Contrasting Autonomic and Hemodynamic Effects of Insulin in Healthy Elderly Versus Young Subjects
Acute increases in plasma insulin produce both sympathoexcitation and vasodilation in normal young adults. Aging is associated with insulin resistance and may alter the sympathetic or the vascular responses to insulin. Therefore, we assessed sympathetic and vascular responses to acute physiological increases in plasma insulin levels in 10 healthy, normotensive elderly (65±2 years) and 12 normal young (27±1 years) subjects matched for body mass index (25±1 kg/m2 in both groups). We measured muscle sympathetic nerve activity (microneurography), FBF (plethysmography), heart rate, and blood pressure and calculated forearm vascular resistance and insulin sensitivity (M value) during a 90-minute hyperinsulinemic/euglycemic clamp. M values were 4.3±0.4 mg·kg−1·min−1 in the elderly and 8.4±1.4 mg·kg−1·min−1 in the young subjects (P<.05). Baseline muscle sympathetic nerve activity was higher in the elderly subjects (33±3 versus 15±2 bursts per minute, P<.05); however, the absolute and percent increases in muscle sympathetic nerve activity were smaller in the elderly than in the young subjects (+10±1 versus +15±1 bursts per minute, or +37±11% versus +110±16%, P<.05). Forearm vascular resistance decreased with insulin from 46±2 to 31±3 units in the young but increased with insulin in the elderly subjects from 37±3 to 47±7 units (P<.05). Heart rate increased in young but not in elderly subjects. Insulin did not change blood pressure in either group. In conclusion, as opposed to vasodilation in young adults, insulin caused vasoconstriction in healthy elderly individuals. The failure of the vasodilator action of insulin in the elderly may permit even modest insulin-induced sympathoexcitation to elicit vasoconstriction. We speculate that the vasoconstrictor response to insulin may further potentiate insulin resistance in the elderly.
There is substantial evidence from human studies that acute elevations in plasma insulin levels cause sympathetic activation1 2 3 4 that could elevate arterial pressure.5 6 7 In healthy young subjects, however, the sympathoexcitation with physiological increases in plasma insulin is offset by vasodilation, and blood pressure does not increase.1 2 3 4
Most studies of the sympathetic and vascular actions of insulin have been performed in normal young adults who were relatively insulin sensitive. It is known that the vasodilator response to insulin is markedly attenuated in insulin-resistant states, such as obesity and type II diabetes.8 9 Aging also is associated with insulin resistance10 11 12 13 and may alter the vascular and/or sympathetic responses to insulin. Furthermore, aging is associated with elevated sympathetic activity,14 15 16 and some studies suggest that elderly individuals may have a greater sympathetic response to various stimuli, such as an oral glucose load.17 18 19 20 Finally, aging is associated with a higher incidence of hypertension.21
Thus, we hypothesized that in elderly individuals, acute hyperinsulinemia may cause enhanced sympathoexcitation and/or impaired vasodilation, which might result in increased blood pressure. We therefore measured the sympathetic, vascular, and blood pressure responses to insulin in healthy, normotensive elderly (≥55 years old) and young (20 to 30 years old) humans.
Subjects were 10 (5 male) elderly healthy normotensive and 12 (7 male) young healthy normotensive adults. Demographic and screening data are shown in Table 1⇓. All subjects underwent a screening history and physical examination, an oral glucose tolerance test, tests to rule out thyroid and kidney dysfunction, and an exercise treadmill test. Body fat content was estimated from measurements of skinfold thickness at seven sites.22 Ideal body weight was estimated from 1983 Metropolitan Life Insurance height and weight tables. No subject was receiving any medication, and all had blood pressures <140/85 mm Hg measured in the sitting position on three different occasions. The exercise treadmill test was negative in all subjects, and none had diabetes. The elderly subjects tended to have a decreased glucose tolerance compared with the young (Table 2⇓). One elderly subject and one young subject had an impaired glucose tolerance according to the National Diabetes Data Group criteria.23 The studies were approved by an Institutional Review Board on Human Investigation, and written informed consent was obtained.
Heart rate was measured continuously by an ECG. Blood pressure was measured by an automatic sphygmomanometer (Life Stat 200, Physio Control Corp) during the last half of each minute. FBF was measured by venous occlusion plethysmography with an air plethysmograph.24 Central venous pressure was measured through an 18.5-gauge polyethylene catheter inserted percutaneously into an antecubital vein and advanced into an intrathoracic vein. Respiration was monitored by a strain-gauge pneumotach.
Intraneural recording techniques were used to obtain multifiber recordings of postganglionic MSNA. A tungsten microelectrode (200-μm-diameter shaft, 1- to 5-μm uninsulated tip) was inserted into a muscle fascicle of the peroneal nerve posterior to the fibular head. A reference electrode was inserted subcutaneously 1 to 3 cm from the recording electrode. Electrodes were connected to a preamplifier (gain, 1000) and amplifier (variable gain, 1 to 99). Neural activity was fed to a bandpass filter (width, 0.7 to 2.0 kHz) and a resistance-capacitance integrating network (time constant, 0.1 second) to obtain a mean voltage neurogram.
Three criteria indicated acceptable MSNA recordings. First, electrical stimulation through the electrode in the nerve elicited only muscle contractions. Second, tapping innervated muscle elicited afferent mechanoreceptor discharge. Third, the neurogram revealed sympathetic bursts that increased during apnea. Evidence that this activity represents efferent MSNA has been reviewed before.25 During the insulin session, in one elderly and one young subject the MSNA recording site was lost after baseline measurements. During the vehicle control session, nerve recordings were obtained from five elderly subjects.
An insulin session was performed in all subjects. Insulin (Humulin, Eli Lilly & Co) diluted in saline with 1 mL of the subject's blood was infused by digital infusion pump (Bard Medsystems). Insulin infusion rates were targeted to increase plasma levels by 420 to 480 pmol/L. The clamp was performed for 90 minutes. The insulin infusion rate was primed for the first 10 minutes with decreasing doses and then 240 pmol·m−2·min−1.1 2 Blood glucose levels were maintained at euglycemic control values by variable infusion of 20% dextrose with an infusion pump (Flo-Gard 6200, Travenol Laboratories).
Six elderly subjects (four male) underwent a vehicle control session in which 0.2% saline was infused in a volume comparable to the dextrose volume of the insulin session. Vehicle control sessions were not performed in young subjects because we recently showed that a 90-minute vehicle infusion does not appreciably alter any of the measured parameters in this population.26
For 3 days preceding the study, dietary carbohydrate intake was adjusted to ≈250 g/d. All studies were performed in the morning after a 12-hour fast. Subjects were placed in the supine position, and intravenous catheters were placed in the right and left antecubital fossae for insulin/glucose or saline infusion and blood sampling, respectively. The central venous catheter was inserted at the left antecubital fossa. FBF was measured in the right arm. A satisfactory sympathetic nerve recording site was then achieved. After instrumentation, two 5-minute periods of baseline measurements (separated by 5 minutes) were obtained. The hyperinsulinemic/euglycemic clamp or the saline vehicle clamp was then performed as described above. Sympathetic nerve activity and hemodynamic measurements were recorded for 5 of every 15 minutes throughout the infusion period. Plasma insulin and plasma glucose were measured every 5 minutes throughout the study. Blood samples for serum potassium and plasma norepinephrine and epinephrine were taken during baseline and at the end of the insulin or vehicle infusion.
Blood glucose was analyzed with a YSI glucose analyzer (model 2300 Stat, Yellow Springs Instruments). Plasma insulin levels were measured in duplicate by the radioimmunoassay described by Yalow and Berson.27 Interassay and intra-assay coefficients of variation in our laboratory are 7.0% and 5.1%, respectively.2 Plasma norepinephrine and epinephrine were assayed by high-performance liquid chromatography with electrochemical detection in a BAS Biophase ODF (C-18) column and a BAS LC-4B detector (Bio Analytical Systems) with a detection threshold of 0.24 nmol/L for norepinephrine and of 0.27 nmol/L for epinephrine. Interassay and intra-assay coefficients of variation are 3.8% and 3.6%, respectively, for norepinephrine and 6.7% and 4.2%, respectively, for epinephrine. Serum potassium was assayed by spectrophotometry.
Tracings of sympathetic neurograms, ECGs, respiration, central venous pressure, and FBF were recorded on a physiological recorder (model 2800S, Gould Inc) at a paper speed of 5 mm/s and also with a MacLab data aquisition system (AD Instruments) on a Macintosh Computer (Apple Inc). Sympathetic bursts were identified by inspection. Interobserver and intraobserver variabilities in identifying bursts average 5.4±0.5% (range, 0% to 20%) and 4.3±0.3% (range, 0% to 14%), respectively.1 2 FBF is expressed as mL·min−1·100 mL forearm volume−1. FVR is calculated by dividing mean arterial pressure (diastolic+1/3 pulse pressure) by flow and is expressed in arbitrary units.
For each parameter (MSNA, heart rate, blood pressure, FBF, and central venous pressure), every 5-minute period of data collection was averaged to a single value. Data are presented as mean±SE.
For statistical analysis, repeated-measures ANOVA, Student's t test, and least-squares linear regression analysis were used where applicable. A value of P<.05 was considered significant. Demographic data were compared by use of an unpaired t test. Data were analyzed by repeated-measures ANOVA with time as within subjects factor and group (elderly versus young) and, in the elderly, session (insulin versus vehicle) as between groups factors. Post hoc comparisons were made by application of planned contrasts (baseline versus 90 minutes of insulin or vehicle infusion). Correlation analyses were performed by least-squares linear regression analysis.
Insulin Infusion Studies
Plasma insulin and glucose levels
Fasting plasma insulin levels averaged 54±12 pmol/L in the elderly and 66±12 pmol/L in the young subjects (Table 3⇓). Steady-state plasma insulin levels were reached after 30 minutes of insulin infusion, averaging 516±54 pmol/L in the elderly and 534±54 pmol/L in the young subjects.
Fasting plasma glucose levels averaged 5.1±0.1 mmol/L in the elderly and 4.7±0.1 mmol/L in the young subjects and did not change significantly during the insulin clamp.
M values (mg·kg−1·min−1 metabolized glucose) calculated from glucose infusion rates during minutes 60 to 90 of insulin infusion were significantly higher in the young subjects (8.4±1.4 versus 4.3±0.4 mg·kg−1·min−1, P<.05), indicating that the elderly subjects were more insulin resistant.
Elderly subjects had higher baseline MSNA than young subjects (33±3 versus 15±2 bursts per minute, P<.05). With insulin, MSNA increased to 43±4 and 31±3 bursts per minute in elderly and young subjects, respectively (P<.05 versus baseline). The absolute as well as the relative magnitude of increase was diminished in the elderly compared with the young subjects (+10±1 versus +15±1 bursts per minute, or +37±11% versus +110±16%, P<.05). In both groups, it appeared that fasting plasma insulin levels correlated positively with baseline MSNA; however, with the number of subjects studied, this did not reach statistical significance in either group (r=.56, P=.09 for the elderly and r=.45, P=.17 for the young subjects).
Baseline FBF was higher and FVR was lower in the elderly subjects. With insulin, FBF decreased in the elderly subjects (from 2.7±0.3 mL·min−1·100 mL−1 during baseline to 2.3±0.2 mL·min−1·100 mL−1 at the end of the clamp, P<.05), whereas it increased in young subjects (from 2.0±0.1 to 3.2±0.4 mL·min−1·100 mL−1, P<.05). Conversely, FVR increased in the elderly (from 37±3 to 47±7 units, P<.05) and decreased in the young subjects (from 46±2 to 31±3 units, P<.05). In the young subjects, the increment in FBF and the decrease in FVR correlated positively with the M values (r=.67, P<.05 and r=.64, P<.05 for increase in FBF and decrease in FVR, respectively), whereas in the elderly there was no significant correlation between these variables (r=.48, P=NS and r=.22, P=NS).
Heart rate, blood pressure, and central venous pressure (Table 3⇑)
At baseline, heart rate was similar in elderly and young subjects. In the young subjects, heart rate increased with insulin (from 62±3 bpm during baseline to 67±3 bpm after 90 minutes of infusion, P<.05), whereas in the elderly subjects, heart rate did not change with insulin (62±2 and 63±3 bpm during baseline and after 90 minutes of infusion, respectively, P=NS).
Compared with young subjects, elderly subjects had higher baseline systolic (126±3 versus 118±2 mm Hg, P<.05) and diastolic (79±2 versus 69±2 mm Hg, P<.05) blood pressures. However, insulin did not significantly alter blood pressure in either group.
Baseline central venous pressure was similar in the young and the elderly subjects. Insulin decreased central venous pressure slightly in both groups.
Plasma norepinephrine and epinephrine
Elderly subjects had significantly higher baseline plasma norepinephrine levels than young subjects (1.89±0.27 versus 1.04±0.14 nmol/L, P<.05) (Fig 2⇑, Tables). In the young subjects, plasma norepinephrine increased significantly with insulin to 1.37±0.18 nmol/L, P<.05, whereas in the elderly subjects, plasma norepinephrine levels did not change significantly with insulin. Plasma epinephrine levels did not increase during the insulin infusion (data not shown).
Serum potassium decreased slightly with insulin in both groups (from 4.1±0.1 to 3.8±0.1 mmol/L in both groups, P<.05); however, no subject developed hypokalemia.
Vehicle Infusion Studies
No variable changed significantly during the vehicle session performed in elderly subjects with the exception of a slight increase in systolic (+6±2 mm Hg), diastolic (+4±2 mm Hg), and mean (+5±2 mm Hg) arterial pressures (P<.05).
There are two major findings in the present study. First, in healthy normotensive but insulin-resistant elderly subjects, acute physiological elevations of plasma insulin caused vasoconstriction, as opposed to the vasodilation observed in young control subjects. Second, insulin-induced sympathetic activation and increases in norepinephrine and heart rate were attenuated in elderly subjects. These data indicate that the insulin-induced vasoconstriction is not due to exaggerated insulin-induced sympathetic activation but rather to a reduction in the vasodilator action of insulin.
Acute increases in plasma insulin cause opposing vasodilation and sympathetic activation in normal young adults, which is consistent with previous reports.1 2 3 4 The novel finding in this study is that despite blunted sympathetic activation in response to insulin, insulin induces vasoconstriction and not vasodilation in healthy normotensive elderly subjects. Previous reports on insulin-mediated vasodilation in elderly humans are equivocal: Boden et al13 observed similar increases in leg blood flow with insulin in elderly and young subjects. In one other recent report of the vascular actions of insulin in the elderly, insulin appears to produce no significant change in calf blood flow in healthy elderly subjects; calf vascular resistance was not reported.12 However, that study did not appear to show vasoconstriction in elderly subjects. We speculate that higher steady-state plasma insulin levels in our subjects might contribute to the different results of our study.
Both the absolute and relative increases in MSNA during insulin were diminished in the elderly subjects. Plasma norepinephrine increased during insulin infusion in the young but not in the elderly subjects. This is consistent with the observations of Minaker et al,28 who used insulin doses even higher than those in our study. Thus, mechanisms other than exaggerated sympathoexcitation or catecholamine release are responsible for the vasoconstrictor effect of insulin in older subjects.
Another, perhaps more plausible explanation for the vasoconstriction in the elderly is that the direct vasodilator action of insulin is markedly attenuated or absent in elderly subjects and therefore overridden by even moderate sympathoexcitation.
Indeed, a series of observations suggest that the vasodilator actions of insulin could be attenuated in the elderly. Two groups recently reported that the vasodilator action of insulin in humans depends on nitric oxide,29 30 presumably of endothelial origin.30 Several lines of evidence show that nitric oxide–dependent vasodilation is impaired with increasing age: Kung and Lüscher31 showed attenuated endothelium-dependent vasodilation due to diminished nitric oxide production in senescent WKY rats. Celermeier et al32 and Corretti et al33 observed that flow-dependent vasodilation, which has been shown to be nitric oxide dependent,34 is impaired in elderly humans. Recently, by intra-arterial infusion of metacholine and sodium nitroprusside, Gerhard et al35 showed a progressive impairment of endothelium-dependent vasodilation with increasing age in healthy humans. Thus, an age-related impairment of endothelial function may be responsible for a failure of insulin-induced vasodilation in elderly individuals.
The insulin-induced vasoconstriction in elderly humans may itself be implicated in insulin resistance in elderly subjects. In the present study, the young subjects with the greatest insulin-induced increase in FBF were the most insulin sensitive. Laakso et al8 9 suggest that the failure of insulin to stimulate skeletal muscle blood flow causes insulin resistance in obesity and type II diabetes by decreasing glucose delivery. A similar mechanism may underlie at least part of the insulin resistance that we and others10 11 12 13 observed in elderly subjects, as suggested by Meneilly et al.12
Baron and Brechtel36 reported that in young adults, the insulin-induced increase in heart rate is paralleled by an increase in cardiac output, which prevents a fall in blood pressure during peripheral vasodilation. In contrast to our findings in young subjects, we did not observe an increase in heart rate with insulin in the elderly subjects. Thus, although we did not measure cardiac output, we speculate that insulin may not increase and may even decrease cardiac output in elderly subjects in the absence of vasodilation. Interestingly, blood pressure did not increase in the elderly subjects, despite insulin-induced sympathetic activation and vasoconstriction. A decrease in cardiac output or vasodilation in other vascular beds (eg, splanchnic bed) may explain why blood pressure did not increase with insulin despite vasoconstriction in the skeletal muscle vascular bed.
Potential Limitations of the Study
First, it may be argued that factors other than age contribute to the observed group differences. Although elderly and young subjects were matched for body mass indexes, the elderly had a higher percent body fat. We found no significant correlation between percent body fat and changes in FVR or MSNA in either young or elderly subjects. Thus, we do not believe that the differences in percent body fat contribute to the differential effects of insulin.
Second, Moan et al37 questioned whether insulin truly causes sympathetic activation, because they reported that a vehicle control experiment produced changes in sympathetic activity similar to those of a hyperinsulinemic/euglycemic clamp. In contrast to these observations, the present study demonstrates that in elderly subjects, a vehicle control experiment has no effect on MSNA. Furthermore, we have reported that vehicle infusion over 90 minutes does not change MSNA in young healthy adults.26 Thus, we believe that the observed changes in the present study are not attributable to time or vehicle.
Third, baseline FVR was lower in the elderly subjects. This, however, is consistent with the finding of Lipsitz et al.38 Because of body composition differences between the two groups, differences in the absolute plethysmographic measure of baseline FBF do not necessarily reflect differences in baseline vascular tone. However, intervention-induced changes in FBF in each group are a reliable measure of intervention-induced changes in vascular tone.24
In healthy young adults, insulin induces sympathetic activation and vasodilation. In contrast, in healthy, insulin-resistant elderly individuals, the increases in MSNA and heart rate are blunted and vasodilation is replaced by vasoconstriction. A blunted vasodilator response to insulin in the elderly may permit even modest insulin-induced sympathoactivation to elicit vasoconstriction. We speculate that the failure of the vasodilator action of insulin may be explained by an age-related impairment of endothelial function and that the overall vasoconstrictor effect of insulin in elderly subjects may further potentiate insulin resistance in this population.
Selected Abbreviations and Acronyms
|FBF||=||forearm blood flow|
|FVR||=||forearm vascular resistance|
|MSNA||=||muscle sympathetic nerve activity|
This work was supported by National Institutes of Health (NIH) grants HL-43514, HL44546, and HL-24962; by Veterans Administration Merit Review funds; by grant RR59 from the General Clinical Research Center Program, Division of Research Resources, NIH; and by a training grant from the German Research Association (Deutsche Forschungsgemeinschaft). We are indebted to Elaine S. Paul for her technical assistance, to the nurses and dietitians of the Clinical Research Center, and to Dr Phyllis Stumbo, Cathy Shenard, and Paul Muhle.
- Received June 11, 1996.
- Revision received July 12, 1996.
- Revision received September 25, 1996.
Anderson EA, Hoffman RP, Balon TW, Sinkey CA, Mark AL. Hyperinsulinemia produces both sympathetic neural activation and vasodilation in normal humans. J Clin Invest. 1991;87:2246-2252.
Rowe JW, Young JB, Minaker KL, Stevens AL, Pallotta J, Landsberg L. Effect of insulin and glucose infusions on sympathetic nervous system activity in normal men. Diabetes. 1981;30:219-225.
Vollenweider P, Randin D, Tappy L, Jequier E, Nicod P, Scherrer U. Impaired insulin-induced sympathetic neural activation and vasodilation in skeletal muscle in obese humans. J Clin Invest. 1994;93:2365-2371.
Landsberg L. Obesity and hypertension: experimental data. J Hypertens. 1992;10(suppl 7):S195-S201.
Reaven GM. Role of insulin resistance in human disease. Diabetes. 1988;37:1595-1607.
Laakso M, Edelman SV, Brechtel G, Baron AD. Decreased effect of insulin to stimulate skeletal blood flow in obese man: a novel mechanism for insulin resistance. J Clin Invest. 1990;85:1844-1852.
Laakso M, Edelman SV, Brechtel G, Baron AD. Impaired insulin-mediated skeletal muscle blood flow in patients with NIDDM. Diabetes. 1992;41:1076-1083.
Rowe JW, Minaker KL, Pallotta JA. Characterization of the insulin resistance of aging. J Clin Invest. 1983;71:1581-1587.
Boden G, Chen X, DeSantis RA, Kendrick Z. Effects of age and body fat on insulin resistance in healthy men. Diabetes Care. 1993;16:728-733.
Ng AV, Callister R, Johnson DG, Seals DR. Age and gender influence on muscle sympathetic nerve activity at rest in healthy humans. Hypertension. 1993;21:498-503.
Iwase S, Mano T, Watanabe T, Saito M, Kobayashi F. Age-related changes of sympathetic outflow to muscles in humans. J Gerontol. 1991;46:M1-M5.
Fleg JL, Tzankoff SP, Lakatta KG. Age-related augmentation of plasma catecholamine during dynamic exercise in healthy males. J Appl Physiol. 1985;59:1033-1039.
Sowers JR, Rubenstein LZ, Stern N. Plasma norepinephrine responses to posture and isometric exercise increase with age in the absence of obesity. J Gerontol. 1983;38:315-317.
Palmer GJ, Ziegler MG, Lake CR. Response of norepinephrine and blood pressure to stress increases with age. J Gerontol. 1978;33:482-487.
National High Blood Pressure Education Program Working Group. National high blood pressure education program working group report on hypertension in the elderly. Hypertension. 1994;23:275-285.
National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes. 1979;28:1039-1057.
Siggard-Anderson J. Venous occlusion plethysmography on the calf. Dan Med Bull. 1970;17:1-68.
Yalow RS, Berson SA. Immunoassay of endogenous plasma insulin in man. J Clin Invest. 1960;39:1157-1175.
Scherrer U, Randin D, Vollenweider P, Vollenweider L, Nicod P. Nitric oxide release accounts for insulin's vascular effects in humans. J Clin Invest. 1994;94:2511-2515.
Steinberg HO, Brechtel G, Johnson A, Fineberg N, Baron AD. Insulin-mediated skeletal muscle vasodilation is nitric oxide dependent. J Clin Invest. 1994;94:1172-1179.
Kung CF, Lüscher TF. Different mechanisms of endothelial dysfunction with aging and hypertension in rat aorta. Hypertension. 1995;25:194-200.
Joannides R, Haefeli W, Linder L, Richard V, Bakkali EH, Thuillez C, Lüscher TF. Nitric oxide is responsible for flow-dependent dilatation of human peripheral conduit arteries in vivo. Circulation. 1995;91:1314-1319.
Gerhard M, Roddy MA, Creager SJ, Creager MA. Aging progressively impairs endothelium dependent vasodilation in forearm resistance vessels of humans. Hypertension. 1996;27:849-853.
Baron AD, Brechtel G. Insulin differentially regulates systemic and skeletal muscle vascular resistance. Am J Physiol. 1993;265:E61-E67.
Moan A, Høieggen A, Nordby G, Eide I, Kjeldsen SE. Evidence against activation of the sympathetic nervous system by insulin. Proc 7th European Meeting on Hypertension, Ricerca Scietifica ed Educazione Permanente. 1995;suppl 103:126. Abstract 552.