Reproducibility of measurements of regional resting and hyperemic myocardial blood flow assessed with PET.

ABSTRACT

UNLABELLED PET with 13N-ammonia permits the noninvasive quantification of myocardial blood flow (MBF) in humans. The present study was done to assess the reproducibility of quantitative blood flow measurements at rest and during pharmacologically induced hyperemia in healthy individuals. METHODS: Thirty healthy volunteers (26 men, 4 women) were studied. Paired measurements of MBF at rest (n = 21), during adenosine (n = 15) and during dipyridamole (n = 7) were performed using a two-compartment model for 13N-ammonia PET. The mean difference between baseline and follow-up blood flow (% difference) was calculated to assess reproducibility. RESULTS: No significant difference was observed between resting blood flow at baseline or follow-up (15.8% +/- 15.8%; p = ns). Baseline and follow-up resting blood flow were linearly correlated (r = 0.63, p < 0.005). Normalization of resting blood flow to the rate pressure product improved the reproducibility significantly (15.8% +/- 15.8% versus 10.1% +/- 10.5%, p < 0.05). Baseline and follow-up hyperemic myocardial blood flow did not differ (11.8% +/- 9.4%; p = ns) and were linearly correlated (r = 0.69, p < 0.0005). CONCLUSION: MBF at rest can be measured reproducibly with 13N-ammonia PET. The individual response to pharmacologic stress appears to be relatively consistent. Thus, serial blood flow measurements with 13N-ammonia PET can be used to quantify the effect of various interventions on MBF and vasodilatory reserve.