Relative sensitivity of polymerase chain reaction assays used for detection of feline herpesvirus type 1 DNA in clinical samples and commercial vaccines

Objective - To determine relative detection rates and detection limits for 6 published polymerase chain reaction (PCR) assays used for detection of feline herpesvirus type 1 (FHV-1) DNA. Sample population - 5 vaccines licensed for use in preventing FHV-1-associated disease; 15 conjunctival biopsy specimens collected from cats with keratitis, conjunctivitis, or both; and a plaque-purified field isolate of FHV-1 cultured in vitro. Procedure - Vaccines and clinical samples were assessed for FHV-1 DNA by use of all 6 assays. Detection rates were calculated by assuming that any sample in which FHV-1 DNA was detected was a true-positive result. Detection limits were estimated by use of serial dilutions of DNA extracted from cultured FHV-1 and 1 clinical sample. Results - Testing by use of all 6 assays resulted in detection of FHV-1 DNA in all 5 vaccines. Testing by use of all 6 assays yielded concordant results for 9 of 15 conjunctival biopsy specimens (8 with negative results and 1 with a positive result). Calculated detection rates for clinical samples ranged from 29% to 86%. Assay sensitivity was ranked similarly by use of detection rate or detection limit. Conclusions and clinical relevance - Testing by use of all assays was equally likely to detect vaccine virus. Therefore, a positive PCR result in a cat may reflect vaccine virus rather than wild-type virus. Test sensitivity as assessed by detection limits and detection rates varied greatly. Because FHV-1 can be shed in clinically normal animals, high detection rate will not necessarily correlate with high diagnostic sensitivity.