Effect of sodium and potassium supplementation on vascular and endothelial function: a randomized controlled trial1–3 Natalie Blanch, Peter M Clifton, Kristina S Petersen, and Jennifer B Keogh
Keywords: endothelial function, potassium, sodium, vascular function, flow-mediated dilatation
INTRODUCTION
Endothelial dysfunction is a predictor of cardiovascular disease (CVD),4 and impaired endothelial function has been associated with increased CVD morbidity and mortality in people with and without pre-existing vascular disease (1, 2). Vascular function or arterial stiffness measures, including the pulse wave velocity (PWV) and central augmentation index (AIx), also provide a noninvasive assessment of CVD risk in the general population (3, 4).
Dietary sodium intake is associated with increased risk of stroke and CVD (5, 6). Increased dietary sodium intake is known to impair endothelial function in fasting and postprandial states (7, 8). A reduction of 3-g Na/d intake can improve endothelial function as assessed by using flow-mediated dilatation (FMD) in the fasting state (9). Furthermore, dietary salt loading is known to impair measures of vascular function in hypertensive populations (10). Conversely, high potassium intake is associated with lower CVD risk and lower blood pressure (BP), particularly in hypertensive populations and populations with higher sodium intake (11). Increased dietary potassium improves fasting FMD independent of BP (12), and a high potassium meal can attenuate a postprandial decrease in FMD (13). The effect of high potassium on vascular function measures is unclear (14). The mechanisms for the effect of sodium and potassium on vascular and endothelial function are currently unknown. However, alterations in sodium and potassium intakes are known to alter biological molecules such as adhesion molecules. Increases in dietary sodium increase endothelin-1 expression (9), and increased dietary potassium reduces E-selectin in the short term (12). Berry et al. (15) also observed a reduction in fasting intercellular adhesion molecule-1 (ICAM-1) after a high-potassium diet. The ratio of sodium to potassium has been identified as a more useful predictor of CVD than is sodium or potassium intake alone (16). Follow-up studies on the Trials of Hypertension Prevention I and II named the ratio between sodium and potassium as 1
From the School of Pharmacy and Medical Science, University of South Australia, Adelaide, Australia. 2 JBK is a fellow of the South Australian Cardiovascular Research Development Program funded by the Heart Foundation and the Government of South Australia. PMC is supported by a National Health and Medical Research Council Principal Research Fellowship. This research was jointly funded through these fellowships and the University of South Australia. 3 Address correspondence to J Keogh, School of Pharmacy and Medical Science, University of South Australia, GPO Box 2471, Adelaide, South Australia 5000, Australia. E-mail: [email protected]. 4 Abbreviations used: AIx, augmentation index; BP, blood pressure; CVD, cardiovascular disease; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; HKHN, high-potassium, high-sodium meal; ICAM-1, intercellular adhesion molecule-1; L-FMC, low flow-mediated constriction; LKHN, lowpotassium, high-sodium meal; LKLN, low-potassium, low-sodium meal; PWV, pulse wave velocity; PWVc-f, carotid-femoral pulse wave velocity; SBP, systolic blood pressure. Received December 17, 2014. Accepted for publication February 26, 2015. First published online March 18, 2015; doi: 10.3945/ajcn.114.105197.
Am J Clin Nutr 2015;101:939–46. Printed in USA. Ó 2015 American Society for Nutrition
939
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
ABSTRACT Background: It is known that increased potassium and reduced sodium intakes can improve postprandial endothelial function. However, the effect of increasing potassium in the presence of high sodium in the postprandial state is not known. Objective: We aimed to determine the effect of high potassium and high sodium on postprandial endothelial function as assessed by using flow-mediated dilatation (FMD) and arterial compliance as assessed by using pulse wave velocity (PWV) and central augmentation index (AIx). Design: Thirty-nine healthy, normotensive volunteers [21 women and 18 men; mean 6 SD age: 37 6 15 y; BMI (in kg/m2): 23.0 6 2.8] received a meal with 3 mmol K and 65 mmol Na (low-potassium, high-sodium meal (LKHN)], a meal with 38 mmol K and 65 mmol Na [high-potassium, high-sodium meal (HKHN)], and a control meal with 3 mmol K and 6 mmol Na (low-potassium, low-sodium meal) on 3 separate occasions in a randomized crossover trial. Brachial artery FMD, carotid-femoral PWV, central AIx, and blood pressure (BP) were measured while participants were fasting and at 30, 60, 90, and 120 min after meals. Results: Compared with the LKHN, the addition of potassium (HKHN) significantly attenuated the postmeal decrease in FMD (P-meal by time interaction , 0.05). FMD was significantly lower after the LKHN than after the HKHN at 30 min (P , 0.01). AIx decreased after all meals (P , 0.05). There were no significant differences in AIx, PWV, or BP between treatments over time. Conclusion: The addition of potassium to a high-sodium meal attenuates the sodium-induced postmeal reduction in endothelial function as assessed by FMD. This trial was registered at http://www.anzctr.org.au/ as ACTRN12613000772741. Am J Clin Nutr 2015;101:939–46.
940
BLANCH ET AL.
a significant predicting factor for CVD (17). Combined sodium and potassium interventions showed a benefit of reducing sodium and increasing potassium on blood pressure and vascular function in some populations (18, 19); however, the effect of modifying the ratio of sodium to potassium on postprandial vascular and endothelial function remains to be determined. The primary hypothesis tested in this study was that the addition of potassium to a high-sodium meal would mitigate the adverse effects of sodium on postprandial FMD. A low-potassium, low-sodium control meal was included in the study design to confirm the known adverse effect of high sodium on postprandial FMD. METHODS
Men and women aged between 18 and 70 y were recruited through personal contact and an advertisement on community notice boards (Figure 1). Inclusion criteria were BMI (in kg/m2) $18 and #30, systolic blood pressure (SBP) ,130 mmHg, diastolic blood pressure (DBP) ,90 mmHg, weight stability in the preceding 6 mo, and no use of an antihypertensive or cholesterol-lowering medication, systemic steroids, nonsteroidal anti-inflammatory drugs, or folate supplementation. Participants were not excluded if they were taking any other vitamin supplements provided that the dose was kept constant for the duration of the study. Exclusion criteria were known metabolic disease such as liver or kidney disease, treated hypertension, known or treated high cholesterol, clinical CVD, and inability to comprehend study protocol. Study methods In a randomized, double-blind, crossover design, 39 participants completed the study on 3 mornings separated by a minimum 7-d washout period. Participants consumed 3 test meals that contained 3 mmol K and 6 mmol Na [low-potassium, low-sodium meal (LKLN)], 3 mmol K and 65 mmol Na [low-potassium, high-
FIGURE 1
Diet analysis Participants completed an online dietary survey (Dietary Questionnaire for Epidemiologic Studies; Cancer Council Victoria) before commencing dietary interventions to determine habitual dietary intakes of potassium and other nutrients. Volunteers recorded their food intakes on the day before the first study and replicated these intakes before subsequent test days. Weight and height On arrival, participants had body height (first visit only) measured to the nearest 0.1 cm with a stadiometer (SECA) while barefoot. Body weight was measured to the nearest 0.05 kg by using calibrated electronic digital scales (SECA) while participants wore light clothing and no footwear. FMD As previously described (13), endothelium-dependent FMD of the right brachial artery was measured in the longitudinal plane
Consolidated Standards of Reporting Trials diagram of the flow of subjects through the study.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
Subjects
sodium meal (LKHN)], or 38 mmol K and 65 mmol Na [highpotassium, high-sodium meal (HKHN)]. Meals consisted of tomato soup (Heinz Big Red Condensed Tomato Salt Reduced Soup; Coles) made according to the manufacturer’s instructions. Sodium chloride (SAXA Plain Table Salt; Coles) was added to LKHN and HKHN, and potassium chloride (Paddymelon Gourmet Foods; The Melbourne Food Ingredient Depot) was added to the HKHN. All other macronutrients and micronutrients were identical between meals. After screening, participants were assigned to a treatment order by using an online-generated balanced random-number allocation sequence (http://www.randomization.com) by a person independent of the study. Participants were asked to fast (no food; 250 mL H2O only) from 2200 the night before each visit and refrain from alcohol, caffeine, and vigorous exercise in the 24 h before each study visit.
POTASSIUM REVERSES SODIUM-INDUCED VASOCONSTRICTION
FMD analysis Ultrasound images were recorded at a rate of 30 frames/s with screen-capture software (Debut Video Capture Software Professional V1.82; NCH Software) without QRS gating because previous studies showed that continuous recording has good agreement with R-wave gated measures when edge-detection software is used (22, 23). FMD video files were analyzed with edge-detection software (Brachial Analyzer for Research V6.1.3; Medical Imaging Application LLC). For both baseline and deflation, a region of interest was defined over a clear section of vessel with care to ensure that the region of interest was the same size and position for both baseline and deflation files. The automated edge-detection feature of the software was used to perform a frame-by-frame analysis to generate artery diameter (mm) values for both baseline and deflation. Baseline was defined as the average of 60-s pre-inflation diameter measures. The peak diameter was determined as the maximum diameter postcuff release. With the use of baseline and peak diameters, the percentage change in diameter was calculated (%FMD). The shear rate was calculated from the maximum blood velocity (V) and brachial artery diameter (D) (24) as follows: 4 3 VOD
ð1Þ
Normalized FMD was calculated by dividing the %FMD by the shear rate (21). The intraobserver CV calculated from baseline measurements (n = 39) was 19.6%. Low flow-mediated constriction (L-FMC) was calculated from the mean arterial diameter during the last 30 s before cuff release and expressed as the percentage of change from baseline (25). Arterial compliance Carotid-femoral PWV (PWVc-f) was measured by using a SphygmoCor device (AtCor Medical). Measurements were taken in the supine position by a single operator (CV = 5%) as previously described (26). Central AIx was measured by using the SphygmoCor device. Measurements were taken in the seated position by a single operator. BP Brachial BP was measured by using the SphygmoCor device. Measurements were taken in the seated position by a trained laboratory technician after participants were seated for 2 min. A
series of measurements were taken, each 1 min apart until 4 consistent measurements were obtained (i.e., SBP within a range of 10 mm Hg and DBP within a range of 5 mm Hg). The first reading was discarded, and 3 consistent measurements were averaged (27). Laboratory analysis Blood samples were collected from a cannula inserted into the left brachial vein into tubes with no additive for the measurement of serum sodium and potassium concentrations and ICAM-1, E-selectin, and endothelin-1. Serum was isolated by centrifugation at 2500 3 g for 10 min, and samples were stored at 2808C until analyzed. Biochemical assays were performed in a single assay on completion of the study. ICAM-1, E-selectin, and endothelin-1 were measured by using Quantikine human immunoassay kits (R&D Systems; Bioscientific) on the basis of the manufacturer’s protocols. An analysis of serum sodium and potassium concentrations was performed at the Institute of Medical and Veterinary Sciences, Adelaide, South Australia. Ethics This study was approved by the University of South Australia’s Human Research Ethics Committee (ethics approval number: 0000031547). All participants gave written informed consent. This trial was registered at http://www.anzctr.org.au/ as ACTRN12613000772741. Statistical analysis The primary outcome for the statistical analysis was the effect of HKHN compared with LKHN on FMD. The secondary outcome was the effect of HKHN compared with LKHN on PWV. On the basis of power calculations from our previous study (13) with 80% power and P , 0.05 to detect a minimum change in FMD of 1.3% absolute, 40 people were required in a crossover study. With 80% power and P , 0.05 to detect a 10% (0.6-m/s) change in PWVc-f and the assumption of a within-patient SD of 0.9 m/s, 20 people were required in a crossover study. All analyses were performed with SPSS 21 for Windows software (SPSS Inc.). Significance was set at P , 0.05. Data were tested for the normality of distribution and to ensure that residuals had approximately constant SDs by using the Kolmogorov-Smirnov test, Q-Q plots, and histograms. An ANOVA with repeated measures (with diet as the within-subject factor) was used to analyze outcomes (with and without covariates including diet order and BP), and we performed a post hoc analysis to assess primary and secondary comparisons. For FMD only, differences from baseline were calculated before the analysis to make meal comparisons easier to see. The primary comparison was HKHN compared with LKHN as reflected by the hypothesis. A secondary comparison was included to show a difference in LKHN and the LKLN control to show the adverse effect of sodium on postprandial FMD. Age, BMI, sex, and habitual potassium intake were also included as covariates. Preliminary analyses were performed to ensure no violation of assumptions of linearity. Pearson correlation analyses were conducted to assess the association of change
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
above the antecubital fossa with an 8.8-MHz linear array transducer (MySono U6; Samsung Medison) according to published guidelines (20, 21). Subjects attended after an overnight fast and lay quietly for 5 min in a quiet, temperature-controlled room before endothelium-dependent FMD measurements were obtained (21). The brachial artery diameter was measured before and after forearm ischemia caused by inflation of a sphygmomanometer cuff applied to the right forearm 2 cm below the olecranon process to 200 mmHg for 5 min. Continuous longitudinal 2-dimensional images of the brachial artery were obtained and digitally recorded during quite rest (1 min) and from 30 s before cuff release and during reactive hyperemia (3 min). All images were stored for offline analysis by a single trained observer who was unaware of the treatments at the time of measurement.
941
942
BLANCH ET AL.
between variables. Data are expressed as means 6 SDs or median (IQRs) as appropriate. RESULTS
Subjects Thirty-nine subjects completed the study, and baseline characteristics are outlined in Table 1. DBP before the LKHN was significantly lower than before the LKLN or the HKHN (P = 0.01). There were no significant differences between other fasting variables between treatments. Brachial artery endothelial function
TABLE 1 Fasting characteristics of participants1
BP SBP and DBP increased after all meals (P , 0.01 for both) (Table 2). There were no significant differences in SBP after each of the test meals and a borderline change in DBP (P-meal 3 time = 0.05). No effect of age, BMI, sex, or habitual potassium intake was observed. Mean arterial pressure increased after all meals (P = 0.03). There were no significant differences in mean arterial pressure after each of the meals.
Arterial compliance measures PWV c-f increased after all meals compared with fasting (P-time effect , 0.01). There were no significant differences in PWVc-f after each of the meals (P-meal 3 time interaction = 0.29 for the HKHN compared with the LKHN). Central AIx and AIx corrected for heart rate at 75 beats/min (n = 34) decreased after each meal (P-time , 0.01 for both). There was no meal 3 time interaction (HKHN compared with LKHN) for AIx (P = 0.20) or AIx corrected for heart rate at 75 beats/ min (P = 0.10). Heart rate decreased after all meals (P-time , 0.01). There were no significant differences in heart rate after each of the meals. Pulse pressure increased after all meals (P-time , 0.01). There were no significant differences in pulse pressure after each of the meals (P-meal 3 time interaction = 0.53).
Baseline Age, y BMI, kg/m2 HR, beats/min SBP, mm Hg DBP, mm Hg MAP, mm Hg Brachial artery diameter, mm FMD, % PWVc-f, m/s AIx, % AIx@75beats/min,2 % PP, mm Hg Serum sodium, mmol/L Serum potassium, mmol/L ICAM-1, mg/L E-selectin, mg/L Endothelin-1, pg/mL
38 23.1 57 115 71 83 3.80 9.5 6.7 12 7 31 140 4.2 131.5 38.3 1.34
6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
16 2.9 9 8 6 7 0.66 3.5 1.1 15 16 4 1 0.2 38.1 22.1 0.36
1 All values are means 6 SDs. n = 39 (21 women and 18 men). AIx, augmentation index; AIx@75beats/min, augmentation index corrected to heart rate at 75 beats/min; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; HR, heart rate; ICAM-1, intracellular adhesion molecule-1; MAP, mean arterial blood pressure; PP, pulse pressure; PWVc-f, carotid-femoral pulse wave velocity; SBP, systolic blood pressure. 2 n = 34 (18 women and 16 men).
Biochemical analysis Serum sodium increased significantly after the LKHN (maximum change was 1.2 6 0.9 mmol at 60 min) and the HKHN compared with the LKLN (P-meal 3 time interaction , 0.01 for both comparisons). There was no difference in serum sodium between the LKHN and the HKHN (P = 0.59). Serum potassium significantly increased after the HKHN (maximum change was 0.6 6 0.3 mmol at 90 min) compared with the LKHN and the control meal (P-meal 3 time interaction , 0.01 for both comparisons). Endothelin-1 decreased after both meals (P-time , 0.01) but was higher after the HKHN than after the LKLN (P-meal 3 time interaction = 0.03). ICAM-1 decreased after both meals (P-time , 0.01) but decreased less after the HKHN than after the LKHN. There were no changes in serum E-selectin after any of the interventions. FMD and serum sodium was negatively correlated at 60 min with the LKHN (r = 20.45, P = 0.01). There were no correlations between FMD and serum potassium, adhesion molecules, or maximum flow. There was a weak inverse correlation between FMD and SBP and DBP at various time points but only with the LKHN diet.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
There was no meal 3 time interaction (P = 0.06) but a significant effect of meal (P , 0.01) for the comparison of the 3 meals (Table 2). The addition of potassium in the HKHN significantly attenuated the postprandial decrease in FMD after the LKHN (P-meal 3 time interaction = 0.02). There was no significant meal 3 time interaction (P = 0.65) or time effect (P = 0.59) for the FMD change from baseline (time 0) between the 3 meals (Figure 2). There was a significant effect of meal (P , 0.01), and FMD decreased by 21.5% from baseline after the LKLN and was further decreased to 22.9% from baseline after the LKHN. When the change in FMD from baseline was compared for the primary hypothesis (LKHN compared with HKHN), there was no meal 3 time interaction (P = 0.57), but the meal effect was significant (P , 0.01). The maximum difference in FMD after the LKHN compared with the HKHN was at 30 min (23.58 6 0.85%; P , 0.01). There was no effect of age, sex, BMI, or habitual potassium intake observed. When FMD was normalized for the shear rate, the meal effect did not change, but meal by time interaction weakened and became
nonsignificant (P-meal effect , 0.01; P-meal 3 time interaction = 0.12 for the LKHN compared with the HKHN). The low flow constriction observed with the LKHN (Table 2) was completely abolished by the HKHN (P-meal 3 time interaction , 0.01) with a difference of 24.95 6 0.87% and P , 0.01 at 60 min. There was no effect of age, BMI, sex, or habitual potassium intake observed.
60
90
120
0
30
60
90
120
0
30
60
HKHN (min)
90
120
6 6 6 6 6 6 6 6 6
0.7 2.1 1 1 1 1 0.2 3 3.2
6 6 6 6 6 6 6 6 6 0.6 2.6 1 2 1 1 0.2 3 3.1
4.4 6 0.05
30 6 1 140 6 0.2
20.1 43.3 53 115 71 82 6.6 6 21
6 6 6 6 6 6 6 6 6 0.9 2.8 1 1 1 1 0.2 3 2.9
4.4 6 0.04
31 6 1 140 6 0.2
20.9 44.1 54 116 70 82 6.7 6 23
6 6 6 6 6 6 6 6 6 0.5 2.4 1 1 1 1 0.2 3 2.8
4.4 6 0.04
31 6 1 140 6 0.2
1.1 49.5 54 117 72 83 6.8 6 23
6 6 6 6 6 6 6 6 6 0.8 2.7 1 1 1 1 0.2 3 3.0
4.2 6 0.05
30 6 1 140 6 0.2
20.9 48.2 58 112 69 81 6.6 10 6
6 6 6 6 6 6 6 6 6 0.8 2.9 1 1 1 1 0.2 3 3.1
4.3 6 0.04
30 6 1 141 6 0.2
21.6 48.6 55 114 71 83 6.6 7 0
6 6 6 6 6 6 6 6 6 0.8 2.7 1 1 1 1 0.2 3 3.1
4.3 6 0.04
30 6 1 141 6 0.2
23.5 48.0 54 115 72 83 6.7 7 22
6 6 6 6 6 6 6 6 6 0.7 2.8 1 1 1 1 0.2 3 3.3
4.4 6 0.04
31 6 1 141 6 0.2
22.0 46.5 54 116 71 83 6.7 6 22 6 6 6 6 6 6 6 6 6 0.8 2.9 1 2 1 1 0.2 3 3.0
4.4 6 0.04
32 6 1 141 6 0.2
21.8 49.7 54 117 72 84 6.8 6 22
6 6 6 6 6 6 6 6 6 0.7 2.4 1 1 1 1 0.2 3 3.1
4.3 6 0.05
30 6 1 140 6 0.3
20.7 43.3 57 114 71 83 6.5 11 6 6 6 6 6 6 6 6 6 6 0.7 2.2 1 2 1 1 0.2 3 3.0
4.7 6 0.05
30 6 1 141 6 0.2
0.7 46.1 55 115 72 84 6.6 10 4
6 6 6 6 6 6 6 6 6
1.0 2.7 1 2 1 1 0.2 3 3.2
4.9 6 0.06
0.3 49.3 53 116 71 82 6.7 8 0
4.8 6 0.05
0.8 2.9 1 1 1 1 0.2 3 3.1
31 6 1 141 6 0.2
6 6 6 6 6 6 6 6 6
31 6 1 141 6 0.2
1.5 50.2 54 115 71 83 6.5 9 1
6 6 6 6 6 6 6 6 6
0.7 3.1 1 1 1 1 0.2 3 3.1
0.03 0.06
0.53 0.50 ,0.01 ,0.01 ,0.01 ,0.01
0.09 ,0.01 0.08 0.11 0.04 ,0.01 0.35 0.62 ,0.01 0.49 0.93 ,0.01 0.05 0.90 ,0.01 0.15 0.70 0.03 0.48 0.82 ,0.01 0.39 ,0.01 ,0.01 0.10 ,0.01 ,0.01
0.06 ,0.01 0.08 0.18
0.54 0.59
0.01 0.08 0.33 0.19 0.01 0.03 0.29 0.20 0.10
0.02 0.03
0.87 ,0.01 0.38 0.40 0.01 ,0.01
0.03 0.94 0.03
4.8 6 0.05 ,0.01 ,0.01 ,0.01 ,0.01
32 6 1 141 6 0.2
2.1 48.5 53 117 72 84 6.8 9 1
P Meal 3 Meal Time Meal 3 time2 effect2 effect2 time3
128.0 6 5.1 116.4 6 4.8 118.5 6 4.8 122.1 6 4.7 119.4 6 4.6 133.1 6 6.4 117.3 6 4.7 116.5 6 4.8 115.3 6 4.5 117.8 6 4.7 128.2 6 5.5 120.7 6 5.4 118.6 6 5.4 118.6 6 5.0 115.3 6 5.0 ,0.01 0.60 33.1 6 4.1 37.5 6 3.9 36.5 6 3.9 37.0 6 3.8 34.4 6 3.3 38.5 6 3.7 40.9 6 3.1 39.7 6 3.3 37.6 6 3.1 39.8 6 3.1 37.9 6 3.5 38.5 6 3.9 38.5 6 3.8 36.3 6 3.3 37.6 6 3.3 1.36 6 0.07 1.23 6 0.06 1.36 6 0.07 1.46 6 0.07 1.47 6 0.07 1.45 6 0.08 1.08 6 0.06 1.20 6 0.05 1.28 6 0.05 1.39 6 0.06 1.26 6 0.06 1.14 6 0.05 1.13 6 0.05 1.24 6 0.06 1.22 6 0.05 0.02
4.3 6 0.04
20.4 44.6 55 115 71 83 6.6 7 21
4.2 6 0.04
0.8 2.4 1 1 1 1 0.2 2 2.6
30 6 1 139 6 0.2
6 6 6 6 6 6 6 6 6
30 6 1 140 6 0.2
21.5 41.2 58 113 70 82 6.5 10 6
8.6 6 0.6 7.5 6 0.7 7.3 6 0.6 7.4 6 0.6 7.1 6 0.6 8.9 6 0.5 6.2 6 0.6 5.8 6 0.4 6.0 6 0.5 7.3 6 0.6 8.8 6 0.5 9.7 6 0.7 9.0 6 0.7 8.8 6 1.0 9.3 6 0.8 3.89 6 0.11 3.91 6 0.12 3.90 6 0.11 3.89 6 0.11 3.84 6 0.11 3.86 6 0.10 3.90 6 0.11 3.93 6 0.11 3.90 6 0.11 3.84 6 0.11 3.89 6 0.11 3.78 6 0.11 3.81 6 0.12 3.79 6 0.12 3.76 6 0.12
30
LKHN (min)
1
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
All values are means 6 SEMs. n = 39 (21 women and 18 men). AIx, augmentation index; AIx@75beats/min, augmentation index corrected to heart rate at 75 beats/min; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; HKHN, high-potassium, high-sodium meal; HR, heart rate; ICAM-1, intracellular adhesion molecule-1; L-FMC, low flow-mediated constriction; LKHN, low-potassium, highsodium meal; LKLN, low-potassium, low-sodium meal; MAP, mean arterial blood pressure; PP, pulse pressure; PWVc-f, carotid-femoral pulse wave velocity; SBP, systolic blood pressure. 2 Repeated-measures ANOVA for 3 meals. 3 Repeated-measures ANOVA for LKHN compared with HKHN. 4 Repeated-measures ANOVA for baseline measurements, P = 0.01. 5 n = 34 (18 women and 16 men).
FMD, % Brachial artery diameter, mm L-FMC, % Shear rate, s21 HR, beats/min SBP, mm Hg DBP,4 mm Hg MAP, mm Hg PWVc-f, m/s AIx, % AIx@75beats/ min,5 % PP, mm Hg Serum sodium, mmol/L Serum potassium, mmol/L ICAM-1, mg/L E-selectin, mg/L Endothelin-1, pg/mL
0
LKLN (min)
TABLE 2 All outcome variables for each meal intervention1
POTASSIUM REVERSES SODIUM-INDUCED VASOCONSTRICTION
943
944
BLANCH ET AL.
DISCUSSION
The main outcome from this study was that the addition of potassium to a high-sodium meal significantly attenuated a sodium-invoked postmeal reduction in endothelial function as assessed by FMD. To our knowledge, this is the first time that combined sodium and potassium has been examined in the postprandial state. The amount of added sodium used in the meals was equivalent to w2 rashers of bacon [or 2 thick sausages (80 g each)]. The amount of added potassium was equivalent to 3 large bananas [or 2 large potatoes (340 g cooked weight) or 0.75 cups dried apricots]. Dickinson et al. (7) reported that a high-sodium meal impaired brachial artery FMD in the postprandial state, which was a finding that was replicated in this study (Figure 2). These findings were also consistent with the deleterious effect of sodium on fasting FMD in the short term (8, 9, 28). We showed that impairment in FMD was correlated with increases in SBP and DBP after the high-sodium meal without potassium. This effect was not seen in the postprandial study by Dickinson et al. (7). Furthermore, we previously showed that a high-potassium meal (without sodium) can attenuate postmeal responses in FMD to a low-potassium meal (13). Previous studies that investigated the effect of increasing potassium on fasting FMD have reported mixed results (12, 15, 29); however, dose appears to play an important role with potassium increases .40 mmol improving endothelial function (14). The current study suggests a high potassium load attenuates postmeal decreases in FMD irrespective of the sodium concentration present. L-FMC increased after the high-sodium meal and decreased after the addition of potassium to the meal. L-FMC measures the endothelial control of vascular tone at rest (30). L-FMC mechanisms are not mediated by nitric oxide release and are thought to be different from those of FMD (30). FMD does not correlate with L-FMC measures, and the clinical relevance of changes in L-FMC remains unknown (25). However, L-FMC has been shown to be attenuated by 3 mo of statin
treatment (31). The assessment of L-FMC combined with FMD provides complimentary information about vascular homeostasis in addition to traditional risk factors (30). We observed no significant differences in SBP between meals. BP increased after all meals. This result was unexpected because previous studies that used the same control meal in similar populations showed a small postmeal decrease in BP at 30 min (13). However, postmeal increases in BP that provoked increases in PWVc-f in response to a high-fat meal were previously described (32). Our study showed no change in PWVc-f between meals; however, we saw a small but significant trend for increased PWVc-f over time. Previous studies showed an improvement in fasting PWVc-f in hypertensive (10, 33, 34) but not normotensive (35) populations with reduced sodium intake. The influence of potassium on vascular function measures is unclear, with BP likely to influence any observed effect (14). Our study showed a significant reduction in central AIx in response to each of the test meals. This result remained after correction for heart rate (augmentation index corrected to heart rate at 75 beats/min). Previous studies showed mixed results in AIx in response to a mixed meal with some studies that showed an increase in AIx (36) and other studies that showed a decrease (37). Furthermore, Dickinson et al. (38) reported a further increase in AIx after a high-sodium meal. Our study showed no meal 3 time interaction for AIx. Previous studies showed no effect of potassium interventions on fasting AIx (15, 39, 40). One salt-substitution study (simultaneously reducing sodium and increasing potassium) showed significant improvements in measures of arterial stiffness (central pulse pressure and pulse wave reflection time) after 12 mo of intervention but did not see an effect on AIx (41). This study showed that a meal that contained 38 mmol K raised serum potassium by 0.6 mmol/L in healthy individuals. We hypothesized that this increase in serum potassium affects the FMD response by one of 2 mechanisms. First, we speculate that
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
FIGURE 2 Mean (6SEM) changes in brachial artery FMD from baseline in response to the consumption of test meals that contained 3 mmol K and 65 mmol Na (LKHN) compared with 38 mmol K and 65 mmol Na (HKHN) and 3 mmol K and 6 mmol Na (LKLN). n = 39 (18 men and 21 women). For the change in FMD from baseline (repeated-measures ANOVA): P-meal 3 time = 0.65; P-meal effect , 0.01; P-time effect = 0.59. For change in FMD from baseline: repeated-measures ANOVA): LKHN compared with HKHN P-meal 3 time interaction = 0.57 (P-meal effect , 0.01; P-time effect = 0.35); LKLN compared with LKHN P-meal 3 time interaction = 0.33 (P-meal effect = 0.01; P-time effect = 0.66). FMD, flow-mediated dilatation; HKHN, high-potassium, high-sodium meal; LKHN, low-potassium, high-sodium meal; LKLN, low-potassium, low-sodium meal.
POTASSIUM REVERSES SODIUM-INDUCED VASOCONSTRICTION
a mechanistic explanation for the effect of potassium and sodium on endothelial function; however, this study provides some insight into the changes in endothelial mediators that occur in response to high-potassium and -sodium loads. In conclusion, this study showed that the amount of potassium present in 3 servings of fruit can significantly attenuate reductions in endothelial function induced by a high-sodium (65-mmol) meal. A transient increase in serum potassium is a plausible mechanism for this response; however, increased potassium does not appear to affect other measures of vascular function. This findings suggests that, in populations in whom reducing sodium intake may be problematic (e.g., primarily of processed foods), increasing potassium intake may reduce CVD risk. Although the long-term implications of this result in the pathogenesis of atherosclerosis and CVD is unknown, this study provides additional evidence that increases in dietary potassium should be encouraged, particularly in the presence of a high-sodium westernized diet. The authors’ responsibilities were as follows—JBK and PMC: developed the hypotheses tested in the study, designed the research (project conception, development of overall research plan, and study oversight), and contributed to statistical analyses and interpretation of data; NB: contributed to the study design, planned and conducted the study, performed vascular measurements (hands-on conduct of experiments and data collection), performed initial statistical analyses, and drafted the manuscript; KSP: assisted in conducting the study and performed some vascular measurements (hands-on conduct of experiments and data collection); and all authors: critically reviewed the manuscript. None of the authors reported a conflict of interest related to the study.
REFERENCES 1. Kitta Y, Obata J-e, Nakamura T, Hirano M, Kodama Y, Fujioka D, Saito Y, Kawabata K-i, Sano K, Kobayashi T, et al. Persistent impairment of endothelial vasomotor function has a negative impact on outcome in patients with coronary artery disease. J Am Coll Cardiol 2009;53:323–30. 2. Heitzer T, Schlinzig T, Krohn K, Meinertz T, Mu¨nzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation 2001;104:2673–8. 3. McEniery CM, Yasmin, Hall IR, Qasem A, Wilkinson IB, Cockcroft JR. Normal vascular aging: differential effects on wave reflection and aortic pulse wave velocity: the Anglo-Cardiff Collaborative Trial (ACCT). J Am Coll Cardiol 2005;46:1753–60. 4. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol 2010;55: 1318–27. 5. Strazzullo P, D’Elia L, Kandala N-B, Cappuccio FP. Salt intake, stroke, and cardiovascular disease: meta-analysis of prospective studies. BMJ 2009;339:b4567. 6. He J, Ogden LG, Vupputuri S, Bazzano LA, Loria C, Whelton PK. Dietary sodium intake and subsequent risk of cardiovascular disease in overweight adults. JAMA 1999;282:2027–34. 7. Dickinson KM, Clifton PM, Keogh JB. Endothelial function is impaired after a high-salt meal in healthy subjects. Am J Clin Nutr 2011; 93:500–5. 8. Dickinson KM, Keogh JB, Clifton PM. Effects of a low-salt diet on flow-mediated dilatation in humans. Am J Clin Nutr 2009;89:485–90. 9. Dickinson KM, Clifton PM, Keogh JB. A reduction of 3 g/day from a usual 9 g/day salt diet improves endothelial function and decreases endotheilin-1 in a randomised cross_over study in normotensive overweigh and obese subjects. Atherosclerosis 2014;233:32–8. 10. Todd AS, MacGinley RJ, Schollum JB, Johnson RJ, Williams SM, Sutherland WH, Mann JI, Walker RJ. Dietary salt loading impairs arterial vascular reactivity. Am J Clin Nutr 2010;91:557–64. 11. Aburto NJ, Hanson S, Gutierrez H, Hooper L, Elliott P, Cappuccio FP. Effect of increased potassium intake on cardiovascular risk factors and disease: systematic review and meta-analyses. BMJ 2013;346:f1378.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
the rise in serum potassium may directly affect the small and intermediate calcium-activated potassium channels of endothelial cells, which are then electrically coupled to the smooth cells (i.e., an endothelium-derived hyperpolarizing factor effect) (42). Second, increased serum potassium may increase nitric oxide release because cellular studies showed an increase in release of nitrites (an index of nitric oxide release) in response to increased extracellular potassium (43). Serum sodium was raised by a maximum of 1.2 mmol/L after the 65-mmol sodium meals. This change in serum sodium was expected because previous studies showed an increase in serum sodium in response to a 65-mmol sodium meal (38). Dickinson et al. (7) hypothesized that the mechanism for the observed decrease in FMD after a high-sodium meal is increased serum sodium resulting in a reduction in nitric oxide bioavailability. Evidence from in vitro studies suggested increases of 5–10 mmol in serum sodium reduce nitric oxide bioavailability and nitric oxide synthase activity (44, 45). We did not measure serum concentrations of nitrate and nitrite in this study because previous studies did not detect changes in nitrate and nitrite concentrations after a 65-mmol Na meal (38). In our study, we observed a relation between BP and FMD after the high-sodium meal that was not seen in the study by Dickinson et al. (7); therefore, the rise in BP could be the mechanism for the reduction in FMD. We observed small but significant differences in serum endothelin-1 and ICAM-1 with diminished postmeal falls with the addition of potassium to the meal, but baseline values were higher with the high-sodium meal, and this finding alone may account for the difference. Endothelin-1 was previously shown to decrease after a reduced sodium diet (9). Furthermore, Liu et al. (46) showed that endothelin-1 decreased after a 7-d low-sodium diet (51.3 mmol/d) and increased after a 7-d high sodium diet (307.7 mmol/d), and the addition of potassium (60 mmol/d for 7 d) attenuated the response to the high-sodium diet. Conversely, Dickinson et al. (38) failed to detect a difference in postprandial endothelin-1 concentrations between a low-sodium meal and a high-sodium meal alone (5 compared with 65 mmol). Previously, reductions in fasting ICAM-1 were reported after a 6-wk high-potassium diet in woman only (15), but we did not see a sexrelated treatment effect in our study. The role of variations in endothelin-1 and ICAM-1 in alterations in endothelial function in the postprandial state are unknown, but they are unrelated to changes in FMD. Limitations of this study included the use of a relatively healthy population sample, which might have limited the potential to see improvements in BP and vascular variables. However, we saw significant improvements in measures of endothelial function after the addition of potassium to the highsodium meal. Furthermore, it is unclear whether these results persist in the longer term, and the effect of high potassium in the presence of high sodium on fasting FMD warrants additional investigation. We acknowledge that, in populations with renal impairment, there may be risk of hyperkalemia by increasing dietary potassium, and in our study, serum potassium remained elevated at 120 min. Another limitation was that we did not adjust for any effect of menstrual cycles of premenopausal women. However, we did not see an effect of sex and did not observe any differences in baseline FMD measurements between treatments; therefore, the menstrual cycle does not appear to have affected our data. It was a limitation that we are unable to provide
945
946
BLANCH ET AL. 30. Poredos P, Jezovnik MK. Testing endothelial function and its clinical relevance. J Atheroscler Thromb 2013;20:1–8. 31. Megnien JL, Simon A, Andriani A, Segond P, Jeannin S, Levenson J. Cholesterol lowering therapy inhibits the low-flow mediated vasoconstriction of the brachial artery in hypercholesterolaemic subjects. Br J Clin Pharmacol 1996;42:187–93. 32. Lithander FE, Herlihy LK, Walsh DM, Burke E, Crowley V, Mahmud A. Postprandial effect of dietary fat quantity and quality on arterial stiffness and wave reflection: a randomised controlled trial. Nutrition Journal 2013 Jul 10 (Epub ahead of print; DOI: 10.1186/1475-2891-1293). 33. Jablonski KL, Fedorova OV, Racine ML, Geolfos CJ, Gates PE, Chonchol M, Fleenor BS, Lakatta EG, Bagrov AY, Seals DR. Dietary sodium restriction and association with urinary marinobufagenin, blood pressure, and aortic stiffness. Clin J Am Soc Nephrol 2013;8: 1952–9. 34. He FJ, Marciniak M, Visagie E, Markandu ND, Anand V, Dalton RN, MacGregor GA. Effect of modest salt reduction on blood pressure, urinary albumin, and pulse wave velocity in white, black, and Asian mild hypertensives. Hypertension 2009;54:482–8. 35. Todd AS, Macginley RJ, Schollum JB, Williams SM, Sutherland WH, Mann JI, Walker RJ. Dietary sodium loading in normotensive healthy volunteers does not increase arterial vascular reactivity or blood pressure. Nephrology (Carlton) 2012;17:249–56. 36. Ahuja KD, Robertson IK, Ball MJ. Acute effects of food on postprandial blood pressure and measures of arterial stiffness in healthy humans. Am J Clin Nutr 2009;90:298–303. 37. Taylor JL, Curry TB, Matzek LJ, Joyner MJ, Casey DP. Acute effects of a mixed meal on arterial stiffness and central hemodynamics in healthy adults. Am J Hypertens 2014;27:331–7. 38. Dickinson KM, Clifton PM, Burrell LM, Barrett PHR, Keogh JB. Postprandial effects of a high salt meal on serum sodium, arterial stiffness, markers of nitric oxide production and markers of endothelial function. Atherosclerosis 2014;232:211–6. 39. Graham UM, McCance D, Young I, Mullan K. A randomised controlled trial evaluating the effect of potassium supplementation on vascular function and the renin-angiotensin-aldosterone system. J Hum Hypertens 2014;28:333–9. 40. Matthesen SK, Larsen T, Vase H, Lauridsen TG, Pedersen EB. Effect of potassium supplementation on renal tubular function, ambulatory blood pressure and pulse wave velocity in healthy humans. Scand J Clin Lab Invest 2012;72:78–86. 41. Hu J, Jiang X, Li N, Yu X, Perkovic V, Chen B, Zhao L, Neal B, Wu Y. Effects of salt substitute on pulse wave analysis among individuals at high cardiovascular risk in rural China: a randomized controlled trial. Hypertens Res 2009;32:282–8. 42. Coleman HA, Tare M, Parkington HC. Endothelial potassium channels, endothelium-dependent hyperpolarization and the regulation of vascular tone in health and disease. Clin Exp Pharmacol Physiol 2004;31: 641–9. 43. Oberleithner H, Callies C, Kusche-Vihrog K, Schillers H, Shahin V, Riethmuller C, Macgregor GA, de Wardener HE. Potassium softens vascular endothelium and increases nitric oxide release. Proc Natl Acad Sci U S A 2009;106:2829–34. 44. Fujiwara N, Osanai T, Kamada T, Katoh T, Takahashi K, Okumura K. Study on the relationship between plasma nitrite and nitrate level and salt sensitivity in human hypertension: modulation of nitric oxide synthesis by salt intake. Circulation 2000;101:856–61. 45. Oberleithner H, Riethmu¨ller C, Schillers H, MacGregor GA, de Wardener HE, Hausberg M. Plasma sodium stiffens vascular endothelium and reduces nitric oxide release. Proc Natl Acad Sci USA 2007;104:16281–6. 46. Liu Z, Peng J, Lu F, Zhao Y, Wang S, Sun S, Zhang H, Diao Y. Salt loading and potassium supplementation: effects on ambulatory arterial stiffness index and endothelin-1 levels in normotensive and mild hypertensive patients. J Clin Hypertens (Greenwich) 2013;15:485–96.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
12. Blanch N, Clifton P, Petersen K, Willoughby S, Keogh J. Effect of high potassium diet on endothelial function. Nutr Metab Cardiovasc Dis 2014;24:983–9. 13. Blanch N, Clifton PM, Keogh JB. Postprandial effects of potassium supplementation on vascular function and blood pressure: a randomised cross-over study. Nutr Metab Cardiovasc Dis 2014;24:148–54. 14. Blanch N, Clifton PM, Keogh JB. A systematic review of vascular and endothelial function: Effects of fruit, vegetable and potassium intake. Nutr Metab Cardiovasc Dis 2014 Oct 15 (Epub ahead of print; DOI: 10.1016/j.numecd.2014.10.001). 15. Berry SE, Mulla UZ, Chowienczyk PJ, Sanders TA. Increased potassium intake from fruit and vegetables or supplements does not lower blood pressure or improve vascular function in UK men and women with early hypertension: a randomised controlled trial. Br J Nutr 2010; 104:1839–47. 16. Intersalt: an international study of electrolyte excretion and blood pressure. Results for 24 hour urinary sodium and potassium excretion. Intersalt Cooperative Research Group. BMJ 1988;297:319–28. 17. Cook NR, Obarzanek E, Cutler JA, Buring JE, Rexrode KM, Kumanyika SK, Appel LJ, Whelton PK, for the Trials of Hypertension Prevention Collaborative Research Group. Joint effects of sodium and potassium intake on subsequent cardiovascular disease: the trials of Hypertension Prevention Follow-up Study. Arch Intern Med 2009;169:32–40. 18. Zhou B, Wang HL, Wang WL, Wu XM, Fu LY, Shi JP. Long-term effects of salt substitution on blood pressure in a rural North Chinese population. J Hum Hypertens 2013;27:427–33. 19. Gilleran G, O’leary M, Bartlett W, Vinall H, Jones A, Dodson P. Effects of dietary sodium substitution with potassium and magnesium in hypertensive type II diabetics: a randomised blind controlled parallel study. J Hum Hypertens 1996;10:517–21. 20. Harris RA, Nishiyama SK, Wray DW, Richardson RS. Ultrasound assessment of flow-mediated dilation. Hypertension 2010;55:1075–85. 21. Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D, Charbonneau F, Creager MA, Deanfield J, Drexler H, Gerhard-Herman M, Herrington D. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery* 1: a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol 2002; 39:257–65. 22. Kizhakekuttu TJ, Gutterman DD, Phillips SA, Jurva JW, Arthur EIL, Das E, Widlansky ME. Measuring FMD in the brachial artery: how important is QRS gating? J Appl Physiol (1985) 2010;109:959–65. 23. Gemignani V, Bianchini E, Faita F, Giannarelli C, Plantinga Y, Ghiadoni L, Demi M. Ultrasound measurement of the brachial artery flowmediated dilation without ECG gating. Ultrasound Med Biol 2008; 34:385–91. 24. Parker BA, Trehearn TL, Meendering JR. Pick your Poiseuille: normalizing the shear stimulus in studies of flow-mediated dilation. J Appl Physiol 2009;107:1357–9. 25. Spiro JR, Digby JE, Ghimire G, Mason M, Mitchell AG, Ilsley C, Donald A, Dalby MC, Kharbanda RK. Brachial artery low-flowmediated constriction is increased early after coronary intervention and reduces during recovery after acute coronary syndrome: characterization of a recently described index of vascular function. Eur Heart J 2011;32:856–66. 26. Hwang MH, Yoo JK, Kim HK, Hwang CL, Mackay K, Hemstreet O, Nichols WW, Christou DD. Validity and reliability of aortic pulse wave velocity and augmentation index determined by the new cuff-based SphygmoCor Xcel. J Hum Hypertens 2014;28:475–81. 27. Ogedegbe G, Pickering T. Principles and techniques of blood pressure measurement. Cardiol Clin 2010;28:571–86. 28. Jablonski KL, Racine ML, Geolfos CJ, Gates PE, Chonchol M, McQueen MB, Seals DR. Dietary sodium restriction reverses vascular endothelial dysfunction in middle-aged/older adults with moderately elevated systolic blood pressure. J Am Coll Cardiol 2013;61:335–43. 29. He FJ, Marciniak M, Carney C, Markandu ND, Anand V, Fraser WD, Dalton RN, Kaski JC, MacGregor GA. Effects of potassium chloride and potassium bicarbonate on endothelial function, cardiovascular risk factors, and bone turnover in mild hypertensives. Hypertension 2010;55:681–8.
Keywords: endothelial function, potassium, sodium, vascular function, flow-mediated dilatation
INTRODUCTION
Endothelial dysfunction is a predictor of cardiovascular disease (CVD),4 and impaired endothelial function has been associated with increased CVD morbidity and mortality in people with and without pre-existing vascular disease (1, 2). Vascular function or arterial stiffness measures, including the pulse wave velocity (PWV) and central augmentation index (AIx), also provide a noninvasive assessment of CVD risk in the general population (3, 4).
Dietary sodium intake is associated with increased risk of stroke and CVD (5, 6). Increased dietary sodium intake is known to impair endothelial function in fasting and postprandial states (7, 8). A reduction of 3-g Na/d intake can improve endothelial function as assessed by using flow-mediated dilatation (FMD) in the fasting state (9). Furthermore, dietary salt loading is known to impair measures of vascular function in hypertensive populations (10). Conversely, high potassium intake is associated with lower CVD risk and lower blood pressure (BP), particularly in hypertensive populations and populations with higher sodium intake (11). Increased dietary potassium improves fasting FMD independent of BP (12), and a high potassium meal can attenuate a postprandial decrease in FMD (13). The effect of high potassium on vascular function measures is unclear (14). The mechanisms for the effect of sodium and potassium on vascular and endothelial function are currently unknown. However, alterations in sodium and potassium intakes are known to alter biological molecules such as adhesion molecules. Increases in dietary sodium increase endothelin-1 expression (9), and increased dietary potassium reduces E-selectin in the short term (12). Berry et al. (15) also observed a reduction in fasting intercellular adhesion molecule-1 (ICAM-1) after a high-potassium diet. The ratio of sodium to potassium has been identified as a more useful predictor of CVD than is sodium or potassium intake alone (16). Follow-up studies on the Trials of Hypertension Prevention I and II named the ratio between sodium and potassium as 1
From the School of Pharmacy and Medical Science, University of South Australia, Adelaide, Australia. 2 JBK is a fellow of the South Australian Cardiovascular Research Development Program funded by the Heart Foundation and the Government of South Australia. PMC is supported by a National Health and Medical Research Council Principal Research Fellowship. This research was jointly funded through these fellowships and the University of South Australia. 3 Address correspondence to J Keogh, School of Pharmacy and Medical Science, University of South Australia, GPO Box 2471, Adelaide, South Australia 5000, Australia. E-mail: [email protected]. 4 Abbreviations used: AIx, augmentation index; BP, blood pressure; CVD, cardiovascular disease; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; HKHN, high-potassium, high-sodium meal; ICAM-1, intercellular adhesion molecule-1; L-FMC, low flow-mediated constriction; LKHN, lowpotassium, high-sodium meal; LKLN, low-potassium, low-sodium meal; PWV, pulse wave velocity; PWVc-f, carotid-femoral pulse wave velocity; SBP, systolic blood pressure. Received December 17, 2014. Accepted for publication February 26, 2015. First published online March 18, 2015; doi: 10.3945/ajcn.114.105197.
Am J Clin Nutr 2015;101:939–46. Printed in USA. Ó 2015 American Society for Nutrition
939
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
ABSTRACT Background: It is known that increased potassium and reduced sodium intakes can improve postprandial endothelial function. However, the effect of increasing potassium in the presence of high sodium in the postprandial state is not known. Objective: We aimed to determine the effect of high potassium and high sodium on postprandial endothelial function as assessed by using flow-mediated dilatation (FMD) and arterial compliance as assessed by using pulse wave velocity (PWV) and central augmentation index (AIx). Design: Thirty-nine healthy, normotensive volunteers [21 women and 18 men; mean 6 SD age: 37 6 15 y; BMI (in kg/m2): 23.0 6 2.8] received a meal with 3 mmol K and 65 mmol Na (low-potassium, high-sodium meal (LKHN)], a meal with 38 mmol K and 65 mmol Na [high-potassium, high-sodium meal (HKHN)], and a control meal with 3 mmol K and 6 mmol Na (low-potassium, low-sodium meal) on 3 separate occasions in a randomized crossover trial. Brachial artery FMD, carotid-femoral PWV, central AIx, and blood pressure (BP) were measured while participants were fasting and at 30, 60, 90, and 120 min after meals. Results: Compared with the LKHN, the addition of potassium (HKHN) significantly attenuated the postmeal decrease in FMD (P-meal by time interaction , 0.05). FMD was significantly lower after the LKHN than after the HKHN at 30 min (P , 0.01). AIx decreased after all meals (P , 0.05). There were no significant differences in AIx, PWV, or BP between treatments over time. Conclusion: The addition of potassium to a high-sodium meal attenuates the sodium-induced postmeal reduction in endothelial function as assessed by FMD. This trial was registered at http://www.anzctr.org.au/ as ACTRN12613000772741. Am J Clin Nutr 2015;101:939–46.
940
BLANCH ET AL.
a significant predicting factor for CVD (17). Combined sodium and potassium interventions showed a benefit of reducing sodium and increasing potassium on blood pressure and vascular function in some populations (18, 19); however, the effect of modifying the ratio of sodium to potassium on postprandial vascular and endothelial function remains to be determined. The primary hypothesis tested in this study was that the addition of potassium to a high-sodium meal would mitigate the adverse effects of sodium on postprandial FMD. A low-potassium, low-sodium control meal was included in the study design to confirm the known adverse effect of high sodium on postprandial FMD. METHODS
Men and women aged between 18 and 70 y were recruited through personal contact and an advertisement on community notice boards (Figure 1). Inclusion criteria were BMI (in kg/m2) $18 and #30, systolic blood pressure (SBP) ,130 mmHg, diastolic blood pressure (DBP) ,90 mmHg, weight stability in the preceding 6 mo, and no use of an antihypertensive or cholesterol-lowering medication, systemic steroids, nonsteroidal anti-inflammatory drugs, or folate supplementation. Participants were not excluded if they were taking any other vitamin supplements provided that the dose was kept constant for the duration of the study. Exclusion criteria were known metabolic disease such as liver or kidney disease, treated hypertension, known or treated high cholesterol, clinical CVD, and inability to comprehend study protocol. Study methods In a randomized, double-blind, crossover design, 39 participants completed the study on 3 mornings separated by a minimum 7-d washout period. Participants consumed 3 test meals that contained 3 mmol K and 6 mmol Na [low-potassium, low-sodium meal (LKLN)], 3 mmol K and 65 mmol Na [low-potassium, high-
FIGURE 1
Diet analysis Participants completed an online dietary survey (Dietary Questionnaire for Epidemiologic Studies; Cancer Council Victoria) before commencing dietary interventions to determine habitual dietary intakes of potassium and other nutrients. Volunteers recorded their food intakes on the day before the first study and replicated these intakes before subsequent test days. Weight and height On arrival, participants had body height (first visit only) measured to the nearest 0.1 cm with a stadiometer (SECA) while barefoot. Body weight was measured to the nearest 0.05 kg by using calibrated electronic digital scales (SECA) while participants wore light clothing and no footwear. FMD As previously described (13), endothelium-dependent FMD of the right brachial artery was measured in the longitudinal plane
Consolidated Standards of Reporting Trials diagram of the flow of subjects through the study.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
Subjects
sodium meal (LKHN)], or 38 mmol K and 65 mmol Na [highpotassium, high-sodium meal (HKHN)]. Meals consisted of tomato soup (Heinz Big Red Condensed Tomato Salt Reduced Soup; Coles) made according to the manufacturer’s instructions. Sodium chloride (SAXA Plain Table Salt; Coles) was added to LKHN and HKHN, and potassium chloride (Paddymelon Gourmet Foods; The Melbourne Food Ingredient Depot) was added to the HKHN. All other macronutrients and micronutrients were identical between meals. After screening, participants were assigned to a treatment order by using an online-generated balanced random-number allocation sequence (http://www.randomization.com) by a person independent of the study. Participants were asked to fast (no food; 250 mL H2O only) from 2200 the night before each visit and refrain from alcohol, caffeine, and vigorous exercise in the 24 h before each study visit.
POTASSIUM REVERSES SODIUM-INDUCED VASOCONSTRICTION
FMD analysis Ultrasound images were recorded at a rate of 30 frames/s with screen-capture software (Debut Video Capture Software Professional V1.82; NCH Software) without QRS gating because previous studies showed that continuous recording has good agreement with R-wave gated measures when edge-detection software is used (22, 23). FMD video files were analyzed with edge-detection software (Brachial Analyzer for Research V6.1.3; Medical Imaging Application LLC). For both baseline and deflation, a region of interest was defined over a clear section of vessel with care to ensure that the region of interest was the same size and position for both baseline and deflation files. The automated edge-detection feature of the software was used to perform a frame-by-frame analysis to generate artery diameter (mm) values for both baseline and deflation. Baseline was defined as the average of 60-s pre-inflation diameter measures. The peak diameter was determined as the maximum diameter postcuff release. With the use of baseline and peak diameters, the percentage change in diameter was calculated (%FMD). The shear rate was calculated from the maximum blood velocity (V) and brachial artery diameter (D) (24) as follows: 4 3 VOD
ð1Þ
Normalized FMD was calculated by dividing the %FMD by the shear rate (21). The intraobserver CV calculated from baseline measurements (n = 39) was 19.6%. Low flow-mediated constriction (L-FMC) was calculated from the mean arterial diameter during the last 30 s before cuff release and expressed as the percentage of change from baseline (25). Arterial compliance Carotid-femoral PWV (PWVc-f) was measured by using a SphygmoCor device (AtCor Medical). Measurements were taken in the supine position by a single operator (CV = 5%) as previously described (26). Central AIx was measured by using the SphygmoCor device. Measurements were taken in the seated position by a single operator. BP Brachial BP was measured by using the SphygmoCor device. Measurements were taken in the seated position by a trained laboratory technician after participants were seated for 2 min. A
series of measurements were taken, each 1 min apart until 4 consistent measurements were obtained (i.e., SBP within a range of 10 mm Hg and DBP within a range of 5 mm Hg). The first reading was discarded, and 3 consistent measurements were averaged (27). Laboratory analysis Blood samples were collected from a cannula inserted into the left brachial vein into tubes with no additive for the measurement of serum sodium and potassium concentrations and ICAM-1, E-selectin, and endothelin-1. Serum was isolated by centrifugation at 2500 3 g for 10 min, and samples were stored at 2808C until analyzed. Biochemical assays were performed in a single assay on completion of the study. ICAM-1, E-selectin, and endothelin-1 were measured by using Quantikine human immunoassay kits (R&D Systems; Bioscientific) on the basis of the manufacturer’s protocols. An analysis of serum sodium and potassium concentrations was performed at the Institute of Medical and Veterinary Sciences, Adelaide, South Australia. Ethics This study was approved by the University of South Australia’s Human Research Ethics Committee (ethics approval number: 0000031547). All participants gave written informed consent. This trial was registered at http://www.anzctr.org.au/ as ACTRN12613000772741. Statistical analysis The primary outcome for the statistical analysis was the effect of HKHN compared with LKHN on FMD. The secondary outcome was the effect of HKHN compared with LKHN on PWV. On the basis of power calculations from our previous study (13) with 80% power and P , 0.05 to detect a minimum change in FMD of 1.3% absolute, 40 people were required in a crossover study. With 80% power and P , 0.05 to detect a 10% (0.6-m/s) change in PWVc-f and the assumption of a within-patient SD of 0.9 m/s, 20 people were required in a crossover study. All analyses were performed with SPSS 21 for Windows software (SPSS Inc.). Significance was set at P , 0.05. Data were tested for the normality of distribution and to ensure that residuals had approximately constant SDs by using the Kolmogorov-Smirnov test, Q-Q plots, and histograms. An ANOVA with repeated measures (with diet as the within-subject factor) was used to analyze outcomes (with and without covariates including diet order and BP), and we performed a post hoc analysis to assess primary and secondary comparisons. For FMD only, differences from baseline were calculated before the analysis to make meal comparisons easier to see. The primary comparison was HKHN compared with LKHN as reflected by the hypothesis. A secondary comparison was included to show a difference in LKHN and the LKLN control to show the adverse effect of sodium on postprandial FMD. Age, BMI, sex, and habitual potassium intake were also included as covariates. Preliminary analyses were performed to ensure no violation of assumptions of linearity. Pearson correlation analyses were conducted to assess the association of change
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
above the antecubital fossa with an 8.8-MHz linear array transducer (MySono U6; Samsung Medison) according to published guidelines (20, 21). Subjects attended after an overnight fast and lay quietly for 5 min in a quiet, temperature-controlled room before endothelium-dependent FMD measurements were obtained (21). The brachial artery diameter was measured before and after forearm ischemia caused by inflation of a sphygmomanometer cuff applied to the right forearm 2 cm below the olecranon process to 200 mmHg for 5 min. Continuous longitudinal 2-dimensional images of the brachial artery were obtained and digitally recorded during quite rest (1 min) and from 30 s before cuff release and during reactive hyperemia (3 min). All images were stored for offline analysis by a single trained observer who was unaware of the treatments at the time of measurement.
941
942
BLANCH ET AL.
between variables. Data are expressed as means 6 SDs or median (IQRs) as appropriate. RESULTS
Subjects Thirty-nine subjects completed the study, and baseline characteristics are outlined in Table 1. DBP before the LKHN was significantly lower than before the LKLN or the HKHN (P = 0.01). There were no significant differences between other fasting variables between treatments. Brachial artery endothelial function
TABLE 1 Fasting characteristics of participants1
BP SBP and DBP increased after all meals (P , 0.01 for both) (Table 2). There were no significant differences in SBP after each of the test meals and a borderline change in DBP (P-meal 3 time = 0.05). No effect of age, BMI, sex, or habitual potassium intake was observed. Mean arterial pressure increased after all meals (P = 0.03). There were no significant differences in mean arterial pressure after each of the meals.
Arterial compliance measures PWV c-f increased after all meals compared with fasting (P-time effect , 0.01). There were no significant differences in PWVc-f after each of the meals (P-meal 3 time interaction = 0.29 for the HKHN compared with the LKHN). Central AIx and AIx corrected for heart rate at 75 beats/min (n = 34) decreased after each meal (P-time , 0.01 for both). There was no meal 3 time interaction (HKHN compared with LKHN) for AIx (P = 0.20) or AIx corrected for heart rate at 75 beats/ min (P = 0.10). Heart rate decreased after all meals (P-time , 0.01). There were no significant differences in heart rate after each of the meals. Pulse pressure increased after all meals (P-time , 0.01). There were no significant differences in pulse pressure after each of the meals (P-meal 3 time interaction = 0.53).
Baseline Age, y BMI, kg/m2 HR, beats/min SBP, mm Hg DBP, mm Hg MAP, mm Hg Brachial artery diameter, mm FMD, % PWVc-f, m/s AIx, % AIx@75beats/min,2 % PP, mm Hg Serum sodium, mmol/L Serum potassium, mmol/L ICAM-1, mg/L E-selectin, mg/L Endothelin-1, pg/mL
38 23.1 57 115 71 83 3.80 9.5 6.7 12 7 31 140 4.2 131.5 38.3 1.34
6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6
16 2.9 9 8 6 7 0.66 3.5 1.1 15 16 4 1 0.2 38.1 22.1 0.36
1 All values are means 6 SDs. n = 39 (21 women and 18 men). AIx, augmentation index; AIx@75beats/min, augmentation index corrected to heart rate at 75 beats/min; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; HR, heart rate; ICAM-1, intracellular adhesion molecule-1; MAP, mean arterial blood pressure; PP, pulse pressure; PWVc-f, carotid-femoral pulse wave velocity; SBP, systolic blood pressure. 2 n = 34 (18 women and 16 men).
Biochemical analysis Serum sodium increased significantly after the LKHN (maximum change was 1.2 6 0.9 mmol at 60 min) and the HKHN compared with the LKLN (P-meal 3 time interaction , 0.01 for both comparisons). There was no difference in serum sodium between the LKHN and the HKHN (P = 0.59). Serum potassium significantly increased after the HKHN (maximum change was 0.6 6 0.3 mmol at 90 min) compared with the LKHN and the control meal (P-meal 3 time interaction , 0.01 for both comparisons). Endothelin-1 decreased after both meals (P-time , 0.01) but was higher after the HKHN than after the LKLN (P-meal 3 time interaction = 0.03). ICAM-1 decreased after both meals (P-time , 0.01) but decreased less after the HKHN than after the LKHN. There were no changes in serum E-selectin after any of the interventions. FMD and serum sodium was negatively correlated at 60 min with the LKHN (r = 20.45, P = 0.01). There were no correlations between FMD and serum potassium, adhesion molecules, or maximum flow. There was a weak inverse correlation between FMD and SBP and DBP at various time points but only with the LKHN diet.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
There was no meal 3 time interaction (P = 0.06) but a significant effect of meal (P , 0.01) for the comparison of the 3 meals (Table 2). The addition of potassium in the HKHN significantly attenuated the postprandial decrease in FMD after the LKHN (P-meal 3 time interaction = 0.02). There was no significant meal 3 time interaction (P = 0.65) or time effect (P = 0.59) for the FMD change from baseline (time 0) between the 3 meals (Figure 2). There was a significant effect of meal (P , 0.01), and FMD decreased by 21.5% from baseline after the LKLN and was further decreased to 22.9% from baseline after the LKHN. When the change in FMD from baseline was compared for the primary hypothesis (LKHN compared with HKHN), there was no meal 3 time interaction (P = 0.57), but the meal effect was significant (P , 0.01). The maximum difference in FMD after the LKHN compared with the HKHN was at 30 min (23.58 6 0.85%; P , 0.01). There was no effect of age, sex, BMI, or habitual potassium intake observed. When FMD was normalized for the shear rate, the meal effect did not change, but meal by time interaction weakened and became
nonsignificant (P-meal effect , 0.01; P-meal 3 time interaction = 0.12 for the LKHN compared with the HKHN). The low flow constriction observed with the LKHN (Table 2) was completely abolished by the HKHN (P-meal 3 time interaction , 0.01) with a difference of 24.95 6 0.87% and P , 0.01 at 60 min. There was no effect of age, BMI, sex, or habitual potassium intake observed.
60
90
120
0
30
60
90
120
0
30
60
HKHN (min)
90
120
6 6 6 6 6 6 6 6 6
0.7 2.1 1 1 1 1 0.2 3 3.2
6 6 6 6 6 6 6 6 6 0.6 2.6 1 2 1 1 0.2 3 3.1
4.4 6 0.05
30 6 1 140 6 0.2
20.1 43.3 53 115 71 82 6.6 6 21
6 6 6 6 6 6 6 6 6 0.9 2.8 1 1 1 1 0.2 3 2.9
4.4 6 0.04
31 6 1 140 6 0.2
20.9 44.1 54 116 70 82 6.7 6 23
6 6 6 6 6 6 6 6 6 0.5 2.4 1 1 1 1 0.2 3 2.8
4.4 6 0.04
31 6 1 140 6 0.2
1.1 49.5 54 117 72 83 6.8 6 23
6 6 6 6 6 6 6 6 6 0.8 2.7 1 1 1 1 0.2 3 3.0
4.2 6 0.05
30 6 1 140 6 0.2
20.9 48.2 58 112 69 81 6.6 10 6
6 6 6 6 6 6 6 6 6 0.8 2.9 1 1 1 1 0.2 3 3.1
4.3 6 0.04
30 6 1 141 6 0.2
21.6 48.6 55 114 71 83 6.6 7 0
6 6 6 6 6 6 6 6 6 0.8 2.7 1 1 1 1 0.2 3 3.1
4.3 6 0.04
30 6 1 141 6 0.2
23.5 48.0 54 115 72 83 6.7 7 22
6 6 6 6 6 6 6 6 6 0.7 2.8 1 1 1 1 0.2 3 3.3
4.4 6 0.04
31 6 1 141 6 0.2
22.0 46.5 54 116 71 83 6.7 6 22 6 6 6 6 6 6 6 6 6 0.8 2.9 1 2 1 1 0.2 3 3.0
4.4 6 0.04
32 6 1 141 6 0.2
21.8 49.7 54 117 72 84 6.8 6 22
6 6 6 6 6 6 6 6 6 0.7 2.4 1 1 1 1 0.2 3 3.1
4.3 6 0.05
30 6 1 140 6 0.3
20.7 43.3 57 114 71 83 6.5 11 6 6 6 6 6 6 6 6 6 6 0.7 2.2 1 2 1 1 0.2 3 3.0
4.7 6 0.05
30 6 1 141 6 0.2
0.7 46.1 55 115 72 84 6.6 10 4
6 6 6 6 6 6 6 6 6
1.0 2.7 1 2 1 1 0.2 3 3.2
4.9 6 0.06
0.3 49.3 53 116 71 82 6.7 8 0
4.8 6 0.05
0.8 2.9 1 1 1 1 0.2 3 3.1
31 6 1 141 6 0.2
6 6 6 6 6 6 6 6 6
31 6 1 141 6 0.2
1.5 50.2 54 115 71 83 6.5 9 1
6 6 6 6 6 6 6 6 6
0.7 3.1 1 1 1 1 0.2 3 3.1
0.03 0.06
0.53 0.50 ,0.01 ,0.01 ,0.01 ,0.01
0.09 ,0.01 0.08 0.11 0.04 ,0.01 0.35 0.62 ,0.01 0.49 0.93 ,0.01 0.05 0.90 ,0.01 0.15 0.70 0.03 0.48 0.82 ,0.01 0.39 ,0.01 ,0.01 0.10 ,0.01 ,0.01
0.06 ,0.01 0.08 0.18
0.54 0.59
0.01 0.08 0.33 0.19 0.01 0.03 0.29 0.20 0.10
0.02 0.03
0.87 ,0.01 0.38 0.40 0.01 ,0.01
0.03 0.94 0.03
4.8 6 0.05 ,0.01 ,0.01 ,0.01 ,0.01
32 6 1 141 6 0.2
2.1 48.5 53 117 72 84 6.8 9 1
P Meal 3 Meal Time Meal 3 time2 effect2 effect2 time3
128.0 6 5.1 116.4 6 4.8 118.5 6 4.8 122.1 6 4.7 119.4 6 4.6 133.1 6 6.4 117.3 6 4.7 116.5 6 4.8 115.3 6 4.5 117.8 6 4.7 128.2 6 5.5 120.7 6 5.4 118.6 6 5.4 118.6 6 5.0 115.3 6 5.0 ,0.01 0.60 33.1 6 4.1 37.5 6 3.9 36.5 6 3.9 37.0 6 3.8 34.4 6 3.3 38.5 6 3.7 40.9 6 3.1 39.7 6 3.3 37.6 6 3.1 39.8 6 3.1 37.9 6 3.5 38.5 6 3.9 38.5 6 3.8 36.3 6 3.3 37.6 6 3.3 1.36 6 0.07 1.23 6 0.06 1.36 6 0.07 1.46 6 0.07 1.47 6 0.07 1.45 6 0.08 1.08 6 0.06 1.20 6 0.05 1.28 6 0.05 1.39 6 0.06 1.26 6 0.06 1.14 6 0.05 1.13 6 0.05 1.24 6 0.06 1.22 6 0.05 0.02
4.3 6 0.04
20.4 44.6 55 115 71 83 6.6 7 21
4.2 6 0.04
0.8 2.4 1 1 1 1 0.2 2 2.6
30 6 1 139 6 0.2
6 6 6 6 6 6 6 6 6
30 6 1 140 6 0.2
21.5 41.2 58 113 70 82 6.5 10 6
8.6 6 0.6 7.5 6 0.7 7.3 6 0.6 7.4 6 0.6 7.1 6 0.6 8.9 6 0.5 6.2 6 0.6 5.8 6 0.4 6.0 6 0.5 7.3 6 0.6 8.8 6 0.5 9.7 6 0.7 9.0 6 0.7 8.8 6 1.0 9.3 6 0.8 3.89 6 0.11 3.91 6 0.12 3.90 6 0.11 3.89 6 0.11 3.84 6 0.11 3.86 6 0.10 3.90 6 0.11 3.93 6 0.11 3.90 6 0.11 3.84 6 0.11 3.89 6 0.11 3.78 6 0.11 3.81 6 0.12 3.79 6 0.12 3.76 6 0.12
30
LKHN (min)
1
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
All values are means 6 SEMs. n = 39 (21 women and 18 men). AIx, augmentation index; AIx@75beats/min, augmentation index corrected to heart rate at 75 beats/min; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; HKHN, high-potassium, high-sodium meal; HR, heart rate; ICAM-1, intracellular adhesion molecule-1; L-FMC, low flow-mediated constriction; LKHN, low-potassium, highsodium meal; LKLN, low-potassium, low-sodium meal; MAP, mean arterial blood pressure; PP, pulse pressure; PWVc-f, carotid-femoral pulse wave velocity; SBP, systolic blood pressure. 2 Repeated-measures ANOVA for 3 meals. 3 Repeated-measures ANOVA for LKHN compared with HKHN. 4 Repeated-measures ANOVA for baseline measurements, P = 0.01. 5 n = 34 (18 women and 16 men).
FMD, % Brachial artery diameter, mm L-FMC, % Shear rate, s21 HR, beats/min SBP, mm Hg DBP,4 mm Hg MAP, mm Hg PWVc-f, m/s AIx, % AIx@75beats/ min,5 % PP, mm Hg Serum sodium, mmol/L Serum potassium, mmol/L ICAM-1, mg/L E-selectin, mg/L Endothelin-1, pg/mL
0
LKLN (min)
TABLE 2 All outcome variables for each meal intervention1
POTASSIUM REVERSES SODIUM-INDUCED VASOCONSTRICTION
943
944
BLANCH ET AL.
DISCUSSION
The main outcome from this study was that the addition of potassium to a high-sodium meal significantly attenuated a sodium-invoked postmeal reduction in endothelial function as assessed by FMD. To our knowledge, this is the first time that combined sodium and potassium has been examined in the postprandial state. The amount of added sodium used in the meals was equivalent to w2 rashers of bacon [or 2 thick sausages (80 g each)]. The amount of added potassium was equivalent to 3 large bananas [or 2 large potatoes (340 g cooked weight) or 0.75 cups dried apricots]. Dickinson et al. (7) reported that a high-sodium meal impaired brachial artery FMD in the postprandial state, which was a finding that was replicated in this study (Figure 2). These findings were also consistent with the deleterious effect of sodium on fasting FMD in the short term (8, 9, 28). We showed that impairment in FMD was correlated with increases in SBP and DBP after the high-sodium meal without potassium. This effect was not seen in the postprandial study by Dickinson et al. (7). Furthermore, we previously showed that a high-potassium meal (without sodium) can attenuate postmeal responses in FMD to a low-potassium meal (13). Previous studies that investigated the effect of increasing potassium on fasting FMD have reported mixed results (12, 15, 29); however, dose appears to play an important role with potassium increases .40 mmol improving endothelial function (14). The current study suggests a high potassium load attenuates postmeal decreases in FMD irrespective of the sodium concentration present. L-FMC increased after the high-sodium meal and decreased after the addition of potassium to the meal. L-FMC measures the endothelial control of vascular tone at rest (30). L-FMC mechanisms are not mediated by nitric oxide release and are thought to be different from those of FMD (30). FMD does not correlate with L-FMC measures, and the clinical relevance of changes in L-FMC remains unknown (25). However, L-FMC has been shown to be attenuated by 3 mo of statin
treatment (31). The assessment of L-FMC combined with FMD provides complimentary information about vascular homeostasis in addition to traditional risk factors (30). We observed no significant differences in SBP between meals. BP increased after all meals. This result was unexpected because previous studies that used the same control meal in similar populations showed a small postmeal decrease in BP at 30 min (13). However, postmeal increases in BP that provoked increases in PWVc-f in response to a high-fat meal were previously described (32). Our study showed no change in PWVc-f between meals; however, we saw a small but significant trend for increased PWVc-f over time. Previous studies showed an improvement in fasting PWVc-f in hypertensive (10, 33, 34) but not normotensive (35) populations with reduced sodium intake. The influence of potassium on vascular function measures is unclear, with BP likely to influence any observed effect (14). Our study showed a significant reduction in central AIx in response to each of the test meals. This result remained after correction for heart rate (augmentation index corrected to heart rate at 75 beats/min). Previous studies showed mixed results in AIx in response to a mixed meal with some studies that showed an increase in AIx (36) and other studies that showed a decrease (37). Furthermore, Dickinson et al. (38) reported a further increase in AIx after a high-sodium meal. Our study showed no meal 3 time interaction for AIx. Previous studies showed no effect of potassium interventions on fasting AIx (15, 39, 40). One salt-substitution study (simultaneously reducing sodium and increasing potassium) showed significant improvements in measures of arterial stiffness (central pulse pressure and pulse wave reflection time) after 12 mo of intervention but did not see an effect on AIx (41). This study showed that a meal that contained 38 mmol K raised serum potassium by 0.6 mmol/L in healthy individuals. We hypothesized that this increase in serum potassium affects the FMD response by one of 2 mechanisms. First, we speculate that
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
FIGURE 2 Mean (6SEM) changes in brachial artery FMD from baseline in response to the consumption of test meals that contained 3 mmol K and 65 mmol Na (LKHN) compared with 38 mmol K and 65 mmol Na (HKHN) and 3 mmol K and 6 mmol Na (LKLN). n = 39 (18 men and 21 women). For the change in FMD from baseline (repeated-measures ANOVA): P-meal 3 time = 0.65; P-meal effect , 0.01; P-time effect = 0.59. For change in FMD from baseline: repeated-measures ANOVA): LKHN compared with HKHN P-meal 3 time interaction = 0.57 (P-meal effect , 0.01; P-time effect = 0.35); LKLN compared with LKHN P-meal 3 time interaction = 0.33 (P-meal effect = 0.01; P-time effect = 0.66). FMD, flow-mediated dilatation; HKHN, high-potassium, high-sodium meal; LKHN, low-potassium, high-sodium meal; LKLN, low-potassium, low-sodium meal.
POTASSIUM REVERSES SODIUM-INDUCED VASOCONSTRICTION
a mechanistic explanation for the effect of potassium and sodium on endothelial function; however, this study provides some insight into the changes in endothelial mediators that occur in response to high-potassium and -sodium loads. In conclusion, this study showed that the amount of potassium present in 3 servings of fruit can significantly attenuate reductions in endothelial function induced by a high-sodium (65-mmol) meal. A transient increase in serum potassium is a plausible mechanism for this response; however, increased potassium does not appear to affect other measures of vascular function. This findings suggests that, in populations in whom reducing sodium intake may be problematic (e.g., primarily of processed foods), increasing potassium intake may reduce CVD risk. Although the long-term implications of this result in the pathogenesis of atherosclerosis and CVD is unknown, this study provides additional evidence that increases in dietary potassium should be encouraged, particularly in the presence of a high-sodium westernized diet. The authors’ responsibilities were as follows—JBK and PMC: developed the hypotheses tested in the study, designed the research (project conception, development of overall research plan, and study oversight), and contributed to statistical analyses and interpretation of data; NB: contributed to the study design, planned and conducted the study, performed vascular measurements (hands-on conduct of experiments and data collection), performed initial statistical analyses, and drafted the manuscript; KSP: assisted in conducting the study and performed some vascular measurements (hands-on conduct of experiments and data collection); and all authors: critically reviewed the manuscript. None of the authors reported a conflict of interest related to the study.
REFERENCES 1. Kitta Y, Obata J-e, Nakamura T, Hirano M, Kodama Y, Fujioka D, Saito Y, Kawabata K-i, Sano K, Kobayashi T, et al. Persistent impairment of endothelial vasomotor function has a negative impact on outcome in patients with coronary artery disease. J Am Coll Cardiol 2009;53:323–30. 2. Heitzer T, Schlinzig T, Krohn K, Meinertz T, Mu¨nzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation 2001;104:2673–8. 3. McEniery CM, Yasmin, Hall IR, Qasem A, Wilkinson IB, Cockcroft JR. Normal vascular aging: differential effects on wave reflection and aortic pulse wave velocity: the Anglo-Cardiff Collaborative Trial (ACCT). J Am Coll Cardiol 2005;46:1753–60. 4. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of cardiovascular events and all-cause mortality with arterial stiffness: a systematic review and meta-analysis. J Am Coll Cardiol 2010;55: 1318–27. 5. Strazzullo P, D’Elia L, Kandala N-B, Cappuccio FP. Salt intake, stroke, and cardiovascular disease: meta-analysis of prospective studies. BMJ 2009;339:b4567. 6. He J, Ogden LG, Vupputuri S, Bazzano LA, Loria C, Whelton PK. Dietary sodium intake and subsequent risk of cardiovascular disease in overweight adults. JAMA 1999;282:2027–34. 7. Dickinson KM, Clifton PM, Keogh JB. Endothelial function is impaired after a high-salt meal in healthy subjects. Am J Clin Nutr 2011; 93:500–5. 8. Dickinson KM, Keogh JB, Clifton PM. Effects of a low-salt diet on flow-mediated dilatation in humans. Am J Clin Nutr 2009;89:485–90. 9. Dickinson KM, Clifton PM, Keogh JB. A reduction of 3 g/day from a usual 9 g/day salt diet improves endothelial function and decreases endotheilin-1 in a randomised cross_over study in normotensive overweigh and obese subjects. Atherosclerosis 2014;233:32–8. 10. Todd AS, MacGinley RJ, Schollum JB, Johnson RJ, Williams SM, Sutherland WH, Mann JI, Walker RJ. Dietary salt loading impairs arterial vascular reactivity. Am J Clin Nutr 2010;91:557–64. 11. Aburto NJ, Hanson S, Gutierrez H, Hooper L, Elliott P, Cappuccio FP. Effect of increased potassium intake on cardiovascular risk factors and disease: systematic review and meta-analyses. BMJ 2013;346:f1378.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
the rise in serum potassium may directly affect the small and intermediate calcium-activated potassium channels of endothelial cells, which are then electrically coupled to the smooth cells (i.e., an endothelium-derived hyperpolarizing factor effect) (42). Second, increased serum potassium may increase nitric oxide release because cellular studies showed an increase in release of nitrites (an index of nitric oxide release) in response to increased extracellular potassium (43). Serum sodium was raised by a maximum of 1.2 mmol/L after the 65-mmol sodium meals. This change in serum sodium was expected because previous studies showed an increase in serum sodium in response to a 65-mmol sodium meal (38). Dickinson et al. (7) hypothesized that the mechanism for the observed decrease in FMD after a high-sodium meal is increased serum sodium resulting in a reduction in nitric oxide bioavailability. Evidence from in vitro studies suggested increases of 5–10 mmol in serum sodium reduce nitric oxide bioavailability and nitric oxide synthase activity (44, 45). We did not measure serum concentrations of nitrate and nitrite in this study because previous studies did not detect changes in nitrate and nitrite concentrations after a 65-mmol Na meal (38). In our study, we observed a relation between BP and FMD after the high-sodium meal that was not seen in the study by Dickinson et al. (7); therefore, the rise in BP could be the mechanism for the reduction in FMD. We observed small but significant differences in serum endothelin-1 and ICAM-1 with diminished postmeal falls with the addition of potassium to the meal, but baseline values were higher with the high-sodium meal, and this finding alone may account for the difference. Endothelin-1 was previously shown to decrease after a reduced sodium diet (9). Furthermore, Liu et al. (46) showed that endothelin-1 decreased after a 7-d low-sodium diet (51.3 mmol/d) and increased after a 7-d high sodium diet (307.7 mmol/d), and the addition of potassium (60 mmol/d for 7 d) attenuated the response to the high-sodium diet. Conversely, Dickinson et al. (38) failed to detect a difference in postprandial endothelin-1 concentrations between a low-sodium meal and a high-sodium meal alone (5 compared with 65 mmol). Previously, reductions in fasting ICAM-1 were reported after a 6-wk high-potassium diet in woman only (15), but we did not see a sexrelated treatment effect in our study. The role of variations in endothelin-1 and ICAM-1 in alterations in endothelial function in the postprandial state are unknown, but they are unrelated to changes in FMD. Limitations of this study included the use of a relatively healthy population sample, which might have limited the potential to see improvements in BP and vascular variables. However, we saw significant improvements in measures of endothelial function after the addition of potassium to the highsodium meal. Furthermore, it is unclear whether these results persist in the longer term, and the effect of high potassium in the presence of high sodium on fasting FMD warrants additional investigation. We acknowledge that, in populations with renal impairment, there may be risk of hyperkalemia by increasing dietary potassium, and in our study, serum potassium remained elevated at 120 min. Another limitation was that we did not adjust for any effect of menstrual cycles of premenopausal women. However, we did not see an effect of sex and did not observe any differences in baseline FMD measurements between treatments; therefore, the menstrual cycle does not appear to have affected our data. It was a limitation that we are unable to provide
945
946
BLANCH ET AL. 30. Poredos P, Jezovnik MK. Testing endothelial function and its clinical relevance. J Atheroscler Thromb 2013;20:1–8. 31. Megnien JL, Simon A, Andriani A, Segond P, Jeannin S, Levenson J. Cholesterol lowering therapy inhibits the low-flow mediated vasoconstriction of the brachial artery in hypercholesterolaemic subjects. Br J Clin Pharmacol 1996;42:187–93. 32. Lithander FE, Herlihy LK, Walsh DM, Burke E, Crowley V, Mahmud A. Postprandial effect of dietary fat quantity and quality on arterial stiffness and wave reflection: a randomised controlled trial. Nutrition Journal 2013 Jul 10 (Epub ahead of print; DOI: 10.1186/1475-2891-1293). 33. Jablonski KL, Fedorova OV, Racine ML, Geolfos CJ, Gates PE, Chonchol M, Fleenor BS, Lakatta EG, Bagrov AY, Seals DR. Dietary sodium restriction and association with urinary marinobufagenin, blood pressure, and aortic stiffness. Clin J Am Soc Nephrol 2013;8: 1952–9. 34. He FJ, Marciniak M, Visagie E, Markandu ND, Anand V, Dalton RN, MacGregor GA. Effect of modest salt reduction on blood pressure, urinary albumin, and pulse wave velocity in white, black, and Asian mild hypertensives. Hypertension 2009;54:482–8. 35. Todd AS, Macginley RJ, Schollum JB, Williams SM, Sutherland WH, Mann JI, Walker RJ. Dietary sodium loading in normotensive healthy volunteers does not increase arterial vascular reactivity or blood pressure. Nephrology (Carlton) 2012;17:249–56. 36. Ahuja KD, Robertson IK, Ball MJ. Acute effects of food on postprandial blood pressure and measures of arterial stiffness in healthy humans. Am J Clin Nutr 2009;90:298–303. 37. Taylor JL, Curry TB, Matzek LJ, Joyner MJ, Casey DP. Acute effects of a mixed meal on arterial stiffness and central hemodynamics in healthy adults. Am J Hypertens 2014;27:331–7. 38. Dickinson KM, Clifton PM, Burrell LM, Barrett PHR, Keogh JB. Postprandial effects of a high salt meal on serum sodium, arterial stiffness, markers of nitric oxide production and markers of endothelial function. Atherosclerosis 2014;232:211–6. 39. Graham UM, McCance D, Young I, Mullan K. A randomised controlled trial evaluating the effect of potassium supplementation on vascular function and the renin-angiotensin-aldosterone system. J Hum Hypertens 2014;28:333–9. 40. Matthesen SK, Larsen T, Vase H, Lauridsen TG, Pedersen EB. Effect of potassium supplementation on renal tubular function, ambulatory blood pressure and pulse wave velocity in healthy humans. Scand J Clin Lab Invest 2012;72:78–86. 41. Hu J, Jiang X, Li N, Yu X, Perkovic V, Chen B, Zhao L, Neal B, Wu Y. Effects of salt substitute on pulse wave analysis among individuals at high cardiovascular risk in rural China: a randomized controlled trial. Hypertens Res 2009;32:282–8. 42. Coleman HA, Tare M, Parkington HC. Endothelial potassium channels, endothelium-dependent hyperpolarization and the regulation of vascular tone in health and disease. Clin Exp Pharmacol Physiol 2004;31: 641–9. 43. Oberleithner H, Callies C, Kusche-Vihrog K, Schillers H, Shahin V, Riethmuller C, Macgregor GA, de Wardener HE. Potassium softens vascular endothelium and increases nitric oxide release. Proc Natl Acad Sci U S A 2009;106:2829–34. 44. Fujiwara N, Osanai T, Kamada T, Katoh T, Takahashi K, Okumura K. Study on the relationship between plasma nitrite and nitrate level and salt sensitivity in human hypertension: modulation of nitric oxide synthesis by salt intake. Circulation 2000;101:856–61. 45. Oberleithner H, Riethmu¨ller C, Schillers H, MacGregor GA, de Wardener HE, Hausberg M. Plasma sodium stiffens vascular endothelium and reduces nitric oxide release. Proc Natl Acad Sci USA 2007;104:16281–6. 46. Liu Z, Peng J, Lu F, Zhao Y, Wang S, Sun S, Zhang H, Diao Y. Salt loading and potassium supplementation: effects on ambulatory arterial stiffness index and endothelin-1 levels in normotensive and mild hypertensive patients. J Clin Hypertens (Greenwich) 2013;15:485–96.
Downloaded from ajcn.nutrition.org at NEW YORK UNIVERSITY BOBST LIBRARY on May 11, 2015
12. Blanch N, Clifton P, Petersen K, Willoughby S, Keogh J. Effect of high potassium diet on endothelial function. Nutr Metab Cardiovasc Dis 2014;24:983–9. 13. Blanch N, Clifton PM, Keogh JB. Postprandial effects of potassium supplementation on vascular function and blood pressure: a randomised cross-over study. Nutr Metab Cardiovasc Dis 2014;24:148–54. 14. Blanch N, Clifton PM, Keogh JB. A systematic review of vascular and endothelial function: Effects of fruit, vegetable and potassium intake. Nutr Metab Cardiovasc Dis 2014 Oct 15 (Epub ahead of print; DOI: 10.1016/j.numecd.2014.10.001). 15. Berry SE, Mulla UZ, Chowienczyk PJ, Sanders TA. Increased potassium intake from fruit and vegetables or supplements does not lower blood pressure or improve vascular function in UK men and women with early hypertension: a randomised controlled trial. Br J Nutr 2010; 104:1839–47. 16. Intersalt: an international study of electrolyte excretion and blood pressure. Results for 24 hour urinary sodium and potassium excretion. Intersalt Cooperative Research Group. BMJ 1988;297:319–28. 17. Cook NR, Obarzanek E, Cutler JA, Buring JE, Rexrode KM, Kumanyika SK, Appel LJ, Whelton PK, for the Trials of Hypertension Prevention Collaborative Research Group. Joint effects of sodium and potassium intake on subsequent cardiovascular disease: the trials of Hypertension Prevention Follow-up Study. Arch Intern Med 2009;169:32–40. 18. Zhou B, Wang HL, Wang WL, Wu XM, Fu LY, Shi JP. Long-term effects of salt substitution on blood pressure in a rural North Chinese population. J Hum Hypertens 2013;27:427–33. 19. Gilleran G, O’leary M, Bartlett W, Vinall H, Jones A, Dodson P. Effects of dietary sodium substitution with potassium and magnesium in hypertensive type II diabetics: a randomised blind controlled parallel study. J Hum Hypertens 1996;10:517–21. 20. Harris RA, Nishiyama SK, Wray DW, Richardson RS. Ultrasound assessment of flow-mediated dilation. Hypertension 2010;55:1075–85. 21. Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D, Charbonneau F, Creager MA, Deanfield J, Drexler H, Gerhard-Herman M, Herrington D. Guidelines for the ultrasound assessment of endothelial-dependent flow-mediated vasodilation of the brachial artery* 1: a report of the International Brachial Artery Reactivity Task Force. J Am Coll Cardiol 2002; 39:257–65. 22. Kizhakekuttu TJ, Gutterman DD, Phillips SA, Jurva JW, Arthur EIL, Das E, Widlansky ME. Measuring FMD in the brachial artery: how important is QRS gating? J Appl Physiol (1985) 2010;109:959–65. 23. Gemignani V, Bianchini E, Faita F, Giannarelli C, Plantinga Y, Ghiadoni L, Demi M. Ultrasound measurement of the brachial artery flowmediated dilation without ECG gating. Ultrasound Med Biol 2008; 34:385–91. 24. Parker BA, Trehearn TL, Meendering JR. Pick your Poiseuille: normalizing the shear stimulus in studies of flow-mediated dilation. J Appl Physiol 2009;107:1357–9. 25. Spiro JR, Digby JE, Ghimire G, Mason M, Mitchell AG, Ilsley C, Donald A, Dalby MC, Kharbanda RK. Brachial artery low-flowmediated constriction is increased early after coronary intervention and reduces during recovery after acute coronary syndrome: characterization of a recently described index of vascular function. Eur Heart J 2011;32:856–66. 26. Hwang MH, Yoo JK, Kim HK, Hwang CL, Mackay K, Hemstreet O, Nichols WW, Christou DD. Validity and reliability of aortic pulse wave velocity and augmentation index determined by the new cuff-based SphygmoCor Xcel. J Hum Hypertens 2014;28:475–81. 27. Ogedegbe G, Pickering T. Principles and techniques of blood pressure measurement. Cardiol Clin 2010;28:571–86. 28. Jablonski KL, Racine ML, Geolfos CJ, Gates PE, Chonchol M, McQueen MB, Seals DR. Dietary sodium restriction reverses vascular endothelial dysfunction in middle-aged/older adults with moderately elevated systolic blood pressure. J Am Coll Cardiol 2013;61:335–43. 29. He FJ, Marciniak M, Carney C, Markandu ND, Anand V, Fraser WD, Dalton RN, Kaski JC, MacGregor GA. Effects of potassium chloride and potassium bicarbonate on endothelial function, cardiovascular risk factors, and bone turnover in mild hypertensives. Hypertension 2010;55:681–8.
