DIABETES TECHNOLOGY & THERAPEUTICS
Volume 3, Number 1, 2001
Mary Ann Liebert, Inc.
Chronic Intermittent Intravenous Insulin Therapy:
A New Frontier in Diabetes Therapy
THOMAS T. AOKI, M.D., EUGEN O. GRECU, M.D., Ph.D.,
MICHAEL A. ARCANGELI, M.S., MAHMOUD M. BENBARKA, M.D.,
PAMELA PRESCOTT, M.D., and JONG HO AHN, M.D.
ABSTRACT
The limited success achieved in controlling diabetes and its complications with conventional
insulin therapy suggests the need for reevaluation of the appropriateness of insulin administration protocols. Indeed, conventional subcutaneous insulin administration produces
slowly changing blood insulin levels and suboptimal hepatocyte insulinization resulting in
impaired hepatic capacity for processing incoming dietary glucose. The novel approach to insulin administration known as chronic intermittent intravenous insulin therapy (CIIIT) delivers insulin in a pulsatile fashion and achieves physiological insulin concentration in the
portal vein. Done as a weekly outpatient procedure combined with daily intensive subcutaneous insulin therapy, this procedure has been shown to (1) significantly improve glycemic
control while decreasing the incidence of hypoglycemic events, (2) improve hypertension control, (3) slow the progression of overt diabetic nephropathy, and (4) reverse some manifestations of diabetic autonomic neuropathy (e.g., abnormal circadian blood pressure pattern, severe postural hypotension, and hypoglycemia unawareness).
INTRODUCTION
A
insulin
deficiency is generally considered as having the pivotal role in the development of diabetes mellitus (DM) and its complications, the
availability of insulin for therapy over the last
eight decades has led to only partial success in
achieving glycemic control and even less in
preventing or limiting the development of the
chronic complications of this disease. Why
have we not done better? As the quality of exogenous insulin has not been an issue since the
advent of human insulins, the appropriateness
of its administration regimens should be carefully reevaluated.
LTHO UGH A RELATIVE OR ABSOLUTE
Normally, insulin is secreted in a pulsatile
fashion1 and in various amounts, in close relationship with meals.2,3 Though not unanimous,
the balance of available experimental evidence
suggests a more potent hypoglycemic effect of
pulsatile insulin as opposed to continuous insulin infusion.4,5 Continuous exposure to insulin and glucagon is known to decrease the
hormones’ metabolic effectiveness on splanchnic glucose production in humans.6 Down-regulation at the cellular level may partially explain the decreased action of steady-state levels
while pulsatile hormone secretion may allow
recovery of receptor affinity or receptor numbers. Intermittent intravenous insulin administration with peaks of insulin concentrations
University of California, Davis, and the Aoki Diabetes Research Institute, Sacramento, California.
111
112
may enhance suppression of gluconeogenesis
and reduce hepatic glucose production (HGP).4
Goodner et al.7 showed in fasting rhesus monkeys that HGP oscillates in synchrony with the
islet cells secretory cycle suggesting that the
greater metabolic effects of pulsatile insulin occur at the liver. Similar results were obtained
in in vitro systems. 5
For induction and maintenance of insulindependent fuel-processing enzyme synthesis
(e.g., hepatic glucokinase, phosphofructokinase, and pyruvate kinase), the hepatocytes require a defined insulin level (200–500 mU/mL
in the portal vein)8,9 concomitant with high glucose levels (“bimolecular signal”).10–12 In nondiabetic subjects, portal insulin concentrations
are 2–3-fold greater than those in the peripheral circulation.8 During the first pass through
the liver, 50% of the insulin is removed,13 pointing to the liver as the principal metabolic target organ of the gastrointestinal tract and the
pancreas. The insulin retained by the hepatocytes may itself be essential for the long-term
effects of insulin on hepatic glucose metabolism as well as growth and de novo enzyme synthesis.14 After oral glucose intake, the liver accounts for an equal or greater proportion of
total net glucose uptake as compared to the periphery.15 Insulin exerts pivotal control of BG
levels through its ability to regulate HGP directly16 or indirectly.17 The traditional subcutaneous (s.c.) insulin administration regimens
used by diabetic patients (a) lack the pulsatile
aspect (s.c. insulin administration achieves
steady or slowly changing plasma concentrations), and (b) do not reach high enough insulin
concentration at the hepatocyte level (e.g., 10 U
regular insulin injected s.c. produce a peak systemic circulation concentration of 30–40
mU/mL and an even lower portal vein concentration of 15–20 mU/mL18 ).
A relative deficiency of insulin at the hepatocyte level leads to impaired capacity for processing of incoming dietary glucose. With the
liver being the target organ of the pancreas, it
appears, therefore, that the primary purpose of
giving insulin to the diabetic patient should not
be to control BG level (“control theory”), but
rather the normalization of hepatic metabolism.
AOKI ET AL.
RATIONALE AND METHODOLOGY
It has been shown that the diabetic patient’s
capacity to oxidize and store exogenous carbohydrate is markedly impaired.19 In the resting
postabsorptive nondiabetic subject, the energy
requirement is met primarily by fat oxidation
reflected by a respiratory quotient (RQ; V
CO2/V O2) of 0.7–0.8 (indirect calorimetry20).
After glucose administration, CO2 production
(and consequently the RQ) increases (0.9–1.0),
indicating that glucose has become the primary
source of energy. In contrast, in the patient with
diabetes mellitus on conventional insulin therapy, no such increase in RQ21,22 or CO2 production23 is observed. The possible fate of ingested glucose is (a) oxidation (liver, brain,
muscle), (b) conversion to fat (liver, muscle,
adipose tissue), (c) storage as glycogen (liver,
muscle) or transamination of intermediary
metabolites to form amino acids (e.g., alanine).
Only the first two processes generate CO2 to increase the RQ. Liver and muscle appear to be
the most active tissues for glucose oxidation. In
1985, Meistas et al.23 showed in nondiabetic
postabsorptive men that resting muscle is not
the source of the increase in CO2 production after ingestion of a 100-g glucose meal. Glucose
utilization by fat is minimal (ca. 2%),24 and adipose tissue production of CO2 after glucose administration has been reported to be insignificant.25 Carbon dioxide production by the
heart,26 kidney,27 brain,28 and gut29 has also
been measured in other mammals, and cannot
account for the large increment in total CO2
production. Thus, both indirect in vivo human
studies23 and direct in vitro experimental data
support the concept that the liver is the source
of the majority of the CO2 production associated with glucose ingestion.
The lack of increase in RQ (and CO2 production) after glucose ingestion in patients
with diabetes on conventional insulin therapy
is consistent with the hypothesis that their liver
is metabolically “dormant” and relatively incapable of oxidizing ingested carbohydrates.
However, this ability to oxidize dietary-derived glucose can be restored to normal after
several days of artificial beta cell-directed insulin administration (Biostator; Fig. 1),22 or
PULSATILE INTRAVENOUS INSULIN THERAPY IN DIABETES
113
FIG. 1. Nonprotein respiratory quotient (npRQ), carbohydrate (CHO), and lipid oxidation rates, and metabolic rates
of normal subjects and diabetic patients on day 0 (conventional insulin therapy) and on day 4 (after 72 h on artificial
B-cell), before (0 time 5 postabsorptive state) and during the 3-h oral 100-g glucose tests. Values shown are X 6 SEM.
(Adapted from Foss et al.22)
with CIIIT within 6–7 h.30 This process was initially called “hepatic activation.” Though not
tried (due to the obvious extreme practical difficulties), it is conceivable that several days of
hyperinsulinemic euglycemic clamp procedure
could also achieve RQ normalization.
This novel approach to insulin administration known as “hepatic activation” or “chronic
intermittent intravenous insulin therapy”
(CIIIT) corrects the above outlined drawbacks
of conventional s.c. insulin therapy as follows:
1. It delivers insulin to the hepatocyte in the
physiologically appropriate portal vein concentrations range (over 200 mU/mL).
2. Insulin is administered in pulses intravenously (i.v.) similar to the physiological
insulin secretion pattern.
In essence, CIIIT uses a novel insulin algorithm consisting of a series of intravenous insulin pulses administered concomitantly with
oral glucose. Pulses of insulin should have
greater potential for hepatic activation than a
continuous infusion of insulin since it is possible to manipulate the rate at which insulin concentration changes, the magnitude of each
pulse, the duration of hepatic insulin exposure,
and to superimpose the pulses on a rising baseline. Any or all of these may be metabolically
important signals The insulin is injected by a
programmed Bionica MD-100 infusion pump
into a forearm vein, and each insulin pulse
achieves a peak venous “free” insulin of at least
200 mU/mL (Fig. 2). The pulses are given during the first hour of a three hour treatment with
three consecutive treatments given each treatment day. Initial CIIIT consists of two consecutive treatment days, followed by weekly treatment days. RQ measurements, using a
Sensormedic Metabolic Measurement Cart
(Sensormedics, Anaheim, CA), are obtained before and throughout the treatment period at 60min intervals. An increase in the RQ to greater
114
AOKI ET AL.
FIG. 2. Peripheral venous free insulin levels following intravenous insulin delivery by Bionica pump (35 mU/kg
body weight, q6 min pulses 3 10).
than 0.90 is used as the index of therapeutic efficacy.21–23 It was postulated that if hepatic activation was achieved and maintained in patients with “brittle” IDDM by this treatment,
the glycohemoglobin Alc (HbAlc) blood levels
and the frequency of hypoglycemic reactions
should decrease.
CIIIT for an average of 41 months. The major
results of the study indicated:
(a) A significant decline in HbAlc from the
baseline of 8.5% to 7.0% at the end of the observation period (p , 0.0003; Fig. 3)
CLINICAL EFFECTS OF CIIIT IN
DIABETES
CIIIT effect on glycemic control
In a study published in 1993,30 the results of
long-term CIIIT on 20 IDDM patients with
“brittle” disease (wide swings in fingerstick
blood glucose concentrations and frequent hypoglycemic episodes despite being on intensive
subcutaneous insulin therapy [ISIT]—four
daily insulin injection regimen—for at least 1
year) were reported. The patients received
FIG. 3. Changes in HbAlc levels and incidence of major
and minor hypoglycemic events with chronic intermittent
intravenous insulin therapy. (Adapted from Aoki et al.30 )
PULSATILE INTRAVENOUS INSULIN THERAPY IN DIABETES
(b) A decline in the frequency of major hypoglycemic events from 3.0 to 0.1/month (p ,
0.0001)
(c) A decline in the frequency of minor hypoglycemic events from 13.0 to 2.4/month (p ,
0.0001)
The last two effects are contrary to the results
of Diabetes Control and Complication Trial
(DCCT) where the improvement in HbAlc was
accompanied by a threefold increase in the frequency of major hypoglycemic events.31 In
many of the study patients, a gradual improvement in awareness to impending hypoglycemia was noted which became apparent after 3–4 months of therapy and led to a marked
decrease in the frequency of major hypoglycemic events. Furthermore, the patients reported increased energy level and required
shorter daily periods of rest leading to increased productivity and quality of life. There
was no significant change in the patients’ body
weight during the study.
The exact mechanism by which CIIIT lowered HbAlc blood levels and decreased the frequency of hypoglycemic events is yet to be determined. Nondiabetic individuals stimulate
hepatic processes, by way of high portal-vein
concentrations of insulin and glucose, with
every meal. The study patients’ livers were
stimulated three times during 1 day of the week
rather than three times daily as in nondiabetic
individuals. The reason for the effectiveness of
CIIIT may lie in the long half-life of glucokinase. The half-life of rat glucokinase is 3–4
days11 and that of the human enzyme may be
longer because of the lower rate of human metabolism.30
CIIIT effect on hypertension
Systemic hypertension is a frequent complication found in both IDDM and NIDDM patients and has proved to be an important risk
factor for the development of microangiopathy32 and macroangiopathy.33 In IDDM, the
prevalence of hypertension increases with the
duration of the disease and develops mainly in
connection with the clinical emergence of
nephropathy.34 There appears to be a positive
115
correlation between plasma glucose elevation
and systolic blood pressure (BP) raising the
possibility that the metabolic disorders associated with diabetes are acting to either cause or
amplify the concurrent BP problem.35 Significant reductions in systolic and diastolic BP
after improvement in metabolic control in diabetic subjects have been reported.36 Conversely, in IDDM subjects studied during controlled insulin withdrawal,37 the worsening of
metabolic control was accompanied by BP increases. These reports suggest that tight metabolic control in diabetic subjects may provide
beneficial effects on BP levels.38 We have recently assessed the effects of CIIIT on BP control in a prospective, randomized, crossover
clinical trial involving 26 IDDM subjects with
hypertension and nephropathy.39 After a stabilization period (ISIT for glycemic control and
normalization of BP on medication), the study
subjects were randomly assigned to control
(ISIT) or treatment (ISIT 1 CIIIT) groups for 3
months and then crossed over into the opposite phase for another 3 months. Addition of
weekly CIIIT during the treatment phase was
the only procedural difference between the
control and treatment phases. Throughout the
control and treatment phases, the BP values
were maintained at the level established in each
subject at the end of the stabilization period
through appropriate adjustment in the antihypertensive medication (AHM) dosage. The
AHM dosage requirements decreased significantly (46%; p , 0.0001) and linearly over time
(p , 0.0058) during the treatment phase while
remaining essentially unchanged during the
control phase (Fig. 4). This study indicates that
CIIIT markedly improves BP control in subjects
with IDDM and hypertension.
Increased vascular smooth muscle (VSM)
tone is the hallmark of the hypertensive state
in both IDDM and NIDDM.40 Human and animal studies suggest a role for insulin in the
regulation of VSM tone.41–43 Such insulin regulation of VSM function may be lost in insulinopenic states. Recent evidence suggests
that insulin causes endothelium-derived nitric
oxide-dependent vasodilation and that insulin’s vasodilating action serves to both amplify insulin’s overall effect to stimulate skele-
116
AOKI ET AL.
FIG. 4. Changes in antihypertensive medication requirements of patient groups A and B during the postrandomization period. (Adapted from Aoki et al.39)
tal muscle glucose uptake and modulate vascular tone.44 It is possible that CIIIT partially
normalizes the “vascular reactivity,” thus lowering the AHM dosage requirements.
CIIIT effect on diabetic nephropathy
Diabetic nephropathy (DN) develops in
35–40% of patients with IDDM,45 and it is the
most common cause of end-stage renal disease
(ESRD) in the United States. 46 It is generally
agreed that DN is the result of hyperglycemia,
whether alone or in combination with other factors.47,48 However, once nephropathy is clinically
overt (macroalbuminuria, decreased glomerular
filtration rate), the degree of metabolic control is
believed to have lost its significance as a risk
factor with other mechanisms having greater
influence.47 Until recently, there was no known
therapeutic strategy able to stop or reverse the
progression to ESDR.45 Suggested conventional
ways to slow the progression of DN are effective antihypertensive therapy49 and reduced
protein intake.50
Several studies have indicated no effect of
tight glycemic control on the progression of
overt nephropathy.51,52 The effect of CIIIT on
the progression of DN was evaluated in a multicenter, retrospective, longitudinal study, involving 31 patients with type 1 diabetes melli-
tus and overt DN, on intensive subcutaneous
insulin therapy (ISIT) and weekly CIIIT.53 All
studied patients were followed weekly for at
least 12 months (average 37 6 4.6 months), and
appropriate adjustments were made weekly in
their insulin dosage (ISIT) and in their antihypertensive medication with the goal of maintaining optimum glycemic control and blood
pressures at or below 140/90 mm Hg for each
patient. All patients had monthly HbAlc
(HPLC) and semiannual creatinine clearance
(CrCl) determinations. The HbAlc levels declined from 8.64 6 0.57% to 7.60 6 0.3% (p 5
0.0062) during the study period. The CrCl remained essentially unchanged (from 46.1 6
2.19 mL/min at baseline, to 46.0 6 3.9 mL/min
at the end of the observation period, with an
average annualized slope increase of 3.39 6 1.5
mL/min/year (p 5 NS). This study suggests
that addition of CIIIT to ISIT in patients with
type 1 DM appears to arrest or markedly reduce the progression of overt diabetic
nephropathy. Similarly favorable results with
CIIIT added to ISIT were reported in a larger
multicenter, prospective, controlled clinical
trial in patients with type 1 DM and overt DN.54
The mechanism by which CIIIT reduces the
deterioration rate of renal function in patients
with overt DN remains to be established. In a
recent study, McLennan et al.48 have shown
PULSATILE INTRAVENOUS INSULIN THERAPY IN DIABETES
that a high glucose concentration inhibits
mesangium degradation and could promote
the mesangium enlargement known to occur in
DN. Enlargement of the mesangium, the most
consistent morphological finding in DN, can
compress the glomerular capillaries and thus
alter intraglomerular hemodynamics.55 Improved glycemic control, as obtained in the
CIIIT treated subjects, could promote mesangium
degradation and improve renal hemodynamics.
Indeed, we already have evidence that CIIIT improves systemic hemodynamics,39 and it is likely
that it has a similar effect on renal hemodynamics (e.g., decrease in glomerular efferent arteriolar vasoconstriction) resulting in the noted
decrease in the progression of DN towards
ESRD. Several animal experiments have demonstrated that glomerular expression of transforming growth factor–b, the key mediator
between hyperglycemia and mesangial cell
stimulation toward overproduction of extracellular matrix, is stimulated by hyperglycemia
and/or hypoinsulinemia.56 Both hyperglycemia
and hypoinsulinemia are improved through
the CIIIT procedure.
CIIIT effect on diabetic autonomic neuropathy
Though clinical symptoms of diabetic autonomic neuropathy (DAN) develop in relatively
few patients, the measurable autonomic defects
are extremely common in diabetes.57 DNA is
known to be associated with a higher morbid-
117
ity and mortality rate,58 as well with elevated
HbAlc levels.59
CIIIT effect on abnormal circadian blood pressure pattern. Several studies have demonstrated a
blunted diurnal variation in blood pressure (BP)
in patients with DAN.60,61 Insufficient BP decline
during the night might be associated with increased target organ damage.62,63 No data are
available regarding the effects of tight glycemic
control or intensive insulin therapy on the abnormal circadian BP pattern in patients with
IDDM. We have recently reported the results of
a randomized controlled clinical study involving
74 IDDM patients on intensive subcutaneous insulin therapy (ISIT)64 of which 36 (group A) underwent, in addition, weekly CIIIT for 3 months.
Group B (controls, n 5 38) continued on ISIT
alone. All study patients were seen weekly by
the investigators and underwent monthly HbAlc
determinations and monthly 24-h ambulatory
BP monitoring. Glycemic control improved significantly in group A subjects (HbAlc declined
from 7.9% to 7.2%, p 5 0.0002), but remained essentially unchanged in group B (control). The
night/day systolic BP ratio decreased from 0.97
to 0.94 (23.10%) in group A and increased from
0.95 to 0.98 (13.16%) in group B subjects (p 5
0.0224), while the night/day diastolic BP ratio
decreased from 0.93 to 0.90 (23.23%) in group A
and increased from 0.91 to 0.94 (13.29%) in
group B subjects (p 5 0.0037; Fig. 5). Consider-
FIG. 5. Changes in mean night/day blood pressure ratio with CIIIT in groups A (treated) and B (control) patients.
(Adapted from Aoki et al.64)
118
ing that all other aspects of therapy were similar for the two patient groups, the further improvement in the glycemic control and the
significant improvement in the abnormal circadian BP pattern noted in group A patients
was likely due to the addition of weekly CIIIT.
The cause of the abnormal circadian BP pattern
is secondary to or associated with abnormal
glucose metabolism and reduced arterial wall
distensibility65,66 as well as the development
of diabetic autonomic neuropathy.67 The improvement/stabilization of this abnormal circadian pattern obtained in the group A patients
might be the result of an improved metabolic
milieu as suggested by the decline in HbAlc,
with possible consequent improvement in arterial distensibility and diabetic autonomic
neuropathy. The practical importance and clinical consequences of the improvement in the
circadian BP pattern is conjectural at present.
Insufficient BP decline during the night might
be associated with increased target organ damage.62,63 The reversal or at least prevention of further deterioration of the abnormal circadian BP
pattern obtained in the patient group treated
with ISIT 1 CIIIT might lessen target organ
damage. Further studies with extended followup are necessary to confirm this possibility.
CIIIT effect on orthostatic hypotension of diabetes.
Orthostatic hypotension (OH) of diabetes, another likely manifestation of DAN,68 is defined
as a decrease in diastolic BP greater than 10 mm
Hg69 or decline of systolic BP greater than 30
mm Hg after 2 min of standing,70 in the presence of adequate blood volume. Diabetic patients with OH have decreased ability to release
norepinephrine, leading to low plasma norepinephrine levels and supersensitive a and b receptors. 71 Diabetic OH is likely due to impaired
sympathetic vasoconstrictor activity, leading to
an impaired compensatory increase in total peripheral resistance upon standing.72 The current conventional therapy for orthostatic hypotension of diabetes is less than satisfactory.73
Even the promising newer agents such as the
somatostatin analogues, have a very limited
duration of action and require frequent injections.74
We have recently reported 75 the correction of
postural hypotension with CIIIT in three pa-
AOKI ET AL.
tients with IDDM and severe disabling postural
hypotension who had previously failed all conventional therapeutic attempts. The patients
underwent weekly CIIIT for 3 months while
continuing their usual regimen of four daily
subcutaneous insulin injections. Before and at
the end of this therapeutic trial the patients had
tilt-table tests, 24-h ambulatory BP measurements, cardiac autonomic function testing, and
HbAlc determinations. Within the first month
of CIIIT, a marked decrease in the intensity and
frequency of postural dizziness was noted in
all subjects, as well as the disappearance of syncopal episodes. After 2 months of CIIIT, the
postural dizziness ceased entirely, and at the
end of the third month of therapy, the subjects
were able to resume their customary activities.
The tilt-table test normalized in two and improved in one patient, HbAlc levels declined,
the initial abnormal circadian BP pattern was
corrected, and the score of the cardiovascular
reflex tests also improved. These results suggest an improvement in the vasoconstrictor
mechanisms in response to postural changes,
possibly as a result of the improvement in diabetic autonomic neuropathy, as indicated by
the normalization of the circadian blood pressure pattern and by the improved cardiovascular reflex score. The improvement in the autonomic neuropathy was likely secondary to
the improved metabolic milieu during CIIIT as
reflected by decreased HbAlc levels. The vascular endothelial cell secretion of the potent
vasoconstrictor endothelin (ET1), has been
shown to be enhanced by insulin in cultured
cells.76 An enhancement of endothelial cells’
ET1 production following exposure to the high
pulsatile insulin levels of CIIIT could have contributed to the improvement of the vasoconstrictor mechanism in our patients.
CONCLUSIONS
The studies briefly reviewed above indicate
that CIIIT improves glycemic control, concomitantly with marked decrease in the frequency of
both major and minor hypoglycemic events and
improved hypoglycemia awareness,30 improves
control of hypertension in diabetes,39 retards
progression of overt nephropathy,53,54 reverses
PULSATILE INTRAVENOUS INSULIN THERAPY IN DIABETES
the abnormal circadian BP pattern,64 and corrects
postural hypotension of diabetes.75
Anecdotal clinical experiences suggest that
CIIIT also improves diabetic peripheral polyneuropathy, diabetic cardiomyopathy,77 diabetic retinopathy, and diabetic foot ulcer healing. Confirmation of these latter observations
requires future larger, prospective clinical
studies.
We believe that the above clinical outcomes
with CIIT are due to weekly pulsatile insulin
administration. As mentioned under the Rationale and Methodology sections, RQ normalization can also be achieved after several days
of artificial beta cell–directed insulin administration (Biostator)22 or, conceivably, after several days of a hyperinsulinemic euglycemic
clamp procedure; however it is clearly impractical to perform weekly clamp procedures for
extended period of time (years) and on a large
scale. Furthermore, though we recognize that
patients’ enthusiasm and enhanced compliance
might be a component of any research program’s good results, the published results with
CIIIT on improving glycemic control, decreased frequency of hypoglycemic events, and
on chronic complications of diabetes (e.g., overt
nephropathy), have been superior to intensive
s.c. insulin therapy (multiple insulin injections
or external insulin infusion pumps31 ).
FIG. 6. Venous free insulin concentrations in 6 type 1
diabetic patients following 5, 25, and 50 milliunits of insulin per kg body weight pulses using the artificial pancreas (GCIIS).
119
FIG. 7. Rapid increases in respiratory quotient (RQ) in
14 type 1 diabetic patients following the administration
of insulin pulses using the Bionica MD-110 infusion pump
during the first day of CIIIT.
HISTORICAL NOTE
Although the use of pulsatile insulin infusions in the treatment of diabetic patients, as
opposed to square wave insulin infusions, had
occurred to us in the 1970s, the need for such
a waveform was not immediately apparent.
During the course of our initial studies with the
artificial pancreas (Biostator, Glucose Controlled Insulin Infusion System, Life Science Division, Ames, Elkhart, IN), which used a square
wave infusion pattern, it quickly became apparent that a more efficient and powerful algorithm was needed. The GCIIS required 72 h
to “activate” a type 1 diabetic patient and
needed a great deal of “hands on” time. Could
this period of enzyme induction be shortened?
The use of insulin pulses to stimulate hepatic
metabolism was revisited. The liver appeared
to respond to (1) the rate of change of insulin
levels, (2) the peak insulin concentration, (3) a
rising baseline (a pseudo second phase insulin
release), and (4) the prompt return to baseline
FIG. 8. Rapid increases in respiratory quotient (RQ) in
six type 2 diabetic patients following the administration
of insulin pulses using the Bionica MD-110 infusion pump
during the first day of CIIIT.
120
insulin levels, that is, wide dynamic range. It
was therefore decided to see if insulin pulses
could more rapidly initiate the synthesis and
activation of hepatic enzymes (e.g., hepatic glucokinase and pyruvate dehydrogenase complex respectively) with the biochemical outcome reflected by an increase in the respiratory
quotient.
The artificial pancreas was first reversed engineered so that all of its functions could be controlled from the keyboard of an independent but
linked Digital Equipment Company computer.
Thus, insulin could be pulsed independent of the
glucose levels, and glucose levels could be continuously monitored. In addition, the amounts of
insulin and glucose infused could be quantified
on a minute to minute basis.
The first question to be answered was
whether the roller pumps used in the GCIIS
could generate insulin pulses. Using this instrument, 5, 25, and 50 milliunits of insulin per
kilogram body weight were successively infused into type 1 diabetic patients (n 5 6)
whose blood glucose levels were maintained at
elevated levels with intermittent oral glucose
administration. Arterial blood samples were
withdrawn at 2-min intervals. As seen in Figure 6, even pulses of 5 milliunits of insulin/kg
body weight could be identified and measured.
The next question to be answered was
whether the insulin pulses generated by the reverse engineered GCIIS were physiologically
and, hence, therapeutically more effective than
square waves infusions of insulin. Using the
respiratory quotient as the endpoint, studies
were performed to determine if insulin pulses
could accelerate the “activation” process in
type 1 diabetic individuals, as reflected by
rapid increases in the baseline respiratory quotient from 0.75 to values equal to or greater than
0.90. Fortunately, the answer was yes, and we
quickly switched to simpler infusion pumps.
Initially, McGaw AccuPro pumps interfaced
to an Epson laptop computer equipped with a
floppy disc drive were used, but the BIONICA
MD-110 pump subsequently replaced these.
Using the latter pump, intravenous insulin
pulses (Fig. 2) were administered to both type
1 and 2 diabetic patients concomitant with oral
glucose ingestion, and the respiratory quotient
responses monitored. In Figure 7, the increases
AOKI ET AL.
in RQ during the first day of pulsatile intravenous insulin therapy in 14 type 1 diabetic patients new to the procedure are shown. In contrast to the modest RQ responses reported in
type 1 diabetic patients after 24 h of GCIIS insulin administration,22 marked increases in RQ
to values consistently greater than 0.90 are seen
within hours of initiating pulsatile intravenous
insulin therapy. In Figure 8, prompt and
marked increases in RQ during the first day of
CIIIT in six type 2 diabetic patients are shown.
The implications of these observations were
enormous. This procedure could be done efficiently, quickly, and relatively inexpensively
on an outpatient basis for both patient care and
research purposes. For these reasons, intravenous insulin pulses were used exclusively in
all our studies, reviewed in this manuscript,
from 1983 to the present.
ACKNOWLEDGMENTS
We wish to thank Michael Bradley, Sheila
Hargrave, and Elsie Soeldner for expert technical assistance.
REFERENCES
1. Lang DA, Matthews DR, Peto J, Turner RC: Cyclic oscillations of basal plasma glucose and insulin concentrations in human beings. N Engl J Med 1979;301:
1023–1027.
2. Polonsky KS: The B-cell in diabetes: from molecular
genetics to clinical research. Diabetes 1995;44:705–717.
3. Polonsky KS, Givens BD, Canter EV: Twenty-fourhour profiles and pulsatile patterns of insulin secretion in normal and obese subjects. J Clin Invest 1988
;81:442–448.
4. Bratush-Marrain PR, Komjati M, Waldhausl WK: Efficacy of pulsatile versus continuous insulin administration on hepatic glucose production and glucose
utilization in type 1 diabetic humans. Diabetes 1986;
35:922–926.
5. Komjati M, Bratush-Marrain PR, Waldhausl W: Superior efficacy of pulsatile versus continuous hormone exposure on hepatic glucose production in vitro.
Endocrinology 1986;118:312–319.
6. Bomboy JD, Lewis SB, Lacy WW, Sinclair-Smith CC,
Liljenquist JE: Transient stimulatory effect of sustained hyperglucagonemia on splanchnic glucose
production in normal and diabetic man. Diabetes
1977;26:177–181.
PULSATILE INTRAVENOUS INSULIN THERAPY IN DIABETES
7. Goodner CJ, Cherrington AD: Hepatic glucose production oscillates in synchrony with the islet secretary cycle in fasting rhesus monkeys. Science 1982;
215:1257–1259.
8. Blackard WG, Nelson NC: Portal and peripheral vein
immunoreactive insulin concentrations before and after glucose infusion. Diabetes 1970;19:302–306.
9. Berger W, Goschke H, Moppert J, Kunzli H: Insulin
concentrations in portal venous and peripheral venous blood in man following administration of glucose, galactose, xylitol and tolbutamide. Horm Metab
Res 1972;5:4–8.
10. Cahill GF Jr, Ashmore J, Renold AE, Hastings AB: Blood
glucose and the liver. Am J Med 1959;26:264–282.
11. Weinhouse S: Regulation of glucokinase in liver. In:
Horecker BL, Stadtman ER, eds. Current Topics in
Cell Research. New York: Academic Press, 1976:1–50.
12. Hers HG, Van Schaftingen E: Fructose 2.6-bisphosphate 2 years after its discovery. Biochem J 1982;206:
2–12.
13. Waldhausl M, Bratush-Marrain P, Gasic S, Korn A,
Nowothy P: Insulin production rate, hepatic insulin
retention, and splanchnic carbohydrate metabolism
after oral glucose ingestion in hyperinsulinemic type
II (non–insulin dependent) diabetes mellitus. Diabetologia 1982;23:6–15.
14. Griffen SC, Russell SM, Katz LS, Nicoll CS: Insulin exerts metabolic and growth-promoting effects by a direct action on the liver in vivo: clarification of the functional significance of the portal vascular link between
the beta-cells of the pancreatic islets and the liver.
Proc Natl Acad Sci USA 1987;84:7300–7304.
15. Felig P, Bergman M: Integrated physiology of carbohydrate metabolism. In: Ellenberg M, Rifkin H, eds.
Diabetes Mellitus: Theory and Practice, 4th ed. New
York: Elsevier 1990:51–59.
16. Hers HG: The control of glycogen metabolism in the
liver. Annu Rev Biochem 1976;45:167–189.
17. Ader M, Bergman R: Peripheral effects of insulin
dominate suppression of fasting hepatic glucose production. Am J Physiol 1990;258:E10220–E1032.
18. Ishida T, Chap Z, Chou J, Lewis RM, Hartley CJ, Entman ML, Field JB: Effect of portal and peripheral venous insulin infusion on glucose production and utilization in depancreatized, conscious dogs. Diabetes
1984;33:984–990.
19. Meyer HU, Curchod B, Maeder E, Pahud P, Jequier
E, Felber JP: Modifications of glucose storage and oxidation in nonobese diabetics, measured by continuous indirect calorimetry. Diabetes 1980;29:752–756.
20. Ferrannini E: The theoretical bases of indirect
calorimetry: a review. Metabolism 1988;37:287–301.
21. Aoki TT, Vlachokosta FV, Foss MC, Meistas MT: Evidence for restoration of hepatic glucose processing in
type I diabetes mellitus. J Clin Invest 1983;71:837–839.
22. Foss MC, Vlachokosta FV, Cunningham IN, Aoki TT:
Restoration of glucose homeostasis in insulin-dependent diabetic subjects. Diabetes 1982;31:46–52.
23. Meistas MT, Vlachokosta FV, Gleason RE, Arcangeli
MA, Aoki TT: Role of muscle in CO2 production af-
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
121
ter oral glucose administration in man. Diabetes
1985;34:960–963.
Bjorntrop P, Berchtold P, Holm J, Larsson B: The glucose uptake inhuman adipose tissue in obesity. Eur J
Clin Invest 1971;1:480–485.
Bjorntorp P, Schlersten T, Gottfries A: Effects of glucose infusion on adipose tissue lipogenesis in man.
Acta Med Scand 1968;183:565–571.
Cruickshank EWH, Startup CW: The effect of insulin
on the respiratory quotient, oxygen consumption,
sugar utilization and glycogen synthesis in the normal mammalian heart in hyper- and hypoglycemia. J
Physiol 1933;77:365–385.
Gregg CM, Cohen JJ, Black AJ, Espeland MA, Feldstein ML: Effects of glucose and insulin on metabolism and function of perfused rat kidney. Am J Physiol 1978;235:F52–F61.
Rowe GC, Maxwell GM, Castillo CA, Freeman DJ,
Crumpton CW: A study in man of cerebral blood flow
and cerebral glucose, lactate and pyruvate metabolism
before and after eating. J Clin Invest 1959;38:2154–2158.
Abumrad NN, Cherrington A, Williams P, Lacy W,
Robin D: Absorption and disposition of a glucose load
in the conscious dog. Am J Physiol 1982;242:E398–
E406.
Aoki TT, Benbarka MM, Okimura MC, Arcangeli MA,
Walter RM Jr, Wilson LD, Truong MP, Barber AR, Kumagai LF: Long-term intermittent intravenous insulin
therapy and type 1 diabetes mellitus. Lancet 1993;342:
515–517.
DCCT Research Group. The effect of intensive treatment of diabetes on the development and progression
of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–986.
Janka HU, Warram JH, Rand LI, Krolewski AS: Risk
factors for progression of background retinopathy in
longstanding IDDM. Diabetes 1989;38:460–464.
Fuller JH, Head J: WHO Multinational Study Group:
Blood pressure, proteinuria and their relationship
with circulatory mortality: the WHO multinational
study of vascular disease in diabetics. Diabetes Metab
1989;15:273–277.
Parving HH, Smidt UM, Frisberg B, Bonnevie-Nielsen
V, Andersen AR: A prospective study of glomerular
filtration rate and arterial blood pressure in insulindependent diabetes with diabetic nephropathy. Diabetologia 1981;20:457–461.
Tuck M: Management of hypertension in the patient
with diabetes mellitus. Am J Hypertens 1988;1:384S–
388S.
Ferris JB, O’Hare JA, Kelleher CCM: Diabetic control
and the renin-angiotensin system, catecholamines
and blood pressure. Hypertension 1985;7:II58–II63.
Gunderson HJG: Peripheral blood flow and metabolic
control in juvenile diabetes. Diabetologia 1974;10:225–
231.
Stern M, Tuck ML: Mechanisms of Hypertension in
diabetes mellitus. In: Laragh JH, Brenner BM, eds. Hypertension Pathophysiology, Diagnosis and Management. Vol. 2. New York: Raven Press, 1990:1689–1702.
122
39. Aoki TT, Grecu EO, Prendergast JJ, Arcangeli MA,
Meisenheimer R: Effect of chronic intermittent intravenous insulin therapy on antihypertensive medication requirements in IDDM subjects with hypertension and nephropathy. Diabetes Care 1995;18:1260–
1265.
40. Levy J, Zemel MB, Sower JR: Role of cellular calcium
metabolism in abnormal glucose metabolism and diabetic hypertension. Am J Med 1989;87:7S–16S.
41. Yagi S, Takata S, Kiyokawa H, Yamamoto M, Noto Y,
Ikeda T, Hattori N: Effects of insulin on vasoconstrictive responses to norepinephrine and angiotensin
II in rabbit femoral artery and vein. Diabetes 1988;
37:1064–1067.
42. Hall JE, Brands MW, Mizelle HL, Gaillard CA, Hildebrandt DA: Chronic intrarenal hyperinsulinemia does
not cause hypertension. Am J Physiol 1991;260:F663–
F669.
43. Anderson EA, Hoffman RP, Balon TW, Sinkey CA,
Mark AL: Hyperinsulinemia produces sympathetic
neuronal activation and vasodilatation in normal humans. J Clin Invest 1991;87:2246–2252.
44. Baron AD: Insulin and the vasculature—old actors,
new roles. J Invest Med 1996;44:406–412.
45. Viberti GC, Yip-Messent J, Morcutti A: Diabetic
nephropathy. Diabetes Care 1992;15:1216–1225.
46. U.S. Renal Data System: USRDS 1989 Annual Data
Report. Bethesda, MD, National Institute of Health,
National Institute of Diabetes and Digestive and Kidney Diseases, August 15 and 16, 1989.
47. Hostetter TH: Diabetic nephropathy. N Engl J Med
1985;312:642–643.
48. McLennan SV, Fisher EJ, Yue D, Turtle JR: High glucose concentration causes a decrease in mesangium
degradation. Diabetes 1994;43:1041–1045.
49. Parving HH, Andersen AR, Smidt UM, Hommel E,
Mathiesen ER, Svendsen PA: Effect of antihypertensive treatment on kidney function in diabetic nephropathy. Br Med J (Clin Res Ed) 1987;294:1443–
1447.
50. Zeller K, Whittaker E, Sullivan L, Raskin P, Jacobson
HR: Effect of restricting dietary protein on the progression of renal failure in patients with insulin-dependent diabetes mellitus. N Engl J Med 1991;324:
78–84.
51. Viberti GC, Bilous RW, Mackintosh D, Bending JJ,
Keen H: Long term correction of hyperglycemia and
progression of renal failure in insulin-dependent diabetes. BMJ 1983;268:598–602.
52. Taft JL, Nolan CJ, Yeung SP, Hewitson TD, Martin
FIR: Clinical and histological correlations of decline
in renal function in diabetic patients with proteinuria.
Diabetes 1994;43:1046–1051.
53. Aoki TT, Grecu EO, Gollapudi GM, Barber AR, Arcangeli MA, Benbarka MM, Prescott P, Meisenheimer
R: Effect of intensive insulin therapy on progression
of overt diabetic nephropathy in patients with IDDM.
Endocrine Practice 1999;5:174–178.
54. Dailey GE, Boden GH, Creech RH, Johnson DG, Glea-
AOKI ET AL.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
son RE, Kennedy FP, Weinrauch LA, Weir M, D’Elia
JA: Effects of pulsatile intravenous insulin therapy
(PIVIT) on the progression of diabetic nephropathy.
Metabolism 2000;49:1491–1495.
Steffes MW, Osterby R, Chavers B, Mauer SM: Mesangial expansion as a central mechanism for loss of kidney function in diabetic patients. Diabetes 1989;38:
1077–1081.
Sharma K, Ziyadeh FM: Hyperglycemia and diabetic
kidney disease: the case for transforming growth factor b as a key mediator. Diabetes 1995;44:1139–1146.
O’Brian IAD, O’Hare JP, Lewin IG, Corral RJM: The
prevalence of autonomic neuropathy in insulin-dependent diabetes mellitus: a controlled study based
on heart rate variability. Q J Med 1986;61:957–967.
Ewing DJ, Campbell IM, Clarke BF: The natural history of diabetic autonomic neuropathy. Q J Med
1980;49:95–108.
DCCT Research group: Factors in the development of
diabetic neuropathy: baseline analysis of neuropathy
in the feasibility phase of Diabetes Control and Complications Trial. Diabetes 1988;37:476–481.
Liniger C, Favre L, Assal J-P: Twenty-four hour blood
pressure and heart rate profiles of diabetic patients
with abnormal cardiovascular reflexes. Diabetic Med
199;8:420–427.
Ikeda T, Matsabaro T, Sato J, Sakamoto M: Circadian
blood pressure variation in diabetic patients with autonomic neuropathy. J Hypertens 1993;11:581–587.
Verdecchia P, Schillaci G, Guerrieri M: Circadian
blood pressure changes and left ventricular hypertrophy in hypertension. Circulation 1990;1:528–536.
O’Brien E, Sheridan J, O’Malley K: Dippers and nondippers. Lancet 1988;II:397.
Aoki TT, Grecu EO, Arcangeli MA, Meisenheimer R:
Effect of intensive insulin therapy on abnormal circadian blood pressure pattern in patients with type 1
diabetes mellitus. Online J Curr Clin Trials [serial online], December 15, 1995, doc. no. 199.
Amar J, Chamotin B, Pelissier M, Garelli J, Salvador M:
Influence of glucose metabolism on nictimeral blood
pressure variability in hypertensives with an elevated
waist hip ratio. Am J Hypertens 1995;8:426–428.
Lehman EO, Gosling RG, Sonksen PH: Arterial wall
compliance in diabetes. Diabet Med 1992;9:114–119.
Ikeda T, Matsabaro T, Sato Y, Sakamoto M: circadian
blood pressure variation in diabetic patients with autonomic neuropathy. J Hypertens 1993;11:581–587.
Dyrberg T, Benn J, Sandahl Christiansen J, Hilsted J,
Nerup J: Prevalence of diabetic autonomic neuropathy measured by simple bed-site tests. Diabetologia
1981;20:190–194.
Shumer M, Burton G, Burton C, Crum D, Pfeiffer MA:
Diabetic autonomic neuropathy. Am J Med 1988;85:
137–146.
Vinik AI, Holland MT, Le Beau JM, Luizzi FJ: Diabetic neuropathies. Diabetes Care 1992;15:1926–1975.
Ziegler MG, Lake CR: Noradrenergic responses in
postural hypotension: implications for therapy. In:
PULSATILE INTRAVENOUS INSULIN THERAPY IN DIABETES
72.
73.
74.
75.
76.
Lake CR, Zigler MG, eds. The catecholamines in psychiatric and neurologic disorders. Woburn, MA: Butterworth, 1985:121–136.
Hilstead J: Pathophysiology in diabetic autonomic
neuropathy: cardiovascular, hormonal, and metabolic
studies. Diabetes 1982;31:730–737.
Green DA, Sima AAF, Feldman EL, Stevens MJ: Diabetic neuropathy screening and management. Contemp Intern Med 1993;5:79–91.
Hoeldtke R, Israel CB: Treatment of orthostatic hypotension with octreotide. J Clin Endocrinol Metab
1989;68:1051–1059.
Aoki TT, Grecu EO, Arcangeli MA: Chronic intermittent intravenous insulin therapy corrects orthostatic
hypotension of diabetes. Am J Med 1995;99:683–684.
Rubin SA, Levin ER: The endocrinology of vasoactive
peptides: synthesis to function. J Clin Endocrinol
Metab 1994;78:6–10.
123
77. Aoki TT, Grecu EO, Arcangeli MA: Reversal of severe
nonischemic dilated cardiomyopathy by intensive intravenous insulin therapy in a patient with NIDDM
[Abstract]. Diabetes 1996;45:1071.
Address reprint requests to:
Thomas T. Aoki, M.D.
Division of Endocrinology
University of California, Davis
4150 V Street
PSSB Suite G400
Sacramento, CA 95817
E-mail: [email protected]
N A L
A R T I C L E
S
Effect of Chronic
Intermittent Intravenous
Insulin Therapy on
Antihypertensive
Medication Requirements
in IDDM Subjects With
Hypertension and
Nephropathy
THOMAS T. AOKI, MD
EUGEN O . GRECU, MD, PHD
MICHAEL A. ARCANGELI, MS
RONALD MEISENHEIMER, MS
JOSEPH J. PRENDERGAST, MD
OBJECTIVE — The prevalence of systemic hypertension is increased in patients
with diabetes. In this prospective, randomized, crossover clinical trial, we assessed the
antihypertensive effects of chronic intermittent intravenous insulin therapy (CHIT) on
insulin-dependent diabetes mellitus (IDDM) subjects with hypertension and nephropathy by monitoring the amount of antihypertensive medication (AHM) required to
maintain the blood pressure (BP) :< 140/90 mmHg.
RESEARCH DESIGN A N D M E T H O D S — After a stabilization period, 26
hypertensive IDDM subjects were randomly assigned to a control or treatment phase
for 3 months and then crossed over into the opposite phase for another 3 months.
Addition of CHIT during the treatment phase was the only procedural difference
between the control and treatment phases.
RESULTS — The AHM dosage requirements for maintenance of the baseline BP
levels decreased significantly (46%; P < 0.0001) and linearly over time (P < 0.0058)
during the treatment phase, while remaining essentially unchanged during the control
phase.
CONCLUSIONS — Our data suggest that CHIT markedly improves BP control, as
evidenced by the significantly reduced AHM dosage requirements in subjects with
IDDM and hypertension, possibly through an improvement in vascular reactivity.
From the University of California (T.T.A.), Davis; Aoki Diabetes Research Institute (E.O.G., M.A.A.), Sacramento; and Pacific Medical Research Services Q.J.P.), Atherton, California.
Address correspondence and reprint requests to Thomas T. Aoki, MD, Division of Endocrinology, University of California, Davis, 4301 X St., FOLB II C, Sacramento, CA 95817.
Thomas T. Aoki is a paid consultant of and holds stock in Advanced Metabolic Systems, Inc. Ronald
Meisenheimer is a paid consultant and Michael A. Arcangeli is an employee of Advanced Metabolic Systems,
Inc.
Received for publication 15 November 1994 and accepted in revised form 30 May 1995.
ACE, angiotensin-converting enzyme; AHM, antihypertensive medication; BP, blood pressure; CHIT,
chronic intermittent intravenous insulin therapy; HPLC, high-performance liquid chromatography; IDDM,
insulin-dependent diabetes mellitus; ISIT, intensive subcutaneous insulin therapy; NIDDM, insulindependent diabetes mellitus; VSM, vascular smooth muscle.
1260
ystemic hypertension is a frequent
complication found in both insulindependent diabetes mellitus (IDDM)
and non-insulin-dependent diabetes mellitus (NIDDM) subjects and has proved to
be an important risk factor for the development of microangiopathy (1,2) and
macroangiopathy (3). The prevalence of
hypertension in diabetic subjects is 1.5-2
times greater than in a matched nondiabetic population (4). In IDDM, the prevalence of hypertension increases with the
duration of the disease (5) and develops
mainly in connection with the clinical
emergence of nephropathy (6). There
seems to be a positive correlation between
plasma glucose elevation and systolic
blood pressure, raising the possibility that
the metabolic disorders associated with
diabetes are acting to either cause or amplify the concurrent blood pressure problem (7). Significant reductions in systolic
and diastolic blood pressure after improvement in metabolic control in diabetic subjects have been reported (8,9).
Similarly, in streptozotocin-induced hypertensive diabetic rats, insulin treatment
ameliorated the increased arterial blood
pressure (BP) (10). Conversely, in IDDM
subjects studied during controlled insulin
withdrawal (11), the worsening of metabolic control was accompanied by blood
pressure increases. These reports suggest
that tight metabolic control in diabetic
subjects may provide beneficial effects on
blood pressure levels (12).
We have recently reported the effect of a new therapeutic approach—
chronic intermittent intravenous insulin
therapy (CHIT)—on glycemic control in
IDDM subjects (13). This new therapy
seemed to have a positive effect on blood
pressure levels in hypertensive IDDM
subjects, and results of a pilot study in
this regard have already been reported in
abstract form (14). Encouraged by these
promising preliminary data, we designed
a prospective, randomized, crossover
clinical trial intended to assess the effects
of CHIT on hypertension control as expressed by the antihypertensive medica-
DIABETES CARE, VOLUME 18,
NUMBER 9,
SEPTEMBER
1995
Aoki and Associates
Table 1—ListofAHMs
Class
1
ACE inhibitors
2
3
4
Calcium channel blockers
Loop diuretics
a2-agonists
tion (AHM) dosage requirements in hypertensive IDDM subjects.
RESEARCH DESIGN AND
METHODS
Selection of subjects
Subjects were selected and randomized
for study if they had: 1) a classical history
of IDDM (i.e., early age of onset, episodes
of diabetic ketoacidosis, brittle glycemic
control); 2) a history of hypertension for
at least 1 year and were taking conventional antihypertensive medication; 3) no
clinical or laboratory evidence of secondary hypertension to any other cause except for diabetic renal disease; or 4) creatinine clearance >15 ml""1 • min" 1 •
1.73 m~ 2 . Subjects were excluded from
the study if they had: 1) a history of noncompliance with drug therapy; 2) severe
underlying chronic disease (e.g., decompensated congestive heart failure, unstable angina pectoris, or a history of myocardial infarction within 6 months of
entering the protocol); 3) a history of hypersensitivity reaction to angiotensinconverting enzyme (ACE) inhibitor peptide-like drugs, calcium channel
blockers, loop diuretics, or clonidine; or
4) severe hypertension that could not be
controlled with a drug combination from
the selected AHM classes (see below) during the 1- to 3-month stabilization period. The clinical trial was approved by
the Institutional Review Boards of the
University of California, Davis; the Aoki
Diabetes Research Institute, Sacramento;
or Sequoia Hospital, Redwood City. All
subjects gave written informed consent
before enrollment into the study.
DIABETES CARE, VOLUME 18,
NUMBER 9,
SEPTEMBER
Specific drug used
Incremental
dosage (mg)
Equivalent
AHMU
Maximum dose
used (mg)
Enalapril
Lisinopril
Calan SR
Lasix
Clonidine
5
5
120
20
Catapres-TTS-1
1
1
1
1
1
20
20
240
120
Catapres-TTS-3
Of 41 subjects enrolled in the
study, 32 completed the prerandomization stabilization period (see below). During the postrandomization period, six
subjects were excluded because of noncompliance with the protocol (four noncompliant with the medication; two
moved out of the area). We report 26 subjects who met all the inclusion criteria
and completed the study.
Prerandomization stabilization
period
Because of the potential for biological
variability in the parameters to be assessed (e.g., HbAlc, systemic BP), all subjects selected for the study underwent a 1to 3-month period of stabilization, during
which optimization of glycemic control
was obtained by uniformly instituting a
regimen of four daily insulin injections
(intensive subcutaneous insulin therapy
[ISIT]), dietary instruction, and frequent
home blood glucose monitoring (AccuChek II meter, Boehringer Mannheim, Indianapolis, IN). Concomitantly, the BP
values were brought under control (at or
just below 140/90 mmHg), by administering low to moderate doses of one AHM
or a combination of AHMs from the
classes listed in Table 1. The AHMs were
introduced in listed order (Table 1) as
needed in each case, unless, on occasion,
side effects to a drug required an alteration in the sequence or dosage. The decision to add a second, third, or fourth
drug from the established list was made
when the maximum dose (Table 1) of the
drug used (e.g., ACE inhibitor) failed to
control BP levels. We considered the lowest likely effective dose of each drug as a
1995
unit (AHM U) and used it in incremental
doses as needed (e.g., 10 mg enalapril = 2
AHM U). The number of AHM units used
expressed each patient's AHM requirements throughout the study. This experimental design was used to decrease the
number of variables that had to be considered. All patients were seen weekly by
the investigators, and appropriate insulin
and AHM dosage adjustments were
made.
Study parameters and methods
Because of the potential influence of dietary sodium intake, body weight
changes, and level of physical activity on
the control of hypertension, all subjects
were initially instructed and reminded
weekly to follow a mildly restricted sodium diet (6-8 g NaCl per day) and to
maintain a stable body weight and level of
physical activity throughout the study period. Diabetic control was assessed by
monthly monitoring of HbA lc , determined by high-performance liquid chromatography (HPLC) performed on a Varian Vista 5500 HPLC system equipped
with a Waters SP-5PW column (15) (nondiabetic HbA lc range 4.0-6.0%). Systemic BP was assessed by monthly 24-h
ambulatory BP monitoring (every 20 min
between 0600 and 2200 and every 30
min between 2200 and 0600; model
90207 [SpaceLabs, Redmond, WAI). The
BP was also measured weekly with the
patient in the sitting position, determined
by the average of three readings taken at
10-min intervals, using the same ambulatory BP monitoring device. Serum and
24-h urine samples were collected for determinations of serum electrolytes, creat-
1261
New therapy for hypertensive IDDM patients
Table 2—Baseline subject characteristics
Physical characteristics
n
Age (years)
Sex (F/M)
Race (Caucasian/Hispanic)
Weight (kg)
Diabetes
IDDM
Duration (years)
HbA lc (end stabilization)
Evidence of nephropathy
Hypertension
History of hypertension and therapy
Duration (years)
Medication units (average)
24-h blood pressure (average baseline)
Group A
Group B
13
43 (26-64)
7/6
11/2
78(46-109)
13
43 (25-67)
11/2
13/—
69 (52-97)
13
23 (13-41)
7.6 ± 0 . 1
13
13
28 (17-54)
7.1 ± 0.08
13
13
3(1-8)
9 ±0.5
132 ± 1/78 ± 0.6
13
2.7 (1-10)
8.5 ± 0.3
128 ± 0.9/75 ± 0.6
Data are n, median (range), or means ± SE.
inine and blood urea nitrogen, creatinine
clearance, and urinary protein excretion.
These tests were performed at a commercial laboratory (Nichols Institute).
Postrandomization protocol
The baseline characteristics of the 26 subjects who completed the study are given
in Table 2. After randomization, all subjects continued on ISIT and their antihypertensive therapy and were seen weekly
by the investigators. Appropriate adjustments were made in their ISIT and AHM
dosages, with the goal of maintaining optimum glycemic control and a stable BP
level for each individual patient. In addition, for the first 3 months, the subjects
randomly assigned to group A underwent
weekly CHIT (treatment phase). This
novel therapeutic approach (13) consists
of 7-10 pulses of intravenous insulin,
along with oral glucose administration,
during the 1st hour of a 3-h treatment;
three treatments are given in a day. During these intravenous insulin pulses, systemic free insulin levels of over 200
/xIU/ml are achieved (unpublished data
from previous experiments). At the end of
this 3-month period, the CHIT was discontinued, and the group A subjects were
1262
crossed over into the control phase for
another 3-month period. The subjects
randomly assigned to group B first entered the control phase for 3 months, after
which they were crossed over into the
treatment phase for another 3-month period (Fig. 1). The only procedural difference between the treatment and the control phases was the addition of weekly
CHIT during the treatment phase.
Throughout the postrandomization period, the BP values were to be
maintained relatively stable at the indi-
vidual baseline level for each subject
through appropriate adjustments in the
AHM dosage. The baseline BP level was
assessed by a 24-h ambulatory BP monitoring procedure performed at the end of
the prerandomization period (mean 24-h
BP). The number of AHM units required
by each subject for controlling his or her
BP levels at the end of the stabilization
period (baseline BP) was considered to be
the 100% AHM requirement for that subject. If a patient whose BP was controlled
on X AHM U at baseline demonstrated a
decline in BP during the treatment phase,
the medication dosage was reduced in the
reversed order of its introduction. First to
be decreased was the class 4 AHM, then
the class 3 AHM, 2 AHM, and last, the
ACE inhibitor until the BP value reverted
to baseline level. The degree of decline in
the AHM requirement was expressed as
percentage decline. Thus, each patient's
AHM dosage requirements during the
postrandomization period were compared to his or her baseline AHM requirements and expressed as a percentage
change.
Statistical analysis
All results are given as means ± SE. Data
were analyzed using BMDP2V, repeated
measures of variance analysis. Differences
were considered significant when P <
0.05.
Post-Randomization
Pre-Randomization
: Treatment Phase
Group B: Treatment Phase
(Stabilization)
Group A: Control Phase
Group B: Control Phase
4
1
" 5™
*
6
(Months)
Figure 1—Protocol: prerandomization (stabilization) period; postrandomization period: patient
groups A and B cross over after 3 months.
DIABETES CARE, VOLUME 18,
NUMBER 9,
SEPTEMBER
1995
Aoki and Associates
discontinued in 5 (19%), partially decreased in 16 (62%), and remained unchanged in 5 (19%) subjects (Table 3).
Therefore, in 21 (81%) study subjects,
the AHM dosage had to be either discontinued or decreased to maintain baseline
BP values during the treatment phase. No
subject required an increase in the AHM
dosage during the treatment phase.
Throughout the postrandomization period, the BP values were maintained relatively stable at individual baseline levels for each subject (Table 4).
No significant differences in gly3
cemic
control,
as expressed by the HbAlc
Month
measurements, were noted between the
I-*-GROUP A • G R O U P B
end of the stabilization period (baseline)
Figure 2—AHM requirements ojpatient groups A and B during the postrandomization period. and completion of the control and treatment phases (7.5 ± 0.06%, 7.5 ± 0.04%,
and 7.4 ± 0.04%, respectively). The total
daily insulin requirements, patients' body
RESULTS
became evident by the end of the 1st weight, hematocrit, and blood urea nitromonth, reaching a nadir by the 3rd month gen, as well as creatinine clearance and
Prerandomization stability and
of this phase. The average AHM dosage urinary protein excretion, were not signifbasal values
decline was on the order of 45.39 ± 11% icantly different at the ends of the stabiliBy the time randomization was per- (P < 0.0005) (Fig. 2).
zation period, control phase, and treatformed, all subjects had stable BP values
The combined average decrease in ment phase for either subject group A or
(average 130 ± 0.4/76 ± 0.3 mmHg) and AHM dosage requirements during the B.
optimized glycemic control for at least 1 treatment phase of both subject groups
month (average HbAlc = 7.4 ± 0.05%). was 46.4% (P < 0.0001) and linear over CONCLUSIONS — Our prospective,
time (P < 0.0058). The degree of decline randomized, crossover clinical trial was
Postrandomization results
in AHM dosage requirements was not designed to evaluate the effect of CHIT on
In group A subjects, the AHM dosage re- uniform in all subjects: it was completely the AHM dosage requirements in hyperquirements for maintenance of the baseline BP levels decreased gradually, reaching a nadir during the 3rd month of the
Table 3—Characteristics of the two subject subgroups segregated based on changes in
treatment phase. The decrease, averaging
their AHM requirements during CHIT
47.3 ± 11%, was highly significant (P <
0.0072). After the discontinuation of
AHM requirements during CHIT
CHIT, a gradual increase in the BP levels,
necessitating a gradual increase in AHM
Decreased or
dosage for maintenance of the baseline BP
Abolished
Characteristics
unchanged
levels, was noted. By the 3rd month of the
5
21
control phase, the AHM dose require- n
Age
(years)
40
±
8
43
± 2
ments had returned close to the baseline
Duration of IDDM (years)
27 ± 7
25 ± 2
level (Fig. 2).
Duration of hypertension (years)
1.8 ± 0.4
3.3 ± 0.4
The group B subjects required no AHM U (average baseline)
4 ± 1
10 ± 1
significant changes in AHM dosages for Diabetic nephropathy
maintenance of the baseline BP levels durAlbuminuria (mg/24 h)
54 ± 7
946 ± 92
ing the control phase. After the crossover
Proteinuria (mg/24 h)
217 ± 16
1401 ± 205
71.4 ± 5
into the treatment phase, a gradual deCrCl (ml/min)
52.5 ± 8
cline in the AHM dosage requirements Data are means ± SE.
DIABETES CARE, VOLUME 18, NUMBER 9, SEPTEMBER 1995
1263
New therapy for hypertensive IDDM patients
Table 4—Average 24-h blood pressure levels
Patient
group
Blood pressure
(mmHg)
A
A
B
B
Systolic
Diastolic
Systolic
Diastolic
Baseline
1 month
2 months
3 months
4 months
5 months
6 months
132
80
135
80
133 ± 3
78 ± 2
125 ± 3
74 ± 2
129
77
130
75
140
82
125
71
140
81
126
73
133
79
135
80
136
78
131
78
±4
±2
±4
±4
±3
±2
±4
±3
±3
±2
±4
±2
±4
±3
±3
±3
±3
±2
±5
±5
±4
±2
±4
±3
Data are means ± SE. Group A patients were in the treatment phase during months 1-3 and in the control phase during months 4-6. Group B patients were in the
control phase during months 1-3 and in the treatment phase during months 4-6.
tensive IDDM subjects. We initially considered using the change in mean arterial
pressure to express the CHIT effect on hypertension control. However, we felt it
would be unethical to potentially subject
our diabetic patients with nephropathy to
uncontrolled hypertension for several
months. Therefore, we elected to bring
the study patients' BP under control with
AHM during the stabilization period and
then express the CHIT effect on BP control
through the required changes in AHM
dosages.
The combined average decrease in
AHM dosage requirements during the
treatment phase of both subject groups
was 46.4% (P < 0.0001) and linear over
time (P < 0.0058). The degree of decline
in AHM dosage requirements was not
uniform in all subjects (Table 3), suggesting an inverse relationship between the
lowering of BP levels through CHIT and
the duration of hypertension, baseline
AHM dosage requirements for BP control,
or the severity of diabetic nephropathy.
Analysis of other potential variables, which might have caused this improvement in our subjects' hypertension
control during the treatment phase, suggested that CHIT itself is responsible for
the changes observed. Throughout the
postrandomization period, there were no
significant differences between the degree
of glycemic control (HbA lc ), body
weight, daily amount of insulin administered, patients' volume status (as suggested by hematocrit and blood urea nitrogen levels), dietary caloric and sodium
intake, or level of physical activity, in ei-
1264
ther subject group A or B. Similarly, the
potential effect of the weekly physiciansubject interaction on the results was
minimized by continuing them during
both treatment and control phases for
each subject. The only consistent difference in the protocol between, the treatment and control phases was the CHIT
procedure itself, performed only during
the treatment phase.
Theoretically, a placebo (saline
solution pulses) CHIT trial period could
have further strengthened the interpretation of our results. However, the use of
saline instead of insulin pulses, while administering large amounts of oral glucose
for several hours, would have lead to unacceptably severe hyperglycemia in our
IDDM subjects. For this reason, the placebo-controlled trial was deemed unethical and was not used.
The exact mechanism by which
CHIT reduces the AHA dosage requirements in hypertensive IDDM subjects remains to be determined. Increased vascular smooth muscle (VSM) tone is the
hallmark of the hypertensive state in both
IDDM and NIDDM (16,17). Human and
animal studies suggest a role for insulin in
the regulation of VSM tone (18-24). Such
insulin regulation of VSM function may
be lost in insulinopenic states. In view of
the positive effects of CHIT on hypertension control in IDDM subjects, it is possible that this therapy partially normalizes
vascular reactivity, thus lowering the
AHM dosage requirements.
We conclude that CHIT improves
BP control as reflected by the significantly
reduced AHM requirements in subjects
with IDDM and hypertension, possibly
through improvement in vascular reactivity. This effect represents a significant additional benefit of CHIT. Furthermore,
due to its reproducibility, it could be used
as a probe for studying biochemical and
biological mechanisms of hypertension in
diabetic subjects in vivo.
Acknowledgments—This study was supported in part by Advanced Metabolic Systems
Corporation and by Boehringer Mannheim
Corporation.
We thank Michael Bradley, Sheila Hargrave,
Muna Beeai, Kristi Jacobs, and Sheila
Franchesci for expert technical assistance.
References
1. Ritz E, Hasslacher C, Tschope W, Koch
M, Mann JFE: Hypertension in diabetes
mellitus. Contrib Nephrol 54:77-85,1987
2. Janka HU, Warram JH, Rand LI,
Krolewski AS: Risk factors for progression
of background retinopathy in longstanding IDDM. Diabetes 38:460-464, 1989
3. Fuller JH, Head J, WHO Multinational
Study Group: Blood pressure, proteinuria
and their relationship with circulatory
mortality: the WHO multinational study
of vascular disease in diabetics. Diabetes
Metab 15:273-277, 1989
4. Caro JF: Diabetes and hypertension. Diabetes Care 16:540-541,1993
5. Peiris AN, Gustafson AB: Current therapeutic concepts in diabetic hypertension.
Diabetes Care 9:409-414,1986
6. Parving HH, Smidt UM, Frisberg B, Bon-
DIABETES CARE, VOLUME 18, NUMBER 9, SEPTEMBER
1995
Aoki and Associates
7.
8.
9.
10.
11.
12.
nevie-Nielsen V, Andersen AR: A prospective study of glomerular nitration rate
and arterial blood pressure in insulindependent diabetes with diabetic nephropathy. Diabetologia 20:457-^461,1981
Tuck M: Management of hypertension in
the patient with diabetes mellitus. Am J
Hypertens L384S-388S, 1988
O'Hare JA, Ferris JB, Twomey BM, Gonggrijp H, O'Sullivan DJ: Changes in blood
pressure, body fluids, circulating angiotensin II and aldosterone with improved
diabetic control. Clin Sri 63:415S-418S,
1982
Ferris JB, O'Hare JA, Kelleher CCM: Diabetic control and the renin-angiotensin
system, catecholamines and blood pressure. Hypertension 7 (Suppl. 2):II58-II63,
1985
Sasaki S, Bunag RD: Insulin reverses hypertension and hypothalamic depression
in streptozotocin diabetic rats. Hypertension 5:24-40, 1983
Gunderson HJG: Peripheral blood flow
and metabolic control in juvenile diabetes. Diabetologia 10:225-231, 1974
Stern N, Tuck ML: Mechanisms of hypertension in diabetes mellitus. In Hypertension Paihophysiology, Diagnosis and Man-
agement. Vol. 2. Laragh JH, Brenner BM,
Eds. New York, Raven, 1990, p. 16891702.
DIABETES CARE, VOLUME 18, NUMBER 9, SEPTEMBER
13. Aoki TT, Benbarka MM, Okimura MC,
Arcangeli MA, Walter RM, Wilson LD,
Truong MP, Barber AR, Kumagai LF:
Long-term intermittent intravenous insulin therapy and type I diabetes mellitus.
lancet 342:515-518, 1993
14. Truong MP, Arcangeli MA, Benbarka
MM, Aoki TT, Grecu EO: The impact of
chronic intermittent intravenous insulin
therapy on blood pressure in patients
with insulin dependent diabetes mellitus
and hypertension (Abstract). Clin Res 41:
65A,1993
15. Ellis G, Diamandis EP, Giesbrecht EE,
Daneman D, Allen LC: An automated
"high-pressure" liquid-chromatographic
assay for hemoglobin A lc . Clin Chem 11:
1746-1752, 1984
16. LevyJ, Zemel MB, Sower JR: Role of cellular calcium metabolism in abnormal
glucose metabolism and diabetic hypertension. AmJMed 87 (Suppl. 6A):7S-16S,
1989
17. Weidman P, Beretta-Picolli C, Keusch G,
Gluck Z, Mujagic M, Grimm M, Meier A,
Ziegler WH: Sodium-volume factor, cardiovascular reactivity and hypotensive
mechanism of diuretic therapy in mild
hypertension associated with diabetes
mellitus. Am] Med 67:779-784, 1979
18. Reddy S, Shehin S, Sowers JR, Dardas G,
Zemel M: 45Ca flux and blood pressure
1995
19.
20.
21.
22.
23.
24.
regulation in streptozotocin-induced diabetic rats. J Vase Med Bid 2:47-51, 1990
Yagi S, Takata S, Kiyokawa H, Yamamoto
M, Noto Y, Ikeda T, Hattori N: Effects of
insulin on vasoconstrictive responses to
norepinephrine and angiotensin II in rabbit femoral artery and vein. Diabetes 37:
1064-1067, 1988
Zemel MB, Reddy S, Shehin SE, Lockette
W, Sowers R: Vascular reactivity in
Zucker obese rats: role of insulin resistance. J Vase Med Bid 2:81-85, 1990
HallJE, Brands MW, Kivlighn SD, Mizelle
HL, Hildebrandt DA, Gaillard CA:
Chronic hyperinsulinemia and blood
pressure regulation. Am J Physiol 258:
F722-F731,1990
HallJE, Brands MW, Mizelle HL, Gaillard
CA, Hildebrandt DA: Chronic intrarenal
hyperinsulinemia does not cause hypertension. Am] Physiol 260:F663-F669,1991
Anderson EA, Hoffman RP, Balon TW,
Sinkey CA, Mark AL: Hyperinsulinemia
produces sympathetic neuronal activation and vasodilatation in normal humans. J Clin Invest 87:2246-2252, 1991
Laakso M, Edelman SV, Brechtel G, Ba-on
AD: Decreased effect of insulin to stimulate skeletal muscle blood flow in obese
man: a novel mechanism of insulin resistance.; Clin Invest 85:1844-1852, 1990
1265
ARTICLE IN PRESS
Molecular Aspects of Medicine ■■ (2015) ■■–■■
Contents lists available at ScienceDirect
Molecular Aspects of Medicine
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / m a m
Review
Pulsatile insulin secretion, impaired glucose tolerance and
type 2 diabetes
Leslie S. Satin a,*, Peter C. Butler b, Joon Ha c, Arthur S. Sherman c
a
Department of Pharmacology, Brehm Diabetes Research Center, University of Michigan Medical School, Ann Arbor, MI 48105, USA
Department of Medicine, David Geffen School of Medicine, UCLA, Los Angeles, CA 90095-7073, USA
c Laboratory of Biological Modeling, National Institutes of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda,
MD 20892, USA
b
A R T I C L E
I N F O
Article history:
Received 11 September 2014
Received in revised form 9 January 2015
Accepted 10 January 2015
Available online
Keywords:
Islets
Oscillations
Insulin pulsatility
Insulin resistance
Diabetes
A B S T R A C T
Type 2 diabetes (T2DM) results when increases in beta cell function and/or mass cannot
compensate for rising insulin resistance. Numerous studies have documented the longitudinal changes in metabolism that occur during the development of glucose intolerance
and lead to T2DM. However, the role of changes in insulin secretion, both amount and temporal pattern, has been understudied. Most of the insulin secreted from pancreatic beta
cells of the pancreas is released in a pulsatile pattern, which is disrupted in T2DM. Here
we review the evidence that changes in beta cell pulsatility occur during the progression
from glucose intolerance to T2DM in humans, and contribute significantly to the etiology
of the disease. We review the evidence that insulin pulsatility improves the efficacy of secreted insulin on its targets, particularly hepatic glucose production, but also examine
evidence that pulsatility alters or is altered by changes in peripheral glucose uptake. Finally,
we summarize our current understanding of the biophysical mechanisms responsible for
oscillatory insulin secretion. Understanding how insulin pulsatility contributes to normal
glucose homeostasis and is altered in metabolic disease states may help improve the treatment of T2DM.
© 2015 Elsevier Ltd. All rights reserved.
Contents
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Introduction .............................................................................................................................................................................................................................. 2
Insulin levels oscillate in fasted humans ........................................................................................................................................................................ 2
Individuals with T2DM have impaired insulin pulsatility ........................................................................................................................................ 2
Abnormal insulin pulsatility and increased insulin resistance ............................................................................................................................... 3
Difficulties measuring insulin pulses under fasting conditions in the peripheral circulation, especially using older methods ...... 4
Pulsatile insulin regulates hepatic function more efficiently .................................................................................................................................. 4
The relationship between pulsatile insulin and insulin resistance ....................................................................................................................... 5
Physiological glucose modifies the amplitude but generally not the frequency of insulin pulse bursts ................................................. 6
Ultradian oscillations and their entrainment by small glucose fluctuations are disrupted in T2DM ....................................................... 8
Pulsatility of other islet hormones ................................................................................................................................................................................... 8
Mechanisms governing the intrinsic pulsatility of an islet ...................................................................................................................................... 8
The metronome hypothesis .............................................................................................................................................................................................. 10
* Corresponding author. University of Michigan Medical School, Brehm Tower, Room 5128, Kellogg Eye Center, 1000 Wall St., Ann Arbor, MI 48105, USA.
Tel.: +7346154084; fax: +7342328162.
E-mail address: [email protected] (L.S. Satin).
http://dx.doi.org/10.1016/j.mam.2015.01.003
0098-2997/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Leslie S. Satin, Peter C. Butler, Joon Ha, Arthur S. Sherman, Pulsatile insulin secretion, impaired glucose tolerance and type 2
diabetes, Molecular Aspects of Medicine (2015), doi: 10.1016/j.mam.2015.01.003
ARTICLE IN PRESS
L.S. Satin et al./Molecular Aspects of Medicine ■■ (2015) ■■–■■
2
13.
14.
15.
How are islets synchronized across the pancreas in vivo? ..................................................................................................................................... 11
Therapeutic implications of insulin pulsatility for pharmacotherapy of Type 2 diabetes and for improving the efficacy of insulin
administration ....................................................................................................................................................................................................................... 12
Summary and conclusions ................................................................................................................................................................................................ 13
Acknowledgements .............................................................................................................................................................................................................. 14
References ............................................................................................................................................................................................................................... 14
1. Introduction
Type 2 diabetes (T2DM) is associated with both a reduction in beta-cell mass and impaired beta-cell function.
Less attention has been paid to beta cell function, which may
begin to decline prior to the reduction in beta cell mass or
the development of T2DM (Rahier et al., 2008). For example,
the early loss of first phase secretion has long been considered a hallmark of T2DM (Nesher and Cerasi, 2002; Straub
and Sharp, 2002; Ward et al., 1986). From a therapeutic
standpoint, improving insulin secretion pharmacologically is a more realistic alternative to stimulating beta cell mass
expansion, in part because the latter is likely to occur on a
much slower time scale than improvements in beta cell function. Even in rodents, where robust changes in beta cell mass
can occur, beta cell function changes more rapidly and more
markedly than mass (Topp et al., 2007). Compensatory
changes in beta cell function would be expected to be even
more important in humans, where mass expansion is two
orders of magnitude slower (Kushner, 2013; Saisho et al.,
2013; Teta et al., 2005).
As is the case for other hormones, insulin is secreted from
the pancreas in a pulsatile manner in both experimental
animals and humans, and in patients with T2DM and other
metabolic disorders the pattern of pulsatile release is disturbed. Thus, soon after the first report that fasting plasma
insulin and glucagon levels oscillate in non-human primates in vivo (Goodner et al., 1977), Turner’s group in the
UK demonstrated that pulsatility occurs in healthy human
subjects (Lang et al., 1979) and found disturbed pulsatility
in subjects with T2DM (Lang et al., 1981).
The main focus of this review is the pulsatile insulin secretion of humans, particularly ‘fast oscillations’ in plasma
insulin that have a period reported to range from 5 to 15
minutes. Readers with a special interest in ultradian insulin
oscillations (period ≈ 80–180 minutes) are directed to other
reviews (Polonsky, 1999).
2. Insulin levels oscillate in fasted humans
Lang and colleagues were the first to report insulin oscillations in the peripheral circulation of fasted but otherwise
healthy human subjects. The oscillations they observed had
a mean period of 15 minutes or so (Lang et al., 1979). Peripheral blood was sampled once per minute for a total
duration of 1–2 hours. An example from their paper shows,
at least initially, clear oscillations in insulin, C peptide, and
glucose concentrations in peripheral blood, as shown in Fig. 1.
The continuous lines depict three-minute moving averages
of the data, while the dashed lines show the raw, unsmoothed
data. Small oscillations in glucose are also apparent in the
lower part of the figure, but are difficult to resolve.
Using records such as this, Lang et al. applied
autocorrelation to aid in pulse detection. Autocorrelation involves creating a mirror image of smoothed time series data,
translating it stepwise along the original data, and then calculating a correlation coefficient for each time point in the
interval. Plots of these coefficients reveal peaks that occur
at multiples of the dominant oscillation period(s). Although Lang et al. and other early studies of in vivo insulin
pulsatility (Hansen et al., 1982) reported an oscillation period
of 10–15 minutes, more recent studies have determined the
in vivo period of insulin oscillations to be closer to 5 minutes.
In a later section, we will discuss why limitations in the technical approaches that were available at the time are likely
to explain the prolonged insulin periods originally reported in this literature.
3. Individuals with T2DM have impaired insulin
pulsatility
Using the same approach, Lang et al. (1981) reported that
individuals with diabetes (their mean fasting glucose was
7 mM) displayed shorter and highly irregular oscillations
having a mean period of 8.8 minutes (vs. controls having a
period of 10.7 minutes). Later studies confirmed the impaired insulin pulsatility of T2DM patients (e.g. Gumbiner
et al., 1996, Hunter et al., 1996, Meier et al., 2013, but see
Lin et al. (2002)).
O’Rahilly et al. (1988) extended these studies in individuals with T2DM to their first-degree relatives who lacked
fasting hyperglycemia, to see whether pulsatility defects
were an early event in the progression to diabetes. Control
subjects were matched by age, gender and BMI to the relatives of patients with diabetes. While the control subjects
exhibited regular insulin oscillations, these were lacking in
the relatives of T2DM subjects. However, the relatives studied
were already glucose intolerant and insulin resistant, and
had reduced first phase insulin secretion, so the loss of
pulsatility may have been secondary to reduced pulse
amplitude, which may have reduced the signal-to-noise
ratio.
In a similar study, Schmitz et al. studied the transition
from normal to abnormal glucose tolerance to fasting hyperglycemia to determine whether the loss of first phase
and pulsatile secretion was due to intrinsic beta cell defects
or glucotoxicity (Schmitz et al., 1997). Insulin action and
insulin pulsatility were measured in healthy offspring of
T2DM patients vs. controls matched by age, gender and BMI.
Approximate entropy (ApEn), a measure of the likelihood
that a similar pattern of observed activity will be repeated
in a given time interval, was used to gauge the level of irregularity of plasma insulin pulses; consistent with the
general sense that entropy quantifies disorder, high ApEn
Please cite this article in press as: Leslie S. Satin, Peter C. Butler, Joon Ha, Arthur S. Sherman, Pulsatile insulin secretion, impaired glucose tolerance and type 2
diabetes, Molecular Aspects of Medicine (2015), doi: 10.1016/j.mam.2015.01.003