Typhus Group Rickettsiosis, Brazilian Amazon

Rickettsia rickettsii infection is the only rickettsiosis included in the list of reportable diseases in Brazil, where typhus group rickettsioses, mainly murine typhus, have been underreported. We report a case of typhus group rickettsiosis with unique ecologic particularities in a patient from the Brazilian Amazon, where, to our knowledge, rickettsioses have not been reported.

cropland areas are associated with high risk for infection (2,3,6). Northeast Jeju Island, which includes SH, AW, and HD, has many farms and wetlands, and the IR of SFTSV in ticks peaked there in July, August, and September. In addition, SFTSV was detected in ticks in winter on Jeju Island, but no SFTS cases were reported in South Korea during winter. The 62 confirmed SFTS cases were statistically significantly associated with higher ambient temperature (22.5°C +4.2°C) compared with patients with negative RT-PCR results for SFTSV (18.9°C +5.7°C; p<0.001) (J.R. Yoo, unpub. data). The optimal temperature range for growth and reproduction of H. longicornis ticks is 20°-24°. Jeju Island maintains a temperature >20° during May-October and is largely a rural and natural environment. We consider this area to have the highest prevalence of SFTS cases and ticks with SFTSV in South Korea.
The results of this study showed that Jeju Island has the highest IR of SFTSV in ticks compared with other regions of South Korea and endemic countries. In addition, we found that the partial small segment of SFTSV in ticks was highly homologous to SFTSV in patients on Jeju Island and that Northeast Jeju Island, which includes SH, is a high-risk area for human SFTS infections. T yphus group rickettsioses are vectorborne infectious diseases that include murine typhus, caused by Rickettsia typhi, and epidemic typhus, caused by R. prowazekii (1). R. typhi is maintained in an enzootic cycle involving small mammals (e.g., Rattus spp. rats and Didelphis spp. opossums) and their ectoparasites, mainly fleas (2). R. typhi is usually transmitted to humans by contamination of the bite site, mucosal or skin abrasions with rickettsia-containing ectoparasite feces, or inhalation in contaminated dust (2). R. prowazekii is transmitted mainly by feces of human clothing lice (Pediculus humanus humanus) or a sylvatic cycle in the United States by contact with ectoparasites of flying squirrels (1).

Typhus Group Rickettsiosis, Brazilian Amazon
In Brazil, there have been few reports of murine typhus, mostly >60 years ago and all from humanmodified landscapes in southeastern or southern regions, far from the Amazon (3). To our knowledge, Brazil has had only 1 case of recrudescent, epidemic typhus (Brill-Zinsser disease) in a refugee from Europe (4). Currently, Brazilian spotted fever (R. rickettsii infection), a tickborne disease, is the only rickettsiosis included in the list of notifiable diseases in Brazil (3). We report a new case of typhus group rickettsiosis in a patient from the Brazilian Amazon.
On August 27, 2019, a 37-year-old man was admitted to a hospital at Porto Trombetas District (Pará State, eastern Amazon region of Brazil) (Figure) because of 4 days of fever, chills, headache, malaise, and productive cough, associated with mild respiratory distress 72 hours after fever onset. His daily work consisted of outdoor herpetologic monitoring activities in Saracá-Taquera National Forest (35 km from Porto Trombetas) during the 2 weeks before disease onset. He reported that while in the forest, he removed an attached tick from his right thigh 10 days before disease onset.
Supplementary oxygen, ceftriaxone, chloramphenicol, and doxycycline were administered, and serologic tests for Borrelia burgdorferi and Rickettsia spp. were performed. Twenty-four hours later, the patient showed clinical stability, major improvement of respiratory status, and no fever. After 7 days of antimicrobial drug treatment, he was discharged with complete resolution of symptoms. A serum sample collected on August 28 (5 days after disease onset) showed negative results by commercial ELISA for B. burgdorferi IgM and IgG. Results of indirect fluorescent antibody assays were nonreactive for R. conorii IgG but positive (titer 1:64) for R. typhi IgG.
To confirm a presumptive diagnosis of rickettsiosis, we collected a second serum sample 2 months later and showed by using an in-house indirect fluorescent antibody assays (5) negative results for IgG (titer <1:64) against 6 Rickettsia species: R. felis; R. bellii; and the spotted fever group agents R. rickettsii, R. parkeri, R. amblyommatis, and R. rhipicephali. This serum sample had a titer of 1:2,048 against R. typhi (Wilmington strain), confirming typhus group rickettsiosis.
There have been previous reports of typhus group rickettsiae in ticks from other areas (6,7). After we considered the tick bite history of the patient, during October 12-14, 2019, we went to areas in Saracá-Taquera National Forest that the patient visited and collected 170 ticks in 7 species. Attempts to detect rickettsial DNA in these ticks showed only R. amblyommatis in Amblyomma cajennense sensu stricto ticks (Appendix, https://wwwnc.cdc.gov/EID/article/26/9/20-1305-App1.pdf).
In a retrospective interview, the patient recalled 3 additional activities during the 2 weeks before disease onset: a daily rest in the forest for a few minutes after work, seeking work materials in a small mammal trap-storage at a facility within the natural forest, and 2 short visits to his mother's house in the urban area of Santarém municipality (Pará State). He denied rat infestation in his own house at Porto Trombetas but reported previous rat infestations in his mother's house.
Borreliosis and rickettsioses have not been confirmed in the Brazilian Amazon (8). The patient's tick bite history before onset of symptoms led clinicians to presume these diagnoses and initiate appropriate antimicrobial drug treatment (e.g., doxycycline) (1), with a successful outcome. Because R. amblyommatis, a possibly nonpathogenic species (1), was the only agent detected in collected ticks, we consider that the furuncle lesion at the site of the tick bite was a pyogenic, localized, skin infection, and not rickettsia related.
Because of similar clinical manifestations and serologic cross-reactions between typhus group rickettsiae (1), we could not confirm the typhus group etiologic agent of this case-patient. We presume R. typhi as a probable infection because he had potential occupational exposures with rodents or other small mammals and their ectoparasites, rather than clothing louse infestation (1), in addition to an absence of neurologic symptoms, which are more common for epidemic typhus than for murine typhus (9,10). Conversely, the fact that the patient had spent most of his time in a forest environment could also implicate a sylvatic cycle of typhus group rickettsia in the Amazon forest.