Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Mar 1.
Published in final edited form as: Circ Cardiovasc Imaging. 2016 Mar;9(3):e003954. doi: 10.1161/CIRCIMAGING.115.003954

Simultaneous Non-Invasive Assessment of Systemic and Coronary Endothelial Function Iantorno et al: Cardiac MRI and Endothelial Function

Micaela Iantorno 1,2,#, Allison G Hays 1,#, Michael Schär 3, Rupa Krishnaswamy 1, Sahar Soleimanifard 3,4, Angela Steinberg 1, Matthias Stuber 3,5, Gary Gerstenblith 1, Robert G Weiss 1,3
PMCID: PMC4839535  NIHMSID: NIHMS754885  PMID: 26919997

Abstract

Background

Normal endothelial function is a measure of vascular health and dysfunction a predictor of coronary events. Nitric Oxide (NO)-mediated coronary artery endothelial function (CEF), as assessed by vasomotor reactivity during isometric handgrip exercise (IHE), was recently quantified noninvasively with MRI. Because the internal mammary artery (IMA) is often visualized during coronary MRI we propose the strategy of simultaneously assessing systemic and coronary endothelial function noninvasively by MRI during IHE.

Methods and Results

Changes in cross-sectional area (CSA) and blood flow (BF) in the right coronary artery (RCA) and the IMA in 25 CAD patients and 26 healthy subjects during IHE were assessed using 3T MRI. In 8 healthy subjects a NO synthase inhibitor was infused to evaluate the role of NO in the IMA-IHE response. Inter-observer IMA-IHE reproducibility was good for CSA (R=0.91) and BF (R=0.91). In healthy subjects, CSA and BF of the IMA increased during IHE and these responses were significantly attenuated by L-NMMA (p<0.01 vs. placebo). In CAD patients, the RCA did not dilate with IHE and dilation of the IMA was less than that of the healthy subjects (p=0.01). The BF responses of both the RCA and IMA to IHE were also significantly reduced in CAD patients.

Conclusions

MRI-detected IMA responses to IHE primarily reflect NO-dependent endothelial function, are reproducible and reduced in CAD patients. Endothelial function in both coronary and systemic (IMA) arteries can now be measured noninvasively with the same imaging technique and promises novel insights into systemic and local factors affecting vascular health.

Keywords: MRI, coronary endothelial function, systemic endothelial function, coronary artery disease

In response to certain stresses the healthy endothelium releases nitric oxide (NO) which induces local vascular smooth muscle dilation, inhibits platelet aggregation, attenuates inflammation, and decreases cellular proliferation1. Endothelial dysfunction is characterized by decreased NO bioavailability, occurs early in the development of atherosclerosis, predicts adverse cardiovascular events2, and is a potential target for medical interventions2-7. Peripheral arteries are more accessible to study than the coronary arteries but there is only a modest correlation between coronary and peripheral endothelial function measures8,9 and the coronary vascular bed differs significantly from the systemic vasculature9,10. Patients with coronary artery disease (CAD), in fact, manifest a paradoxical coronary artery vasoconstrictor response to endothelial-dependent stressors that normally cause vasodilatation11 whereas the brachial arteries of CAD patients vasodilate less in response to endothelial dependent stressors than do those of healthy individuals8,12. A recent meta-analysis identified a variable association between cardiovascular events and endothelial dysfunction, depending on whether a central or peripheral vascular bed was studied13. Moreover, studies that compared vasoreactivity of the brachial and the coronary vascular beds were performed using different imaging modalities at different times8. The measurement of endothelial function in systemic arteries that do not develop atherosclerosis, like the brachial artery, provides information about systemic vascular health, whereas CEF measures offer insights into the contributions of systemic and local coronary factors including the presence of coronary atherosclerosis. Ideally, endothelial function would be assessed in systemic and coronary arteries at the same time using the same endothelial-dependent stressor and imaging technology13.

Recently, the combination of new non-invasive 3T coronary MRI methods and isometric handgrip exercise (IHE), an endothelial-dependent stressor, has been reported as a means to noninvasively and reproducibly quantify CEF11,14-16. Furthermore, we recently demonstrated in vivo that the coronary artery response to IHE is reproducible and primarily NO-mediated because it is blocked by the nitric oxide synthase (NOS) inhibitor monomethyl-L-arginine (L-NMMA) in healthy subjects17.

The internal mammary artery (IMA) is a systemic vessel that rarely develops atherosclerosis10, is often used as a coronary artery graft, and has been used to study systemic endothelial function18-20. Because the right and left IMA are visible in many coronary MR images, especially in axial planes which also intersect the right coronary artery (RCA), we posited that measurements of the vasodilator and blood flow responses of the IMA could be obtained at the time of CEF measures and serve as an index of systemic endothelial function. We tested the hypotheses that: 1) the IMA vasoreactive response to IHE is NO-dependent, i.e. the response can be blocked by L-NMMA, a nitric oxide synthase (NOS) inhibitor and thus reflects NO-mediated endothelial function, 2) the IMA-IHE response is reproducible, 3) IMA endothelial function is reduced in CAD patients compared to that of healthy subjects, and 4) among CAD patients, the endothelial-dependent IMA vasoreactive response to IHE differs from the coronary response.

METHODS

Participants

The protocol was approved by the Institutional Review Board of Johns Hopkins Medicine and complies with the Declaration of Helsinki. An Investigational New Drug Application (IND) was obtained from the Food and Drug Administration (#119574) for the administration of L-NMMA. All participants provided written informed consent. All subjects were outpatients with no known contraindications to MRI. Healthy subjects were those without history of CAD and for those over the age of 50 years with an Agatston coronary artery calcium score <10 by computed tomography or an exercise stress test negative for inducible ischemia. CAD patients were individuals with stable coronary artery disease documented on prior coronary x-ray angiography or computed tomography angiography (stenosis of 30% to 70%). The segment of the coronary artery selected for MRI measures of area, velocity and blood flow in patients with CAD had no more than a 30% luminal stenosis.

Study protocol

All participants underwent MRI in the morning after an overnight fast (>8 hours) and prior to administration of any prescribed vasoactive medications. MR images were taken perpendicular to a proximal or mid well-visualized linear segment of the IMA and native RCA that had not undergone prior intervention or had a significant stenosis. To ensure that slice orientation was perpendicular to the coronary and internal mammary artery, double oblique scout scanning was performed as previously reported11. Both the RCA and an IMA were imaged for cross sectional area or flow velocity in cross-section during the same single breath-hold cine sequence although in some instances they were imaged in different sequences if not parallel to one another. Either the left or the right IMA was chosen, depending on which was parallel to the RCA segment. All acquisitions were performed during a period of minimal motion during the cardiac cycle visually determined from cine axial images. Baseline images were acquired at rest for cross-sectional RCA and IMA area and velocity measurements, followed by repeat imaging at the same anatomic locations during 4-7 minutes of continuous IHE at 30% of maximum grip strength15. IHE was performed using an MRI-compatible handgrip dynamometer (Stoelting, Wood Dale, IL, USA) under direct observation and coaching by a research nurse. Heart rate and blood pressure were measured throughout during the study using a noninvasive and MRI-compatible electrocardiogram and blood pressure monitor (In vivo; Precess, Orlando, FL). The rate pressure product (RPP) was obtained as previously described15. The MRI endpoints were: cross sectional area (CSA), flow velocity (FV) and blood flow (BF) measures as described below.

To test the study hypotheses, we conducted 2 protocols outlined here:

1) Role of NO in the IMA vasomotor response to IHE: the L-NMMA Protocol

To assess whether the IHE-induced vasoreactive changes of the IMA are NO-mediated and hence endothelial dependent, 8 healthy individuals underwent IMA imaging before and during IHE while intravenous saline (placebo) was infused. After 10 minutes of post-IHE recovery, without repositioning, each subject then received an intravenous infusion of the NO synthase inhibitor, L-NMMA, at a dose of 0.3 mg/kg/min, as previously described21. A new set of baseline IMA images was obtained 5-10 minutes following initiation of the L-NMMA infusion. Each subject then performed a second IHE during continued L-NMMA infusion, and IMA imaging was repeated at the same location. The entire L-NMMA infusion lasted 15-22 minutes and the entire MRI-CEF-L-NMMA protocol lasted about 60 minutes.

2) IMA and RCA MRI vasomotor responses to IHE in healthy volunteers and patients with CAD

Healthy subjects (n=26) and CAD patients (n=25) were consecutively enrolled and both the right coronary artery (RCA) and an IMA were imaged in cross-section for CSA and FV at rest, followed by repeat imaging during IHE. Inter-observer reproducibility analysis of IMA measurements (CSA, FV, BF) was performed in a randomly selected subset of both healthy volunteers (N=6) and CAD patients (n=5) and none from the subset were excluded from analysis.

MRI

A commercial human 3.0 Tesla whole-body MR scanner (Achieva, Philips, Best, NL) with a 32-element cardiac coil for signal reception was used. Cross-sectional anatomical and flow velocity encoded spiral MRI were obtained using single breath-hold cine sequences15. MRI parameters for anatomical imaging were: repetition time (TR)=18ms, echo time (TE)=2.1ms, radio frequency (RF) excitation angle=20°, acquisition window=13ms, 17-21 spiral interleaves/cine frame, spatial resolution=0.89×0.89×8.0 mm3, and breath-hold duration ~14–24s. MRI parameters for the velocity measurements were: TR=40ms, TE=3.5ms, RF excitation angle=20°, acquisition window=33ms, 9-11 spiral interleaves/cine frame, spatial resolution=0.8×0.8×8.0mm3, velocity encoding=35-75cm/s, and breath-hold duration=14-24s.

Image analysis

Baseline and IHE stress images were analyzed for RCA and IMA CSA using semi-automated software (Cine version 3.15.17, General Electric, Milwaukee, WI, USA). A circular region-of-interest around the RCA and the IMA was traced during a period of least coronary motion over three sequential images. The three values (measured in mm2) were then averaged. The computer algorithm employed an automated full width half maximum algorithm for CSA measurements. For flow measurements, velocity measurements of the same baseline and IHE stress RCA and IMA images were made using commercially available software (QFLOW Version 3.0, Medis, The Netherlands). A region of interest was traced using semi-automated software around a cross-sectional RCA to obtain peak diastolic coronary flow velocity and of the IMA for peak systolic flow velocity (mean velocity of lumen pixels at peak flow), both referred to as FV (flow velocity) hereafter. Velocity was measured in cm/s and coronary and IMA blood flows (=BF, in mL/min) were calculated and converted to units of mL/minute using the adapted equation: cross-sectional area x peak FV × 0.322. Segments with poor image quality (blurring due to artifact/patient motion) on either baseline or stress exams were excluded from analysis.

Statistics

The data were tested for normality using the Shapiro-Wilk test. Parametric (Student’s t-test) and non parametric testing (Wilcoxon signed rank test for paired data and Wilcoxon rank sum test for non paired data) were used when appropriate for normally distributed and not normally distributed data, respectively, to compare the response to IHE from baseline measurements in the RCA and IMA of healthy subjects and patients with CAD for area, velocity and flow measurements. A paired t-test was used to compare IHE-induced IMA changes during placebo to those during L-NMMA infusion. For comparisons among four groups, one-way ANOVA and Kruskal-Wallis were used for parametric and non-parametric comparisons, respectively, with Bonferroni adjustment used for pair-wise comparisons. Linear regression analysis was performed to assess inter-reader reproducibility and the mean differences were displayed with Bland-Altman plots23. Reproducibility was also assessed using intra-class correlation coefficient (ICC)24. Statistical significance was defined as a two-tailed p-value ≤0.05. Results are presented as mean±standard error of the mean (SEM), unless otherwise indicated.

Results

All subjects completed the study. An example of typical changes seen in area and velocity with IHE in a healthy volunteer is shown for the IMA and RCA in Fig. 1. Modest increases in cross-sectional area (CSA) and larger increases in blood flow velocity (shown by darker pixels in the velocity encoded phase contrast images indicating increased velocity in the caudal direction) occur during IHE in healthy subjects.

Fig 1. MR anatomical and flow velocity images of right coronary artery and internal mammary artery at rest and during isometric handgrip exercise.

Fig 1

Open in a new tab

A) Scout scan obtained parallel to the RCA and IMA in a healthy subject is shown together with the location for cross sectional imaging of the two vessels (red line). B) Cross-sectional image perpendicular to the RCA (green box) and IMA (red box) is shown. Magnified cross-sectional image of the RCA and the IMA (shown by green and red boxes, respectively) at rest (C) and during IHE (D). Magnified flow velocity image of the IMA in the same subject is shown at rest (E) and during IHE (F) during systole. Magnified coronary flow velocity image of the RCA in the same subject is shown at rest (G) and during IHE (H) in diastole. The signal phase is proportional to flow velocity with the darker pixels in the velocity phase contrast images during IHE indicating higher velocity in the caudal direction of the IMA and RCA.

1) Role of nitric oxide in mediating the IMA response to IHE

Subject Characteristics and Hemodynamic Effects of L-NMMA

Eight healthy subjects underwent IMA imaging with the L-NMMA protocol (age: 30±4 years). The baseline RPP increased significantly with IHE (p<0.0001) and returned to baseline during the recovery period (p=0.8). During L-NMMA infusion mean RPP was not different from that prior to L-NMMA. The increase in RPP during IHE was similar in the absence and presence of L-NMMA.

L-NMMA infusion blocks the vasodilatory response of the IMA to IHE

The IMA responses to IHE in 8 healthy subjects during placebo and L-NMMA infusion are presented in Figure 2B.There is an approximate 15% increase in CSA, 30% increase in coronary FV and almost 50% increase in CBF during placebo infusion (Fig 2B). IMA CSA increased from a baseline: of 8.5±0.6mm2 to: 9.8±0.7mm2;p<0.001) during IHE. However, there was no significant increase in IMA CSA when IHE was repeated during L-NMMA infusion (2nd baseline: 9.0±0.8mm2 vs. L-NMMA-IHE: 9.2±0.8mm2;p=0.2). In relative terms, %CSA change with IHE was 15.5±2.2% with placebo vs. 2.3±1.3% with L-NMMA (p<0.001).

Fig 2. Intravenous infusion of monomethyl-L-arginine, a nitric oxide synthase inhibitor, blocks isometric handgrip exercise-induced vasodilation and the increase in blood flow in the internal mammary artery of healthy volunteers.

Fig 2

Open in a new tab

A) Diagram illustrating MRI L-NMMA study. B) Vasoreactive changes during IHE for placebo (blue striped) and L-NMMA (yellow) infusions showing that L-NMMA blocks IMA vasodilation and increase in velocity and blood flow with IHE. Comparisons of placebo- and L-NMMA responses were performed with a paired t-test. Abbreviations: L-NMMA: monomethyl-L-arginine; IHE: Isometric Handgrip Exercise

L-NMMA infusion blocks the IMA increase in blood velocity and flow with IHE

Peak systolic velocity and blood flow in the IMA significantly increased with IHE during placebo infusion. Velocity increased from a baseline of 21.4±3.3cm/sec to 27.4±4.2cm/sec (p<0.01) with IHE while blood flow increased from a baseline of 50.2±3.1ml/min to 74.1±4.7ml/min (p<0.0001) with IHE during placebo. L-NMMA infusion did not change baseline IMA flow but completely blocked the IHE-induced increases in velocity and flow (Fig 2). This suggests that the normal vasoactive IMA response to IHE is predominantly NO-mediated.

2) Reproducibility of IMA measurement

Changes of the IMA CSA, FV and BF during IHE were analyzed by two observers (MI, AH). The results strongly correlated for the %CSA, %FV and %BF changes with IHE (Fig 3A-C). The Bland-Altman analysis (Fig 3D-F) and intra-class correlation coefficients (ICC) for %CSA, %FV %BF change with stress in the IMA (ICC=0.89, 0.97 and 0.93, respectively) indicated excellent confidence of agreement and little variability between the two measures.

Fig 3. Inter-observer Reproducibility of Noninvasive MRI Measures of the IMA.

Fig 3

Open in a new tab

(A-C) Linear regression showing strong correlation between % area change with IHE between observer 1 and observer 2 (A), % velocity (B) and % flow change (C) between first and second observer. D-F) Bland-Altman plots indicate good confidence of agreement and little variability between the two measures. Lines represent 95% confidence of agreement.

3) IMA and RCA MRI vasomotor responses to IHE in healthy volunteers and CAD patients

Subject Characteristics

We compared the responses of 26 consecutive healthy subjects (age:45±3.5 years) and 25 patients with CAD (61±1.5 years). The study subject characteristics are summarized in the Table. To age-match healthy subjects with CAD patients, the results from an older subset of the original healthy individuals (n=12, 61±3 years) were also compared to those from the CAD patients (61±1.5 years; p=NS) and those data appear in the Supplemental Materials section.

Hemodynamic effects of IHE in healthy volunteers and patients with CAD

IHE induced significant and similar hemodynamic changes in healthy subjects and CAD patients. In the healthy group, we observed a mean 35.4±4.6% increase in RPP which was not different than the mean 28.7±3.9% RPP change with IHE in CAD patients (p=0.3 vs. healthy subjects).

Coronary and IMA area changes

Both the RCA and IMA in healthy subjects dilated significantly in response to IHE (p<0.001 and p<0.0001 from baseline, respectively, Fig4). In contrast, the RCA in CAD patients did not vasodilate in response to IHE (p=0.4 for CSA compared to baseline), as we previously reported11,15. However, the IMA did vasodilate with IHE in CAD patients (p<0.0001 from baseline) although significantly less than that in healthy volunteers (%IMA change from baseline:16.4±2.5% in healthy vs 9.1±1.6% in CAD patients; p=0.02; Fig4B). The RCA CSA response to IHE was significantly less than the IMA response in CAD patients (CAD patients:RCA %CSA change with IHE:-0.9±1.7% vs IMA %CSA change with IHE:+9.1±1.6%;p<0.001,Fig4B), but not in healthy subjects (healthy: RCA %CSA change with IHE:11.7±2.0% vs IMA %CSA change with IHE:16.4±2.5%;p=NS).

Fig 4. Peripheral and Coronary Endothelial Function in Healthy Volunteers and Patients with CAD-CSA Change.

Fig 4

Open in a new tab

A) Protocol diagram illustrating MRI study. B) Summary results for mean CSA changes in the IMA (striped bars) and RCA (solid bars) during IHE for healthy volunteers (blue) and patients with CAD (red). Analysis performed with Kruskal-Wallis testing with Bonferroni adjustment for pair-wise comparisons.

Coronary and IMA velocity and blood flow measures

For the IMA, both FV and BF increased significantly with IHE in healthy individuals (p<0.001). In contrast, in CAD patients, we observed an attenuated but significant increase in IMA BF with IHE from baseline (p<0.001). The IMA changes in CAD patients were less than those of healthy individuals (p<0.001 CAD IMA vs. healthy IMA; Fig5B). Consistent with prior reports, the RCA vasoreactive responses were characterized by a significant increase in FV and BF in healthy subjects, but not in CAD patients with IHE (Fig 5B,C)11,15,16. When comparing the two vascular beds, the coronary BF response was less than the IMA response in CAD patients, but did not reach statistical significance (p=0.07, Fig 5C). When the results on healthy subjects and CAD patients were combined into a single group, there was a statistically significant relationship between IMA and RCA IHE-responses (Fig 6). However the correlations were not generally significant when the groups were considered separately, suggesting that the IMA-RCA correlation is highly influenced by group differences between healthy subjects and CAD patients rather than by a close fundamental relationship between IMA and RCA responses.

Fig 5. Peripheral and Coronary Endothelial Function in Healthy Volunteers and Patients with CAD-Flow Change.

Fig 5

Open in a new tab

A) Protocol diagram illustrating MRI study. Summary results for mean peak systolic velocity changes in the IMA and peak diastolic velocity changes in the RCA (B) and for mean blood flow changes with IHE (C) during IHE for healthy subjects (blue) and CAD patients (red). Analysis performed with Kruskal-Wallis testing with Bonferroni adjustment for pair-wise comparisons.

Fig 6. Relationship between Coronary and Internal Mammary Artery Endothelial Function.

Fig 6

Open in a new tab

Individual results for IMA and RCA responses to IHE for CSA (A), flow velocity (B), and blood flow (C) changes in healthy subjects (blue) and CAD patients (red). When all participants were combined into a single group there was a statistically significant relationship between IMA and RCA responses (although the significance of the CSA relationship depended on one point remote from the others). However, the correlation between IMA and RCA responses were not significant for each group considered alone, suggesting the overall correlation was primarily related to group differences between healthy subjects and CAD patients rather than a close fundamental relationship between IMA and RCA responses.

Discussion

Abnormal vasomotor responses of systemic and coronary arteries to endothelial-dependent stressors predict subsequent cardiovascular events but the direction, magnitude and prognostic value of the responses often differ between systemic and coronary arteries8,9,25,26. Because systemic and coronary endothelial function measures are typically obtained at different times, with different endothelial-dependent stressors, and assessed with different techniques8,14, it is difficult to know whether different responses between systemic and coronary arteries represent true disparities in local vascular biology or are simply due to differences in the stimulus, means of detection and/or conditions at the time of study. In this paper we describe the first non-invasive means to simultaneously assess endothelial function in systemic and coronary circulations. The measures of systemic (IMA) and coronary (RCA) endothelial function are obtained at the same time, in response to the same stimulus, and detected with the same imaging technology. We demonstrate that the IMA response to IHE is indeed predominantly NO-mediated, reproducible between observers, differs between healthy subjects and those with CAD, and differs from the response of the coronary arteries in CAD patients.

Role of nitric oxide in mediating the IMA response to IHE

We recently reported that the coronary response to IHE is measurable with MRI11,15,16 and mediated by NO in healthy individuals17. Because it was not previously known whether the IMA response to IHE is NO-mediated27, we report here that the NOS inhibitor L-NMMA abolishes the normal IMA vasodilatory and blood flow IHE responses in healthy subjects (Fig 2). Our results are consistent with prior studies indicating that L-NMMA blocks approximately 70% of the brachial artery macrovascular vasodilatory response in healthy subjects28,29. Likewise, the microvascular peripheral response during IHE is also blocked by L-NMMA infusion28. Together these findings in healthy subjects demonstrate that IHE is a predominately NO-mediated endothelial dependent stressor for both the coronary and systemic circulations. Our assumption that NO also acts in the CAD population is reasonable, although not directly demonstrated. We did not administer L-NMMA to CAD patients because L-NMMA reduces NO synthesis and one would expect minimal or no vascular effect in an already severely NO-deficient state like CAD, especially when monitoring the coronary vasoactive dilatory and/or flow effects of IHE that are already absent in CAD patients. In addition, L-NMMA may pose risk in CAD patients. Thus continuous IHE as described here can be used with MRI or potentially other noninvasive imaging modalities to simultaneously probe coronary and systemic NO-mediated endothelial function.

IHE-induced vasoreactivity of the IMA is reduced in CAD patients compared to healthy subjects

We observe a significant reduction of IHE-induced vasoreactivity of the IMA in CAD patients compared to that of healthy subjects (Figs 4 and 5). These observations are in line with prior in vitro studies10,30 as well as in vivo studies showing an impaired IMA vasodilatory response to acetylcholine infusion in CAD patients as compared to controls31. Berkenboom et al showed that selective infusion of the IMA of L-arginine, a NO precursor, reversed the impaired IMA response to acetylcholine in CAD patients, suggesting that decreased NO plays a critical role in the etiology of the diminished IMA response in those patients31. The IMA vasoreactive response generally resembles the brachial response in direction and magnitude in diseased states32.

Systemic vs. coronary endothelial function: differences in vascular properties

Among CAD patients, we observe that coronary arteries fail to dilate and sometimes vasoconstrict in response to IHE whereas the IMA responds with reduced vasodilation, similar to the attenuated brachial response during forearm cuff occlusion11,15,32. One obvious difference between the coronary and IMA vascular beds is that the latter does not develop atherosclerosis while the coronary arteries do10. Furthermore, the vasoreactive responses of coronary and systemic vessels were previously compared, albeit at different time points and imaging modalities, and the correlation between systemic and coronary endothelial function was not strong8,9,14. When the endothelial responses of the IMA and RCA were compared for individual subjects (Fig 6), the correlation was not significant for healthy subjects or for CAD patients alone, although significant when combined. One prior study showed that intrarterial infusion of L-arginine did not affect acetylcholine-induced vasodilation in the coronaries of healthy individuals and CAD patients while it augmented the acetylcholine-induced increase in forearm blood flow in the two groups9. Thus, the response to acetylcholine infusion differed between coronary and systemic arteries suggesting that mechanisms for vasodilation may vary between the vascular beds9. Differences in vascular properties between coronary and systemic vessels may be due to a variable amount of NO production and/or bioavailability. Finally, although atherosclerosis is often regarded as a systemic process, the often-disparate vasoreactive responses of coronary and systemic arteries suggest differences in local milieu10. Prior observations that coronary arteries display a heterogeneous coronary endothelial response depending on the degree of atherosclerotic disease11, 15 suggest that local factors may contribute to local atherosclerotic plaque formation despite the exposure of all coronary segments to identical systemic factors. Therefore the study of these two vascular territories may shed important insights into their relationship and how external and local factors may influence them in different ways. Because endothelial function is often considered a “barometer” of vascular health and responds rapidly to protective strategies, we postulate that this MRI technique could be used in the future to guide therapy (e.g. lowering LDL not to an LDL number but until endothelial function improves), predict future events in at risk populations, and, importantly, to rapidly test the ability of new strategies to improve systemic and coronary vascular health.

Limitations

Sample size is relatively modest in the L-NMMA study but yet large enough to show highly statistically significant responses. Although the RCA was the only coronary artery studied here with the IMA, it is possible to acquire coronary endothelial function measures of the left coronary artery system during the same IHE session11,15,16. While phasic blood flow changes during the cardiac cycle may differ between RCA and LAD, the latter does not lie orthogonal to the axial plane like the IMA and RCA and thus the LAD would require an additional breath-hold acquisition. Local coronary factors likely play a major role in influencing macrovascular regional CEF but we cannot exclude the possibility, based on the current data alone, that differences in distal NO production in the microvasculature could affect shear stress and thereby proximal coronary artery changes to IHE. Regardless, this new approach offers a reproducible measure of NO-mediated endothelial function in coronary and systemic vascular territories concurrently with the same stressor, under the same conditions and detected with the same imaging technology. In future studies it would be useful to compare these measures of systemic endothelial function derived from IMA to those of other, more commonly studied arteries (i.e. brachial and femoral) and to determine the extent to which these measures are associated with future cardiovascular events.

Conclusions

In summary we report here the first non-invasive approach for concurrently measuring systemic and coronary vascular endothelial function. The IMA-IHE response is predominantly NO dependent, can be measured non-invasively with MRI simultaneously with CEF measures, and is reproducible both in healthy volunteers and CAD patients. Importantly, the IMA-IHE response in CAD patients differs significantly from that in healthy subjects. This noninvasive approach promises a more complete assessment of vascular health than measures of endothelial function in a single vascular territory and will enable the systematic study of interventions designed to improve endothelial function over time.

Supplementary Material

Supplemental Material

Table.

Demographics

Characteristics Healthy
Volunteers
(n=26)		 Healthy
Volunteers
Age
Matched
with CAD
Patients
(n=12)		 CAD
Patients
(n=25)
Age (years), Mean ± SEM 45 ± 3.5 61 ± 3 61 ± 1.5
Male Gender (%) 8 (31) 5 (42) 21 (84)
Previous MI 0 0 5 (20%)
PCI/Stents 0 0 8 (32)
CABG surgery 0 0 0
Dyslipidemia 2 (7) 2 (17) 24 (96)
HTN 2 (7) 2 (17) 24 (96)
Diabetes 0 0 3 (12)
ACE-inhibitor 1 (4) 1 (8) 15 (60)
Statin 2 (7) 2 (17) 24 (94)
Beta-blocker 0 0 17 (68)
ASA 2 (7) 2 (17) 24 (96)
 
Open in a new tab

Abbreviations: ACE=angiotensin converting enzyme, ASA=Aspirin, CABG=coronary artery bypass graft, CAD= coronary artery disease, HTN=Hypertension, MI=myocardial infarction, PCI=percutaneous intervention.

Clinical Perspective.

Endothelial dysfunction is characterized by decreased nitric oxide bioavailability, occurs early in the development of atherosclerosis, predicts adverse cardiovascular events, and is a potential target for medical interventions. Thus endothelial function is considered a “barometer” of vascular health. Coronary artery and peripheral artery endothelial function are often, however, weakly correlated and it is not clear whether the heterogeneous responses are due to biologic differences between the vascular territories or to the dissimilar techniques and stressors used to obtain the measures. This study demonstrates, for the first time, that non-invasive MRI measures of coronary artery and internal mammary artery (IMA) endothelial function can be assessed at the same time, in response to the same stressor (handgrip exercise) and detected using the same imaging modality. Moreover, the IMA responses are reproducible and primarily mediated by nitric oxide. Thus it is now possible to simultaneously assess systemic vascular health in vessels that do not typically develop atherosclerosis (IMA) along with coronary vascular health in vessels that do develop atherosclerosis and to do so noninvasively. Future trials evaluating the impact of new therapeutic interventions on both peripheral and coronary vascular health can use this as a noninvasive, reproducible means to quantify changes in nitric oxide mediated endothelial function and thereby enhance assessment of the impact of these interventions on cardiovascular risk, pathogenesis and disease progression in patients with coronary artery disease.

Acknowledgments

Sources of Funding

This work was supported by the National Institutes of Health (HL120905, HL125059), the American Heart Association (11SDG5200004), the Swiss National Science Foundation Grant 320030-143923, the PJ Schafer Award, and the Clarence Doodeman Endowment of Johns Hopkins.

Footnotes

Disclosures

None.

References

  • 1.Deanfield JE, Halcox JP, Rabelink TJ. Endothelial function and dysfunction: Testing and clinical relevance. Circulation. 2007;115:1285–1295. doi: 10.1161/CIRCULATIONAHA.106.652859. [DOI] [PubMed] [Google Scholar]
  • 2.Schachinger V, Britten MB, Zeiher AM. Prognostic impact of coronary vasodilator dysfunction on adverse long-term outcome of coronary heart disease. Circulation. 2000;101:1899–1906. doi: 10.1161/01.cir.101.16.1899. [DOI] [PubMed] [Google Scholar]
  • 3.Ludmer PL, Selwyn AP, Shook TL, Wayne RR, Mudge GH, Alexander RW, Ganz P. Paradoxical vasoconstriction induced by acetylcholine in atherosclerotic coronary arteries. The New England Journal of Medicine. 1986;315:1046–1051. doi: 10.1056/NEJM198610233151702. [DOI] [PubMed] [Google Scholar]
  • 4.Halcox JP, Schenke WH, Zalos G, Mincemoyer R, Prasad A, Waclawiw MA, Nour KR, Quyyumi AA. Prognostic value of coronary vascular endothelial dysfunction. Circulation. 2002;106:653–658. doi: 10.1161/01.cir.0000025404.78001.d8. [DOI] [PubMed] [Google Scholar]
  • 5.Hambrecht R, Wolf A, Gielen S, Linke A, Hofer J, Erbs S, Schoene N, Schuler G. Effect of exercise on coronary endothelial function in patients with coronary artery disease. The New England Journal of Medicine. 2000;342:454–460. doi: 10.1056/NEJM200002173420702. [DOI] [PubMed] [Google Scholar]
  • 6.Nitenberg A, Chemla D, Antony I. Epicardial coronary artery constriction to cold pressor test is predictive of cardiovascular events in hypertensive patients with angiographically normal coronary arteries and without other major coronary risk factor. Atherosclerosis. 2004;173:115–123. doi: 10.1016/j.atherosclerosis.2003.12.030. [DOI] [PubMed] [Google Scholar]
  • 7.Suwaidi JA, Hamasaki S, Higano ST, Nishimura RA, Holmes DR, Jr., Lerman A. Long-term follow-up of patients with mild coronary artery disease and endothelial dysfunction. Circulation. 2000;101:948–954. doi: 10.1161/01.cir.101.9.948. [DOI] [PubMed] [Google Scholar]
  • 8.Anderson TJ, Uehata A, Gerhard MD, Meredith IT, Knab S, Delagrange D, Lieberman EH, Ganz P, Creager MA, Yeung AC, Selwyn AP. Close relation of endothelial function in the human coronary and peripheral circulations. Journal of the American College of Cardiology. 1995;26:1235–1241. doi: 10.1016/0735-1097(95)00327-4. [DOI] [PubMed] [Google Scholar]
  • 9.Hirooka Y, Egashira K, Imaizumi T, Tagawa T, Kai H, Sugimachi M, Takeshita A. Effect of l-arginine on acetylcholine-induced endothelium-dependent vasodilation differs between the coronary and forearm vasculatures in humans. Journal of the American College of Cardiology. 1994;24:948–955. doi: 10.1016/0735-1097(94)90854-0. [DOI] [PubMed] [Google Scholar]
  • 10.Otsuka F, Yahagi K, Sakakura K, Virmani R. Why is the mammary artery so special and what protects it from atherosclerosis? Annals of Cardiothoracic Surgery. 2013;2:519–526. doi: 10.3978/j.issn.2225-319X.2013.07.06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Hays AG, Kelle S, Hirsch GA, Soleimanifard S, Yu J, Agarwal HK, Gerstenblith G, Schar M, Stuber M, Weiss RG. Regional coronary endothelial function is closely related to local early coronary atherosclerosis in patients with mild coronary artery disease: Pilot study. Circulation. Cardiovascular Imaging. 2012;5:341–348. doi: 10.1161/CIRCIMAGING.111.969691. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Green DJ, Dawson EA, Groenewoud HM, Jones H, Thijssen DH. Is flow-mediated dilation nitric oxide mediated?: A meta-analysis. Hypertension. 2014;63:376–382. doi: 10.1161/HYPERTENSIONAHA.113.02044. [DOI] [PubMed] [Google Scholar]
  • 13.Cardona A, Kondapally Seshasai SR, Davey J, Arrebola-Moreno AL, Ambrosio G, Kaski JC, Ray KK. A meta-analysis of published studies of endothelial dysfunction does not support its routine clinical use. International journal of Clinical Practice. 2015;69:649–658. doi: 10.1111/ijcp.12630. [DOI] [PubMed] [Google Scholar]
  • 14.Iantorno M, Weiss RG. Using advanced noninvasive imaging techniques to probe the links between regional coronary artery endothelial dysfunction and atherosclerosis. Trends in Cardiovascular Medicine. 2014;24:149–156. doi: 10.1016/j.tcm.2013.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Hays AG, Hirsch GA, Kelle S, Gerstenblith G, Weiss RG, Stuber M. Noninvasive visualization of coronary artery endothelial function in healthy subjects and in patients with coronary artery disease. Journal of the American College of Cardiology. 2010;56:1657–1665. doi: 10.1016/j.jacc.2010.06.036. [DOI] [PubMed] [Google Scholar]
  • 16.Hays AG, Stuber M, Hirsch GA, Yu J, Schar M, Weiss RG, Gerstenblith G, Kelle S. Non-invasive detection of coronary endothelial response to sequential handgrip exercise in coronary artery disease patients and healthy adults. PloS One. 2013;8:e58047. doi: 10.1371/journal.pone.0058047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Hays AG, Iantorno M, Soleimanifard S, Steinberg A, Schar M, Gerstenblith G, Stuber M, Weiss RG. Coronary vasomotor responses to isometric handgrip exercise are primarily mediated by nitric oxide: A noninvasive MRI test of coronary endothelial function. American Journal of Physiology. Heart and Circulatory Physiology. 2015;308:H1343–1350. doi: 10.1152/ajpheart.00023.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.He GW. Endothelial function related to vascular tone in cardiac surgery. Heart, Lung & Circulation. 2005;14:13–18. doi: 10.1016/j.hlc.2004.11.003. [DOI] [PubMed] [Google Scholar]
  • 19.Nishioka H, Kitamura S, Kameda Y, Taniguchi S, Kawata T, Mizuguchi K. Difference in acetylcholine-induced nitric oxide release of arterial and venous grafts in patients after coronary bypass operations. The Journal of Thoracic and Cardiovascular Surgery. 1998;116:454–459. doi: 10.1016/S0022-5223(98)70011-X. [DOI] [PubMed] [Google Scholar]
  • 20.Werner GS, Wiegand V, Kreuzer H. Effect of acetylcholine on arterial and venous grafts and coronary arteries in patients with coronary artery disease. European Heart Journal. 1990;11:127–137. doi: 10.1093/oxfordjournals.eurheartj.a059668. [DOI] [PubMed] [Google Scholar]
  • 21.Tawakol A, Forgione MA, Stuehlinger M, Alpert NM, Cooke JP, Loscalzo J, Fischman AJ, Creager MA, Gewirtz H. Homocysteine impairs coronary microvascular dilator function in humans. Journal of the American College of Cardiology. 2002;40:1051–1058. doi: 10.1016/s0735-1097(02)02069-7. [DOI] [PubMed] [Google Scholar]
  • 22.Doucette JW, Corl PD, Payne HM, Flynn AE, Goto M, Nassi M, Segal J. Validation of a doppler guide wire for intravascular measurement of coronary artery flow velocity. Circulation. 1992;85:1899–1911. doi: 10.1161/01.cir.85.5.1899. [DOI] [PubMed] [Google Scholar]
  • 23.Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1:307–310. [PubMed] [Google Scholar]
  • 24.Bartko JJ. The intraclass correlation coefficient as a measure of reliability. Psychological Reports. 1966;19:3–11. doi: 10.2466/pr0.1966.19.1.3. [DOI] [PubMed] [Google Scholar]
  • 25.Vita JA, Treasure CB, Nabel EG, McLenachan JM, Fish RD, Yeung AC, Vekshtein VI, Selwyn AP, Ganz P. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation. 1990;81:491–497. doi: 10.1161/01.cir.81.2.491. [DOI] [PubMed] [Google Scholar]
  • 26.Matsuzawa Y, Li J, Aoki T, Guddeti RR, Kwon TG, Cilluffo R, Widmer RJ, Gulati R, Lennon RJ, Lerman LO, Lerman A. Predictive value of endothelial function by noninvasive peripheral arterial tonometry for coronary artery disease. Coronary Artery Disease. 2015;26:231–238. doi: 10.1097/MCA.0000000000000208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Klocke FJ. Epicardial coronary artery vasomotion. Journal of the American College of Cardiology. 2010;56:1666–1667. doi: 10.1016/j.jacc.2010.05.048. [DOI] [PubMed] [Google Scholar]
  • 28.Takahashi N, Ishibashi Y, Shimada T, Sakane T, Ohata S, Sugamori T, Ohta Y, Inoue S, Nakamura K, Shimizu H, Katoh H, Murakami Y. Impaired exercise-induced vasodilatation in chronic atrial fibrillation--role of endothelium-derived nitric oxide. Circulation Journal : Official Journal of the Japanese Circulation Society. 2002;66:583–588. doi: 10.1253/circj.66.583. [DOI] [PubMed] [Google Scholar]
  • 29.Wray DW, Witman MA, Ives SJ, McDaniel J, Fjeldstad AS, Trinity JD, Conklin JD, Supiano MA, Richardson RS. Progressive handgrip exercise: Evidence of nitric oxide-dependent vasodilation and blood flow regulation in humans. American Journal of Physiology. Heart and Circulatory Physiology. 2011;300:H1101–1107. doi: 10.1152/ajpheart.01115.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Dalaklioglu S, Golbasi I, Ogutman C. Comparative effects of preoperative angiotensin-converting enzyme in-hibitor, statin and beta-blocker treatment on human internal mammary artery reactivity in patients with coronary artery disease: A pilot study. The Open Cardiovascular Medicine Journal. 2013;7:69–75. doi: 10.2174/1874192401307010069. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Berkenboom G, Crasset V, Giot C, Unger P, Vachiery JL, LeClerc JL. Endothelial function of internal mammary artery in patients with coronary artery disease and in cardiac transplant recipients. American Heart Journal. 1998;135:488–494. doi: 10.1016/s0002-8703(98)70326-6. [DOI] [PubMed] [Google Scholar]
  • 32.Kitta Y, Obata JE, Nakamura T, Hirano M, Kodama Y, Fujioka D, Saito Y, Kawabata K, Sano K, Kobayashi T, Yano T, Nakamura K, Kugiyama K. Persistent impairment of endothelial vasomotor function has a negative impact on outcome in patients with coronary artery disease. Journal of the American College of Cardiology. 2009;53:323–330. doi: 10.1016/j.jacc.2008.08.074. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
NIHMS754885-supplement-Supplemental_Material.pdf (904.3KB, pdf)

RESOURCES