(Circulation. 1997;96:1185-1191.)
© 1997 American Heart Association, Inc.
Articles |
Do High Proinsulin and C-Peptide Levels Play a Role in Autonomic Nervous Dysfunction?
Power Spectral Analysis in Patients With Non–Insulin-Dependent Diabetes and Nondiabetic Subjects
From the Departments of Clinical Physiology (J.P.T., M.J.M., E.A.L.), Clinical Nutrition (L.K.N., M.I.J.U.), and Internal Medicine (L.K.N.), Kuopio University Hospital and University of Kuopio, Kuopio, Finland, and the Division of Clinical Epidemiology (S.M.H., H.J.J.M.), Department of Medicine, University of Texas Health Science Center at San Antonio, San Antonio, Tex.
Correspondence to Jari Töyry, MD, Department of Clinical Physiology, Kuopio University Hospital, PO Box 1777, FIN-70210 Kuopio, Finland. E-mail jtoyry{at}messi.uku.fi
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background Immunoreactive insulin has been shown to predict the development of parasympathetic autonomic neuropathy. It is possible that constituents of immunoreactive insulin could explain this association. In this cross-sectional study, the relationship of specific insulin, C-peptide, and proinsulin with autonomic nervous dysfunction was evaluated in 57 NIDDM patients and 108 control subjects.
Methods and Results The frequency-domain analysis of heart rate variability was determined by using spectral analysis from stationary regions of registrations while the subjects breathed spontaneously in a supine position. Total power was divided into three frequency bands: low (0 to 0.07 Hz), medium (MFP, 0.07 to 0.15 Hz), and high (HFP, 0.15 Hz to 0.50 multiplied by the frequency equal to the mean RR interval). In NIDDM patients, total power, the three frequency bands (P<.001 for each), and the MFP/HFP ratio (P=.016), which expresses sympathovagal balance, were reduced compared with control subjects. Fasting proinsulin (rs=-.324, P=.014 for diabetics and rs=-.286, P=.003 for control subjects), C-peptide (rs=-.492, P<.001 for diabetics and rs=-.304, P=.001 for control subjects), and total immunoreactive insulin (rs=-.291, P=.028 for diabetics and rs=-.228, P=.017 for control subjects) were inversely related to MFP/HFP. For proinsulin and C-peptide the results did not change after controlling for the effects of age, body mass index, and fasting glucose.
Conclusions Both proinsulin and C-peptide levels were significantly associated with the sympathovagal balance of autonomic nervous function in NIDDM patients and control subjects, but this study cannot determine whether these compounds are directly involved in autonomic nervous dysfunction.
Key Words: diabetes mellitus • nervous system, autonomic • insulin
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Cardiac autonomic neuropathy is known to worsen the prognosis of patients with insulin-dependent diabetes mellitus.1 2 In our recent study3 a sharp increase in the frequency of autonomic neuropathy was found 5 years after diagnosis in patients with NIDDM, and autonomic nervous dysfunction predicted cardiovascular mortality in these patients. Furthermore, high immunoreactive plasma insulin levels were associated with the development of autonomic neuropathy in NIDDM.3 The finding that plasma insulin could contribute to the damage of the autonomic nervous system is intriguing since insulin may have a direct influence on sympathetic nervous system activity, at least in the short-term.4 This question is also relevant in terms of a possible effect of plasma insulin on cardiovascular morbidity and mortality.5 6 7 8 However, in NIDDM, conventional RIA measurements of plasma insulin cross-react with proinsulin. Temple et al9 have provided suggestive evidence that the major part of circulating immunoreactive insulin in NIDDM patients is composed of proinsulin and its conversion intermediates. Other studies have confirmed the disproportionate elevation of proinsulin in NIDDM10 11 12 13 14 15 and possibly also in subjects with impaired glucose tolerance.16 17 The prevailing view is that elevated proinsulin is a marker of impending ß-cell failure,18 and its biological functions, if any, are poorly known. Recent studies have challenged the old dogma19 and suggested that C-peptide could also play a physiological role by stimulating glucose transport in skeletal muscle and ameliorating autonomic nervous dysfunction. Therefore, one could speculate that our previous results3 concerning the relation of plasma immunoreactive insulin to the autonomic nervous dysfunction in patients with NIDDM could also be ascribed to insulin per se (specific insulin) as well as to proinsulin or even to C-peptide.
In our earlier study,3 a conventional deep breathing test for parasympathetic and an orthostatic test for sympathetic autonomic nervous function were used.20 In the present study, PSA of HRV was introduced to assess autonomic function under various conditions. Its advantage is the simultaneous assessment of the sympathetic and parasympathetic components of autonomic nervous function.21 22 23 The HF component of spectral HRV is almost exclusively mediated by vagal activity, and the MF component gives a measure of sympathetic activity with some influence from vagal activity. The sympathovagal balance, on the other hand, may be assessed by examining either MF/HF or the normalized MF and HF components of spectral analysis of HRV.24
The aim of the present study was to examine the role of proinsulin, specific insulin, and C-peptide in the pathogenesis of autonomic nervous dysfunction as evaluated by PSA of HRV in well-characterized NIDDM patients and nondiabetic control subjects.3
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study is a cross-sectional analysis of a 10-year follow-up study of patients with newly diagnosed NIDDM and control subjects.25 The baseline study population consisted of 133 patients with newly diagnosed NIDDM aged 45 to 64 years and 144 randomly selected nondiabetic control subjects of the same age group. The selection of both groups was carried out from May 1, 1979, through December 31, 1981. Both groups were recruited from a defined area of 180 000 inhabitants in the county of Kuopio in eastern Finland.26 Approval for the study was given by the ethics committee of Kuopio University Central Hospital. Informed consent was given by all the subjects studied.
The diabetic patients (70 men and 63 women) were referred to the study
by general practitioners working in community health
centers in the survey area. The diagnosis of diabetes was made in the
clinical setting and was confirmed by an oral glucose tolerance test
using diagnostic criteria recommended by the World Health
Organization Expert Committee on Diabetes Mellitus.27
Subjects with fasting blood glucose levels >7.0 mmol/L for >6
months as well as subjects with secondary diabetes, hypothyroidism,
hyperthyroidism, alcoholism, renal insufficiency, overt carcinoma, or
those in institutional care were not eligible for the study. All the
diabetic patients were nonketotic at the time of diagnosis. The
nondiabetic control population of the same age group (62 men and 82
women) was selected from the population register using random number
tables. The formation, representativeness, and methods
of the baseline examination are available.26 The 10-year
examination was performed between October 1991 and December
1992.25 28 NIDDM patients receiving insulin treatment
(n=18) were excluded from the study. Characteristics of the present
cross-sectional study population are shown in Table 1
.
|
Clinical and Biochemical Characteristics
The following examinations and measurements were done in 1991
through 1992: clinical history (including data on medication),
anthropometric measurements, blood pressure, resting ECG, oral glucose
tolerance test, levels of fasting plasma glucose, plasma insulin,
proinsulin, plasma C-peptide, glycosylated hemoglobin A1C,
serum lipids and lipoproteins, urinary albumin excretion, and
autonomic nervous function tests. BMI was calculated as body weight in
kilograms divided by height in meters squared. Blood pressure was
measured after a 5-minute rest in the sitting position (cuff size,
12.5x40.0 cm). SBP and DBP levels were measured to the nearest 2
mm Hg. The mean of three recordings was used in the
analysis. A conventional 12-lead resting ECG was recorded
for each subject and interpreted according to the Minnesota
code.29 An oral glucose tolerance test was performed after
a 12-hour overnight fast according to World Health Organization
recommendations27 by using a 75-g glucose dose. Samples
for plasma glucose, serum insulin, and C-peptide determinations were
taken before the test (fasting) and at 1 and 2 hours afterward. Samples
for serum insulin determinations were placed in prechilled tubes,
centrifuged, and stored without delay at -70°C until
analyzed. Proinsulin and specific insulin were analyzed
from fasting samples only.
Laboratory Methods
Venous plasma glucose was analyzed by using a glucose
oxidase method (Glucose Auto & Stat HGA-1120 Analyzer, Daiichi
Co). Serum immunoreactive insulin was analyzed by a using
double-antibody RIA (Phasedeph, Pharmacia). The detection limit of the
assays was 15.0 pmol/L, and the coefficient of variation between
duplicate aliquots measured at the same time was 5.1% to 6.1%. Serum
specific insulin concentrations were measured by using a commercial
double-antibody RIA (human insulin–specific RIA method, Linco
Research) in which cross-reactivity with proinsulin is <0.2%. The
lower limit of sensitivity of the Linco assay is 14.4 pmol/L. The
intra-assay coefficient of variation was 4.5%, and the interassay
coefficient of variation was <10%.30 Fasting proinsulin
concentrations were measured by using a nonequilibrium RIA method in
the laboratory of Dr S.M. Haffner, San Antonio, Tex.30
This method was modified slightly to improve the sensitivity at low
concentrations of proinsulin. Antibody was obtained from Linco
Research. The polyclonal antibody used in this assay (168AB) recognizes
a proinsulin-specific epitope formed by the intact A-chain–C-peptide
junction. In this assay, the potency of human insulin and C-peptide is
<0.1% that of proinsulin. Under nonequilibrium conditions,
A-chain–C-peptide junctional cleaved forms of proinsulin are <1% as
potent as intact proinsulin, whereas B-chain–C-peptide junctional
cleaved forms, such as des 31,32 proinsulin, have a cross-reactivity
>95%. Because des 31,32 is the major circulating form of split
proinsulin (
95%), the proinsulin RIA method reported here provides
an estimate for the total concentration of proinsulin
(intact+B–C-junctional cleaved forms) in plasma. The intra-assay
coefficient of variation ranged from 6% to 21% using controls
prepared at 5, 50, and 250 pmol/L.31 Plasma C-peptide was
determined by using an RIA (125-I; Incstar) in which cross-reactivity
with proinsulin is 
4.0%. Serum and lipoprotein lipids were
determined from 12-hour fasting samples. Lipoproteins were
analyzed by using enzymatic methods after
ultracentrifugation and precipitation.32
Glycosylated hemoglobin A1C was measured by using liquid
cation-exchange chromatography (normal range, 4.0% to
6.0%). Urinary albumin excretion was measured from timed
overnight urine samples by using kinetic rate nephelometry on an array
protein analyzer using microalbumin reagent (Beckman);
the lower limit of the assay is 2.0 mg/L.
Classification for the Diagnosis of MI
The definite MI class consisted of patients with major Q-QS
abnormalities (Minnesota code 1.1 to 1.2), those who had suffered an MI
verified at hospital, or both. All patient records were checked to
verify the correct diagnosis of MI.25
Autonomic Function Tests
ECGs (Rigel MultiCare 302, Rigel Research Ltd) and continuous
noninvasive arterial pressure signals from the middle
finger (Finapress, Ohmeda, Inc) were recorded and
simultaneously analog-to-digital converted with a temporal
resolution of 200 Hz per channel and an amplitude resolution of 12
bits.33 The converted signals were stored in an IBM PC/AT
compatible microcomputer. A software QRS-detection algorithm modified
from Engelese and Zeelenberg34 was used to define R peaks
of QRS complexes with an accuracy of better than 2 ms. Beat-to-beat RR
intervals (no ectopic beats were included) and systolic and
diastolic arterial pressures were recorded.
All data acquisition and analyses were performed with a
menu-driven software package (CAFTS, Medikro Ltd).
The quiet breathing test was done in the morning between 9 and 10 AM in the supine position after a 5-minute resting period. The subjects were asked to refrain from using ß-blocking agents and diuretics for 24 hours before the study and to not use alcohol for 48 hours before the study. In the quiet breathing test the subjects breathed freely, maintaining normal tidal volume, for 5 minutes. The time-domain analysis of HRV was assessed by calculating the root mean squared successive difference by using the formula of von Neumann et al.35 The frequency-domain analysis of HRV was determined by using spectral analysis of HRV. Spectral estimations of RR interval variability were obtained from stationary regions of registrations. The mean numbers of sampled RR intervals were 231 (range, 61 to 255) for diabetic and 228 (range, 57 to 255) for control subjects. After detrending of the signals (first degree), a least-mean-square autoregressive model with a model order of 18 was used to obtain a power spectral estimate of RR interval variability. Total power (variance) was divided into three frequency bands: LFP (0.0 to 0.07 Hz), MFP (0.07 to 0.15 Hz), and HFP (0.15 Hz to 0.50 multiplied by the frequency equal to the mean RR interval). Signal powers in the three frequency bands were calculated as integrals under the respective power spectral density function and were expressed in absolute units in milliseconds squared. In addition, the MFP/HFP ratio and the frequency band/total power ratio (ie, normalized unit for each frequency band) were calculated.
Statistical Analysis
Statistical analyses were conducted with the SPSS/PC+
program (SPSS Inc). Results are expressed as mean±SD. Normality of the
distributions was assessed both graphically and with a goodness-of-fit
test. The spectral parameters, immunoreactive insulin,
specific insulin, and proinsulin data were analyzed after
logarithmic transformation to improve skewness and kurtosis, but
untransformed units are presented. The differences between the
two groups were assessed by using Student's t test or
2 test. Spearman's rank correlation coefficient
(rs) was calculated to assess the association of
selected variables with MF/HF or normalized MF and HF components.
The proportion of the variability was calculated as
100xrS2 with a 95% CI. Age, BMI,
and fasting glucose were regarded as potential confounding factors in
stepwise multiple regression analyses between fasting
proinsulin and MF/HF, C-peptide and MF/HF, and specific insulin and
MF/HF, respectively. Age, BMI, and presence of diabetes were regarded
as covariates in the ANCOVA concerning the effect of specific insulin,
proinsulin, and C-peptide on MF/HF for combined cohorts. The 25%
cumulative frequency calculated from control subjects was considered as
an abnormal MF/HF (21%). Multiple logistic regression analyses
were also performed to determine any independent associations of
specific insulin, proinsulin, and C-peptide with MF/HF in the combined
cohorts. A probability value <.05 was considered significant. For
technical reasons, complete data were not obtained from all subjects.
Because subjects with atrial fibrillation at the time of autonomic
testing (n=3) were excluded from the analysis, the number of
subjects examined varied slightly from test to test.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Clinical Characteristics
No statistically significant differences were found in sex, BMI, SBP, DBP, or fasting specific insulin or fasting C-peptide levels between NIDDM patients and control subjects, but the diabetic patients were somewhat older than control subjects. The supine RR interval was shorter in diabetic patients than control subjects (P=.008). The use of ß-blocking agents or diuretics and a history of MI were more frequent in diabetic patients than control subjects. Fasting immunoreactive insulin and fasting proinsulin levels were higher in NIDDM patients (Table 1
). At the 10-year examination 24.6% of NIDDM
patients were treated with diet and 75.4% with oral antidiabetic drugs
(all insulin-treated subjects were excluded).
Autonomic Nervous Function Tests
Total power, every spectral component (LFP, MFP, and HFP), and
MF/HF were significantly lower in the quiet breathing test in NIDDM
patients than control subjects (Table 2). Fig 1 gives a typical PSA curve for an NIDDM patient and a
control subject. MF/HF did not differ during the quiet breathing test
between men and women in either NIDDM patients or control subjects
(P=.591 and P=.561, respectively). Total power
and each spectral component were significantly lower in women than men
among NIDDM patients (P<.001 to .001 for each), but such a
difference could not be found among control subjects (P=.425
to .945). Total power, LFP, MFP, HFP, and MF/HF did not differ between
those with and without a history of MI among NIDDM patients or control
subjects (P=.255 to .893). Total power, LFP, MFP, HFP, and
MF/HF did not differ between those who used ß-blocking medication and
those who did not among either NIDDM patients or control subjects
(P=.083 to .978). Among NIDDM patients, LFP
(P=.018) and MF/HF (P=.024) were significantly
lower in those using compared with those not using diuretics,
but such a difference could not be found in total power, MFP, and HFP
in both groups and in LFP and MF/HF in control subjects
(P=.067 to .933).
|
|
Root mean squared successive difference was significantly lower in the
quiet breathing test in NIDDM patients than control subjects (Table 2).
Factors Associated With MF/HF or Normalized MF or HF Components of
Spectral Analysis
Table 3 summarizes the main associations with MF/HF
during the quiet breathing test. No associations with age, BMI, SBP,
DBP, or urinary albumin excretion were seen. HDL
cholesterol associated positively with MF/HF in NIDDM
patients, and LDL triglycerides were negatively associated
with MF/HF in control subjects (P=.015 and
P=.011, respectively). Fasting plasma glucose and total
immunoreactive insulin levels associated negatively with MF/HF in both
groups (Table 3). Fasting plasma proinsulin and C-peptide levels were
also negatively associated with MF/HF in both groups, and fasting
specific insulin in control subjects, even after controlling for
the effects of age, BMI, and fasting glucose (by multiple stepwise
regression analysis, r=-.323, P=.014 and
r=-.211, P=.031 for proinsulin and
r=-.397, P=.002 and r=-.274,
P=.004 for C-peptide in NIDDM patients and control subjects,
respectively, and r=-.226, P=.021 for specific
insulin in control subjects; Fig 2). Fasting proinsulin
explained 10.5% (95% CI, 0.5% to 29.1%) of the MF/HF variability in
NIDDM patients and 8.2% (95% CI, 1.0% to 20.5%) in control
subjects. The proportion of the variability of MF/HF for C-peptide was
24.2% (95% CI, 6.9% to 44.7%) in NIDDM patients and 9.2% (95% CI,
1.5% to 21.7%) in control subjects; that for fasting specific insulin
was 6.6% (95% CI, 0.5% to 18.2%) in control subjects.
|
|
; n=57), rS=-0.256, P=.008
for control subjects (
; n=105), and
rS=-0.211, P=.007 for combined
cohorts; (b) proinsulin level (logarithmic transformation):
rS=-0.324, P=.014 for NIDDM patients
(n=57), rS=-0.286, P=.003 for
control subjects (n=105), and rS=-0.349,
P<.001 for combined cohorts; and (c) C-peptide level:
rS=-0.492, P<.001 for NIDDM
patients (n=56), rS=-0.304, P=.001
for control subjects (n=108), and rS=-0.398,
P<.001 for combined cohorts.Fasting plasma specific insulin (F=3.725, P=.055), proinsulin (F=5.315, P=.022), and C-peptide (F=20.231, P<.001; all by ANCOVA) levels were significantly higher in those subjects with low MF/HF (an MF/HF of 21% was used as a cutoff limit, ie, 25% cumulative frequency from control subjects) when NIDDM patients and control subjects were combined, even after controlling for the effects of age, BMI, and the presence of diabetes. Multiple logistic regression analyses gave the same results for combined cohorts (data not shown).
Fasting total immunoreactive insulin (rs=-.293, P=.027), fasting proinsulin (rs=-.333, P=.011), and fasting C-peptide (rs=-.395, P=.003) levels were negatively associated with the normalized MF component in NIDDM patients, and the same was true with fasting C-peptide levels in control subjects (rs=-.187, P=.052). Fasting immunoreactive insulin, proinsulin, and C-peptide levels were not associated with the normalized HF component in either group (P=.291 to .889).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this cross-sectional study of control subjects and NIDDM patients 10 years after diagnosis, we demonstrated a marked decrease in the autonomic nervous function in NIDDM patients compared with nondiabetic subjects as assessed by PSA of HRV. This finding is in line with our previous observations based on conventional autonomic nervous function tests.3 Interestingly, in the present study both proinsulin and C-peptide levels showed reasonably high and consistent associations with the sympathovagal balance of autonomic nervous function in NIDDM patients and control subjects. The finding that endogenous proinsulin or C-peptide could play a role in the regulation of autonomic nervous function or might even be involved in the pathogenesis of autonomic nervous damage has not been previously reported.
Assessment of cardiovascular autonomic nervous function can be made from the spectral analysis of HRV.21 22 23 The parasympathetic component of autonomic nervous function is located around the peak (high) frequency of 0.2 Hz in response to atropine.36 37 38 The sympathetic component lies around the peak frequency of 0.1 Hz (named as medium or low frequency), even though this frequency band is partly related to the degree of parasympathetic function.39 In an animal model, the direct electrical stimulus of either the sympathetic or parasympathetic parts of the autonomic nervous system modulated both the "sympathetic" and "parasympathetic" frequency bands of PSA of HRV.40 The balance between the sympathetic and parasympathetic components of autonomic nervous function can be assessed with MF/HF.39 The effect of different interventions on sympathovagal balance has been assessed with MF/HF or the normalized MF or HF components of spectral analysis of HRV.24 An autoregressive model for short-term recordings was used in our study protocol.41 Furthermore, the stationary regions used in the spectral analysis are data segments free from ectopic beats and artifacts.42 The effects of ß-blocking agents and diuretics as confounders could be excluded because their use did not show any association with the sympathetic or parasympathetic components of autonomic nervous function either in NIDDM patients or control subjects. In the present study, moreover, these drugs were withdrawn before the autonomic testing.
The interpretation of PSA of HRV is based on the assumption that the time intervals of impulses after their origination in the sinoatrial node are static in vivo. This implies that no changes in the rate of rapid ventricular depolarization or the duration of ventricular electrical systole and diastole occur and further that the change in the thoracic resistive load will not affect the time period of the propagating impulse. This suggests that the time interval variability recognized at the body surface potentials (ie, R waves) is possibly based on autonomic nervous regulation. We cannot exclude the possibility that high proinsulin or high C-peptide levels could affect the mechanisms contributing to impulse-conducting time intervals at or after the sinoatrial node in addition to autonomic nervous function.
In the early phase of NIDDM preserved sympathetic response to exogenous physiological hyperinsulinemia has been found by using the [3H]norepinephrine tracer method,4 but there are little data on the effect of endogenous insulin on the autonomic nervous system. We have demonstrated3 that high immunoreactive insulin predicts the development of parasympathetic neuropathy as assessed with time-domain analysis (ie, expiration-to-inspiration ratio during deep breathing), but no significant association with sympathetic neuropathy was found as assessed with an SBP decrease during orthostatic testing. These tests do not, however, account for the instantaneous components of autonomic nervous function. Therefore, one could speculate that the effect of immunoreactive insulin or its precursors might alter the sympathovagal balance as evaluated with a more sensitive method like PSA of HRV. Accordingly, high total immunoreactive insulin levels, and even more consistently, high proinsulin and C-peptide levels, were associated with sympathovagal balance toward the decreased sympathetic autonomic function, as suggested by their negative correlation with the sympathetic frequency band. Although the NIDDM patients of the present study also had parasympathetic nervous damage,3 they could still have a relative increase in the parasympathetic component of autonomic regulation, but there are no reliable methods to show this.
A novel finding of our study was that proinsulin and C-peptide were associated with autonomic nervous dysregulation as estimated by MF/HF not only in patients with NIDDM but also in nondiabetic control subjects. Exclusion of subjects with impaired glucose tolerance from control subjects did not alter the main results (data not shown), nor was this association explained by confounding factors (age, sex, obesity, or prevailing plasma glucose levels). Note that the statistical effects of plasma proinsulin and C-peptide were, if anything, more consistent than that of plasma specific insulin, and correlation coefficients were reasonably high, suggesting that circulating proinsulin and C-peptide could indeed have a biological relationship with the autonomic nervous system. The finding that proinsulin and C-peptide were similarly associated with the autonomic nervous function in NIDDM patients and control subjects suggests that these compounds may have a role not only in the development of autonomic nervous damage, eg, diabetic autonomic neuropathy, but also in the regulation of autonomic nervous balance in normal subjects. The mechanism of this relationship, however, is obscure, and we were not able to find any experimental data to support our finding.
Elevated plasma proinsulin is associated with dyslipidemia and hypertension in diabetic43 and nondiabetic44 subjects. High proinsulin is considered a marker of impending ß-cell failure, and the relationship of circulating proinsulin to cardiovascular risk factors could be that of an innocent bystander. Furthermore, the possibility that our finding is epiphenomenal cannot be ruled out, since insulin secretion from the ß cells is regulated by the autonomic nervous system,45 and abnormalities in its regulation may become manifest as impaired processing of insulin and high circulating levels of proinsulin. However, this is not in line with our previous finding3 that hyperinsulinemia predicts the development of autonomic neuropathy as assessed by time-domain analysis (ie, expiration-to-inspiration ratio) in patients with NIDDM. Since C-peptide reflects insulin secretion capacity,28 it was not a surprise that a high C-peptide level, like hyperinsulinemia, was related to autonomic nervous function tests in the present study. This study cannot, however, resolve the question of which of these compounds, if any, has a direct role in the development of autonomic nervous dysfunction. Nevertheless, on the basis of the present results, we hypothesize that proinsulin or C-peptide levels are involved in the dysfunction of the autonomic nervous system in NIDDM patients and even in normal elderly subjects. If so, this offers a unique mechanism by which hyperinsulinemia, besides its potential role in atherogenesis, could contribute to excess cardiovascular mortality, since autonomic nervous dysfunction also predicted cardiovascular mortality in our original study population of NIDDM patients.3
In conclusion, autonomic nervous function as assessed by PSA of HRV is decreased in NIDDM patients after 10 years of duration of diabetes compared with control subjects of the same age group. The sympathovagal balance of autonomic nervous function is consistently associated with high proinsulin and high C-peptide levels in both NIDDM patients and control subjects and also with high specific insulin levels in control subjects.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported by grants from the Finnish Foundation of Diabetes Research and the Council for Health Sciences, Academy of Finland, to Dr Uusitupa.
|
Footnotes |
|---|
1 Present address: Research Institute of Military Medicine, Helsinki, Finland.
Received December 23, 1996; revision received March 12, 1997; accepted March 18, 1997.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Ewing DJ, Campbell IW, Clarke BF. The natural history of diabetic autonomic neuropathy. Q J Med. 1980;49:95-108.
2. Sampson MJ, Wilson S, Karagiannis P, Edmonds M, Watkins PJ. Progression of diabetic autonomic neuropathy over a decade in insulin-dependent diabetics. Q J Med. 1990;278:635-646.
3. Töyry JP, Niskanen LK, Mäntysaari MJ, Länsimies EA, Uusitupa MIJ. Occurrence, predictors, and clinical significance of autonomic neuropathy in NIDDM: ten-year follow-up from the diagnosis. Diabetes. 1996;45:308-315.[Abstract]
4. Tack CJJ, Smits P, Willemsen JJ, Lenders JWM, Thien T, Lutterman JA. Effects of insulin on vascular tone and sympathetic nervous system in NIDDM. Diabetes. 1996;45:15-22.[Abstract]
5. Pyörälä K. Relationship of glucose tolerance and plasma insulin to the incidence of coronary heart disease: results from two population studies in Finland. Diabetes Care. 1979;2:131-141.[Abstract]
6. Welborn TA, Wearne K. Coronary heart disease incidence and cardiovascular mortality in Busselton with reference to glucose and insulin concentrations. Diabetes Care. 1979;2:154-160.[Abstract]
7. Ducimetiere P, Eschwege E, Papoz L, Richard JL, Claude JR, Rosselin G. Relationship of plasma insulin levels to the incidence of myocardial infarction and coronary heart disease mortality in a middle-aged population. Diabetologia. 1980;19:205-210.[Medline] [Order article via Infotrieve]
8.
Despres J-P, Lamarche B, Mauriege P, Cantin P,
Dagenais GR, Moorjani S, Lupien P-J.
Hyperinsulinemia as an independent risk factor for
ischemic heart disease. N Engl J
Med. 1996;334:952-957.
9. Temple RC, Clark P, Schneider A, Naki DK, Yudkin JS, Hales CN. Insulin deficiency in non-insulin dependent diabetes. Lancet. 1989;1:293-295.[Medline] [Order article via Infotrieve]
10. Ward WK, LaCava EC, Paquette TL, Beard JC, Wallum BJ, Porte D. Disproportionate elevation of immunoreactive proinsulin in type 2 (non-insulin-dependent) diabetes mellitus and in experimental insulin resistance. Diabetologia. 1987;30:698-702.[Medline] [Order article via Infotrieve]
11. Yoshioka N, Kuzuya T, Matsuda A, Taniguchi M, Iwamato Y. Serum proinsulin levels at fasting and after oral glucose load in patients with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia. 1988;31:355-360.[Medline] [Order article via Infotrieve]
12. Temple RC, Clark P, Schneider A, Nagi DK, Yudkin JS, Hales CN. Radioimmunoassay may overestimate insulin in non-insulin dependent diabetics. Clin Endocrinol. 1990;32:689-693.[Medline] [Order article via Infotrieve]
13.
Saad MF, Kahn SE, Nelson RG, Pettitt DJ, Knowler WC,
Schwartz MW, Kowalyk S, Bennett PM, Porte D. Disproportionately
elevated proinsulin in Pima Indians with non-insulin dependent diabetes
mellitus. J Clin Endocrinol Metab. 1990;70:1247-1253.
14. Porte D. Banting Lecture 1990: ß cells in type II diabetes mellitus. Diabetes. 1991;40:166-180.[Abstract]
15. Davies MJ, Metcalfe J, Gray IP, Day JL, Hales CN. Insulin deficiency rather than hyperinsulinaemia in newly diagnosed type II diabetes mellitus. Diabet Med. 1993;10:305-312.[Medline] [Order article via Infotrieve]
16. Reaven GM, Chen YDI, Hollenbeck CB, Hseuh WH, Ostrega D, Polonsky KS. Plasma insulin, C-peptide, and proinsulin concentrations in obese and non-obese individuals with varying degrees of glucose tolerance. J Clin Endocrinol Metab. 1993;76:44-48.[Abstract]
17. Davies M, Rayman G, Gray IP, Day JL, Hales CN. Insulin deficiency and increased plasma concentrations of intact and 32/33 split proinsulin in subjects with impaired glucose tolerance. Diabet Med. 1993;10:313-320.[Medline] [Order article via Infotrieve]
18. Hales CN. The pathogenesis of NIDDM. Diabetologia. 1994;37(suppl 2):S162-S168.
19. Wahren J, Johansson B-L, Wallberg-Henriksson H. Does C-peptide have a physiological role? Diabetologia. 1994;34(suppl 2):S99-S107.
20. Ewing JE, Martyn CN, Young RJ, Clarke BF. The value of cardiovascular autonomic function tests: 10 years experience in diabetes. Diabetes Care. 1985;5:491-498.
21.
Akselrod S, Gordon D, Ubel FA, Schannon DC, Barger AC,
Cohen RJ. Power spectrum analysis of heart rate
fluctuation: a quantitative probe of beat-to-beat
cardiovascular control. Science. 1981;213:220-223.
22.
Akselrod S, Gordon D, Madwed JB, Snidman NC, Shannon
DC, Cohen RJ. Hemodynamic regulation:
investigation by spectral analysis. Am J
Physiol. 1985;249:H867-H875.
23.
Parati G, Saul JP, Di Rienzo M, Mancia G.
Spectral analysis of blood pressure and heart rate variability
in evaluating cardiovascular regulation: a critical
appraisal. Hypertension. 1995;25:1276-1286.
24. Fei L. Effects of pharmacological interventions on heart rate variability: animal experiments and clinical observations. In: Malik M, Camm AJ, eds. Heart Rate Variability. New York, NY: Futura Publishing Co, Inc; 1995:275-291.
25. Uusitupa MIJ, Niskanen LK, Siitonen O, Voutilainen E, Pyörälä K. Ten-year cardiovascular mortality in relation to risk factors and abnormalities in lipoprotein composition in type 2 (non-insulin-dependent) diabetic and non-diabetic subjects. Diabetologia. 1993;36:1175-1184.[Medline] [Order article via Infotrieve]
26. Uusitupa M, Siitonen O, Pyörälä K, Aro A, Hersio K, Penttilä I, Voutilainen E. The relationship of cardiovascular risk factors to the prevalence of coronary heart disease in newly diagnosed type 2 (non-insulin-dependent) diabetes. Diabetologia. 1985;28:653-659.[Medline] [Order article via Infotrieve]
27. WHO Expert Committee on Diabetes Mellitus. Second Report. Technical Report Series, No. 646. Geneva, Switzerland: World Health Organization; 1980.
28. Niskanen L, Karjalainen J, Siitonen O, Uusitupa M. Metabolic evolution of type 2 diabetes: a 10-year follow-up from the time of diagnosis. J Int Med. 1994;236:263-270.[Medline] [Order article via Infotrieve]
29. Rose GA, Blackburn H. Cardiovascular survey methods. WHO Monograph Series, No. 56. Geneva, Switzerland: World Health Organization; 1968.
30. Bowsher RR, Wolny JD, Frank BH. A rapid and sensitive immunoassay for the measurement of proinsulin in human serum. Diabetes. 1992;41:1084-1090.[Abstract]
31. Haffner SM, Stern MP, Miettinen H, Gingerich R, Bowsher RR. Higher proinsulin and specific insulin are both associated with a parental history of diabetes in non-diabetic Mexican American subjects. Diabetes. 1995;44:1156-1160.[Abstract]
32. Penttilä IM, Voutilainen E, Laitinen O, Juutilainen P. Comparison of different analytical and precipitation methods for the direct estimation of serum high-density lipoprotein cholesterol. Scand J Clin Lab Invest. 1981;41:353-360.[Medline] [Order article via Infotrieve]
33. Tahvanainen K, Länsimies E, Tikkanen P, Hartikainen J, Kärki T, Lyyra T, Mäntysaari M. Microcomputer-based monitoring of cardiovascular functions in simulated microgravity. Adv Space Res. 1992;12:1227-1236.
34. Engelese WAH, Zeelenberg C. A single scan algorithm for QRS-detection and feature extraction. In: Ripley KL, Ostrow HG, eds. IEEE Computers in Cardiology. Long Beach, Calif: IEEE Computer Society; 1979:37-42.
35. Von Neumann J, Kent RH, Bellinson HR, Hart BI. The mean square successive difference. Ann Math Stat. 1941;12:153-162.
36. Fouad FM, Tarazi RC, Ferrario CM, Fighaly S, Alicandri C. Assessment of parasympathetic control of heart rate by a noninvasive method. Am J Physiol. 1984;246:H838-H842.
37.
Billman GE, Dujardin JP. Dynamic changes in
cardiac vagal tone as measured by time series analysis.
Am J Physiol. 1990;258:H896-H902.
38. Hayano J, Sakakibara Y, Yamada A, Yamada M, Mukai S, Fujinami T, Yokoyama K, Watanabe Y, Takata K. Accuracy of assessment of cardiac vagal tone by heart rate variability in normal subjects. Am J Cardiol. 1991;67:199-204.[Medline] [Order article via Infotrieve]
39. Akselrod S. Components of heart rate variability: basic studies. In: Malik M, Camm AJ, eds. Heart Rate Variability. New York, NY: Futura Publishing Co, Inc; 1995:147-163.
40. Hedman A. Short-term Oscillations in Cardiovascular Control System. Kuopio, Finland: University of Kuopio; 1995. Thesis.
41. Kautzner J, Hnatkova K. Correspondence of different methods for heart rate variability measurement. In: Malik M, Camm AJ, eds. Heart Rate Variability. New York, NY: Futura Publishing Co, Inc; 1995:119-126.
42. Kamath MV, Fallen EL. Correction of the heart rate variability signal for ectopic and missing beats. In: Malik M, Camm AJ, eds. Heart Rate Variability. New York, NY: Futura Publishing Co, Inc; 1995:75-85.
43. Nagi DK, Hendra TJ, Ryle AJ, Cooper TM, Temple RC, Clark PM, Schneider AE, Hales CN, Yudkin JS. The relationship of concentrations of insulin, intact proinsulin, and 32-33 split proinsulin with cardiovascular risk factors in type 2 (non-insulin-dependent) diabetic subjects. Diabetologia. 1990;33:532-537.[Medline] [Order article via Infotrieve]
44. Haffner SM, Mykkänen L, Stern MP, Valdez RA, Heisserman JA, Bowsher RR. Relationship of proinsulin and insulin to cardiovascular risk factors in non-diabetic subjects. Diabetes. 1993;42:1297-1302.[Abstract]
45. Macdonald IA. The sympathetic nervous system and its influence on metabolic function. In: Bannister R, Mathias CJ, eds. Autonomic Failure: A Textbook of Clinical Disorders of the Autonomic Nervous System. Oxford, UK: Oxford University Press; 1992:197-211.
This article has been cited by other articles:
 |
|
 |
|
  N. R. J. C. Dunbar Insulin-Mediated Increase in Sympathetic Nerve Activity Is Attenuated by C-Peptide in Diabetic Rats Experimental Biology and Medicine, January 1, 2004; 229(1): 80 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J W Sayer, B Marchant, S V Gelding, J A Cooper, and A D Timmis Autonomic dysfunction is related to impaired pancreatic beta cell function in patients with coronary artery disease Heart, February 1, 2000; 83(2): 210 - 216. [Abstract] [Full Text]
|
|
|
|
 |
|
 S. M. Haffner, R. D'Agostino Jr, L. Mykkanen, C. N. Hales, P. J. Savage, R. N. Bergman, D. O'Leary, M. Rewers, J. Selby, R. Tracy, et al. Proinsulin and Insulin Concentrations in Relation to Carotid Wall Thickness : Insulin Resistance Atherosclerosis Study Stroke, August 1, 1998; 29(8): 1498 - 1503. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||