Background
Fungal species are responsible for 40%–50% of all microbial keratitis cases. Due to the low amount of extracted DNA in ocular Formalin‐fixed Paraffin‐embedded (FFPE) samples, selecting a reliable molecular method is a substantial issue in this field.
Methods
Sixty‐six samples were collected via the penetrating keratoplasty (PK) technique. Histopathology assays were performed using hematoxylin–eosin (H&E) and periodic acid Schiff (PAS) staining methods. The ITS1/ITS4 and ITS1/ITS2 primer pairs were used in a semi‐nested polymerase chain reaction (PCR) to target the universal internal transcribed spacer (ITS) region. Some PCR results were validated through sequencing.
Results
Fungal DNA was detected in 44 of 66 samples (66.7%), and histopathology was positive for 41 of 66 samples (62.1%). Of 41 histopathologically proven fungal‐positive cases, 39 were PCR‐positive (95%). Moreover, of 44 PCR‐positive samples, 39 (88.6%) were histopathology‐positive, and 5 (11.3%) were histopathology‐negative. Totally in 39 cases (59%), both histopathology and PCR yielded positive results. The Kappa agreement rate between the two diagnostic methods, including histopathology and PCR, was 0.77. Sensitivity, specificity, positive predictive value, and false predictive value were reported as 88.64%, 90.9%, 95.12%, and 80%, respectively.
Conclusion
As we reached the acceptable Kappa agreement rate, we concluded that applying the semi‐nested PCR assay is a promising method for supporting the evidence by histopathology. Finally, we suggest targeting more specific gene regions using primer pairs that amplify smaller amplicon sizes and surveying novel molecular methods such as NGS to achieve higher sensitivity and Kappa agreement rates.