The present study examined the association of myocardial perfusion reserve index (MPRI) in cardiovascular magnetic resonance (CMR) with coronary microvascular dysfunction (CMD) and serum levels of markers of inflammation or endothelial activation. Twelve patients with typical angina pectoris without coronary artery disease were enrolled in this study, and CMR perfusion was analyzed using a steady-state-free-precession sequence with 3 short axis slices per heartbeat during first pass of 0.025 mmol Gadolinium-DTPA/kg body weight. The upslope of myocardial signal intensity curves was used to calculate MPRI. CMD was assessed by intracoronary Doppler flow measurement and biplane angiography. Both MPRI and CMD were assessed during endothelium-independent stimulation with intravenous adenosine and during endothelium-dependent stimulation with intracoronary infusion of acetylcholine. Serum values of soluble CD40 ligand (sCD40L), interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-alpha), soluble intercellular adhesion molecule-1 (sICAM-1), and C-reactive protein (CRP) were measured. Impaired MPRI correlated significantly with a decrease in coronary blood flow reserve after both endothelium-dependent (p = 0.033) and endothelium-independent (p = 0.022) stimulation. Serum levels above the median of all normal ranged biomarkers sCD40L, TNF-alpha, IL-6, sICAM-1 and CRP were associated with an impaired MPRI for stimulation with adenosine as well as acetylcholine. In multivariable analyses, sCD40L (p < 0.001) and TNF-alpha (p = 0.011) were significantly associated with a decrease in MPRI on adenosine, as were TNF-alpha (p = 0.016) and sICAM-1 (p = 0.022) for a decrease in MPRI on acetylcholine. MPRI on adenosine significantly correlated with MPRI on acetylcholine (p < 0.001). Therefore, the present study demonstrates safety and feasibility of an intracoronary infusion of acetylcholine during CMR perfusion analysis, thus allowing direct assessment of endothelial dependent vasomotor function at the myocardial level by CMR. Furthermore, we show that an impaired myocardial perfusion reserved in CMR is associated with established biomarkers of early atherosclerosis and significantly correlated with CMD. CMR combined with adenosine could be proposed as a non-invasive tool to evaluate CMD.