Salt Sensitivity Is Associated With an Increased Mortality in Both Normal and Hypertensive Humans

Abstract

While a variety of studies have demonstrated the heterogeneity of the blood pressure response to salt loading and depletion, none has examined the relationship of these responses to mortality. We conducted a follow‐up study of 430 normal and 278 hypertensive subjects studied as long ago as 28 years by techniques to assess their blood pressure responses to salt and volume expansion and depletion (salt sensitivity). We found that 123 (21 %) of our study subjects had died in the interval since the initial study. The following measures were found to be significantly associated with the risk for death: blood pressure (systolic, diastolic, mean arterial, and pulse pressures), hypertension, age, salt sensitivity, baseline renin levels (inverse), and body mass index (but not body weight). Normotensive, saltsensitive subjects had a survival that was no different than that of hypertensive subjects. Only initially normotensive, salt‐resistant individuals had improved survival. Studies are continuing in order to elucidate the mechanisms by which salt sensitivity, even in normotensive subjects, may lead to reduced survival.

Hypertension, as well as a variety of other factors, is known to be associated with an increased risk for disability and death from cardiovascular disease. While many other risk factors for cardiovascular disease are also frequently found in hypertensive patients (i.e. dyslipidemia, insulin resistance and diabetes mellitus, left ventricular hypertrophy, and cigarette smoking), the role of another frequent finding in hypertensives, salt sensitivity of blood pressure, has not been previously identified as a risk factor for increased mortality. Moreover, salt sensitivity of blood pressure has also been recognized in normotensive subjects, ¹ but the long‐term implications of this finding have not been previously recognized. We have recently had the opportunity to conduct a follow‐up study of a large cohort of normal and hypertensive subjects in whom the blood pressure responses to salt and volume loading and depletion were initially characterized as long as 28 years previously. ¹ , ² We found that such factors as systolic, diastolic, mean arterial, and pulse pressure, as well as the conventionally defined hypertension, age, renin suppression, and body mass index were all associated with reduced survival. Our findings indicate that salt sensitivity, even among initially normotensive subjects, is associated with increased risk for death.

METHODS

Our study population of 708 subjects ranged in age from 18–80 years (34±15, means±standard deviation) when initially studied and included 313 (44.2%) women and 178 (25%) African Americans, 278 (39.3%) of whom were known to be hypertensive at the time of initial study, and 338 (50%) who were found to be salt sensitive. ¹ At the time of the initial study, all hypertensives had discontinued antihypertensive medications for at least 2 weeks and no women were receiving estrogen therapy. Blood pressure responses to salt and volume loading and depletion were assessed by previously described techniques. ¹ For the purposes of the present analyses, we used blood pressure measured at baseline, before administration of saline, at 8 a.m., with the subject seated.

The studies were approved by the Indiana University School of Medicine‐Clarian Institutional Review Board and informed consent was obtained from all subjects at the initial study. For the followup study, information regarding vital status was obtained from the subject, from next of kin, and/or from public death registries (the National Death Index and the Indiana State Board of Health Vital Statistics Bureau). Subjects were located by hospital records, by public directories, and by using the services of a locating firm (Find People Fast). A form letter inviting participation in follow‐up studies (including a health questionnaire and a brief outpatient study) was sent to the last known current address. Health information and cause of death were obtained from death certificates and from the next of kin, when available. Subjects were classified as alive, dead, or status unknown on the basis of the most recent information.

Raw data identifying subjects by name was kept in a locked room with limited access. Statistical analyses were performed using t tests, the Fisher exact test, forward stepwise logistic regression, and forward stepwise Cox proportional hazards regression. Kaplan‐Meier survival curves were constructed and compared with the use of a log rank test. Data are presented as means±standard deviation. A value of p<0.05 was considered significant. For hazard analyses, we present odds ratios (OR) and 95% confidence intervals (CI).

RESULTS

We obtained information on 596 (85%) of the subjects from the original cohort of 708 subjects, of whom 123 (20.6%) had died. The 112 (15%) subjects who could not be located or about whom no information could be obtained were significantly younger (28.6±10.1 vs. 34.8±15.0 years; p<0.001) and more likely to be African American (44.6% vs. 22.1%; p<0.001) than those who were located. The subjects who had died were significantly (p<0.001) older (48±14 vs. 31±13 years), and were more likely to have been hypertensive (63.4% vs. 33.8%), and salt sensitive (73.4% vs. 44.9%) at the time of the initial study than those who survived. There were no significant differences in the race or gender distributions of survivors compared to decedents. Subjects who died were found to have a higher initial body mass index (26.9±5.2 vs. 25.5±5.6; p=0.027), but were not different in body weight (75.5±15.2 vs. 73.5±17.1 kg; p=0.238).

The men were significantly taller (p<0.001) than the women, but there were no height differences between decedents and survivors.

Baseline renin values were significantly (p<0.01) lower in those who subsequently died than in the survivors (5.5±5.3 mm Hg vs. 7.1±6.2 mm Hg ng/mL/180 min). Those who had died had significantly (p<0.001 for all) higher blood pressures when initially studied (systolic, 136±28 mm Hg vs. 119±15 mm Hg; diastolic, 87±17 mm Hg vs. 76±13 mm Hg; mean arterial, 103±20 mm Hg vs. 90±14 mm Hg; and pulse pressure 49±15 mm Hg vs. 43±11 mm Hg). Forward stepwise logistic regression analysis with death as an outcome yielded four significant variables: baseline mean arterial blood pressure (OR, 1.03; CI, 1.01–1.04; p<0.001), age at initial study (OR, 1.08; CI, 1.06–1.10; p<0.001), salt sensitivity (OR, 1.73; CI, 1.02–2.94; p=0.042), and male gender (OR, 1.91; CI, 1.15–3.17; p=0.012). Cox proportional hazards regression was used to examine time to death and yielded age at initial study (OR, 1.06; CI, 1.04–1.07; p<0.001) and baseline diastolic pressure (OR, 1.03; CI, 1.01–1.04; p<0.001) as independent predictors. A Kaplan‐Meier survival analysis indicated a hazard ratio of 0.56 (CI, 0.38–0.82; p=0.002), demonstrating that being normotensive and salt‐resistant at the time of initial study significantly improved survival. ² The survival curve for normotensive, saltsensitive subjects was no different from that of those who were hypertensive at the time of the initial study, both salt sensitive and salt resistant.

DISCUSSION

This unique follow‐up study of almost 30 years in a large group of normotensive and hypertensive subjects of varied demographic characteristics has provided evidence for salt sensitivity of blood pressure as a novel risk factor for mortality. Our findings have confirmed long‐recognized factors, such as age, gender, body mass index, and blood pressure as being associated with an increased risk for death. In addition, we also confirm the emerging importance of pulse pressure as a newly recognized risk factor for mortality. Our finding of an increased risk of salt sensitivity of blood pressure among initially normotensive subjects is a striking new finding. Previous reports in a small group of Japanese hypertensives suggest an increase in cardiovascular events in salt‐sensitive subjects. ³ However, to our knowledge, this is the first study to demonstrate such an increase in risk for those who were initially normotensive.

What mechanisms may have been operative to induce this increase in risk attributable to salt sensitivity of blood pressure? Salt sensitivity of blood pressure is known to increase with increasing age. ⁴

However, even after consideration of the effect of age, salt sensitivity had an independent contribution to the risk for mortality, as indicated by the stepwise logistic regression analysis. Salt sensitivity of blood pressure is also more frequently observed in hypertensive than in normotensive persons. ¹ Again, the effect of salt sensitivity was independent of blood pressure, according to the stepwise regression analysis. Hypertensive (but not normotensive) African Americans are also more frequently found to be salt sensitive, ¹ but race did not influence the risk for mortality in our study. We have previously reported, in a small group of our subjects, that salt sensitivity was associated with a significantly greater increase in blood pressure over a 10‐year period than was salt resistance. ⁴ These findings, as well as epidemiologic observations, ⁵ suggest a link between salt sensitivity of blood pressure and the age‐related increase in blood pressure typical of industrialized (i.e., high dietary salt intake) societies. While it is not possible to ascertain the blood pressure levels of our study subjects before their deaths, it is tempting to speculate that the enhanced risk of death related to salt sensitivity of blood pressure, even among initially normotensive persons, may have been related to an increased blood pressure subsequent to their initial study. In support of this contention is the observation that roughly one half (60) of the deaths were from cardiovascular causes. When this subgroup was analyzed separately, prior salt sensitivity was found in 73.7%, compared to 47.6% of those who died of non‐cardiovascular causes, which is a significant (p<0.001) difference.

Despite previous reports indicating that renin levels are associated with an increased risk for myocardial infarction, ⁶ we could not confirm a direct relationship between renin and mortality in our study. To the contrary, the relationship between renin and mortality was inverse in our study, consistent with relative suppression of renin in salt‐sensitive subjects. ¹ Further studies are in progress to assess the link between renin levels and fatal and nonfatal myocardial infarction.

What are the clinical implications of these findings? These data indicate that the identification of salt‐sensitive individuals, even prior to the development of hypertension, may permit a beneficial reduction in blood pressure through reduction of dietary salt intake. Moreover, the recent findings of the Dietary Approaches to Stop Hypertension (DASH)‐II study ⁷ indicate that dietary salt restriction in combination with increased intake of fresh fruits and vegetables and low‐fat dairy products can reduce blood pressure, and that the regimen is relatively easy to follow. Ultimately, studies to identify genetic markers associated with salt sensitivity offer the best approach to the identification of subjects most likely to benefit from such approaches. In the meantime, salt sensitivity can be identified by practical approaches, such as careful observation of the blood pressure response to a trial of diuretic administration over a period of 7–10 days or to reduced dietary salt intake (aiming for 80 mmol/day), monitored by measurement of sodium in a 24‐hour urine sample (at steady‐state in individuals without renal disease or taking medications affecting renal function or sodium handling, urinary excretion of sodium is equal to intake).

Acknowledgments: I should like to acknowledge the expert professional assistance of my colleagues, Dr. Naomi S. Fineberg, Dr. S. Edwin Fineberg, Dr. Morris Weinberger, Urban Wagner, RN, and Connie Clark, RN in the design, conduct, and analyses of these studies. Support for these studies was provided, in part, by grants from the US Public Health Service, by grants HL‐14159 (Specialized Center of Research [SCOR]‐Hypertension), HL‐57826, RR‐00750 (General Clinical Research Center), and DK‐20542 (Diabetes Research and Training Center) and by the Department of Veterans Affairs Career Development Program.