Comparison of exercise and pharmacological stress gated SPECT in detecting transient left ventricular dysfunction

Transient left ventricular contractile dysfunction (TLVD) is observed owing to post-exercise stunning in patients with coronary artery disease (CAD). Pharmacological stimulation differs from exercise stress because it does not cause demand ischemia. The aim of this study was to determine whether TLVD could also be seen after pharmacological stress (dipyridamole).

Of the patients in whom gated single-photon emission computed tomography (GSPECT) was performed in our institution from January 2004 to April 2007, 439 subjects with known or suspected CAD were included in the study. GSPECT was performed for all patients following exercise (group I, n = 220) or pharmacological stress (group II, n = 219) according to a 2-day (stress-rest) protocol after injection of Tc-99m methoxyisobutyl-isonitrile (MIBI). Stress, rest, and difference (stress-rest value) left ventricular ejection fractions (SLVEF, RLVEF, and DLVEF) and transient ischemic dilatation (TID) ratio were derived automatically. Summed stress score, summed rest score, and summed difference score (SDS) for myocardial perfusion were calculated using a 20-segment model and a five-point scoring system. An SDS > 3 was considered as ischemic. On the basis of the perfusion findings, patients were subdivided into a normal (group A, n = 216) and ischemia group (group B, n = 223). DLVEF and perfusion scores of all groups were compared. Relationships between DLVEF and perfusion, and between TID ratio and DLVEF were also evaluated.

Stress-induced ischemia was observed in 223 of 439 patients (50.8%). In group A, the difference between stress and rest LVEF values was not significant (P = 0.670 and P = 0.200 for groups IA and IIA, respectively). However, LVEF was significantly decreased after stress compared with rest values for group B (P < 0.0001 for groups IB and IIB). TLVD (< or =-5% for DLVEF) was observed in 20 of 216 (9%) and 81 of 223 subjects (36%) in patients in groups A and B, respectively (P < 0.0001). In group I, we found TLVD in 46 of 119 (39%) and 12 of 101 (12%) subjects, in patients with and without ischemia, respectively (P < 0.0001). On the other hand, in group II, TLVD was detected in 35 of 104 (34%) and 8 of 115 (7%) patients with and without ischemia, respectively (P < 0.0001). And also, we found significant good correlations between TID ratios and DLVEF values in four subgroups (r = -0.55, r = -0.62, r = -0.59, and r = -0.41; for groups IA, IB, IIA, and IIB, respectively, P < 0.0001 for all).

Dipyridamole is believed to be less likely than exercise to induce ischemia. However, in this study, TLVD after stress was observed following not only exercise but also pharmacological stress, consistent with ischemia.