Background and objectives: The diagnosis of mucocutaneous herpes simplex virus (HSV) is hampered by suboptimal sensitivity of virus culture and atypical clinical morphology.
Goals: To compare the diagnostic usefulness of the polymerase chain reaction (PCR) and virus culture.
Study design: Consecutive samples from 246 patients at an urban sexually transmitted diseases clinic were tested for HSV by both PCR and virus culture.
Results: Only 59% of HSV-positive samples were correctly diagnosed by the clinician; 11% had an atypical appearance. HSV-positive lesions were more often vesiculoulcerative or crusted than HSV-negative lesions, and of shorter median duration. Thirty-one samples were PCR positive and virus culture negative; these were often from crusted or older lesions. However, PCR was negative in 27 instances in which HSV was diagnosed clinically, of which 2 were vesicular and 15 ulcerative.
Conclusions: HSV PCR is more rapid and sensitive than virus culture for diagnosis of mucocutaneous lesions. The data suggesting that PCR may be suboptimally sensitive need to be further investigated.