Background Acute febrile illness (AFI) represent a significant health challenge in the Peruvian Amazon basin population due to their diverse etiologies and the unavailability of specific on-site diagnostic methods, resulting in underreporting of cases. In Peru, one of the most endemic regions to dengue and leptospirosis is Madre de Dios, a region also endemic to emergent bacterial etiologic agents of AFI, such as bartonellosis and rickettsiosis, whose prevalence is usually underreported. We aimed to molecularly identify the presence of Leptospira spp., Bartonella bacilliformis, and Rickettsia spp. by Polymerase Chain Reaction in serum samples from patients with AFI from Puerto Maldonado-Madre de Dios in Peru. Methods Serum samples from patients with acute febrile illness were analyzed by real-time PCR for detecting the presence of Bartonella bacilliformis, Leptospira spp. and Rickettsia spp. Results Bartonella bacilliformis was the most prevalent bacteria identified in 21.6% (30/139) of the samples, followed by Leptospira spp. in 11.5% (16/139) and Rickettsia spp. in 6.5% (9/139) of the samples. No co-infections were observed between these bacteria. The most frequent symptoms associated with fever among all groups, were headaches, myalgias, and arthralgias. We found no statistically significant differences in the clinical presentation between patients infected with each bacterium. Conclusions In a previous study, we shown the presence of dengue, chikungunya, Zika and oropouche virus. We were able to identify these pathogens in 29.5% of all the samples, with chikungunya and OROV as the most frequently found in 9.4 and 8.6% of all the samples, respectively. In this study we show that B. bacilliformis (21.6%), Leptospira spp. (11.5%) and Rickettsia spp. (6.5%) accounted for the main etiologies of AFI in samples from Puerto Maldonado-Madre de Dios, Perú. Our analysis of their clinical presentation, further shows the importance of implementing more sensitive and specific on-site diagnostic tools in the national surveillance programs.This study confirms that the un-specificity of signs and symptoms is not only associated with arboviral infections, but also with the clinical presentation of endemic bacterial infections.

Peer Reviewed