Microcirculation Function in Non-ST-Elevation Myocardial Infarction After the Index Event and at Follow-Up Assessment

Abstract

Background

The assessment of coronary physiology is seldom considered in cases of non-ST-elevation myocardial infarction (NSTEMI). This study aimed to characterize coronary physiology and determine the incidence of microvascular dysfunction in the myocardial infarction (MI) culprit coronary artery during the acute phase of NSTEMI and subsequent follow-up evaluation.

Methods

This study included 30 patients hospitalized for NSTEMI. A physiological assessment of the MI culprit coronary artery was performed using fractional flow reserve, coronary flow reserve (CFR), and index of microcirculatory resistance (IMR). At a median of 7 months after MI, patients underwent repeated physiological assessment of the same coronary artery.

Results

Microvascular dysfunction identified using CFR (< 2.0) was frequently present (60%) during the acute phase of NSTEMI, whereas severe microcirculatory dysfunction (IMR > 40) was uncommon (17%). Over time, a significant reduction occurred in the prevalence of abnormal CFR values (< 2.0; 60% vs 26%, P = 0.013) and extensive microvascular resistance (IMR > 40; 17% vs 4%, P = 0.03) at the follow-up evaluation. Patients were categorized according to their CFR and IMR results. In patients with abnormal CFR (< 2.0) and normal IMR (< 25), reduced CFR is attributable to elevated resting coronary blood flow rather than diminished hyperemic flow.

Conclusions

In the acute phase of NSTEMI, reduced CFR was commonly observed in the MI culprit coronary artery, whereas severe microcirculatory dysfunction was infrequent.

Non-ST-segment elevation myocardial infarction (NSTEMI) is the predominant form of acute coronary syndrome (ACS), which is characterized by diverse pathophysiological mechanisms and varying risk levels for mortality and major adverse cardiac events (MACE).¹

Despite the efficacy of contemporary reperfusion therapy in effectively restoring epicardial coronary flow, patients frequently experience microcirculatory dysfunction, which is associated with adverse outcomes.² Moreover, most studies on coronary physiology in ACS have focused on patients with ST-segment elevation myocardial infarction (STEMI),³ and on the value of fractional flow reserve (FFR),⁴ disregarding the prevalence of microvascular dysfunction following NSTEMI.

Notwithstanding the current hemodynamic data on patients with ACS, a comprehensive description of epicardial and microvascular functions in NSTEMI and their behavior over time has not yet been established. This study aimed to characterize coronary physiology and determine the incidence of microvascular dysfunction in the culprit coronary artery during the early phase of NSTEMI and during subsequent followup.

Methods

Study design and settings

We conducted a single-centre consecutive study of 30 patients hospitalized for NSTEMI who were enrolled between January 1 and April 30, 2020. Patients were considered eligible for inclusion if they were aged > 18 years and had a clinical diagnosis of NSTEMI, defined according to the universal definition of myocardial infarction (MI) criteria (patients with chest discomfort in the absence of ST-segment elevation or left bundle branch block on electrocardiogram (ECG) and elevation of high-sensitivity cardiac troponin (hsTn) level, with at least one value above the 99th percentile of the upper reference limit)⁵ and provided written informed consent prior to enrollment. The key exclusion criteria were as follows: intolerance to adenosine or contrast media; baseline mean arterial pressure < 60 mm Hg; baseline heart rate < 50 beats per minute; severe chronic kidney disease (glomerular filtration rate < 30 mL/min per m²); concomitant valve disease (greater than mild on echocardiography) or other structural cardiomyopathy; recent acute coronary syndrome (< 6 months); coronary artery disease (CAD) anatomy considered for coronary artery bypass graft (CABG), or ischemic myocardial injury in the context of a mismatch between oxygen supply and demand (type 2 MI).⁵ Repeat coronary angiography was performed within 6 to 10 months of the index angiography procedure. The MI culprit coronary artery was determined by angiography by the operator and later reviewed by an independent physician, combining data from 12-lead ECG, echocardiography, and coronary angiography or intracoronary imaging (Fig. 1). Coronary angiography findings of acute plaque rupture, acute vessel occlusion, or intracoronary imaging suggestive of an acute coronary lesion were considered key features for establishing type 1 MI and the culprit coronary artery.^1,2,5

Figure 1.

Flow diagram of the study. Values in square brackets are interquartile range. ACS, acute coronary syndrome; CABG, coronary artery bypass grafting; CFR, coronary flow reserve; CKD, chronic kidney disease; FFR, fractional flow reserve; IMR, index of microcirculatory resistance; MI, myocardial infarction; MINOCA, MI with nonobstructive coronary arteries; PCI, percutaneous coronary intervention.

Demographic, anthropometric, clinical, laboratory, echocardiographic, and angiographic characteristics were recorded, including follow-up information on all-cause death and MACE, defined as cardiovascular death, MI readmission, and heart failure–related hospital admission. All data were anonymized prior to statistical analyses, and the study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the local hospital ethics committee (CE-023/2017).

Data collection

Upon hospital admission, blood samples were collected from each patient for routine biochemical testing. The VITROS Troponin I ES Assay (Ortho Clinical Diagnostics Rochester, NY) was used to measure troponin levels. The 99th percentile for sensitivity and detection of this test was 0.034 ng/mL, whereas the lower limit was 0.012 ng/mL. Re-elevation of hsTn levels after percutaneous coronary intervention (PCI) was not regarded as an exclusion criterion in this study and was considered in patients with decreasing or stable biomarker levels, in whom a new troponin-level increase occurred within 24 hours after PCI (> 10x upper limit of normal).⁶ Peripheral venous blood samples were obtained upon admission and throughout the patient's hospitalization to obtain laboratory data. Blood samples for troponin analysis were obtained upon admission and at 12 and 24 hours after MI. Subsequently, measurements were performed daily until hospital discharge. The maximum troponin level observed during hospitalization was designated as peak hsTn.

Standard 12-lead ECGs were obtained at admission and during the hospital stay. The presence of ST-segment deviation, T-wave, Q-wave, conduction, and rhythm abnormalities was recorded. Routine transthoracic echocardiography was performed during the index hospital admission (interquartile range [IQR] 2.7 days after admission) and throughout follow-up using a Vivid 7 ultrasound device (GE Healthcare, Horton, Norway). Echocardiographic studies with standard views were performed according to established guidelines, with the left ventricular ejection fraction (LVEF) calculated using Simpson’s method.⁷ The Global Registry of Acute Coronary Events risk score for unstable angina/NSTEMI was calculated according to the guidelines.¹ Clinical follow-up after hospital discharge was performed using the patient’s hospital records, the general practitioner’s records, questionnaires, and the national registry of vital status.

Procedure

Coronary angiography during hospitalization was regarded as a mandatory inclusion criterion for the study. The baseline medications administered to the study participants are presented in Table 1. However, attending physicians were given the discretion to administer vasoactive agents when such were deemed clinically necessary before coronary angiography. A myocardial revascularization procedure was considered for those with significant obstructive CAD, which was defined as a ≥ 70% diameter narrowing of a major coronary artery (or, in the case of the left main coronary artery, an obstruction of ≥ 50% of its diameter).² The decision to consider coronary surgery or PCI (culprit-only or complete revascularization) was based on contemporary evidence and guidelines,² and according to the local standard clinical practice. The TIMI (thrombolysis in myocardial infarction) frame count was used to assess slow and/or no-reflow phenomena on coronary angiography.⁸ After coronary angiography and PCI of the culprit vessel, a hemodynamic assessment was performed using the Coroventis CoroFlow Cardiovascular System (Abbott Laboratories, Chicago, IL) with a pressure–temperature sensor-tipped guidewire, the Abbott Pressure Wire X Guidewire. In accordance with the research protocol, an intracoronary bolus of 1 mg isosorbide dinitrate was administered 2-3 minutes before coronary physiology under resting conditions was evaluated. Following pressure wire equalization, it was advanced to the distal third of the MI culprit coronary artery to simultaneously measure the coronary pressure at both the proximal (Pa) and distal (Pd) artery locations. Transit time was measured using three 3-mL injections of heparinized saline at room temperature and averaged to obtain the mean transit time (Tmn). At baseline, the inverse of Tmn served as a surrogate for coronary blood flow (CBF = 1/Tmn_resting).⁹ Subsequently, adenosine-mediated hyperemia was induced via peripheral venous infusion at a dose adjusted for body weight (140 μg/kg/min).¹⁰ Under conditions of pharmacologic hyperemia, several measurements were calculated. These included hyperemic CBF (CBF_hyperemia = 1/Tmn_hyperemia), the index of microcirculatory resistance (IMR = Pd_hyperemia×Tmn_hyperemia), and coronary flow reserve (CFR = [1/Tmn_hyperemia] / [1/Tmn_resting]). To assess cardiac work, the rate pressure product (RPP = systolic blood pressure × heart rate) and normalized CBF (CBF_norm = [1/Tmn] / RPP) were determined.¹¹ Microvascular responses to adenosine were used to categorize patients. A CFR < 2.0 was considered low, and an IMR of ≥ 25 was considered high.^9,10 Regarding the subsequent physiological assessment, patients were instructed to discontinue any vasoactive medications for 5 days prior to the repeat coronary procedure, which was scheduled for a time between 6 to 10 months after hospital discharge, utilizing the same procedural methodology as previously described.

Table 1.

Baseline characteristics of the study cohort

Baseline characteristics	Value Measure
	n = 30		n = 30
Male sex	23 (77)	Medication at the index event
Age, y	65 [58, 74]	Aspirin / P2Y12 receptor inhibitor	25 (83)
HTN	23 (77)	Oral anticoagulant	4 (13)
Dyslipidemia	28 (93)	Statin	28 (93)
Diabetes mellitus	16 (53)	ACE inhibitor/ARB/ARNI	28 (93)
Smoker/previous smoker	13 (43)	Beta-blocker	23 (77)
BMI, kg/m2	28 [26, 31]	Risk scores
GFR < 45 mL/min per 1.73 m2	0	GRACE6m score	122 [108, 139]
Previous MI and/or PCI	5 (17)	GRACE6m death risk, %	8 [5, 14]
Previous CABG	0	Coronary angiography
Previous HF admission	5 (17)	1–vessel CAD	14 (47)
Creatinine, mg/dL	0.8 [0.7, 0.9]	2–vessel CAD	9 (30)
NTproBNP, pg/mL	1218 [342, 1921]	3-vessel CAD	7 (23)
Peak hsTn, ng/mL	7121 [1267–14,777]	Culprit coronary artery
ECG ST-segment deviation	16 (53)	Left descending artery/branches	9 (30)
ECG negative T-wave	11 (37)	Circumflex/branches	9 (30)
LVEF, %	57 [48, 63]	Right coronary artery/branches	12 (40)

Values are n (%) or median [interquartile range].

ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; BMI, body mass index; CABG, coronary artery bypass grafting; CAD, obstructive coronary artery disease; ECG, 12-lead electrocardiogram; GFR, glomerular filtration rate; GRACE, Global Registry of Acute Coronary Events; HF, heart failure; hsTn, high-sensitivity cardiac troponin; HTN, arterial hypertension; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NTproBNP, N-terminal proB-type natriuretic peptide; PCI, percutaneous coronary intervention; P₂Y₁₂, adenosine diphosphate (ADP)-receptor antagonist.

Statistical analysis

Continuous variables are presented as mean ± standard deviation for normally distributed variables and median values (25th-75th percentiles) for abnormally distributed variables. Categorical variables were presented as counts and proportions. The normality and homogeneity of the variances were tested using the Kolmogorov-Smirnov and Levene tests. Categorical variables were compared using the χ² test or Fisher’s exact test, and the Student t test for normally distributed data and the Mann-Whitney test for non-normally distributed data. Paired Student t tests were used to evaluate the differences in continuous physiological variables between the baseline and follow-up coronary procedures. The Spearman correlation coefficient was used to investigate the associative relationships. To evaluate the differences in CBF, IMR RPP, and normalized RPP during rest and hyperemia, a repeated-measures analysis of variance was employed. This analysis was conducted for both the initial and follow-up coronary procedures across 4 patient groups, as follows: (i) low CFR and low IMR; (ii) low CFR and high IMR; (iii) high CFR and low IMR; and (iv) high CFR and high IMR groups. When a significant interaction was detected, a post-hoc analysis was conducted using the Bonferroni adjustment. Statistical analyses were performed using SPSS version 22.0 (SPSS, Chicago, IL). P-values with a 2-sided α-level of < 0.05 were considered statistically significant.

Results

Table 1 presents the baseline characteristics of the study cohort (N = 30). The mean length of hospital stay was 3.1 ± 0.9 days. At admission, in the standard 12-lead ECG, most NSTEMI patients (N = 29; 97%) presented with ischemic changes (ST-segment and/or T-wave changes). Multivessel obstructive CAD was found in 16 patients (53%), with the right coronary artery (RCA) being the most common MI culprit coronary artery (N = 12; 40%). Regarding patients with multivessel obstructive disease, 9 (56%) underwent revascularization of the nonculprit coronary arteries during the hospital stay. The main reason for not undergoing complete revascularization during the hospital stay was diffuse coronary disease of the nonculprit artery. In addition, 2 patients (13%) had their myocardial revascularization deferred during the follow-up period: 1 patient underwent PCI for chronic coronary occlusion (15 months after hospital admission), and 1 patient was referred for CABG (14 months after the index ischemic event) due to recurrent angina related to previous stent restenosis.

No in-hospital mortality occurred during the index admission. The median clinical follow-up period was 50 [IQR 42-54] months, and a 3-year follow-up was achieved in 93% (N = 28) of the cohort. Three patients (10%) died during the follow-up period, 1 in the first 6 months (cardiovascular death) and the other after 42 months of follow-up (1 cardiovascular and 1 noncardiovascular death). There were 2 MI readmissions (7%), and no hospital admissions for heart failure occurred during the follow-up period.

Physiological assessment of the culprit coronary artery at the index event and follow-up

Coronary angiography procedures (N = 30) were performed mostly on the first day of hospitalization (median, 1 [IQR 0, 2] days). The MI culprit coronary arteries were the left anterior descending artery (N = 9; 30%), the circumflex artery (N = 9; 30%), and the RCA (N = 12; 40%). During the initial procedure, PCI of the culprit coronary artery typically involved the placement of 1 stent (median, 1 [IQR 1, 1] stent; Tables 2 and 3). The stents used had a median diameter of 2.75 [IQR 2.5, 3.0] mm and a median stent length of 23 [IQR 15,33] mm. During PCI for acute MI, a single case of the slow reflow phenomenon was observed, which persisted until the end of the procedure. This patient demonstrated the highest microvascular resistance in the cohort during the initial and subsequent coronary procedures (CFR_initial = 1.5 / IMR_initial = 110; CFR_follow-up = 3.4 / IMR_follow-up = 66). No additional complications were reported in relation to the PCI.

Table 2.

Physiological assessment of the myocardial infarction culprit coronary artery at the index event and follow-up

Measure	Index event (N = 30)	Follow-up (N = 27)	P
Resting Pa, mm Hg	86 [74, 93]	85 [81, 94]	ns
Resting Pd, mm Hg	84 [62, 90]	82 [74, 88]	ns
Resting Tmn, s	0.70 [0.38, 1.15]	0.72 [0.43, 1.0]	ns
Resting Pd/Pa	0.95 [0.91, 0.98]	0.95 [0.90, 0.99]	ns
Rate pressure product	6058 [4709, 7098]	6280 [5094, 7114]	ns
Normalized coronary blood flow	2.3 [1.5, 4.1]	2.0 [1.5, 4.3]	ns
Hyperemia Pa, mm Hg	79 [64, 84]	79 [73, 92]	ns
Hyperemia Pd, mm Hg	71 [54, 81]	70 [63, 81]	ns
Hyperemia Tmn, s	0.31 [0.19, 0.55]	0.27 [0.17, 0.38]	ns
Flow fractional reserve	0.90 [0.86, 0.99]	0.92 [0.83, 0.99]	ns
Flow fractional reserve < 0.8	5 (17)	4 (15)	ns
Coronary flow reserve	1.8 [1.4, 2.7]	2.6 [1.8, 3.6]	0.079
Coronary flow reserve < 2.0	18 (60)	7 (26)	0.013
IMR	21 [12, 34]	20 [12, 30]	ns
adjIMR	22 [14, 33]	19 [12, 31]	ns
IMR ≥ 25 and > 40	11 (37), 5 (17)	7 (26), 1 (4)	ns, 0.03
Resistive reserve ratio	2.3 [1.6, 3.2]	2.8 [1.8, 4.3]	ns

Values are median [interquartile range] or n (%), unless otherwise indicated.

adjIMR, adjusted IMR; IMR, index of microcirculatory resistance; ns not significant; Pa, proximal coronary artery pressure; Pd, distal coronary artery pressure; Tmn, mean transient time.

Table 3.

Physiological assessment according to myocardial infarction culprit coronary arteries at the index event and follow-up

Left anterior descending	Index event	Follow-up	P
	N = 9	N = 9
Resting Pa, mm Hg	83 [73, 103]	85 [83, 94]	ns
Resting Pd, mm Hg	77 [57, 98]	80 [71, 91]	ns
Resting Tmn, s	0.79 [0.45, 1.15]	0.71 [0.53, 0.96]	ns
Rate pressure product	5701 [5092, 7197]	6800 [5600, 7392]	ns
Normalized coronary blood flow	2.0 [1.6, 4.0]	1.5 [1.9, 3.6]	ns
Hyperemia Pa, mm Hg	83 [61, 90]	93 [75, 100]	ns
Hyperemia Pd, mm Hg	63 [46, 80]	80 [62, 80]	ns
Hyperemia Tmn, s	0.53 [0.32, 0.75]	0.36 [0.2, 0.37]	0.073
Flow fractional reserve	0.87 [0.73, 0.90]	0.85 [0.83, 0.90]	ns
Flow fractional reserve < 0.8 (dimensionless ratios)	2 (22)	1 (11)	ns
Coronary flow reserve	1.4 [1.3, 1.7]	2.2 [1.9, 2.8]	0.048
Coronary flow reserve < 2.0 (dimensionless ratios)	7 (78)	3 (33)	ns
IMR	30 [24, 40]	24 [12, 32]	ns
IMR ≥ 25 U and > 40 U	2 (22), 2 (22)	4 (44), 0	ns, ns
Resistive reserve ratio	2.0 [1.6, 2.7]	2.7 [1.8, 2.8]	ns
Low CFR and low IMR; low CFR and high IMR∗	2 (22); 6 (66)	2 (22), 1 (11)	—
High CFR and low IMR; high CFR and low IMR∗	0; 1 (11)	3 (33), 3 (33)	—
Circumflex	N = 9	N = 8
Resting Pa, mm Hg	86 [77, 99]	85 [82, 96]	ns
Resting Pd, mm Hg	86 [73, 99]	84 [82, 88]	ns
Resting Tmn, s	0.59 [0.28, 0.75]	0.69 [0.40, 1.16]	ns
Rate pressure product	6450 [5320, 7841]	6020 [4479, 6417]	ns
Normalized coronary blood flow	2.8 [2.2, 5.5]	2.0 [1.4, 5.5]	ns
Hyperemia Pa, mm Hg	78 [72, 86]	79 [67, 85]	ns
Hyperemia Pd, mm Hg	74 [66, 86]	70 [67, 81]	ns
Hyperemia Tmn, s	0.20 [0.16, 0.28]	0.19 [0.17, 0.43]	ns
Flow fractional reserve	0.99 [0.88, 1.0]	0.98 [0.86, 1.00]	ns
Flow fractional reserve < 0.8 (dimensionless ratios)	1 (11)	1 (13)	ns
Coronary flow reserve	2.7 [1.6, 3.7]	2.6 [1.7, 4.5]	ns
Coronary flow reserve < 2.0 (dimensionless ratios)	4 (44)	2 (25)	ns
IMR	17 [12, 22]	14 [12, 30]	ns
IMR ≥ 25 U and > 40 U	1 (11), 0	2 (25), 0	ns
Resistive reserve ratio	2.5 [1.3, 4.5]	3.3 [1.6, 6.0]	ns
Low CFR and/or low IMR; low CFR and/or high IMR∗	3 (33); 1 (11)	2 (25); 0	—
High CFR and/or low IMR; high CFR and/or low IMR∗	5 (55); 0	5 (63); 1 (13)	—
Right coronary artery	N = 12	N = 10
Resting Pa, mm Hg	86 [60, 91]	82 [78, 94]	ns
Resting Pd, mm Hg	82 [58, 88]	78 [74, 85]	ns
Resting Tmn, s	0.95 [0.36, 1.5]	0.96 [0.25, 1.6]	ns
Rate pressure product	5280 [4300, 6370]	5986 [4212, 7050]	ns
Normalized coronary blood flow	2.9 [1.4, 4.3]	2.4 [1.1, 5.7]	ns
Hyperemia Pa, mm Hg	78 [58, 84]	77 [68, 78]	ns
Hyperemia Pd, mm Hg	65 [55, 72]	69 [61, 78]	ns
Hyperemia Tmn, s	0.42 [0.18, 0.77]	0.30 [0.16, 0.41]	ns
Flow fractional reserve	0.92 [0.90, 0.97]	0.97 [0.81, 1.00]	ns
Flow fractional reserve < 0.8	1 (8)	1 (10)	ns
Coronary flow reserve	1.9 [1.6, 2.7]	3.4 [1.1, 4.2]	ns
Coronary flow reserve < 2.0	6 (50)	2 (20)	ns
IMR	20 [13, 44]	19 [9, 31]	ns
IMR ≥ 25 and > 40	6 (50), 3 (25)	2 (20), 1 (10)	ns, ns
Resistive reserve ratio	2.6 [1.8, 3.6]	3.4 [1.1, 6.0]	ns
Low CFR and/or low IMR; low CFR and/or high IMR∗	4 (33); 3 (25)	3 (30); 0	—
High CFR and/or low IMR; high CFR and/or low IMR∗	2 (17); 3 (25)	4 (40); 3 (30)	—

Values are median [interquartile range] or n (%), unless otherwise indicated.

CFR, coronary flow reserve; IMR, index of microcirculatory resistance; ns, not significant; Pa, proximal coronary artery pressure; Pd, distal coronary artery pressure; Tmn, mean transient time.

^∗Microcirculation function groups, according to CFR < 2.0 and IMR ≥ 25 values.

During the follow-up period, most patients underwent repeat coronary angiography (N = 27; 90%). One patient died before the next procedure, and 2 refused repeat (elective) coronary catheterization. Repeat coronary angiography was performed at a median of 7 [IQR 7, 10] months after the index procedure. Cardiac workload, as assessed using RPP, did not differ significantly between the initial and subsequent coronary procedures (Table 2).

At the initial procedure, the resting Tmn was 0.70 [IQR 0.38, 1.15] seconds, which was comparable to the resting Tmn measured during the follow-up procedure (0.72 [IQR 0.43, 1.0] seconds; P = 0.635). Similarly, the hyperemia Tmn measured at the index event (0.31 [IQR 0.19, 0.55] seconds) showed no significant change from the hyperemia Tmn value observed in the follow-up period (0.27 [IQR 0.17, 0.38] seconds; P = 0.102). No statistically significant correlation was observed (P > 0.05) between stent number, length, or size and resting/hyperemia Tmn in the initial or follow-up coronary procedures. The median FFR was 0.90 [IQR 0.86, 0.99] at the initial procedure and 0.92 [IQR 0.83, 0.99] at follow-up (P = 0.942). The median IMR was 21 [12, 34] at baseline and 20 [12, 30] at follow-up (P = 0.210). No significant changes were observed in either FFR or IMR measurements between the initial and subsequent coronary procedures (P = 0.942 and P = 0.210, respectively). However, the median CFR was 1.8 [1.4, 2.7] at baseline and 2.6 [1.8, 3.6] at follow-up, showing a tendency toward CFR improvement over time (P = 0.079). Our analysis revealed no statistically significant correlation (P > 0.05) between the number, length, or size of the implanted stents and the FFR, CFR, or IMR measured during the initial or follow-up coronary procedures (Table 2; Fig. 2).

Figure 2.

(A) Fractional flow reserve (FFR), (B) coronary flow reserve (CFR), and (C) index of microcirculatory resistance (IMR) measurements, at the index event and follow-up; values are median [interquartile range]: (A) Index event: 0.90 [0.86, 0.95]; follow-up: 0.86 [0.83, 0.97]; (B) Index event: 1.8 [1.4, 2.7]; follow-up: 2.6 [1.8, 3.6]; (C) Index event: 21 [12, 34]; follow-up: 20 [12, 30].

Table 3 presents the physiological coronary assessment results according to the MI culprit coronary arteries. Overall, the circumflex artery exhibited no significant changes in microvascular function over time, and severe microvascular resistance (IMR > 40) was not observed in these subjects. In relation to the RCA, numerical improvements observed in hyperemia Tmn, CFR, and IMR measurements during the follow-up period were not statistically significant (P > 0.05). The left anterior descending artery was the culprit coronary artery that most frequently presented with impaired CFR, with 78% of the subjects demonstrating a CFR < 2.0. Follow-up measurements indicated a trend toward decreased hyperemia Tmn (P = 0.073) and a statistically significant improvement in CFR over time (P = 0.048). In addition, Supplemental Table S1 presents the individual physiological measurements of the participants who did not undergo subsequent coronary procedures during the follow-up period.

At the index coronary catheterization, abnormal CFR measurements (< 2.0) were observed in 60% (N = 18) of the cases, high microvascular resistance (IMR ≥ 25) was observed in 37% (N = 11), and extensive microcirculatory dysfunction (IMR > 40) was observed in 17% (N = 5) of the MI culprit coronary arteries. Ten patients (33%) exhibited low CFR and/or low IMR, 9 (30%) presented low CFR and/or high IMR, 7 (23%) exhibited high CFR and/or low IMR, and 4 (13%) presented high CFR and/or high IMR. Patients with low CFR and/or low IMR exhibited significantly higher resting CBF than all other physiological groups (P < 0.05). Individuals with low CFR and low IMR had greater hyperemic CBF than those with low CFR and/or high IMR (P < 0.001) and high CFR and/or high IMR (P = 0.012), after adjusting for RPP (Fig. 3A). Additionally, these patients demonstrated comparable hyperemic CBF (4.5 ± 1.6 vs 5.2 ± 1.5, P = 0.337) and IMR (15.6 ± 5.1 vs 15.4 ± 5.4, P = 0.962) to those with “normal” microvascular status (high CFR and/or low IMR) after adjusting for cardiac workload (Fig. 3A). Across all patient groups, resting CBF was correlated strongly with hyperemic CBF (r = 0.77; P < 0.001).

Figure 3.

(A) Normalized Coronary Blood Flow under resting and hyperemia conditions, at the initial coronary physiological procedure. (B) Normalized Coronary Blood Flow under resting and hyperemia conditions, at follow-up coronary physiological procedure. bpm, beats per minute; CFR, coronary flow reserve; IMR, index of microcirculatory resistance.

Over time, the prevalence of abnormal CFR values (< 2.0; 60% vs 26%, P = 0.013) and very high microvascular resistance (IMR > 40) significantly improved (17% vs 4%, P = 0.03) at the follow-up evaluation. During the follow-up assessment, the patients were categorized as follows: 8 (30%) exhibited low CFR and/or low IMR, 1 (4%) had low CFR and/or high IMR, 12 (44%) exhibited high CFR and/or low IMR, and 6 (22%) exhibited high CFR and/or high IMR. The group with low CFR and/or low IMR displayed a higher resting CBF than the other physiological groups (Fig. 3B; P < 0.05). A comparison of the low CFR and/or low IMR group to the “normal” physiological group (high CFR and/or low IMR) showed that no significant differences were observed in hyperemic CBF (4.6 ± 2.3 vs 5.4 ± 2.0, P = 0.455) or IMR (13.0 ± 6.9 vs 14.3 ± 5.6, P = 0.685) after adjusting for cardiac workload (Fig. 3B). Across all patient groups, resting CBF was correlated with hyperemic CBF (r = 0.65; P = 0.001).

Subsequently, the patients were classified as having elevated resting CBF based on the highest quartile of the CBF values. Among those identified as having low CFR and/or low IMR, 6 (60%) in the initial assessment and 5 patients (63%) in the follow-up coronary procedure exhibited the highest resting CBF in the cohort.

Cardiac biomarkers and events and physiological assessment of the culprit coronary artery

At the index event, the median peak troponin level was 7121 [1267, 14777] ng/mL, and the median peak N-terminal proB-type natriuretic peptide level was 1218 [342, 1921] pg/mL. Peak troponin levels were higher in patients with multivessel disease than they were in those with 1-vessel CAD (14,060 vs 5961 ng/mL, P = 0.04). The incidence of re-elevation of troponin levels after PCI was 17% (N = 5). Past medical history, stent number and/or dimensions, N-terminal proB-type natriuretic peptide and hsTn levels, re-elevation of troponin after PCI, LVEF, all-cause death, and MACE were not significantly correlated with FFR, CFR, or IMR measurements at hospital admission or during the follow-up period. However, peak troponin levels during hospital admission showed a positive trend with the IMR (r = 0.405, P = 0.061) at follow-up assessment. During the initial coronary procedure, an IMR ≥ 25 was associated with anterior MI (P = 0.030), whereas an IMR > 40 was associated with slow reflow phenomena (P = 0.029). An IMR > 40 at follow-up was significantly associated with higher peak troponin levels (9399 vs 35,000 ng/mL; P = 0.013) during hospital admission. In addition, the occurrence of slow reflow in the index PCI was associated with an IMR > 40 in the follow-up assessment (P < 0.001).

Discussion

This study describes the coronary physiological assessment of patients with type 1 NSTEMI at baseline and during the follow-up assessment. We used a multimodal assessment with FFR, CFR, and IMR of the culprit coronary artery following PCI and repeated the physiological evaluation later in the follow-up period, to assess how coronary circulation recovered after the ischemic event. In our cohort, multivessel CAD was observed often (53%), and abnormal CFR (< 2.0) was frequently present (60%); however, extensive microvascular resistance (IMR > 40) was an unusual finding (17%). Over time, we found a significant reduction in the prevalence of patients exhibiting low CFR and severe microvascular resistance (IMR > 40). In cases of abnormal CFR (< 2.0) and normal IMR (< 25), patients with NSTEMI presented with increased resting CBF rather than compromised microvascular vasodilatation.

Different pathological mechanisms can be observed in NSTEMI, which often display clinical and prognostic differences.^1,12 In type 1 NSTEMI, rupture or erosion of an atherothrombotic plaque occurs within the affected coronary artery. This condition typically presents as a nonocclusive thrombus in the MI culprit coronary artery and results in subendocardial injury and/or necrosis of the heart tissue.⁵ Our research protocol employed various criteria to assess MI mechanisms and enhance consistency in recruiting patients with type 1 NSTEMI. In this clinical setting, invasive coronary physiological assessment with FFR was validated previously in a post hoc analysis of medically stabilized ACS in the Fractional Flow Reserve Versus Angiography for Multivessel Evaluation (FAME) trial¹³ and later in the Fractional flow reserve versus angiography in guiding management to optimise outcomes in non-ST segment elevation myocardial infarction (FAMOUS-NSTEMI) trial,⁴ which compared angiography-guided and FFR-guided approaches in patients with NSTEMI. However, an invasive microcirculatory approach with CFR and IMR to inform treatment decisions is not the standard of care in clinical practice.²

CFR represents the vasodilator capacity of a vascular bed during hyperemia and measures coronary flow through both the epicardial vessels and microvasculature. An impaired CFR is associated with an increased risk of all-cause mortality and MACE in patients with established cardiovascular disease.¹⁴ Studies reporting the prognostic value of CFR have been derived largely from cohorts of chronic coronary syndromes, mostly using noninvasive cardiac imaging. However, 2 previously published invasive studies provided CFR prognostic data for patients with NSTEMI. De Waard et al. invasively measured CFR and hyperemic microvascular resistance in 176 ACS patients, although only 26% had NSTEMI.¹⁵ They reported that a reduced CFR (< 1.5) predicted death and heart failure hospitalization and was associated with cardiac magnetic resonance–defined microvascular injury. In addition, Murai et al. evaluated the prognostic efficacy of physiological indices and reported that a CFR < 2.0 had weak but significant prognostic power in a cohort of 83 patients with NSTEMI.¹⁶ Recently, in a STEMI study, Flores et al. reported that the CFR of the nonculprit artery measured in the acute phase was a powerful prognostic marker for the occurrence of MACE.¹⁷ None of the aforementioned studies reported physiological measurements according to the MI culprit coronary artery (Table 3) or conducted subsequent physiological assessments during the follow-up period.

IMR is a thermodilution technique that allows for the quantitative assessment of microcirculatory resistance in the coronary artery territory. In acute MI, microvascular blockage, extravascular compression from interstitial edema, and intramyocardial bleeding can increase microcirculatory resistance, limiting the maximal coronary flow. An IMR of < 25 is typically used to define patients without microvascular dysfunction. Abnormal IMR values correlate inconsistently with infarct size and microvascular obstruction in the literature, and IMR prognostic data are derived largely from STEMI studies that typically present transmural myocardial damage.¹⁸ Nonetheless, patients with MI presenting with an IMR > 40 (higher specificity for microvascular dysfunction) are more likely to present with microvascular obstruction on cardiac imaging and MACE during follow-up assessment.¹⁹ In our study, only approximately one-third of patients (37%) presented with abnormal microcirculatory resistance (IMR ≥ 25) in the culprit coronary artery. Microcirculatory resistance was not elevated when the circumflex artery was the MI culprit coronary artery, possibly due to the characteristic subendocardial ischemic injury of NSTEMI and the smaller infarct size of the circumflex myocardial blood supply.²⁰ During follow-up assessment, we did not find an overall improvement in IMR values. However, a significant reduction in severe microcirculatory impairment (IMR > 40) was observed over time (17% vs 4%, P = 0.03; Table 2).

No association was observed between death and/or MACE and microvascular hyperemic indices in our study. This lack of association may be attributed to the small sample size and the relatively minor extent of myocardial damage following NSTEMI, as evidenced by the well-preserved LVEF (median, 57%) at the time of hospital admission. Nevertheless, surrogate markers of more severe cardiac injury, such as anterior MI, peak troponin levels, and the occurrence of slow reflow during the initial coronary procedure, were predictive of increased microcirculatory resistance. This finding underscores the potential incremental risk stratification value of IMR in identifying myocardial scar burden following MI.

The present study demonstrates that microvascular dysfunction frequently is observed during the acute phase of NSTEMI and often persists in post-MI patients. Therefore, cardioprotective interventions that have the potential to mitigate damage from myocardial ischemia and reperfusion may improve outcomes associated with microvascular dysfunction.³ Most published data have not shown significant clinical benefits from therapies aimed at enhancing adenosine or nitric oxide action. Similarly, several therapies used in the MI setting, such as P₂Y₁₂ or glycoprotein II_b/III_a inhibitors, and fibrinolytics, which showed cardioprotective effects in preclinical models, frequently failed to replicate their efficacy in clinical trials.¹⁸ However, the most promising approach for achieving cardioprotection in MI patients may involve a combination of mechanical ischemic postconditioning and pharmacologic intervention, in addition to reperfusion.²¹

In our cohort, microvascular dysfunction was identified mostly by an abnormal CFR (< 2.0). This condition tended to improve over time, likely due to microvascular changes, as FFR remained unchanged during the follow-up period, and the occurrence of new angiographic coronary lesions was infrequent. Nevertheless, because CFR represents the ratio of hyperemic to resting CBF, the question of whether the result is attributable to an increased resting CBF, a decreased hyperemic CBF, or a combination of both factors cannot be answered. Through a concurrent analysis of CFR and IMR data, patients were classified into 4 distinct physiological categories, revealing subtypes of microvascular dysfunction. Patients with both low CFR and low IMR demonstrated significantly elevated resting CBF compared with other physiological categories, even after adjusting for cardiac workload. Moreover, hyperemic CBF was comparable to the measurements obtained in the “normal” category (high CFR and/or low IMR). Our results suggest that the low CFR and/or low IMR group did not experience compromised microvascular vasodilation. A higher baseline CBF, rather than a decreased hyperemic CBF, may be the underlying factor contributing to the reduced CFR in this group. Our findings are consistent with those of a recent study by Nardone et al. involving 199 individuals with chest pain and nonobstructive coronary artery disease.¹¹ They reported that participants with low CFR and/or low IMR demonstrated increased CBF while maintaining hyperemic CBF. The underlying causes of elevated resting CBF in patients with low CFR and/or low IMR remain uncertain and may be attributable to diminished cardiac efficiency or uncoupled coronary circulation.¹¹ Previous studies have demonstrated that individuals with low CFR and/or low IMR exhibit increased resting CBF, which is attributed to enhanced baseline production of nitric oxide or insulin resistance.^22,23 Moreover, therapies aimed at enhancing vasodilatation are unlikely to be effective or may be detrimental in patients with increased resting CBF. A potentially more suitable therapeutic approach should focus on addressing reduced myocardial efficiency (eg sodium glucose co-transporter 2 [SGLT2] inhibitors) or uncoupled CBF (eg, antiplatelet agents).^24,25 Consequently, our findings suggest that CBF evaluation should be consistently incorporated into the physiological assessment of the coronary arteries.

Study limitations

This study constituted a preliminary NSTEMI investigation conducted with a limited sample size. This limitation increases the risk of type II errors, as reduced statistical power may impede the detection of potential relationships between microvascular function and the outcomes. Nonetheless, our data provide a comprehensive invasive physiological evaluation of the coronary microcirculation following NSTEMI in the MI culprit coronary artery, offering a dynamic perspective on post-MI microcirculatory changes over time. An additional limitation is the influence of medication administered prior to hospital admission, although most patients were already receiving therapies that could potentially affect coronary microcirculation (antiplatelet drugs, angiotensin-converting enzyme inhibitors or aldosterone receptor antagonists [ACEi/ARA], statins, and beta-blockers) at the time of the initial coronary assessment.²³ After hospital discharge, patients received medical treatment in accordance with standard MI guidelines. Moreover, the possibility of variations in microcirculatory function arising from its evaluation immediately after PCI (acute phase) or during the follow-up period (stable phase) is important to consider. Finally, the research protocol did not include a physiological assessment before PCI, which prevented researchers from determining how PCI affects microcirculatory function. Nevertheless, conducting an FFR assessment after PCI and during the follow-up period facilitated the evaluation of modifications in the coronary artery disease burden, which could potentially influence CFR and IMR measurements during the follow-up period.

Conclusions

In the acute phase of NSTEMI, reduced CFR was observed commonly in the MI culprit coronary artery, whereas severe microcirculatory dysfunction was infrequent.