Dobutamine stress echocardiography and technetium-99m-tetrofosmin/fluorine 18-fluorodeoxyglucose single-photon emission computed tomography and influence of resting ejection fraction to assess myocardial viability in patients with severe left ventricular dysfunction and healed myocardial infarction.

The purpose of this study was to compare 2 different techniques--dobutamine-atropine stress echocardiography (DSE) and dual-isotope simultaneous acquisition (technetium-99-m-tetrofosmin/fluorine 18-fluorodeoxyglucose) single-photon emission computed tomography (DISA-SPECT)--for assessment of viable myocardium. One hundred ten patients (mean age 55 +/- 9 years) with left ventricular (LV) dysfunction (mean LV ejection fraction 27 +/- 13%) underwent both DISA-SPECT and DSE. A 16-segment scoring model was adopted for both techniques. Four types of wall motion during DSE were assessed: (1) biphasic, improvement at low dose (10 microg/kg/min) with worsening at high dose; (2) worsening, deterioration without initial improvement; (3) sustained, persistent or late improvement; and (4) no change. Viability criteria were biphasic, worsening, and sustained improvement with DSE. Viability criteria with DISA-SPECT were normal perfusion and metabolism (normal), concordantly mildly reduced perfusion and metabolism (subendocardial scar), or severely reduced perfusion and increased metabolism (mismatch). Myocardium was considered nonviable with DSE in case of unchanged wall motion, or moderate reduction or absence in both technetium-99m-tetrofosmin perfusion and fluorodeoxyglucose uptake with DISA-SPECT. Of 1,756 of 1,760 analyzable LV segments, 1,373 (78%) had severe wall motion abnormalities at baseline (severe hypokinesia, akinesia, or dyskinesia). Of these abnormal segments, 282 (21%) were considered viable during DSE (63 [5%] with biphasic response, 47 [3%] with ischemia, and 172 [13%]) with sustained improvement, whereas 1,091 (79%) were considered nonviable. With DISA-SPECT, 396 (29%) segments were considered viable (312 [23%] with matched perfusion/metabolism and 84 [6%] with mismatch), whereas 977 segments (71%) were considered nonviable. Both techniques showed agreement for viability in 201 segments and 896 were concordantly classified as nonviable. Disagreement was present in 276 segments of which 195 (71%) were nonviable with DSE and viable with DISA-SPECT. Overall agreement between the 2 techniques was 81% (kappa 0.46) in a subgroup of patients with an ejection fraction <25% 78% (kappa 0.39). Thus, DSE and DISA-SPECT show good agreement for assessing viable myocardium not influenced by resting ejection fraction. DSE underestimated the amount of viable tissue compared with DISA-SPECT.