 5 mm) and blood, serum, cerebrospinal fluid (0.5 mL), and bronchoalveolar lavage (1 mL) samples were obtained and sent to the National Reference Center for Rickettsiae (Marseille, France) for analysis. Total genomic DNA was extracted (Biorobot EZ1 Workstation; QIAGEN, Courtaboeuf, France) from 200 μL of each sample and used as template in a real-time PCR, which used primers and probes targeting a 47-kDa outer membrane protein gene, as described (4). Blood, skin biopsy, and bronchoalveolar lavage samples were positive for O. tsutsugamushi; the cerebrospinal fluid sample was negative. The serum sample was positive for O. tsutsugamushi by indirect immunofluorescence assay (serotypes Gilliam, Kuroki, Sennetsu, and Kawasaki) (5) and positive for O. tsutsugamushi IgM. Oral doxycycline (200 mg/day) was started on hospital day 7; the fever resolved 4 days later. For culture, the positive samples were directly inoculated into monolayers of L929 cells, as described (6). Cultures of blood and skin biopsy samples were negative, but O. tsutsugamushi was isolated from the bronchoalveolar lavage sample after 40 days of culture (Figure); 500 μL of bronchoalveolar lavage fluid was used for culture. We performed PCR amplification and sequencing of the isolate, targeting a 372-bp fragment of the 56-kDa protein gene, and compared the sequences with O. tsutsugamushi 56-kDa protein–encoding gene sequences available in GenBank (7). The sequences showed 99% similarity with strains Jin/2012 and Zhou/2013 (GenBank accession nos. {"type":"entrez-nucleotide","attrs":{"text":"KJ001159","term_id":"594615391","term_text":"KJ001159"}}KJ001159 and {"type":"entrez-nucleotide","attrs":{"text":"KJ001163","term_id":"594615400","term_text":"KJ001163"}}KJ001163, respectively), which were obtained from febrile patients in Zhejiang Province, China, and have not been linked to a reference serotype (Technical Appendix Figure). In light of the test results and the patient’s recent travel to Laos, she was given a diagnosis of O. tsutsugamushi infection–associated ARDS. Figure Orientia tsutsugamushi (arrows) in culture of bronchoalveolar lavage fluid from a patient with acute respiratory distress syndrome (Diff-Quick stain, VWR International, France). Original magnification ×100. Our isolation of O. tsutsugamushi in bronchoalveolar lavage fluid from a patient with scrub typhus shows that this bacterium can be present in such samples. We also showed that skin biopsy and bronchoalveolar lavage samples can be used for the diagnosis of scrub typhus. To be suitable for culture, samples must be collected as early as possible in the disease course. In this case, blood and skin biopsy samples were obtained late in the disease, which may explain why O. tsutsugamushi was not isolated from these samples. Endothelial cells are the target cells of O. tsutsugamushi in the lung (8), and it has been proposed that ARDS in scrub typhus is associated with a cytokine increase as part of the immune response to O. tsutsugamushi infection (9). Rickettsial diseases are increasingly being diagnosed in international travelers: one report showed that 2% of imported fevers are caused by rickettsioses, and hospitalization was necessary for the 38% of O. tsutsugamushi–infected travelers (10). The diagnosis of rickettsial infections is challenging because many physicians are unfamiliar with these diseases. However, the diagnosis of scrub typhus in patients with ARDS is critical for initiating appropriate and timely doxycycline treatment. In the case reported here, a diagnosis of scrub typhus was not suspected even though the patient had compatible exposure and travel histories and clinical findings consistent with the disease. The delay in diagnosis led to a life-threatening condition for the patient. Physicians in areas where scrub typhus is nonendemic should have a high index of suspicion for rickettsial infections in patients with recent travel histories to areas where the disease is endemic and consider treatment with tetracyclines whenever rickettsial infection is suspected. Furthermore, the potential for aerosol transmission of O. tsutsugamushi from patients with scrub typhus–associated ARDS to health care workers should be evaluated. Technical Appendix: Phylogenetic tree for Orientia tsutsugamushi 56-kDa protein–encoding gene sequences obtained from GenBank. Click here to view.(77K, pdf)

To the Editor. Q fever is a worldwide zoonosis caused by an obligate intracellular bacterium, Coxiella burnetii (1). Q fever endocarditis is associated with surgery for 15%–73% of patients, causes death for 5%–65% of patients, and induces a large number of relapses when the endocarditis is inadequately treated (1). The most serious risk factor for endocarditis is a substantial underlying valvulopathy, but progression to endocarditis is also found in patients with clinically silent, previously undiagnosed, valvulopathies (1). Since the 1960s, Q fever has been recognized as a public health problem in Saudi Arabia, and studies have shown that coxiellosis occurs in livestock (2,3). Only a few cases of Q fever endocarditis in Saudi Arabia have been reported (4–6). We report 2 new cases of Q fever endocarditis and detection of a new C. burnetii genotype in this country. The first case was detected in 2007 in a 45-year-old man in Saudi Arabia who had fever, pneumonia, and asthenia. A transesophageal echocardiogram showed endocarditis. Results of an immunofluorescence assay were positive for C. burnetii; phase I titers for IgG, IgM, and IgA were 51,200, 100, and 25, respectively, and phase II titers were 102,400, 200, and 50, respectively. Serum and blood samples were negative for C. burnetii by real-time PCR for the IS1111 and the IS30A spacers (7). For each sample, the quality of DNA extraction was verified by real-time PCR for a housekeeping gene encoding β-actin (7). The aortic valve was surgically replaced, and C. burnetii–specific PCR results for the valve were positive. According to multispacer sequence typing (8), this C. burnetii isolate was a new genotype, MST51 (Figure). A C. burnetii isolate was cultured from the valve of this patient by the shell-vial method that used human embryonic lung cells (7). IgG anticardiolipin testing results were negative (9). The patient was given 200 mg oral doxycycline daily and 200 mg oral hydroxychloroquine 3 times daily for 18 months. Figure Neighbor-joining tree of Coxiella burnetii genotypes determined by multispacer sequence typing. Arrow indicates new genotype in Saudi Arabia. The second case was detected in 2012 in a 13-year-old boy in Saudi Arabia who had tetralogy of Fallot, a prosthetic pulmonary valve, 2 intracardiac stents, and long-term fever. Serologic testing results were positive for C. burnetii; phase I titers for IgG, IgM, and IgA were 51,200, 400, and 200, respectively, and phase II titers were 102,400, 800, and 400, respectively. Whereas serum and blood samples were negative for C. burnetii by real-time PCR for the IS1111 and the IS30A spacers, the β-actin control was positive (cycle threshold 16% of patients with acute Q fever have endocarditis and that ≈16%–37% of patients with Q fever endocarditis could have had previous symptomatic acute Q fever infection (1). Thus, many cases of endocarditis might be avoided if patients with acute Q fever receive antimicrobial drugs as prophylaxis (1). For patients >40 years of age, transthoracic echocardiography should be performed because of the increased prevalence of valvulopathy and Q fever endocarditis in this population (9). As a result, more studies are needed to determine whether our data can affect local clinical practice.

To determine the presence of Bartonella quintana in head and body lice from persons in different locations in Ethiopia, we used molecular methods. B. quintana was found in 19 (7%) genotype C head lice and in 76 (18%) genotype A body lice. B. quintana in head lice was positively linked to altitude (p = 0.014).

Cat-scratch disease is seasonal in the United States and Japan; but no data are available from Europe. To assess the seasonality of the disease in France, we analyzed lymph node biopsy specimens collected during 1999–2009. Most (87.5%) cases occurred during September–April and peaked in December.

To the Editor: The genus Sphingomonas was proposed by Yabuuchi et al. in 1990 (1) and amended by Takeuchi et al. in 1993 (2). It now has been subdivided into 4 separate genera: Sphingomonas sensu stricto, Sphingobium, Novosphingobium, and Sphingopyxis. The bacteria of the genus Sphingomonas are yellow-pigmented, nonfermenting, gram-negative bacilli with a single polar flagellum; they are widely distributed in the natural environment, especially in water and soil (3). These bacteria are characterized by the presence of a unique sphingoglycolipid with the long-chain base—dihydrosphingosin, ubiquinone 10 (Q-10), and 2-hydroxymyristic acid (2-OH C14:0)—and the absence of 3-hydroxy fatty acids (4). S. mucosissima was isolated and identified in 2007 by Reddy and Garcia-Pichel from biologic soil crust samples collected from sandy arid soil in the US Colorado Plateau (5). Sphingomonas spp. are opportunistic pathogens and have recently been implicated in a variety of community-acquired and nosocomial infections, considered to originate from contaminated hospital equipment or manipulation of some medical devices (3). The survival of Sphingomonas spp. in indoor dust particles as aerosols and their resistance to many disinfecting and toxic chemicals may explain their ability to colonize medical devices such as mechanical ventilators, catheters, and bronchofiberoscopes (6). In the past few years, these organisms, in particular S. paucimobilis, have been implicated in a variety of community-acquired and nosocomial infections. We report a case of S. mucosissima bacteremia in a patient with sickle cell disease. In February 2008, a 17-year-old woman with homozygous sickle cell anemia was hospitalized when her condition suddenly became worse. The patient had undergone a splenectomy in 1992 and a cholecystectomy in February 2007. Four days after admission, she had a fever of 38.7°C. Two aerobic blood specimens, drawn on the fifth day of her hospitalization, yielded gram-negative bacilli after a 24-hour incubation. The gram-negative bacilli were positive for catalase and oxidase but remained unidentified by API 20NE strip (bioMerieux, Marcy l’Etoile, France). MICs of antimicrobial drugs were determined for the gram-negative bacilli by using an Etest assay (AB BIODISK, Solna, Sweden) on Mueller-Hinton medium. MICs were 1 μg/mL for cefotaxime, 1 μg/mL for amoxicillin–clavulanic acid, 2–3 μg/mL for vancomycin, 0.064 μg/mL for imipenem, 4–5 μg/mL for ceftazidime, 1 μg/mL for amikacin, 3 μg/mL for ciprofloxacin, and 0.047 μg/mL for trimethoprim-sulfamethoxazole. DNA was extracted from 1 colony by using a QIAamp Tissue kit (QIAGEN, Hilden, Germany) as described by the manufacturer. A 16S rDNA sequence was obtained (1,410 bp) by using the fD1 (5′-AGAGTTTGATCCTGGCTCAG-3′) and rP2 (5′-ACGGCTACCTTGTTACGACTT-3′) primer pair (7,8). Using BLAST version 2.2.9 software (www.ncbi.nlm.nig.gov/BLAST), we determined that this sequence showed 98% similarity with the 16S rDNA sequence of S. mucosissima (GenBank accession no. {"type":"entrez-nucleotide","attrs":{"text":"AM229669","term_id":"87080454","term_text":"AM229669"}}AM229669). A phylogenetic neighbor-joining tree resulting from comparison of sequences of the 16S rDNA genes of Sphingomonas spp. was made with the MEGA 3.1 software (www.megasoftware.net). This analysis confirmed that the isolate belonged to S. mucosissima. Initial treatment of intravenous administration of ceftriaxone was begun. The fever resolved after 1 day and the patient’s condition improved. Treatment was stopped after 5 days, and the patient remained apyretic. Two S. mucosissima isolates were recovered from 2 different blood-culture samples drawn 24 hours apart, which suggests that S. mucosissima was not just a transient organism but indeed was responsible for the patient’s septicemia. Phenotypic identification of the gram-negative bacilli failed because the definite bacterial species S. mucosissima was not included in the API database (http://industry.biomerieux-usa.com/industry/food/api/apiweb.htm) used for the phenotypic identification. However, the isolates’ biochemical characteristics were consistent with those previously reported for S. mucosissima (5) (Table). Final identification was achieved by comparing the almost complete 16S rDNA sequence with homologous sequences deposited in GenBank. Table Biochemical characteristics of the previously reported Sphingomonas mucosissima isolate ({"type":"entrez-nucleotide","attrs":{"text":"AM229669","term_id":"87080454","term_text":"AM229669"}}AM229669) and the isolate from this study We believe that the patient’s intravenous catheter was the source of the infection because she did not have wound infections and cultures of her urine were negative for infectious agents. Antimicrobial drug treatment, selected on the basis of an in vitro S. mucosissima susceptibility profile, facilitated the patient’s recovery. This case report illustrates that the pathogenic potential of S. mucosissima should be considered in diagnosis in such cases because the organism can cause bacteremia in patients, primarily in those with underlying debilitating conditions and those who have undergone medical interventions.