Salt sensitivity is associated with obesity, and increased cardiovascular morbidity and mortality. We investigated whether treatment of obesity and its associated metabolic abnormalities corrects salt sensitivity and restores impaired nitric oxide (NO) metabolism characteristic of salt sensitivity. Twenty, otherwise, healthy obese salt-sensitive subjects completed a 12-month program of caloric restriction, aerobic exercise and metformin. Two salt sensitivity tests were performed, that is at baseline and end of program. Lifestyle-metformin treatment decreased weight (9.8±0.3 kg), body mass index (3.9±0.2 kg/m2), waist (11.5±0.5 cm), systolic blood pressure (SBP) (8.6±0.4 mm Hg), diastolic blood pressure (DBP) (5.5±0.4 mm Hg), triglyceride (40±5 mg/dl), fasting (8.3±1 μIU/ml) and post-load (20±4 μIU/ml) insulin levels, and salt sensitivity. Going from a high-sodium (∼300 mmol) to a low-sodium diet (∼30 mmol of sodium/day) lowered SBP/DBP by 14.7±1.7/7.4±0.9 mm Hg at baseline and by 8.6±1.9/3.2±1.2 mm Hg after treatment (P<0.001). More importantly, blood pressure (BP) sensitivity to customary levels of dietary salt (∼150 mmol of sodium/day) was abolished by the lifestyle-metformin treatment. Differences in SBP/DBP between usual and low salt averaged 11±1/8±1 mm Hg before treatment, and 3±1/1±0.5 mm Hg after treatment (P<0.001). At baseline, NO-metabolite excretion was inhibited during high salt; this impairment was corrected by the lifestyle-metformin treatment. In conclusion, acquired correctable factors play an important role in the pathogenesis of salt sensitivity associated with obesity. Correction of salt sensitivity may account for the BP lowering induced by weight reduction. Restoration of the inability to increase or sustain NO production in response to high salt could account for the correction of salt sensitivity induced by the lifestyle-metformin treatment.
Salt sensitivity is characterized by an increased reactivity of blood pressure (BP) to changes in salt intake, and is associated with increased cardiovascular disease events and reduced survival.1, 2, 3, 4, 5 Although the pathogenesis of salt sensitivity is unknown, both genetic and environmental factors appear to interact to determine the exaggerated BP response to changes in salt intake.6, 7, 8, 9, 10, 11, 12, 13, 14, 15 Increased reactivity of BP to salt is commonly observed in subjects with obesity and associated cardiovascular risk factors.6, 7, 8, 9, 10 Insulin resistance, hypertension, increased sympathetic activity, endothelial dysfunction and abnormal nitric oxide (NO) bioactivity are among the list of factors that have been proposed to directly or indirectly lead to salt sensitivity related to obesity.2, 4, 6, 16, 17, 18 NO is involved in renal sodium homeostasis and inhibition of NO synthesis is associated with impaired natriuresis and salt sensitivity.19 Impairment in NO production has been observed in salt-sensitive subjects when exposed to a high sodium diet.14, 18 Lifestyle changes and metformin have been shown to reduce adiposity, improve insulin sensitivity, lower BP, reduce sympathetic activity and improve endothelial function.17, 20 To determine the role of acquired factors in the pathogenesis of salt sensitivity, we investigated whether treatment of obesity and associated metabolic abnormalities by means of a combined lifestyle-metformin program corrects salt sensitivity. In addition, we investigated whether the impairment in NO production induced by high salt in salt-sensitive individuals could be corrected by the lifestyle modification-metformin treatment. This work provides additional information on the mechanisms of elevated BP associated with obesity.
Materials and methods
A total of 103 voluntary adult subjects, 18–70 years of age, of either sex, living in the city of Caracas and attending our Center for Detection and Treatment of Silent Cardiovascular and Metabolic Risk Factors during 2003–2005 were evaluated and screened for study enrollment. Subjects were included if were salt sensitive, had a body mass index (BMI) >27 but lower than 39 kg/m2 and a waist circumference >102 cm for men and >92 cm for women according to the NCEP-ATP III guidelines for abdominal obesity.21 Subjects were excluded if older than 70 years, or had a history of coronary artery disease, heart failure, valvular heart disease, stroke, transient ischemic attacks, arteriosclerosis obliterans, renal or hepatic dysfunction, active disease states, severe hypertension, oral contraceptive use and serum creatinine concentration >1.5 mg/dl.
Of the 103 subjects, 48 were salt resistant, 28 had an intermediate sensitivity to salt and 27 were salt sensitive. Of the 27 salt-sensitive subjects, two had BMI <27 kg/m2, one had elevated liver function tests, and one moved out of town the week after completing the first salt-sensitivity testing. Thus, 23 met the inclusion and exclusion criteria. These subjects were invited to participate in a 12-month program of caloric restriction, aerobic exercise and metformin. Of the 23, three refused to participate and 20 were enrolled in the study. The main objective was to determine whether weight loss and correction of metabolic abnormalities associated with obesity corrects salt sensitivity. The research protocol was approved by the Central University Hospital of the city of Caracas, and by the Nova Southeastern University Review Boards. All participants gave written informed consent.
Complete history, physical examination and laboratory investigations, including hematology, chemistry, fasting lipid panel, fasting and post-load (75 g D-glucose) glucose and insulin levels, liver function tests, urinalysis, microalbuminuria and urinary sodium in 24 h urine samples, were obtained at baseline and at 12±1 months into the program. All medications were discontinued 4 weeks and antihypertensive medications 8 weeks before entering the study.
The goals of the lifestyle program were to achieve and maintain a weight reduction of at least 7% of clinical body weight through a nutritionist-designed healthy low-calorie diet (1800–2000 for men and 1600–1800 for women), where carbohydrates comprised 50–60%, fat 30% and proteins 20–30% of total calories. In addition, subjects were asked to engage in aerobic physical activity of moderate intensity, such as brisk walking, to a goal of 150 min per week. Exercise was to start slow and built-up over the first 2 months. Metformin was started at a daily dose of 500 mg for 1 week, increased to 500 mg twice daily for an additional week, reaching a maximal dose of 500 mg three times daily by the third week. Individualized adjustments in dose escalation were made based on drug tolerability. Participants were seen on a monthly basis. Body weight, BP, heart rate, general patient's health and program compliance were assessed, and benefits associated with the program were discussed and reinforced (behavior modification).
Salt-sensitivity testing was conducted twice in all subjects, once at the beginning of the study and again 12 months into the program. For the high-salt diet, subjects were instructed to continue with their dietary habits and to take 10 1-g tablets of sodium chloride daily for 1 week (∼17 g of salt/day). The high-salt period was followed by a 1-week period of low-salt diet designed to provide 2.5 g/day. The low-salt diet was formulated to contain the same caloric intake as the high-salt diet. BP readings were obtained before and at the end of each of the salt periods. If the difference in mean BP between high and low sodium weeks was equal to or greater than 10 mm Hg, the patient was deemed salt sensitive. Salt resistance was defined as increases of less than 3 mm Hg, with no changes or decreases in mean BP. Intermediate values classified subjects as of intermediate salt sensitivity.1, 4, 18, 22 Urinary sodium excretion values were obtained for assessing levels of salt intake and compliance with sodium diets.
BP was measured with a standard mercury sphygmomanometer and the cuff size was optimized for arm circumference. The same cuff and sphygmomanometer was used to measure the BP in a specific patient. Because BP was the main variable of the study, resting BP was best attained with patients lying comfortably rather than sitting for 30 min. The average of three consecutive BP readings not greater than 4 mm Hg apart from each other was employed. During the salt-sensitivity testing, the investigators obtaining the BP readings were not aware of the salt intake of the study subjects, nor of the subjects' BP at baseline. Results were provided to the study coordinator who entered the BP data in the subject's data collection form. Heart rate was estimated from a 1-min pulse assessed after the last BP measurement. Overall adiposity was assessed by body weight and BMI. BMI was calculated as the ratio of body weight in kilograms to height in meters squared (kg/m2). Waist circumference was measured in the standing position midway between the highest point of the iliac crest and the lowest point of the costal margin in the mid-axillary line. All anthropometric measurements reflected the average of two measurements.
Determination of nitrites and nitrates in urine samples
Nitrites plus nitrates were quantified employing a modification of a commercially available NO assay kit (Oxford Biomedical Research Inc., Oxford, MI, USA). To minimize dietary intake of nitrates and nitrites, 3-days before starting the high and low-salt diets, and during the 7 days of the high and low-salt diets, patients were asked to restrain from canned foods, black tea, meat and meat derivatives, and from processed food. Twenty-four-hour urine samples were collected at the end of the high- and low-sodium diets (10 days restriction from nitrate and nitrite containing foods). Urine was frozen at –60°C until required for assay. Urine was diluted before the assay. After precipitation of the protein content, the nitrates present in the urine supernatant were reduced 1:1 to nitrites by incubation with metallic cadmium beads for 24 h. The total nitrite concentration was then estimated by the Griess reaction using a multiwell microplate for reading of sample absorbance at 540 nm. This value represents the total amount of urine NO end products (nitrite+nitrate). Urine samples were processed in duplicates, and an internal standard was also employed for interassay differences.
Descriptive statistics were generated for the study population using mean (and s.e.m.) for continuous variables and proportions for dichotomous variables. Two-sample comparison for continuous variables was analyzed with the Student's t-test or paired t-test when appropriate. Triglyceride, insulin and microalbuminuria concentrations were log-transformed for statistical analysis and back-transformed for reporting. Linear regression analysis and multiple regression analysis were used to assess the relationships between continuous variables. All probability tests were two-sided. Differences were considered significant at values of P<0.05. All statistical analysis was performed with SPSS version 11.0 (SPSS Inc., Chicago, IL, USA).
Obese salt-sensitive individuals had average BP in the pre-hypertensive range, and microalbuminuria and triglyceride levels in the upper normal range (Table 1). Urinary sodium excretion averaged 159 mmol/day (nearly 9.3 g of salt/day). The combined lifestyle-metformin program achieved significant reductions in weight (9.8±0.3 kg), BMI (3.9±0.2 kg/m2), waist (11.5±0.5 cm), systolic blood pressure (SBP) (8.6±0.4 mm Hg), diastolic blood pressure (DBP) (5.5±0.4 mm Hg), triglyceride (40±5 mg/dl) and in fasting (8.3±1 μIU/ml) and post-load (20±4 μIU/ml) insulin levels. Urinary sodium excretion was 12% lower at end of treatment that at baseline (P<0.05; Table 1).
As expected, BP of obese salt-sensitive individuals was highly dependent on urinary sodium (Figure 1). Combined lifestyle-metformin treatment significantly reduced the sensitivity of the BP to salt (Figure 1). At baseline, salt-sensitivity testing, that is going from a high-salt diet (∼300 mmol of sodium/day) to a low-salt diet (∼30 mmol of sodium/day), induced large BP changes (Table 2). These BP changes were reduced by approximately 50% after weight loss. More importantly, the sensitivity of the BP to customary dietary salt (∼150 mmol of sodium/day) was corrected by the treatment (Figure 1). The differences in SBP/DBP between usual and low-salt intake levels averaged 11±1/8±1 mm Hg before treatment and only 3±1/1±0.5 mm Hg after treatment (P<0.001). Furthermore, at the low-sodium diet (∼30 mmol of sodium/day), the BP of obese salt-sensitive subjects at baseline was no different from that obtained on the same individual after weight loss and correction of metabolic abnormalities (Figure 1). The lifestyle-metformin treatment failed to decrease the BP below the low levels achieved by salt restriction (Figure 1).
The only variable that predicted the correction of salt sensitivity was the level of BP obtained at usual (r2=0.52; P<0.0003) or at high-salt intake (r2=0.59; P<0.0001). Obesity, serum lipids, fasting and post-load insulin and glucose levels, and urinary sodium or potassium excretion did not predict the improvement in salt sensitivity achieved with the combined lifestyle-metformin program.
The urinary excretion of NO metabolites (nitrates+nitrites) was dependent on the level of salt intake (Figure 2). Lower NO metabolite levels were encountered during high than during the low-salt diet. This effect was observed both for the absolute amount of NO metabolites excreted in 24 h (μmol/day), as well as for the NO-metabolite levels corrected by urinary creatinine concentrations (μmol NO/mg creatinine). The lifestyle-metformin treatment did not affect the urinary excretion of NO metabolites during the low-salt diet; however, it prevented the reduction in NO metabolite levels associated with the high-salt diet (Figure 2).
Dietary restriction, aerobic exercise and/or metformin individually are known to induce weight loss, correct metabolic abnormalities associated with central obesity, and reduce development of type II diabetes mellitus.17, 23, 24, 25, 26 In the current study, a 1-year combined program of intensive lifestyle changes and metformin induced a 13% decrease in body weight, BMI and waist circumference, and a 7% reduction in BP levels. Serum triglyceride and the insulin/glucose ratio were decreased by 30%. Interestingly, these changes were associated with marked reduction in salt sensitivity. The sensitivity of the BP to 130–150 mmol of sodium/day, which is the usual dietary salt intake of our study subjects, was practically abolished by the lifestyle-metformin treatment program. Of clinical significance was the observation that after weight loss and correction of metabolic abnormalities, BP levels at usual levels of salt intake were not significantly different from those attained with intense salt restriction. Only one study had previously assessed the effects of weight loss on salt sensitivity.6 In largely obese teenagers, weight reduction (8% of basal weight) achieved through a 20-week program of caloric restriction was found to correct salt sensitivity. These two observations (Rocchini et al.6 and current study) indicate that correctable acquired factors play an important role in the pathogenesis of salt sensitivity associated with obesity.
The present study was not designed to compare the effects of lifestyle changes versus metformin in improving salt sensitivity. Therefore, from the current data, it is not possible to determine which of the procedures was responsible for correcting salt sensitivity. Our goal was to improve the metabolic abnormalities of the obese salt-sensitive individuals to test whether salt sensitivity was correctable by interventions known to reduce adiposity, insulin resistance, and associated lipid abnormalities. As indicated above caloric restriction, increase physical activity and metformin have been shown independently to reduce cardiovascular risk factors and the development of type II diabetes.17, 23, 24, 25, 26 However, the lifestyle interventions have been shown to be more effective than metformin in preventing development of type II diabetes mellitus.26
A substantial body of evidence supports the concept that pre-hypertension and hypertension are associated with obesity; however, the reason for this association in still controversial, and may be multifactorial.15, 16, 25, 27, 28, 29, 30, 31 Increased salt sensitivity may be one of the reasons of why obese individuals have an increase prevalence of elevated BP. Adiposity, by a yet unknown mechanisms, would make the individual more salt sensitive increasing his/her BP due to dietary salt. When adiposity is reduced, BP will drop due to decreased salt sensitivity. In fact, in our obese salt sensitive subjects, the lifestyle-metformin program induced a marked decreased in BP reactivity to dietary salt and an average drop of 9 and 6 mm Hg in SBP and DBP, respectively. In support of the role of salt sensitivity in the pathogenesis of obesity-related high BP, is our observation that baseline BP levels on the low-salt diet were not reduced by the lifestyle-metformin intervention. Weight loss and correction of metabolic abnormalities only lowered BP that was elevated by customary or high levels of salt intake. These findings account for the strong association observed between salt sensitivity and baseline BP (present study). In addition to decreased salt sensitivity, healthier dietary habits and reduced caloric intake are often associated with decreased dietary salt intake. In our study group, dietary salt during treatment was modestly reduced (∼1 g/day) compared with baseline intake levels. Consequently, a combination of diminished salt sensitivity and a modest reduction in salt intake may account for the BP lowering observed with the lifestyle-metformin program employed in this study.
The study design employed does not allow determination of the relative contributions of diet, exercise and metformin to the correction of salt sensitivity observed. In addition, we do not know which of the changes, that is reduced weight, improved insulin sensitivity, correction of dyslipidemia, or others, is directly or indirectly responsible for the correction of salt sensitivity. Since weight loss simply achieved by caloric restriction corrected salt sensitivity in obese adolescents,6 it is feasible that reduced adiposity would also be the main factor by which the lifestyle-metformin treatment corrected salt sensitivity in our study. However, how reduced adiposity translates into decreased salt sensitivity is unclear. Endothelial dysfunction, insulin resistance, hyperinsulinemia, decreased sympathetic activation, decreased NO production, hyperaldosteronism, increased endothelin levels, among others, could either alone or in combination, induce a state of salt sensitivity and thus increase subject's BP.2, 6, 16, 17, 18, 19, 20, 23, 30, 31, 32, 33, 34 Correction of one or several of these factors may be responsible for the correction of salt sensitivity induced by the lifestyle-metformin treatment. Experiments in dogs suggest that increase in sympathetic activity rather than insulin resistance is responsible for the increase in BP associated with obesity.28 The lack of association observed between the magnitude of reduction in salt sensitivity and the fasting and post-load levels of glucose and insulin observed in this study are also against the role of insulin in the pathogenesis of salt sensitivity.
A limitation of the study design was the lack of a control group of untreated obese salt-sensitive individuals. However, and because of the increased cardiovascular and metabolic risk of these individuals and the long duration of the study, it was considered unethical to include such a group. A control group would have also added additional strength to the conclusion that the decrease in salt sensitivity was due to the intervention and not to within subject variability of salt sensitivity. However, such a possibility is rather unlikely not only because of the magnitude of the changes observed, but also to the consistency of the results, namely, every single study subject was much less salt sensitive after the intervention. In addition, the reliability of dietary salt-sensitivity testing had been previously assessed by repeatedly studying the effects of a high and low-salt diet. Reliability of classification of salt sensitivity as described by the kappa statistic was 0.87 implying an almost perfect strength of agreement between the first and second salt-sensitivity testing.35
NO has been shown to play a role in animal models of salt sensitivity.14, 18, 19, 32, 34, 36 NO is involved in the control of intrarenal hemodynamics and sodium homeostasis, and inhibits Cl-reabsorption at the ascending thick loop of Henle to induce natriuresis.15, 35 Treatment with an NO-synthesis inhibitor induces salt sensitivity, whereas administration of L-arginine and/or of an NO-donor corrects salt sensitivity.32, 34 However, assessing NO production in humans is not without difficulties. Urinary and plasma levels of NO metabolites, and measurements of endothelial-dependent vasodilation have been used to assess NO bioactivity in humans.1, 14, 19, 37 However, the relative contribution of systemic and of renal NO to urinary NO metabolite excretion is unclear. Downregulation of kidney endothelial36 and neuronal38, 39, 40 NO synthases has been found associated with marked reductions in NO metabolite excretion, suggesting that renal NO production is an important contributor to urinary NO metabolite excretion. In addition, endothelial dysfunction was also found associated with impairment in NO metabolite excretion.37, 40 Further, increases in inducible NO synthase activity in response to bladder inflammation has also been shown to increase in urinary NO metabolite excretion.41
Taking into account these limitations, in this study, urinary NO metabolite levels were employed as a surrogate marker of NO bioactivity. As previously shown,14, 19 we report that in salt-sensitive individuals a high-salt diet increases BP and inhibits urinary NO metabolite excretion, and that these effects are reverted by salt restriction. It thus appears that obese salt-sensitive individuals are unable to increase NO production in response to salt intake. Our findings support results obtained in animal models of obesity, such as the Zucker rats, fructose-fed rats and rats fed a high fat-refined carbohydrate diet. In these animal models, obesity is associated with high BP, salt sensitivity and impaired NO bioactivity.37, 38, 39 Altogether, these findings suggest that obesity and its associated metabolic abnormalities impair endothelial and/or renal NO production leading to sodium retention and salt sensitivity. Both salt sensitivity and NO impairment were corrected by treating adiposity and its associated metabolic abnormalities (present study). Unfortunately, we cannot determine whether the improvement in NO levels is responsible for correcting salt sensitivity, or vice versa, if the increase in NO results from correcting salt sensitivity.
In summary, we demonstrated that an intensive lifestyle-metformin treatment program decreases adiposity, improves its associated metabolic abnormalities, corrects the increased reactivity of BP to sodium and thus lowers BP in obese salt-sensitive individuals. We propose that BP lowering was most likely due to decrease in salt sensitivity. Improvement in salt sensitivity was also associated with correction of high salt-induced inhibition of NO production, a characteristic of salt-sensitive individuals. To the extent that urinary NO metabolite levels reflect the activity of the endogenous NO system, our results suggest that salt sensitivity in obese individuals may result from an inability to increase or sustain NO production in response to high salt. By restoring this response, the lifestyle-metformin treatment could ameliorate salt sensitivity. In conclusion, our findings support the view that acquired correctable factors play an important role in the pathogenesis of salt sensitivity associated with obesity, and that correction of central obesity, in addition to improving the associated metabolic abnormalities, markedly reduces the reactivity of the BP to salt, thus lowering the subject's BP.