Comparison of coronary flow reserve feasibility and values in different stress echocardiography protocols: dobutamine, dipyridamole, exercise and rapid pacing

Background and aims To compare the feasibility of Coronary Flow Velocity Reserve (CFVR) assessment in the left anterior descending (LAD) artery in four types of stress echocardiography (SE): dobutamine (DOB), dipyridamole (DIP), rapid pacing (PAC) and bicycle exercise (EXE). Methods and results We subjected 369 patients (mean age: 67±11 years) to SE with DOB (up to 40 mcg/kg/min, n=230), DIP (0.84 mg/kg, n=73), PAC (n=22) or EXE (n=44). CFVR was measured as the ratio of peak diastolic coronary flow velocity (CFV) during exercise, pharmacological stress or pacing and peak diastolic CFV at rest in distal or mid LAD. The feasibility was excellent during PAC (100%), DOB (95%) and DIP (95%) and lower during EXE (73%, p<0.01 vs other groups) when assessed in all consecutive patients. When assessed in patients having readable LAD flow at rest the feasibility achieved even higher values, being however still lowered in EXE group, see Figure 1. In multivariate analysis only the EXE protocol was a predictor of LAD flow loss during SE, with OR = 8.23 (95% CI 2.17 – 31.33), p=0.0019. CFVR was lower with PAC (1.8±0.4) as compared to DIP (2.2±0.6, p=0.0061) and DOB (2.2±0.6, p=0.0025), but similar to EXE (2.0±0.6, p=0.178), and correlated best with the peak heart rate in EXE and PAC, see Figure 2. Conclusion CFVR in LAD can be obtained during all forms of SE, but the feasibility is significantly higher with PAC and pharmacological tests as compared to EXE, which was identified in our study as the independent predictor of the loss of LAD flow recording at the peak of stress test. Moreover, CFVR values were the lowest in PAC group which however encompassed the older patients with more advanced coronary artery disease. Significant correlation between HR and CFVR observed in EXE and PAC suggests that in this type of SE cut-off value of CFVR should be probably indexed to maximal HR achieved. FUNDunding Acknowledgement Type of funding sources: None. Figure 1 Figure 2

Abstract 1077: Non-Invasive MRI Assessment of Coronary Endothelial-Dependent Vasoreactivity

Endothelial-dependent coronary artery vasoreactivity is an important indicator of vascular function and is traditionally measured invasively. We hypothesized that coronary vasoreactivity can be measured non-invasively and abnormal function detected using high-field MRI. We studied 16 healthy adults (mean age ± SD 41.1±14.1 years, seven men) and eight patients with coronary artery disease (CAD, mean age 56.6±11.3, two men). To measure endothelial-dependent vasoreactivity, 3T coronary MRI was performed before and during continuous isometric hand-grip exercise, a known endothelial-dependent stressor. Cross-sectional coronary artery area and peak diastolic coronary flow velocities were quantified. In healthy adults, coronary arteries significantly dilated and increased in flow velocity with stress (baseline vs. stress area: 13.0±3.0 vs.14.9±4.1mm2, p<0.001; and velocity: 23.4±7.7 vs. 31.2±11.2 cm/s, p<0.0001). In CAD patients, the coronary arteries did not dilate (baseline vs. stress area: 18.4±5.2 vs. 17.5±4.5mm2, p=0.11) and decreased in flow velocity with stress (24.2±9.2 vs. 20.6±5.9cm/s, p=0.04). Relative changes are shown in Figure 1 . We provide first evidence that MRI and isometric handgrip exercise, an established endothelial-dependent stressor, non-invasively detects significant changes in two common indices of coronary endothelial vasoreactivity: coronary artery area and flow velocity. This novel technique demonstrates physiologic coronary artery vasodilation and increased flow velocity in healthy subjects in response to handgrip stress, and their pathologic absence in patients with CAD. Figure 1. Percent change in coronary artery area and peak diastolic coronary flow velocity from baseline to stress (isometric handgrip exercise) in healthy adults (n=16) and CAD patients (n=8). (*,† p<0.0001 healthy vs. CAD)

Reduction of coronary flow reserve in patients with increased aortic stiffness

Aortic stiffness is thought to affect coronary blood flow, but little is known about its influence on coronary flow reserve (CFR). The objective of the present study was to investigate the relationship between aortic stiffness and CFR in matched patients with and without increased aortic stiffness. Stress transoesophageal echocardiography (TEE) as the CFR measurement and coronary angiography were performed in all cases. Increased aortic stiffness was defined if elastic modulus Ep > 680 mmHg. The following patient populations free of coronary artery disease were compared: 36 subjects with normal aortic distensibility and 19 age-, sex-, and risk factor-matched patients with increased aortic stiffness. CFR was significantly reduced in patients with increased aortic stiffness as compared with cases with normal aortic distensibility (2.64 ± 1.16 vs. 2.12 ± 0.58, p <0.01). Hyperaemic diastolic flow velocities were reduced in patients with increased aortic stiffness (129.5 ± 36.6 cm/s vs. 102.1 ± 39.8 cm/s, p <0.05). Negative correlations were found between Ep and hyperaemic diastolic coronary flow velocity (r = –0.41, p < 0.01) and CFR (r = –0.21, p < 0.05). CFR is reduced in patients with increased aortic stiffness and negative correlations exist between these functional parameters.

Acute and specific collagen type I degradation increases diastolic and developed tension in perfused rat papillary muscle

Collagen degradation is suggested to be responsible for long-term contractile dysfunction in different cardiomyopathies, but the effects of acute and specific collagen type I removal (main type in the heart muscle) on tension have not been studied. We determined the diastolic and developed tension length relations in isometric contracting perfused rat papillary muscles (perfusion pressure 60 cmH2O) before and after acute and specific removal of small collagen struts with the use of purified collagenase type I. At 95% of the maximal length (95% Lmax), diastolic tension increased 20.4 ± 8.1% ( P < 0.05, n = 6) and developed tension increased 15.0 ± 6.7% after collagenase treatment compared with time controls. Treatment increased the diastolic muscle diameter by 7.1 ± 3.4% at 95% Lmax, whereas the change in diameter due to contraction was not changed. Diastolic coronary flow and normalized coronary arterial flow impediment did not change after collagenase treatment. Electron microscopy revealed that the number of small collagen struts, interconnecting myocytes, and capillaries was reduced to ∼32% after treatment. We conclude that removal of the small collagen struts by acute and specific collagen type I degradation increases diastolic and developed tension in perfused papillary muscle. We suggest that diastolic tension is increased due to edema, whereas developed tension is increased because the removal of the struts poses a lower lateral load on the cardiac myocytes, allowing more myocyte thickening.

Modified heart-lung preparation for the evaluation of systolic and diastolic coronary flow in rats

A modified heart-lung preparation of the rat, which permits measuring systolic and diastolic coronary flow separately and enables coronary compliance to be evaluated, is described. The systemic circulation was substituted by a shunt circuit, and the elastic properties of the arterial tree were mimicked by a rubber balloon. Systolic and diastolic coronary flow was evaluated from the pulmonary and aortic flow signal. Integrated phasic pulmonary flow represented right ventricular stroke volume. Integrated phasic systolic aortic flow represented left ventricular stroke volume minus that volume flowing into the coronary arteries during systole, because the aortic flow probe had to be inserted distal to the origin of the coronary vessels. Because right and left ventricular stroke volume was identical under steady-state conditions, the difference between systolic pulmonary and systolic aortic flow resulted in systolic coronary flow. Diastolic coronary flow was measured by means of the retrograde flow through the aortic flow probe. Coronary compliance was calculated according to Frank's windkessel model from coronary resistance and from central diastolic aortic pressure, which decayed exponentially after switching out the rubber balloon and the shunt circuit. It could be shown that the proportion of systolic to diastolic coronary flow depends on coronary compliance.