Dear Editor, Epidemic retinitis (ER), also known as post-fever retinitis, is a distinct clinical entity characterized by multifocal retinitis, macular edema, and an association with recent febrile illness.1 A rickettsial etiology has been widely postulated based on epidemiological patterns, serological associations, and favorable response to doxycycline therapy.1–3 However, definitive microbiological confirmation has remained elusive. Conventional diagnostics such as Weil–Felix test suffer from limited specificity and lack of documentation of fourfold raise in titers after the fever.1 Recent advances in metagenomic sequencing have raised expectations of pathogen identification in intraocular fluids. We studied ER cases with anterior chamber reaction that underwent aqueous tap for panel-based next-generation sequencing (NGS) validated for ocular fluid. The panel screened for multiple organisms (https://www.infexn.com/pathogens-detected.html) included several rickettsial and typhoid species: Rickettsia conorii, Rickettsia typhi, Rickettsia bellii, Salmonella enterica subsp. enterica serovar Paratyphi A, Salmonella enterica subsp. enterica serovar Typhimurium, and enterica serovar Typhi. Surprisingly, none of the samples demonstrated the presence of these genomes. This finding aligns with prior work by Arunasri et al., where metagenomic approaches on vitreous samples failed to identify a consistent pathogenic organism in post-fever retinitis and instead demonstrated a dysbiotic microbial milieu rather than a relevant causative agent.4,5 In our cohort, NGS returned positive only in three patients. The results were considered false positives as the detected organisms were represented by low read counts, lacked antimicrobial resistance gene signatures, and were not consistent with known intraocular pathogens. Furthermore, all patients demonstrated classical ER morphology and showed resolution with standard therapy, without targeted antimicrobial therapy against the detected organisms Table 1. These findings strongly suggested background contamination or commensals.Table 1: Clinical correlation of aqueous humor NGS findings in ERA critical limitation of the NGS platform used in our patients was its panel-based design. While it included rickettsial and typhoid organisms, it did not screen for several viruses previously associated with ER, including dengue, chikungunya, and West Nile virus. This restricted coverage may introduce a selection bias in pathogen detection. In contrast, whole-genome metagenomic sequencing may offer broader pathogen detection but remains limited by cost, turnaround time, and interpretative complexity. Clinicians must exercise caution in interpreting positive NGS results, especially when they are discordant with clinical presentation. Additionally, the negative results raise several possibilities: Aqueous humor may not be the optimal sample for detection of intracellular organisms such as rickettsial or enteric pathogens; alternatively, ER in this cohort may not be attributable to rickettsial etiology despite supportive serology (Weil–Felix test in two cases), or the panel-based NGS approach itself may be limited by restricted organism coverage and low microbial yield in small-volume aqueous samples. In conclusion, our study, in concordance with previous reports,4,5 did not identify rickettsial or typhoid organisms in ER using aqueous humor panel based NGS. These findings highlight both the biological and technical limitations of current NGS platforms and underscore the need for cautious interpretation. Until more robust evidence emerges, clinical judgment supported by characteristic morphology and therapeutic response remains central to the management of ER. Financial support and sponsorship: Nil. Conflicts of interest: There are no conflicts of interest.
Kawali et al. (Wed,) studied this question.