Abstract
Quantification of myocardial perfusion reserve (MPR) using vasodilator stress cardiac magnetic resonance is increasingly used to detect coronary artery disease. However, MPR can also be altered because of changes in microvascular function. We aimed to determine whether MPR can distinguish between ischemic cardiomyopathy (IC) secondary to coronary artery disease and non-IC (NIC) with microvascular dysfunction and no underlying epicardial coronary disease. A total of 60 patients (mean age 65 ± 14 years, 30% women), including 31 with IC and 29 with NIC, were identified from a pre-existing vasodilator stress cardiac magnetic resonance registry. Short-axis cine slices were used to measure left ventricular ejection fraction (LVEF) using the Simpson method of disks. MPR index (MPRi) was determined from first-pass myocardial perfusion images during stress and rest using the upslope ratio, normalized for the arterial input and corrected for rate pressure product. Patients in both groups were divided into subgroups of LVEF ≤35% and LVEF >35%. Differences in MPRi between the subgroups were examined. MPRi was moderately correlated with LVEF in patients with NIC (r = 0.53, p = 0.03), whereas the correlation in patients with IC was lower (r = 0.32, p = 0.22). Average LVEF in NIC and IC was 34% ± 8% and 35% ± 8%, respectively (p = 0.63). MPRi was not significantly different in IC compared with NIC (1.17 [0.88 to 1.61] vs 1.23 [1.07 to 1.66], p = 0.41), including the subgroups of LVEF (IC: 1.20 ± 0.56 vs NIC: 1.15 ± 0.24, p = 0.75 for LVEF ≤35% and IC: 1.35 ± 0.44 vs NIC: 1.58 ± 0.50, p = 0.19 for LVEF >35%). However, MPRi was significantly lower in patients with LVEF ≤35% compared with those with LVEF>35% (1.17 ± 0.40 vs 1.47 ± 0.47, p = 0.01). Similar difference between LVEF groups was noted in the patients with NIC (1.15 ± 0.24 vs 1.58 ± 0.50, p = 0.006) but not in the patients with IC (1.20 ± 0.56 vs 1.35 ± 0.44, p = 0.42). MPRi can be abnormal in the presence of left ventricular dysfunction with nonischemic etiology. This is a potential pitfall to consider when using this approach to detect ischemia because of epicardial coronary disease using myocardial perfusion imaging.
Introduction
Abnormalities in myocardial perfusion reserve (MPR) are known to be associated with coronary artery disease (CAD). In the ISCHEMIA study, patients with stable ischemic heart disease and left ventricular (LV) dysfunction (LVD) had worse outcomes than patients without LVD. Moreover, patients with LVD assigned to an initial invasive treatment strategy aimed at coronary revascularization had better clinical outcomes compared with a conservative strategy without initial coronary revascularization; whereas no such difference was present in patients without LVD.1 In patients with depressed LV ejection fraction (LVEF), assessment of myocardial perfusion can potentially be used to differentiate ischemic from nonischemic heart disease as the etiology of the LVD and thus guide treatment decisions. However, abnormal MPR may be present not only in LVD because of obstructive CAD, but also in the setting of nonischemic cardiomyopathy (NIC) because of underlying microvascular dysfunction.2,3 Indeed, previous studies have demonstrated an impairment of myocardial perfusion in patients with dilated cardiomyopathy4,5 and that the reduction of MPR is correlated with LVEF.3 Cardiac magnetic resonance (CMR) is commonly used to evaluate patients with LVD. It is considered the reference standard for quantifying LVEF, and the burden and pattern of myocardial fibrosis. When combined with vasodilator stress perfusion imaging, CMR can also be used to quantify MPR.6,7 Semiquantitative approaches, such as the upslope ratio, for determining MPR from stress CMR images have been validated against invasive fractional flow reserve8,9 and shown to be independently associated with the occurrence of major adverse cardiac events.10,11 In this study, we aimed to determine whether MPR measured using stress perfusion CMR can distinguish between ischemic cardiomyopathy (IC) and NIC, and how this ability is influenced by the underlying LVEF.
Methods
We identified 60 patients, including 31 with IC and 29 with NIC, from a pre-existing registry of patients who underwent vasodilator stress CMR at the University of Chicago. All patients provided written informed consent before their imaging to be included in this registry, which was approved by the institutional review board. The definition of IC was based on the following criteria: (1) history of obstructive epicardial coronary disease, including previous myocardial infarction or coronary artery revascularization; (2) LVEF <50% using CMR measurement; and (3) evidence of coronary artery luminal stenosis ≥70% in 1 or more epicardial vessels on invasive or computed tomography coronary angiography. The definition of NIC was based on the following criteria: (1) enlarged LV end-diastolic volume indexed to body surface area and reduced LVEF (<50%); (2) absence of significant CAD or late gadolinium enhancement (LGE) pattern consistent with previous myocardial infarction; and (3) no evidence of hypertrophic and infiltrate cardiomyopathy, significant primary valvular heart disease, or congenital heart disease. Patient demographics and clinical history were extracted from the electronic medical records.
CMR imaging was performed using a 1.5-T scanner (Achieva, Philips, Best, The Netherlands) with a 5-element phased array cardiac coil. First-pass perfusion CMR images were acquired approximately 60 seconds after the administration of a vasodilator (regadenoson 0.4 mg, Lexiscan; Astellas Pharma, Northbrook, Illinois, USA) using 0.05 to 0.1 mmol/kg of gadolinium-based contrast agent, depending on renal function. Imaging was performed on 3 short-axis LV slices using a hybrid gradient echo-epi pulse sequence (voxel size ~2.5 × 2.5 mm, slice thickness 10 mm, flip angle 20°, repetition time 5.9 ms, echo time 2.5 ms, EPI Factor 5, delay time 80 ms, and SENSE Factor 1.3). The vasodilator effect was reversed with aminophylline (75 mg). Retrospectively-gated cine images were subsequently obtained using a steady-state free precession sequence, during approximately 5-second breath holds (repetition time 2.9 ms; echo time 1.5 ms; flip angle 60°; temporal resolution 30 to 40 ms). Standard long-axis views were obtained, including 4-, 2-, and 3-chamber planes. In addition, short-axis slices spanning the entire left and right ventricles from base to the apex (slice thickness 8 mm; gap 2 mm) were obtained. Thereafter, perfusion at rest was acquired using the same pulse sequence and contrast dosing used for stress imaging. After a 5-minute delay, LGE imaging was also performed in the same short- and long-axis views as the cine images, using a T1-weighted gradient echo pulse sequence with a phase sensitive inversion recovery reconstruction (TR 4.5 ms, TE 2.2 ms, TI 250 to 300 ms, flip angle 30°, flip angle 5°, voxel size 2 × 2 × 10 mm, SENSE Factor 2). An inversion time between 250 and 300 ms was used to achieve nulling of normal myocardium. Heart rate (HR), blood pressure, and symptoms were monitored throughout the examination.
Using the short-axis cine images, the LV and right ventricular (RV) end-diastolic and end-systolic frames were identified and the Simpson method of disks was used to calculate LV end-diastolic mass, LV and RV end-diastolic and end-systolic volumes, and the corresponding ejection fraction using commercial software (Medis, Leiden, The Netherlands).
MPR index (MPRi), a semiquantitative marker of myocardial perfusion,12 was measured using commercial software (CardioAI, Arterys, San Francisco, California). The LV and RV endocardial and epicardial boundaries were traced on the 3 short-axis stress and rest perfusion images, initially by the software and then, adjusted as necessary by an expert user. An LV blood pool region of interest was manually traced in the midslice of both rest and stress images. Once segmentation was verified, a bull’s-eye plot of the segmental perfusion at rest upslope of the LV myocardium divided by the upslope of the blood pool was displayed along with the corresponding time-intensity curves. The same was done for the stress perfusion images. Finally, an upslope ratio (i.e., MPRi) was calculated from the previously mentioned stress and rest perfusion values for each of the American Heart Association 16 segments.13 The highest 2 numbers and lowest 2 numbers were excluded from the 16 segments to account for outlier values that are more likely to represents errors in myocardial segmentation rather than actual regional variation in myocardial perfusion. Global rest upslope ratio, global stress upslope ratio, and MPRi were calculated as the average of the remaining segments (Figure 1). To correct for potential differences in HR and blood pressure between patients, the MPRi was normalized to rate pressure product (RPP) measured during resting conditions: Corrected MPRi = MPRi / rest RPP × 10,000. Corrected MPRi is referred to as MPRi throughout the manuscript for simplicity.
Figure 1.
Example of stress perfusion images and the corresponding time intensity curves obtained at stress and rest in a patient with NIC with LVEF 42% (A), and in a patient with IC with LVEF 42% (B).
Continuous variables were tested for normal distribution and presented as mean ± standard deviation when normally distributed or as median with interquartile ranges when not normally distributed. Intergroup differences were tested using t test or the Mann-Whitney U tests, respectively. Categoric variables were presented as absolute numbers with percentages and tested using the chi-square test. Patient cohorts (both IC and NIC) were divided into 2 clinically relevant groups each, according to LVEF ≤35% and LVEF >35%.14 Linear regression analysis with Pearson correlation coefficients was performed to test the relation between MPRi and LV function and size parameters. Intraobserver reproducibility was tested for MPRi on 20 randomly selected patients with measurements made 3 months apart to prevent memory recall and quantified using intraclass correlation coefficients. Values of p <0.05 were considered significant. Analysis was performed using SPSS software (version 23.0, Chicago, Illinois).
Results
Patient characteristics are shown in Table 1, along with the key imaging findings. Thirty-one patients with IC had a definite evidence of obstructive coronary disease based on invasive coronary angiography or coronary computed tomography. In the NIC group, 29 patients were diagnosed with dilated cardiomyopathy, 15 patients (52%) did not have a specifically identified underlying etiology (i.e., idiopathic), 2 patients (7%) had alcohol-associated cardiomyopathy, 1 patient (3%) had LV noncompaction cardiomyopathy, 7 patients (24%) had chemotherapy-associated cardiomyopathy, and 4 patients (14%) had postmyocarditis-associated cardiomyopathy.
Table 1.
Population baseline parameters and comparison between IC and NIC
| Mean ± SD or medium (interquartile range) or n (%) | ||||
|---|---|---|---|---|
| Overall (n=60) | IC (n=31) | NIC (n=29) | P value | |
| Age (yrs) | 66 (51,73) | 67 (62,74) | 59 (45,73) | 0.112 |
| Women | 24 (40%) | 9 (29%) | 15 (52%) | 0.114 |
| BMI (kg/m2) | 28±5 | 27±5 | 29 | 0.127 |
| BSA (m2) | 1.9±0.3 | 1.9±0.2 | 2.0 | 0.100 |
| Smoker (current or former) | 35 (58%) | 21 (68%) | 14 (48%) | 0.067 |
| CAD | 32 (53%) | 31 (100%) | 1 (3%) | <0.001 |
| Hypertension | 43 (72%) | 23 (77%) | 20 (67%) | 0.390 |
| Hyperlipidemia | 32 (53%) | 23 (77%) | 9 (30%) | <0.001 |
| Diabetes | 19 (32%) | 7 (23%) | 12 (40%) | 0.165 |
| ECG | ||||
| LBBB | 11 (18%) | 4 (13%) | 7 (23%) | 0.317 |
| PVC | 23 (38%) | 11 (37%) | 12 (40%) | 0.791 |
| Medications | ||||
| ACEI/ARB/ARNI | 45 (75%) | 21 (70%) | 24 (80%) | 0.371 |
| Beta-blocker | 49 (82%) | 24 (80%) | 25 (83%) | 0.739 |
| Aldosterone antagonist | 17 (28%) | 7 (23%) | 10 (33%) | 0.390 |
| Diuretic | 31 (52%) | 15 (50%) | 16 (53%) | 0.796 |
| Revascularization History | ||||
| Prior PCI | 16 (27%) | 16 (52%) | 0 | <0.001 |
| Prior CABG | 11 (18%) | 11 (35%) | 0 | <0.001 |
| Stress CMR hemodynamics | ||||
| Stress heart rate (beats/min) | 99±13 | 96±14 | 102±10 | 0.061 |
| Stress SBP (mmHg) | 126±23 | 126±24 | 125±21 | 0.864 |
| Stress RPP | 12392±2695 | 11830 (9760,14007) | 12210 (10983,14427) | 0.343 |
| Rest heart rate (beats/min) | 76 (66,88) | 73±13 | 81±13 | 0.020 |
| Rest SBP (mmHg) | 127±20 | 131±22 | 123±18 | 0.129 |
| Rest RPP | 9081 (8008,11357) | 8978 (7772,11639) | 9256 (8488,10752) | 0.502 |
| CMR parameters | ||||
| LVEF (%) | 35±9 | 35±8 | 34±8 | 0.630 |
| LVEDV (ml) | 235±59 | 232±61 | 239±56 | 0.645 |
| LVEDVi (ml/m2) | 119±27 | 116±30 | 121±23 | 0.471 |
| LVESV (ml) | 151 (122,185) | 148 (119,186) | 152 (125,184) | 0.484 |
| LVESVi (ml/m2) | 89 (75,116) | 74 (61,87) | 76 (62,93) | 0.637 |
| LVM (g) | 131±39 | 125±36 | 137±41 | 0.233 |
| LVMi (g/m2) | 66±16 | 62±14 | 69±18 | 0.098 |
| RVEF (%) | 51 (43,57) | 49±10 | 48±12 | 0.727 |
| RVEDV (ml) | 161 (139,197) | 159 (129,190) | 162 (141,209) | 0.434 |
| RVEDVi (ml/m2) | 85 (75,96) | 83 (66,94) | 87 (77,100) | 0.735 |
| RVESV (ml) | 79 (59,106) | 79 (54,100) | 74 (61,116) | 0.528 |
| RVESVi (ml/m2) | 41 (30,52) | 41 (29,48) | 41 (32,55) | 0.849 |
| LGE | 41 (68%) | 29 (94%) | 12 (41%) | <0.001 |
| Ischemic pattern | 29 (48%) | 29 (94%) | 0 | <0.001 |
| Non-ischemic pattern | 17 (28%) | 4 (13%) | 13 (45%) | <0.001 |
| Perfusion defect | 29 (48%) | 24 (77%) | 5 (17%) | <0.001 |
| MPRi | 1.22 (1.03,1.64) | 1.17 (0.88,1.61) | 1.23 (1.07,1.66) | 0.409 |
Data expressed as mean ± standard deviation for normally distributed parameters or as median with interquartile ranges when not normally distributed.
The IC and NIC subgroups had similar age, gender, body mass index, medication usage, and cardiovascular risk factor profile, with the exception of hyperlipidemia. Compared with IC, no difference in LV parameters was seen in the NIC group. The NIC group had a higher HR at rest. The IC group had a higher prevalence of LGE and vasodilator-induced perfusion defect (Table 1).
Patients in the subgroup of LVEF ≤35% had higher stress HR, HR at rest, and blood pressure at rest than those with LVEF >35% (Table 2). With the exception of hyperlipidemia, the proportion of cardiovascular risk factors was similar for both groups. The prevalence of LGE and vasodilator-induced perfusion defects was also similar for both groups. The left ventricle was larger, and the right ventricular ejection fraction was lower in patients with LVEF ≤35%.
Table 2.
Comparison of patient characteristics and imaging parameters between LVEF≤35% and LVEF>35%
| LVEF≤35% (n=30) | LVEF>35% (n=30) | P value | |
|---|---|---|---|
| Age (yrs) | 62±15 | 61±15 | 0.80 |
| Gender, female | 13 (43%) | 11 (37%) | 0.40 |
| BMI (kg/m2) | 27±5 | 29±5 | 0.09 |
| BSA (m2) | 1.9±0.3 | 2.0±0.2 | 0.05 |
| Smoker (current or prior) | 14 (47%) | 21 (70%) | 0.06 |
| CAD | 15 (50%) | 16 (53%) | 0.50 |
| Hypertension | 22 (73%) | 21 (70%) | 0.50 |
| Hyperlipidemia | 11 (37%) | 21 (70%) | 0.01 |
| Diabetes | 9 (30%) | 10 (33%) | 0.50 |
| ECG | |||
| LBBB | 6 (20%) | 5 (17%) | 0.50 |
| PVC | 11 (37%) | 12 (40%) | 0.50 |
| Stress CMR hemodynamics | |||
| Stress heart rate(beats/min) | 103±10 | 95±14 | 0.02 |
| Stress SBP (mmHg) | 126±24 | 126±22 | 1.00 |
| Stress RPP | 12,916±2,878 | 11,869±2,434 | 0.13 |
| Rest heart rate(beats/min) | 85 (76,94) | 66 (62,76) | 0.00 |
| Rest SBP (mmHg) | 130 (105,142) | 123 (115,139) | 0.65 |
| Rest RPP | 10128 (8897,12808) | 8488 (7694,9889) | 0.01 |
| CMR parameters | |||
| LVEF, (%) | 28 (23,32) | 42 (38,44) | <0.01 |
| LVEDV, ml | 241±62 | 229±56 | 0.42 |
| LVEDVi (ml/m2) | 126±28 | 111±24 | 0.03 |
| LVESV (ml) | 177±53 | 134±36 | <0.01 |
| LVESVi (ml/m2) | 91±26 | 66±15 | <0.01 |
| LVM (g) | 137±40 | 126±38 | 0.25 |
| LVMi (g/m2) | 70±17 | 61±15 | 0.04 |
| RVEF (%) | 45±12 | 52±9 | 0.01 |
| RVEDV (ml) | 158±41 | 174±47 | 0.16 |
| RVEDVi (ml/m2) | 86 (66,94) | 84 (76,99) | 0.64 |
| RVESV (ml) | 80 (62,110) | 79 (54,101) | 0.56 |
| RVESVi (ml/m2) | 42 (32,62) | 39 (30,47) | 0.28 |
| LGE | 20 (67%) | 21 (70%) | 0.10 |
| Ischemic pattern | 15 (50%) | 14 (47%) | 0.50 |
| Non-ischemic pattern | 10 (30%) | 7 (23%) | 0.50 |
| Perfusion defect, n (%) | 17 (57%) | 14 (47%) | 0.15 |
| MPRi | 1.17±0.43 | 1.47±0.48 | 0.01 |
Data expressed as mean ± standard deviation for normally distributed parameters or as median with interquartile ranges when not normally distributed.
The global MPRi was lower than 2.0 for all patients. MPRi was not different between the IC and NIC groups. In subgroup analysis, compared with the LVEF >35% group, patients with LVEF ≤35% had reduced MPRi (1.47 ± 0.47 vs 1.17 ± 0.40, p <0.01). Patients with NIC who had an LVEF ≤35% had lower MPRi than those with LVEF >35% (1.15 ± 0.24 vs 1.58 ± 0.50, p <0.01); however, no difference in MPRi was seen for patients with IC with LVEF ≤35% versus LVEF >35% (1.20 ± 0.56 vs 1.35 ± 0.44, p = 0.42) (Figure 2). Additionally, there was no relation between MPRi and LV end-diastolic volume and LV end-systolic volume, or LV end-diastolic mass. The MPRi was lower in segments with LGE than in those without LGE (1.13 ± 0.54 vs 1.39 ±0.60, p = 0.04) (Figure 3).
Figure 2.
Top: Comparisons of MPRI based on etiology of cardiomyopathy: IC versus NIC for full cohort (a), and for subgroups with LVEF<35% (b), and with LVEF>35% (c). Bottom: Comparisons of MPRi based on LV function: LVEF ≤35% versus LVEF>35% for the entire cohort(d), and for subgroups with NIC (e) and with IC (f). Note: Dashed horizontal lines and whiskers in panels (a) represent medians and inter-quartile ranges, while in panels (b), (c), (d), (e) and (f) – means and standard deviations.
Figure 3.
Comparison of MPRi between segments with LGE and segments without LGE. Note: Dashed horizontal lines and whiskers represent medians and inter-quartile ranges.
MPRi was moderately correlated to LVEF in the NIC group (r = 0.53, p <0.001). However, there was only weak correlation between the 2 parameters in the IC group and the overall cohort (r = 0.36, p = 0.01) (Figure 4). For all patients, the MPRi was positively correlated to LV stroke volume (r = 0.53, p <0.001) and negatively correlated to stress HR (r = 0.44, p = 0.001), stress RPP (r = 0.36, p = 0.01), systolic blood pressure (SBP) at rest (r = 0.41, p = 0.001), HR at rest (r = 0.47, p <0.001), and RPP at rest (r = 0.63, p <0.001). For NIC, MPRi was negatively correlated to stress HR (r = 0.53, p = 0.003), HR at rest (r = 0.67, p <0.001), and RPP at rest (r = 0.66, p <0.001). For IC, MPRi was negatively correlated to stress HR (r = 0.45, p = 0.01), stress RPP (r = 0.40, p = 0.03), SBP at rest (r = 0.50, p = 0.005), HR at rest (r = 0.40, p = 0.03), and RPP at rest (r = 0.62, p <0.001).
Figure 4.
Linear regression plots representing the relationship between MPRi and LVEF in IC and NIC.
The intraclass correlation for the intraobserver variability of MPRi was intraclass correlation coefficients = 0.85.
Discussion
Stress myocardial perfusion imaging is commonly performed in clinical practice to differentiate ischemic from NIC. In this study, we quantified MPRi using stress perfusion CMR to evaluate the relation between MPRi and LVEF in IC and NIC. There was no difference in MPRi between IC and NIC. However, in patients with NIC, MPRi was correlated with LVEF, although this relation was not seen in patients with IC. The reason for this finding is that IC is associated with significant regional reductions in MPRi as a consequence of a significant upstream epicardial coronary artery stenosis, independently of LVEF. However, in NIC, there are multiple mechanisms of abnormal MPRi, all related to the microvascular structure and function. As LVEF worsens, increases in myocardial resistive forces occur because of increased LV pressure, LV dilation, and hypertrophy. Simultaneously, anatomic changes, such as perivascular fibrosis and reduced capillary density also occur. Each of global changes would be expected to worsen microvascular function, which was quantified in this study in terms of MPRi.
MPRi is not only altered in response to epicardial CAD but also by changes in coronary microvascular function that may occur because of underlying myocardial fibrosis, endothelial dysfunction, mismatch of oxygen demand and supply, and microvascular remodeling.15,16 Indeed, Zhou et al11 demonstrated that MPRi, rather than obstructive CAD, was an independent imaging marker to predict major adverse cardiac events in patients with ischemic symptoms. Few studies have evaluated the effects of reduced LVEF on MPRi. Majmudar et al showed that coronary flow reserve (CFR) measured using positive emission tomography was reduced in patients with both IC and NIC with LVEF ≤45%. In their cohort, 82% of patients had an abnormal CFR (<2.0). Additionally, their study also demonstrated that abnormal CFR was associated with poor prognosis but did not evaluate the relation between CFR and LVEF.8 In our study, we found that MPRi, which is the stress perfusion CMR equivalent to CFR, was also abnormal (defined as MPRi <217) in nearly all patients with IC (94%) and NIC (86%). Furthermore, the microvascular dysfunction associated with NIC may not be readily noted on visual inspection of the stress perfusion images alone. In our cohort, only 5 of patients (17%) with NIC had visually evident stress-induced perfusion defect despite that 86% of them had abnormal MPRi. The best cut-off value for defining abnormal MPRi has yet to be determined. Thomson et al18 have suggested that an MPRi cut-off value of <1.84 detected coronary microvascular dysfunction. In our patients with NIC, 83% of the group still had an MPRi <1.84.
Zhang et al enrolled patients with reduced LVEF (<50%) to undergo the stress CMR for the evaluation of myocardial perfusion and LGE. They found that CMR-detected myocardial ischemia and LGE provided prognostic value for cardiovascular outcomes, independently of LVEF. In this study, ischemia was assessed qualitatively rather than quantitatively. However, the cohort included cardiomyopathy with diverse etiology, and only a limited number of patients had severely reduced LVEF (<30%).19 As implied by the previous discussion, consideration of only qualitatively evident perfusion abnormalities could conceal abnormal myocardial perfusion caused by coronary microvascular dysfunction. The relation between MPRi and LVEF has not been fully elucidated.20 Gulati et al3 demonstrated the association between LVEF and global MPRi in patients with dilated cardiomyopathy using quantitative stress CMR perfusion measurements. They noted a trend toward lower MPR in patients with LGE. LVEF was identified as independent determinant of MPR. Our study expands on their findings by showing that impaired MPRi was prevalent in both IC and NIC, but it was associated with LVEF only in NIC. A potential explanation for this discrepancy is that in IC, obstructive coronary artery stenosis, presence of myocardial scar, and microvascular dysfunction may all contribute to LV dysfunction, weakening the relation between LVEF and MPRi.
Conversely, in NIC, MPRi was related to LVEF, which is consistent with the findings of Gulati et al.3 It has been suggested that abnormalities in MPRi in NIC may be due to the effects of underlying myocardial damage and interstitial fibrosis21 on myocardial perfusion and cellular metabolism.22 Other factors that might be responsible for the abnormal myocardial perfusion seen in NIC23 include coronary endothelial dysfunction,24 reduced coronary capillary bed volume,25 and pressure overload on the coronary microvascular vessels.26 Additionally, elevated JR at rest and SBP could result in a normal physiologic decrease in the MPRi. This is because some of the available perfusion reserve is being utilized to maintain the myocardial blood flow at rest required to meet the increased myocardial oxygen demand created by elevations in the HR at rest and blood pressure. In our study, we attempted to account for this physiologic response by normalizing MPR to the RPP at rest.
Stress myocardial perfusion imaging using single-photon emission computed tomography, positron emission tomography, or CMR is commonly performed to differentiate IC from NIC. Our data clearly show that LV dysfunction is significantly associated with reductions in MPR even in patients with NIC. This association significantly impairs the ability of myocardial perfusion imaging to differentiate IC from NIC. If a noninvasive approach to differentiating between these 2 disease states is desired, it may be more useful to directly visualize the coronary anatomy using coronary computed tomography27,28 or to directly visualize the presence of myocardial infarction related scar using LGE imaging.29,30
Although our study revealed an association between LVEF and MPRi in NIC, an important limitation of our study is that we did not determine whether the reduction in LVEF is the cause of the abnormal MPRi or whether the reductions in MPRi are contributing to the underlying LV dysfunction. Another limitation is that this was a retrospective, single-center study, performed in a small sample of patients referred for vasodilator stress CMR. Larger, multi-center studies are needed to confirm our findings. Finally, we assessed MPR using semiquantitative upslope approach, rather than a fully quantitative approach capable of measuring myocardial blood flow in absolute terms of ml/g/min.31,32 The use of absolute myocardial blood flow analysis rather than MPRi could provide greater insight into the relation between LVEF and blood flow at rest versus the hyperemic myocardial blood flow.
One could potentially view as a limitation of our study the fact that 1 patient in the NIC group had CAD (Table 1). However, this patient had nonobstructive coronary stenosis in the LAD artery with luminal narrowing of 40% to 50% and showed a nonischemic LGE pattern. We believe that this was incidental nonobstructive CAD in the setting of NIC. Excluding this patient from analysis did not substantially change the key findings of our study.
In conclusion, our findings indicate that MPRi can be abnormal in the presence of LV dysfunction even in the absence of ischemic heart disease. This is a potential pitfall to consider when using myocardial perfusion imaging to differentiate between ischemic and NIC.
Acknowledgments
This project was supported by the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH) (Bethesda, Maryland) through grant number 5UL1TR002389-02 that funds the Institute for Translational Medicine (ITM) (Chicago, Illinois). Dr. H. Patel was funded by a T32 Cardiovascular Sciences Training (Chicago, Illinois) Grant (5T32HL7381). Dr. Kawaji, was funded by a K25 Grant (HL141634) (Bethesda, Maryland).
Footnotes
Disclosures
Dr. A. Patel has received research support from Philips, Arterys, CircleCVI, and Neosoft. The remaining authors have no conflicts of interest to declare.
References
- 1.Lopes RD, Alexander KP, Stevens SR, Reynolds HR, Stone GW, Piña IL, Rockhold FW, Elghamaz A, Lopez-Sendon JL, Farsky PS, Chernyavskiy AM, Diaz A, Phaneuf D, De Belder MA, Ma YT, Guzman LA, Khouri M, Sionis A, Hausenloy DJ, Doerr R, Selvanayagam JB, Maggioni AP, Hochman JS, Maron DJ. Initial Invasive versus conservative management of stable ischemic heart disease in patients with a history of heart failure or left ventricular dysfunction: insights from the ISCHEMIA trial. Circulation 2020;142:1725–1735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bravo PE, Di Carli MF, Dorbala S. Role of PET to evaluate coronary microvascular dysfunction in non-ischemic cardiomyopathies. Heart Fail Rev 2017;22:455–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gulati A, Ismail TF, Ali A, Hsu LY, Gonçalves C, Ismail NA, Krishnathasan K, Davendralingam N, Ferreira P, Halliday BP, Jones DA, Wage R, Newsome S, Gatehouse P, Firmin D, Jabbour A, Assomull RG, Mathur A, Pennell DJ, Arai AE, Prasad SK. Microvascular dysfunction in dilated cardiomyopathy: a quantitative stress perfusion cardiovascular magnetic resonance study. JACC Cardiovasc Imaging 2019;12:1699–1708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Canetti M, Akhter MW, Lerman A, Karaalp IS, Zell JA, Singh H, Mehra A, Elkayam U. Evaluation of myocardial blood flow reserve in patients with chronic congestive heart failure due to idiopathic dilated cardiomyopathy. Am J Cardiol 2003;92:1246–1249. [DOI] [PubMed] [Google Scholar]
- 5.Rijnierse MT, de Haan S, Harms HJ, Robbers LF, Wu L, Danad I, Beek AM, Heymans MW, van Rossum AC, Lammertsma AA, Allaart CP, Knaapen P. Impaired hyperemic myocardial blood flow is associated with inducibility of ventricular arrhythmia in ischemic cardiomyopathy. Circ Cardiovasc Imaging 2014;7:20–30. [DOI] [PubMed] [Google Scholar]
- 6.Bhave NM, Freed BH, Yodwut C, Kolanczyk D, Dill K, Lang RM, Mor-Avi V, Patel AR. Considerations when measuring myocardial perfusion reserve by cardiovascular magnetic resonance using regadenoson. J Cardiovasc Magn Reson 2012;14:89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hubert A, Seitz A, Pereyra VM, Bekeredjian R, Sechtem U, Ong P. Coronary artery Spasm: the Interplay between endothelial dysfunction and vascular smooth muscle cell hyperreactivity. Eur Cardiol 2020;15:e12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Barmeyer AA, Stork A, Muellerleile K, Schofer AK, Tiburtius C, Koester R, Heitzer T, Adam G, Meinertz T, Lund GK. Comparison of quantitative coronary angiography and first-pass perfusion magnetic resonance imaging for the detection of an impaired coronary perfusion in nonsevere coronary stenosis. J Magn Reson Imaging 2008;27:1005–1011. [DOI] [PubMed] [Google Scholar]
- 9.Nagel E, Klein C, Paetsch I, Hettwer S, Schnackenburg B, Wegscheider K, Fleck E. Magnetic resonance perfusion measurements for the noninvasive detection of coronary artery disease. Circulation 2003;108:432–437. [DOI] [PubMed] [Google Scholar]
- 10.Doyle M, Weinberg N, Pohost GM, Bairey Merz CN, Shaw LJ, Sopko G, Fuisz A, Rogers WJ, Walsh EG, Johnson BD, Sharaf BL, Pepine CJ, Mankad S, Reis SE, Vido DA, Rayarao G, Bittner V, Tauxe L, Olson MB, Kelsey SF, Biederman RW. Prognostic value of global MR myocardial perfusion imaging in women with suspected myocardial ischemia and no obstructive coronary disease: results from the NHLBI-sponsored WISE (Women's Ischemia Syndrome Evaluation) study. JACC Cardiovasc Imaging 2010;3:1030–1036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhou W, Lee JCY, Leung ST, Lai A, Lee TF, Chiang JB, Cheng YW, Chan HL, Yiu KH, Goh VK, Pennell DJ, Ng MY. Long-term prognosis of patients with coronary microvascular disease using stress perfusion cardiac magnetic resonance. JACC Cardiovasc Imaging 2021;14:602–611. [DOI] [PubMed] [Google Scholar]
- 12.Schulz-Menger J, Bluemke DA, Bremerich J, Flamm SD, Fogel MA, Friedrich MG, Kim RJ, von Knobelsdorff-Brenkenhoff F, Kramer CM, Pennell DJ, Plein S, Nagel E. Standardized image interpretation and post-processing in cardiovascular magnetic resonance –2020 update: Society for Cardiovascular Magnetic Resonance (SCMR): Board of Trustees Task Force on Standardized Post-Processing. J Cardiovasc Magn Reson 2020;22:19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jerosch-Herold M, Seethamraju RT, Swingen CM, Wilke NM, Stillman AE. Analysis of myocardial perfusion MRI. J Magn Reson Imaging 2004;19(6):758–770. [DOI] [PubMed] [Google Scholar]
- 14.McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, Falk V, Filippatos G, Fonseca C, Gomez-Sanchez MA, Jaarsma T, Køber L, Lip GY, Maggioni AP, Parkhomenko A, Pieske BM, Popescu BA, Rønnevik PK, Rutten FH, Schwitter J, Seferovic P, Stepinska J, Trindade PT, Voors AA, Zannad F, Zeiher A, Committee for Practice Guidelines ESC. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012;33:1787–1847. [DOI] [PubMed] [Google Scholar]
- 15.Duncker DJ, Koller A, Merkus D, Canty JM Jr. Regulation of coronary blood flow in health and ischemic heart disease. Prog Cardiovasc Dis 2015;57:409–422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Herrmann J, Kaski JC, Lerman A. Coronary microvascular dysfunction in the clinical setting: from mystery to reality. Eur Heart J 2012;33:2771–2782b. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chih S, Macdonald PS, Feneley MP, Law M, Graham RM, McCrohon JA. Reproducibility of adenosine stress cardiovascular magnetic resonance in multi-vessel symptomatic coronary artery disease. J Cardiovasc Magn Reson 2010;12:42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Thomson LE, Wei J, Agarwal M, Haft-Baradaran A, Shufelt C, Mehta PK, Gill EB, Johnson BD, Kenkre T, Handberg EM, Li D, Sharif B, Berman DS, Petersen JW, Pepine CJ. Bairey Merz CN. Cardiac magnetic resonance myocardial perfusion reserve index is reduced in women with coronary microvascular dysfunction. A National Heart, Lung, and Blood Institute-sponsored study from the women's ischemia syndrome evaluation. Circ Cardiovasc Imaging 2015;8:e002481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ge Y, Antiochos P, Steel K, Bingham S, Abdullah S, Chen YY, Mikolich JR, Arai AE, Bandettini WP, Shanbhag SM, Patel AR, Farzaneh-Far A, Heitner JF, Shenoy C, Leung SW, Gonzalez JA, Shah DJ, Raman SV, Ferrari VA, Schulz-Menger J, Stuber M, Simonetti OP, Kwong RY. Prognostic value of stress CMR perfusion imaging in patients with reduced left ventricular function. JACC Cardiovasc Imaging 2020;13:2132–2145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Slivnick JA, Zareba KM, Truong VT, Liu E, Barnes A, Mazur W, Binkley P. Impairment in Quantitative Microvascular Function in Non-Ischemic Cardiomyopathy as Demonstrated Using Cardiovascular Magnetic Resonance. ResearchSquare. Available at: https://assets.researchsquare.com/files/rs-496006/v1/b70f630b-ade5-4e2a-809c-93b85274c613.pdf?c=1631882670. Accessed on Oct 14, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Masci PG, Schuurman R, Andrea B, Ripoli A, Coceani M, Chiappino S, Todiere G, Srebot V, Passino C, Aquaro GD, Emdin M, Lombardi M. Myocardial fibrosis as a key determinant of left ventricular remodeling in idiopathic dilated cardiomyopathy: a contrast-enhanced cardiovascular magnetic study. Circ Cardiovasc Imaging 2013;6:790–799. [DOI] [PubMed] [Google Scholar]
- 22.Masci PG, Marinelli M, Piacenti M, Lorenzoni V, Positano V, Lombardi M, L'Abbate A, Neglia D. Myocardial structural, perfusion, and metabolic correlates of left bundle branch block mechanical derangement in patients with dilated cardiomyopathy: a tagged cardiac magnetic resonance and positron emission tomography study. Circ Cardiovasc Imaging 2010;3:482–490. [DOI] [PubMed] [Google Scholar]
- 23.Neglia D, Liga R. Absolute myocardial blood flow in dilated cardiomyopathy: does it matter? JACC Cardiovasc Imaging 2019;12:1709–1711. [DOI] [PubMed] [Google Scholar]
- 24.Drzezga AE, Blasini R, Ziegler SI, Bengel FM, Picker W, Schwaiger M. Coronary microvascular reactivity to sympathetic stimulation in patients with idiopathic dilated cardiomyopathy. J Nucl Med 2000;41:837–844. [PubMed] [Google Scholar]
- 25.Crea F, Camici PG, Bairey Merz CN. Coronary microvascular dysfunction: an update. Eur Heart J 2014;35:1101–1111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Taqueti VR, Di Carli MF. Coronary microvascular disease pathogenic mechanisms and therapeutic options: JACC State-of-the-Art Review. J Am Coll Cardiol 2018;72:2625–2641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Reynolds HR, Shaw LJ, Min JK, Page CB, Berman DS, Chaitman BR, Picard MH, Kwong RY, O'Brien SM, Huang Z, Mark DB, Nath RK, Dwivedi SK, Smanio PEP, Stone PH, Held C, Keltai M, Bangalore S, Newman JD, Spertus JA, Stone GW, Maron DJ, Hochman JS. Outcomes in the ISCHEMIA trial based on coronary artery disease and ischemia severity. Circulation 2021;144:1024–1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Taylor AJ, Cerqueira M, Hodgson JM, Mark D, Min J, O’Gara P, Rubin GD, American College of Cardiology Foundation Appropriate Use Criteria Task Force, Society of Cardiovascular Computed Tomography, American College of Radiology, American Heart Association, American Society of Echocardiography, American Society of Nuclear Cardiology, North American Society for Cardiovascular Imaging, Society for Cardiovascular Angiography and Interventions, Society for Cardiovascular Magnetic Resonance, Kramer CM, Berman D, Brown A, Chaudhry FA, Cury RC, Desai MY, Einstein AJ, Gomes AS, Harrington R, Hoffmann U, Khare R, Lesser J, McGann C, Rosenberg A, Schwartz R, Shelton M, Smetana GW, Smith SC. ACCF/SCCT/ACR/AHA/ASE/ASNC/NASCI/SCAI/SCMR 2010 appropriate use criteria for cardiac computed tomography. A report of the American College of Cardiology Foundation Appropriate Use Criteria Task Force, the Society of Cardiovascular Computed Tomography, the American College of Radiology, the American Heart Association, the American Society of Echocardiography, the American Society of Nuclear Cardiology, the North American Society for Cardiovascular Imaging, the Society for Cardiovascular Angiography and Interventions, and the Society for Cardiovascular Magnetic Resonance. J Am Coll Cardiol 2010;56:1864–1894. [DOI] [PubMed] [Google Scholar]
- 29.McCrohon JA, Moon JC, Prasad SK, McKenna WJ, Lorenz CH, Coats AJ, Pennell DJ. Differentiation of heart failure related to dilated cardiomyopathy and coronary artery disease using gadolinium-enhanced cardiovascular magnetic resonance. Circulation 2003;108:54–59. [DOI] [PubMed] [Google Scholar]
- 30.Patel AR, Kramer CM. Role of cardiac magnetic resonance in the diagnosis and prognosis of nonischemic cardiomyopathy. JACC Cardiovasc Imaging 2017;10:1180–1193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Patel AR, Antkowiak PF, Nandalur KR, West AM, Salerno M, Arora V, Christopher J, Epstein FH, Kramer CM. Assessment of advanced coronary artery disease: advantages of quantitative cardiac magnetic resonance perfusion analysis. J Am Coll Cardiol 2010;56:561–569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Patel AR, Salerno M, Kwong RY, Singh A, Heydari B, Kramer CM. Stress cardiac magnetic resonance myocardial perfusion imaging: JACC Review Topic of the Week. J Am Coll Cardiol 2021;78:1655–1668. [DOI] [PMC free article] [PubMed] [Google Scholar]