Development of a risk score for patients with ischaemic cardiomyopathy.

BACKGROUND: Ischaemic cardiomyopathy is a leading cause of heart failure and is associated with a poor prognosis. AIM: To evaluate predictors of major adverse cardiovascular events (MACE) and to develop a risk score for the disease. METHODS: All patients with ischaemic cardiomyopathy referred to a tertiary hospital between 2010 and 2018 for stress-rest gated single-photon emission computed tomography (SPECT) were included retrospectively (n=747). Clinical and gated SPECT-derived variables were analysed as predictors of MACE, a combined endpoint of cardiovascular mortality, heart failure hospitalization or myocardial infarction during follow-up. A multivariable Cox model using backwards stepwise regression with competing risks was used to select the best parsimonious model. RESULTS: After a median follow-up of 4.7 years, 313 patients had MACE (41.9%). Independent predictors of MACE were previous heart failure admission, worsening angina or dyspnoea, estimated glomerular filtration rate ≤60mL/min/1.73 m2, age>73 years, diabetes, atrial fibrillation, end-diastolic volume index>83mL/m2 and>12% of scarred myocardium. A risk score ranging from 0 to 12 classified patients as at intermediate risk (event rate of 4.0 MACE per 100 person-years), high risk (11.3 MACE per 100 person-years) or very high risk (27.8 MACE per 100 person-years). The internally validated area under the curve was 0.720 (95% confidence interval 0.660-0.740) and calibration was adequate (Hosmer-Lemeshow test P=0.28) for MACE. CONCLUSIONS: In patients with ischaemic cardiomyopathy, a simple risk score using dichotomic and readily available variables obtained from clinical assessment and gated SPECT accurately predicts the risk of MACE.

Impact of revascularization guided by functional testing in ischaemic cardiomyopathy.

AIMS: The burden of ischaemia is a risk factor for adverse outcomes in ischaemic cardiomyopathy (ICM) but is not systematically tested when deciding on revascularization. Limited data exists in patients with ICM regarding the interaction between ischaemia and early coronary revascularization (ECR). This study sought to determine if the burden of ischaemia modifies the outcomes of ECR in ICM. METHODS AND RESULTS: Consecutive patients with ICM (left ventricular ejection fraction < 40%) with a stress-rest gated single-photon emission computed tomography (N = 747) were followed-up for ECR and major cardiovascular events (MACEs, cardiovascular death, myocardial infarction, or heart failure hospitalization). A 1:1 matched population was selected using a propensity score for ECR. The interaction between ischaemia and ECR was evaluated in the matched cohort. In the initial cohort, 131 patients underwent ECR. Of them, 109 were matched to non-ECR patients. After a median follow up of 4.1 years, 102 (46.8%) patients experienced a MACE. The effect of revascularization on MACE was dependent of the percent of ischaemia (P for the interaction at 10% ischaemia = 0.021), so that a trend towards a decreased risk of MACE was seen in patients with >10% of ischaemia [hazard ratio (HR) = 0.59 (0.30-1.18)], whereas a non-significant increase of MACE was observed in those with <10% ischaemia (HR = 1.67 [0.94-2.96]). CONCLUSIONS: In a contemporary cohort of patients with ICM, the beneficial effects of ECR may be mediated by the percent of ischaemia. This study supports stress testing in ICM and an ischaemia-guided approach for ECR.

Q waves in ischemic cardiomyopathy.

Q waves may be observed in the absence of non-viable tissue. However, their scintigraphic translation in patients with ischemic cardiomyopathy (ICM) has not been properly assessed. This study sought to establish the determinants of Q waves in the absence of non-viable tissue and the diagnostic accuracy in this population. A retrospective study enrolling 487 consecutive patients (67.0 [57.4 - 75.4] years), with ICM, LVEF < 40% and narrow QRS who underwent stress-rest 99 m-Tc SPECT was conducted. A 17-segment model for myocardium was used: Myocardium was divided in basal (1 to 6), mid (7 to 12), apical (13 to 16) and apex (17) segments. Non-viable tissue was defined as a severe perfusion defect without systolic thickening. Patients with Q waves (65.7%) had more non-viable tissue, more extensive scar and less ischemia. Q waves had a moderate correlation with non-viable tissue (AUC = 0.63) and were associated with the extension of the scar. After excluding patients with non-viable tissue in any myocardial segment, Q waves were observed in 51.9% of the patients, of which 78.1% had a scar fulfilling viability criteria. The presence of Q waves was associated with the location of these scars in a base-to-apex axis (OR = 1.88 [1.35-2.62] for segment towards the apex) and their extent (OR = 1.19 [1.05 - 1.35] for each segment). In patients with ICM, Q waves discriminate poorly viable from non-viable tissue. Q waves in this population may be due to extensive scars fulfilling viability criteria located in apical segments.