Effects of left ventricular size on the accuracy of diastolic parameters derived from myocardial perfusion SPECT: comparison with tissue Doppler echocardiography.

Background: Impaired left ventricular (LV) diastolic function is a sensitive and early sign of myocardial ischemia. We evaluated the effects of LV size on the accuracy of diastolic parameters derived from SPECT.

Methods: The study population consisted of 151 patients with known or suspected coronary artery disease who underwent both SPECT and transthoracic echocardiography. Peak filling rate (PFR), one-third mean filling rate (1/3 MFR) and the ratio of time to PFR to the RR interval (TPFR/RR) were calculated by quantitative gated SPECT. Peak early mitral annular velocity (e') was used as the reference standard of LV diastolic function.

Results: There were 43 patients with end-systolic volume (ESV) of ≤10 ml, 43 patients with ESV of 11-20 ml and 65 patients with ESV of >20 ml. There were significant differences in PFR (p < 0.001), 1/3 MFR (p < 0.001) or TPFR/RR (p = 0.01) among the 3 groups. These diastolic parameters were increased with decreased LV size. In overall patients, PFR (r = 0.24, p = 0.003) and 1/3 MFR (r = 0.31, p < 0.001) were positively, and TPFR/RR (r = -0.23, p = 0.004) was inversely correlated with e'. Multivariate linear regression analyses showed that male gender (β = -0.14, p = 0.07; β = -0.16, p = 0.04), ESV (β = -0.63, p < 0.001; β = -0.45, p < 0.001) and e' (β = 0.36, p < 0.001; β = 0.40, p < 0.001) were significant factors associated with PFR or 1/3 MFR. Multivariate linear regression analysis also showed that ESV (β = -0.17, p = 0.03) and e' (β = -0.21, p = 0.01) were significant factors associated with TPFR/RR.

Conclusions: Our data suggest that PFR, 1/3 MFR and TPFR/RR derived from SPECT are correlated with e' as the reference standard of LV diastolic function, but are overestimated in small-sized heart. LV size should be taken into consideration when interpreting these diastolic parameters.