Blood pressure and retinal small arteries

Abstract

The clinical usefulness and relevance of the pioneering work of Keith et al., categorizing the hypertensive retinopathy into four groups with increasing severity, has been questioned because of insufficient reproducibility of grade I and II retinopathy [1]. Since retinal vessels offer the unique opportunity to visualize body's microvasculature directly, noninvasively, repeatedly and safely in vivo, advanced imaging techniques and (semi)-automatic analyzing methods were introduced to overcome these criticisms and to detect reliably early retinal vascular alterations. One of these new promising approaches, introduced by our study group about 10 years ago, assesses both functional (perfusion) and retinal arteriolar structural parameters of the retinal circulation by using scanning laser Doppler flowmetry (SLDF) with automatic full-field perfusion imaging analyses [2]. This approach allows the noninvasive assessment of both the outer and inner diameter of retinal arterioles in vivo, and thus, early vascular remodeling of retinal arterioles by calculating wall-to-lumen ratio (WLR) [3]. Vascular remodeling, characterized by an increased WLR, of small resistance and large arteries, has been identified as one of the early processes that occurs in response to increased blood pressure (BP) and reflects early hypertensive end-organ damage [4]. By using this method, we could show that hypertensive patients with a history of cerebrovascular events have an increased WLR of retinal arterioles compared to both treated hypertensive patients and normotensive individuals. Moreover, treated hypertensive patients with poor BP control have higher WLR than treated hypertensive patients with good BP control [3]. In addition, never-treated hypertensive patients revealed higher WLR than normotensive individuals [5]. It is noteworthy to stress that WLR of retinal arterioles was closely correlated with the micromyographic measurement of media-to-lumen ratio of subcutaneous small arteries, which has been documented to predict cardiovascular events [6]. Moreover, this technique is now mentioned as a new technological approach in the 2013 European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC) Guidelines for the management of arterial hypertension [7]. In this issue of the Journal of Hypertension, the reputable study group of Wong presented data about another development analyzing noninvasively retinal vasculature. The relationship of BP on retinal vascular network was analyzed in a multiethnic population, hence addressing also potential race/ethnic variations [8]. Assessment of retinal network was based on a computer-assisted program, elaborated by the authors, capable of performing analysis of the recently newly developed parameter retinal vascular fractal dimension (Df) that has been proposed to be a global measure of the geometric complexity. After adjustments, it was shown that retinal Df was inversely correlated to BP level. Interestingly, this was evident in patients with uncontrolled or untreated hypertension, but not in controlled hypertension. These observations were true in all three ethnic groups (Chinese, Malay and Indian participants, a total of 3876 participants). Taken together, these results seem to be in accordance with the above-mentioned findings of our group using a different technology. Previously, the same study group used a different approach analyzing the effect of BP on retinal vasculature. By examining retinal calibers, they observed that the retinal arteriolar caliber was narrower in uncontrolled/untreated hypertensive patients as compared to both controlled hypertensive patients and normotensive individuals. After adjustment of confounding factors, a gradual BP increase was inversely associated with a decrease in arteriolar caliber. In line with the current study, this was evident in all three included ethnic groups (Chinese, Malay and Indians) [9]. More recently, the same study group reported an association of different quantitative and qualitative parameters of retinal microvascular abnormalities with elevated BP that was based on the Singapore Malay Eye study. Moreover, the results suggest that smaller retinal Df combined with the evidence of straighter retinal arterioles may represent a possible marker of a poor BP control in treated hypertensive patients [10]. There is no doubt that the study group led by Wong is a highly acknowledged expert in the field of retinal examinations, initiating and analyzing large-scale important studies (e.g. The Multi-Ethnic Study of Atherosclerosis and Blue Mountains Eye Study) [11,12]. However, the increasing number of different quantitative and qualitative parameters of retinal vasculature (though with similar findings in the Asian population) may be confusing and leave the question open which may be the most representative parameter. By using only one parameter of retinal structural alterations, namely retinal Df, in the present study, the authors miss the opportunity to differentiate whether retinal DfI is the most promising and reliable parameter to detect early retinal involvement in the clinical course of hypertensive state. Another important point was not addressed by the study group. Nowadays, the importance of central hemodynamics is obvious. For example, the Conduit Artery Function Evaluation (CAFE) study, conducted in hypertensive patients, has shown that the noninvasively measured central BP is superior to brachial BP in predicting cardiovascular outcomes [13]. Indeed, we could recently demonstrate in patients across a wide range of BP values that both peripheral and central SBP and pulse pressure correlated with WLR assessed by SLDF. Of note, however, central pulse pressure showed the strongest relation with wall-to-lumen ratio of retinal arterioles. In contrast, among others, peripheral mean arterial pressure was not at all correlated with WLR [14]. In the current study, no difference between antihypertensive drugs on vascular remodeling was found. This may at least be partly as a result of the cross-sectional nature of the study and lack of knowledge about the duration of antihypertensive drug therapy. Hence, it is quite speculative whether the used type of medication may result in different effects on microvascular remodeling. Previous analyses suggest that effective antihypertensive treatment is capable of reversing these vascular adaptive processes [15] and different antihypertensive drugs have disparate effects on microvascular structure [16]. Last but not least, the most intriguing question, however, remains unanswered whether (and which of the) noninvasive assessments of the retinal vasculature represents a reliable tool assessing long-term changes in clinical practice. ACKNOWLEDGEMENTS Conflicts of interest There no conflicts of interest.