Predictive value for adverse cardiac events of left ventricular diastolic function measured by gated myocardial perfusion imaging

Objective To study the predictive value for adverse cardiac events (ACE) of left ventricular diastolic function measured by G-MPI. Methods Patients assessed by 2 days rest-stressed G-MPI during March 2012 to May 2013 in Beijing Hospital were collected and observed for the cumulative survival rate without ACE after G-MPI. ACE included cardiac death, non-fatal MI and late revascularization (60 d after the examination). Using quantitative gated SPECT (QGS) to get the LVEF and left ventricular peak filling rate (PFR). Using 17 segments, 5-point scoring system to get the summed stress score (SSS). Cox regression analysis was used to identify the independent predictors for ACE, and Kaplan-Meier method was used to investigate the cumulative survival rate without ACE. χ2 test was also used. Results A total of 11 patients who underwent early revascularization (within 60 d after G-MPI) were excluded and 139 patients (76 males, 63 females; average age 68.73 years) were assigned to different groups for outcome analysis, and 9 (6.5%, 9/139) patients had ACE in the whole group. Cox regression analysis showed that SSS, LVEF, PFR were the independent predictors for ACE (all P<0.05). Patients with PFR<2.1 EDV/s had a higher incidence of ACE than those with PFR≥2.1 EDV/s (14.3%(7/49) vs 2.2%(2/90), χ2=7.63, P<0.05). Kaplan-Meier survival analysis showed that the cumulative survival rate without ACE was lower in patients with PFR<2.1 EDV/s than that in patients with PFR≥2.1 EDV/s (67.7% vs 95.0%), and lower in patients with LVEF<50% than that in patients with LVEF≥50%(32.0% vs 90.8%), and also lower in patients with SSS≥8 than those with SSS<8(62.0% vs 94.7%; χ2 values: 11.92, 11.27, 10.40, all P<0.01). Combining PFR with SSS and LVEF respectively, Kaplan-Meier survival analysis showed that the cumulative survival rate without ACE in patients with LVEF≥50% was higher than that in patients with LVEF<50% when PFR<2.1 EDV/s (76.7% vs 30.8%), and also higher in patients with SSS<8 than that in patients with SSS≥8 when PFR≥2.1 EDV/s (100% vs 72.2%; χ2 values: 7.14, 13.09, both P<0.01). Conclusion PFR of left ventricular diastolic function measured by G-MPI is one of the independent predictors for ACE, and it could effectively improve the predictive value for ACE if combined with LVEF and SSS. Key words: Coronary artery disease; Prognosis; Myocardial perfusion imaging; Tomography, emission-computed, single-photon; Ventricular function; Technetium Tc 99m sestamibi