Is examination of retinal vessels still useful in hypertensive patients?

Introduction

The microcirculation plays a key role in the pathophysiology of hypertension [1, 2]. It is the primary site of increased vascular resistance, which is the major haemodynamic hallmark of hypertension. During early stages in the development of hypertension, small arteries (diameter 150 to 300 micrometres) and arterioles (diameter 10 to 150 micrometres) usually constrict due to various nervous or endocrine and autocrine mechanisms. In the long run, these functional forms of vascular narrowing become more structural [2, 3]. This later stage is characterised by inward hypertrophy of small arteries and arterioles as well as gradual rarefaction of both arterioles and capillaries [2, 3]. These structural alterations not only raise blood pressure to permanently hypertensive levels, but also cause serious perfusion problems in various organs, such as the brain, kidneys and heart, with a subsequent increased mortality and morbidity risk [2, 3]. The microcirculation controls the delivery of oxygen and nutrients to nearly every cell in the body. In major diseases, such as hypertension, diabetes, cardiac ischaemia and dementia, this delivery is hampered by the deterioration of microvascular function [2, 4].

The treatment of hypertension is aimed primarily at lowering blood pressure, although the real goal should actually also include the reduction of target organ damage. This latter goal is not always reached because of technical and financial limitations to obtaining information on the degree of target organ damage. The assessment of target organ damage is thus far mostly centred around an overall measurement of renal and cardiac function or biochemical parameters that give an indication of this function. Direct assessment of microvascular function and structure is still largely lacking. In this short review I shall argue that the retina offers an easily accessible tissue to follow microvascular changes and potential damage during the onset and treatment of hypertension.

Assessment of microvascular structure and function

The most direct way of assessing microvascular structure and function is direct imaging of these small vessels. Intravital microscopy has been used by many investigators in animal models of human disease. This work has revealed important pathophysiological events leading to these diseases [5, 6]. However, such methods have only limited possibilities for use in humans and can only be applied in tissues without or with only limited skin protection, such as the finger nailfold, tongue or eye. Actually, ophthalmologic inspection, with a focus on the retina, has long been a standard procedure in the diagnostic work-up of hypertensive or diabetic patients. A semi-quantitative grading system was used to classify the degree of microvascular damage in this ophthalmological inspection method. In the 1960s, fluorescent angiography was introduced to image retinal vessels. Although this approach is still used in ophthalmology as a diagnostic test of patency and leak of retinal vessels, it is not suited for a more generalised study of cardiovascular target organ damage because of its invasive approach.

A more elaborate, but also more quantitative method to assess human microcirculation has been nailfold capillaroscopy, in which skin capillaries are observed using a microscope with an epi-illumination system [5, 6]. Abnormal microvascular patterns have been observed and quantified using this method in diseases affecting the skin microcirculation, such as systemic sclerosis, diabetes and hypertension. New techniques for video microscopic examinations of more deeply located tissue have been introduced in recent decades [7, 8]. These techniques are based on the use of orthogonal polarisation spectral or sidestream dark field imaging. Such devices use the principle that green light illuminates the depth of a tissue and that the scattered light is absorbed by haemoglobin of red blood cells contained in the vasculature. Hand-held cameras allow microcirculatory observations to be made during surgical procedures, but also in the non-hospital circumstances of epidemiological studies.

In addition to the (video)microscopy techniques to investigate the structure and function of the microcirculation, there has been growing interest in advanced perfusion imaging technologies, such as laser Doppler flowmetry, positron emission tomography (PET), magnetic resonance imaging (MRI) and angiography [6, 9]. Laser Doppler is particularly suited to study microvascular reactivity and has been important in studies on the role of the endothelium as well as in studying mechanisms underlying flow-mediated changes in (micro)vascular reactivity. Specific retinal laser Doppler flowmetry was developed in Heidelberg in the 1990s [9]. This technology is based on a combination of confocal scanning laser tomography and laser Doppler flowmetry. It has been used successfully by Schmieder and co-workers in hypertension research in the past 15 years [9]. PET has turned out to be a useful tool in cardiac pathophysiology studies in combination with molecular studies on cardiac metabolism. MRI has become a major technology in the study of vascular structure and function in several target organs; however, its resolution is still not high enough to assess the smallest vessels within the microcirculation. Finally, mechanistic knowledge on the role of microvessels in conditions such as hypertension and diabetes has profited greatly from the use of isolated small arteries. This technique had already been applied in experimental animal models for a long time and was introduced in human studies towards the end of the 20th century [3]. Although there are clear limits to the use of isolated small arteries in clinical studies, they have provided new insights into the molecular and cellular mechanisms of microvascular disease in diabetes, hypertension and other cardiovascular diseases.

Retinal imaging

The greatest technological innovation in recent decades in large-scale studies on the role of the microcirculation in (cardiovascular) disease has undoubtedly been advanced retinal imaging. Hypertensive retinopathy was described as early as the 19th century and was, for a long time, a qualitative marker of the degree of target organ damage in hypertension. Since loss of visual function was not regarded as a primary morbid event in hypertension, attention gradually shifted throughout the 20th century to other organs, more specifically the kidney, heart and brain. However, some 20 years ago a revival of retinal imaging was initiated by Wong and co-workers [10, 11] in collaboration with many centres worldwide. They introduced a relatively easy to use non-mydriatic video camera with advanced software to analyse retina microvascular network recordings off-line in great detail. This combination allows the measurement of microvascular diameters, wall (:) lumen ratios, vessel numbers, branching angles and vessel tortuosity. Furthermore, these imaging techniques have substantial reproducibility and can be used repeatedly in the same individuals for follow-up studies.

Since their original introduction in the early 21st century, retinal microvascular advanced imaging systems have developed further. The Heidelberg scanning laser Doppler flowmetry (SLDF) technology has already been mentioned above. Other new developments are optical coherence tomography (OCT) and OCT angiography [12, 13], which have now even reached primary care health centres for diagnosis of sight-threatening retinal disease; however, they can also be used to observe ocular biomarkers to detect systemic disease and follow their progression during treatment. This setting permits the collection of large data sets on ocular vascular parameters. A recent paper introduced the term “oculome” when using these data sets to characterise systemic disease [13]. Another important new development is adaptive optics ophthalmoscopy (AOO), which allows a near histological evaluation of the retinal microvasculature with a resolution of just a few micrometres [14, 15]. AOO is based on a commercially available flood-illumination retinal camera in combination with advanced semi-automatic software for image processing. The technique has been proven to be highly reproducible and is particularly suited for investigating structural microvascular changes since it provides a direct measure of the vessel wall. A recent review in the journal Blood Pressure discusses and compares these recent technological developments in retinal microvascular research in depth [9].

Retinal microvascular changes in hypertension

The various new technologies for retinal microvascular investigation have permitted the study of a range of relevant parameters, including vascular diameters, vascular density, branching architecture, tortuosity, focal narrowing and perfusion-related parameters. The early studies with these new technologies were performed primarily in hypertensive and diabetic patients, since the role of the microcirculation in these diseases had been firmly established at that time, although on the basis of more invasive and technically complex methods. The earlier studies basically confirmed previous microvascular observations in animal models of hypertension and observations in hypertensive humans. In both cases, hypertension is associated with arteriolar narrowing and increased wall (:) lumen ratio. Furthermore, a loss of small arterioles and capillaries (rarefaction) is a hallmark of even early stages of hypertension. The idea was that retinal imaging could give a new source of information, in particular in long-term epidemiological follow-up studies. A number of population-based studies between 2000 and 2020 indeed confirmed this idea. A review of these studies can be found in a 2012 paper by Liu et al and a 2019 follow-up paper by Cheung and co-workers [11, 16]. One of the interesting conclusions of these meta-analyses is that preclinical changes in the microcirculation and in large arteries have some shared but mainly unique pathways associated with cardiovascular disease. The retinal microvascular changes can already be observed in early stages of the development of hypertension and have a predictive power for its later development and related cardiovascular risk. Apart from heart-related cardiovascular risk, they are also early signs of brain-related risks, such as the development of (vascular) dementia. The latter finding is not surprising since the retina microvascular network is often regarded as a marker of brain vascular development and possible abnormalities.

In addition to these population-based studies, several authors have also focused on more mechanistic questions, such as early life vascular changes [17] or genetic determinants of microvascular function and structure in hypertension [18]. One of the major challenges for the years to come will be to unravel which molecular mechanisms determine the genetic influence of vascular development and how these mechanisms lead to sustained hypertension later in life. For this purpose, studies on changes in the (micro)vascular wall will be necessary. Newer adaptations of the retinal arteriolar morphology and perfusion assessment, such as SLDF and AOO, have been introduced recently that may allow such studies.

Retinal microcirculation as a marker and prognostic factor for cardiovascular disease

In a recent review, Allon et al [19] summarised the results of studies included in PubMed and Embase databases from 1 January 2000 to 1 January 2020 on retinal microvascular signs in patients with various cardiac diseases, including acute coronary syndrome, coronary artery disease, heart failure, and conduction abnormalities. Their overall conclusion from this extensive literature analysis is that the retinal microvasculature can provide essential data about concurrent cardiac disease status and predict future risk of cardiac-related events. One of the most interesting findings of their review was that microvascular diameter, in particular narrower arterioles in combination with wider venules, correlates with the incidence of acute coronary syndrome, especially in women, and can predict its occurrence. This gender discrepancy supports the hypothesis that microvascular dysfunction plays a greater role in the pathogenesis of coronary heart disease in women than in men. Another important conclusion from this systematic review is that, in addition to microvascular diameters, other retinal microvascular structural characteristics, such as branching angles and tortuosity, are associated with heart disease and death. This fascinating aspect of the microcirculation and its underlying molecular determinants will no doubt be a major area of research for years to come. It may also lead to novel possible targets for the prevention or treatment of cardiovascular disease and its treatment.

Conclusion

After a long history of ophthalmologic observations in patients with hypertension, this diagnostic procedure almost disappeared from routine medical care at the end of the 20th century. Major technological innovations around the year 2000 simplified and improved the way the retinal microcirculation can be studied. Many studies in the last two decades based on these innovations proved the value of the retinal microcirculation for assessment of microvascular changes in hypertensive patients and predicting long-term cardiovascular outcomes. Although the title of this short paper ends with a question mark, the conclusion finishes with an exclamation mark: examination of retinal vessels is useful in hypertensive patients!