• Background: The presence of herpetic DNA has been shown in diseased and healthy corneal tissue. A clinical correlation with the activity of the disease has not yet been demonstrated. This study was done to evaluate the use of DNA amplification for HSV-1 from different sites for the clinical prognosis after corneal grafting. • Patients and methods: Eighteen patients with herpetic keratitis, 8 patients with other forms of keratitis, and 15 patients with corneal disease unrelated to herpes undergoing penetrating keratoplasty were investigated. From these, aqueous humor was obtained at the time of surgery. The excised cornea was divided into three parts for paraffin embedding, 24 h tissue culture and preparation of minced tissue. All samples were processed for HSV-1 glycoprotein D PCR followed by Southern blot and DNA hybridization. • Results: In the herpes group, target DNA was detected in 4/18 aqueous humor samples, 7/16 minced tissue preparations, 6/18 explant culture fluid samples and 4/15 paraffin sections. In the control groups of other keratitis and nonherpetic eye disease, respectively, target DNA was found in 0/5 and 2/12 aqueous humor samples, 1/6 and 0/12 minced tissue preparations, 0/8 and 0/15 explant culture fluid samples and in 1/6 and 1/14 paraffin sections. Five of six patients in whom herpes DNA was detected in the short-term tissue culture experienced an episode of herpes reactivation within 4 months after transplantation, whereas only one of the remaining patients in all three groups did so (p=0.0005). • Conclusion: A slow viral replication may be responsible for early recurrence of herpetic keratitis after corneal grafting. Detection of herpetic DNA in short-term tissue cultures from explant tissues may help to define the patients at risk.