| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2007;49:1032.)
© 2007 American Heart Association, Inc.
Original Articles |
From the Division of Nephrology, Department of Medicine, University of California San Francisco.
Correspondence to Olga Schmidlin, University of California San Francisco 1291 Moffitt Hospital, Box 0126, San Francisco, CA 94143-0126. E-mail olga{at}picasso.ucsf.edu
| Abstract |
|---|
|
|
|---|
5 mm Hg. We determined whether dietary NaCl loading induces greater increases in external Na+ balance, plasma volume, and cardiac output in SS, compared with any in SR subjects, and differential changes in systemic vascular resistance (SVR) that could account for the pressor differences between SS and SR subjects. Using impedance cardiography, we measured cardiac output and SVR daily at 4-hour intervals throughout the last 3 days of a 7-day period of low NaCl intake (30 mmol per day) and throughout a subsequent 7-day period of NaCl loading (250 mmol per day). In the 11 SS subjects, compared with the 12 SR subjects, NaCl loading induced no greater increases in Na+ balance, body weight, plasma volume, and cardiac output. Yet, from days 2 to 7 of NaCl loading, changes of MAP in SS diverged progressively from those in SR. From days 2 to 4, progressive increases of MAP in SS subjects reflected importantly impaired decreases of SVR, as judged from "normal" decreases of SVR in SR subjects. In SS and SR subjects combined, changes in both MAP and SVR on day 2 strongly predicted changes in MAP on day 7. In many normotensive blacks, vascular dysfunction is critical to the initiation of a pressor response to dietary NaCl.
Key Words: blood pressure sodium chloride electrolyte balance vascular resistance cardiac output
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
Setting
Participants were admitted to the General Clinical Research Center at University of California San Francisco for a 2-week course of study. Throughout the study, participants ate a eucaloric metabolic diet, with calories derived from protein 16%, carbohydrates 50%, and fat 34%, and drank deionized water. Per 70 kg of body weight (BW), the basal diet provided 30 mmol of Na+ and 45 mmol of K+. Physical activity was limited to daily walks on the centers 1 floor.
Intervention (NaCl Load)
Week 1 served as the baseline period. Throughout week 2, the basal diet was supplemented with NaCl, 220 mmol per 70 kg of BW per day, added to the diet as table salt and broth for a total daily NaCl intake of 250 mmol/70 kg per day (but
300 mmol per day).
Assessment of Salt Sensitivity
With an automated oscillometric device (Dinamap, Criticon Inc), programmed to obtain 5 readings within 5 minutes, BP was measured daily every 4 hours after 5 minutes of supine rest; an average daily BP was calculated. To determine salt sensitivity, the average mean arterial pressure (MAP) of days 5, 6, and 7 during NaCl restriction was subtracted from the average MAP of days 5, 6, and 7 during NaCl loading. Salt sensitivity was defined as an NaCl-induced increase in MAP of
5 mm Hg and salt resistance as an increase of <5 mm Hg. This is a more rigorous criterion than that used in our previous study in which we compared the frequency of salt sensitivity in normotensive whites with that in blacks.11
Hemodynamic Outcomes
Throughout the final 3 days of the low-NaCl period and throughout the 7-day period of NaCl loading, CO was measured at 4-hour intervals (between 6:00 AM and 10:00 PM, immediately after BP measurements) using impedance cardiography (BioZ ICG monitor, Cardiodynamics). Impedance cardiography sensors were placed on the neck and chest according to the manufacturers instructions. Data were averaged over 30-beat intervals for 5 minutes. SVR was calculated as (MAPCVP)/CO, where CVP is the central venous pressure, assumed to be 6 mm Hg, and MAP is the MAP measured during acquisition of impedance data. Daily averages of CO and SVR were calculated.
A meta-analysis of 154 studies published between 1966 and 1997, comparing impedance cardiography with a variety of reference methods, such as Fick and thermodilution, showed an overall correlation of 0.82.12 Impedance cardiography has excellent intrasubject reproducibility.13 In the current study, during 3 days of baseline measurements, the mean within-subject coefficient of variation for average daily values of CO was 2.8%, of SVR was 3.1%, of heart rate (HR) was 2.3%, and of MAP was 1.4%.
Metabolic Outcomes
BW was measured daily at 6:00 AM. Spontaneously voided urine was collected daily over 24-hour periods and analyzed for Na+ and creatinine. Net external balance of Na+ was calculated from its dietary intake and urinary output. On day 7 of the baseline period, and on days 2 and 7 of the NaCl-loading period, blood samples were obtained at 9:00 AM to determine levels of serum electrolytes, creatinine, total protein, hematocrit, plasma renin activity, and aldosterone by standard techniques. Change in plasma volume was estimated from NaCl-induced changes in both serum protein concentration and hematocrit values.
Data Analysis
Effects of NaCl loading in salt-sensitive subjects, termed SS, and in salt-resistant subjects, termed SR, respectively, were calculated as the percentage of change from baseline values where baseline value is defined as the average value of days 5, 6, and 7 of the low-NaCl period. NaCl-induced hemodynamic and metabolic changes on days 2 and 7 of NaCl loading are predetermined primary outcomes. Unpaired and paired t tests, respectively, were used for between-group (SS versus SR) and within-group (NaCl effect) comparisons. Nonparametric tests were used when variances were unequal. Relationships between variables were explored using regression analysis. Data are presented as mean and 95% confidence interval. The null hypothesis was rejected at P<0.05. P values were adjusted for multiple comparisons using the Bonferroni method. Analyses were carried out using Statistica (Statsoft Inc).
| Results |
|---|
|
|
|---|
MAP 10.4 mm Hg (6.9 to 14 mm Hg); 12 (52%) were SR, with mean
MAP 0.2 mm Hg (1.2 to 1.6 mm Hg). SS subjects were slightly older than SR subjects, but SS and SR subjects did not differ from each other with respect to admission BP, serum electrolytes, and creatinine (Table 1). All of the subjects had a BMI of <31 and, using the Modification of Diet in Renal Disease study group study equation,14 a calculated glomular filtration rate of >60 mL/min per 1.73 m2. At the end of the low-NaCl period, SS and SR subjects were not different with respect to MAP (SS: 85 [80 to 90] mm Hg; SR: 90 [84 to 96] mm Hg), CO (SS: 5.3 [4.9 to 5.8] L/min; SR: 5.5 [4.9 to 6.1] L/min), HR (SS: 65 [58 to 71] bpm; SR 72 [67 to 78] bpm), stroke volume (SS: 84 [75 to 93] mL; SR: 78 [65 to 92] mL), and SVR (SS: 15.0 [13.7 to 16.3] mm Hg/L per minute; SR: 16.0 [14.3 to 17.7] mm Hg/L per minute).
|
In SR subjects, NaCl loading induced a transient but significant decrease in MAP during the first 2 days of its initiation. MAP then increased progressively to values not different from those at baseline. In contrast, in SS subjects, MAP increased progressively from day 2 of NaCl loading (Figure 1A).
MAP on day 2 was 4.7 mm Hg in SR subjects and +2.4 mm Hg in SS subjects (P<0.01).
MAP on day 7 was +1.4 mm Hg in SR subjects and +14.4 mm Hg in SS subjects (P<0.001; Table 2).
|
|
In SR subjects, NaCl loading induced a highly significant decrease in SVR that reached its nadir, 14.7% below baseline, on day 2. In SS subjects, NaCl loading induced a much smaller, although significant, decrease in SVR that reached its nadir, 5.9%, on day 1 (Figure 1B). On day 2 in SS subjects, the difference from baseline, 4%, was not significant. In SR subjects, SVR returned to levels not different from baseline by day 6 of the NaCl load, whereas in SS subjects, SVR rose significantly to 8.2% above baseline by day 5 and remained significantly increased thereafter. On all days but day 1, changes in SVR differed highly significantly between SR and SS subjects (P<0.01 on days 2 and 7, respectively; Table 2). On each day from day 2 to 7, the average increase in SVR in SS subjects, 3.1% per day (1.6% to 4.5%), closely approximated that in SR subjects, 2.6% per day (1.5% to 3.6%), indicating that the slopes of changes in SVR were parallel in the 2 groups.
In both SS and SR subjects, NaCl loading induced near immediate, similar, transient increases in CO (Figure 1C and Table 2). In most subjects, the maximum increase occurred on day 2 or 3 and amounted in SS subjects to 10.5% (6.6% to 15.2%) above baseline and in SR subjects to 11.6% (6.8% to 16.3%; P value not significant). In both SS and SR, values of CO on day 7 of NaCl loading were significantly lower than those on day 3 and not significantly different from baseline.
In both SS and SR subjects, the increase in CO resulted from an increase in stroke volume, which, in both groups, increased progressively during the initial 3-day period of NaCl loading and, thereafter, remained increased above baseline by 10% (6% to 14%) in SS subjects and by 15% (7% to 24%) in SR subjects (SS versus SR, P value not significant). With NaCl loading in both groups, HR decreased progressively over time by 6% (2% to 10%) in SS subjects and by 9% (4% to 15%) in SR subjects (SR versus SS, P value not significant; Table 2).
The 7-day net cumulative Na+ balance was slightly but significantly greater in SR subjects, 673 (560 to 785) mmol, than SS subjects, 493 (441 to 545) mmol (P<0.05; Figure 1D). The NaCl-induced increase in BW was slightly, although not significantly, greater in SR than SS subjects (Figure 1E). In both groups, changes in BW, as well as Na+ balance, were largest in the first 2 days of, and persisted throughout, the NaCl-loading period (Table 2). By day 2 of NaCl loading, plasma protein concentration had decreased highly significantly from its baseline value in both SR subjects, 11% (13% to 8%) and SS subjects, 11% (15% to 8%; SR versus SS, P value not significant; Figure 2A). This decrease persisted unchanged on day 7 in both SR subjects, 12.5% (17% to 8%), and SS subjects 11% (14% to 8%), indicating a similarly substantial and persisting expansion of plasma volume in both groups. In both groups, NaCl loading induced a decrease in hematocrit values of
12% on both days 2 and 7 (Figure 2B).
|
During NaCl restriction but not during NaCl loading, serum levels of Na+ and Cl were significantly lower in SR than in SS subjects. In SR subjects, NaCl loading induced significant increases in Na+ and Cl on both days 2 and 7. In SS subjects, NaCl loading induced a significant increase in Cl but not in Na+ on day 7. During NaCl restriction, levels of serum K+ were similar in SS and SR subjects. In SS but not in SR subjects, NaCl loading induced a significant decrease in serum K+. Baseline values and NaCl-induced changes in plasma renin activity and aldosterone on days 2 and 7 of NaCl loading did not differ between groups.
In all of the subjects combined, changes in MAP on day 2 were strongly predictive of those on day 7 (R=0.756; P<0.001), as well as of the average pressor response of days 5, 6, and 7 (Figure 3A). In fact, the pressor response at 6:00 PM on day 1, that is, 9 hours after initiating the NaCl load, when subjects had ingested only
150 mmol of NaCl per 70 kg of BW, was strongly predictive of the pressor response on day 7 (R=0.726; P<0.001), as well as of the average pressor response of days 5, 6, and 7 (Figure 3B). Similarly, the NaCl-induced changes in SVR on day 2 were highly predictive of the pressor response on day 7 (R=0.713; P<0.001), as well as of the average pressor response of days 5, 6, and 7 (Figure 3C). Changes in CO were not predictive of changes in MAP.
|
During the last 3 days of NaCl loading, SVR was increased in all 11 of the SS subjects. In 2 of them, the increase in SVR was accompanied by a slightly greater increase in CO, and in 2 SS subjects, SVR and CO were increased similarly (Figure 4). In all but 2 SR subjects, SVR was decreased. In SR subjects, but not in SS subjects, there was a strong inverse relationship between the changes in CO and SVR (R=0.873).
|
| Discussion |
|---|
|
|
|---|
However, over the initial 3-day period of NaCl loading in the current study, SVR decreased sharply in the SR subjects but changed little in the SS subjects. Indeed, a decrease in MAP attended the initial decrease in SVR in SR subjects. Similarly discordant effects on SVR and MAP1618 have also been noted in the first 3-day period of NaCl loading in animal models of salt-sensitive hypertension and salt-resistant controls. That SVR might be an important determinant of the initial pressor effect of dietary NaCl does not comport with the traditional nephrogenic formulation of salt sensitivity.4,5 This formulation holds that during the initial several days of NaCl loading, SVR remains normal, that is, little changed from baseline, hence, not determining of the initial pressor effect of dietary NaCl, sharp increases in CO fully accounting for that effect.4,5 The main observations cited in support of this formulation are those of Guyton et al19 in their classic studies of the 70% nephrectomized dog in which NaCl loading in an amount "about five times normal" induced immediate increases in MAP and CO several times those of the current study. Over the first several days in these studies, SVR changed relatively little, decreasing slightly, compared with the 40% increase in CO, which fully accounted for a similar increase in MAP during this initial period of NaCl loading. Subsequently, CO progressively decreased to values approaching those at baseline, while a reciprocally increasing SVR became the dominant determinant of the sustained pressor effect of dietary NaCl. The increase in SVR is formulated to be the normal autoregulatory response to an increase in CO that excessively perfuses tissue. This increase in SVR is the only way formulated that SVR can be a determinant of the pressor effect of NaCl.
However, in the studies of the 70% nephrectomized dog, control observations were not reported. In later studies of 2 nonsurgical animal models of salt-sensitive hypertension, the angiotensin IIinfused dog16 and the Dahl SS rat,17 Cowley and his colleagues observed that, in both the normal canine control and the Dahl SR rat, there occurred NaCl-induced increases in CO similar to those induced in the salt-sensitive animals. In further accord with the current observations, in both salt-resistant control animals, the increases in CO were offset by a "reciprocal reduction" in SVR, which accounted for the nonoccurrence of a NaCl-induced pressor effect in the normal dog16 and a transient decrease in BP in the Dahl SR rat.17 Ganguli et al18 made similar observations in Dahl rats. In the SR animals, the observation that the NaCl-induced increase in CO did not evoke an increase in SVR would "appear to contradict the theory of autoregulation."16 If it is assumed that the "reciprocal reduction" of SVR observed in the SR animals represents a normal vasodilatory response to dietary NaCl, the observations taken as a whole permit the inference that, in the SS animals, an impaired vasodilatory response to NaCl loading can be a major determinant initiating the pressor effect of dietary NaCl.20
The current studies are the first comparing SS and SR normotensive (or hypertensive) humans with respect to the effect of NaCl loading on the time courses of changes in BP, CO, and SVR. Thus, given that SVR decreased sharply during the early course of NaCl loading in the SR subjectsas might define a normal vasodilatory response in this population of study subjects, under the conditions of this studythe observation in the SS subjects that SVR changed little during this period does not indicate that SVR remained normal but suggests instead a systemic vascular dysfunction that is expressed as an impaired vasodilatory response to dietary NaCl loading.
At pointed odds with the traditional formulation of salt sensitivity, the initial pressor effect of dietary NaCl depends on such a vascular dysfunction. Only in combination with it could the NaCl-induced transient increase in CO have induced an initial pressor effect in the SS. Because dietary NaCl induced similar transient increases in CO in the SS and SR subjects, dietary NaCl could not have induced its initial pressor effect had it also induced a decrease in SVR like that it induced in the SR subjects. In single-point echocardiographically derived measurements of SVR made on the last day of a 4-day period of NaCl loading, Sullivan et al21 observed this variable to decrease in SR subjects and to change relatively little in SS subjects. Because at that time, as in the current study, CO was similarly moderately increased in SR and SS subjects, the authors concluded the pressor effect of NaCl in SS subjects reflected impairment of a normal NaCl-induced decrease in SVR. Indeed, in the currently studied SR subjects, significant decreases in SVR persisted on days 4 and 5 of NaCl loading but not thereafter, when SVR was substantially increased in the SS subjects, and that increase was, on average, the major determinant of the pressor effect of NaCl. It might be argued that, in the SR subjects, an NaCl-induced decrease in SVR caused the increase in CO by entraining renal retention of NaCl that expands plasma volume. However, in the normal dog16 and in the SR humans studied by Sullivan et al,21 as in the currently studied SR subjects, HR did not increase with NaCl loading when SVR decreased, as such a mechanism would predict, at least early in the course of NaCl loading. Indeed, in the currently studied SR subjects, as in the SS subjects, HR decreased with NaCl loading.
By the end of the NaCl-loading period, SVR was increased in all 11 of the SS subjects, but in 2 of these, a persisting increase in CO was the dominant determinant of the increase in MAP. In another 2 SS subjects, the increase in MAP appeared to be determined as much by an increase in CO as by that in SVR. Thus, within the SS group, there may be subgroups that differ qualitatively or temporally from each other with respect to hemodynamic response to NaCl loading, much as proposed by Sullivan et al,21 who suggested that some SS subjects are "output responders," whereas others are "resistance responders."
Yet, in most of the currently studied subjects, a major vascular determinacy of the pressor effect of dietary NaCl loading would seem to operate from its outset. In all of the subjects combined, the changes induced in SVR and in MAP on day 2 of NaCl loading are highly predictive of the changes induced in MAP on day 7. Indeed, on the first day of NaCl loading in all of the subjects, the change in MAP occurring at 6:00 PM, relative to the MAP occurring at this time during the preceding 3-day period of NaCl restriction, is also highly predictive of the change in MAP induced by NaCl loading on day 7. In the SR subjects, NaCl loading induced over the initial 2-day period a progressive decrease in both SVR and MAP, indicating the operation of a robust antipressor systemic vasodilatation in response to an increasing plasma volume and CO. By contrast, by day 2 of NaCl loading in the SS subjects, SVR and MAP were not significantly changed from baseline despite similar increases in plasma volume and CO. Yet, from day 2 to 7 of NaCl loading, SVR increased in parallel in SR and SS subjects, which suggests that by day 2, the pressor effect of dietary NaCl in the SS subjects is modulated by, and importantly determined by, a strong opposing vasodilatory effect of dietary NaCl. The current observations provide evidence that in many healthy normotensive blacks, an impaired vasodilatory response to dietary NaCl can be a major pathogenic factor in the pressor effect of dietary NaCl. A "vasogenic" initiation of salts pressor effect might in some instances have a sympathetic or "hormonal-sympathetic" basis, several possible mechanisms of which have been described.10,20,2224
That Na+ balance in the SS was not greater than that in the SR might seem inconsistent with the capacity of an increase in renal perfusion pressure to effect an increase in urinary Na+ excretion. This capacity assumes a relatively unhindered transmission of the perfusion pressure. It has been reported that in blacks with either SS hypertension25 or normotensive salt sensitivity,8 but not in those without salt sensitivity, NaCl loading over a week induces a decrease in renal blood flow and hence an increase in renal vascular resistance. In the salt-sensitive normotensive subjects, the reported extent of the decrease in renal blood flow and the increase in renal vascular resistance varied directly with that of the pressor effect of NaCl. Thus, with such an NaCl-induced renal impairment in the current study, pressure natriuresis might be quite subtle and take the form of a less positive Na+ balance than that occurring without the pressor effect in SS subjects, as the current data might suggest. Such a renal impairment could well be part of a more general vascular dysfunctional response to dietary NaCl loading in the salt-sensitive state.
Limitations
The current studies were performed in a selected group of people, normotensive middle-aged blacks, and thus the results may not apply to other populations or instances of salt sensitivity. However, for all of the animal models of salt sensitivity in which a salt-resistant control group was studied along with the salt-sensitive group, results similar to the current ones were reported.1618
We calculated SVR from measured values of MAP and CO and an arbitrary constant value of CVP, as is commonly done, particularly in healthy individuals in whom a relatively modest increase in circulatory volume over several days is unlikely to induce a substantial increase in CVP. In calculating SVR, other investigators have also made the tacit assumption that CVP or right atrial pressure does not change with NaCl loading.1618 Yet, in Dahl SS but not in Dahl SR rats, right atrial pressure increased slightly but significantly by 2.8 mm Hg when NaCl intake was increased 16-fold for 5 days and by 1.5 mm Hg when it was increased 7-fold,26 as in the current study. Such a small NaCl-induced increase of CVP in SS subjects but not in SR subjects would have resulted in smaller NaCl-induced group differences in SVR than those reported. However, negation of NaCl-induced group differences in SVR would require substantially greater increases in CVP, given the fact that group differences in NaCl-induced changes of BP were >10 mm Hg.
Perspectives
That the initiation of some human salt sensitivity associates more with diminished (initial) systemic vasodilation in response to dietary NaCl loading than with an abnormally enhanced renal reclamation of Na+ and Cl expands the scope of potential mechanisms of human salt sensitivity and thereby the range of potential preventive and therapeutic approaches to much of human hypertension.
| Acknowledgments |
|---|
Sources of Funding
The studies were carried out in the General Clinical Research Center, Moffitt/MZ Hospital, University of California, San Francisco, with funds provided by the National Center for Research Resources (M0 RR-00079), US Public Health Service.
This research was supported in addition by National Institutes of Health/National Heart, Lung, and Blood Institute grant RO1-HL64230 and gifts from the Emil Mosbacher, Jr, Foundation.
Disclosures
None.
Received November 15, 2006; first decision December 8, 2006; accepted February 21, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. Schmidlin, A. Forman, A. Sebastian, and R. C. Morris Jr Sodium-Selective Salt Sensitivity: Its Occurrence in Blacks Hypertension, December 1, 2007; 50(6): 1085 - 1092. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2007 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |