Diagnostic Accuracy of a CRISPR-Based Assay in Smear- and Culture-Negative Fungal Keratitis.

Abstract

IMPORTANCE: Diagnosing fungal keratitis (FK) in patients with negative smear and culture results remains clinically challenging, highlighting the need for alternative diagnostic approaches. OBJECTIVE: To determine the diagnostic accuracy of the clustered regularly interspaced short palindromic repeats (CRISPR)-based Rapid Identification of Mycoses using CRISPR (RID-MyC) assay for detecting FK in patients with negative smear and culture results using in vivo confocal microscopy (IVCM) as the reference standard. DESIGN, SETTING, AND PARTICIPANTS: This prospective diagnostic accuracy study was conducted from December 2024 to March 2025 at Aravind Eye Hospital, a tertiary ophthalmology referral hospital in Coimbatore, India. Consecutive patients clinically suspected to have microbial keratitis with negative smear and culture results were eligible for inclusion. Data were analyzed from March 2025 to June 2025. INTERVENTIONS: All included participants underwent corneal scraping for RID-MyC assay and imaging by IVCM. MAIN OUTCOMES AND MEASURES: The primary outcomes were sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic concordance of the RID-MyC assay compared with IVCM results. RESULTS: Of 245 consecutive patients clinically suspected to have microbial keratitis, 82 were smear negative. After exclusions due to contraindications or positive subsequent cultures, 41 patients with smear- and culture-negative results were ultimately included in the final analysis. Of these 41 patients (mean [SD] age, 51.0 [14.6] years; 21 [51.2%] women), RID-MyC demonstrated sensitivity of 82.1% (95% CI, 63%-94%) and specificity of 76.9% (95% CI, 46%-95%). Positive predictive value was 88.5% (95% CI, 74%-95%) and negative predictive value was 66.7% (95% CI, 46%-82%). Concordance between RID-MyC and IVCM was observed in 33 cases (80.5%). Notably, prior antifungal treatment was most frequent (4 of 5 [80%]) among patients with positive IVCM but negative RID-MyC results. Conversely, all patients (3 of 3 [100%]) with negative IVCM but positive RID-MyC findings had smaller, peripheral, or paracentral lesions. CONCLUSIONS AND RELEVANCE: In this diagnostic study, in patients with smear- and culture-negative FK, the RID-MyC assay showed good diagnostic accuracy comparable with IVCM and was feasible in all cases, including those in whom imaging was not possible. With its rapid turnaround and minimal equipment needs, RID-MyC may serve as a practical adjunct to conventional diagnostics, particularly in high-burden, resource-limited settings where IVCM is unavailable or contraindicated.