0021-972X/97/$03.00/0
6
Journal of Clinical Endocrinology and Metabolism
Copyright © 1997 by The Endocrine Society
Alterations in the Glucose-Stimulated Insulin Secretory
Dose-Response Curve and in Insulin Clearance in
Nondiabetic Insulin-Resistant Individuals
CLARE N. O. JONES, DEE PEI, PATRICIA STARIS, KENNETH S. POLONSKY,
Y. D.-IDA CHEN, AND GERALD M. REAVEN
Departments of Medicine, Stanford University School of Medicine, Stanford, California 94305;
University of Chicago, Pritzker School of Medicine, Chicago, Illinois 60637; and Shaman
Pharmaceuticals, Inc., South San Francisco, California 94080
ABSTRACT
Plasma glucose and insulin responses to a graded iv infusion of
glucose were compared in two groups of glucose-tolerant women di-
vided on the basis of their insulin sensitivity. Resistance to insulin-
mediated glucose disposal was measured using the insulin suppres-
sion test, and the women studied were chosen to represent the highest
and lowest quartiles of insulin resistance seen in the normal popu-
lation. The sensitivity of the pancreatic b-cell to glucose was assessed
by measuring the glucose, insulin, and C peptide concentrations in
response to continuous graded iv infusions of glucose at rates of 1, 2,
3, 4, 6, and 8 mg/kgzmin for 40 min each. In addition, insulin secretion
rates in response to the graded glucose infusion, calculated over each
sampling period, were derived from deconvolution of peripheral
plasma C peptide concentrations, using a two-compartment model of
C peptide kinetics and standard parameters for C peptide clearance.
Although plasma glucose concentrations were only slightly higher
throughout the glucose infusion, the insulin concentrations were ap-
P REVIOUS STUDIES have demonstrated a highly signif-
icant relationship in normoglycemic individuals be-
tween resistance to insulin-mediated glucose disposal and
the plasma insulin response to oral glucose (1–3). Although
the plasma glucose concentration is increased in normal in-
dividuals proportionate to the degree of resistance to insulin-
mediated glucose disposal, the plasma insulin response to
oral glucose is increased out of proportion to the coexisting
plasma glucose concentration (1–3). Thus, the pancreatic
b-cell appears to adapt to and compensate for the chronic
state of insulin resistance by secreting more insulin in re-
sponse to a given increment in plasma glucose than would
be the case in an insulin-sensitive individual. In this study we
have used the graded glucose infusion protocol devised by
Polonsky and associates (4, 5) to test the hypothesis that the
increase in the insulin secretory response that is observed in
insulin-resistant subjects is due at least in part to a shift to the
left in the glucose-stimulated insulin secretory dose-response
curve.
Received September 23, 1996. Revision received February 3, 1997.
Accepted February 19, 1997.
Address all correspondence and requests for reprints to: G. M.
Reaven, M.D., Shaman Pharmaceuticals, Inc., 213 East Grand Avenue,
South San Francisco, California 94080-4812.
* This work was supported by NIH Research Grants HL-08506,
DK-31842, and DIC-20595 DRTC.
1834
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 October 2015. at 05:27 For personal use only. No other uses without permission. . All rights reserved.
Vol. 82, No.
Printed in U.S.A.
proximately doubled in the insulin-resistant subjects. When ex-
pressed as a function of the molar increments in plasma glucose
achieved during the glucose infusion studies, the insulin-resistant
women had a 90% higher (684 6 55 vs. 360 6 36 pmol/Lzmmol/L; P ,
0.001) total integrated plasma insulin response as the glucose con-
centration was increased from 5 to 9 mmol/L. However, the total
integrated insulin secretory rate was only increased by 37% (1494 6
133 vs. 1093 6 125 pmol/mmol/Lzmin; P , 0.05) in the insulin-resis-
tant group. This discrepancy suggested that insulin clearance was
lower in the insulin-resistant subjects, and the calculation of this
value, as the ratio of the total secretion of insulin to the area under
the plasma insulin curve, was significantly lower in the insulin-re-
sistant group (1.25 6 0.05 vs. 1.87 6 0.16 L/minzm2; P , 0.005). These
results show that the hyperinsulinemia of insulin resistance results
from an increase in insulin secretion secondary to a shift to the left
of the glucose-stimulated insulin response curve as well as a decrease
in insulin clearance. (J Clin Endocrinol Metab 82: 1834 –1838, 1997)
Subjects and Methods
The experimental population consisted of 20 nonobese, nondiabetic,
healthy women, less than 60 yr of age, recruited from the San Francisco
Bay area by advertisements in local newspapers and selected from a
larger group of volunteers on the basis of their degrees of insulin re-
sistance. They were in good general health on the basis of history,
physical examination, complete blood count, routine biochemical
screening, and electrocardiogram. There was a family history of non-
insulin-dependent diabetes in a first degree relative in 4 of the insulin-
sensitive and 5 of the insulin-resistant women. The insulin-sensitive
group contained 1 woman of Asian origin, 1 Latina, and 9 non-Hispanic
whites, whereas the insulin-resistant group contained 1 Asian, 2 Latinas,
and 6 non-Hispanic whites. None of the women had a history of ges-
tational diabetes or obesity. This project was approved by the Stanford
human subjects committee, and all women gave informed consent.
After an overnight fast, blood was drawn for measurement of plasma
glucose (6) and insulin (7) concentrations before and 30, 60, 120, and 180
min after the ingestion of a 75-g oral glucose challenge. Only volunteer
subjects with a normal glucose tolerance test by the National Diabetes
Data Group criteria were included in the study (8).
Resistance to insulin-mediated glucose disposal was estimated by a
modification (9) of the original insulin suppression test (IST) (10). After
an overnight fast, iv catheters were placed in a superficial antecubital
vein in each arm. One arm was used for a continuous 180-min infusion
of glucose (240 mg/m2zmin), somatostatin (5 mg/min), and insulin (25
mU/m2zmin). Venous blood samples for glucose and insulin determi-
nations were obtained from the contralateral arm every 30 min (to 150
min) and then every 10 min for the last 30 min of the infusion. The mean
of these last 4 values was used to calculate the steady state plasma
glucose (SSPG) and insulin (SSPI) concentrations. Under these experi-
mental conditions, endogenous insulin secretion was suppressed by
somatostatin, and the SSPI concentration achieved was comparable in all
 INSULIN RESISTANCE AND INSULIN SECRETION 1835

individuals. The SSPG provided a measure of insulin-mediated glucose SSPI values were also somewhat higher in the insulin-resis-
disposal: the higher the SSPG, the more insulin resistant the individual. tant individuals (280 6 15 vs. 220 6 13 pmol/L; P , 0.005)
Volunteers with a SSPG concentration above 9 mmol/L were classified
as being insulin resistant, whereas those with values below 5 mmol/L
despite the fact that a similar amount of insulin was infused
were classified as being insulin sensitive. This differentiation corre- in both groups.
sponds to the upper and lower quartiles, respectively, of SSPG values Plasma glucose and insulin concentrations before and after
observed in nondiabetic individuals studied by our research group (3). the oral glucose challenge are shown in Fig. 2. Although all
On the basis of these criteria, 9 insulin-resistant and 11 insulin-sensitive
women were selected for further study. These 2 groups were matched
volunteers had normal glucose tolerance, the total area under
for age (44 6 4 vs. 40 6 2 yr) and body mass index (24.6 6 1.1 vs. 23.3 6 the glucose response curve was 25% greater in the insulin-
0.8 kg/m2). resistant women (21.5 6 0.35 vs. 16.3 6 0.67 mmol/Lzh; P ,
Pancreatic b-cell function was quantified by determining the insulin 0.001). However, the plasma insulin response to oral glucose
and C peptide responses to graded iv infusions of glucose (4, 5). After
was more than twice as high in the insulin-resistant group
an overnight fast, iv catheters were placed in a superficial antecubital
vein in each arm. One arm was used for the infusion of 20% glucose at (1115 6 160 vs. 544 6 60 pmol/Lzh; P , 0 01).
a rate 1 mg/kgzmin, followed by infusions of 2, 3, 4, 6, and 8 mg/kgzmin. The mean plasma glucose, insulin, and C peptide concen-
Each infusion rate was administered for a period of 40 min. Venous trations achieved at each stage of the graded glucose infusion
blood samples for glucose, insulin, and C peptide determinations were shown in Fig. 3 indicate that the insulin-resistant women had
obtained from the contralateral arm at fasting and then 10, 20, 30, and
40 min into each glucose infusion period. a slightly higher plasma glucose (averaging ;0.7 mmol/L
Insulin secretion rates over each sampling period were derived by higher) at any given glucose infusion rate (P 5 0.06, by
deconvolution of peripheral plasma C peptide concentrations, using a two-way ANOVA). Despite the relative similarity in plasma
two-compartment model of C peptide kinetics and standard parameters glucose concentrations, plasma insulin concentrations at the
for C peptide clearance estimated for each subject, taking into account
body surface area and age (11, 12). For each subject, the mean glucose end of each glucose infusion rate were more than twice as
and insulin levels during each stage of the graded glucose infusion were high in the insulin-resistant subjects (P , 0.001, by two-way
used to plot a graph of the insulin-glucose dose-response curve. To ANOVA). The relative increase in the plasma C peptide
compare each subject’s plasma insulin response at the same plasma concentrations was less marked, being approximately 70%
glucose level, the best-fit quadratic curve (least squares fit using Mi-
crosoft Deltagraph, Seattle, WA) was drawn through the data. The
higher in the insulin-resistant women (P , 0.001, by two-way
insulin concentration at molar increments of plasma glucose beginning ANOVA).
at 5 mmol/L can then be obtained by interpolation. The same technique Figure 4 shows plasma insulin concentrations and insulin
applied to the insulin secretion rate yields a plot of the insulin secretion secretion rates at molar increments in plasma glucose. The
rate at molar increments of plasma glucose.
Values for continuous variables are expressed as the mean 6 se. The
insulin-resistant subjects had higher plasma insulin concen-
SSPG and SSPI values were compared using two-tailed unpaired Stu- trations and insulin secretion rates as the plasma glucose
dent’s t tests. The areas under the glucose and insulin curves after the concentration was increased above 5 mmol/L, and in both
oral glucose challenge were calculated using the trapezoidal method, cases, the increases were statistically significant (P , 0.001,
and statistical analysis was performed using two-tailed unpaired Stu-
by ANOVA). The magnitude of these differences can be
dent’s t tests. Differences between the glucose and insulin concentrations
after the oral glucose challenge and the insulin secretory response to iv evaluated by comparing the total integrated responses as the
glucose in the two groups were compared by two-way ANOVA, using plasma glucose concentration was increased from 5 to 9
SAS software (SAS Institute, Cary, NC). mmol/L. In the case of plasma insulin (Fig. 4, left panel), there
was an approximately 90% increase in the total integrated
Results response of the insulin-resistant individuals (684 6 55 vs.
The SSPG and SSPI of the two groups during the IST are 360 6 36 pmol/Lzmmol/L; P , 0.001, by t test), whereas the
shown in Fig. 1. The SSPG values were over 3 times higher increase in the insulin secretion rate was approximately half
in the insulin-resistant women, with no overlap between the as great in these subjects (1494 6 133 vs. 1093 6 125 pmol/
two groups (10.95 6 0.61 vs. 3.42 6 0.14 mmol/L; P , 0.001). mmol/Lzmin; P , 0.05, by t test).

FIG. 1. SSPG (A) and SSPI (B) concen-

trations at the end of the IST in the
insulin-resistant (f) and insulin-sensi-
tive ( ) groups.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 October 2015. at 05:27 For personal use only. No other uses without permission. . All rights reserved.
1836 JONES ET AL. JCE & M • 1997
Vol 82 • No 6

FIG. 2. Plasma glucose (A) and insulin

(B) concentrations before and after a
75-g oral glucose challenge in the insu-
lin-resistant (zzz•zzz) and insulin-sensi-
tive (—•—) groups.

FIG. 3. Plasma glucose (A), insulin (B), and C peptide (C) concentrations at each stage of the graded glucose infusion in the insulin-resistant
(zzz•zzz) and insulin-sensitive (—•—) groups. Note that the x-axis is drawn to be linear with increasing glucose infusion rate rather than with
time.

The observation that the increase in the plasma insulin state MCRex, 0.54 6 0.02 vs. 0.64 6 0.03 L/minzm2; P , 0.02).
concentration was relatively greater than the increase in the Although both methods for calculating insulin clearance in-
insulin secretion rate suggested that the insulin-resistant sub- dicate that the process was significantly reduced in the in-
jects also had a decrease in the rate of insulin clearance, a sulin-resistant subjects, the quantitative relationships were
possibility supported by the fact that they also had higher not the same. This is to be expected because the first calcu-
SSPI concentrations (see Fig. 1). One way to estimate insulin lates the clearance of endogenously produced insulin se-
clearance during the graded glucose infusions is to calculate creted into the portal circulation and thus exposed to exten-
the ratio of the total production of insulin to the area under sive first pass metabolism by the liver before reaching the
the peripheral insulin curve (13). This nonsteady state esti- systemic circulation. The second represents the clearance of
mate of endogenous (end) insulin metabolic clearance rate exogenously infused insulin when endogenous secretion is
(MCR) was significantly lower in the insulin-resistant sub- suppressed by somatostatin, a situation in which the portal
jects (MCRend adjusted for body surface area: resistant vs. and arterial concentrations of insulin are identical, and clear-
sensitive, 1.25 6 0.05 vs. 1.87 6 0.16 L/minzm2; P , 0.005). ance by the liver and other peripheral tissues (largely the
It is also possible to estimate insulin clearance during the kidneys) occurs in parallel. The two are related in the fol-
steady state conditions of the insulin suppression test. In this lowing way: MCRend 5 MCRex/(1 2 Eh), where Eh indicates
instance, the insulin clearance will equal the rate of infusion the hepatic extraction ratio (14). As the hepatic extraction
of insulin divided by the SSPI concentration. This calculation ratio is approximately 60% at physiological insulin concen-
also indicated that exogenous (ex) insulin clearance of the trations (15), we would expect endogenous insulin clearance
insulin-resistant subjects was significantly lower (steady to be approximately 2.5 times the exogenous clearance; our

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 October 2015. at 05:27 For personal use only. No other uses without permission. . All rights reserved.
 INSULIN RESISTANCE AND INSULIN SECRETION 1837

FIG. 4. Plasma insulin concentrations

(A) and insulin secretion rates (B) in
response to molar increments in the
plasma glucose concentration during
the graded glucose infusion in the in-
sulin-resistant (zzz•zzz) and insulin-sen-
sitive (—•—) groups.

results agree fairly well with this. Thus, regardless of the that insulin clearance was decreased in insulin-resistant in-
method employed, there appears to be a defect in the re- dividuals when estimated by either approach is consistent
moval of insulin from plasma in insulin-resistant subjects. with previous demonstrations of a decrease in insulin clear-
ance in two different ethnic groups at high risk for future
Discussion diabetes [Mexican-Americans (19) and African-Americans
The results in Fig. 2 show that the plasma insulin response (20)], in nondiabetic off-spring of patients with noninsulin-
to an oral glucose load is increased relatively more than the dependent diabetes (21), and in obesity (22).
coexisting plasma glucose concentration in insulin-resistant In conclusion, it is apparent from the results presented that
compared to insulin-sensitive subjects. However, the contri- the plasma insulin response to a glucose challenge is a com-
bution of b-cell sensitivity to glucose is difficult to analyze in plex function not only of the coexisting plasma glucose con-
this situation because the glycemic stimulus to the b-cells is centration, but of both the b-cell sensitivity to glucose and the
not matched, and other factors, such as variable gastric emp- rate of insulin clearance, which, in turn, are related to the
tying and unquantified neural and enteric stimuli, will com- degree of resistance to insulin-mediated glucose disposal.
plicate the interpretation. The graded glucose infusion al- Both the enhanced b-cell sensitivity to glucose and the re-
lows comparison of the insulin secretory responses at duced insulin clearance of insulin-resistant subjects may be
matched plasma glucose concentrations and in response to a viewed as an adaptive response, permitting them to maintain
slowly rising plasma glucose level within the physiological normal glucose tolerance in response to a glucose challenge.
range. Using this technique, it is clear that the insulin secre- Although it is tempting to regard this as an autoregulatory
tory response is significantly higher in insulin-resistant sub- mechanism, a cross-sectional study such as this cannot dis-
jects. As such, these data support the view that an adaptive sect which abnormality is primary and which is an adapta-
response occurs in the b-cells of insulin-resistant subjects, tion to it. Finally, two important caveats must be made con-
permitting them to maintain normal glucose tolerance by cerning our results. In the first place, the technique used to
responding in an accentuated manner to a glucose challenge. calculate insulin secretion used a model containing param-
Consistent with the idea of an adaptive response on the part eters for C peptide kinetics that were estimated for each
of the b-cell to variations in insulin-mediated glucose dis- subject rather than measured individually. However, this
posal is the situation in exercise-trained subjects, in whom a approach should give estimates of insulin secretion rates that
decrease in the insulin response to both oral and iv glucose differ by only 10 –12% for each individual and 1–2% for
appears to represent an adaptation to their enhanced insulin group means from those obtained with individual parame-
sensitivity (16 –18). ters, even in a sample heterogeneous in terms of insulin
The results presented also demonstrated a decrease in resistance (12). Secondly, insulin secretion was measured at
insulin clearance in insulin-resistant women when deter- the end of each 40 min of a stepped glucose infusion, and it
mined both under the steady state condition of exogenous is possible that longer infusion periods will lead to different
infusion of insulin during the IST and during the nonsteady estimates of the b-cell secretory response to incremental in-
state endogenous production of insulin in the grade glucose creases in plasma glucose. On the other hand, unless the
infusion. The fact that the estimated decrease in clearance dynamics of insulin secretion and clearance changed dispro-
was greater when based on the endogenous vs. the exoge- portionately in one of the two experimental groups, the qual-
nous approach (33% vs. 15%) may also be due to incomplete itative nature of our findings would not be confounded.
suppression of the b-cell by somatostatin. As we did not Thus, we believe it likely that the hyperinsulinemia of insulin
measure C peptide levels during the IST, we cannot resolve resistance results from an increase in insulin secretion sec-
this issue for certain, although past experience with the same ondary to a shift to the left of the glucose-stimulated insulin
infusion rate of somatostatin suggests that C peptide levels response curve as well as from a decrease in insulin
are indeed satisfactorily suppressed (9). In any event, the fact clearance.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 October 2015. at 05:27 For personal use only. No other uses without permission. . All rights reserved.
1838
References
1. Hollenbeck CB, Chen N, Chen Y-DI, Reaven GM. 1984 Relationship between
the plasma insulin response to oral glucose and insulin-stimulated glucose
utilization in normal subjects. Diabetes. 33:460 – 463.
2. Hollenbeck C, Reaven GM. 1987 Variations in insulin-stimulated glucose
uptake in healthy individuals with normal oral glucose tolerance. J Clin En-
docrinol Metab. 64:1169 –1173.
3. Reaven GM, Brand RJ, Chen Y-D, et al. 1993 Insulin resistance and insulin
secretion are determinants of oral glucose tolerance in normal individuals.
Diabetes. 42:1324 –1332.
4. Byrne MM, Sturis J, Polonsky KS. 1995 Insulin secretion and clearance during
low-dose graded glucose infusion. Am J Physiol. 268:E21–E27.
5. Byrne MM, Sturis J, Clement K, et al. 1994 Insulin secretion abnormalities in
subjects with hyperglycemia due to glucokinase mutations. J Clin Invest.
93:1120 –1130.
6. Kadish AH, Litle RL, Sternberg JC. 1968 A new and rapid method for the
determination of glucose by measurement of rate of oxygen consumption. Clin
Chem. 14:116 –131.
7. Hales CH, Randle PJ. 1963 Immunoassay of insulin and insulin-antibody
precipitate. Biochem J. 88:137–146.
8. National Diabetes Data Group. 1979 Classification and diagnosis of diabetes
mellitus and other categories of glucose intolerance. Diabetes. 28:1039 –1057.
9. Harano Y, Hidaka H, Takatsuki K, et al. 1978 Glucose, insulin and soma-
tostatin infusion for determination of insulin sensitivity in vivo. Metabolism.
27:1449 –1459.
10. Shen S-W, Reaven GM, Farquar J. 1970 Comparison of impedance to insulin-
mediated glucose uptake in normal subjects and in subjects with latent dia-
betes. J Clin Invest. 49:2151–2150.
11. Polonsky KS, Licinio-Paixao J, Given BD, et al. 1986 Use of biosynthetic
human C-peptide in the measurement of insulin secretion rates in normal
volunteers and type I diabetic patients. J Clin Invest. 77:98 –105.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 October 2015. at 05:27 For personal use only. No other uses without permission. . All rights reserved.
JONES ET AL. JCE & M • 1997
Vol 82 • No 6
12. Van Cauter E, Mestrez F, Sturis J, et al. 1992 Estimation of insulin secretion
rates from C-peptide levels– comparison of individual and standard kinetic
parameters for C-peptide clearance. Diabetes. 41:368 –377.
13. Polonsky KS, Given BD, Hirsch L, et al. 1987 Quantitative study of insulin
secretion and clearance in normal and obese subjects. J Clin Invest. 81:435– 441.
14. Pilo A, Navalesi R, Ferrannini E. 1976 Insulin kinetics after portal and pe-
ripheral injection of I-insulin. Data analysis and modeling. Am J Physiol.
230:1626 –1629.
15. Ferrannini E, Wahren J, Faber OK, et al. 1983 Splanchnic and renal metabolism
of insulin in human subjects: a dose-response study. Am J Physiol.
244:E517–E527.
16. Zawalich W, Maturo S, Felig P. 1982 Influence of physical training on insulin
release and glucose utilization by islet cells and liver glucokinase activity in
the rat. Am J Physiol. 6:E464 –E469.
17. Mikines KJ, Sonne B, Tronier B, Galbo H. 1989 Effects of training and
detraining on dose-response relationship between glucose and insulin secre-
tion. Am J Physiol. 256:E588 –E596.
18. Pigon J, Karlsson J, Ostenson C-G. 1994 Physical fitness and insulin sensitivity
in human subjects with a low insulin response to glucose. Clin Sci. 87:187–192.
19. Haffner SM, Stern MP, Watanbe RM, et al. 1992 Relationship of insulin
clearance and secretion to insulin sensitivity in non-diabetic Mexican Amer-
icans. Eur J Clin Invest. 22:147–153.
20. Osei K, Schuster DP. 1994 Ethnic differences in secretion, sensitivity, and
hepatic extraction of insulin in black and white Americans. Diabetes Med.
11:755–762.
21. Cozzolino D, Sessa G, Salvatore T, et al. 1995 Hyperinsulinemia in offspring
of non-insulin-dependent diabetes mellitus patients: the role played by ab-
normal clearance of insulin. Metabolism. 44:1278 –1282.
22. Meistas MT, Margolis S, Kowarski AA. 1983 Hyperinsulinemia of obesity is
due to decreased clearance of insulin. Am J Physiol. 245:E155–E159.