Comprehensive assessment of systemic arteriosclerosis in relation to the ocular resistive index in acute coronary syndrome patients

This study aimed to investigate the relationship between ocular vascular resistance parameters, evaluated by laser speckle flowgraphy (LSFG), and systemic atherosclerosis, renal parameters and cardiac function in acute coronary syndrome (ACS) patients. We evaluated 53 ACS patients between April 2019 and September 2020. LSFG measured the mean blur rate (MBR) and ocular blowout time (BOT) and resistivity index (RI). 110 consequent patients without a history of coronary artery disease who visited ophthalmology as a control group. Significant positive correlations were observed between ocular RI and systemic parameters in ACS patients, including intima-media thickness (r = 0.34, P = 0.015), brachial-ankle pulse-wave velocity (r = 0.41, P = 0.002), cystatin C (r = 0.32, P = 0.020), and E/e’ (r = 0.34, P = 0.013). Ocular RI was significantly higher in the ACS group than in the control group in male in their 40 s (0.37 ± 0.02 vs. 0.29 ± 0.01, P < 0.001) and 50 s (0.36 ± 0.02 vs. 0.30 ± 0.01, P = 0.01). We found that the ocular RI was associated with systemic atherosclerosis, early renal dysfunction, and diastolic cardiac dysfunction in ACS patients, suggesting that it could be a useful non-invasive comprehensive arteriosclerotic marker.


Relationship between cardiac parameters and mean ocular RI and BOT.
Differences in systemic atherosclerotic, renal and cardiac parameters, and ocular parameters between ACS patients with and without retinopathy. ACS patients with (n = 32) and without (n = 21) retinopathy, diagnosed by the existence of soft exudate in the retina, were similar in systemic atherosclerosis, renal, and cardiac parameters. They were also similar in the ocular parameters (RI, 0.40 ± 0.10 vs. 0.41 ± 0.08, P = 0.71; BOT, 48.7 ± 5.7 vs. 49.2 ± 4.0, P = 0.77). Table 3 shows the characteristics of the control group subjects (n = 110). We compared the RI and BOT values between the ACS and control groups in each www.nature.com/scientificreports/ age category of 40-49 years, 50-59 years, 60-69 years, and ≥ 70 years, respectively, divided by sex (Fig. 5). Significant differences were found only among males, in the RI values in the 40-49 years and 50-59 years categories.

Discussion
This study had two main findings. First, the ocular RI obtained by LSFG in ACS patients was significantly associated with atherosclerosis and systemic arteriosclerosis. These trends were similar but weaker for ocular BOT. Second, the RI value in the ACS group was significantly higher in the male aged 40 s to 50 s group than in the same age male control group.
Clinical value of ocular RI and BOT measured by LSFG. Retinal blood flow is adequately regulated by ocular perfusion pressure and vascular resistance in response to various physiological stimuli 12 . Variations in vascular resistance are generally influenced by vascular endothelial function and smooth muscle elasticity. LSFG allows for quantitative estimation of blood flow in the ONH 13  www.nature.com/scientificreports/ question and presents several important findings. The ocular RI correlated positively with systemic (IMT and baPWV), renal (BUN, cystatin C, U-β2MG, and L-FABP), and cardiac (E/e') parameters, with BOT showing a much weaker association, suggesting that ocular RI may be more appropriate index to systemic parameters than BOT, at least in ACS patients. Carotid artery assessment by ultrasonography effectively detected the presence of carotid and other atherosclerotic diseases 14 . Increased mean IMT is a major cardiovascular risk factor 15 . baPWV, the most widely used measure of arterial stiffness, was shown to be a strong predictor of future cardiovascular events such as ACS 16 . Previous studies showed that the BOT on LSFG was associated with baPWV and IMT thickening 17 , but the relation between ocular RI and baPWV or IMT was not assessed before. Our results suggest that ocular RI, rather than BOT, could help identify the progression of systemic atherosclerosis in such high burden atherosclerotic patients as ACS patients.
Renal dysfunction is associated with an increased risk of cardiovascular morbidity and mortality. Renal diseases such as chronic kidney disease (CKD) reportedly progress with arteriosclerosis 18,19 . The cardiorenal syndrome is a disorder of the heart and kidney, in which one organ reportedly causes dysfunction of the other. The presence of renal disorder in ACS patients was reported to affect their prognosis 20 . Cystatin C, U-β2MG, and L-FABP are known to be better biomarkers for diagnosing acute kidney disease, which could transition to CKD, than creatinine and eGFR 21 . BOT obtained by LSFG in previous studies was correlated with renal function, and patients with CKD were found to experience an ocular circulatory disorder 22 .
Evaluation of diastolic function after ACS is important because it correlates with infarct size. Diastolic dysfunction has a high risk of death and is associated with poor prognosis independently of left ventricular systolic function 23,24 . The E/e' ratio is an echocardiographic index used to assess left ventricular diastolic function. It was associated with left ventricular diastolic pressure (and mean pulmonary artery wedge pressure) earlier in the diastole 25 . Previous studies have stated that optic disc BOT obtained by LSFG is significantly correlated with left ventricular diastolic function (E/e' ratio) in healthy persons 26,27 . Our data showed a similar finding, but the www.nature.com/scientificreports/ correlation coefficient was higher for RI than for BOT, suggesting that ocular RI assessment after ACS might be useful for determining the presence of diastolic dysfunction.

Role of transient retinopathy in the ACS patients.
We found that 21 (40%) ACS patients had mild and transient retinopathy with soft exudate. Kinoshita et al. 28 reported that cotton wool spots developed within two months in more than half of the patients with AMI undergoing PCI and then tended to become quiescent without treatment. We found no difference in systemic atherosclerosis, renal, or cardiac parameters between patients with and without soft exudate in our study. Although the patient characteristics in the two studies were different, the results are comparable; the occurrence of soft exudate was unrelated to any ocular parameters measured by LSFG. The soft exudate is known as the representative of ischemic disorders in the local site of the ocular artery. Therefore, our findings and those of this previous study suggest that it may not always directly reflect systemic vascular resistance or atherosclerosis, especially for ACS patients in the subacute phase.
Clinical implications. Our results indicated a clinical importance that the non-invasive evaluation of ocular blood flow may be a robust marker for promoting the comprehensive evaluations for not only retinal but also systemic arteriosclerosis in ACS patients. Also, the high RI value in middle age patients of 40-60 years can be a potential to predict the vulnerability to cardiovascular disease.

Study limitations
This study has several limitations. First, this study was conducted in a single institution; the sample size was small because we focused on ACS patients without hemodialysis or assisted circulation apparatus. Particularly, for the comparison of RI and BOT values between the ACS and control groups, the number of subjects was small in each age category. The subject number who had a significant difference in the RI among of the age categories of 40-59 years (ACS n = 17 vs. control n = 44) might have been statistically acceptable. Second, because most of our ACS patients were male, sex differences could have had some influence on our results, as reported previously 29 . We at least indicated a descriptive result of comparing the male and female RI and BOT values between the ACS and control groups (Fig. 5) to show the sex difference. Third, control group data were collected from the subjects who had visited the Division of Ophthalmology in our hospital with no ocular disorders in the examined eye, so the absence of cardiovascular disease was assessed by the subject clinical record, but not by CAG or coronary CT image. Detailed coronary information was not obtained because control subjects did not have a history of episodes of ACS. Finally, the dates of a comprehensive assessment by ophthalmic examination and other modalities might have affected our results. We excluded ACS patients with severe systemic conditions such as hemodialysis or assisted circulation apparatus to minimize this effect. In addition, all assessments were www.nature.com/scientificreports/ performed in ACS patients who were considered to have good physical condition within 1 week after a blanking period following ACS onset.

Conclusion
Ocular vascular assessment by LSFG in ACS patients revealed that ocular RI was strongly associated with systemic parameters, i.e., atherosclerosis and early renal and cardiac diastolic dysfunction whereas these associations were weak for BOT. Therefore, ocular RI might be a useful non-invasive and comprehensive arteriosclerotic marker in ACS patients.

Methods
We studied 58 patients admitted to the coronary care unit of Nihon University Itabashi Hospital for ACS between April 1 2019, and September 30 2020. All patients underwent PCI on admission day, followed by a visit to the Department of Ophthalmology (mean, 11 ± 5 days after ACS) once their general conditions were stable. The dates for comprehensive assessment by LSFG, ultrasonographic imaging of the carotid artery, brachial-ankle pulse-wave velocity, and transthoracic echocardiography were set within 1 week between the 4 modalities. Patients were excluded from this study if they had glaucoma, uveitis, optic neuropathy, or retinal or choroidal vascular disease, left ventricular (LV) ejection fraction < 30%, or underwent hemodialysis, assisted circulation apparatus or previous intraocular surgery or if they were unable to visit the Department of Ophthalmology on foot. After filtering, 53 patients met the study criteria. LSFG control data were collected from 110 consecutive healthy subjects without any ocular diseases in the examined eye and no history of cardiovascular disease, from whom informed consent was obtained for the evaluation of LSFG. Control subjects were chosen among normal subjects aged 40 years or older who visited the Division of Ophthalmology in our hospital. The institutional review board of Nihon University Itabashi Hospital approved this cross-sectional study, and all participants provided their informed consent for participating in the study. The study was conducted following the tenets of the Declaration of Helsinki. www.nature.com/scientificreports/ Measurement of the carotid intima-media thickness. One to two weeks after admission for ACS, we performed high-resolution ultrasonographic imaging of the carotid artery with an EUB-8500 device (Hitachi, Co. Ltd., Tokyo, Japan), using B-scan mode and a probe frequency of 7.5 MHz. Measurements were performed with the participants in the supine position, and their head slightly turned away from the sonographer. The procedures involved scanning the near and far walls of the carotid artery 1 cm proximally and distally to the carotid bulb in the longitudinal view, and the average of the maximum values on both sides 30 was used as the mean IMT for data analysis.
Measurements of brachial-ankle pulse-wave velocity. The baPWV was measured 1-2 weeks after admission for ACS, using a volume-plethysmographic device (baPWV/ABI; Nihon Colin Co., Tokyo, Japan) that simultaneously recorded heart sounds, electrocardiograms, and blood pressure at the left and right brachia and ankles. After the patients rested for a minimum of 5 min and while lying supine, pulse volume waveforms were recorded non-invasively from over the brachial and tibial arteries, and the two waveforms time delay (T) between the feet was measured. The distance (D) covered by the waves was estimated as the distance measured between the two recording sites. baPWV was calculated as follows: baPWV = D (cm)/T(s).   Echocardiographic parameters. Echocardiographic parameters were obtained during the stable phase (mean, 6 ± 4 days after admission for ACS). Echocardiography was performed with the patient in the supine position, using Vivid 7 or Vivid E9 cardiovascular ultrasonographic systems (GE Healthcare, Milwaukee, WI, USA), operated by experienced sonographers who were blinded to patient data. Echocardiographic measurements were performed following the American Society of Echocardiography guideline 31 . Briefly, LV diastolic diameter (LVDd), LV systolic diameter (LVDs), and left atrial diameter were measured on the parasternal longaxis view. LV ejection fraction was measured using the modified Simpson's method in apical 4-and 2-chamber views. The LV mass was calculated with the formula derived from the data of the American Society of Echocardiography 32 . The LV mass index was calculated as the ratio of the LV mass-to-body surface area. Transmitral flow velocity curves were recorded to measure the peak early (E) and late (A) diastolic velocities. Tissue Doppler imaging at the mitral annulus level was obtained in the septal position to measure the early (e') and late (A') diastolic myocardial velocities, as previously described 28 .
Ocular fundus examinations. All patients underwent a baseline ophthalmic evaluation by a welltrained ophthalmologist (HY) before the ocular blood flow measurement. All patients had good visual acuity (VA > 20/20) and normal intraocular pressure (IOP < 20 mmHg). IOP was monitored by applanation tonometry (Haag Streit, Bern, Switzerland).
Laser speckle flowgraphy measurements. After the pupils were dilated with a 0.5% tropicamide eye drop, a commercially available LSFG-NAVI system (Softcare Co., Ltd., Fukutsu, Japan) was used to measure ocular circulation at the ONH. The principles of LSFG were previously described in detail 12 . Briefly, the LSFG images were obtained from a 21° section centered on the optic disc. This observation field comprised 750 pixels (width) × 360 pixels (height). The mean blur rate (MBR) was calculated from the moving erythrocytes illuminated by an 830-nm wavelength diode laser beam. The MBRs were expressed in arbitrary units and were considered an indicator of the relative erythrocyte velocity. A total of 118 MBR images were recorded from the ONH area over 4 s. Using accompanying analysis software (LSFG Analyzer, Version 3.3.3.0; Softcare Co., Ltd., Fukutsu, Japan) 22 , a grayscale map of the still images was automatically created by averaging the MBR images (Fig. 2a). MBR can be determined for the entire ONH area (referred to as MA, "mean MBR of the entire area") or, separately, for vessels (MV, "mean MBR of vascular area") and tissue (MT, "mean MBR of tissue area"). The changing MBR pulse wave corresponding to each cardiac cycle was obtained on the analysis screen and displayed after being normalized to one pulse (Fig. 2b). The pulse wave analysis in the ONH circulation was performed on this screen, as was previously described 13 . Briefly, the RI was calculated as the ratio of the difference between the maximum and minimum MBR (labeled A) to the maximum MBR. The number of frames showing one-half the value of A was designated B, and the number of frames showing one cardiac cycle was labeled C. The following formula: BOT = 100 × (B)/(C) was used to analyze the pulse wave in the optic nerve head circulation 33 . We measured MBR, RI, and BOT of the ONH area twice in each eye, and the average values calculated by the LSFG software were used for statistical analysis. All pulse wave analyses in this study were based on the pulse waveform obtained from the MV (corresponding to the large retinal vessels within the ONH area). www.nature.com/scientificreports/ Statistical analyses. Continuous data are expressed as mean ± standard deviation if normally distributed or as median (interquartile range) if otherwise. Comparisons were performed using Student's t-test or Mann-Whitney U test. Categorical data are expressed as numbers and percentages and were compared using the chisquared test or Fisher's exact test. Linear regression analysis with Spearman rank-order correlation coefficients was used to assess the correlations between the variable, including systemic atherosclerosis, renal, and transthoracic echocardiography cardiac parameters with the LSFG parameters. Statistical significance was defined as a 2-tailed P-value < 0.05. Statistical analyses were performed using JMP Version 14.0 (SAS Institute, Cary, NC, USA).