ABSTRACT: Arthropod-borne viruses (arboviruses) represent a threat to global public health, especially in the tropical and subtropical regions of the world. More than 150 arboviruses can infect humans; they cause mainly febrile illness, although hemorrhagic complications and diseases affecting the central nervous system (SNC) can also be observed. Arboviruses represent a threat to Brazil and, therefore, a permanent surveillance of these viruses is required to timely reduce the risk of epidemic outbreaks. The Brazilian Amazon region is where the highest number of arboviruses has been detected in the world. Besides, malaria is also endemic in the Amazon region, with a significant predominance of Plasmodium vivax. It is often difficult to differentiate between malaria and arboviral diseases, as they share similar clinical features and laboratory findings, mainly undifferentiated fever. This study aimed to estimate possible viral etiology in patients with febrile syndrome negative for Plasmodium infection, in the Brazilian Amazon. We initially analyzed serum samples of 124 participants with a DNA microarray platform designed for the detection of arboviruses and viruses transmitted by small mammals, but no virus was detected. Then, the serum samples of 76 participants were analyzed with a deep New Generation Sequencing, which showed evidence of the presence of only one arbovirus, the Zika virus in only one pool of 9 serum samples. This result is in contrast with our hypothesis, showing that arboviruses are not frequent in suspected malaria cases in Manaus, Brazil. Other viruses instead of arboviruses were found in this study. Primate erythrovirus 1 was the virus most frequently found virus in the suspected malaria patients, followed by Enterobacteria phage lambda. Besides, we detected, in a lower frequency, the Pegivirus C. In addition to the exogenous viruses, we also detected human endogenous retrovirus in all pools. Due to the high number of viruses that are important in the differential diagnosis of malaria, cost-effective and simple high throughput methods are required, helping molecular surveillance of misdiagnosed viral infections. Further studies with more robust sample sizes in other areas in the Amazon are needed.