Hendrik L. Booysen

Indexes of Aortic Pressure Augmentation Markedly Underestimate the Contribution of Reflected Waves Toward Variations in Aortic Pressure and Left Ventricular Mass

Although indexes of wave reflection enhance risk prediction, the extent to which measures of aortic systolic pressure augmentation (augmented pressures [Pa] or augmentation index) underestimate the effects of reflected waves on cardiovascular risk is uncertain. In participants from a community sample (age >16), we compared the relative contribution of reflected (backward wave pressures and the reflected wave index [RI]) versus augmented (Pa and augmentation index) pressure wave indexes to variations in central aortic pulse pressure (PPc; n=1185), and left ventricular mass index (LVMI; n=793). Aortic hemodynamics and LVMI were determined using radial applanation tonometry (SphygmoCor) and echocardiography. Independent of confounders, RI and backward wave pressures contributed more than forward wave pressures, whereas Pa and augmentation index contributed less than incident wave pressure to variations in PPc (P<0.0001 for comparison of partial r values). In those <50 years of age, while backward wave pressures (partial r=0.28, P<0.0001) contributed more than forward wave pressures (partial r=0.15, P<0.001; P<0.05 for comparison of r values), Pa (partial r=0.13, P<0.005) contributed to a similar extent as incident wave pressure (partial r=0.22, P<0.0001) to variations in LVMI. Furthermore, in those ≥50 years of age, backward wave pressures (partial r=0.21, P<0.0001), but not forward wave pressures (P=0.98), while incident wave pressure (partial r=0.23, P<0.0001), but not Pa (P=0.80) were associated with LVMI. Pa and augmentation index underestimated the effect of wave reflection on PPc and LVMI in both men and women. Thus, as compared with relations between indexes of aortic pressure augmentation and PPc or LVMI, strikingly better relations are noted between aortic wave reflection and PPc or LVMI.

Reflected rather than forward wave pressures account for brachial pressure-independent relations between aortic pressure and end-organ changes in an African community.

Aims: To determine whether brachial blood pressure (BP)-independent relations between aortic pressure and cardiovascular damage are better explained by reflected (backward) (Pb) or forward (Pf) wave pressure effects. Methods: In 1174 participants from a community of African ancestry, we assessed central aortic pulse pressure (PPc), Pb, and Pf (radial applanation tonometry, SphygmoCor) as well as left ventricular mass index (LVMI) (n = 786), aortic pulse wave velocity (PWV) (n = 1019), carotid intima-media thickness (IMT) (n = 578), transmitral early-to-late left ventricular diastolic velocity (E/A) (n = 779) and estimated glomerular filtration rate (eGFR) (n = 1174). Results: Independent of mean arterial pressure and confounders, PPc, and both Pb and Pf were associated with end-organ measures or damage (P < 0.05 to P < 0.0001). With adjustments for brachial PP and confounders, Pb remained directly associated with LVMI (partial r = 0.09, P < 0.01), PWV (partial r = 0.28, P < 0.0001), and IMT (partial r = 0.28, P < 0.0001), and inversely associated with E/A (partial r = −0.31, P < 0.0001) and eGFR (partial r = −0.14, P < 0.0001). Similar relations were noted with the presence of end-organ damage (P < 0.05 to P < 0.0001). In contrast, with adjustments for brachial PP and confounders, Pf no longer retained direct relations with LVMI, PWV, and IMT or inverse relations with E/A and eGFR. Adjustments for Pb, but not Pf, diminished brachial PP-independent relationships between PPc and end-organ measures. Independent relations between Pb, but not Pf and end-organ measures, were largely attributed to Pb accounting for most of the variation in brachial-to-aortic PP amplification. Conclusions: In communities of African ancestry, brachial BP-independent relations between aortic pressure and end-organ changes are largely attributed to an impact of reflected rather than forward wave pressures.

A Mismatch Between Aortic Pulse Pressure and Pulse Wave Velocity Predicts Advanced Peripheral Arterial Disease

OBJECTIVES To determine whether increases in central aortic pulse pressure (PPc), but decreases in carotid-femoral pulse wave velocity (PWV) predict the presence of advanced peripheral arterial disease (PAD). METHODS Applanation tonometry and vascular ultrasound were employed to assess carotid-femoral PWV, PPc, and carotid intima media thickness (IMT) in 136 patients of African ancestry with chronic critical lower limb ischaemia (CLI) and in 1,030 randomly selected healthy adults of African ancestry, 194 of whom were age- and sex matched (controls). RESULTS With adjustments for confounders, compared with age- and sex-matched controls, participants with CLI had an increased carotid IMT (p = .0001) and PPc (p < .0001), but a markedly reduced PWV (m/second) (CLI = 5.7 ± 3.7, controls = 8.6 ± 3.4, p < .0001). PWV was correlated with PPc in controls (r = .52, p < .0001), but not in CLI (r = -.06). A PPc/PWV mismatch index showed increased values in participants with CLI over the full adult age range assessed. With carotid IMT, PPc, or aortic augmentation index in the same regression model, an increase in the PPc/PWV mismatch index was independently associated with CLI (p < .0001) and a PPc/PWV value upper 95% confidence interval in the community sample predicted CLI (odds ratio = 32 [6-169], p < .0001). PPc/PWV predicted CLI with a similar level of performance and accuracy and a greater specificity (98%) than that of IMT (82%). CONCLUSION In CLI, while PPc increases, carotid-femoral PWV is markedly reduced. A PPc/PWV mismatch may be a new risk marker for advanced PAD.

Differential relationships of systolic and diastolic blood pressure with components of left ventricular diastolic dysfunction.

Aims: To determine whether SBP or DBP is best associated with different components of left ventricular diastolic dysfunction. Methods: In 241 randomly selected participants, echocardiographic left ventricular diastolic function was assessed from early-to-atrial (E/A) transmitral velocity and E/e′ where e′ represents myocardial tissue lengthening velocity in early diastole as measured at the mitral annulus. Relationships between diastolic function and blood pressure (BP) were assessed from brachial and central aortic (radial applanation tonometry and SphygmoCor software) measurements. Results: Independent of confounders, brachial DBP (partial r = –0.21, P < 0.002), but not SBP (partial r = –0.09, P = 0.18), was associated with E/A and the relationship between brachial DBP and E/A persisted with adjustments for brachial (P < 0.002) or aortic (P < 0.05) SBP. Although aortic SBP was independently associated with E/A, this relationship did not persist with adjustments for DBP (partial r = –0.05, P = 0.44). In contrast, both brachial (partial r = 0.34, P < 0.0001) and aortic (partial r = 0.34, P < 0.0001) SBP were independently associated with E/e′, effects that persisted with adjustments for DBP (P < 0.0001), although independent relationships between DBP and E/e′ did not persist with adjustments for brachial or aortic SBP (P = 0.17–0.57). In quartiles of DBP or SBP within normal-to-high normal ranges, multivariate adjusted E/A was decreased and E/e′ increased as compared with those with optimal BP values (P < 0.05 to P < 0.005). Conclusion: Both SBP and DBP are important determinants of separate components of left ventricular diastolic dysfunction and these effects are noted even within normotensive BP ranges. DBP may be as important as SBP in the transition to diastolic dysfunction.

Aortic, but not brachial blood pressure category enhances the ability to identify target organ changes in normotensives.

Aims: We sought to determine whether within normal/high-normal blood pressure (BP) ranges (120–139/80–89 mmHg), aortic BP may further refine BP-related cardiovascular risk assessment, as determined from target organ changes. Methods: In 1169 participants from a community sample of African ancestry, 319 (27%) of whom had a normal/high-normal BP, aortic BP was determined using radial applanation tonometry and SphygmoCor software, and target organ changes assessed from carotid-femoral pulse wave velocity (PWV) (n = 1025), estimated glomerular filtration rate (eGFR) (n = 944), and left ventricular mass indexed to height2.7 (LVMI) (n = 690). Results: Normal versus high-normal BP categories failed to differentiate between those participants with a BP above optimal values with versus without multivariate-adjusted target organ changes. However, in those with a normal/high-normal BP with aortic SBP values that were less than 95% confidence interval of healthy participants with optimal BP values (45% of those with a normal/high-normal BP), no unadjusted or multivariate adjusted target organ changes were noted. In contrast, those with a normal/high-normal BP with aortic SBP values that exceeded optimal thresholds, demonstrated unadjusted and multivariate adjusted increases in PWV and LVMI and decreases in eGFR (P < 0.05 to P < 0.005 after multivariate adjustments). Conclusion: In contrast to normal versus high-normal BP categories which do not clearly distinguish normotensives with from those without organ damage, noninvasively determined aortic BP measurements may refine the ability to detect those with a normal/high-normal BP at risk of BP-related cardiovascular damage.

Aortic Pulse Pressure Amplification Imputed From Simple Clinical Measures Adds to the Ability of Brachial Pressure to Predict Survival

BACKGROUND Although aortic-to-brachial pulse pressure amplification (PPamp) may offer prognostic information beyond brachial blood pressure (BP), this approach is limited in resource-limited settings. We aimed to derive an equation to impute central aortic PP (PPc) from simple clinical measures and assess whether imputed PPamp adds to the ability of brachial BP to predict mortality. METHODS An imputation equation for PPc, incorporating brachial PP, age, mean arterial pressure, and pulse rate, was identified from multivariate modeling of the factors associated with radial applanation tonometry-derived (measured) PPc in 1,179 community participants and validated in a clinical sample of 351 patients. We applied the equation to ambulatory awake BP and pulse rate values in a separate group of 4,796 patients referred for ambulatory monitoring and evaluated the impact on all-cause mortality. RESULTS Imputed PPc values closely approximated measured PPc (r (2) = 0.96, mean difference ± (2 × SD) = 1.4±6.2mm Hg). In adjusted Cox proportional models including adjustments for awake brachial PP during 47,111 person-years of follow-up, where 648 patients died, hazards ratio for all-cause mortality per SD of awake PPamp was 0.79 (95% confidence interval (CI): 0.68-0.93, P < 0.005). The hazards ratio for brachial PP with (1.49, CI = 1.36-1.64, P < 0.0001) or without (1.46, CI = 1.35-1.59, P < 0.0001) PPamp in the model was similar. Awake PPamp also predicted survival independent of awake brachial systolic BP (P < 0.0001). CONCLUSIONS PPc imputed from simple clinical assessments closely approximates measured PPc. PPamp derived from imputed PPc adds to the ability of brachial BP to predict survival. In resource-limited settings, an imputation equation may be employed to approximate aortic BP and enhance risk prediction.

Carotid Intima-Media Thickness in African Patients with Critical Lower Limb Ischemia Infected with the Human Immunodeficiency Virus

Background: The extent to which Human Immunodeficiency Virus (HIV) is associated with increases in carotid Intima-Media Thickness (IMT) independent of conventional cardiovascular risk factors is unclear. Hence, we evaluated whether independent of conventional risk factors, an increased carotid IMT occurs in African HIV infected patients with chronic Critical Limb Ischemia (CLI). Methods: Carotid IMT was measured in 217 sequentially recruited patients with CLI, 25 of whom were HIV positive and in 430 randomly selected controls from a community sample. Results: As compared to HIV negative patients with CLI, HIV positive patients were younger (49 ± 10 vs. 64 ± 11 years, p<0.0001) and had a markedly lower prevalence of hypertension and diabetes mellitus (p<0.0001), but a similar proportion of patients smoked (76% vs 67%). However, as compared to patients with CLI who were HIV negative, HIV positive patients had a similar increase in carotid IMT (HIV positive= 0.75 ± 0.14 mm; HIV negative= 0.79 ± 0.14 mm; Controls= 0.64 ± 0.15, p < 0.0001 versus Controls) even after adjustments for age, sex and conventional risk factors (HIV positive= 0.75 ± 0.13 mm; HIV negative=0.73 ± 0.15 mm, Controls=0.66 ± 0.15, p < 0.005). IMT was similarly increased in HIV positive patients with CLI as compared to controls when assessed in men, smokers, and black African patients only (p < 0.05-0.0001), or in those who were receiving highly active antiretroviral therapy (n=12, 0.74 ± 0.10 mm) as compared to those not receiving therapy (0.75 ± 0.15 mm). As compared to controls, the age- sex- and conventional risk factor-adjusted odds of having an IMT ≥ 0.8 mm was similarly increased in patients with CLI who were HIV positive (odds ratio= 8.89, CI= 2.79-28.32, p= 0.0002) as those who were HIV negative (odds ratio= 2.70 CI= 1.51-4.81, p < 0.001). Conclusion: These results suggest that despite being of a younger age, with or without conventional risk factor adjustments, marked increases in carotid IMT in HIV in Africa are a risk factor for CLI.

Time to the peak of the aortic forward wave determines the impact of aortic backward wave and pulse pressure on left ventricular mass

Aim: To determine the degree to which an extended time to the peak of the aortic forward wave or early wave reflection time enhance associations between aortic backward wave pressure and hence central aortic pulse pressure (PPc) and left ventricular mass index (LVMI). Methods: In 701 adult participants from a community sample either receiving no antihypertensive therapy or receiving low-dose thiazide diuretic monotherapy for at least a year (the major therapy employed), we assessed aortic haemodynamics (SphygmoCor software and wave separation analysis; AtCor Medical, West Ryder, New South Wales, Australia) and LVMI (echocardiography). Results: An interaction between time to the peak of the aortic forward wave and aortic backward wave pressure was independently associated with aortic augmented pressure (P < 0.01), PPc (P < 0.005), LVMI (P < 0.01), and LV hypertrophy (LVH; P = 0.01). The time to the peak of the aortic forward wave–aortic backward wave pressure interaction translated into a stepwise increase in the independent association between aortic backward wave pressure and aortic augmented pressure or PPc across quartiles of time to the peak of the aortic forward wave (P < 0.05 to < 0.0001 for comparison of slopes of relations). Furthermore, the time to the peak of the aortic forward wave–aortic backward wave pressure interaction translated into an increase in the independent association between PPc or aortic backward wave pressure and LVMI (P < 0.05 to < 0.001 for comparison of slopes and strength of relations) or LVH (P < 0.05 for comparisons of odds ratios), but not between forward wave pressures and LVMI or LVH across quartiles of time to the peak of the aortic forward wave. A markedly better ability of aortic backward wave pressure and PPc, but not forward wave pressures to detect LVH was noted in the highest as compared with the first three quartiles of time to the peak of the aortic forward wave (P < 0.05). In contrast, reflection time failed to influence the impact of aortic backward wave pressure or PPc on LVMI. Conclusions: Time to the peak of the aortic forward wave, but not early wave reflection markedly influences the impact of aortic backward wave pressure and hence aortic pulse pressure on LVMI and LVH in adults.

Chronic kidney disease epidemiology collaboration-derived glomerular filtration rate performs better at detecting preclinical end-organ changes than alternative equations in black Africans.

Aim: To identify whether the more recently developed equation for estimated glomerular filtration rate (eGFR) [Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI)] is more closely associated with end-organ changes than previous equations in a group of black African descent. Methods: In 1221 randomly recruited participants of black African ancestry in South Africa, we evaluated serum creatinine concentrations, echocardiographic left ventricular mass index (n = 833), carotid-femoral (aortic) pulse wave velocity (PWV) (n = 1053) and carotid intima–media thickness (n = 633). We calculated eGFR from the Jelliffe, five Cockcroft–Gault, Salazar–Corcoran, Modification of Diet in Renal Disease (MDRD) and CKD-EPI equations. Results: After multivariate adjustments, eGFR calculated from all formulae was inversely associated with left ventricular mass index (P < 0.0001) and PWV (P < 0.05 to <0.001), but not with carotid intima–media thickness (P > 0.08). However, although eGFR determined from all equations except Cockcroft–Gault lean body weight or adjusted body weight was independently associated with left ventricular hypertrophy (n = 390 of 833), CKD-EPI-derived eGFR, but not eGFR determined from alternative equations, was independently associated with an increased PWV (n = 88 of 1053). eGFR derived from the CKD-EPI and MDRD equations showed a better performance (area under the receiver operator characteristic curve) for the detection of left ventricular hypertrophy (P < 0.0005) than eGFR determined from alternative equations. Conclusions: In black Africans, eGFR derived from the CKD-EPI equation is better at detecting end-organ measures than eGFR derived from either the MDRD or alternative equations. To enhance risk prediction in black African communities, eGFR calculated from the CKD-EPI equation may be preferred to other equations.

Intrafamilial Aggregation and Heritability of Aortic Reflected (Backward) Waves Derived From Wave Separation Analysis

BACKGROUND Although aortic wave reflection may be inherited, the extent to which indexes of wave reflection derived from wave separation analysis (reflected (backward) wave index (RI) and pressure (Pb)) show intrafamilial aggregation and heritability is uncertain. We therefore aimed to determine the intrafamilial aggregation and heritability of RI and Pb and compare these with indexes of pressure augmentation. METHODS Aortic Pb, RI, augmented pressure (Pa), and augmentation index (AIx) were determined using radial applanation tonometry and SphygmoCor software in 1,152 participants of 315 families (111 father-mother, 705 parent-child, and 301 sibling-sibling pairs) from an urban developing community of black Africans. Heritability estimates were determined from Statistical Analysis for Genetic Epidemiology software. RESULTS With appropriate adjustments, significant correlations were noted between parent-child pairs for Pb and Pa (P < 0.05 for all), but not for RI (P = 0.50) or AIx (P = 0.90) and between sib-sib pairs for Pb and Pa (P < 0.05), but not for RI (P = 0.54) or AIx (P = 0.14). No correlations for indexes of wave reflection were noted between fathers and mothers (P > 0.57). After adjustments, Pb (h2 = 0.24±0.07) and Pa (h2 = 0.23±0.07) (P < 0.001 for both) but not RI (h2 = 0.04±0.06, P = 0.27) or AIx (h2 = 0.10±0.07, P = 0.07) showed significant heritability. CONCLUSIONS Aortic reflected (backward) waves derived from either wave separation (Pb) or pulse wave analysis (Pa) show a similar degree of intrafamilial aggregation and heritability, but the use of RI or AIx may underestimate reflected wave effects.