See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/256469277

Insulin Stacking Versus Therapeutic Accumulation: Understanding the

Differences

Article in Endocrine Practice · September 2013

DOI: 10.4158/EP13090.RA · Source: PubMed

CITATIONS READS
72 7,815

2 authors:

Tim Heise Luigi Meneghini

Profil University of Texas Southwestern Medical Center
414 PUBLICATIONS 16,476 CITATIONS 136 PUBLICATIONS 2,600 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Tim Heise on 28 December 2015.

The user has requested enhancement of the downloaded file.

 Review Article
INSULIN STACKING VERSUS THERAPEUTIC ACCUMULATION:
UNDERSTANDING THE DIFFERENCES
Tim Heise, MD1; Luigi F. Meneghini, MD2
ABSTRACT
Objective: The build-up in insulin levels following
repeated injection of prandial insulin at close intervals—
referred to as insulin stacking—can increase the risk of
hypoglycemia. With the development of basal insulins with
a half-life >24 hours and a duration of action >40 hours,
clinicians may be concerned about stacking when these
long-acting formulations are administered once daily. The
objective of this review is to clarify the difference between
inappropriate insulin stacking when shorter-acting insulin
formulations are repeatedly used to correct hyperglycemia
and the appropriate accumulation of long-acting insulin
formulations dosed to steady-state pharmacokinetic (PK)
profiles.
Methods: Relevant literature on insulin stacking,
glucose-clamp studies, and clinical studies of insulin, in
conjunction with the clinical experience of the authors,
were used to present an overview of insulin PK proper-
ties and the effects of appropriate and inappropriate dos-
ing intervals on steady-state conditions and likely clinical
outcomes.
Results: Clinical studies confirm theoretical PK prin-
ciples showing that unwanted insulin stacking (excessive
Submitted for publication February 18, 2013
Accepted for publication July 31, 2013
From 1Profil, Neuss, Germany, and 2University of Miami, Miller School of
Medicine, Miami, Florida.
Address correspondence to Dr. Tim Heise, Profil Institut für
Stoffwechselforschung GmbH, Hellersbergstr. 9, 41460 Neuss, Germany.
E-mail: tim.heise@profil.com.
Published as a Rapid Electronic Article in Press at http://www.endocrine
practice.org on September 6, 2013. DOI:10.4158/EP13090.RA
To purchase reprints of this article, please visit: www.aace.com/reprints.
Copyright © 2014 AACE.
concentrations) for basal insulin can be avoided by follow-
ing recommended dosing and titration algorithms. Long-
acting basal insulins need more time to reach steady state
than shorter-acting basal insulins but then show reduced
peak-trough fluctuations, translating into more consistent
biologic action over a 24-hour period.
Conclusion: The unwanted stacking and consequent
hypoglycemia that can occur when correctional doses of
rapid-acting insulin are administered at close intervals
does not occur when long-acting basal insulins are dosed
in appropriate amounts and adjusted at appropriate time
intervals (e.g., insulin stacking, insulin administration,
diabetes, ultralong duration of action, hypoglycemia every
three or more days), allowing for pharmacologic steady-
state accumulation. (Endocr Pract. 2014;20:75-83)
Abbreviations:
NPH = neutral protamine Hagedorn; PD = pharmaco-
dynamic; PK = pharmacokinetic
INTRODUCTION
Excessive accumulation of insulin in the circulation,
often referred to as “stacking,” is typically discussed in
conjunction with the administration of rapid-acting insulin
to correct hyperglycemia. Hirsch (1,2) described stacking
as “the practice of providing correctional doses of insulin
before a prior dose of prandial insulin (or the peak action
of neutral protamine Hagedorn, [NPH]) has had its full
effect,” whereas Bequette (3), in the context of continu-
ous subcutaneous insulin infusion, described stacking as
a situation in which “previously infused insulin still has
an effect on future glucose values.” Physicians, nurses,
and diabetes educators may also be familiar with stacking
described as the “bolus-on-board” issue (4-7). Figure 1 dia-
grammatically illustrates a representative stacking scenario
using a profile that approximates a rapid-acting insulin
(top panel, A) or regular human insulin (bottom panel, B).
Administration of a correctional bolus of insulin before the
ENDOCRINE PRACTICE Vol 20 No. 1 January 2014 75
76 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1)

prior prandial dose has been completely absorbed leads to GOALS OF BASAL INSULIN THERAPY
an acute overlap in insulin action (residual from prior dose
plus correctional dose) and a greater than desired insulin The goal of basal insulin replacement therapy is to
concentration. The actual insulin available thus exceeds reproduce the physiologic 24-hour secretion of insulin that
the required amount, increasing the risk of hypoglycemia. occurs naturally via continuous small pulses from pancre-
It is important to keep in mind that the duration of atic beta-cells, keeping glucose levels stable in the fasting
action of prandial insulins (particularly when used in high state. Intermediate-acting basal insulins, such as NPH, do
doses) can considerably exceed the time of oral glucose not reproduce the endogenous (physiologic) glucose-low-
(carbohydrate) absorption (8). Each of these definitions/ ering effect needed for proper basal insulin replacement
concepts carries the implication that the additional insu- due to substantial (and unwanted) peaks in biologic activ-
lin added to a previous dose might result in inappropri- ity and insufficient duration of biologic action (9,14,15).
ate insulin build-up to levels that can precipitate hypogly- Avoiding treatment-associated hypoglycemia and hyper-
cemia. Thus, the term “stacking” is used here to describe glycemia with NPH insulin requires more frequent dos-
unintended insulin increases to inappropriately high levels, ing and complex self-management, often limiting patients’
usually acutely. ability to maintain stable glycemic ranges (16-19). For
New-generation basal insulins with a half-life >24 those patients with a strong surge in blood glucose at dawn
hours and a duration of action >40 hours are currently in
clinical development (9-13), with one, insulin degludec,
currently entering clinical use in some nations. This desir-
able pharmacokinetic (PK) property, which should allow
for reliable once-daily dosing for all patients, has raised
the concern of whether daily administration of a basal insu-
lin with a duration of action that exceeds 24 hours could
result in excessive accumulation of insulin in the circula-
tion, thereby leading to an increased risk in the frequency,
duration, or severity of hypoglycemia. Based on clinicians’
experience with other drugs, this is not an illogical concern.
Concerns about the inappropriate dosing of rapid-
acting insulin, either from pump infusion or subcutaneous
injection, are not quite analogous to the appropriate build-
up or accumulation that must occur for a long-acting basal
insulin to reach a stable, steady-state concentration, which
we will term steady-state accumulation. The PK goals of
therapy for these two insulin classes are quite different. For
rapid-acting insulin administered as intermittent boluses,
the goal is to achieve appropriate single-dose kinetics with
a comparatively large peak to trough ratio, so that insu-
lin concentration rises rapidly to prevent an excessive
postprandial glucose increase and drops quickly to avoid
late postprandial hypoglycemia. By contrast, with a basal
insulin dosed once daily, the goal is to achieve appropri-
ate steady-state kinetics with a low peak to trough ratio,
whereby over a 24-hour period the rate of insulin uptake
and elimination by target tissues equals the rate of insu-
lin absorption. This ensures that the amount of insulin
available in circulation over this period of time (24 hours)
remains constant and predictable.
This paper is intended to review the goals of basal
insulin therapy and the clinical properties and PK of basal
insulin in order to better understand why a long half-life
Fig. 1. Schematic representation of how inappropriate stacking
and appropriate dosing, both in terms of amount and inter-
could occur with too-early administration of a hypothetical rapid-
val, allow for the desirable accumulation of insulin neces- acting insulin. If a corrective dose of insulin is administered while
sary to achieve the flat profile and low peak to trough ratio a previous dose of (prandial) insulin is still available, this can
important for optimal basal replacement. result in excessive glucose-lowering action and hypoglycemia.
 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1) 77
(20-22), the option of a variable basal delivery by insulin
pump or supplementing treatment with nighttime NPH can
be considered.
Thus, appropriately replacing insulin needs requires
individual adjustments, which often depend on a subject’s
lifestyle, dietary preferences, and endogenous glucose
production. Improved PK/pharmacodynamic (PD) proper-
ties of new insulins therefore can only be one part of an
individual optimization in insulin therapy. Because of the
variable lifestyle of a patient, even a complete physiologic
(flat) basal insulin substitution might still result in some
fluctuations in fasting blood glucose levels.
The long-acting basal insulin analogs (insulin detemir
and insulin glargine) have improved PK profiles compared
with NPH (9,15,23,24) due to their prolonged biologic
action and low peak to trough ratios (14). Furthermore,
their duration of action at steady state, which is dose
dependent, approaches 24 hours in most people with type
2 diabetes (14,25) but may not last a full 24 hours in all
patients, especially those with type 1 diabetes and/or those
requiring lower insulin doses. Newer insulin analogs cur-
rently in clinical development, such as insulin degludec
and LY2605541 (pegylated insulin lispro), might offer
additional benefits due to an even longer duration of action.
For example, insulin degludec demonstrates glucose-low-
ering activity beyond 42 hours, translating into lower bio-
logic fluctuations when dosed once daily in both type 1 and
type 2 diabetes (9,26-28). In two phase 1 studies in healthy
subjects (29) and one phase 1 study in people with type 2
diabetes (13), LY2605541 appeared to have a half-life in
the range of 45 to 76 hours and a duration of action well
beyond 24 hours.
In one sense, concerns about stacking-related prob-
lems with insulins having a longer duration of action may
seem paradoxical, as existing data for both insulin glargine
and insulin detemir show no indication of more frequent
or severe hypoglycemic episodes compared with NPH. On
the contrary, at equivalent degrees of glycemic control,
the data consistently show significantly decreased risks
of symptomatic, nocturnal, and overall hypoglycemia in
patients with type 2 diabetes treated with either insulin
detemir or insulin glargine (30,31). Recent data for insulin
degludec also indicate a significantly lower risk of noctur-
nal hypoglycemia versus insulin glargine in type 1 diabetes
(P = .021) (11) and lower risk of both nocturnal (P = .04)
and overall (P = .036) hypoglycemia in type 2 diabetes
compared with those treated with insulin glargine (P = .04)
(10).
INSULIN PHARMACOKINETICS
Addressing clinical concerns about insulin accumu-
lation when duration of action exceeds 24 hours requires
clarification of how basal insulin is absorbed and elimi-
nated and how it reaches a stable, steady state following
repeated once-daily dosing. The gold standard for obtaining
data describing insulin absorption, elimination, and plasma
concentration over time (i.e., PK) and the glucose-lowering
effect (i.e., PD) is the glucose-clamp study (14,15,24,32).
In these studies, fasted subjects are administered a spe-
cific dose of insulin while glucose is infused intravenously
at a rate sufficient to maintain a predetermined glucose
concentration (i.e., the “clamped” glucose level), usually
around 90 to 100 mg/dL. The glucose-lowering effect of
insulin over time is estimated by the rate at which glucose
is infused to maintain the clamped level (glucose infusion
rate, usually reported as mg/kg/min). Plasma insulin con-
centration can also be measured via serial venous sampling
during the clamp to estimate PK characteristics. Typical
PK and PD parameters derived from these clamp studies
are described and shown in relation to the time-action pro-
file of representative basal insulin in Figure 2.
Two PK concepts that are important to emphasize are
the steady-state condition and the peak to trough ratio. The
steady-state condition is a state of equilibrium wherein the
plasma concentration of insulin remains within a constant
range each day because the 24-hour elimination rate equals
the 24-hour absorption rate; hence, insulin levels do not
increase further with repeated equivalent doses at appro-
priately spaced intervals. The peak to trough ratio is the
ratio of the highest to lowest biologic activity of insulin
at steady state. Trough levels (occasionally called residual
levels) are measured just prior to the next insulin injection.
The peak to trough ratio indicates how consistently the
glucose-lowering effect is sustained over the dose interval;
the smaller the peak to trough ratio the more steady (con-
sistent) the biologic action over time.
Several things should be kept in mind when reviewing
the time-action curve shown in Figure 2. In the distribution
phase, during which insulin is absorbed from the subcuta-
neous depot into the circulation, there is a slight separation
between the maximum insulin concentration (reflecting
absorption) and the time of maximum glucose-lowering
action (reflecting the time required for insulin to access
target tissues, bind to receptors, and elicit metabolic activ-
ity). After being absorbed from the subcutaneous depot and
entering circulation, insulin is rapidly taken up and metab-
olized by the large pool of insulin receptors available in
target tissues such as the liver and muscle. In fact, the elim-
ination of intravenously injected human insulin has a half-
life of approximately 5 to 10 minutes (33,34). Elimination,
predominantly through receptor interaction and internal-
ization, is much more rapid than absorption; accordingly,
the reported half-life of insulin reflects absorption half-life
rather than elimination half-life. Thus, the terminal half-
life of insulin is determined by the rate of absorption (rate-
limiting step) rather than the rate of elimination. This PK
property is referred to as “flip-flop kinetics” (35,36).
The terminal plasma half-life (t½) of a drug is the
time it takes for 50% of the drug to be eliminated from the
78 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1)

Fig. 2. Representative PK/PD profile for basal insulin after single-dose administration. Cmax: Peak insulin concentration
reached; Tmax: Time when peak insulin concentration is reached; AUC-INS0-24: area under the PK curve corresponds to
total amount of insulin (analogue) absorbed over 24 hours; t½ (terminal plasma half-life): time it takes for 50% of the
drug to be eliminated from the plasma in its terminal phase (i.e., after Cmax has been reached and the drug is eliminated
at a steady rate). Peak to trough ratio: difference between the peak and minimum drug concentration. Trough levels
are pharmacokinetic concentrations measured just prior to the next insulin injection; glucose infusion rate (GIR)max:
time of maximum glucose infusion rate (the glucose infusion rate is the indicator of blood glucose-lowering effect in
glucose clamp studies) and corresponds to peak insulin action. AUC-GIR0-24: area under the PD curve corresponds to
total amount of glucose infused over 24 hours, which is an indicator of the total insulin effect in a treatment day. AUC
= area under the curve; GIR = glucose infusion rate; PD = pharmacodynamic; PK = pharmacokinetic.

plasma and is particularly important when evaluating its
potential for accumulation (15). Because the plasma con-
centration decreases by 50% with each half-life, at least 4
half-lives are required for a drug to be nearly completely
eliminated (50 + 25 + 12.5 + 6.25% = 93.75%) (15). If
insulin is readministered before the complete elimination
of a prior dose, this will inevitably lead to accumulation
(i.e., an overlap of the PK/PD of 2 or more injections, and
therefore to higher PK levels than observed with a single
injection alone). How much PK levels will increase is
determined by the dose (amount) of each injection (obvi-
ously the higher the dose, the higher the PK levels) and the
dosing frequency (the more frequent the dose, the higher
the PK levels). Although the PK levels reached in steady
state are dose dependent, the time to reach steady state is
solely dependent on the half-life of a drug. As a rule of
thumb, the common assumption is that it takes about 4 to 5
half-lives to reach steady state. This, however, depends on
the definition of “steady state.” Whereas after 5 half-lives
the PK levels reach 99% of the constant concentration seen
at steady state, after 3 half-lives the PK levels will already
approach 90% of the steady-state concentration (Fig. 3),
which some authors regard as the threshold for defining
“clinical” steady state (37).
When the t½ of an insulin is about 24 hours, conve-
niently, this implies that 50% of the depot concentration
will be absorbed into the circulation and subsequently
eliminated each day. With a hypothetical daily dose of
10 U required for fasting glucose control, appropriate
steady-state accumulation of insulin will occur until suf-
ficient insulin accumulates in the injection depot (20 U),
such that a 50% elimination each day is equal to a 10 U
elimination and 10 U of circulating insulin with biologic
activity remaining. This is when a steady-state condition
is reached. As long as the dose remains at 10 U, no fur-
ther accumulation is possible because 10 U will always be
eliminated each day (steady state has been achieved). If
the dose were to be increased further, to 15 U daily, for
example, then a modest amount of steady-state accumula-
tion would resume until a new steady state was reached.
 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1) 79

Fig. 3. Example of time-to-reach steady state without inappropriate accumulation of basal insulin using a simplified one-
compartment model (10 U, with t½ ~ 24 hours). Because dosing frequency is approximately equal to half-life, insulin only
accumulates until steady state is reached, at which time the daily injected dose is balanced by elimination. s.c. = subcutaneous;
U = units of insulin; t½ = half-life.

Repeat administration of a drug prior to its complete
elimination will lead to greater steady-state accumula-
tion, but will also reduce the peak-trough fluctuations. On
the other hand, when the dosing interval is greater than
the half-life, steady-state accumulation will be less pro-
nounced, but the peak-trough fluctuations will be greater.
Optimally basal insulins should have a nearly peakless PK/
PD profile. A once-daily basal insulin with a half-life >24
hours should have less of a peak effect compared with an
insulin with a shorter half-life.
STEADY STATE AND HALF-LIVES
OF BASAL INSULINS
The steady-state principles described above have
been demonstrated in a PK/PD study in people with type
1 diabetes using the long-acting basal insulin analog, insu-
lin detemir (25). In a glucose clamp study involving 25
subjects, insulin detemir was dosed twice daily over 7 to
14 days (mean daily dose, 0.36 ± 0.11 U/kg). PK levels
increased from zero to a mean maximum level of 2,217 ±
960 pmol/L after the second injection on the first treatment
day (25). Thereafter, however, there was no substantive
further increase in PK levels, so that after 7 to 14 days of
treatment, trough levels remained at the same level as peak
concentrations. Compared with steady-state data, the aver-
age insulin concentrations on the first treatment day were
lower and peak to trough fluctuations were higher. Similar
results were observed for the PD effects (25). These results
nicely confirm the theoretical considerations above. With
a dosing interval around the half-life, the insulin concen-
tration will increase (i.e., accumulation will occur until
steady state is reached) and peak to trough fluctuations
will decrease. When steady state is achieved, no further
increase in insulin concentration will occur unless the dose
amount is changed or the dosing interval shortened.
80 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1)

Similar results were obtained for insulin glargine
when given once daily over 7 or 14 days in people with
type 1 diabetes (38,39). At steady state, insulin glargine
had a slightly higher mean peak concentration and greater
PD effects on the last treatment day compared with the
first treatment day (38,39). In general, investigations of
glargine’s PK properties are hampered by the lack of a
commercially available assay specific for insulin glargine
and its metabolites. This might explain why an earlier study
did not detect any steady-state accumulation with repeated
dosing of insulin glargine over 11 days (40). However, a
recent investigation using a specific assay determined the
half-life of insulin glargine to be 12.5 hours (26), so that
with once-daily dosing, one would certainly expect accu-
mulation to a steady state but with greater peak to trough
ratios than if it had a longer half-life (see the basal insulin
curves in Fig. 4 A).
The same randomized, double-blind, two-period,
crossover trial also investigated the PK properties of
insulin degludec under steady-state conditions in 66 peo-
ple with type 1 diabetes (26). Patients received insulin
degludec in a fixed dose (0.4, 0.6, or 0.8 U/kg) once daily
for 8 days. A euglycemic glucose clamp was performed on
treatment day 8. Consistent with the observed half-life of
about 25 hours, insulin degludec concentrations increased
over the first treatment days and reached steady state in
2 to 3 days without additional steady-state accumulation
thereafter. In type 2 diabetes, at steady state, the PK and
PD profiles of insulin degludec were virtually peakless, as
indicated, for example, by an equal distribution of the PK
levels and the PD effect between the first and the second
12 hours postdosing (indicated by a ratio between the area
under the curve [AUC]0-12 h, steady state [SS] and the AUCtotal,
SS of approximately 0.5) (Fig. 5).

Fig. 4. Hypothetical examples of profiles of insulins with various half-lives; accumulation from first dose to steady state (top
panel, A) and perturbations following various types of common dosing errors as indicated by arrows, when introduced at steady
state (bottom panel, B). When interpreting the effects of double-dosing (bottom row), it is important to note that different pharma-
cokinetic scales have been used for the rapid-acting and basal insulin curves. Fluctuations in insulin concentration (and therefore
glucose-lowering action) are greatest, and dosing errors have the most acute effects, with basal insulin having a shorter half-life
(e.g., 6 hours) and duration of action. Fluctuations are dampened and dosing errors have less acute effects after stacking with
insulin formulations having a longer half-life/duration of action. The half-lives for basal insulin shown in the figure correspond
approximately to those of neutral protamine Hagedorn (6 hours) (41), insulin glargine (12.5 hours) (26), and insulin degludec (25
hours) (26).
 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1) 81
Another basal insulin currently in clinical develop-
ment is a pegylated insulin lispro (LY2605541, Humalog-
PEG). In a phase 1 study, fixed doses (3, 4.5, 6, and 9 nmol/
kg) of LY2605541 were given once daily for 14 days to 32
patients with type 2 diabetes who previously used insulin
(13). Two 24-hour euglycemic clamps on days 1 and 14
and interim PK sampling indicated that steady state had
been reached by 7 to 10 days, with a low peak to trough
ratio and a duration of action >24 hours.
It is important to carefully consider the time it takes
for basal insulin to reach steady-state levels when making
dose adjustments. Increasing the dose of an insulin prior
to that insulin achieving steady state can result in overin-
sulinization and potentially in hypoglycemia. This is illus-
trated in Figure 5, in which we show that clinical steady
state (defined as trough levels reaching 90% of the plateau
concentration) is usually achieved after 3 to 4 times the
insulin’s half-life with once-daily dosing (Fig. 4 A). A basal
insulin with a longer duration of action will take more time
to reach steady state, but once at steady state, the insulin
will have lower peak to trough ratios than a basal insu-
lin with a shorter duration of action. Thus, at steady state,
shorter-acting basal insulins will show stronger fluctua-
tions in PK concentrations than longer-acting formulations
when administered at comparable dosing intervals.
Figure 4 B shows examples of how errors in dosing
for hypothetical rapid-acting and basal insulins with dif-
ferent half-lives, when introduced at steady state, might
affect PK profiles and thereby glucose-lowering action.
The problem of inappropriate changes in PD effect can be
avoided if dose adjustments are performed at appropriate
intervals. For example, if a basal insulin is titrated against
fasting plasma glucose levels and the time taken for that
basal insulin to reach a new steady state after a dose adjust-
ment were 2 to 3 days, then it may be unwise to adjust the
dose more than twice per week. If insulin dose adjustments
were made every 2 days, this would not allow concentra-
tions to reach steady state between dose adjustments, with
a risk of “overshooting” the fasting glucose target with too
much insulin on board and possibly result in hypoglyce-
mia for some patients. With low peak to trough ratios and
long duration of action with a basal insulin, occasional dos-
ing errors should not result in clinically relevant adverse
effects.
This is in contrast to the undesirable stacking that
occurs with administration of excess prandial insulin,
wherein the high peak to trough ratio could result in dan-
gerous glucose-lowering over a short period (e.g., 2 to
6 hours; Fig. 1). The difference in peak to trough ratios
translates into the acuteness of the glucose-lowering (bio-
logic) activity for insulin. For example, a dosing error for
a basal insulin with a low peak to trough ratio would have
limited acute impact on blood glucose effect because the
effect would be distributed more or less evenly over many
hours (>42 hours in the case of insulin degludec), and the
Fig. 5. At steady state, insulin degludec produces a dose-propor-
tional and near-constant pharmacodynamic effect over a 24-hour
period. Reproduced with permission from (27). GIR = glucose
infusion rate.
increase in peak effect would still be relatively minor when
compared to the effect of dosing errors with a prandial
insulin. In support of this observation, iatrogenic overdos-
ing has not been a major concern when titrating basal insu-
lin with an ultralong duration of action to reach glycemic
targets (10-12).
CONCLUSION
Steady-state accumulation is a normal part of the PK
process that enables a long-acting insulin to reach a stable,
steady-state condition with subsequently minimal fluc-
tuation in insulin levels, and therefore, glucose-lowering
activity. This should not be confused with the potentially
dangerous acute overaccumulation (stacking) that occurs
when a corrective dose of rapid-acting insulin is admin-
istered to correct hyperglycemia before the previous dose
has been eliminated. Basal insulins with an ultralong dura-
tion of action with a half-life of at least 24 hours and a
low peak to trough ratio do not pose any special concern
regarding the inappropriate accumulation of insulin, so
long as appropriate dosing and adjustment intervals are
respected.
DISCLOSURE
Dr. Luigi F. Meneghini is a consultant for Novo
Nordisk and Sanofi-Aventis and is on advisory boards
for Novo Nordisk and Halozyme. He receives research
support from Pfizer, MannKind, Boehringer Ingelheim,
and Sanofi-Aventis. Dr. Tim Heise’s institution received
research support from Astellas, Bayer, Becton-Dickinson,
Biocon, Biodel, Boehringer Ingelheim, GlaxoSmithKline,
Roche, Johnson & Johnson, Eli Lilly, Novo Nordisk,
Prosidion, Sanofi-Aventis, and SkyePharma. Dr. Heise is
82 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1)
on the advisory boards of and has received speaker hono-
raria and travel grants from Boehringer Ingelheim and
Novo Nordisk.
ACKNOWLEDGMENT
We thank Gabrielle Parker and Gary Patronek for edit-
ing and writing assistance, supported by Novo Nordisk.
REFERENCES
1. Hirsch IB. Insulin analogues. N Engl J Med. 2005;352:
174-183.
2. DeWitt DE, Hirsch IB. Outpatient insulin therapy in type
1 and type 2 diabetes mellitus: a scientific review. JAMA.
2003;289:2254-2264.
3. Bequette BW. Glucose clamp algorithms and insulin
time-action profiles. J Diabetes Sci Technol. 2009;3:
1005-1013.
4. Zisser H, Robinson L, Bevier W, et al. Bolus calculator:
a review of four ”smart” insulin pumps. Diabetes Technol
Ther. 2008;10:441-444.
5. Walsh J, Roberts R. The Bolus on Board (BOB or Unused
Insulin). Available at: http://www.diabetesnet.com/diabe-
tes-control/rules-control/bolus-board. Accessed December
7, 2012.
6. Walsh J, Roberts R, Bailey T. Guidelines for optimal
bolus calculator settings in adults. J Diabetes Sci Technol.
2011;5:129-135.
7. Walsh J, Roberts R. An accurate DIA prevents excessive
insulin stacking. Available at: http://www.diabetesnet.com/
about-diabetes/insulin/insulin-action-time/duration-insu-
lin-action. Accessed December 6, 2012.
8. Nosek L, Roggen K, Heinemann L, et al. Insulin aspart
has a shorter duration of action than human insulin over
a wide dose-range. Diabetes Obes Metab. 2013;15:77-
83.
9. Owens DR. Insulin preparations with prolonged effect.
Diabetes Technol Ther. 2011;13(suppl 1):S5-S14.
10. Garber AJ, King AB, Del Prato S, et al. Insulin degludec,
an ultra-longacting basal insulin, versus insulin glargine in
basal-bolus treatment with mealtime insulin aspart in type
2 diabetes: (BEGIN Basal-Bolus Type 2): a phase 3, rando-
mised, open-label, treat-to-target trial non-inferiority trial.
Lancet. 2012;379:1498-1507.
11. Heller S, Buse J, Fisher M, et al. Insulin degludec, an
ultra-longacting basal insulin, versus insulin glargine in
basal bolus treatment with mealtime insulin aspart in type
1 diabetes (BEGIN Basal-Bolus Type 1): a phase 3, ran-
domised, open-label, treat-to-target non-inferiority trial.
Lancet. 2012;379:1489-1497.
12. Zinman B, Philis-Tsimikas A, Cariou B, et al. Insulin
degludec versus insulin glargine in insulin-naive patients
with type 2 diabetes: a 1-year, randomized, treat-to-tar-
get trial (BEGIN Once Long). Diabetes Care. 2012:35:
2464-2471.
13. Heise T, Howey DC, Sinha VP, Choi SL, Mace KF.
Steady-state pharmacokinetics (PK) and glucodynamics
(GD) of the novel, long-acting basal insulin LY2605541
dosed once-daily (QD) in patients with type 2 diabetes mel-
litus (T2DM). Diabetes. 2012;61(suppl 1):A256. Abstract.
14. Heise T, Pieber TR. Towards peakless, reproducible and
long-acting insulins. An assessment of the basal analogues
based on isoglycaemic clamp studies. Diabetes Obes
Metab. 2007;9:648-659.
15. Arnolds S, Kuglin B, Kapitza C, Heise T. How pharma-
cokinetic and pharmacodynamics principles pave the way
for optimal basal insulin therapy in type 2 diabetes. Int J
Clin Pract. 2010;64:1415-1424.
16. Cryer PE. Hypoglycaemia: the limiting factor in the
glycaemic management of Type I and Type II Diabetes.
Diabetologia. 2002;45:937-948.
17. Cryer PE. The barrier of hypoglycemia in diabetes.
Diabetes. 2008;57:3169-3176.
18. Fidler C, Elmelund Christensen T, Gillard S.
Hypoglycemia: an overview of fear of hypoglycemia,
quality-of-life, and impact on costs. J Med Econ. 2011;14:
646-655.
19. Frier BM. How hypoglycaemia can affect the life of a per-
son with diabetes. Diabetes Metab Res Rev. 2008;24:87-92.
20. Atiea JA, Ryder RR, Vora J, et al. Dawn phenomenon:
its frequency in non-insulin-dependent diabetic patients on
conventional therapy. Diabetes Care. 1987;10:461-465.
21. Bolli GB, Gerich JE. The “dawn phenomenon”--a com-
mon occurrence in both non-insulin-dependent and insu-
lin-dependent diabetes mellitus. N Engl J Med. 1984;310:
746-750.
22. Carroll MF, Schade DS. The dawn phenomenon revisited:
implications for diabetes therapy. Endocr Pract. 2005;11:
55-64.
23. Heise T, Nosek L, Rønn BB, et al. Lower within-sub-
ject variability of insulin detemir in comparison to NPH
insulin and insulin glargine in people with type 1 diabetes.
Diabetes. 2004;53:1614-1620.
24. Porcellati F, Bolli GB, Fanelli CG. Pharmacokinetics and
pharmacodynamics of basal insulins. Diabetes Technol
Ther. 2011;13(suppl 1):S15-S24.
25. Bott S, Tusek C, Jacobsen LV, et al. Insulin detemir
under steady-state conditions: no accumulation and con-
stant metabolic effect over time with twice-daily admi-
nistration in subjects with Type 1 diabetes. Diabet Med.
2006;23:522-528.
26. Heise T, Hövelmann U, Nosek L, Bøttcher SG, Granhall
C, Haahr H. Insulin degludec: two-fold longer half-life
and more consistent pharmacokinetic profile than insulin
glargine. Diabetologia. 2011;54(suppl 1):1046. Abstract.
27. Heise T, Nosek L, Bøttcher SG, Hastrup H, Haahr H.
Ultra-long-acting insulin degludec has a flat and stable
glucose-lowering effect in type 2 diabetes. Diabetes Obes
Metab. 2012;14:944-950.
28. Heise T, Hermanski L, Nosek L, Feldman A, Rasmussen
S, Haahr H. Insulin degludec: four times lower pharmaco-
dynamic variability than insulin glargine under steady-state
conditions in type 1 diabetes. Diabetes Obes Metab. 2012;
14:859-864.
29. Sinha VP, Howey DC, Soon DKW, et al. Single-dose
pharmacokinetics (PK) and glucodynamics (GD) of the
novel, long-acting basal insulin LY2605541 in healthy sub-
jects. Diabetes. 2012;61(suppl 1):A273.
30. Horvath K, Jeitler K, Berghold A, et al. Long-acting
insulin analogues versus NPH insulin (human isophane
insulin) for type 2 diabetes mellitus. Cochrane Database
Syst Rev. 2007;2:CD005613.
31. DeVries JH, Nattrass M, Pieber TR. Refining basal insu-
lin therapy: what have we learned in the age of analogues?
Diabetes Metab Res Rev. 2007;23:441-454.
32. Heise T, Heinemann L. Rapid and long-acting analogues as
an approach to improve insulin therapy: An evidence-based
medicine assessment. Curr Pharm Des. 2001;7:1303-1325.
 Insulin Stacking: Good or Bad?, Endocr Pract. 2014;20(No. 1) 83
33. Williams RF, Gleason RE, Soeldner JS. The half-life
of endogenous serum immunoreactive insulin in man.
Metabolism. 1968;17:1025-1029.
34. Frost DP, Srivastava MC, Jones RH, Nabarro JD,
Sonksen PH. The kinetics of insulin metabolism in diabe-
tes mellitus. Postgrad Med J. 1973;49(suppl 7):949-954.
35. Yáñez JA, Remsberg CM, Sayre CL, Forrest ML,
Davies NM. Flip-flop pharmacokinetics—delivering a
reversal of disposition: challenges and opportunities during
drug development. Ther Deliv. 2011;2:643-672.
36. Boxenbaum H. Pharmacokinetics tricks and traps: flip-
flop models. J Pharm Pharm Sci. 1998;1:90-91.
37. Rowland M, Tozer TN. Clinical Pharmacokinetics and
Pharmacodynamics – Concepts and Applications. 4th ed.
Philadelphia, PA: Lippincott Williams & Wilkins; 2007.
38. Porcellati F, Rossetti P, Ricci NB, et al. Pharmacokinetics
and pharmacodynamics of the long-acting insulin analog
View publication stats
glargine after 1 week of use compared with its first admi-
nistration in subjects with type 1 diabetes. Diabetes Care.
2007;30:1261-1263.
39. Bock G, Wutte A, Köhler G, Korsatko S, Semlitsch B,
Pieber TR. Pharmacodynamics and pharmacokinetics of
long-acting insulin analogues detemir and glargine after
7 days of use and after its first administration in subjects
with type 1 diabetes. Diabetologia. 2008;51(suppl 1):S390.
Abstract.
40. Heise T, Bott S, Rave K, Dressler A, Rosskamp R,
Heinemann L. No evidence for accumulation of insulin
glargine (LANTUS): a multiple injection study in patients
with Type 1 diabetes. Diabet Med. 2002;19:490-495.
41. Kølendorf K, Bojsen J. Kinetics of subcutaneous NPH
insulin in diabetics. Clin Pharmacol Ther. 1982;31:494-500.