Epicardial adipose tissue thickness as a predictor of impaired microvascular function in patients with non-obstructive coronary artery disease.

OBJECTIVE

To determine if increased epicardial adipose tissue (EAT) measured by cardiac CT could be associated with impaired myocardial flow reserve (MFR) in patients with non-obstructive coronary artery disease (CAD).

BACKGROUND

Studies have shown that EAT volume is related to epicardial obstructive CAD, myocardial ischemia and major adverse cardiac events. However, the association between EAT with coronary microvascular dysfunction and impaired MFR has not been well clarified.

METHODS

Consecutive patients who underwent Rb-82 positron emission tomography (PET), coronary artery calcium (CAC) scoring and non-invasive coronary computed tomography angiography (CCTA) were screened. PET scans were analysed for standard myocardial perfusion (MPI) and MFR. CCTA results were analysed and only patients with non-obstructive CAD (<50% luminal diameter stenosis) were included. EAT thickness and volumes were measured from CT scans.

RESULTS

Of 137 patients without obstructive CAD by CCTA and with normal Rb-82 PET relative MPI, 26 (19.0%) patients had impaired MFR < 2 and 87 (64%) patients had CAC. EAT(thickness), EAT(volume) and CAC values were higher in patients with impaired MFR < 2 than those with normal MFR ≥ 2 (6.7 ± 1.6 mm vs 4.4 ± 1.0 mm, P < .0001; 119.0 ± 25.3 cm(3) vs 105.8 ± 30.5 cm(3), P < .04 and 508.9 ± 554.3 vs 167.8 ± 253.9, P < .0001, respectively). However, EAT(thickness) had a stronger negative correlation with MFR than EAT(volume) and CAC (r = -0.78 vs r = -0.25 and ρ = -0.32, P < .0001). With multivariable logistic regression analysis, only EAT(thickness) was independently associated with impaired MFR (OR 20.7, 95% CI 4.9-87.9, P < .0001). Importantly, the receiver-operator characteristic (ROC) curves demonstrated a superior performance of EAT(thickness) vs EAT(volume) and EAT(thickness) vs CAC in detecting impaired MFR (AUC: 0.945 vs 0.625, difference between AUC: 0.319, P < .0001; AUC: 0.945 vs 0.710, difference between AUC: 0.235, P < .0006, respectively). On ROC curve analysis, an EAT(thickness) cut-off value > 5.6 mm was optimal in detecting impaired MFR with a sensitivity and specificity of 81% and 92%, respectively.

CONCLUSIONS

Increased EAT appears to be associated with impaired MFR. This parameter may help improve detection of patients at risk of microvascular dysfunction.