Neural control, cardiac stress and retinal vascular dynamics: the SABPA study

Abstract

Motivation: Structural and functional similarities exist between the retinal, cerebral and the coronary vasculature. Specifically, the retinal and cerebral vasculature show striking anatomical, functional and autoregulatory similarities –maintaining constant blood pressure (BP), despite changes in systemic BP. The latter ensures adequate perfusion and the preservation of blood-ocular or blood-brain barriers. However, the ability to maintain these autoregulatory capacities diminishes with increases in BP. Sympathetic nervous system (SNS) hyperactivity, in response to both chronic and acute stressors, may be a risk factor for hypertension, ischemic stroke and other cardiovascular diseases (CVD), specifically in Africans. These conditions primarily affect the microvasculature. The retina is the only location where the microvasculature’s structure and function can be observed non-invasively. Great controversy exists regarding the legitimacy of external factors which may influence or determine myogenic, neurogenic and neurovascular coupling mechanisms – thereby impacting on the retinal vasculature’s ability to autoregulate. Studies investigating SNS hyperactivity, markers of cardiac stress and neurotrophins’ association with retinal vasculature are scarce or lacking all-together. A so-called brain-retina-heart link may be explored as a mechanistic pathway within the context of CVD, myocardial ischemia and stroke. Aims: The primary aim of this study was to evaluate the influence of SNS hyperactivity, cardiac stress and neurotrophic factors on the retinal vasculature’s structure and function within a South African, bi-ethnic cohort. This novel approach mandates a step-by-step evaluation and exploration of the brain-retina-heart link, providing a more holistic, preliminary mechanistic identification and translating to mainly stroke-risk susceptibility. The Sympathetic activity and Ambulatory Blood Pressure in Africans (SABPA) study was ideal in addressing these aims, as it is the only study in Sub-Saharan Africa which is designed to investigate the brain-heart link, by way of retinal vascular structure and function, cardiac stress markers, affiliated neurotrophic factors, cardiovascular and stroke risk, in a bi-ethnic cohort. Additionally, various investigations within the SABPA study have shown that the SABPA-Africans present with a SNS hyperactive status – hence this cohort is ideal to investigate the effects of SNS hyperactivity on retinal vascular structure and function. Methodology: This study is an affiliate of the SABPA study. Participants were 409 African and Caucasian teachers recruited from the North West Province (Dr Kenneth Kaunda Educational District), South Africa, all of similar socio-economic, educational and healthcare status. All participants in phase 1 (2007 – 2008) were invited to take part in phase 2 (2011 – 2012). Of the initial 409 participants, 359 reported back for phase 2 of the study. Only participants who had participated in phase 2 were included, as no retinal vessel analyses were performed during phase 1. From these participants, n=41 poor/ unusable retinal vessel recordings, n=1 epilepsy were excluded from all analyses. Additional exclusions were made in the separate manuscripts to align with the set hypotheses of each. Groups were stratified according to ethnicity as evident by statistically significant interactions between major variables. The dynamics vessel analyses (DVA) was used to assess retinal vessel structure (central retinal arteriolar equivalent (CRAE), central retinal venular equivalent (CRVE) and arteriolar-to-venular ratio) and function (arteriolar and venular maximum dilation, arteriolar maximum constriction, time to dilate and to constrict and area-under the curve (AUC) analyses during flicker-light-induced-provocation (FLIP). Systemic cardiac stress markers measured in serum included cardiac troponin T (cTnT) and amino-terminal pro-B-type natriuretic peptide (NT-proBNP). Heart-rate-variability (HRV) parameters included gold-standard time- and frequency-domain parameters measured during FLIP. Brain-derived neurotrophic factor (BDNF) was measured in serum samples. Retinal venular widening, arterio-venous nicking, cTnT values above 4.2ng/mL, NT-proBNP levels below the age and gender-specific reference ranges and BDNF levels below 1.51ng/mL were applied to determine the University of California (UCLA) Stroke-risk score to assess increased 10-year stroke-risk probability. A priori co-variates generally included age, body surface area, serum cotinine, gamma-glutamyl transferase, glycated haemoglobin, total cholesterol/ HDL cholesterol ratio, tumour-necrotic factor alpha, hypertensive/diabetic retinopathy, 24-hour pulse pressure and vessel segment diameter (the latter in all DVA analyses). Additional adjustments were made for CRVE in CRAE models and vice versa. Statistical calculations included T-tests, analysis of covariance and Chi-square tests. Multiple linear regression analyses calculated independent relationships between major variables, while odds ratios (OR) determined stroke-risk probability based on the UCLA stroke-risk score, independent of a priori co-variates. Receiver operating characteristics (ROC) determined the optimal BDNF cut-point predictive of retinopathy. General Population Profile: Overall, Africans persistently presented with wider retinal venules, smaller arterio-venous ratio and greater arteriolar and venular dilation in response to FLIP, than Caucasians. Africans displayed a poorer cardio-metabolic profile. Africans revealed lower NT-proBNP levels, slightly higher BDNF levels, yet similar cTnT values were observed between ethnic groups. The African group had higher: 24H-BP (hypertensive average) and pulse pressure, pre-and-post FLIP BP, intra-ocular pressure and diastolic ocular perfusion pressure values, hypertensive/diabetic retinopathy and prevalence of AV-nicking compared to Caucasians. Africans also showed a greater decrease in 24H-HRV and HRV during FLIP in both time and-frequency domain parameters, than their Caucasian counterparts. Results and Conclusions: The main results and conclusions for the three manuscripts prepared as part of this thesis are as follows: 1. Retinal vasculature reactivity during flicker-light-provocation, cardiac stress and stroke risk in Africans: The SABPA study: In Africans, a reduced retinal arteriolar calibre and attenuated arteriolar dilation during FLIP was associated with higher cTnT levels. Their larger retinal-venular calibre and attenuated arteriolar dilation during FLIP were associated with lower NT-proBNP levels. Again, exclusively in Africans, increased cardiac stress, wider venular calibres and retinal arteriovenous-nicking, predicted an increased 10-year stroke risk. None of these associations were evident in the Caucasian group. Increased cardiac stress, mainly attributed to well-documented SNS hyperactivity in Africans, may contribute to diminished myogenic responses. Hence observable changes in the retinal vasculature may serve as markers for the identification and prediction of cardio-systemic and cerebral vascular morbidities and risks – thereby establishing a brain-retina-heart link. 2. Heart rate variability, the dynamic nature of the retinal microvasculature and cardiac stress: Providing insight into the brain-retina-heart link: The SABPA study Africans displayed wider retinal venules and attenuated HRV during FLIP. FLIP elicited both increased SNS activity and decreased parasympathetic activity (SDNN and rMSSD time domain analyses) and modulation (gold-standard LFnu, HFnu frequency analyses) in this bi-ethnic cohort. In Africans, decreased HRV during FLIP accompanied greater arteriolar and venular responses and elevated systemic levels of cTnT, implying that the SNS exerted a significant effect on the smooth-muscle tone of the retinal vasculature, either directly or indirectly. SNS hyperactivity may contribute to disrupted retinal autoregulation. These findings may exemplify central control by the brain on all systemic regulatory functions, across all vascular beds. 3. Low Brain-derived-neurotrophic-factor reflects attenuated retinal vascular functionality and increased stroke risk: The SABPA study Lower systemic BDNF levels – the most abundant neurotrophic factor in the retina – were observed in the total cohort compared to the normal reference ranges. However, low follow-up BDNF levels were only associated with attenuated retinal arteriolar responses in the African group. Attenuated arteriolar constriction responses were exclusively observed in the Africans. Although low follow-up BDNF levels were also observed in Caucasians, BDNF’s direct vasodilatory effect appeared to be more pronounced in this group. However, low BDNF levels predicted retinopathy and increased stroke risk, irrespective of race or gender. Attenuated arteriolar constriction responses as well as stroke risk may indicate a diminished neuro-protective effect of BDNF in the SABPA cohort. Sustained SNS hyperactive state (supported by decreased heart rate variability, during FLIP, associating with lower BDNF levels), as observed in the Africans, may lead to blunted or diminished BDNF action, resulting in lower BDNF levels. Due to a sustained SNS hyperactive state, and high pressure conditions (lower BDNF levels were associated with increased IOP in this African cohort), BDNF’s direct action on vascular-smooth-muscle cells may be altered, modifying arteriolar vascular resistance. This may contribute to disturbed neurovascular coupling as well as increased stroke risk, particularly in Africans. These findings further exemplify the brain’s central, neurogenic control function exerted on autoregulatory processes, regardless of the vascular bed. General conclusion: The current study exemplifies the possible role of SNS hyperactivity, regarding the control of retinal vascular dynamics, specifically in Africans. This role is emphasized as SNS hyperactivity and the subsequent modified hemodynamic responses observed during such high-pressure conditions, chronic cardiac stress and diminished neuroprotective effects may possibly directly contribute to impaired retinal autoregulation. This sustained high-pressure system may not only lead to a decrease in and eventual diminishing of myogenic control mechanisms, but also indicate an altered neurogenic response and neurovascular coupling, all ultimately contributing to autoregulation. The presence of pre-existing SNS hyperactivity, and additional SNS activity provoked during FLIP may further increase SNS activity and modulation. During a stressor, like FLIP, an existing hyperactive SNS (as is repeatedly evident in the SABPA Africans) will further be challenged, thereby causing further reduction in HRV indicating greater SNS activity and modulation, in an attempt to maintain adequate perfusion. Existing SNS hyperactivity-driven, pressure-induced endothelial dysfunction and impaired smooth-muscle responses may manifest as attenuated retinal vessel dilatory and constriction responses. The latter also indicates that BDNF’s direct action on vascular smooth-muscle cells might be altered, attenuating arteriolar vascular resistance and contributing to disturbed neurovascular coupling. This increased, sustained SNS activity/modulation may lead to a prolonged vessel response, even after the stressor has ceased, indicating difficulty in re-establishing proper SNS balance, once this balance has been chronically disturbed. As a consequence the autoregulatory capacity of the retinal microvasculature may be impaired and the risk for stroke, retinopathy and possibly glaucoma and optic nerve damage will be increased within the SABPA African population.