Introduction:
Timely identification of infectious agents is critical for optimal clinical interventions. Despite substantial advancement, many cases remain undiagnosed using conventional testing because these are disease-specific and biased to “prevalent” infections, according to clinician's experience. Hence, new or emerging pathogens are likely to be missed unless they present with a clinically distinct phenotype. In this study, we utilised unbiased genomics-based diagnostics to identify infectious pathogens in a patient cohort with acute febrile illness who remained undiagnosed when they were discharged from the hospital.
Methods:
Sixty plasma samples were sourced from the Colombo Dengue study, and these patients presented with symptoms similar to dengue fever and were treated as such but were later found to be negative for dengue by serology and RT-PCR. Total nucleic acid (TNA) were extracted from plasma and divided into two portions: one for DNA and the other for RNA library preparation. RNA was reverse transcribed utilising Takara SMARTer Universal Low Input RNA Kit (Catalog number 634940). DNA and RNA libraries were sequenced on the high-throughput NovaSeq 6000 platform and generated data subsequently analysed through the CZID (Chan-Zuckerberg ID) pipeline.
Results:
Fourteen of the 60 patients had significant quantities of viral genomic material (in terms of depth and coverage across the genome) for Chikungunya (n=9) and Dengue viruses (n=5). In addition, another 35 samples had probable bacterial infection (e.g., Klebsiella, Pseudomonas), but given the patchy coverage across the large bacterial genomes, these need to be reconfirmed with specific serological tests.
Conclusion:
Our findings suggest that in Sri Lanka, chikungunya infection can masquerade as dengue and that unbiased genomic diagnosis can inform clinicians of routinely missed clinical diagnoses.