| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Monell Chemical Senses Center (M.P.), Philadelphia, Pennsylvania 19104; and Department of Medicine (R.T.) and Monell Chemical Senses Center and Department of Medicine (K.L.T.), University of Pennsylvania, Philadelphia, Pennsylvania 19104
Address all correspondence and requests for reprints to: Dr. Karen Teff, Monell Chemical Senses Center, 3500 Market Street, Philadelphia, Pennsylvania 19104. E-mail: kteff{at}pobox.upenn.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective of this study was to determine whether experimentally induced glucose intolerance decreases HRV in a circadian relevant manner in healthy individuals.
Design: This was a within-subject, randomized design study with subjects infused for 48 h with saline (50 ml/h) or 15% glucose (200 mg/m2·min). HRV was evaluated using time domain measurements taken over the 48-h period. Blood pressure and heart rate were monitored, and blood samples were taken.
Setting: This study was performed at the General Clinical Research Center of the Hospital of the University of Pennsylvania.
Patients: Sixteen healthy subjects (eight men and eight women; 18–30 yr old; mean body mass index, 21.7 ± 1.6 kg/m2) were studied.
Results: After glucose infusion, mean plasma glucose was increased by 16.8% (P < 0.0001), and plasma insulin was increased by 99.4% (P < 0.0001) compared with after saline infusion. Significant decreases in homeostasis model assessment indicated a decrease in insulin sensitivity (saline, 0.52 + 0.13; glucose, 0.34 + 0.12; P < 0.0001). The nocturnal root mean square successive difference, an index of cardiac vagal activity, was significantly decreased (P < 0.01), and nocturnal HR (P < 0.001) and blood pressure were significantly elevated (saline, 107.4 ± 2.7; glucose, 112.5 ± 3.3 mm Hg; P < 0.05) compared with the saline control. The change in homeostasis model assessment due to glucose infusion was significantly correlated with the change in root mean square successive difference (r = 0.48; P < 0.01).
Conclusions: Prolonged mild hyperinsulinemia disrupts the circadian rhythm of cardiac autonomic activity. Early changes in the neural control of cardiac activity may provide a potential mechanism mediating the pathophysiological link between impaired glucose tolerance and cardiovascular disease.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Insulin resistance and compensatory hyperinsulinemia have been proposed in the etiology of metabolic disease (8) and, in fact, are also associated with dysregulation of the autonomic nervous system (9, 10). Increases in muscle SNS activity are positively correlated with waist circumference and indices of insulin resistance (11, 12). Furthermore, elevated insulin levels can increase SNS activity, as demonstrated by systemic increases in norepinephrine turnover (13) and muscle SNS activity (14, 15). To investigate the effect of hyperinsulinemia on HRV, a number of studies have used acute insulin infusions, typically the euglycemic, hyperinsulinemic clamp (16, 17, 18). The findings from these studies are inconsistent, possibly because of the relatively short length of the glucose infusions (2–6 h) and the limited duration of the HRV measurements, which has typically been in the 5- to 15-min range.
In clinical populations, 24-h periods of monitoring are frequently used to fully assess cardiovascular activity (19). The longer period of monitoring allows a more fully integrated view of cardiovascular function and provides insight into nocturnal activity, the time of day at which HRV is at its highest. This is because cardiovascular autonomic activity exhibits a circadian rhythm with systolic blood pressure (SBP) and HR, indices of cardiac SNS, elevated during the day and low at night, whereas parasympathetic activity, as measured by HRV, is low during the day and increased at night. Recent data suggest that both the absence of a decrease in nocturnal levels and a premature early morning rise in blood pressure (BP) are related to increased target organ damage and cardiovascular events (20, 21). Thus, chronobiology may provide insight into the pathophysiological state of the cardiovascular system (22, 23). The objective of the present study was to determine whether prolonged physiological increases in plasma insulin and/or glucose induced by a 48-h glucose infusion could significantly alter the circadian profile of HRV in normal healthy individuals independently of changes in weight.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Sixteen lean subjects (eight men and eight women), 18–30 yr of age (mean, 22.3 ± 3.7 yr), with body mass indices ranging from 18.1 to 24 kg/m2 (mean, 21.7 ± 1.6 kg/m2) and a percent body fat of 16.9 ± 5.4% (range, 7–18% for men and 16–26% for women), participated in this study. Subjects were recruited through newspaper advertisements, and each subject gave written informed consent before entering the study. Two women initiated, but then dropped out of, the study, and their data were not included in the dataset. Of the final group, hormonal data were collected for all 16 subjects, but HRV and HR data were only collected for 15 of the subjects, because a malfunction occurred in one of the Holter monitors on 1 d. After a telephone interview to assess eligibility, subjects underwent a physical examination, including an electrocardiogram and medical history to ensure they had no chronic illnesses, including diabetes, hypertension, and abnormal heart rhythms. Smokers, subjects taking prescription medications, and subjects with a family history of diabetes or hypertension were also excluded from the study. A blood sample was drawn after an overnight fast to evaluate clinical blood chemistry, and subjects were permitted to participate if fasting plasma glucose was less than 110 mg/dl, hemoglobin level was more than 12 mg/dl, and BP was less than 140/90 mm Hg. To control for the effect of the menstrual cycle, all experimental testing in women took place within the follicular phase of the menstrual cycle. These studies were approved by the institutional review board of the University of Pennsylvania.
Experimental protocol
Each subject underwent two experimental conditions administered in a randomized fashion over a 2-month period: 1) 48-h glucose infusion and 2) 48-h saline infusion. On the evenings before the experimental days, subjects arrived at the General Clinical Research Center (GCRC) at the Hospital of the University of Pennsylvania at 1700 h. Subjects were given dinner at 1800 h and a snack at 2000 h, after which they remained fasting until the following morning. During their stay in the GCRC, subjects were permitted to eat three meals and a snack per day (ad libitum), but were only allowed food prepared by the GCRC kitchen and selected from a preset menu. Subjects were permitted to walk around the GCRC, but were not allowed to leave the unit. Physical activity was generally limited and remained constant throughout the 48-h period.
At 0700 h, after an overnight fast, two iv catheters were inserted in opposite arms, and an iv infusion of either saline (0.9% saline at a rate of 50 ml/h) or glucose (15% dextrose solution at a rate of 200 mg/m2·min) was initiated and sustained for a 48-h period. One iv line was used for infusion, and the other for blood sampling. Glucose infusions were supplemented with 8 mEq/liter potassium phosphate to prevent hyperinsulinemia-induced hypokalemia. Upon initiation of the infusion at 0700 h, subjects were connected to a Holter monitor for measuring HRV and HR. Blood samples were taken approximately every 2 h (except between 2300 and 0700 h, when they were collected every 4 h). BP measurements were made at the time of blood withdrawal, i.e. every 2 h during the time subjects were awake (eight measurements) and every 4 h during the sleep periods (two measurements) using a Dynamap (Datascope Corp., Montvale, NJ). BP was measured in the dominant arm while the subjects were supine. Blood was collected in tubes containing EDTA, and the samples were kept on ice for not longer than 1 h. Samples were then centrifuged and stored at –70 C for later assay. Upon initiating the infusion, two sequential 24-h urine collections took place from 0600 h one morning to 0600 h the following morning. A 15-ml aliquot of each 24-h urine sample was taken and processed for measurement of urinary catecholamines.
HRV
HRV was measured using an ambulatory digital Holter monitor (RZ153 series, seven-electrode, three-lead mode, Rozinn Electronics, Inc., Glendale, NY), which collected 48-h electrocardiographic recordings. Each subject underwent continuous monitoring with three-channel recordings throughout the study. All recordings were registered on memory cards and contained at least 47 h of analyzable electrocardiographic data (mean registration time, 47:45 ± 2:03 h). The Holter electrocardiogram (ECG) leads were sampled at a continuous 180 samples/sec (digitizing rate). Recordings were downloaded to a Holter for Windows analysis system (Rozinn Electronics, Inc.) for analysis after the recording was completed. Continuous data were analyzed by the software and binned into 5-min segments.
R wave beats (QRS complexes) and RR intervals were identified. Artifactual noise and ectopic beats (supraventricular and ventricular) were recognized and eliminated from the analysis by the Holter software. The RR intervals preceeding and succeeding intervals with noise or ectopic beats were also excluded from the analysis. The instantaneous heart period function during those periods of excluded data was estimated by linear interpolation. At least 80% of the RR interval data in a 5-min epoch had to constitute normal RR intervals or that epoch was excluded from the total analysis (1). To find and correct errors not detected by commercial instruments, all recordings were visually examined and manually overedited to verify beat classification. Individual cycle lengths between the QRS complexes were determined, and variables describing the statistical distribution of the set of all cycle lengths, such as the mean and SD, were calculated over the whole recording or segments of the recordings (i.e. night vs. day).
Data were analyzed either by acquiring a mean reading of the entire 48-h period of recording or by averaging 5-min segments of the recording over 1-h periods. This was done by transporting the data into Microsoft Excel (Microsoft Corp., Redmond, WA) and averaging every 12 successive 5-min intervals to obtain hourly mean values of the HRV parameters. To investigate the effect of experimental manipulation on the circadian rhythm of HRV, different time periods were analyzed: 1) hourly over the 48-h period (see Fig. 2
), 2) active (0700–2300 h) vs. sleep periods (2300–0700 h), and 3) 12-h blocks (see Table 1; 0700–1900 h, day; 1900–0700 h, night). Night to day differences were determined by subtracting the average daytime values from the average nighttime values.
|
|
Biochemical analysis
Plasma immunoreactive insulin was measured in duplicate by a double-antibody RIA. The antibodies were purchased from Linco Research, Inc. (St. Charles, MO). The insulin antibody used is the human-specific antibody from Linco with no cross-reactivity (<0.2%) to human proinsulin or the primary circulating split form. Analysis of insulin was performed by the Diabetes Research Center of the University of Pennsylvania. The interassay coefficient of variation for insulin is 8%, and the intraassay variation is 6%. Plasma glucose was analyzed by an automated glucose/lactate monitor (YSI, Inc., Yellow Springs, OH) at the Monell Chemical Senses Center (Philadelphia, PA). Free fatty acids were measured by enzymatic assay using the Wako kit (Wako Chemicals, Richmond, VA) at the Monell Chemical Senses Center. Urinary norepinephrine and epinephrine were measured by HPLC with electrochemical detection at the GCRC of the Hospital of the University of Pennsylvania.
Statistical analysis
Significant differences in plasma glucose and insulin were determined by comparing 48-h means of each variable and conducting a paired t test. Using repeated measures ANOVA, each HRV time domain variable was analyzed to determine whether there were significant time, treatment, or time by treatment interactions. Post hoc analysis was then conducted on comparable time points using either Tukey’s or Bonferroni post hoc analysis. Twelve- and 24-h mean segments of time domain variables, day-night (and night-day) differences, as well as 24-h collections of urinary catecholamines were also evaluated using repeated measures ANOVA, followed by post hoc testing. Mean night and mean day values for SBP were compared between treatments using a dependent t test, because the number of measurements taken during the night was not equal to that taken during the day. Statistical significance for all measurements was P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The prolonged glucose infusions resulted in an approximately 16% increase in plasma glucose over the 48-h period compared with saline infusion. Mean glucose levels over 48 h were 96.3 ± 6.8 mg/dl during the saline infusion compared with 112.6 ± 6.7 mg/dl during the glucose infusion (Fig. 1, left graph; P < 0.0001). These relatively modest increases in plasma glucose fell within the range of impaired glucose tolerance and were associated with sustained increases in plasma insulin.
|
, right graph; P < 0.0001). During the saline infusion, mean glucose and insulin levels were similar on d 1 and 2. In contrast, after glucose infusion, mean glucose levels were significant lower on 2 than on d 1 (P < 0.001), and mean insulin levels were elevated (P < 0.01). Three hours after termination of the infusion, plasma glucose levels had returned to baseline (83.8 ± 7.2 mg/dl, glucose vs. 81.8 ± 7.4 mg/dl, saline). Plasma insulin levels remained slightly elevated but within the normal range 3-h after termination of the glucose infusion (glucose, 14.6 ± 4.6 µU/ml; saline, 9.4 ± 2.6 µU/ml; P < 0.0001). Free fatty acid levels were completely repressed by the glucose infusion due to the hyperinsulinemia (saline, 0.23 ± 0.06 mM; glucose, 0.12 ± 0.03 mM glucose; P < 0.00003; data not shown). Effect of 48-h glucose and saline infusions on indices of HRV during infusion periods
Time domain measurements for each 12-h period (0700–1900 and 1900–0700 h) during the saline and glucose infusions are shown in Table 1. Illustrated in the table are four indices of HRV, RMSSD and pNN50, two measures of parasympathetic activity, and SDNN and SDANN, mixed measures of parasympathetic and sympathetic activity. Significant differences were found between day and night values for all parameters on d 1 and 2 of the saline infusion, illustrating the circadian nature of the measurements. In contrast, although significant diurnal differences were found on d 1 of the glucose infusion, no significant differences in the day compared with the night values were observed on d 2 of the glucose infusion. Thus, by the second day of the glucose infusion, circadian differences were attenuated.
RMSSD, the most specific index of cardiac vagal activity, exhibited a typical circadian pattern, being lower during the day and higher at night during both days of saline infusion (Fig. 2). However, as shown in Fig. 2 and in Table 1, although there were no significant differences in the RMSSD circadian rhythm between the glucose and saline infusions during the first 24 h of the glucose infusion, by the second 24 h of the infusion, the nocturnal rise was blunted [treatment by time interaction: F(47, 611) = 2.21; P < 0.0001]. In fact, night to day differences in RMSSD (Fig. 3) were significantly attenuated on the second day of the glucose infusion compared with the saline infusion [treatment by time interaction: F(3, 39) = 6.17; P = 0.001]. To be able to compare the RMSSD time periods to those of the BP measurements (see below), we also analyzed RMSSD over the active (0700–2300 h) vs. sleep (2300–0700 h) periods. Similar to those during the 12-h periods, RMSSD values were significantly higher during the sleep period compared with the active period during the saline infusion, and a significant decrease in RMSSD was found during the second day of the glucose infusion [treatment by time interaction: F(3, 39) = 6.2; P < 0.002; data not shown]. Thus, all three methods of analysis, 1) hourly (Fig. 2), 2) 12-h time blocks (Table 1), and 3) awake (0700–2300 h) vs. sleep periods (2300–0700 h), demonstrate that the glucose infusion significantly decreased cardiac vagal activity during the second night.
|
). Effect of 48-h glucose and saline infusions on BP and HR during infusion periods
Figure 4 (left graph) illustrates the circadian variation in SBP. Although a decrease in nocturnal SBP was evident during the saline infusion, the magnitude of decline was relatively small compared with previous reports (4, 20). Although significant time effects were found [F(3, 45) = 6.63; P < 0.001], post hoc analysis only revealed statistically significant day-night differences on the night of the second day of the saline infusion (Table 2). The modest decline at night was most likely due to the fact that BP was measured during blood withdrawal, so the subject was awake during these readings, causing increased nocturnal measurements.
|
|
, right graph). No significant treatment differences were found in diastolic BP on d 1 (saline: d 1, 65.5 ± 1.3; night 1, 63.5 ± 2.1; glucose: d 1, 63.4 ± 1.4; night 1, 62.4 ± 1.9) or d 2 (saline: d 2 63.8 ± 1.3; night 2, 60.1 ± 1.3; glucose: d 2, 64.4 ± 1.5; night 2 62.4 ± 1.4).
On d 1, mean HRs were 68.1 ± 8.4 and 63.4 ± 7.6 beats per minute (bpm) during the saline infusion and 67.6 ± 10.7 and 62.4 ± 9.8 bpm during the glucose infusion. As shown in Table 3, significant day-night differences in HR were observed during both days of the saline infusion. Significant increases in HR were observed on the second day of the 48-h glucose infusion. The nocturnal decline in HR was blunted after the glucose infusion compared with the saline infusion and was significantly greater than on the saline day (P < 0.001). Day-night differences in HR were also significantly smaller on the second day of hyperglycemia (Fig. 5, left; saline, 5.8 ± 2.8 bpm; glucose, 1.7 ± 4.6 bpm; P < 0.003).
|
|
No significant differences were found in the 24-h urinary catecholamine measurements during saline and glucose infusions. Urinary norepinephrine levels were 16.9 ± 9.5 (d 1) and 20.3 ± 20.4 (d 2) during the saline infusion and 21.3 ± 9.7 (d 1) and 24.1 ± 9.5 (d 2) during the glucose infusion. Urinary epinephrine levels were 4.7 ± 4.7 (d 1) and 6.5 ± 5.2 (d 2) during the saline infusion and 4.5 ± 3.8 (d 1) and 6.1 ± 2.7 (d 2) during the glucose infusion. Urinary dopamine levels were 167.4 ± 89.7 (d 1) and 197.3 ± 159.2 (d 2) during the saline infusion and 161.2 ± 76.1 (d 1) and 226.9 ± 169.8 (d 2) during the glucose infusion.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Alterations in cardiovascular activity were most likely elicited by increases in circulating insulin and/or glucose or an induction of insulin resistance. The exact mechanisms mediating the observed effects are not known, although evidence exists for each of these factors acting independently on autonomic activity. Epidemiological studies suggest that HRV is inversely associated with plasma glucose (4, 29), although the results of acute manipulations of glucose levels are inconsistent. One study demonstrated significant alterations in indices of cardiovascular function in healthy control subjects independently of changes in plasma insulin (30), but another study found no significant effect of elevated glucose levels (31). Short-term infusions of insulin, such as occur during a hyperinsulinemic, euglycemic clamp, have also been shown to decrease cardiac PNS activity (18) or, alternatively, increase the ratio of low to high frequency oscillations, which has been interpreted as an increase in cardiac sympathovagal balance (32). A few studies using similar methodologies have reported no changes in HRV variables during hyperinsulinemic, euglycemic conditions (16, 17).
Infusions of glucose and/or insulin for longer than 5 h have, to date, not been used to evaluate the effect on cardiovascular autonomic activity. In the present study, significant changes in HRV were only observed on the second day of the glucose infusion, and the initial attenuation occurred during the night, when cardiac vagal activity is typically elevated. These data suggest that significant decreases in HRV did not take place until cardiac tissue had been exposed to elevated levels of glucose and insulin for an extended period of time. Pathak et al. (33) demonstrated that insulin decreases mRNA expression of M2-muscarinic receptors in rat atrial cardiomyocytes in a dose and time-dependent manner, with maximal effects after 18 h of insulin incubation. Thus, in humans, in whom glucose levels were only modestly elevated compared with the in vitro studies, a period of 36 h was necessary to elicit changes in autonomic activity.
Prolonged exposure of cardiac tissue to elevated levels of circulating insulin may explain the previously reported relationship between insulin sensitivity and cardiac autonomic activity. A number of studies have shown that insulin sensitivity, measured by a hyperinsulinemic, euglycemic clamp, is correlated with RMSSD in healthy subjects (34). A similar relationship was demonstrated during experimentally induced stress in a group of controls and first-degree relatives of type 2 diabetics (9), in whom decreased insulin sensitivity was correlated with decreased HRV. However, Bergholm et al. (18) demonstrated that only insulin-sensitive subjects exhibited a decrease in cardiac vagal activity in response to an elevation in plasma insulin, whereas no changes in HRV were observed in insulin-insensitive subjects. In the present study, the 48-h glucose infusion elicited a metabolic profile of glucose intolerance and may have induced an insulin-resistant state in lean, healthy control subjects. As indicated by homeostasis model assessment (HOMA), an indirect measurement of insulin sensitivity, the prolonged glucose infusion significantly decreased insulin sensitivity even when measured 3 h after terminating the glucose infusion (Table 4). All subjects participating in the experiment exhibited a decrease in HOMA in response to the glucose infusion. Thirteen of 15 individuals showed a decline in cardiac vagal activity, as indicated by a decrease in RMSSD. We found that the HOMAs calculated from insulin and glucose values at 0300 h were significantly correlated with the RMSSD night-day differences at the same time (Fig. 6; r = 0.48; P < 0.001). In contrast, no significant correlations were found between glucose or insulin values at the same time point and RMSSD values. These data suggest that the observed decrease in HRV may be due to the induction of insulin resistance by the prolonged glucose infusion.
|
|
and 3). We found that the glucose infusion elicited a significant nocturnal increase in BP, such that nighttime SBP did not decline and remained in the same range as normal daytime values. No significant changes were found in diastolic BP, suggesting that an increase in cardiac output occurred independently of a change in vascular resistance. A diurnal profile of BP in which the nocturnal decrease is either attenuated or absent is generally termed a non-dipper pattern (37, 38), and several studies have highlighted the clinical significance of these disturbances, which are associated with a progression in end-organ damage in hypertensive patients (21, 39). The elicitation of a non-dipper profile coupled with a significant increase in HR after a relatively modest experimental manipulation is surprising, because this type of patterning would not be expected in a healthy, lean, nonhypertensive individual. Despite increases in BP and HR, which are usually considered to be sympathetically mediated responses, we found no significant increases in the HRV measures of mixed parasympathetic and sympathetic activities. In fact, only time domain measurements considered to be primarily mediated by the vagus nerve (RMSSD and pNN50) were significantly lower after the glucose infusion compared with the saline infusion control. Furthermore, no significant increases in urinary catecholamines were observed during the glucose infusion. These data suggest that the SNS may not have been activated to the same extent as PNS was inhibited and raise the question as to what branch of the autonomic nervous system is mediating the observed acute increases in BP and HR. Withdrawal of cardiac vagal activity can occur independently of SNS activation (40); therefore, the PNS may be responsible for the initial changes in cardiovascular responses in response to an acute metabolic challenge.
The 48-h glucose infusion into lean, control subjects elicits a metabolic profile resembling that of an obese, glucose-intolerant individual, i.e. significant elevations in plasma insulin and leptin (data not shown) with mild elevations in plasma glucose levels, blunted nocturnal HRV, and mild elevations in HR and BP. However, the observed changes in cardiovascular autonomic activity were elicited, independently of an increase in weight or body adiposity. These findings, occurring after only 48 h of mild hyperglycemia, have profound implications with respect to the effects of peripheral metabolism on cardiovascular activity. Although it is well established that diabetes increases the risk of coronary heart disease and myocardial infarction, our new findings suggest that one of the very first changes in cardiovascular function may occur at the level of the nervous system. Although one would assume the rapid return of a normal cardiovascular profile in normal healthy individuals, in individuals with metabolic disease, this may not be the case. The majority of cardiac events occur in the hours between 0500–0700 h (27), a time at which SNS activity is rapidly increasing, and PNS activity is decreasing. Repeated metabolic challenges, such as during overeating, may prematurely decrease cardiac parasympathetic activity, rendering the obese individual more susceptible to cardiac events. Future studies should focus on the neural contribution to circadian cardiovascular function, because disruption of the circadian cardiovascular profile may be an early pathophysiological marker for impaired glucose tolerance.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online January 4, 2006
Abbreviations: BP, Blood pressure; bpm, beats per minute; ECG, electrocardiogram; HOMA, homeostasis model assessment; HR, heart rate; HRV, HR variability; pNN50, percentage of difference between adjacent normal RR intervals that are greater than 50 msec; PNS, parasympathetic nervous system; RMSSD, root mean square successive difference; SBP, systolic blood pressure; SDANN, SD of the mean of normal R-R intervals for each 5-min period of the ECG recording; SDNN, SD of all normal R-R intervals; SNS, sympathetic nervous system.
Received August 3, 2005.
Accepted December 21, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Pletinck, F. Verbeke, L. Van Bortel, C. Dequidt, D. Vijt, W. Van Biesen, and R. Vanholder Acute central haemodynamic effects induced by intraperitoneal glucose instillation Nephrol. Dial. Transplant., December 1, 2008; 23(12): 4029 - 4035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Teff Visceral Nerves: Vagal and Sympathetic Innervation JPEN J Parenter Enteral Nutr, September 1, 2008; 32(5): 569 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Jehn, D. J. Brotman, and L. J. Appel Racial Differences in Diurnal Blood Pressure and Heart Rate Patterns: Results From the Dietary Approaches to Stop Hypertension (DASH) Trial Arch Intern Med, May 12, 2008; 168(9): 996 - 1002. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
