Hospital Practice

ISSN: 2154-8331 (Print) 2377-1003 (Online) Journal homepage: http://www.tandfonline.com/loi/ihop20

Optimizing Inpatient Glycemic Control with Basal-

Bolus Insulin Therapy

R. Daniel Pollom MD

To cite this article: R. Daniel Pollom MD (2010) Optimizing Inpatient Glycemic Control with
Basal-Bolus Insulin Therapy, Hospital Practice, 38:4, 98-107

To link to this article: http://dx.doi.org/10.3810/hp.2010.11.346

Published online: 13 Mar 2015.

Submit your article to this journal

Article views: 5

View related articles

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ihop20

Download by: [University of Pennsylvania] Date: 09 November 2015, At: 10:46


Optimizing Inpatient Glycemic Control
with Basal-Bolus Insulin Therapy
R. Daniel Pollom, MD 1
1
Diabetes Care Center, Community
Health Network, Indianapolis, IN
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
Correspondence: R. Daniel Pollom, MD,
Medical Director,
Diabetes Care Center,
Community Health Network,
8205 E. 56th St.,
Suite 100,
Indianapolis, IN 46216.
Tel: 317-621-4044
Fax: 317-621-4050
E-mail: dpollom@ecommunity.com
98
CLINICAL FEATURES
Abstract: Hyperglycemia is highly prevalent in the acute-care setting and is associated with
an increased risk of morbidity and mortality. Evidence suggests that glycemic control in this
population is suboptimal, due in part to continued use of nonphysiologic sliding-scale insulin
strategies without scheduled basal insulin doses or prandial insulin with concomitant correction
doses. Although the ineffectiveness and risks of sliding-scale insulin regimens have been
criticized for decades, sliding-scale insulin is still the most commonly prescribed subcutaneous
insulin regimen among inpatients. Improving inpatient management requires the use of scheduled
basal-bolus insulin therapy that includes basal insulin, nutritional insulin, and supplemental, or
correctional, insulin. Insulin analogs are the preferred insulins, as they provide a more physiologic
action than human insulin regimens, are associated with a lower risk of hypoglycemia, and
are more convenient to administer than human insulins. Standardized insulin protocols and
subcutaneous insulin order sets are critical components of effective inpatient glycemic control.
Although preliminary data have demonstrated that inpatient diabetes management programs
involving basal-bolus insulin therapy are effective and well tolerated, more research is needed.
Keywords: hyperglycemia; in-hospital; inpatient; insulin; sliding-scale; basal-bolus
Introduction
The rapid increase of type 2 diabetes in the community has been well documented.
This rise in the prevalence of diabetes and its risk factors has resulted in a substantial
increase in the prevalence and impact of hyperglycemia and diabetes among
hospitalized patients.1 In one study of  2000 patients consecutively admitted to
a community teaching hospital, the incidence of hyperglycemia was 38%,2 while
data from a 2007 survey suggested that 22% of all inpatient days in the hospital
were incurred by patients with diabetes.3
However, not all patients with hyperglycemia will have previously diagnosed
diabetes. For every 2 patients who have diagnosed diabetes, 1 additional patient will
develop newly observed hyperglycemia.2 In these patients, hyperglycemia results
from undiagnosed diabetes; temporary conditions, such as the stress of hospitaliza-
tion, illness, or surgery; or the use of certain medications or parenteral nutrition.4
The presence of hyperglycemia and suboptimal glycemic control is a strong
predictor of negative outcomes among inpatients in the intensive care unit (ICU).
Patients with hyperglycemia face a higher risk of mortality. Mortality rates in
the ICU can be as high as 40% for patients with glucose levels  300 mg/dL and
up to 35% for patients with levels  200 mg/dL.5 Higher rates of infection and
complications, and a higher likelihood of requiring surgical procedures and longer
lengths of hospital stay are often reported among critically ill and noncritically ill
inpatients with hyperglycemia.1,5–11
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331

Moreover, the costs associated with inpatient care of
patients with diabetes are substantial, accounting for nearly
half of the $174 billion in total medical expenditures.3 Because
patients with hyperglycemia are more likely to have worse
outcomes and longer hospital stays than normoglycemic
patients,12 effective inpatient glycemic control is critical.
Current State of Inpatient
Glycemic Control
Considerable evidence suggests that inpatient diabetes is
poorly controlled.13 Approximately 90% of hospitalized
patients with diabetes using insulin therapy experience
glycemic values  200 mg/dL. In 37% of those patients using
insulin, such elevations will persist for  3 days.13
Glycemic control is frequently overlooked in general med-
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
icine and surgical units. Poor control and associated adverse
outcomes then become a quality-of-care issue.1 Challenges
to achieving effective glycemic control include the fear of
hypoglycemia, a lack of knowledge regarding adverse effects
of hyperglycemia, and competing care priorities among insti-
tutions and patients.14 Several factors specific to inpatient care
can also complicate glycemic management, such as frequent
changes among health care teams, insulin requirements, other
medication changes, and delays in timely nutritional intake.14
Insulin therapy is the most effective agent for achieving
and maintaining glycemic control in the hospital.12,15 Despite
the availability of numerous protocols and guidelines for
inpatient insulin use, clinicians continue to use sliding-scale
insulin therapy.16 Although its definitions vary, sliding-scale
insulin usually includes the use of a regular human insulin
in response to hyperglycemia without any scheduled basal
or prandial insulin. This method of glucose regulation has
been shown to be ineffective. Therefore, a renewed focus on
effective strategies for managing hyperglycemia and system-
atic changes to address poor glycemic control are needed.
The Impact of Hyperglycemia
and Its Management
Although hyperglycemia is linked to numerous adverse con-
sequences, the results of interventional studies on the impact
of intensive therapies to normalize glycemic control have been
inconsistent.12 Early interventional studies demonstrated that
tight glycemic control using blood glucose targets of 80 to
110 mg/dL were associated with improved hospital outcomes
in critical-care settings.17,18 Less stringent glycemic control with
a target of  140 mg/dL was also associated with significant
benefits, including a 6.2% absolute reduction and a 29.3% rela-
tive reduction in hospital mortality among patients in the ICU.19
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331
Optimizing Inpatient Glycemic Control
However, the aggressive target of between 80 and
110 mg/dL has been difficult to achieve in subsequent studies
without increasing the risk of severe hypoglycemia.20–22
Moreover, recent studies and meta-analyses have not found
consistent reductions in mortality with intensive glycemic
control outside of the surgical ICU.20,23,24 These studies also
demonstrated that severe hypoglycemia was more common in
the intensively treated group,23 as has been shown in outpatient
studies on the impact of intensive glycemic control.25
Together, these data suggest that the benefits of strict
glycemic control in hospitalized patients (with a target of
80–110 mg/dL) may be offset, at least in part, by increased
rates of hypoglycemia, largely due to the absence of frequent,
effective glucose monitoring.12 In clinical practice, a number
of factors may contribute to increased rates of hypoglycemia,
including inaccuracies of glucose monitoring using point-
of-care devices, inadequate intravenous (IV) insulin algo-
rithms, and failure to adjust insulin regimens after changes
in patient conditions, such as changes in nutritional intake
or renal function.
These recent results have prompted the American Diabetes
Association (ADA) and the American Association of
Clinical Endocrinologists (AACE) to issue an updated
consensus statement on inpatient glycemic management.12
The ADA/AACE consensus statement also suggests that
the risk of hypoglycemia associated with inpatient glycemic
control can be minimized by relaxing glycemic targets and
implementing standardized insulin protocols.12 According
to the statement, blood glucose levels in critically ill
patients should now be maintained between 140 and 180
mg/dL. Targets  110 mg/dL in this population are not
recommended.12 Although no prospective studies have been
conducted in noncritically ill patients, the ADA/AACE
statement recommends premeal glucose targets  140
mg/dL in conjunction with random blood glucose targets
 180 mg/dL in this population.12 Although there are no
large prospective clinical trials sufficiently powered to
detect differences in outcomes, there are prospective stud-
ies examining the efficacy and tolerability of basal-bolus
insulin therapy outside of the ICU.26,27
Treatment Options for Inpatient
Hyperglycemia
Insulin therapy is the preferred method for achiev-
ing glycemic control in the hospital setting. 12 Oral
secretagogues offer little flexibility or opportunity for
titration and contribute to hypoglycemia when meals are
delayed or missed altogether.15,28 Although metformin is
99
 R. Daniel Pollom

an important component of outpatient diabetes therapy, Table 1. Time-Action Profiles of Subcutaneous Insulin Therapies
many hospitalized patients may have contraindications to Insulin or Insulin Onset of Time to Duration
metformin or clinical factors that increase the risk of lactic Analog Action Peak Action of Action

acidosis, a potentially fatal complication of metformin.15 Short-acting human

insulin
Thiazolidinediones can increase intravascular volume,
Regular human 30–60 min 2–3 h 8–10 h
which can be problematic for inpatients with diabetes who insulin31
are predisposed to heart failure or who exhibit hemody- Intermediate-acting
namic changes, and can require weeks to show efficacy.15 insulin
Therefore, noninsulin agents should be discontinued in Neutral protamine 2–4 h 4–10 h 12–18 h
most hospitalized patients, with the possible exception Hagedorn31
of selected stable patients who are expected to regularly Long-acting insulin
consume meals.12 Detemir32,33 0.8–2 h Peakless Up to 24 h
31
Glargine 2–4 h Peakless 20–24 h
Intravenous Insulin Rapid-acting, mealtime
Continuous IV insulin infusion is the most effective bolus insulins
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015

method for controlling hyperglycemia in most critical-care Aspart31 5–15 min 30–90 min 4–6 h

patients.12 Indications for IV insulin include an NPO status Lispro31 5–15 min 30–90 min 4–6 h
34
in type 1 diabetes; general preoperative, intraoperative, Glulisine 15 min 30–90 min 5.3 h

and postoperative care; post-organ transplantation; total

parenteral nutrition therapy; labor and delivery; myocardial
infarction; and stroke. 12,15 Other indications for IV
insulin include hyperglycemic hyperosmolar syndrome,
diabetic ketoacidosis, and hyperglycemia during high-
dose glucocorticoid therapy. Although many hospitals do
not allow the use of IV insulin outside of the ICU, some
evidence suggests that it can be used safely.29 In either
setting, the administration of IV insulin should be directed
by validated protocols that specify predefined adjustments
in insulin infusion rates based on glycemic fluctuations.12
Because IV insulin protocols do not provide adequate
mealtime insulin coverage, subcutaneous administration of
a rapid-acting insulin analog at meals may be reasonable.30
Subcutaneous Basal-Bolus Insulin
Therapy
Scheduled basal-bolus therapy has numerous theoretical
and actual benefits among hospitalized patients and is the
preferred method for achieving and maintaining glycemic
control in patients with diabetes or hyperglycemia outside
of critical care.12 Scheduled inpatient subcutaneous insulin
consists of 3 components: basal insulin, nutritional insulin,
and supplemental, or correctional, insulin. Ideally, inpatient
subcutaneous insulin therapy should effectively mimic
insulin physiology in healthy individuals, covering both
basal and nutritional insulin needs with a minimal risk
of hypoglycemia. The onset, peak, and duration of action
of currently available subcutaneous insulin therapies are
described in Table 1.31–34
Basal Insulin
Basal insulin is used in inpatient insulin therapy to control
fasting hepatic glucose production and insulin-induced glucose
uptake in peripheral tissues. It accounts for approximately
50% of a patient’s total daily insulin requirement. A patient’s
insulin sensitivity is best defined by the daily basal insulin
dose. Basal and prandial doses of insulin should be adjusted
daily rather than scheduled as standing orders. Standing
basal insulin orders do not offer the dosing flexibility needed
to avoid both hypoglycemia and marked glucose elevations
associated with changes in the patient’s clinical condition
and calorie intake.
Insulin options that can be used to provide basal insulin
coverage among patients receiving subcutaneous insulin
include the long-acting basal insulin analogs insulin detemir
and insulin glargine, and the intermediate-acting insulin
NPH.15 However, NPH has several pharmacologic limitations
in outpatients, including variable onset of action, a relatively
short duration of action for true basal coverage, distinct and
unpredictable peak of action, and considerable intrapatient
variability in insulin action. Long-acting insulin analogs,
such as insulin detemir and insulin glargine, were developed
to more closely approximate normal basal insulin secretion.
As shown in Table 1, these agents have a duration of action
of up to 24 hours, enabling once-daily dosing without a
distinct peak in insulin action, and provide more reproducible
insulin action with less intrapatient variability.31,33,35–38 While
pharmacological data comparing NPH with long-acting

100 © Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331

insulin analogs among hospitalized patients are rare, the
pharmacokinetic and pharmacodynamic advantages of basal
insulin analogs observed among outpatients are also likely
to be apparent among inpatients.
Recent trials by Umpierrez et al26,27 have compared the use
of insulin analogs with human insulin regimens in the hospital
setting. Patients either on glargine/glulisine or detemir/aspart
were compared with patients on NPH/regular human insulin.
There was no significant difference in mean daily glucose
levels or the incidence of hypoglycemia in the detemir/aspart
study.26 The second multicenter study compared insulin
glargine/glulisine with NPH/regular human insulin following
the use of IV insulin to treat diabetic ketoacidosis. Between
groups there was no difference in the mean daily glucose,
although a significantly lower rate of hypoglycemia was seen
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
in those on glargine/glulisine compared with NPH/regular
insulin (15% vs 41%; P = 0.03).27 According to the Society of
Hospital Medicine, the dose of NPH should be decreased by
one-third to one-half in patients who switch to NPO status.39
Because NPH and regular human insulin are commonly used
in the acute-care setting, it is reassuring that the Umpierrez
studies26,27 demonstrated both the efficacy and safety of a
human insulin combination.
Nutritional Insulin
In the hospital, nutrition may be provided as parenteral or
enteral nutrition, calorie-dense supplements, or IV dextrose
along with discrete meals.15 Nutritional insulin in the hospital
is the quantity of insulin needed to cover each of these
nutritional sources. If patients are eating meals, nutritional
insulin is the same as prandial insulin.
The amount of prandial insulin used can be calculated
based on the amount of basal insulin that a patient requires.
Prandial insulin coverage can be given in doses that range
from 10% to 30% of the total basal dose. If a patient eats only
a portion of his or her meal, the prandial dose is often cut
by 50% or not given at all. For hospitals that prefer match-
ing prandial insulin to grams of carbohydrate consumed, an
insulin-to-carbohydrate ratio can be calculated by doubling
the basal dose used that day and dividing that number into
500.40 The result will provide the grams of carbohydrate
covered by 1 unit of insulin. This method has been used suc-
cessfully in 5 of our community-based acute-care hospitals
for patients with type 1 and type 2 diabetes. Such an approach
allows insulin management to be patient-specific and avoids
the pitfalls associated with sliding-scale treatment alone.
Options for nutritional insulin therapy include rapid-acting
insulin analogs (eg, insulin aspart, insulin lispro, and insulin
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331
Optimizing Inpatient Glycemic Control
glulisine) and short-acting regular human insulin. While
there are few data comparing the efficacy and tolerability of
short-acting regular human insulin with insulin analogs as
nutritional insulin, use of short-acting regular human insulin
as nutritional insulin has significant limitations. For example,
regular human insulin has a delayed onset of action, which
necessitates that it be injected at least 30 to 45 minutes before
eating.38 This requirement is especially inconvenient in the
hospital setting, when mealtimes may vary and the quantity of
food eaten is unpredictable. It also has a long duration of action
(5–8 hours), which can increase the risk of late postprandial
hypoglycemia.31,38 Theoretically, the rapid-acting insulin analogs
should enable dosing either pre- or immediately post-meal.
This allows for greater mealtime flexibility and prandial dose
adjustments based on the quantity of food ingested.
Correctional (Supplemental) Insulin
Correction-dose insulin refers to insulin used to supplement
the usual dose of mealtime insulin.15,41 Due to the more favor-
able pharmacologic profile of rapid-acting insulin analogs and
their ability to provide greater postprandial glycemic control
with less hypoglycemia compared with regular human insu-
lin, rapid-acting analogs should be used as the supplemental
insulin of choice.38,41 The correction dose for a rapid-acting
insulin analog is calculated using the “1800 Rule.” If the
total daily insulin dose is divided into 1800, the result is the
mg/dL drop one would expect by administering 1 unit of a
rapid-acting analog. For example, if a patient’s total daily
insulin dose is 50 U, it is expected that 1 unit of a rapid-acting
insulin analog would decrease blood glucose by 36 mg/dL
(1800/50). Therefore, if the patient had a blood glucose read-
ing of 210 mg/dL and blood glucose goal of  140 mg/dL
(a difference of 70 mg/dL), he or she would require 2 units
of insulin. The total daily dose can be predicted by counting
the total units required in the past 24 hours or by doubling
the basal dose chosen for the day in question.
Although correction-dose insulin is often confused with
sliding-scale insulin, they are quite different.15,41 Correction-
dose insulin is an important component of complete insulin
management and is used along with prandial and basal
insulin on an as-needed basis. Sliding-scale insulin is used
without scheduled insulin and without regard to mealtimes
or insulin sensitivity.
Sliding-Scale Insulin: A Recipe
for Poor Glycemic Control
In patients with diabetes or hyperglycemia outside of critical
care, subcutaneous therapy is the preferred method for
101
 R. Daniel Pollom
achieving and maintaining glycemic control.12 However,
the most commonly prescribed subcutaneous insulin regimen
among inpatients continues to be sliding-scale insulin,42 a
reactive approach to insulin therapy that is nonphysiologic
and has no evidence of effectiveness.
Sliding scales can result in rapid changes in blood glucose
levels that exacerbate hyperglycemia and promote hypogly-
cemia (Figure 1).15,43–45 The use of a one-size-fits-all sliding
scale makes no attempt to mimic normal physiology or pro-
vide for the normal basal insulin needs of a given patient.
Prandial insulin coverage matched to the patient’s intake of
carbohydrate is notably absent from such regimens.
Sliding scales without basal insulin coverage also con-
tain no provisions for variations in insulin resistance. In
patients with significant insulin deficiency, sliding-scale
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
insulin is more likely to result in significant hyperglycemia,
ketosis, ketoacidosis, or hypoglycemia.41 Moreover, orders
for sliding-scale insulin often do not consider the relation-
ship between meal intake, glucose excursions, and prandial
insulin requirements.41 Sliding scales used without prandial
insulin will invariably omit insulin if the glucose is normal
premeal, leading to recurrent postprandial hyperglycemia.46
Clinical Studies on Sliding-Scale Insulin
Although the theoretical and actual limitations of sliding-
scale insulin are widely reported, only a few studies have
investigated its efficacy. A prospective cohort study of admitted
adult patients reported that sliding-scale insulin was used in
76% of patients and did not improve overall glycemia during
Figure 1. Mean glycemic variability seen in a group of patients treated with sliding-scale insulin therapy. Reprinted with permission from Pharmacotherapy.45
350
300
Mean glucose level (mg/dl)
250
200
150
100
50
0
0 20 40
Hour of therapy
102
hospitalization.46 In this study, patients treated solely with
sliding-scale insulin had blood glucose levels  300 mg/dL
3 times more often than patients treated with other therapies.
Interestingly, in 80% of patients, diabetes orders prescribed
at admission were not modified during the patient’s hospital
stay, despite his or her poor glycemic control.
A second prospective cohort study reported that sliding-
scale insulin alone was associated with a mean glucose level
20 mg/dL higher per patient day compared with scheduled insulin
therapy when adjusted for clinical factors.16 A retrospective
observational analysis of 80 consecutive inpatients receiving
sliding-scale insulin therapy reported that only 12% of sliding-
scale insulin injections decreased blood glucose values to within
the target range, while 84% of injections were subtherapeutic.45
Despite the persistent elevations in blood glucose in this popu-
lation, neither the timing nor dose of the insulin was adjusted
in 81% of patients. Only 2% to 10% of patients in this study
achieved glycemic control using sliding-scale insulin.
Why Does Sliding-Scale Insulin Persist?
Although the clinical literature has discouraged the use of
sliding-scale insulin for  40 years,42 its use persists in both
academic and community medical centers.41 It continues to be
passed down from attending physicians to residents and from
residents to medical students, possibly due to tradition and its
simplicity and ease of use.41 Despite its convenience, not a single
study has shown that sliding-scale insulin without scheduled
insulin improves glycemic control or clinical outcomes com-
pared with basal-bolus subcutaneous insulin.42
60 80 100 120
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331

Efficacy of Basal-Bolus Versus
Sliding-Scale Insulin
Only a few clinical trials have compared the efficacy and
safety of scheduled basal-bolus with sliding-scale insulin
therapy. For example, in an early retrospective study of
47 patients admitted for ketoacidosis, patients treated with
sliding-scale insulin had a median blood glucose level of
262 mg/dL and length of stay of 6.3 days.47 In contrast,
patients treated with scheduled regular human insulin and
NPH therapy had a median blood glucose level of 200 mg/dL
and a length of stay of 4.4 days.
Sliding-scale insulin and scheduled basal-bolus insulin
therapy have also been compared prospectively in a random-
ized trial of 130 insulin-naïve patients with type 2 diabetes.48
In this study, patients were randomly assigned to receive
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
basal-bolus therapy with insulin glargine and insulin glulisine
or a standard sliding-scale insulin protocol. Patients random-
ized to sliding-scale insulin received regular human insulin
4 times daily for glucose levels  140 mg/dL.
Patients who received basal-bolus therapy experienced
significantly greater improvement in glycemic control
compared with those treated with sliding-scale insulin
(P  0.01), as shown in Figure 2.48 The mean difference in
blood glucose between groups was 27 mg/dL (P  0.01).
In this study, 66% of patients treated with basal-bolus therapy
achieved the mean glucose target of  140 mg/dL compared
with 38% of those treated with sliding-scale insulin. How-
ever, the mean daily dose of insulin was significantly higher
with the basal-bolus regimen compared with the sliding-scale
group. Despite maximal use of sliding-scale insulin doses
in this study, blood glucose remained  240 mg/dL in 14%
Figure 2. Changes in blood glucose concentrations among insulin-naïve inpa-
tients with type 2 diabetes treated with sliding-scale or basal-bolus insulin therapy.
Copyright 2007 American Diabetes Association. From Diabetes Care, Vol. 30, 2007;
2181–2186. Reprinted with permission from The American Diabetes Association.48
Basal bolus insulin
240
Sliding scale insulin
Blood glucose (mg/dL)
220
*
*
200 * †
†
† †
180
160
140
120
100
Admit 1 2 3 4 5 6 7 8 9 10
Days of therapy
*
P  0.01; †P  0.05.
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331
Optimizing Inpatient Glycemic Control
of patients treated with sliding-scale insulin. Once these
insulin-naïve patients were switched to basal-bolus insulin,
glycemic control rapidly improved (Figure 3).48
Despite the greater glycemic control observed with
basal-bolus therapy, the rates of hypoglycemia were low and
comparable among groups (0.4% of blood glucose readings
were  60 mg/dL with basal-bolus therapy compared with
0.2% with sliding-scale insulin). There were no blood glucose
assessments  40 mg/dL.48
Role of Standardized Subcutaneous
Insulin Order Sets/Protocols
Standardized subcutaneous insulin order sets are needed to
decrease an institution’s reliance on sliding-scale insulin.
These order sets must promote the use of scheduled insulin
therapy with basal insulin, prandial insulin, and correctional
insulin.49,50 Due to their more physiologic pharmacologic
profile compared with human insulin therapies, insulin
analogs should be the preferred agents in these order sets.1
Ideally, order sets should include provisions for special
patient circumstances, such as end-stage renal disease and
hepatic failure, and for those receiving steroids or enteral
nutrition.1 Oral or parenteral steroid use may require IV
insulin for optimal and timely glucose control. In such cases,
IV insulin is the most efficient dose-finding strategy and
eliminates days of excessive hyperglycemia. Subcutaneous
insulin protocols should also establish a target range
for blood glucose levels, provide recommendations for
standardizing blood glucose monitoring, coordinate blood
glucose testing with nutrition delivery and insulin admin-
istration, and prompt clinicians to consider discontinuing
oral antidiabetic agents.14 Safe and effective insulin orders
are reviewed daily and adjusted based on a patient’s glucose
response. Use of standing basal and prandial insulin subjects
patients to repeated episodes of hypoglycemia as well as
persistent hyperglycemia. All insulin orders should include
appropriate call orders so that treatment can be adjusted in
a timely fashion to prevent recurrent glucose elevations or
repeated episodes of hypoglycemia. Insulin protocols can
also be useful sources of diabetes education and serve as
a tool to help clinicians individualize treatment regimens
at discharge.
Impact of Standardized
Subcutaneous Insulin Protocols
Recent studies have evaluated the impact of initiating
detailed subcutaneous insulin protocols and order sets on
inpatient glycemic management.49,50
103
 R. Daniel Pollom
Figure 3. Mean blood glucose concentration in patients with blood glucose  240 mg/dL despite increasing doses of regular insulin using the sliding-scale protocol before
and after switching to basal-bolus insulin (P  0.05). Copyright 2007 American Diabetes Association. From Diabetes Care, Vol. 30, 2007;2181–2186. Reprinted with permis-
sion from The American Diabetes Association.48
300
280
260
Blood glucose (mg/dL)
240
220
200
180
160
140
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
120
100
Admit 1
In one before-and-after trial of 169 inpatients with diabetes
or hyperglycemia outside of the ICU, investigators evaluated
the impact of initiating a detailed subcutaneous insulin proto-
col, computerized admission order set, and case-based educa-
tion on glycemic control, insulin use, and length of stay.49 After
these measures were introduced, the fraction of mean blood
glucose readings between 60 and 180 mg/dL increased from
59% to 65%, length of stay decreased from 112.2 to 86 hours,
the use of nutritional insulin increased from 40% to 75%, and
the percentage of patients who used only sliding-scale insulin
decreased from 29% to 8% (P  0.05 for all changes).
A prospective observational study of 9314 non-ICU patients
with diabetes or documented hyperglycemia evaluated the
impact of structured insulin orders and an insulin management
algorithm on basal insulin use and glycemic control.50 In this
study, the number of uncontrolled patient stays (defined as a
day-weighted mean blood glucose  180 mg/dL) decreased
by 21% (from 4372 to 3465 days; P  0.005). The percentage
of patient days with hypoglycemia (defined as blood glucose
 60 mg/dL) decreased by 32%.
Recommendations for Special Situations
Some clinical situations and conditions, such as hyperglycemia
with glucocorticoid use and management of hyperglycemia in
patients using insulin pumps at admission, may be too complex
to be covered in a short order set and should be detailed in a
more comprehensive protocol.14
104
2 3 4 1 2 3 4 5 6 7
Days of therapy
Glucocorticoid-Induced Hyperglycemia
The association between glucocorticoid use and hyperglycemia
is well established. Although the impact of glucocorticoids
on blood glucose varies significantly from patient to patient
and depends on the type of steroid used and the frequency of
dosing, hyperglycemia induced by glucocorticoids character-
istically peaks 8 to 12 hours after administration.14 Because
glucocorticoids typically have their greatest effect on post-
prandial rather than fasting glucose, the best insulin regimens
for managing glucocorticoid-induced hyperglycemia may use
proportionally less basal insulin and more rapid-acting or nutri-
tional insulin.14 NPH administered twice daily may be useful
in treating the midday glucose elevations commonly seen with
daily morning doses of prednisone. When prednisone is dosed
at 8 AM, glucose levels tend to be at their highest from late
morning until early evening. NPH can be titrated in the morn-
ing to help reduce elevated glucoses at midday and adminis-
tered at a lower dose at the second daily dose to accommodate
the declining insulin needs, as the hyperglycemic effects of
the steroid effect wanes in late evening and through the night.
Patients on Insulin Pumps
Currently, there is no standard on how to best treat patients
with continuous subcutaneous insulin infusion (CSII), or
insulin pumps, in the hospital setting. The choice to continue
or discontinue CSII is complicated by the degree of illness
(and cognition) of the patient and his or her consequent ability
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331

to manage CSII, and the hospital staff’s level of comfort
with CSII. For these reasons, hospitals should have a written
policy on how best to manage patients using CSII. One
option is to allow patients who are able to manage their own
pumps to continue using them if a nurse records all insulin
dosing on the medical record. Other hospitals require referral
to an endocrinologist or endocrine team, or even demand
that patients remove the pump and use multiple injections
instead (often, unfortunately, with sliding-scale regimens).
If patients are required to stop using the pump, basal insulin
and multiple doses of nutritional insulin (preferably insulin
analogs) or IV insulin should be used as an immediate
replacement.
Practical Considerations
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
for Optimizing Inpatient
Glycemic Control
The American College of Endocrinology and the ADA
Task Force on Inpatient Diabetes have issued several
recommendations for optimizing inpatient glycemic
management.51 First, because hyperglycemia must be rec-
ognized before it is treated, the Task Force recommends
that elevated blood glucose be tested for in all hospitalized
patients. In addition, the Task Force suggests that using a
multidisciplinary team approach to diabetes management
enhances the recognition and treatment of hyperglycemia.
This multidisciplinary team should include medical staff,
nurses, case managers, pharmacists, dietitians, quality-
improvement staff, information systems personnel, and
administrators.
Preparing for Discharge
Optimal inpatient diabetes management also includes early
and proactive planning for a smooth transition to outpa-
tient care, with appropriate diabetes therapies and man-
agement strategies. Hospitalization offers a tremendous
opportunity to help patients improve the daily management
of their diabetes and achieve long-term improvements in
glycemic control. All patients requiring use of insulin in
the acute-care setting should have their HbA1c determined
at the time insulin treatment is initiated. The HbA1c will
identify those patients with a diagnosis of diabetes (HbA1c
 6.5) as opposed to those with stress hyperglycemia.
Elevated HbA1c levels on admission will help clinicians
target the population of patients with suboptimal out-
patient control. These patients can benefit from having
appropriate changes made to their outpatient medications
as well as referrals for outpatient teaching services.
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331
Optimizing Inpatient Glycemic Control
Cost-Effectiveness in Inpatient
Glycemic Control
In developing its updated consensus statement on inpatient
glycemic control, the ADA/AACE considered data on the
cost-effectiveness of inpatient glycemic control. Although
data on this topic are somewhat limited, preliminary
evidence suggests that controlling hyperglycemia can
provide cost savings.7,12,52,53 Cost savings with inpatient
glycemic control are primarily attributable to reductions
in laboratory, pharmacy, and radiology costs; inpatient
complications; ventilator days; lengths of stay in the ICU;
and overall lengths of stay in the hospital.12
Summary
Hyperglycemia is commonly seen among hospitalized
patients. Observational studies have shown that
hyperglycemia is associated with increased morbidity,
mortality, length of stay, and increased referrals to outpa-
tient extended-care facilities. Umpierrez et al2 found that
hyperglycemia was present in 38% of patients admitted to
the hospital, of whom one-third had no history of diabetes
before admission, and that newly discovered hypergly-
cemia was associated with a higher in-hospital mortality
rate and lower functional outcome not only in critically ill
patients admitted to the ICU but also in patients admitted
to general medicine or surgical wards.
Although sliding-scale insulin has been used for
decades, existing evidence suggests that it is ineffective
and may negatively affect outcomes in patients with hyper-
glycemia. This reactive approach to managing hypergly-
cemia does not include a basal insulin component, does
not account or control for nutritional intake, and should
be avoided.
Instead, subcutaneous basal-bolus insulin therapy is
preferred. Although there are few prospective, random-
ized, and controlled studies among inpatients comparing
basal-bolus therapy with sliding-scale insulin, current
evidence demonstrates that basal-bolus therapy provides
greater glycemic control compared with sliding-scale
insulin, with a low risk of hypoglycemia.
Several strategies have been recommended to enhance
inpatient glycemic control, including the use of a multi-
disciplinary team approach to diabetes management, the
implementation of standardized insulin order sets and
protocols, and education of all staff members involved in
the care of patients with hyperglycemia. Data indicate that
insulin protocols and order sets improve glycemic control,
optimize insulin therapy, and reduce length of hospital
105
 R. Daniel Pollom
stay. Additional research on the use of basal-bolus insulin
therapy to optimize clinical outcomes in noncritically ill
patients is needed.
Acknowledgments
The author wishes to thank Nicole Cooper of DesignWrite,
LLC for medical writing and editorial assistance. Funding
to support the preparation of this manuscript was provided
by Novo Nordisk, Inc.
Conflict of Interest Statement
R. Daniel Pollom, MD discloses conflicts of interest with
Eli Lilly, Merck, Novo Nordisk, Inc., Roche Diagnostics,
and sanofi-aventis US, Inc.
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
References
1. American Association of Clinical Endocrinologists, American Diabetes
Association. Inpatient diabetes and glycemic control: a call to action
conference. Available at: http://www.aace.com/meetings/consensus/
IIDC/IDGC0207.pdf. Accessed September 27, 2010.
2. Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi
AE. Hyperglycemia: an independent marker of in-hospital mortal-
ity in patients with undiagnosed diabetes. J Clin Endocrinol Metab.
2002;87(3):978–982.
3. American Diabetes Association. Economic costs of diabetes in the U.S.
in 2007. Diabetes Care. 2008;31(3):596–615.
4. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet.
2009;373(9677):1798–1807.
5. Krinsley JS. Association between hyperglycemia and increased hospital
mortality in a heterogeneous population of critically ill patients. Mayo
Clin Proc. 2003;78(12):1471–1478.
6. Pittas AG, Siegel RD, Lau J. Insulin therapy for critically ill hospitalized
patients: a meta-analysis of randomized controlled trials. Arch Intern
Med. 2004;164(18):2005–2011.
7. Furnary AP, Wu Y, Bookin SO. Effect of hyperglycemia and continuous
intravenous insulin infusions on outcomes of cardiac surgical procedures:
the Portland Diabetic Project. Endocr Pract. 2004;10(suppl 2):
21–33.
8. McAlister FA, Majumdar SR, Blitz S, Rowe BH, Romney J, Marrie TJ.
The relation between hyperglycemia and outcomes in 2,471 patients
admitted to the hospital with community-acquired pneumonia. Diabetes
Care. 2005;28(4):810–815.
9. Latham R, Lancaster AD, Covington JF, Pirolo JS, Thomas CS. The
association of diabetes and glucose control with surgical-site infections
among cardiothoracic surgery patients. Infect Control Hosp Epidemiol.
2001;22(10):607–612.
10. Turchin A, Matheny ME, Shubina M, Scanlon JV, Greenwood B,
Pendergrass ML. Hypoglycemia and clinical outcomes in patients with
diabetes hospitalized in the general ward. Diabetes Care. 2009;32(7):
1153–1157.
11. Baker EH, Janaway CH, Philips BJ, et al. Hyperglycaemia is associated
with poor outcomes in patients admitted to hospital with acute
exacerbations of chronic obstructive pulmonary disease. Thorax.
2006;61(4):284–289.
12. Moghissi ES, Korytkowski MT, Dinardo M, et al. American Association
of Clinical Endocrinologists and American Diabetes Association
consensus statement on inpatient glycemic control. Diabetes Care.
2009;32(6):1119–1131.
13. Wexler DJ, Meigs JB, Cagliero E, Nathan DM, Grant RW. Prevalence
of hyper- and hypoglycemia among inpatients with diabetes: a national
survey of 44 U.S. hospitals. Diabetes Care. 2007;30(2):367–369.
106
14. Maynard G, Wesorick DH, O’Malley C, Inzucchi SE. Subcutaneous
insulin order sets and protocols: effective design and implementation
strategies. J Hosp Med. 2008;3(5 suppl):S29–S41.
15. Clement S, Braithwaite SS, Magee MF, et al. Management of diabetes
and hyperglycemia in hospitals. Diabetes Care. 2004;27(2):553–591.
16. Schnipper JL, Barsky EE, Shaykevich S, Fitzmaurice G, Pendergrass ML.
Inpatient management of diabetes and hyperglycemia among general
medicine patients at a large teaching hospital. J Hosp Med. 2006;1(3):
145–150.
17. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy
in critically ill patients. N Engl J Med. 2001;345(19):1359–1367.
18. Malmberg K, Norhammar A, Wedel H, Rydén L. Glycometabolic
state at admission: important risk marker of mortality in convention-
ally treated patients with diabetes and acute myocardial infarction.
Long-term results from the diabetes and insulin-glucose infusion
in acute myocardial infarction (DIGAMI) study. Circulation.
1999;99(20):2626–2632.
19. Krinsley JS. Effect of an intensive glucose management protocol
on the mortality of critically ill adult patients. Mayo Clin Proc.
2004;79(8):992–1000.
20. Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional
glucose control in critically ill patients. N Engl J Med. 2009;360(13):
1283–1297.
21. Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy
and pentastarch resuscitation in severe sepsis. N Engl J Med.
2008;358(2):125–139.
22. Preiser JC, Devos P, Ruiz-Santana S, et al. A prospective randomised
multi-centre controlled trial on tight glucose control by intensive insulin
therapy in adult intensive care units: the Glucontrol study. Intensive Care
Med. 2009;35(10):1738–1748.
23. Wiener RS, Wiener DC, Larson RJ. Benefits and risks of tight glucose
control in critically ill adults: a meta-analysis. JAMA. 2008;300(8):
933–944.
24. Griesdale DE, de Souza RJ, van Dam RM, et al. Intensive insulin therapy
and mortality among critically ill patients: a meta-analysis including
NICE-SUGAR study data. CMAJ. 2009;180(8):821–827.
25. Skyler JS, Bergenstal R, Bonow RO, et al. Intensive glycemic control and
the prevention of cardiovascular events: implications of the ACCORD,
ADVANCE, and VA diabetes trials: a position statement of the American
Diabetes Association and a scientific statement of the American College
of Cardiology Foundation and the American Heart Association. Diabetes
Care. 2009;32(1):187–192.
26. Umpierrez GE, Hor T, Smiley D, et al. Comparison of inpatient insulin
regimens with detemir plus aspart versus NPH plus regular in medical
patients with type 2 diabetes. J Clin Endocrinol Metab. 2009;94(2):
564–569.
27. Umpierrez GE, Jones S, Smiley D, et al. Insulin analogs versus human
insulin in the treatment of patients with diabetic ketoacidosis: a
randomized controlled trial. Diabetes Care. 2009;32(7):1164–1169.
28. American Diabetes Association. Standards of medical care in
diabetes—2010. Diabetes Care. 2010;33(suppl 1):S11–S61.
29. Vinik R, Jones RE, Pendleton RC, Ku SY. Safety and efficacy of an
insulin infusion protocol designed for the non-intensive care setting.
Endocr Pract. 2009;15(7):682–688.
30. Smiley D, Rhee M, Peng L, et al. Safety and efficacy of continuous
insulin infusion in noncritical care settings. J Hosp Med. 2010;5(4):
212–217.
31. Hirsch IB. Insulin analogues. N Engl J Med. 2005;352(2):174–183.
32. 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(5):648–659.
33. Plank J, Bodenlenz M, Sinner F, et al. A double-blind, randomized,
dose-response study investigating the pharmacodynamic and pharma-
cokinetic properties of the long-acting insulin analog detemir. Diabetes
Care. 2005;28(5):1107–1112.
34. Levy P. Insulin analogs or premixed insulin analogs in combination with
oral agents for treatment of type 2 diabetes. Med Gen Med. 2007;9(2):12.
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331

35. Heise T, Nosek L, Ronn BB, et al. Lower within-subject variability of
insulin detemir in comparison to NPH insulin and insulin glargine in
people with type 1 diabetes. Diabetes. 2004;53(6):1614–1620.
36. Heinemann L, Linkeschova R, Rave K, Hompesch B, Sedlak M, Heise T.
Time-action profile of the long-acting insulin analog insulin glargine
(HOE901) in comparison with those of NPH insulin and placebo. Diabetes
Care. 2000;23(5):644–649.
37. Owens DR, Coates PA, Luzio SD, Tinbergen JP, Kurzhals R.
Pharmacokinetics of 125I-labeled insulin glargine (HOE 901) in healthy
men: comparison with NPH insulin and the influence of different
subcutaneous injection sites. Diabetes Care. 2000;23(6):813–819.
38. DeWitt DE, Hirsch IB. Outpatient insulin therapy in type 1 and type 2
diabetes mellitus: scientific review. JAMA. 2003;289(17):2254–2264.
39. Wesorick D, O’Malley C, Rushakoff R, Larsen K, Magee M. Management
of diabetes and hyperglycemia in the hospital: a practical guide to sub-
cutaneous insulin use in the non-critically ill, adult patient. J Hosp Med.
2008;3(5 suppl):S17–S28.
40. Walsh J, Roberts R. Pumping Insulin: Everything You Need for Success on
a Smart Insulin Pump. 4th ed. San Diego, CA: Torrey Pines Press; 2006.
41. Hirsch IB. Sliding scale insulin—time to stop sliding. JAMA.
Downloaded by [University of Pennsylvania] at 10:46 09 November 2015
2009;301(2):213–214.
42. Umpierrez GE, Palacio A, Smiley D. Sliding scale insulin use: myth or
insanity? Am J Med. 2007;120(7):563–567.
43. Levetan CS, Magee MF. Hospital management of diabetes. Endocrinol
Metab Clin North Am. 2000;29(4):745–770.
44. Browning LA, Dumo P. Sliding-scale insulin: an antiquated approach
to glycemic control in hospitalized patients. Am J Health Syst Pharm.
2004;61(15):1611–1614.
45. Golightly LK, Jones MA, Hamamura DH, Stolpman NM, McDermott MT.
Management of diabetes mellitus in hospitalized patients: efficiency
and effectiveness of sliding-scale insulin therapy. Pharmacotherapy.
2006;26(10):1421–1432.
© Hospital Practice, Volume 38, Issue 4, November 2010, ISSN – 2154-8331
Optimizing Inpatient Glycemic Control
46. Queale WS, Seidler AJ, Brancati FL. Glycemic control and sliding
scale insulin use in medical inpatients with diabetes mellitus. Arch
Intern Med. 1997;157(5):545–552.
47. Gearhart JG, Duncan JL 3rd, Replogle WH, Forbes RC, Walley EJ.
Efficacy of sliding-scale insulin therapy: a comparison with prospective
regimens. Fam Pract Res J. 1994;14(4):313–322.
48. Umpierrez GE, Smiley D, Zisman A, et al. Randomized study of
basal-bolus insulin therapy in the inpatient management of patients
with type 2 diabetes (RABBIT 2 trial). Diabetes Care. 2007;30(9):
2181–2186.
49. Schnipper JL, Ndumele CD, Liang CL, Pendergrass ML. Effects
of a subcutaneous insulin protocol, clinical education, and
computerized order set on the quality of inpatient management of
hyperglycemia: results of a clinical trial. J Hosp Med. 2009;4(1):16–27.
50. Maynard G, Lee J, Phillips G, Fink E, Renvall M. Improved inpa-
tient use of basal insulin, reduced hypoglycemia, and improved
glycemic control: effect of structured subcutaneous insulin orders
and an insulin management algorithm. J Hosp Med. 2009;4(1):3–15.
51. ACE/ADA Task Force on Inpatient Diabetes. American College of
Endocrinology and American Diabetes Association consensus statement
on inpatient diabetes and glycemic control. Endocr Pract. 2006;12(4):
458–468.
52. van den Berghe G, Wouters PJ, Kesteloot K, Hilleman DE.
Analysis of healthcare resource utilization with intensive insulin
therapy in critically ill patients. Crit Care Med. 2006;34(3):612–616.
53. Krinsley JS, Jones RL. Cost analysis of intensive glycemic control in
critically ill adult patients. Chest. 2006;129(3):644–650.
107