Relationship between insulin sensitivity, hyperinsulinemia, and insulin-mediated sympathetic activation in normotensive and hypertensive subjects

AJH

1996;9:1172-1178

Relationship Between Insulin Sensitivity, Hyperinsulinemia, and Insulin-Mediated Sympathetic Activation in Normotensive and Hypertensive Subjects Carlos Arauz-Pacheco, Dan Lender, Peter G. Snell, Beverley Huet, Luis C. Ramirez, Laura Breen, Pablo Mora, and Philip Raskin

The adrenergic response to high physiological hyperinsulinemia was studied in 39 hypertensive subjects (28 men and 11 women) and 25 normal volunteers (15 men and 10 women), using the euglycemic clamp technique. Control studies using 0.45% saline infusions (sham studies) were also performed. Before and during the clamp procedure, plasma norepinephrine (NE) and epinephrine (E) were measured using a high performance liquid chromatographic method (HPLC). The association between the increment in NE and E levels and insulin sensitivity, steadystate insulin level during the clamps, waist to hip ratio (WHR), baseline NE levels and gender was studied. NE levels increased during the hyperinsulinemic period (mean increase 46 _+ 6 pmol P < .001 v baseline and P < .01 v sham studies). E levels did not differ between the insulin clamps and the sham studies. Insulin sensitivity was not significantly associated with the increment in NE. Hypertensive subjects had a

higher NE increase than the normotensive individuals (55 _+ 7 v 3 0 _+ 10 pmol, P = . 0 3 ) , but also had higher insulin levels during the clamps (839 + 43 v 522 _+ 38 pmol, P < . 0 0 1 ) . Insulin levels accounted for most of the differences in NE increase between the normotensive and hypertensive groups. Gender, adiposity and WHR were also associated with NE increment. We conclude that the insulin mediated sympathetic activation is not affected in the presence of decreased insulin sensitivity for glucose utilization. The greater degree of sympathetic activation observed in hypertensive subjects is a function of the level of insulinemia obtained during the clamps. © 1996 American Journal of Hypertension, Ltd. Am J Hypertens 1996;9:11721178

epidemiological association between hypertension and increased plasma insulin levels or decreased insulin sensitivity is well known. ~ 3 This association is attri-

buted, in part, to the close relationship between obesity (a hyperinsulinemic and insulin resistant condition) and hypertension. 2 Evidence that hypertension is associated with elevated plasma insulin levels and

Received December 4, 1995. Accepted May 28, 1996. From the Department of Internal Medicine, Clinical Diabetes Division (CA-P, DL, LCR, LB, PM, PR), the Cardiology Division (PS), and the General Clinical Research Center {BH), the University of Texas Southwestern Medical Center at Dallas, Dallas, Texas. Supported m part by grant #M01RR00633 (General Chnlcal Research Center).

Prehmlnary results were presented at the 65th Scientific Sessions of the American Heart Association, New Orleans, Louisiana, 1992. Address correspondence and reprint requests to Philip Raskm, MD, Professor of Internal Medicine, The Universaty of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 752358858

A

n

© 1996 by the Amertcan Journal of Hype~tenslon. Ltd Pubhshed by Elsevler Science, lHc

Insulin resistance, hypertension, norepinephrine, blood, euglycemic clamps, sympathetic nervous system. KEY WORDS:

0895-7061/96/$15 O0 PI1 S0895-7061( 96 )00256-7

AJH-DECEMBER 1996-VOL. 9, NO. 12, PART 1

INSULIN SENSITIVITY AND SYMPATHETIC ACTIVATION 1173

insulin resistance i n d e p e n d e n t l y of obesity has been published. 3-9 It is still uncertain what the mechanism(s) is for the association between hyperinsulinemia and elevated blood pressure levels. Insulin infusions enhance the renal reabsorption of s o d i u m at a renal tubular level. 1° By decreasing urinary s o d i u m excretion, hyperinsulinemia could result in an elevation of blood pressure. Another p r o p o s e d mechanism by which insulin m a y affect the blood pressure levels is through the stimulation of the sympathetic nervous system. Studies by Rowe et al 1] s h o w e d a dose d e p e n d e n t increase in plasma norepinephrine levels during euglycemic hyperinsulinemia in normal humans. Some of these subjects manifested h e m o d y n a m i c changes characteristic of a h y p e r a d r e n e g i c state. Using m i c r o n e u r o g r a p h y , A n d e r s o n et a112 demonstrated, that during hyperinsulinemia, the rate of firing of the muscular sympathetic nervous system increases significantly. Recent studies have analyzed the relationship between sympathetic hypertension and insulin-mediated sympathetic activation. Morgan et al has shown that spontaneously hypertensive rats have increased adrenal sympathetic nerve responses to insulin c o m p a r e d to normotensive Wistar-Kyoto rats, but had similar lumbar sympathetic nerve responses. ~3 Lembo et al reported that forearm norepinephrine release in increased in hypertensive subjects during systemic hyperinsulinemia. ~4 To further s u p p o r t the hypothesis that sympathetic activation plays a role in the d e v e l o p m e n t of hypertension in insulin resistant subjects, insulin mediated sympathetic activation should be preserved in individuals w h o are resistant to insulin's effects on glucose metabolism w h e n c o m p a r e d to insulin sensitive subjects. The p u r p o s e of this study was to evaluate the hypothesis that hypertensive subjects (expected to be insulin resistant) have similar sympathetic responses

to insulin, w h e n c o m p a r e d to normotensive (insulin sensitive) subjects. SUBJECTS A N D M E T H O D S

A total of 64 volunteers were studied. Of these, 39 were hypertensive and 25 were normotensive. The general characteristics of the study subjects are presented in Table 1. The purpose and methods of the study were explained to the subjects by the investigators, and all the participants gave their informed, written consent. The study was approved by the Institutional Review Board for H u m a n Studies of the University of Texas Southwestern Medical Center at Dallas. The inclusion criteria for the hypertensive subjects included: 1) presence of chronic hypertension with an average of sitting diastolic blood pressure between 95 and 114 m m Hg on two readings, 3 weeks after discontinuation of all antihypertensive medications; 2) age 21 or older; 3) weight less than 40% over the ideal b o d y weight; 4) absence of angina pectoris, previous myocardial infarction, cardiac failure, arrhythmia, or renal dysfunction; 5) the use of other concomitant medication that could affect insulin sensitivity; 6) fasting plasma glucose < 110 m g / d L . The normotensive subjects were all 21 years of age or older, less than 40% over the ideal b o d y weight and free of any significant disease and took no concomitant medications. All subjects had an initial laboratory screening panel that included a complete blood count and a chemistry panel including fasting glucose and lipids, electrolytes, liver function tests, serum creatinine, and uric acid. Patient Selection

EXPERIMENTAL PROTOCOL

The s t u d y participants were admitted to the General Clinical Research Center at Parkland Memorial Hospital the night prior to the studies, where they were assured of a nonstressful environment. On the morning of the studies a euglycemic, hyperinsulinemic clamp study was p e r f o r m e d as follows. A small plas-

TABLE 1. CHARACTERISTICS OF THE STUDY SUBJECTS Normotensive Men

No. Age (yr) BMI Body fat (%) WHR HTN (years) FPG BP

50 29.9 28.9 0.97

15 ___3 ± 0.9 ___2.0 _+ 0.06

104 ± 2 118/74

Hypertensive Women

55 29.0 37.7 0.79

10 ± 3 + 1.2 + 1.8 + 0.07

99 ± 3 115/73

Men

Women

28 55 ± 2 29.4 ± 0.6 28.1 ± 1.3 0.97 +_ 0.04 8.6 _+ 2.2 103 - 2 151/98

11 53 -- 3 27.3 + 1.1 35.4 ± 2.4 0.83 ± 0.06 9.5 + 2.4 96 + 2 140/100

Abbreviations BP, mean blood pressure m each group (mm/Hg); BMI, body mass index (weight (kg)/helght (m) 2); FPG, j-astmg plasma glucose (mg/ dL ) ; HTN, hypertenszon duratwn; WHR, wrest to-lup ratio.

1174

ARAUZ-PACHECO

AIH-DECEMBER 1996-VOL. 9, NO. 12, PART 1

ET AL

60 800 40

600 Go

ILl

400

z _m 20

200 o__.

o

~

o -200

Normotensive

Hypertensive

Sham

[mChange m NE E":ISSI~

FIGURE 1. Increment in norepinephrme (NE) levels and steady-state insulin ( SSI) levels in normotensive and hypertensive subjects during euglycemic, insulin clamps and during 0.45% saline infuswns (sham clamps). During the clamps, elevations in NE occurred in both normo and hypertensive groups ( f P < .01 v sham clamps). Hypertensive subjects had higher NE levels during hyperinsulinemia compared to normotensive subjects (* P = .03). Hypertensive subjects developed higher insulin levels during the insulin clamp studies (**P ~ .001) (probably reflecting decreased insulin clearance1°~: ). During the sham studies the insulin levels remained low (at baseline level).

tic catheter was placed in a forearm vein in a retrograde fashion to obtain blood samples. The hand was placed in a warm box at 65°C. A long, thin plastic catheter was placed in an antegrade fashion in a antecubital vein for the infusion of insulin and 20% dextrose (17.7% glucose). The plasma insulin level was raised acutely and a continuous infusion of insulin was maintained at 40 mU / square meter of body surface a r e a / m i n u t e over 2 h. Dextrose was infused at a variable rate according to the results of plasma blood glucose levels measured at 5-min intervals on samples using an enzymatic glucose analyzer. The rates of glucose infusion were determined using the protocol of DeFronzo et al. 15 With this methodology, the plasma glucose levels were maintained at the basal level with less than 5% variation from this level. Harvard Apparatus infusion pumps Model 2205 (Harvard Apparatus, South Natick, MA) were used during the procedure. Insulin levels were measured in blood samples obtained every 10 min during the procedure using a sensitive radioimmunoassay with a 2% to 5% intraassay coefficient of variation (CV) and a 5% to 10% interassay CV. The mean insulin level between min 60 and 120 was defined as steady-state insulin level. Insulin sensitivity (IMGU) was calculated by dividing the glucose infusion rates during min 90 to 120 of the clamp divided by the insulin level during the clamp in order to correct for the intersubject variability in the insulin levels during the clamps. IMGU was expressed in milligrams of glucose / kilograms of fat free mass (FFM)/minute. During insulin infusions, at the levels used in this study, the hepatic glucose output

has previously been shown to be completely supressed in nondiabetic subjects. 4 Body density was obtained in all subjects using underwater weight. Percent of body fat was derived from body density and fat free mass was calculated by subtracting fat mass from total body weight. WaisHo-hip ratio was calculating by dividing the minimum circumference at the level of the umbilicus and the maximum circumference at the level of the gluteus. Plasma NE and E levels were measured using a high-performance liquid chromatographic method (Nichols Laboratories, San Juan Capistrano, CA). Duplicate samples containing 10 cc each were obtained at 0, 60, and 120 min. The samples were collected in ice-cold tubes, centrifuged immediately and frozen at -20°C prior to assay (the average storage time was 2 weeks, and no sample was stored for more than 4 weeks). The sensitivity of the NE assay is 20 pmol with interassay CV of 1.9% to 2.9% and intraassay CV of 2.4%. The sensitivity of the E assay was 20 pmol with interassay CVs of 6.4% to 7.9% and intraassay CV of 7.8%. Since the NE levels at 60 and 120 rain were virtually identical, the averages of the NE and E levels at 60 and 120 min were calculated and the incremental changes in NE and E were obtained by subtracting the values at 0 min from the average of 60 and 120 min in each subject's samples. NE and E levels were also drawn after half-normal (0.45 NaC1) infusions reproducing the clamp environment in a sample of 10 subjects (five normotensive and five hypertensive individuals) in order to control for the nonspecific stress effects on the NE levels (sham studies). A total of 46 study subjects underwent two clamps at identical insulin infusion rates (40 m U / m 2 min -t) as part of a different protocol. The intrasubject coefficient of variation in IMGU and steady state insulin levels during the clamps, were calculated from these studies and were found to be 12% and 13% respectively. Systemic arterial blood pressure and pulse rate was measured noninvasively every 10 min during the procedure using an external monitor. STATISTICAL ANALYSIS The statistical analysis was performed with BMDP software (BMDP Statistical Software, Los Angeles, CA). Data are expressed as mean ± SEM. For baseline variables, change in NE and E levels, M-values, and SSI, comparisons between normotensive and hypertensive subjects and between the insulin clamps and the sham studies were performed using one-sample t tests. A P = .05 was considered statistically significant. The relationship between change in NE level and IMGU, percent of body fat, WHR, gender, hypertensive and normotensive status and SSI, was assessed with multiple regression models. As hypertension status and SSI were highly interrelated, an additional comparison of the in-

INSULIN SENSITIVITY A N D SYMPATHETIC ACTIVATION

A J H - D E C E M B E R 1996 VOL. 9, NO. 12, P A R T 1

TABLE 2. M E T A B O L I C A N D

ADRENERGIC

PARAMETERS

DURING

INSULIN

Normotensive

79 573 104 102 6.2 288 38 41 16

CLAMPS

Hypertensive

Men

Insulin 0' Insulin 90-120' Glucose 0' Glucose 90-120' IMGU Norepi O' A Norepi Epineph O' A Epineph

1175

+ 14 + 48 + 2 + 3 + 0.5 _+ 21 + 12 ± b ± 7

Women

Men

46 + 11 405 ± 52 99 ± 3 97 +_ 3 7.9 ± 0.5 261 ± 34 17 + 13 32 ± 6 10 ± 6

90 ± 7 832 ± 49* 103 ± 2 105 +_ 3 4.6 _+ 0.3t 219 _+ 16"* 55 + 8 35 ± 3 14 ± 4

Women

70 -860 ± 95 ± 96 ± 6.6 + 323 ± 57 ± 44 ± 22 ±

14 83* 2 4 0.5t 35 11§ 9 11

Insulin, norepmephrme (NorepO and epmephrtne (Epmeph) m pmol, glucose m mg/dL, * P < 001 versus normotenszve; ~ P < 05 versus no~motensive men and women and hypertensive women. § P ~- 05 versus normotens&e women

cremental change in NE between subjects in the low, m e d i u m , and high tertiles of SSI in the s t u d y g r o u p as a whole was performed. Also, the changes in NE in subjects in the low (below the median) and high (above the m e d i a n ) SSI levels in the hypertensive and normotensive subgroups were compared. A N O V A was used in the s u b g r o u p comparisons. RESULTS

H y p e r t e n s i v e and n o r m o t e n s i v e subjects w e r e c o m p a rable in terms of age, b o d y fat content, waist-to-hip ratio, fasting insulin, and fasting glucose levels (Table 1). D u r i n g the hyperinsulinemic, euglycemic c l a m p studies the m e a n glucose levels was m a i n t a i n e d at 98 + 0.4 m g / d L ( m e a n + s t a n d a r d error of the mean, SEM). The percent coefficient of variation in glucose levels d u r i n g the clamps: 3.1%. The m e a n insulin level obtained d u r i n g the c l a m p studies (SSI) w a s 720 + 35 pmol. H y p e r t e n s i v e subjects had higher SSI d u r i n g the c l a m p s than did n o r m o t e n s i v e individuals (839 _+ 43 v 522 _+ 39 pmol, P < .001, h y p e r t e n s i v e versus n o r m o t e n s i v e ) (Figure 1 ). H y p e r t e n s i v e subjects had significantly lower I M G U than n o r m o t e n s i v e individuals (5.1 ± 0.3 v 6.9 ± 0.4 m g / k g / m i n / p m o l ; P < .001, h y p e r t e n s i v e versus n o r m o t e n s i v e ) .

TABLE 3. M U L T I P L E R E G R E S S I O N A N A L Y S I S D E P E N D E N T V A R I A B L E : C H A N G E IN N E LEVELS Parameter

Estimate

Standard Error

t

P

Constant Gender SSI Percent of body fat Waist-to-hip ratio

177 71.4 0.050 -2.59 344

81 22.7 0.018 0.76 109

2.18 3.15 2.79 -3.40 3.17

.03 .003 .007 .001 .002

Regression P v a l u e .00005, R ~ = 0.34, Adjusted R e = (; 30. NE, norepmcphrmc: SSI, steady statc ltlstdl~l Icvcls

NE levels s h o w e d a significant increase d u r i n g the h y p e r i n s u l i n e m i c c l a m p s in both h y p e r t e n s i v e and n o r m o t e n s i v e subjects ( m e a n increment f r o m all the studies: 46 ± 6 p m o l , P < .001 v baseline, a n d P < .01 v s h a m studies), the increase was greater in the h y p e r t e n s i v e g r o u p c o m p a r e d to the n o r m o t e n s i v e (55 + 7 p m o l v 30 ± 10 pmol; P = .03, h y p e r t e n s i v e versus n o r m o t e n s i v e ) . No significant change in NE levels was o b s e r v e d in the s h a m studies ( m e a n change - 5 _+ 7 pmol, P < .2) (Figure 1). In multiple regression analysis, I M G U and baseline NE level did not s h o w a significant association with the change in NE. H y p e r t e n s i o n status, gender, SSI, percent of b o d y fat, and W H R s h o w e d a significant association with the d e p e n d e n t variable (change in NE) in univariate analysis. H o w e v e r , in multiple regression analysis, after the steady-state insulin level was i n t r o d u c e d in the model, h y p e r t e n s i o n status no longer was associated w i t h NE level (Table 2). The best m o d e l for the multiple regression analysis included the variables gender, SSI, p e r c e n t a g e of b o d y fat and W H R (Table 3). Together, these variables explained 30% of the variance in NE r e s p o n s e to insulin. A d o s e - r e s p o n s e effect w a s o b s e r v e d w h e n the incremental change in NE w a s a n a l y z e d according to tertile of SSI (Figure 2), with the subjects in the highest tertile of SSI h a v i n g levels 2.5 times higher than the subjects in the lowest tertile (P = .014). H y p e r t e n s i v e subjects with high SSI h a d a higher i n c r e m e n t in NE than n o r m o t e n s i v e patients with low SSI (Figure 3 P < .01 ). N o r m o t e n s i v e subjects with high SSI had similar NE increments w h e n c o m p a r e d with h y p e r t e n s i v e subjects with low SSI (41 + 10 v 47 _+ 9 pmol, n o r m o tensive with high SSI versus h y p e r t e n s i v e with low SSI); they also h a d similar SSI levels (651 ± 15 v 658 ± 38 p m o l ) (Figure 3). E p i n e p h r i n e levels increased d u r i n g the insulin infusion ( m e a n i n c r e m e n t 16 ± 7 p m o l ) but also increased d u r i n g the s h a m studies ( m e a n increment 10

1176 ARAUZ-PACHECOET AL

8o ~

60

/

& I.U

z 40 .E_ (1)

~

20

Low(408+/-26)

Intermediate(678+/-12) High(1029+/-43) SSI tertiles (pMol)

FIGURE 2. Incremental change in NE levels in the study group as a whole according to tertile of steady-state insuhn level (SSI). The change in NE levels mcreased with the SSI tertile. The differences between the group in the high tertile and the one in the low tertile were statistically significant (*P = .014).

_+ 5 pmol). No significant differences in E levels, was found between the insulin clamps and the sham studies (P > .2). In the multiple regression analysis, change in E levels did not correlate significantly with any of the other variables studied. No changes in blood pressure and pulse rate were observed during the clamp studies in normotensive or hypertensive individuals. DISCUSSION This study shows that insulin mediated NE stimulation is independent of the degree of metabolic insulin sensitivity as determined by the IMGU. That is, individuals with decreased insulin sensitivity (that is, insulin resistant), have increases in NE levels similar to insulin sensitive subjects when they are infused similar amounts of insulin (at least acutely and at the high physiological levels used in this study). The association between the absolute insulin levels obtained during the infusions and the NE increment in this study, suggests that a dose dependent pattern of response is present in the insulin mediated adrenergic stimulation phenomenon, within the high physiological insulin levels induced by our protocol. Hypertensive subjects had a greater degree of NE stimulation in this study, however, they also had higher insulin levels during the clamps. When insulin levels during the clamp were taken into account, the effects of hypertension status on NE increment disappear. The fact that normotensive and hypertensive subjects with similar SSI levels had almost identical NE responses, reinforces the hypothesis that the higher insulin level observed in hypertensive subjects, and not the hypertensive status per se, is associated with the greater NE increments seen in the hypertensive group in this study. The higher insulin levels, in the hypertensive group, compared to an age and obesity matched nor-

AJH-DECEMBER 1996 VOL. 9, NO. 12, PART 1

motensive group, receiving identical insulin infusions, most likely reflects decreased insulin clearance. Several groups have recently reported decreased metabolic insulin clearance in hypertensive patients. 16'17 In this study, we assessed the insulin-mediated sympathetic activation using venous plasma catecholamine levels. This method has some limitations as reviewed recently by Hjemdahl. is Plasma levels of catecholamines reflect the net effect of norepinephrine and epinephrine overflow from sympathetic synapsis and the adrenal medulla and the rate of clearance of these substances. Also, the regional effect of a given physiological stimulus is highly heterogeneous as Morgan et a113 and others have shown. 19Nevertheless, NE levels in venous samples obtained from the forearm correlated well with muscular sympathetic nerve recordings obtained with microneurography and reported by Wallin et al. 2° Thus, the measurements obtained in this study can be considered, with limitations, as a reflection of the local (forearm) muscular sympathetic activity. The mechanism of stimulation of the adrenergic nervous system by insulin is not k n o w n . 21"22 A central stimulatory effect has been postulated. The ventromedial hypothalamus has been involved in sensing insulin. 22 2-Deoxy-D-glucose, a nonmetabolizable glucose analog, results in abolition in the feeding-related increases in sympathetic activity. Glucose metabolism at this level seems to be necessary for the signal to be expressedY In obese and hypertensive subjects, metabolic insulin resistance (low IMGU) is the consequence of a reduced nonoxidative glucose metabolism at the muscular level, glycogen storage is the main pathway affected; glucose oxidation in the muscle is

80

~;

6o

r~ v LU

z 40 c_ ~ 0

2o

NL(346+/-34)

i

NH(658+/-30) HL(651+/-15) Subgroup (SSI pMol)

HH(1037+/-53)

FIGURE 3. Incremental change in NE levels in the hypertensive and normotensive group according to steady-state insulin (SSI) level ( below and above median ). The hypertensive subgroup with high SSI (HH) had significantly higher NE increment compared with the normotenstve subgroup with low SSI (NL) (*P < .01 ). The hypertensive group with low SSI (HL) and the normotenstve group with high SSI ( NH) had similar NE increments and szmllar SSI levels (see text).

AJH-DECEMBER 1996-VOL. 9, NO 12, PART 1

INSULIN SENSITIVITY AND SYMPATHETIC ACTIVATION

2. usually n o r m a l in this setting. The specificity of the defect in metabolic insulin resistance m a y explain w h y other insulin actions (that is, s y m p a t h e t i c activa3. tion) are not affected in insulin resistant subjects like the ones in this study. The results of our s t u d y m a y be relevant in ex4. plaining the relationship b e t w e e n insulin resistance and consequent h y p e r i n s u l i n e m i a , and arterial h y p e r tension. Since insulin resistant subjects d e v e l o p 5. higher insulin levels after a meal than do insulin sensitive subjects with similar glucose tolerance, it seems likely that they w o u l d h a v e a greater adrenergic re6. sponse. Plasma insulin levels in individuals with norm a l glucose tolerance can v a r y four- to fivefold dep e n d i n g on the degree of insulin sensitivity. 24 Thus, lean subjects can h a v e a p e r i p h e r a l insulin level of 7. a p p r o x i m a t e l y 300 p m o l and obese ones can h a v e values in the range of 1500 pmol. If the relationship between insulin levels and NE that we describe here is 8. present in the p o s t p r a n d i a l state, one should expect a higher s y m p a t h e t i c stimulation in the subjects with a greater insulin response. 9. Body fat showed a negative correlation with NE increment during the clamps. This association has been identified by other investigators and it has been sug- 10. gested that it plays a role in the association between obesity and hypertension. 24 Another potential explanation of these findings is the possibility that the negative 11. association between b o d y fat and NE increment is due to regression to the m e a n of the baseline NE levels. We did not observe any significant change in blood 12. pressure levels or heart rate during the clamp studies in normotensive and hypertensive subjects. This can be attributed to the vasodilatory effects of insulin ~-5-27 that c o m p e n s a t e for the increase in the n o r e p i n e p h 13. rine levels or that the m o d e s t increase in NE levels w e r e not e n o u g h to elevate the levels of blood pressure. Also, since w e u s e d an external blood pressure 14. monitor, it is likely that the sensitivity of the m e a s u r e m e n t s p e r f o r m e d w a s not e n o u g h to detect significant but m i n o r changes. Differences m i g h t h a v e been seen had we m e a s u r e d intraarterial pressure directly. The 15 possibility exists that the changes p r o d u c e d b y h y p e r insulinemia in h u m a n s are seen only with chronic stimulation. More research is necessary to clarify the 16. relationship b e t w e e n insulin and s y m p a t h e t i c activation p e r f o r m i n g studies at different insulin and glucose levels and also d u r i n g chronic studies. The latter, 17. h o w e v e r , m a y be difficult in h u m a n s . 18. ACKNOWLEDGMENTS We wish to acknowledge Philip Wyrick, Donald Tunstall, and Kevin Sullivan for their technical assistance, and Jcyekki Keys for excellent secretarial assistance. 19. REFERENCES 1. Welborn TA, Breckenridge A, Dolelry CT, et al: Seruminsulin in essential hypertension and in peripheral vas- 20. cular disease. Lancet 19661336-1337.

1177

Van Italli TB: Health implications of overweight and obesity in the United States. Ann intern Med 1985; 103:983-988. Modan M, Halkin H, Almog S, et al: Hyperinsulinemia: a link between hypertension, obesity and glucose intolerance. J Clin Invest 1985;75:809-817. Ferranninni E, Buzzigoli G, Bonnadonna R, et al: Insulin resistance in essential hypertension. N Engl J Med 1987;317:350-357. Shen-D-C, Shieh S-M, Fuh MM-T, et al: Resistance to insulin-stimulated-glucose uptake in patients with hypertension. J Clin Endoc Metab 1988;66:580-583. Marigliano A, Tedde R, Sechi LA, et al: Insulinemia and blood pressure. Relationships in patients with primary and secondary hypertension and with and without glucose metabolism impairment. 1990;3:521-526. Swilocki AL, Hoffman BB, Reaven GM: Insulin resistance, glucose intolerance and hyperinsulinemia in patients with hypertension. 1989;2:419-423. Falkner B, Hulman S, Tennenbaum J, Kushner H: Insulin resistance and hypertension in young black men. Hypertension 1990; 16:708- 711. Pollare T, Lithel H, Berne C: Insulin resistance is a characteristic feature of hypertension independent of obesity. Metabolism 1990;39:167-174. DeFronzo RA, Cooke R, Andres R, et al: The effect of insulin on renal handling of sodium, potassium, calcium and phosphate in man. J Clin Invest 1975;55:845855. RouTeJW, Young JB, Minaker KL, et al: Effect of insulin and glucose infusions on sympathetic nervous activity in normal man. Diabetes 1981;30:219-225. Anderson EA, Hoffman RP, Balon TW, et al: Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal humans. J Clin Invest 1991; 87:2246-2252. Morgan DA, Balon TW, Ginsberg BH, Mark AL: Non uniform regional sympathetic responses to hyperinsulinemia in rats. Am J Physiol 1993;264:R423-R427. Lembo G, Napoli R, Cepaldo B, et al. Abnormal sympathetic overactivity evoked by insulin in the skeletal muscle of patients with essential hypertension. J Clin Invest 1992;90:24-29. DeFronzo RA, Tobin J, Andres R: Glucose clamp technique: a method for quantifying insulin resistance and secretion. Am J Physiol 1979;237:E214-E223. Salvatore T, Cozzolino D, Glunta R, et al: Decreased insulin clearance as a feature of essential hypertension. J Clin Endocrinol Metab 1992;74:144-148. Swislocki AL: Impaired insulin clearance in essential hypertension. J Hum Hypertens 1994;8:185-190. Hjendahl P: Plasma catecholamines. Analytical challenges and physiological limitations. Balliere's Clin Endocrinol Metab 1993;7:307-353. Esler M, Jennings G, Leonard P, et al: Contribution of individual organs to total noradrenaline release in humans. Acta Physiol Scand Suppl 1984; 527:11-16. Wallin BC, Morlin C, Hjendahl P: Muscle sympathetic activity and venous plasma noradrenaline concentra-

1178

ARAUZ-PACHECO ET AL

tions during static exercise in normotensive and hypertensive subjects. Acta Physiol Scand 1987; 129:489-497. 21.

22.

23.

Landsberg L, Young JB: Insulin-mediated glucose metabolism in the relationship between dietary intake and sympathetic nervous system activity. Int J Obesity 1985;9 (suppl 2):63-68. Young JB, Landsberg L: Impaired suppression of sympathetic activity during fasting in the gold thioglucose treated mouse. J Clin Invest 1990;65:1086-1094. Rappaport EB, Young JB, Landsberg L: Dissociation of sympathetic nervous system (SNS) and adrenal medullary responses to 2-deoxy-D-glucose (2-DG). Clin Res 1980;28:403A.

AJH DECEMBER 1996 VOL. 9, NO. 12, PART 1

24.

Peiris AN, Mueller RA, Struve MF, et al: Relationship of androgenic activity to splanchnic insulin metabolism and peripheral glucose utilization in premenopausal women. J Clin Endocrinol Metab 1987;64:162-169.

25.

Brands MW, Mizelle HL, Gaillard CA, et al: The hemodynamic response to chronic hyper-insulinemia in conscious dogs. Am J Hypertens 1991;4:164.

26.

Edelman SV, Laakso M, Wallace P, et al: Kinetics insulin-mediated and non-insulin mediated glucose uptake in humans. Diabetes 1990;39:955-964.

27. Laakso M, Edelman SV, Olefsky JM, et al: Kinetics of in vivo muscle insulin-mediated glucose in human obesity. Diabetes 1990; 39:965 -974.