To compare the quantitative parameters of myocardial blood flow based on SPECT imaging and those determined by PET imaging in coronary microvascular disease (CMVD) animal models, in order to clarify the accuracy and feasibility of SPECT quantitative analysis in CMVD.

Seven Saanen milk goats (either male or female; (20±5) kg), were selected for establishing CMVD animal models by microsphere embolization. Dynamic myocardial perfusion imaging (DMPI) with one-day method of resting + ATP stress ⁹⁹Tc^m-methoxyisobutylisonitrile (MIBI) SPECT was performed before and after the modeling, respectively. One-day method of resting + ATP stress ¹³N-ammonia PET DMPI was performed after the modeling. The quantitative parameters determined by SPECT and PET after the modeling, including stress myocardial blood flow (SMBF), resting myocardial blood flow (RMBF) and myocardial flow reserve (MFR), were compared by paired t test. Parameters based on SPECT after modeling were compared with those of baseline levels. Bland-Altman analysis was applied to access the agreement between SPECT and PET.

Four of the seven experimental goats were fully imaged. The RMBF(ml·g^-1·min^-1; 1.52±0.27 vs 1.29±0.20), SMBF(ml·g^-1·min^-1; 0.74±0.19 vs 0.99±0.26), and MFR (0.53±0.16 vs 0.76±0.10) of the left ventricle (global) obtained by SPECT and PET in CMVD models were not significantly different (t values: 3.121, 1.195, 1.930, all P>0.05). Among left anterior descending branch (LAD), left circumflex (LCX) and right coronary artery (RCA), the RMBF, SMBF and MFR values quantified by SPECT and PET were neither statistically significant (t values: 0.182-2.734, all P>0.05). Bland-Altman analysis showed the quantitative parameters measured by SPECT and PET DMPI in left ventricle, LAD, LCX, RCA had a good consistency. The difference between the two methods for determining RMBF was up to 0.63 ml·g^-1·min^-1, and that of SMBF was up to 0.66 ml·g^-1·min^-1. All points are within the 95% confidence limit; MFR differs at most by 0.56, and 14/16 points were within 95% confidence limit. The RMBF (ml·g^-1·min^-1) of left ventricle measured by SPECT after modeling was not significantly different from that before modeling (1.52±0.27 vs 1.57±0.36; t=0.166, P>0.05); the SMBF (ml·g^-1·min^-1) and MFR after modeling were significantly lower than those before modeling (0.74±0.19 vs 2.34±0.89, 0.53±0.16 vs 1.39±0.31, t values: 3.836, 6.309, both P<0.05). Similar results were found when comparing the parameters of LAD/LCX/RCA after modeling with those before modeling (RMBFt values: 0.191, 0.235, 0.195, all P>0.05; SMBF/MFRt values: 0.411-19.911, all P<0.05).

The blood flow quantitative parameters measured by SPECT imaging have a good consistency with those based on PET imaging, and the myocardial blood flow quantitative analysis of SPECT can evaluate the blood flow perfusion of CMVD.

Coronary artery disease; Myocardial perfusion imaging; Tomography, emission-computed, single-photon; Technetium Tc 99m sestamibi; Positron-emission tomography; Ammonia; Goats