PURPOSE: The aim of this study was to assess the ability of real-time breath-hold-triggered myocardial perfusion imaging (MPI) using a novel cadmium-zinc-telluride (CZT) gamma camera to discriminate artefacts from true perfusion defects. METHODS: A group of 40 patients underwent a 1-day (99m)Tc-tetrofosmin pharmacological stress/rest imaging protocol on a conventional dual detector SPECT gamma camera with and without attenuation correction (AC), immediately followed by scanning on an ultrafast CZT camera with and without real-time breath-hold triggering (instead of AC) by intermittent scanning confined to breath-hold at deep inspiration (using list mode acquisition). We studied the use of breath-hold triggering on the CZT camera and its ability to discriminate artefacts from true perfusion defects using AC SPECT MPI as the reference standard. Myocardial tracer uptake (percent of maximum) from CZT was compared to AC SPECT MPI by intraclass correlation and by calculating Bland-Altman limits of agreement. RESULTS: AC of SPECT MPI identified 19 apparent perfusion defects as artefacts. Of these, 13 were correctly identified and 4 were partially unmasked (decrease in extent and/or severity) by breath-hold triggering of the CZT scan. All perfusion defects verified by SPECT MPI with AC were appropriately documented by CZT with and without breath-hold triggering. This was supported by the quantitative analysis, as the correlation (r) of myocardial tracer uptake between CZT and AC SPECT improved significantly from 0.81 to 0.90 (p<0.001) when applying breath-hold triggering. Similarly, Bland-Altman limits of agreement were narrower for CZT scans with breath-hold triggering. CONCLUSION: This novel CZT camera allows real-time breath-hold triggering as a potential alternative to AC to assist in the discrimination of artefacts from true perfusion defects.