Although numerous clinical markers and risk factors have been identified, novel clinical markers and approaches are required to assess patient risk stratification with HFrEF. Relying solely on retinal vasculature as a biomarker can complicate the evaluation of patients with HF. On the other hand, the ability to examine the choroidal layers with swept source or EDI-OCT offers new research opportunities for patients with HFrEF and the ischemic type of HFrEF, which is the leading cause of LV dysfunction in adults worldwide. 16–19
SFCT is commonly used in research trials. Still, it reflects only the whole choroidal structure and may be affected by various factors such as age, axial length, intraocular pressure, and systolic blood pressure. 20–22 It also fails to distinguish vascular from stromal components. 23 Therefore, since 2014, the focus of investigations has been switched to finding a method for choroidal structure analysis with a better distinction between the lumen and the stroma. 24 The term “CVI” was introduced as the ratio of the luminance area to the total choroidal area and reported to be around 65% in healthy eyes. 14 It has been proved that CVI is a more informative, reliable, and steady index of proportional change in the choroidal vasculature as it is relatively resistant to changes in other physiological parameters. 14,25
Similar to other organs, the choroid could be susceptible to any chronic cardiac processes. Extreme hypertensive retinopathy is associated with hypertensive choroidopathy and choroidal thickening. Hypercholesterolemia is linked to choroidal thickening, whereas smoking, ocular ischemia syndrome, and systemic hypertension are linked to a thinner choroid. Alternatively, data addressing the association between CT and carotid artery stenosis and diabetes mellitus are conflicting. 9, 26–30
Comparing 56 eyes of 56 patients with chronic HF with 56 eyes of 56 individuals of the same sex, Altinkaynak et al 31 discovered that the mean SFCT value was statistically considerably lower in patients with chronic HF, and it was positively correlated with EF. They hypothesized that a low cardiac output might lead to vasoconstriction development in peripheral vessels like orbital and choroidal arteries to maintain critical organs’ blood supply, causing choroidal ischemia and retinal pigment epithelium atrophy. Rakusiewicz and colleagues 18 discovered that children with congestive HF due to dilated cardiomyopathy had a thinner CT at all measured locations. They found that chronic HF impacted CT and that this parameter might be useful for monitoring the clinical progression of dilated cardiomyopathy in children. Nonetheless, another study by Alur et al 32 found neither a significant difference in SFCT between patients and controls nor a correlation between SFCT and EF. Although there is no clear explanation for the divergent results, they may be explained by the wide range of EF, dissimilar treatments received by each study’s patients, and the impact of diuretics and angiotensin-converting enzyme inhibitors on SFCT. 33
HFrEF is accompanied by peripheral vasoconstriction, which ensures appropriate perfusion and oxygenation to the heart, brain, and other vital tissues. 3,34 It is suggested that this compensatory vasoconstriction affects the choroidal arteries and, thus, lowers SFCT. 31 When choroidal vasoconstriction persists, the ensuing chronic ischemia may result in the atrophy of the retinal pigmented epithelium, which exacerbates SFCT reduction. 29,30,35,36 Consequently, as shown in the current study, HFrEF can be represented in choroidal thinning, and HF can be reflected in the thinning of the choroid.
Regardless of SFCT abnormalities in individuals with HFrEF, it is unclear which component of the choroid (the stroma or the lumen of the choroidal vessels) is most affected by HFrEF. To our knowledge, the present study is the first to evaluate CVI in the setting of HFrEF. We demonstrated that CVI was decreased in the HFrEF group, but the association was not statistically significant before and after age adjustment (P = 0.118 and P = 0.098, respectively), suggesting that diminishing CT in HFrEF can be associated with the shrinkage of both components of the choroid (ie, the stroma and the lumen).
Any defect or disruption in the choroid or its blood flow may result in degenerative changes and neovascularization. 22 In contrast to the retinal vasculature, which utilizes autoregulation to maintain relatively constant blood flow despite variations in ocular perfusion pressure, the regulation of blood flow in the choroid is complex. 37 It is now well established that the choroid is not a passive vascular bed in which a decrease in perfusion pressure leads to a linear decrease in blood flow. 6,38,39 The choroid has neural control and is supplied by sympathetic and parasympathetic nerve fibers, which may help the choroid to compensate for blood pressure alterations through vasoconstriction or dilation. 40,41 There is also a compensatory myogenic mechanism in the choroid: smaller resistance arteries can change smooth muscle contraction and tone to maintain vessel wall stretch in the desired range. 42 It seems that the choroid has a weak but significant ability to autoregulate its blood flow in response to changes in ocular perfusion pressure. 43
Previous studies have reported that patients with HF have reduced retinal vessel density compared with the normal population. 44 It has been hypothesized that the mentioned peripheral vasoconstriction in response to a low cardiac output is the pathophysiological mechanism that triggers this finding. 44–46 Therefore, retinal vessel density as measured by OCTA may provide an insight into the global microperfusion and hemodynamic state of patients with chronic HF. 47,48 The results of the current study are in accordance with previous studies on patients with HFrEF in terms of reduced retinal VD in comparison with the normal population.19,49 Since the blood supply to the retina is physiologically kept constant by specific autoregulation mechanisms, such a decrease in VD in patients with chronic HF can also be justified by insufficiency to autoregulate the retinal blood supply in these patients. 19
The main advantage of our study is its use of a noninvasive technology that enables the reproducible and quantitative evaluation of abnormalities in the retinal and choroidal microvasculature in a homogeneous group of patients with HFrEF. This study does come, however, with several limitations. Firstly, the relatively small sample size might affect the results. Secondly, the cross-sectional design of the current study precluded longitudinal change assessment in the retinal and choroidal vasculatures. Thirdly, the study was conducted in a tertiary center, which may have led to the recruitment of more chronic or complicated patients and, thus, selection bias. Fourthly, the duration of chronic HF was not specified, which might have affected the results. Finally, the influence of individually started HF treatment on the obtained results concerning retinal VD and the choroidal vasculature should not be ruled out.
We believe that our findings demonstrate a new multidisciplinary pathway in the diagnosis and management of HFrEF, which is a syndrome affecting literally every organ in the human body. Accordingly, we advocate retinal examination as a promising endpoint in future studies.
In conclusion, we demonstrated a reduction in retinal VD in both the superficial and deep capillary plexuses as well as SFCT in patients with HFrEF, although CVI was not changed significantly. Further long-term observations of larger numbers of patients will determine whether these biomarkers may be useful in clinical practice in patients with HFrEF.