Comparative Evaluation of LAMP and PCR for HSV Detection in Corneal Samples from Patients with Suspected and Confirmed Herpetic Keratitis

Purpose. Accurate detection of herpes simplex virus (HSV) in ocular infections, particularly in herpetic keratitis (HSK), is essential for effective clinical management. This study compared the analytical and clinical performance of loop-mediated isothermal amplification (LAMP) and polymerase chain reaction (PCR) for detecting HSV DNA in corneal specimens from patients with varying clinical presentations. Methods. The analytical sensitivity of LAMP and PCR was initially evaluated using serial dilutions of HSV-1 viral stock. Clinical testing was performed on two types of corneal samples from 92 adult patients. The first group, the clinical manifestation group (n = 53), included patients with ocular symptoms but no epithelial defects, with specimens obtained through corneal surface swabbing. The second group, the keratitis group (n = 39), comprised patients with confirmed HSK characterized by epithelial loss, with specimens collected via corneal scraping. The keratitis cases were further divided into epithelial, stromal (including ulcerative and non-ulcerative), and unspecified subtypes based on a retrospective review of clinical documentation. All samples were analyzed using both LAMP and PCR methods. Results. Analytical sensitivity testing showed detection limits of ≥10 infectious units (IU) for LAMP and ≥25 IU for PCR. Across the clinical cohorts, the proportion of patients positive by both LAMP and PCR was approximately threefold higher in the keratitis group than in the clinical manifestation group (43.6% vs. 15.1%, p = 0.0066). LAMP-only positives were observed in both groups but were more frequent in the keratitis group. No patient tested positive by PCR alone. Double-negative results were nearly twice as common in the clinical manifestation group (45.3%) as in the keratitis group (23.1%). Within the keratitis subtypes, concordant positive results were most frequent in epithelial and ulcerative stromal HSK, whereas negative results were concentrated in non-ulcerative and not-specified cases. Conclusions. Differences in detection rates across clinical groups and HSK subtypes reflect underlying variation in viral load, lesion accessibility, and sampling conditions. LAMP demonstrated higher analytical sensitivity and detected additional cases missed by PCR, supporting its utility as a rapid and sensitive diagnostic tool for ocular HSV, particularly in early-stage or mild disease where viral load may be low. Integrating molecular findings with careful clinical assessment remains essential for accurate diagnosis.