Virulent forms of Newcastle disease virus (NDV) are a major concern for poultry producers around the world and the rapid diagnosis of an outbreak is crucial to any control program. A validated real-time reverse transcription-PCR test (fusion test) directed at the fusion-cleavage site of NDV was developed to differentiate virulent Newcastle disease virus strains from those of low virulence, however one virulent isolate, Dove/Italy/2736/2000, escaped detection during the initial evaluation of the test. The objectives of this study were to determine how this isolate differed from other detectable isolates, to identify other isolates that may fail detection, and to characterize the effect of specific probe-site mutations on the fusion test at a range of annealing temperatures. Using a virulent NDV isolate (Game fowl/US(CA)/2002) as a backbone that has 100% identity to the fusion-test probe, specific changes were made to the fusion-test probe-site to reflect the unique mismatches found in Dove/Italy/2736/2000 and other selected regions of the probe. Mutated clones with mismatches unique to Dove/Italy/2736/2000 at positions 6, 13, and 14 were not detected until annealing temperatures were lowered to 50 degrees C. Those detected at 58 degrees C contained 1-2 mismatches (position 1 and 6, 13 and 14, or 14 only) although increased cycle threshold values compared to the parent clone indicated decreased sensitivity. Data from this study predicts that the fusion test may fail to detect some viruses among lineage 4b and potential solutions to identify this subset of viruses include lowering the annealing temperature or modifying the probe.