Although myocardial interstitial fibrosis has been considered to play a pathogenic role in chronic heart failure (HF), the role of perivascular fibrosis, another form of fibrosis, remains to be elucidated.

We examined 64 consecutive patients with non-ischemic HF caused by hypertrophic cardiomyopathy (HCM, n=16), hypertensive heart disease (HHD, n=11), or dilated cardiomyopathy (DCM, n=37), diagnosed by both cardiac catheterization and endomyocardial biopsy (right ventricular side of the interventricular septum) in the Tohoku University Hospital between January 2001 and April 2009. We calculated the collagen volume fraction (CVF) and perivascular fibrosis ratio (PFR) in biopsy samples and also examined Thrombolysis in Myocardial Infarction (TIMI) frame count to evaluate coronary blood flow.

There was no significant correlation between CVF and PFR (r²=0.0007). Although CVF was comparable among HCM, HHD, and DCM (1.11±1.04, 1.89±1.61, and 1.41±1.48, respectively), PFR was significantly higher in HCM than in DCM (1.78±1.09 vs. 1.23±0.44, p<0.05). PFR was not correlated with cardiac function parameters, such as left ventricular (LV) ejection fraction, cardiac output, LV end-diastolic pressure, LV end-diastolic volume, aortic pressure, or pulmonary artery pressure. However, PFR was significantly correlated with coronary flow in the left anterior descending coronary artery (as evaluated by TIMI frame count) (r²=0.3351, p<0.0001, in all-cases combined population), but not with that in the left circumflex or right coronary artery. This correlation remained significant in a logistic regression model tested in 7 variables (body mass index, PVR, CVF, presence of hypertension, dyslipidemia, diabetes mellitus, and atrial fibrillation).

These results indicate that coronary perivascular fibrosis is associated with the impairment of coronary blood flow although not associated with interstitial fibrosis or cardiac function, suggesting that it can be a new therapeutic target to improve coronary microcirculation.