May 2023
Volume 12, Issue 5
Open Access
Retina  |   May 2023
Correlation Between Coronary and Retinal Microangiopathy in Patients With STEMI
Author Affiliations & Notes
  • Anna-Maria Sideri
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Menelaos Kanakis
    School of Medicine, University of Patras, University Eye Clinic, Rion University Hospital, Patras, Greece
  • Andreas Katsimpris
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Aristotelis Karamaounas
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Dimitrios Brouzas
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Petros Petrou
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Evangelia Papakonstaninou
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Konstantinos Droutsas
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Stylianos Kandarakis
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Georgios Giannopoulos
    3rd Department of Cardiology, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece
  • Ilias Georgalas
    School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital of Athens, Athens, Greece
  • Correspondence: Ilias Georgalas, School of Medicine, National and Kapodistrian University of Athens, 1st University Eye Clinic, G. Gennimatas General Hospital, 154 Mesogeiion ave., Athens 115 27, Greece. e-mail: [email protected] 
Translational Vision Science & Technology May 2023, Vol.12, 8. doi:https://doi.org/10.1167/tvst.12.5.8
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Anna-Maria Sideri, Menelaos Kanakis, Andreas Katsimpris, Aristotelis Karamaounas, Dimitrios Brouzas, Petros Petrou, Evangelia Papakonstaninou, Konstantinos Droutsas, Stylianos Kandarakis, Georgios Giannopoulos, Ilias Georgalas; Correlation Between Coronary and Retinal Microangiopathy in Patients With STEMI. Trans. Vis. Sci. Tech. 2023;12(5):8. https://doi.org/10.1167/tvst.12.5.8.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: To investigate the morphological and functional correlation between microvascular retinal changes in optical coherence tomography angiography (OCTA) and the microvascular coronary circulation in patients with ST elevation myocardial infarction (STEMI) coronary heart disease (CHD).

Methods: A total of 330 eyes from 165 participants (88 cases and 77 controls) were enrolled and imaged. Superficial capillary plexus (SCP) and deep capillary plexus (DCP) vascular density was measured in the central (1 mm) and perifoveal (1–3 mm) areas and in the superficial foveal avascular zone (FAZ) and choriocapillaris (3 mm). These parameters were then correlated to the left ventricular ejection fraction (LVEF), and the number of affected coronary arteries.

Results: Decreased vessel densities in the SCP and DCP and choriocapillaris were positively correlated to the LVEF values (P = 0.006, P = 0.026, and P = 0.002, respectively). No statistically significant correlation between the SCP and DCP central area or FAZ area was found. Regarding the number of affected vessels, significant negative correlations were revealed for the SCP and DCP central vessel densities (P < 0.001 and P < 0.001, respectively) and the SCP perifoveal vascular density (P = 0.009).

Conclusions: OCTA vascular indices are significantly correlated with morphological and functional parameters in patients with STEMI CHD. SCP vascular density especially seems to be a promising biomarker for the extent of both macrovascular damage (number of affected coronary arteries) and microvascular damage, as mirrored in the decreased LVEF at admission.

Translational Relevance: OCTA vascular indices offer a valuable insight into the microvascular status of coronary circulation.

Introduction
Coronary heart disease (CHD) is a significant and growing cause of morbidity and mortality worldwide, despite the progress made regarding diagnosis and treatment. The disease can either present as a progressive and gradually debilitating form or emerge suddenly without prodromal symptoms as an acute coronary syndrome. ST segment elevation myocardial infarction (STEMI), or transmural infarct, represents about one third of the total cases. Known risk factors include hypertension, diabetes mellitus, hypercholesterolemia, and smoking.1 
CHD has both a macrovascular and a microvascular component that contribute to the disease pathophysiological changes.2 Despite the increasing interest in the role of microvascular damage in CHD, procedures available to cardiologists for the direct depiction of cardiac microcirculation are invasive, expensive, time consuming, operator dependent, and not readily available.3 The complex problem of risk stratification could probably be addressed indirectly through the study of retinal microcirculation. 
In the past, fundoscopy has provided evidence toward the correlation of retinal macrovascular changes and CHD.4,5 Today, the advent of optical coherence tomography angiography (OCTA) offers a detailed view of the retinal capillary circulation. Although a relationship between retinal capillary function and coronary microcirculation seems probable, current published evidence in the field is scarce. In this context, our study aimed to investigate the correlation between coronary and retinal microangiopathy as depicted in OCTA in patients with STEMI and subsequently the potential of OCTA vascular indices as CHD biomarkers. 
Methods
The study was designed as a prospective cross-sectional study and was conducted from October 2019 to September 2021 in the Cardiology Intensive Care Unit of G. Gennimatas General Hospital on patients with STEMI. The study was conducted in accordance with the tenets of the Declaration of Helsinki and local regulations. All participants signed an informed consent document. Our study was approved by our institutional review board and the local ethics committee (protocol no. 1819004358). 
Eligible hemodynamically stable patients were transferred under cardiologic supervision to the University Eye Clinic within the hospital. Patients were then subjected to OCTA to image the superficial capillary plexus (SCP) and deep capillary plexus (DCP), as well as choroidal circulation, and underwent ophthalmologic examinations (slit-lamp, including fundoscopy and refraction). Exclusion criteria were previous retinal vascular disease of any kind (e.g., degenerative macular disease, diabetic retinopathy or maculopathy, retinal vascular occlusion), history of ocular inflammation or trauma, intraocular hypertension or glaucoma, history of vitrectomy, presence of epiretinal membrane or macular hole, refractive error ≥ ±3 diopters. Eye conditions reported by the participants were confirmed during the ophthalmologic examination (e.g., previous ocular surgery, macular degeneration). Newly discovered conditions during the examination were documented and evaluated according to the exclusion criteria. 
A control group consisted of consecutive healthy individuals (i.e., without clinically manifest coronary artery disease and over the age of 30 years to match the age range of the STEMI cases) who were visiting the Ophthalmology Department outpatient clinic for reasons unrelated to the exclusion criteria (e.g., refraction). 
The presence of coronary artery stenosis and the number of affected branches were calculated according to the standard definition of angiographically significant heart disease of at least 50% severity in a major coronary artery or a major branch (at least 2 mm in diameter). 
All patients were subjected to revascularization procedures within the first 24 hours. OCTA was performed after revascularization using the DRI OCT Triton Swept Source Optical Coherence Tomograph (Topcon, Tokyo, Japan). The machine acquires 100,000 A-scans per second with transverse and axial resolutions of 20 µm and 7 µm, respectively, by utilizing a laser with a central wavelength of 1050 nm. Each OCTA B-scan is 3 × 3 mm, centered on the fovea, and consists of 320 clusters of four B-scans. All cases were examined within the first 2 days from admission and under tropicamide 0.5%–induced mydriasis. A 3 × 3-mm macular scan was obtained for both eyes of each patient. Only images without artifacts and with scan quality over 40 (OCTA scan quality 71.13 ± 14.9 vs. 74.35 ± 35 in controls and cases, respectively; P = 0.17) were included in the study. 
The vascular plexus segmentation and artifact reduction were generated by a native automated layer segmentation algorithm (IMAGEnet 6 version 1.14). The quality of the segmentation and the presence of projection, banding, or motion artifacts were also checked manually by two authors (E.K., A.S.); in cases of disagreement, a third author (I.G.) made the final decision. Vascular indices recorded included the foveal avascular zone (FAZ) size and the vascular densities of the SCP and DCP, as well as the choriocapillaris (CC) layer. The native OCTA software was utilized for the study measurements. For segmentation, the native 1-mm central sector was adopted, and the perimacular area was handled as a whole (circular ring ranging from 1–3 mm). OCTA 3 × 3-mm images of the SCPs and DCPs in controls and cases with different numbers of coronary vessels involved are presented in Figure 1
Figure 1.
 
(A–D) OCTA 3 × 3 mm images of the SCP and FAZ from a control participant (A), a patient with STEMI with one affected vessel (B), a patient with STEMI with two affected vessels (C), and a patient with STEMI with three affected vessels (D). (E–H) Corresponding OCTA images of the DCPs of the same patients: the control participant (E) and the patients with STEMI with one, two, or three affected vessels, respectively (F–H).
Figure 1.
 
(A–D) OCTA 3 × 3 mm images of the SCP and FAZ from a control participant (A), a patient with STEMI with one affected vessel (B), a patient with STEMI with two affected vessels (C), and a patient with STEMI with three affected vessels (D). (E–H) Corresponding OCTA images of the DCPs of the same patients: the control participant (E) and the patients with STEMI with one, two, or three affected vessels, respectively (F–H).
The rationale behind the option not to adopt the classic four Early Treatment Diabetic Retinopathy Study (EDTRS) sectors was that, although the effect on the capillary network should be by nature equal in the four quadrants, random variability could otherwise obscure valid results. Thus, averaging the vascular density in the four quadrants seemed prudent. Vascular density was measured as vascular length per unit of area (mm/mm2; i.e., mm−1). For the CC layer, the analysis was performed on the whole 3-mm area, as the CC layer has a homogeneous structure and uniform appearance in OCTA irrespective of the area under consideration. Furthermore, the ETDRS sectorial approach was created on the basis of the superficial retinal vasculature, as no accurate depiction of the DCP, SCP, and CC was available at the time. 
Also recorded were patient demographic data, medications (beta-blockers, calcium channel blockers, aspirin, statins, angiotensin-converting enzyme [ACE] inhibitors/angiotensin receptor blockers [ARBs], or anticoagulants), ancillary tests (hematocrit, high-density lipoprotein [HDL]/low-density lipoprotein [LDL], blood glucose, and creatinine), and cardiovascular parameters such as type of STEMI, occluded artery localization, and left ventricular ejection fraction (LVEF). Excel (Microsoft, Redmond, WA) was utilized to collect the data, and statistical calculations were performed using R 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria). Power analysis and sample size calculations (prior to the recruitment of the participants) were conducted using an independent-sample t-test aimed at detecting a difference of 2 mm−1 in vessel density between the two groups. Assuming a standard deviation of 5% in vessel density (based on previous studies), at least 63 patients in each study group would be required to detect a difference in OCTA metrics of 2% with 0.80 power. 
Descriptive statistics of the study population were reported using percentage values for categorical variables and mean ± standard deviation for continuous variables. Because OCTA parameters from both eyes of participants were assessed, we used linear mixed-model analysis with a random intercept on the individual level to calculate the relationship between OCTA metrics and myocardial infarction. By using mixed-model analysis, we adjusted for the relationship between the OCTA metrics of different eyes from the same individuals. P values less than 0.05 were considered statistically significant. 
Results
A total of 330 eyes from 165 participants (88 cases and 77 controls) were enrolled and imaged. The demographic and clinical characteristics of both patients and controls are presented in Table 1. The two groups were similar in age (P = 0.72), but there were more females in the control group (P = 0.02). There were also significant differences between the STEMI and control groups regarding smoking habits (P < 0.01), dyslipidemia (P < 0.01), treatment with statins (P < 0.01), and incidence of diabetes mellitus (P < 0.01), as well as the presence of a family history of CHD (P < 0.01). Among the STEMI cases, a significant portion were under antihypertensive medication. More specifically, 61 received beta-blockers (69%), 11 received calcium channel blockers (12%), and 21 received ACE inhibitors/ARBs (31%). Also among the STEMI group, 11 were on anticoagulant therapy (12%) and 56 were receiving aspirin (64%). Ancillary tests at admission for the STEMI cases are also presented in Table 1
Table 1.
 
Demographic and Clinical Characteristics
Table 1.
 
Demographic and Clinical Characteristics
Clinical characteristics of the patients with STEMI are presented in Table 2. The mean LVEF was 46.02% ± 8.89%. As for the localization of the STEMI, 40 were anterior (45%), 44 were inferior (50%), and four were lateral (5%). The number of patients per number of affected coronary vessels is also presented in Table 2
Table 2.
 
Clinical Characteristics of STEMI Cases at Admission
Table 2.
 
Clinical Characteristics of STEMI Cases at Admission
The majority of patients (46 patients, 53%) had one-vessel disease, with the left anterior descending (LAD) artery being affected in 21 of them and the right coronary artery (RCA) in another 21. Four patients had significant coronary disease in a major branch (diagonal/acute marginal/ramus intermedius). Twenty patients had two-vessel disease; in 15 of them, the LAD artery and RCA were affected. Finally, 22 patients had angiographic involvement of all three major coronary arteries (25%). 
The association estimates of the OCTA vascular density measurements with LVEF at hospital admission in patients with STEMI are presented in Table 3. Regarding the perifoveal area of SCP, the vessel density was 47.12 mm–1 (range, 41.36–54.08), and there was a statistically significant (P = 0.006) correlation with the LVEF (β coefficient = 0.06; 95% confidence interval [CI], 0.02–0.10). The same was true for the perifoveal area of the DCP, for which the density was 57.40 mm–1 (range, 48.17–66.19), and there was a statistically significant (P = 0.026) correlation with the LVEF (β coefficient = 0.10; 95% CI, 0.01–0.18). Analysis of the CC layer revealed a vascular density of 68.54 mm–1 (range, 46.65–79.43), and there was a statistically significant (P = 0.002) correlation with the LVEF (β coefficient = 0.28; 95% CI, 0.11–0.45). On the other hand, no statistically significant correlations were detected with regard to the central area of the SCP (P = 0.958), the central area of the DCP (P = 0.798), or FAZ size (P = 0.740). Scatterplots of these correlations are presented in Figure 2. The quantitative data for the correlation of retinal capillary plexuses vessel density as assessed with OCTA and the number of involved coronary arteries are presented in Table 4. This table also includes the vascular density data for the control group indicating no involved coronary arteries. 
Table 3.
 
Association Estimates of OCTA Parameters With LVEF at Admissiona
Table 3.
 
Association Estimates of OCTA Parameters With LVEF at Admissiona
Figure 2.
 
Scatterplots of LVEF (%) versus OCTA vascular indices. (A) LVEF versus central SCP vessel density (mm−1), (B) LVEF versus perifoveal SCP vessel density (mm−1), (C) LVEF versus central DCP vessel density (mm−1), (D) LVEF versus perifoveal DCP vessel density (mm−1), (E) LVEF versus CC vessel density, and (F) LVEF versus FAZ size (µm2).
Figure 2.
 
Scatterplots of LVEF (%) versus OCTA vascular indices. (A) LVEF versus central SCP vessel density (mm−1), (B) LVEF versus perifoveal SCP vessel density (mm−1), (C) LVEF versus central DCP vessel density (mm−1), (D) LVEF versus perifoveal DCP vessel density (mm−1), (E) LVEF versus CC vessel density, and (F) LVEF versus FAZ size (µm2).
Table 4.
 
OCTA Parameters by Number of Affected Vesselsa
Table 4.
 
OCTA Parameters by Number of Affected Vesselsa
Regarding the SCP, in both the perifoveal and the central area a significant correlation between vascular density and the number of the involved arteries was discovered (P = 0.009 and P < 0.001, respectively). More specifically, for the perifoveal area, median vessel densities for the controls and patients with one, two, or three affected vessels were 47.66 mm–1 (range, 40.87–54.08), 47.75 mm–1 (range, 41.88–54.08), 46.88 mm–1 (range, 42.67–52.10), and 46.21 mm–1 (range, 41.36–48.92), respectively (β coefficient = −0.40; 95% CI, −0.69 to −0.11). As for the DCP, there was a statistically significant correlation only for the central area (P < 0.001), where the vessel densities for the controls and patients with one, two, or three affected vessels were 19.91 mm–1 (range, 8.25–31.93), 19.38 mm–1 (range, 12.03–25.08), 17.78 mm–1 (range, 8.14–26.37), and 16.61 mm–1 (range, 12.43–24.87), respectively (β coefficient = −1.05; 95% CI, −1.61 to −0.49). 
In the DCP perifoveal area, no statistically significant correlation was discovered (P = 0.129). The same was true for the CC layer (P = 0.064). The FAZ area was significantly increased with respect to the number of affected coronary arteries (P < 0.001). The median vessel densities for the controls and patients with one, two, or three affected vessels were 169.76 mm–1 (range, 51.00–473.62), 287.16 mm–1 (range, 51.44–438.54), 281.37 mm–1 (range, 63.11–431.19), and 280.97 mm–1 (range, 212.19–417.31), respectively (β coefficient = 38.90; 95% CI, 26.57–51.24). 
Discussion
Heart microvascular abnormalities are now considered a significant component of CHD. The term coronary microvascular disease3 has been coined to describe these alterations comprised of a combination of structural and functional abnormalities that can elicit symptoms such as angina and dyspnea even in the absence of a manifest obstructive plaque. Non-obstructive coronary atherosclerosis and microvascular ischemia come with an increased risk of major adverse cardiovascular events. Obviously, coronary microvascular disease is practically always present in patients with obstructive CHD, as endothelial and coronary vasomotor dysfunction are both present through the early stages of atherosclerosis long before the emergence of large epicardial vessels obstruction. 
Today, cardiology faces the challenge to look deeper into heart microcirculation, which is not an easy task, as existing methods come with certain disadvantages, such as their being invasive, operator dependent, technically challenging, not readily available, or costly. Noninvasive methods include positron emission tomography, cardiac magnetic resonance, and Doppler coronary flow reserve velocity; invasive methods include invasive coronary flow reserve, index of microvascular resistance, and wave intensity analysis.3 All of these represent indirect approaches that conclude the presence of CMD by evaluating functional and morphologic parameters affected by the microvascular dysfunction. 
A strong connection between the microcirculations of various organs can be assumed in the face of common risk factors, such as diabetes, hyperlipidemia, hypertension, or smoking, and is supported by a robust body of literature. Accordingly, such a proposed connection also applies to the early stages of vascular disease, as in cases of diabetes mellitus without diabetic retinopathy and the early stages of arterial hypertension. 
Ophthalmological examinations have long been a handy tool for cardiologists to assess the effect of hypertension on retinal vessels, but with modern imaging techniques this collaboration has the potential to be extended to new levels. Alan et al.6 presented evidence for the ability of OCTA to predict the emergence of iodinated contrast medium–induced acute kidney injury in patients with acute coronary syndrome subjected to coronary angiography (area under ROC curve = 0.745; 95% CI, 0.649 –0.841; P < 0.001). These authors suggested that the kidney, retina, and heart share a similar microvascular background. This is further supported by the work of Yeung et al.7 in which individuals with chronic kidney disease had significantly reduced vessel densities in the SCP and DCP. Similarly, the correlation between diabetic retinopathy and renal diabetic insult is well established.8,9 In patients treated for systemic hypertension, rarefaction of retinal capillary plexuses is associated with higher blood pressure and lower glomerular filtration rate.10 The effect of blood pressure on the CC is not clear, as flow voids appear to be increased11,12 in some studies and decreased in others.13 
Microvascular damage is also regarded as the underlying cause of congestive heart failure (CHF). Retinopathy in color fundus images has been shown by Wong et al.5 to be an independent predictor of CHF in a large cohort. Rakusiewicz et al.14 observed a significant decrease in SCP vascular density in children with CHF in the course of dilated cardiomyopathy. Nägele et al.15 also reported impaired retinal microvascular dilatation as a response to flicker light in a prospective observational study of 74 CHF patients (mean LVEF, 37% ± 12.8%). 
In this pilot study, we investigated the hypothesis that microvascular retinal changes can reflect the microvascular status of coronary circulation by searching for correlations of retinal microvascular indices (namely, capillary vessel density and FAZ size) to morphologic (number of affected coronary vessels) and LVEF parameters in patients with STEMI. Notably, OCTA vascular indices are remarkably repeatable and stable even in the intensive care unit setting, as demonstrated by Courtie et al.16 In particular, decreased SCP vessel density has been associated with higher concentrations of osteoprotegerin and angiopoietin-2 in patients hospitalized for acute coronary syndrome versus controls.17 
It is widely appreciated that the severity of left ventricular dysfunction as expressed by the reduction in LVEF is a significant short- and long-term prognostic factor in patients with STEMI.18,19 In our study, we detected a significant correlation between retinal capillary circulation parameters and LVEF. This correlation was detected in the perifoveal area of both the SCP and DCP and the CC layer. In contrast, no correlation was detected between LVEF and the central SCP and DCP area, or FAZ size. This may imply a more profound effect of the factors associated with reduced LVEF on the more vascularized perifoveal area, given that a significant portion of the central area is covered by a central avascular zone, which is devoid of vessels; thus, any differences in vascular density are inevitably reduced. 
Retinal blood flow is efficiently autoregulated, by a mechanism mainly influenced by local factors,20 allowing for vascular resistance adaptation to perfusion pressure changes, thus maintaining a constant blood flow.21 Similarly, an autoregularory component seems to be present in choroidal circulation but not to the same extent as in the retina.2224 In the EYE-MI pilot study,25 LVEF at admission was inversely associated with inner vessel density (i.e., SCP vessel density in all four quadrants). Low cardiac output has also been associated with decreased subfoveal choroidal thickness,26 lower diastolic velocities, and higher resistance indexes in ophthalmic artery Doppler ultrasound.27 
In our group of 88 patients with STEMI, the LVEF mean value was 46.02% ± 8.89%. Thus, we have reason to believe that, in our cases, the probability of autoregulation exhaustion was unlikely, because according to the inclusion criteria all of them were hemodynamically stable. This point of view is also supported by the findings of Arnould et al.28 in a small subgroup of 13 STEMI cases with severely altered LVEF at admission (42% ± 8%). The authors did not find a correlation between acute or chronic hemodynamic variables and SCP parameters at baseline or at the 3-month follow-up, when LVEF improved to 50% ± 9%, suggesting a wide range of effectiveness for retinal blood flow autoregulation. 
Our results also suggest a significant correlation between the number of affected coronary vessels and the retinal capillary network (control cases were assigned 0 affected vessels). The correlation was statistically significant for the central SCP and DCP areas and the FAZ. The difference in the vascular density for the perifoveal area was also statistically significant with regard to the SCP but not the DCP, suggesting a tendency for the central area (especially the SCP) and the FAZ size to be more closely correlated to the extent of the coronary vessel number involvement (in the LVEF, the same tendency was found toward the perifoveal area). 
The aforementioned results are partially in accordance with those of Wang et al.,29 who demonstrated a general trend of significantly decreased vessel density and flow area in the SCP but not DCP in a group of 158 patients with CHD, although in our study we included patients with STEMI patients who had more severe CHD. It seems, then, that the SCP is affected prior to and to a further extent than the DCP. Similarly, the increased DCP vascular density described by the aforementioned study may represent the stage of loss of capillary regulatory mechanisms that precedes capillary rarefaction, as described below. In accordance with our findings, in a recent study Chua et al.30 discovered a significant correlation between SCP vessel density and adverse cardiac remodeling markers in hypertensive patients who underwent cardiovascular magnetic resonance imaging. 
Although our study is a pilot study, we will attempt to offer a pathophysiologic explanation that could be of clinical significance, given that the number of involved vessels is a structural variable by nature, whereas LVEF is a functionality variable. It is therefore possible that the central area defects can better mirror the large coronary vessel defects, as central area defects appear later in the course of microvascular disease. In diabetic patients, the appearance of defects in vascular density precedes disruption of the FAZ. This has been reported in several studies regarding young patients with type 1 diabetes31 that have found that a phase of capillary dilatation precedes the capillary dropout around the FAZ.3234 Similar findings have been reported regarding DCP capillary diameter.35,36 Reduced LVEF probably is more strongly connected to the extent of microvascular damage and reperfusion injury. 
It seems that low perifoveal SCP vascular density is associated with coronary vascular disease, and these findings are consistent with previous reports.25 Such observations point to the assumption that retinal microvascular defects are probably related to the cardiovascular risk profile in our patients. The process may evolve through an initial stage of epicardial- and non-endothelium–dependent microvascular lesions, gradually exhausting the coronary reserve. Subsequently, ischemia-related vascular dysfunction, reperfusion injury, and distal emboli can lead to a major coronary event.25,37 Our findings regarding the potentially significant predictive value of OCTA indices in CHD are further supported by the study by Arnould et al.,38 who used a machine-learning Bayesian classifier on OCTA retinal microcirculation data and reported that they achieved good results in predicting cardiovascular risk scores. 
We acknowledge that there are several limitations to our study. As we included only hemodynamically stable patients with STEMI, the study is subject to selection bias. Only the native DRI OCT Triton software was utilized for the removal of artifacts. No axial length measurement of the eyes was performed (which could increase the accuracy of our measurements through OCTA-scale correction), because, for safety reasons, we tried to minimize the total exam time. Another parameter that was not evaluated was the presence of carotid artery stenosis, which has a documented association with severe CHD39 and could also interfere with OCTA vascular density and FAZ measurements. Stratification of the participants according to their antihypertensive treatment was not performed. Our association estimates regarding LVEF and OCTA metrics cannot be extrapolated to the general population, as LVEF measurements in the control group were not performed due to resource constraints. Finally, the control group included statistically fewer men than the patient groups, and the study and control populations were comprised only of Caucasians, thus limiting the possibility of extrapolating our results to other populations. 
In conclusion, the ability to apply a fast, safe, accurate, and repeatable procedure such as OCTA vascular indices to the biomarkers of CHD in STEMI may represent a leap forward in our understanding of the role of microvascular coronary dysfunction in the acute coronary syndrome, resulting in more accurate risk assessments and prognoses. The findings of the present study, which should be considered as preliminary proof-of-concept evidence, correlate markers of more advanced heart disease (including severity of angiographic coronary artery disease and LVEF) with retinal microangiopathy, suggesting that ocular microangiopathy reflects heart microangiopathy; thus, OCTA-defined retinal microangiopathy may reflect heart microangiopathy. The strength of this relationship should, of course, be evaluated by further studies, which could answer the question of whether an abnormal retinal OCTA should trigger a thorough cardiovascular evaluation in asymptomatic individuals with no known heart disease. 
Acknowledgments
Disclosure: A.-M. Sideri, None; M. Kanakis, None; A. Katsimpris, None; A. Karamaounas, None; D. Brouzas, None; P. Petrou, None; E. Papakonstaninou, None; K. Droutsas, None; S. Kandarakis, None; G. Giannopoulos, None; I. Georgalas, None 
References
Virani SS, Alonso A, Aparicio HJ, et al. Heart disease and stroke statistics—2021 update: a report from the American Heart Association. Circulation. 2021; 143: e254–e743. [CrossRef] [PubMed]
Camici PG, Crea F. Coronary microvascular dysfunction. N Engl J Med. 2007; 356: 830–840. [CrossRef] [PubMed]
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. [CrossRef] [PubMed]
Hsu CY, Lee CM, Chou KY, et al. The association of diabetic retinopathy and cardiovascular disease: a 13-year nationwide population-based cohort study. Int J Environ Res Public Health. 2021; 18: 8106. [CrossRef] [PubMed]
Wong TY, Rosamond W, Chang PP, et al. Retinopathy and risk of congestive heart failure. JAMA. 2005; 293: 63–69. [CrossRef] [PubMed]
Alan G, Guenancia C, Arnould L, et al. Retinal vascular density as a novel biomarker of acute renal injury after acute coronary syndrome. Sci Rep. 2019; 9: 8060. [CrossRef] [PubMed]
Yeung L, Wu IW, Sun CC, et al. Early retinal microvascular abnormalities in patients with chronic kidney disease. Microcirculation. 2019; 26: e12555. [CrossRef] [PubMed]
Lin H-T, Zheng C-M, Wu Y-C, et al. Diabetic retinopathy as a risk factor for chronic kidney disease progression: a multicenter case-control study in Taiwan. Nutrients. 2019; 11: 509. [CrossRef] [PubMed]
Park HC, Lee Y-K, Cho A, et al. Diabetic retinopathy is a prognostic factor for progression of chronic kidney disease in the patients with type 2 diabetes mellitus. PLoS One. 2019; 14: e0220506. [CrossRef] [PubMed]
Chua J, Chin CWL, Hong J, et al. Impact of hypertension on retinal capillary microvasculature using optical coherence tomographic angiography. J Hypertens. 2019; 37: 572–580. [CrossRef] [PubMed]
Chua J, Chin CWL, Tan B, et al. Impact of systemic vascular risk factors on the choriocapillaris using optical coherence tomography angiography in patients with systemic hypertension. Sci Rep. 2019; 9: 5819. [CrossRef] [PubMed]
Polak K, Polska E, Luksch A, et al. Choroidal blood flow and arterial blood pressure. Eye (Lond). 2003; 17: 84–88. [CrossRef] [PubMed]
Spaide RF. Choriocapillaris flow features follow a power law distribution: implications for characterization and mechanisms of disease progression. Am J Ophthalmol. 2016; 170: 58–67. [CrossRef] [PubMed]
Rakusiewicz K, Kanigowska K, Hautz W, Ziółkowska L. The impact of chronic heart failure on retinal vessel density assessed by optical coherence tomography angiography in children with dilated cardiomyopathy. J Clin Med. 2021; 10: 2659. [CrossRef] [PubMed]
Nägele MP, Barthelmes J, Ludovici V, et al. Retinal microvascular dysfunction in heart failure. Eur Heart J. 2018; 39: 47–56. [CrossRef] [PubMed]
Courtie EF, Kale AU, Hui BTK, et al. Stability of OCT and OCTA in the intensive therapy unit setting. Diagnostics (Basel). 2021; 11: 1516. [CrossRef] [PubMed]
Hannappe M-A, Arnould L, Méloux A, et al. Vascular density with optical coherence tomography angiography and systemic biomarkers in low and high cardiovascular risk patients. Sci Rep. 2020; 10: 16718. [CrossRef] [PubMed]
Perelshtein Brezinov O, Klempfner R, Zekry SB, Goldenberg I, Kuperstein R. Prognostic value of ejection fraction in patients admitted with acute coronary syndrome: a real world study. Medicine (Baltimore). 2017; 96: e6226. [CrossRef] [PubMed]
Stolfo D, Cinquetti M, Merlo M, et al. ST-elevation myocardial infarction with reduced left ventricular ejection fraction: insights into persisting left ventricular dysfunction. A pPCI-registry analysis. Int J Cardiol. 2016; 215: 340–345. [CrossRef] [PubMed]
Delaey C, Van De Voorde J. Regulatory mechanisms in the retinal and choroidal circulation. Ophthalmic Res. 2000; 32: 249–256. [CrossRef] [PubMed]
Wu LT, Wang JL, Wang YL. Ophthalmic artery morphological and hemodynamic features in acute coronary syndrome. Invest Ophthalmol Vis Sci. 2021; 62: 7. [CrossRef]
Luksch A, Polska E, Imhof A, et al. Role of NO in choroidal blood flow regulation during isometric exercise in healthy humans. Invest Ophthalmol Vis Sci. 2003; 44: 734–739. [CrossRef] [PubMed]
Schmidl D, Boltz A, Kaya S, et al. Comparison of choroidal and optic nerve head blood flow regulation during changes in ocular perfusion pressure. Invest Ophthalmol Vis Sci. 2012; 53: 4337–4346. [CrossRef] [PubMed]
Schmidl D, Schmetterer L, Witkowska KJ, et al. Factors associated with choroidal blood flow regulation in healthy young subjects. Invest Ophthalmol Vis Sci. 2016; 57: 5705–5713. [CrossRef] [PubMed]
Arnould L, Guenancia C, Azemar A, et al. The EYE-MI pilot study: a prospective acute coronary syndrome cohort evaluated with retinal optical coherence tomography angiography. Invest Ophthalmol Vis Sci. 2018; 59: 4299–4306. [CrossRef] [PubMed]
Altinkaynak H, Kara N, Sayın N, Güneş H, Avşar Ş, Yazıcı AT. Subfoveal choroidal thickness in patients with chronic heart failure analyzed by spectral-domain optical coherence tomography. Curr Eye Res. 2014; 39: 1123–1128. [CrossRef] [PubMed]
Almeida-Freitas DB, Meira-Freitas D, Melo LA, Jr, Paranhos A, Jr, Iared W, Ajzen S. Color Doppler imaging of the ophthalmic artery in patients with chronic heart failure. Arq Bras Oftalmol. 2011; 74: 326–329. [CrossRef] [PubMed]
Arnould L, Guenancia C, Gabrielle PH, et al. Influence of cardiac hemodynamic variables on retinal vessel density measurement on optical coherence tomography angiography in patients with myocardial infarction. J Fr Ophtalmol. 2020; 43: 216–221. [CrossRef] [PubMed]
Wang J, Jiang J, Zhang Y, Qian YW, Zhang JF, Wang ZL. Retinal and choroidal vascular changes in coronary heart disease: an optical coherence tomography angiography study. Biomed Opt Express. 2019; 10: 1532–1544. [CrossRef] [PubMed]
Chua J, Le TT, Sim YC, et al. Relationship of quantitative retinal capillary network and myocardial remodeling in systemic hypertension. J Am Heart Assoc. 2022; 11: e024226. [CrossRef] [PubMed]
Sousa DC, Leal I, Moreira S, et al. Optical coherence tomography angiography study of the retinal vascular plexuses in type 1 diabetes without retinopathy. Eye (Lond). 2020; 34: 307–311. [CrossRef] [PubMed]
Chen YJ, Khouri AS, Zarbin MA, Szirth BC. Early retinal microvascular abnormalities in young adults with type 1 diabetes mellitus without clinically evident diabetic retinopathy. Retina. 2021; 41: 1478–1486. [CrossRef] [PubMed]
Rosen RB, Andrade Romo JS, Krawitz BD, et al. Earliest evidence of preclinical diabetic retinopathy revealed using optical coherence tomography angiography perfused capillary density. Am J Ophthalmol. 2019; 203: 103–115. [CrossRef] [PubMed]
Li T, Jia Y, Wang S, et al. Retinal microvascular abnormalities in children with type 1 diabetes mellitus without visual impairment or diabetic retinopathy. Invest Ophthalmol Vis Sci. 2019; 60: 990–998. [CrossRef] [PubMed]
Mastropasqua R, Toto L, Mastropasqua A, et al. Foveal avascular zone area and parafoveal vessel density measurements in different stages of diabetic retinopathy by optical coherence tomography angiography. Int J Ophthalmol. 2017; 10: 1545–1551. [PubMed]
Zhu TP, Li EH, Li JY, et al. Comparison of projection-resolved optical coherence tomography angiography-based metrics for the early detection of retinal microvascular impairments in diabetes mellitus. Retina. 2020; 40: 1783–1792. [CrossRef] [PubMed]
Niccoli G, Scalone G, Lerman A, Crea F. Coronary microvascular obstruction in acute myocardial infarction. Eur Heart J. 2016; 37: 1024–1033. [CrossRef] [PubMed]
Arnould L, Guenancia C, Bourredjem A, et al. Prediction of cardiovascular parameters with supervised machine learning from Singapore “I” vessel assessment and OCT-angiography: a pilot study. Transl Vis Sci Technol. 2021; 10: 20. [CrossRef] [PubMed]
Kallikazaros I, Tsioufis C, Sideris S, Stefanadis C, Toutouzas P. Carotid artery disease as a marker for the presence of severe coronary artery disease in patients evaluated for chest pain. Stroke. 1999; 30: 1002–1007. [CrossRef] [PubMed]
Figure 1.
 
(A–D) OCTA 3 × 3 mm images of the SCP and FAZ from a control participant (A), a patient with STEMI with one affected vessel (B), a patient with STEMI with two affected vessels (C), and a patient with STEMI with three affected vessels (D). (E–H) Corresponding OCTA images of the DCPs of the same patients: the control participant (E) and the patients with STEMI with one, two, or three affected vessels, respectively (F–H).
Figure 1.
 
(A–D) OCTA 3 × 3 mm images of the SCP and FAZ from a control participant (A), a patient with STEMI with one affected vessel (B), a patient with STEMI with two affected vessels (C), and a patient with STEMI with three affected vessels (D). (E–H) Corresponding OCTA images of the DCPs of the same patients: the control participant (E) and the patients with STEMI with one, two, or three affected vessels, respectively (F–H).
Figure 2.
 
Scatterplots of LVEF (%) versus OCTA vascular indices. (A) LVEF versus central SCP vessel density (mm−1), (B) LVEF versus perifoveal SCP vessel density (mm−1), (C) LVEF versus central DCP vessel density (mm−1), (D) LVEF versus perifoveal DCP vessel density (mm−1), (E) LVEF versus CC vessel density, and (F) LVEF versus FAZ size (µm2).
Figure 2.
 
Scatterplots of LVEF (%) versus OCTA vascular indices. (A) LVEF versus central SCP vessel density (mm−1), (B) LVEF versus perifoveal SCP vessel density (mm−1), (C) LVEF versus central DCP vessel density (mm−1), (D) LVEF versus perifoveal DCP vessel density (mm−1), (E) LVEF versus CC vessel density, and (F) LVEF versus FAZ size (µm2).
Table 1.
 
Demographic and Clinical Characteristics
Table 1.
 
Demographic and Clinical Characteristics
Table 2.
 
Clinical Characteristics of STEMI Cases at Admission
Table 2.
 
Clinical Characteristics of STEMI Cases at Admission
Table 3.
 
Association Estimates of OCTA Parameters With LVEF at Admissiona
Table 3.
 
Association Estimates of OCTA Parameters With LVEF at Admissiona
Table 4.
 
OCTA Parameters by Number of Affected Vesselsa
Table 4.
 
OCTA Parameters by Number of Affected Vesselsa
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×