Marinobufagenin and markers of early cardiovascular risk in a young black and white population: The African-PREDICT study

Abstract

There have been many arguments about the harmful effect of increased or low salt intake and its concurrent role in cardiovascular health. While an overwhelming amount of research has focused on the relationship between salt and blood pressure, more attention has been brought to novel markers associated with increased salt consumption and their role in cardiovascular pathophysiology. One such biomarker is the steroidal hormone and Na+/K+-ATPase inhibitor, marinobufagenin (MBG), released in response to the sodium induced angiotensin-aldosterone-sympatho-excitatory pathway. In vitro and animal model studies have shown MBG infusion to promote microvascular hyperpermeability, cardiovascular fibrosis and cardiac hypertrophy. These findings were concomitant with increased plasma MBG and urinary MBG excretion in animals. Studies in humans however are far more limited, with the majority of studies investigating relationships between cardiovascular risk markers and MBG (urinary and plasma MBG) being performed in populations with reported pathologies. Information on relationships between markers of cardiovascular risk and MBG in young healthy adults is scarce. If we are able to determine associations between established markers of cardiovascular risk and MBG at an early age and prior to cardiovascular pathologies, especially in a population consuming excessive amounts of salt, this would reinforce the need to implement sodium reduction strategies in an effort to reduce cardiovascular risk.Furthermore, there is a lack of research pertaining to MBG and the possible diverse physiological or pathophysiological roles thereof in men and women or black and white ethnic groups, respectively. There are various studies indicating increased salt-sensitivity in women as well as black populations. It is therefore also possible that these specific groups may be more sensitive to the cardiovascular effects of MBG and may be at a greater risk. The central aim of this study was to investigate the role of 24hr urinary MBG as a potential early marker in the development of cardiovascular disease. This was done by exploring the relationships of 24hr urinary MBG with established markers of early cardiovascular risk in young healthy adults. This study specifically investigated these relationships within respective sex (men and women) and ethnic groups (black and white) so to bring forth new evidence and add to a body of literature where information in these respective populations are scant. Taking into consideration the young age, as well as the peak cardiovascular health of the participants of this study, new evidence of possible relationships between 24hr urinary MBG and early markers of cardiovascular risk may highlight the adverse role of MBG on the cardiovascular system prior to the onset of disease. The original research study included in this thesis made use of the data of the first 711 consecutively enrolled participants from the African Prospective study on the Early Detection and Identification of Cardiovascular disease and Hypertension (African-PREDICT) (42% men and 51% black) with complete 24hr urinary data. The African-PREDICT study enrolled black and white, men and women (between the ages of 20-30 years), from diverse communities and socio-economic backgrounds from in and around the Potchefstroom area (North-West Province, South Africa). All individuals who took part in the study voluntarily underwent health screening prior to inclusion into the African-PREDICT study. The following criteria were used to determine eligibility for inclusion into the study: participants were normotensive (office blood pressure <140/90 mmHg); HIV uninfected; had not been previously diagnosed with any chronic illnesses (self reported); and were not using chronic medication. Also, none of the women who participated in the study were pregnant or lactating at the time that measurements were performed. All participants gave written informed consent. Individuals who met the inclusion criteria were invited back to take part in the advanced measurements of the African-PREDICT study. Simultaneous ambulatory blood pressure monitoring (Card(X)plore device Meditech, Budapest, Hungary, British Hypertension Society (BHS)) and 24hr electrocardiography (ECG) (Cardio Visions 1.15.2 Personal Edition software, Meditech, Budapest, Hungary) were done to determine 24hr blood pressure (including night-time systolic blood pressure dipping status) and heart rate variability. Data from the 24hr ECG time and frequency domains included low frequency heart rate variability (LF HRV), high frequency heart rate variability (HF HRV), LF/HF heart rate variability ratio, and the standard deviation of normal-to-normal intervals (SDNN). Large artery stiffness was measured as carotid-femoral pulse wave velocity (cfPWV)(Sphygmocor® XCEL device, AtCor Medical Pty. Ltd., Sydney, Australia), and microvascular function determined as the peak retinal artery dilation in response to a light flicker provocation (Dynamic Retinal Vessel Analyzer (DVA), Imedos Systems, Jena, Germany). The central retinal artery (CRAE) and central retinal vein equivalent (CRVE) were also determined using static retinal images. Indices of left ventricular structure (left ventricular mass index (LVMi)) and function (stroke volume index (SVi), cardiac output index (COi), left atrial to aortic ratio (LA:Ao), mitral valve E to A ratio (E:A), ratio of mitral peak velocity of early filling (E) to early diastolic mitral annular velocity (e’) (E:e’)) were determined by means of transthoracic echocardiography (General Electric Vivid E9 device, GE Vingmed Ultrasound A/S, Horten, Norway). The current recommendation of the World Health Organization refers to 24hr urine sampling as the golden standard for determining the average population wide salt intake. In this study, 24hr urinary biomarkers included MBG, sodium, potassium, creatinine and albumin. Estimated salt intake was calculated from 24hr urinary sodium excretion. Estimated glomerular filtration rate (eGFR) and albuminuria were used as indices of renal function. Additional biochemical analyses were performed using serum samples to measure aldosterone (RIA Aldosterone Kit, Beckman Coulter, Immunotech, Radiova, Czech Republic), interleukin-6 (IL-6) (high sensitivity Quantikine ELISA kit), C-reactive protein (CRP), creatinine, high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), total cholesterol (TC), triglycerides, gamma glutamyltransferase (GGT), glucose (Cobas Integra 400plus, Roche, Basel Switzerland) and cotinine (chemiluminescence method on the Immulite, Siemens, Erlangen, Germany). For each manuscript we tested for an interaction of sex and ethnicity on the relationships between MBG and cardiovascular risk markers. Group stratifications for further statistical analyses were done accordingly. The Kolmogorov-Smirnov test was conducted to test for the normal distribution of data. Normally distributed data were shown as the mean and standard deviation, whereas non-Gaussian distributed data were logarithmically transformed and presented as geometric means with the 5th and 95th percentiles. Independent T-tests, analyses of covariance and chi-square tests were used to compare the respective means and proportions. We performed Pearson, partial and multivariate regression analyses to demonstrate relationships between 24hr urinary MBG and markers of early cardiovascular risk. A more detailed outline of the group stratifications and statistical analyses are provided in the respective chapters. Basic characteristics Based on interaction testing, we performed statistical analyses and group comparisons in men and women. Sensitivity analysis for ethnicity was also performed, although our results seemed to be more sex-specific than ethnic-specific. The mean estimated salt intake for this population was 7.69 g/day, with men consuming approximately 8.32 g/day (2.42; 21.2 g/day, 5th and 95th percentiles; N=296) and women 7.27 g/day (2.70; 19.2 g/day, 5th and 95th percentiles; N=415) (p<0.003). Expectedly, men also had increased mean 24hr urinary MBG excretion (4.13 nmol/day) compared to women (2.69 nmol/day) (p<0.001). Mean 24hr urinary MBG excretion was higher in white compared to black men (4.59 vs. 3.70 nmol/day, p<0.001), but similar for white and black women (2.83 vs. 2.56 nmol/day, p=0.11). While salt intake did not differ between black and white men, black women (7.83 g/day) consumed more salt than their white counterparts (6.68 g/day) (p=0.007). Men from this study demonstrated higher 24hr blood pressure, cfPWV and LF HRV (p<0.001) compared to women, while women had elevated 24hr heart rate and HF HRV (p<0.001). When comparing the characteristics of participants within the lowest quartile of MBG excretion to those in the highest quartile, adults with excessive MBG excretion (quartile four) had higher 24hr systolic blood pressure, LVMi, EDVi and SVi (all p<0.050). We also compared the cardiovascular, retinal microvascular and biochemical profiles of dippers (night-time blood pressure dipping >10%) and non-dippers (night-time blood pressure dipping <10%). Night-time systolic blood pressure was significantly higher in non-dippers (113 ± 9.01 mmHg) compared to dippers (105 ± 8.99 mmHg) (p<0.001), with non-dippers exhibiting a mean night-time systolic blood pressure dipping percentage of only 5.9%. Non-dippers had narrower CRAE compared to dippers, although CRVE and retinal microvascular reactivity to a stressor did not differ. There was also no difference in the estimated salt intake or 24hr urinary MBG excretion between the two groups. We firstly reiterated results from previous studies that included older or hypertensive adults, by demonstrating a strong correlation between estimated salt intake and 24hr urinary MBG (r=0.49; p<0.001) in the young healthy men and women from the African-PREDICT study. Secondly, we found that only in women, cfPWV increased significantly across increasing quartiles of 24hr urinary MBG excretion independent of mean arterial pressure and estimated salt intake (p=0.001). The positive association persisted after performing multiple regression analyses adjusting for several covariates (Adj. R2=0.23; std. β=0.15; p=0.002). In addition, LVMi associated positively with MBG excretion only in women after partial (r=0.38; p<0.001) and multi-variable regression analyses (Adj. R2=0.06; std. β=0.127; p=0.015). In the total group, we found significant relationships of MBG with systolic blood pressure, indices of early target organ damage and SVi in unadjusted analyses. MBG correlated positively with systolic blood pressure (r=0.20; p<0.001), LVMi (r=0.21; p<0.001), EDVi (r=0.20; p<0.001), SVi (r=0.10; p=0.008) and negatively with eGFR (r=-0.08; p=0.031). However, after adjusting for age, sex and ethnicity, the negative relationship between MBG and eGFR lost significance. After performing multivariate adjusted regression analyses the relationships of MBG with LVMi, EDVi and SVi became borderline significant. Still, LVMi was positively associated with MBG excretion in the highest MBG excretion quartile (Adj. R2=0.20; std. β=0.15; p=0.043; N=165). Thirdly, we indicated that both estimated salt intake and plasma aldosterone contribute positively to multiple regression models with MBG excretion as the main dependent variable in men and women (both p<0.001). However, contrasting associations of MBG with indices of autonomic activity were evident between women and men, with MBG excretion associating positively with LF HRV in women (Adj. R2=0.33; β=0.11; p=0.030) and negatively with HF HRV in men (R2=0.36; β=0.12; p=0.034). Lastly, we demonstrated in young healthy adults with ambulatory blood pressure <140/90 mmHg, who exhibit a non-dipping night-time blood pressure pattern, that MBG excretion was associated with reduced peak retinal artery dilation (Adj. R2=0.34; β=-0.26; p<0.001).To a lesser extent, estimated salt intake was also inversely associated with peak retinal artery dilation (Adj. R2=0.30; β=-0.14; p=0.051), but this relationship was significantly confounded by MBG excretion (Adj. R2=0.33; β=-0.015; p=0.86). Taking into consideration the known positive relationship between MBG and salt-sensitivity, and the literature linking black ethnicity with increased salt-sensitivity, we performed additional sensitivity analyses for the interaction of ethnicity on the relationships of MBG with cardiovascular risk factors. While we saw clear sex differences in the relationships of MBG with cfPWV and LVMi, we found no interaction of ethnicity on the latter relationships. We did, however, find that the relationships between MBG and indices of autonomic activity were only evident in black women and men, but not their white counterparts. MBG excretion associated positively with LF HRV in women (Adj. R2=0.38; β=0.13; p=0.036) and negatively with HF HRV in men (Adj. R2=0.40; β=0.18; p=0.045). For the first time in a young human cohort, this study demonstrated significant associations between markers of increased cardiovascular risk and elevated 24hr urinary MBG excretion. Additionally, our results support previous findings, only demonstrated in animals, of the possible involvement of the sodium-induced angiotensinergic-sympatho-excitatory pathway in MBG stimulation. In young healthy adults, we found positive associations between known predictors of increased cardiovascular risk (large artery stiffness, increased left ventricular mass and reduced retinal microvascular function in non-dippers) and 24hr urinary MBG excretion. Our results suggest that MBG may play a harmful role in the early development of cardiovascular disease, especially in populations consuming a habitual high salt diet. Although we found no association between MBG and kidney function it is possible that MBG may have an adverse effect on the kidneys at a later stage in life. Notably, our results were predominantly evident in women who are reportedly more salt-sensitive compared to men. We therefore propose that women are likely more sensitive to the cardiovascular effect of MBG and may be at greater risk when consuming increased amounts of salt. However, the lack of relationships between MBG and markers of early cardiovascular risk in black adults was unforeseen. Especially since black ethnicity is associated with salt-sensitivity.This study, furthermore, demonstrated positive relationships of estimated salt intake and aldosterone with 24hr urinary MBG excretion in men and women. However, only in women did we find a positive association between indices of sympathetic activity and 24hr urinary MBG excretion. We were able to replicate previous findings from animals in a human cohort, namely that MBG was associated with components of the proposed sodium induced angiotensinergic-sympatho-excitatory pathway.Despite the cross-sectional study design being a main limitation and prohibiting conclusions on causality, this study provides valuable insights into the possible early pathophysiological role of MBG in humans. In addition, taking into consideration the results of this study and the comprehensive literature linking salt intake and MBG, the study’s main findings support the implementation of sodium reduction strategies in South Africa.