Moghissi

Hospital Management of Diabetes
Etie S. Moghissi, MD, FACP, FACEa,*,
Irl B. Hirsch, MDb,c,*
a
Inpatient Diabetes and Metabolic Control Task Force,
American Association of Clinical Endocrinologists, 501 East Hardy Street,
Suite 110, Inglewood, CA 90301, USA
b
Division of Metabolism, Endocrinology, and Nutrition, Department of Medicine,
University of Washington Medical Center, 1959 NE Paciﬁc Street,
Box 356176, Seattle, WA 98195, USA
c
Diabetes Care Center, University of Washington Medical Center,
Seattle, WA, USA
Diabetes remains a major cause of death and disability and is a growing
global concern. The age-adjusted prevalence of type 2 diabetes in the United
States increased by more than 55% from 1990 to 2000 [1]. It is estimated
that one out of three individuals born in 2000 will develop diabetes in their
lifetime [2].
Chronic complications of diabetes, especially cardiovascular disease,
result in hospitalization in many patients with diabetes. In 2001, more than
4.6 million hospitalizations were associated with diabetes, accounting for
nearly 17 million hospital days at a cost of over $40 billion [3].
Mounting observational and interventional evidence consistently indi-
cates that hyperglycemia in the hospital setting is associated with increased
mortality and morbidity and that meticulous glycemic control can improve
clinical outcomes [4–7]. The purpose of this article is to review the evidence,
discuss the importance of striving for good glycemic control in the hospital
setting, and emphasize the need for additional outcome research studies to
further examine the currently recommended in-hospital glycemic guidelines.
Dr. Morghissi is the former Vice-Chair of the Statement on Inpatient Diabetes and
Metabolic Control writing panel of the American Association of Clinical Endocrinologists.
* Corresponding authors. Division of Metabolism, Endocrinology, and Nutrition,
Department of Medicine, University of Washington Medical Center, 1959 NE Pacific Street,
Box 356176, Seattle, WA 98195 (I.B. Hirsch).
E-mail addresses: emoghissi@pol.net (E.S. Moghissi); ihirsch@u.washington.edu (I.B.
Hirsch).
0889-8529/05/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.ecl.2004.11.001 endo.theclinics.com
Endocrinol Metab Clin N Am
34 (2005) 99–116

The article concludes with a discussion of strategies for achieving tight
glycemic targets.
Hyperglycemia in the hospital setting
Hyperglycemia is a frequent ﬁnding in hospitalized patients. With the
escalating incidence of diabetes in the United States, the number of patients
with diabetes who require hospitalization has also increased. Between 1980
and 2001, the age-adjusted hospital discharge rate for diabetes as any-listed
diagnosis in the general population increased over 49% from 109.0 to 162.4
per 10,000 in the general population [8]. A rapidly growing body of evidence
clearly suggests that meticulous glycemic control improves clinical out-
comes, including reduced mortality and infection rates [4–7]; however, until
recently there were no established guidelines or standards for management
of diabetes in the hospital setting.
In a recent consensus conference cosponsored by the American
Association of Clinical Endocrinologists, American College of Endocrinol-
ogy, American Diabetes Association, Endocrine Society, American Asso-
ciation of Diabetes Educators, American Heart Association, American
Society of Anesthesiologists, Society of Critical Care Medicine, Society of
Hospital Medicine, and Society of Thoracic Surgeons, a position statement
was developed by the American College of Endocrinology with recom-
mendations for glycemic targets in the hospital setting [9]. During the 2-day
conference, experts reviewed extensive data and deliberated regarding the
most appropriate targets for glycemic control in the hospital setting [9]. The
upper limits of normal for glycemic targets are shown in Box 1.
Role of intravenous insulin therapy in critically ill patients
Hyperglycemia is very common in critically ill patients in the intensive
care unit (ICU), even among patients with no prior history of diabetes. This
Box 1. Upper limits for glycemic targets
Intensive care unit
110 mg/dL (6.1 mmol/L)
Noncritical care units
Preprandial
110 mg/dL (6.1 mmol/L)
Maximal glucose
180 mg/dL (10.0 mmol/L)
Data from American College of Endocrinology. Position statement on inpatient
diabetes and metabolic control. Endocr Pract 2004;10:77–82.
100 MOGHISSI & HIRSCH

‘‘stress hyperglycemia’’ is believed to be due to insulin resistance in the liver
and the skeletal muscle. Until recently, it was felt that mild to moderate
degrees of hyperglycemia was inconsequential or perhaps even adaptive to
ensure there was enough substrate to provide enough energy.
Van den Berghe and colleagues [5] challenged this notion in a prospective,
randomized control study. They studied 1548 mechanically ventilated adults
who were admitted to a surgical ICU. These patients were randomized to
receive either intensive insulin therapy with target blood glucose of 80 to 110
mg/dL or conventional therapy to maintain blood glucose between 180 and
200 mg/dL; insulin infusion was initiated in the conventional group if blood
glucose exceeded 215 mg/dL. The primary outcome measure was all-cause
mortality in the ICU. Secondary outcome measures included in-hospital
mortality, duration of ICU stay, and need for ICU ventilator support for
more than 14 days.
The intensive insulin therapy group showed a significant reduction in
ICU mortality compared with the conventional therapy group, 4.6% and
8% respectively (P 0.04). The mortality reduction was only evident in
patients who stayed in the ICU more than 5 days, 10.6% versus 20.2%
(P = 0.005). In addition, the intensive therapy group had significant
reductions in in-hospital mortality (ÿ34%), sepsis (ÿ46%), acute renal
failure requiring dialysis or hemofiltration (ÿ41%), rate of transfusion
(ÿ50%), and polyneuropathy (ÿ44%).
Impact of hyperglycemia and intravenous insulin infusion on the outcome of
myocardial infarction
More than 80% of deaths associated with diabetes are from cardiovas-
cular disease, of which 75% are a result of coronary artery disease [10].
Although the mortality from coronary artery disease has declined in the
general population, this decline has not been as significant in individuals with
diabetes. Insulin resistance and abnormal glucose tolerance is even more
common in patients with acute coronary syndrome as evidenced by a recent
study showing that only one third of all patients had a normal glucose
tolerance test 3 months after the acute event [11]. These results are consistent
with a 2002 prospective study by Norhammar and colleagues [12], who
looked at outcomes in 181 consecutive patients admitted to the coronary
care units of two hospitals in Sweden with acute myocardial infarction but
no diagnosis of diabetes and a blood glucose concentration of less than
200 mg/dL (11.1 mmol/L). Subjects underwent standardized oral glucose
tolerance tests (75-g glucose load) at discharge and again 3 months later.
Fifty-eight of 164 (35%) and 58 of 144 (40%) individuals had impaired
glucose tolerance at discharge and after 3 months, respectively, whereas 51 of
164 (31%) and 36 of 144 (25%) had previously undiagnosed diabetes
mellitus.
101
DIABETES: HOSPITAL MANAGEMENT

It is well established that the short- and long-term prognosis of diabetic
patients sustaining acute myocardial infarction is poor and is associated with
increased mortality and morbidity [13]. A meta-analysis of 15 studies reported
that hyperglycemia (glucose > 110 mg/dL), with or without a prior history of
diabetes, is associated with increases in in-hospital mortality and congestive
heart failure in patients admitted for acute myocardial infarction [14].
The DIGAMI study demonstrated that the unfavorable long-term
prognosis could be improved by insulin treatment [4], extending findings
from previous reports [15–17]. In the DIGAMI study, 620 patients with
acute myocardial infarction and hyperglycemia (with and without prior
history of diabetes) were randomized to either intravenous (IV) insulin-
glucose infusion for at least 24 hours, followed by multidose subcutaneous
insulin treatment for at least 3 months or to a control group, which received
conventional treatment that generally included sulfonylurea therapy. The
conventional treatment group was used as the control arm; therapy was left
to the discretion of the treating physicians. Standard therapy for acute
myocardial infarction was applied to all subjects. The baseline character-
istics were similar between the two groups. The primary end point of the
study was 3-month mortality; the secondary end point was 1-year mortality.
Results showed that the mortality at 1 year was significantly reduced in
the infusion group (19%) compared with the control group (26%). The
long-term results (3.4 years of follow-up) showed a persistent relative
mortality reduction of 25% (P = 0.011) in the insulin-treated group, which
corresponds to an absolute mortality reduction of 11%. Risk reduction was
even more significant in patients who were considered ‘‘low risk’’: a group of
272 patients with no prior history of insulin therapy. In this group, there was
a 58% reduction in mortality at discharge (P 0.05), a 50% reduction at 12
months, and a 45% reduction at 3.4 years (P = 0.004).
Impact of insulin therapy on nonglycemic parameters of metabolic control
Insulin promotes glucose oxidation, which is beneficial in ischemic
situations [18]. Increased levels of circulating free fatty acids (FFAs) are
a common finding in a setting of myocardial ischemia [19]. FFA oxidation is
detrimental to the myocardium because of an increased oxygen demand and
to a direct inhibition of glucose oxidation [20]. In addition, increased FFA
use during ischemia causes accumulation of FFA metabolites. These in turn
are toxic to the myocardium and can provoke arrhythmia and exacerbate
mechanical dysfunction [19]. Furthermore, the Paris Prospective Study
reported FFAs as a predictor of sudden death [21], and the infusion of FFAs
can induce ventricular fibrillation [19,22].
Proinflammatory cytokines are also likely involved in the morbidity and
mortality of severe illness [23,24]. These cytokines appear to promote acute
thrombosis, sepsis, heart failure, the cachexia of malignancy, and are
102 MOGHISSI & HIRSCH

involved in the pathogenesis of atherosclerosis. The regulation of these
cytokines (and the acute phase proteins they modulate) is complex and not
fully understood. However, one fact remains clear: exogenous insulin, when
infused to normal or near-normal blood glucose levels, can inhibit some of
these macrophage and monocyte products [25–27].
There are two other possible mechanisms through which insulin may
independently improve outcomes in critically ill patients. First, it suppresses
the growth factors involved in acute thrombosis, which may be important
during acute myocardial infarction [28]. Second, it stimulates endothelial
nitric oxide synthase, which results in the synthesis of nitric oxide and
potentially results in vascular vasodilation [29].
Impact of hyperglycemia and intravenous insulin infusion on cardiac
surgery outcome
The prevalence of diabetes among patients undergoing coronary artery
bypass graft (CABG) is as high as 28%, and diabetes is an independent risk
factor for CABG-related death [30]. In addition, postoperative hyperglyce-
mia (200 mg/dL) is a predictor of infectious complications in patients
undergoing coronary artery surgery [31] and is associated with surgical-site
infection [32]. In a study of 1000 patients undergoing cardiothoracic surgery
at a large university hospital setting, hyperglycemia in the first 48 post-
operative hours was associated with a twofold higher rate of surgical-site
infection compared with the normoglycemic group [32].
It has been shown that IV insulin during the postoperative period im-
proves outcome. Zerr and colleagues [33] were able to demonstrate that use
of IV insulin in the first 3 postoperative days significantly reduces morbidity
in cardiac surgery patients, specifically in reducing deep sternal wound
infection. This retrospective study involved 1585 patients with diabetes who
underwent cardiac surgery between 1987 and 1993. The study looked at
the rate of deep wound infection in patients treated before and after
September 1991, when a protocol of postoperative continuous IV insulin
to maintain a blood glucose level of less than 200 mg/dL was initiated.
Patients treated with IV insulin infusion to maintain glucose levels below
200 mg/dL had a significantly lower incidence (P 0.02) of deep wound
infection, from 2.8% before the implementation of IV insulin therapy to
0.74% the third year after the implementation. In a subsequent publication,
Furnary and colleagues [34] reported that continuous insulin infusion
added an independently protective effect against death (odds ratio 0.50,
P = .005) to the constellation of risk factors described in the Society of
Thoracic Surgeons risk model [30]. They concluded that diabetes per se is
not a true risk factor for death after CABG; rather, it is the underlying
glycometabolic state of the myocardium that independently affects post-
operative mortality.
103

Relationship between glycemic control and stroke outcome
Observational studies suggest a correlation between blood glucose level,
mortality and morbidity, and functional recovery in patients who have
suffered a stroke. Capes and colleagues [35] performed a meta-analysis of 26
studies published between 1996 and 2000 and reported that in patients with
no history of diabetes who had an ischemic stroke, even a moderately
elevated glucose level ([110 mg/dL) is associated with a threefold higher
risk of in-hospital or 30-day mortality and an increased risk of poor func-
tional recovery compared with those with lower glucose levels. This meta-
analysis was limited by several factors, including the use of pooled studies
with different inclusion and exclusion criteria, definitions of hyper-
glycemia, and concomitant treatment; nevertheless, the strong association
between admission hyperglycemia and poor prognosis suggests that glucose
level is an important prognostic factor for morbidity and mortality after
stroke.
Other studies suggest similar correlation between blood glucose level and
mortality and morbidity in stroke patients [36,37]. These observational
studies highlight the need for interventional controlled trials to investigate
the impact of targeted glycemic control on the outcome of acute stroke both
in patients with and without a known history of diabetes.
Hyperglycemia and infection
The association between hyperglycemia and infection has long been
recognized [38,39]. Pomposelli and coworkers [38] found that early post-
operative glucose control predicts nosocomial infection in diabetic patients.
In this study, 97 patients with diabetes undergoing surgery were studied. On
postoperative day 1, a single glucose level of more than 220 mg/dL was
shown to be a sensitive predictor of nosocomial infection at a rate 2.7 times
higher than in patients who had blood glucose levels below 220 mg/dL. In
a separate evaluation that excluded patients with minor urinary tract infec-
tion from the analysis, the rate of severe infection such as sepsis, pneumonia,
and wound infection was increased to 5.7%.
In 1499 patients studied between 1987 and 1994, Furnary and colleagues
[6] found that hyperglycemia in the first 48 hours following surgery was an
independent risk factor for deep sternal wound infection; patients with
glucose levels above 200 mg/dL had a risk 2.2 times higher than those
patients with glucose levels of less than 200 mg/dL. Control of hyper-
glycemia with IV insulin in these patients resulted in a 66% reduction of this
serious complication to a rate similar to that of nondiabetic subjects.
The association between hyperglycemia and increased infection may be
attributed to impaired immune function. Phagocyte dysfunction appears to
be the primary problem, including impaired adherence, chemotaxis, phago-
104 MOGHISSI HIRSCH

cytosis, bacterial killing, and respiratory burst [40–50]. Other abnormalities
of impaired immune function associated with hyperglycemia include
nonenzymatic glycation of immunoglobulins and reduced T-lymphocyte
populations [51]. The effect on peripheral lymphocytes associated with
hyperglycemia is reversed when glucose levels are lowered [52].
Good glycemic control reduces cost and length of hospital stay
Although the association between hyperglycemia and adverse outcomes
is well supported [53–56], prospective studies for using glucose reduction as
a means of reducing length of hospitalization are few, and cost-effectiveness
analyses are even fewer in number.
Several studies have shown a relationship between improved inpatient
glucose control and decreased hospital length of stay [34,57–62]. Furnary
and colleagues [62] showed that length of stay was increased by 1 day for
each 50-mg/dL increase above 150 mg/dL in the mean glucose during the
first 3 postoperative days (P 0.001). An intensive IV insulin regimen was
associated with 23% reduction in length of hospital stay. This resulted in
a net savings of more than $680 per patient.
In a nonrandomized, retrospective review, Levetan and associates [58]
reported that diabetes team consultation resulted in a 56% shorter
hospitalization than those patients with no consultation and proposed
a cost saving of $2353 per patients who were managed by the diabetes team.
From the studies available, the cost benefit of improved glycemic control
appears to be apparent; however, more rigorous analyses of cost effective-
ness are warranted.
The use of intravenous insulin in the hospital
In the ICU, the data support the use of IV insulin both for critically ill
patients (especially those who have had recent surgery) and for those with
hyperglycemia during an acute myocardial infarction [9,63]. There is also
general consensus that IV insulin should be used for metabolically unstable
patients with widely fluctuating blood glucose levels, regardless of their
location in the hospital [9,63]. However, IV insulin infusions are infre-
quently used on general medical and surgical wards. Typical ‘‘sliding scale’’
insulin is not effective in the manner that it is typically used [64], and yet it is
part of our medical culture [65].
IV insulin infusions have been used to manage diabetes for more than 20
years and have a proven record of safety and efficacy [66]. An insulin drip
has more predictable bioavailability, is easier and quicker to titrate, and is
safer than subcutaneous insulin in patients who are not eating or who may
be suddenly switched to no-eating status.
105

There are a wide variety of insulin infusion protocols, but there are no
studies in which algorithms are compared with one another. It is apparent
that protocols have to be adapted to each specific hospital environment and
that no one protocol would be likely to be effective at all hospitals. As an
example, Appendix 1 presents the protocol used by the author at the
University of Washington Medical Center in Seattle, Washington [67].
Several aspects of the IV insulin protocol need to be emphasized. First,
there must be enough bedside glucose testing to ensure optimum patient
safety. The ideal frequency of glucose testing has not been systematically
studied. At the University of Washington, we generally initiate the protocol
with hourly glucose measurements, at least until the blood glucose is stable
[67]. Next, the protocol should include some mechanism for changing
infusion rate if there have been significant alterations in blood glucose.
Some of the most widely used protocols have this feature [68,69].
Next, enough glucose needs to be provided to prevent both hypoglycemia
and starvation ketosis. Again, the ideal amount has not been systematically
studied, but most authors suggest 5 to 10 g each hour. If ketonuria occurs
despite well-controlled blood glucose levels, the amount of glucose needs to
be increased.
The University of Washington Medical Center is a 400-bed teaching
hospital where numerous IV insulin infusion protocols have been used since
1991. Each service had their own protocol, but in July of 2002 the insulin
infusion algorithms were all standardized to a protocol adapted from the
Markovitz report (see Appendix 1) [68]. Instead of introducing the protocol
throughout the entire hospital at the same time, each nursing unit was
transitioned sequentially until they were comfortable with the new protocol.
This took over a year to accomplish, but we found satisfaction from nurses
and physicians to be high, and the protocol was found to be quite safe [70].
Compared with the old protocol, which resulted in over 40% of blood
glucose values above 180 mg/dL, we found only 16.7% of blood glucose
levels greater than this value. Likewise, hypoglycemia was rare with the
new protocol. Only 3% had one blood glucose measurement below
40 mg/dL, and 16% had one reading below 60 mg/dL. On the old proto-
col, this compares to 14% and 30% below 40 mg/dL and 60 mg/dL,
respectively.
It is clear that for these types of protocols to work effectively, there
must be an understanding of the general philosophies of insulin therapy
among endocrinologists, cardiologists, surgeons, anesthesiologists, primary
care physicians, nurses, and hospital pharmacists. Often a significant
amount of education will be required for those not very familiar with
insulin therapy.
According to the American College of Endocrinology position statement
on inpatient diabetes and metabolic control [9], implementation of these
protocols requires assessment of hospital systems for safety and quality of
care; adjustments may be required for appropriate provision of diabetes care,
106 MOGHISSI HIRSCH

including timely delivery of meal trays, point-of-care blood glucose testing,
and the administration of diabetes medications. It is also recommended that
nursing staff receive adequate and ongoing in-service training on the spe-
cialized needs of the inpatient with diabetes, especially with regard to insulin
therapy. A team approach to implementation is also recommended. In addi-
tion to the physician, the team may include specialty staff such as a certified
diabetes educator. Diabetes educators and nursing staff should collaborate in
providing basic ‘‘survival skills’’ when needed to allow for a safe discharge.
Finally, discharge planning should be initiated well in advance, exploring
community resources and arranging for follow-up of diabetic issues.
The use of subcutaneous insulin therapy in the hospital
When addressing subcutaneous (SC) insulin therapy, it is important that
the definitions of insulin therapy are appreciated. Differences between
physiologic and nonphysiologic insulin replacement, basal and prandial
insulin replacement, and the difference between a supplement and an
adjustment need to be clear [66]. The insulin supplement (or ‘‘correction
dose’’) needs to be differentiated from sliding scale insulin [34,66].
Neutral protamine Hagedorn and regular insulin have been traditionally
used for basal and prandial glucose coverage, respectively. We believe that
the use of current insulin analogs minimizes hypoglycemia and increases
flexibility [71]. It is important to appreciate that even though studies have
not been performed, rapid-acting analogs (lispro or aspart) should
theoretically be more effective and safer as correction dose insulin than
regular insulin. This is because their more rapid onset and shorter duration
of action makes them more predictable. By the same token, it is our opinion
that, although using a pure basal insulin (eg, insulin glargine) to treat acute
hyperglycemia will result in failure, glargine does provide a much more
predictable basal delivery of insulin compared with neutral protamine
Hagedorn or ultralente insulin.
Appendix 1 presents a sample SC order form that was developed and
implemented at the University of Washington Medical Center to improve
safety and efficacy of SC insulin [67]. Compared with the IV insulin infusion
protocols, there is much less experience with these SC protocols. However,
our protocol has resulted in several advantages, including some that were
unexpected (Appendix 2). First, like all hospital protocols, having a uniform
SC insulin algorithm results in standardization of therapy. Any problems
can be rectified and there can be immediate quality improvements in the
entire hospital system. Second, the protocol is designed to improve glucose
control to better match the new guidelines [9]. Third, and perhaps most
importantly, an SC insulin protocol that differentiates basal insulin
replacement, prandial insulin replacement, and correction dose insulin can
educate both younger and older physicians in current strategies for insulin
107

use [71]. This last benefit was unexpected. Over the years there have been
numerous attempts to better educate physicians about both the basics and
the subtleties of insulin therapy, but surveys have reported that physicians
do not have a good understanding of how to best use insulin, especially in
the United States [72,73]. By using an SC protocol for hospitalized patients
on a daily basis, physicians learn how to adjust insulin appropriately,
including the best use of correction dose insulin. We believe that in relatively
short amounts of time, medical and surgical residents of all specialties could
become quite knowledgeable about the use of SC insulin by simply working
with the protocol for their hospitalized patients. Only prospective trials will
be able to confirm these beliefs.
Summary
The evidence continues to strengthen our understanding that improved
glycemic control with the use of insulin therapy may significantly improve
morbidity and mortality in hospitalized patients with hyperglycemia, with or
without a previous diagnosis of diabetes. However, many questions remain
concerning the impact and relative contributions of blood glucose and
insulin per se. Nevertheless, the publication of numerous and consistent
studies have made it clear that the topic of glycemic management in the
hospital requires a larger priority among clinicians caring for these patients.
The recently published guidelines by the American Association of Clin-
ical Endocrinologists are the first formal recommendations on this topic [9],
but national guidelines for blood glucose levels cannot take into account all
of the different challenges facing different hospitals. This suggests that each
institution will require individualization of protocols even though the
ultimate metabolic goals are identical. Furthermore, it is not realistic to
expect those unfamiliar with diabetes therapy to appreciate all of the nuances
and vagaries of insulin treatment. Like any medical treatment, a significant
amount of time will need to be invested by the providers involved with the
care of these patients before a mastery of the therapy can be achieved.
Nevertheless, because the rewards to our patients can be significant, we need
to strive to improve the systems where we work. Individual clinicians with
vast experience in diabetes care cannot be successful for the inpatient with
diabetes unless the hospital has systems in place to effectively and efficiently
facilitate the management of the metabolic needs of this population.
The main challenge now is the safe and effective implementation of these
guidelines in both small and large hospitals given the limited level of re-
sources available in today’s medical environment. Therefore, our single most
important recommendation is to ensure that all clinicians involved in the
management of these patients are in agreement about general philosophies of
diabetes management. We would recommend that there are ‘‘champions’’ for
each discipline: endocrinology, cardiology, anesthesiology, surgery, nursing,
108 MOGHISSI HIRSCH

and pharmacy, all of which have developed hospital-specific guidelines for
glycemic management. These recommendations can be slowly adapted, one
unit at a time, until the entire hospital has transitioned to a more ‘‘diabetes-
friendly’’ environment. The ultimate goal of well-controlled glycemia with
minimal hypoglycemia should be possible for most hospitals, and we hope
this review will assist clinicians in achieving this objective.
We await additional outcome research with carefully controlled studies to
confirm the value of these recommendations at different levels of glycemic
control. We believe that we can already state with confidence that the
preliminary evidence shows that, like outpatient diabetes management,
metabolic control matters during acute illness.
Appendix 1. Example of a standardized intravenous insulin infusion
General guidelines
Goal blood glucose level: usually 80 to 180 mg/dL (80 to 110 mg/dL for
the ICU)
Standard drip: 100 U/100 mL 0.9% NaCl by way of an infusion device
(1 U/1 cc)
Surgical patients who have received an oral diabetes medication within
24 hours should start when blood glucose level is above 120 mg/dL.
All other patients can start when blood glucose level is 70 mg/dL or
more
Insulin infusions should be discontinued when a patient is eating and
has received first dose of subcutaneous insulin
Intravenous fluids
Most patients will need 5 to 10 g of glucose per hour
D5W or D5W1/2NS at 100 to 200 mL per hour or equivalent (total
parenteral nutrition, enteral feeds, and so forth)
Initiating the infusion
Algorithm 1: start here for most patients.
Algorithm 2: for patients not controlled with algorithm 1, or start here if
status post (s/p) CABG, s/p solid organ transplant or islet cell
transplant, receiving glucocorticoids, or patient with diabetes receiving
more than 80 U/d of insulin as an outpatient
Algorithm 3: for patients not controlled with algorithm 2; no patients
start here without authorization from the endocrine service
Algorithm 4: for patients not controlled with algorithm 3; no patients
start here
109

Patients not controlled with the above algorithms require an endocrine
consult.
Moving from algorithm to algorithm
Moving up: an algorithm failure is defined as a blood glucose level
outside the goal range (see above goal), and the blood glucose does not
change by at least 60 mg/dL within 1 hour
Moving down: when blood glucose level is less than 70 mg/dL 2
Patient monitoring
Goal blood glucose level: 80 to 180 mg/dL
Check capillary blood glucose level every hour until it is within goal
range for 4 hours, then decrease to every 2 hours for 4 hours; if it
remains stable, decrease to every 4 hours
Hourly monitoring may be indicated for critically ill patients even if they
have stable blood glucose
Treatment of hypoglycemia (blood glucose level 60 mg/dL)
Discontinue insulin drip AND
Give D50W IV (patient awake, 25 mL [1/2 amp]; patient not awake, 50
mL [1 amp])
Recheck blood glucose level every 20 minutes and repeat 25 mL of
D50W IV if it is below 60 mg/dL; restart drip once blood glucose is
above 70 mg/dL 2 checks; restart drip with lower algorithm (see
Moving Down)
Algorithm 1 Algorithm 2 Algorithm 3 Algorithm 4
BG Units/hour BG Units/hour BG Units/hour BG Units/hour
70 Off 70 Off 70 Off 70 Off
70–109 0.2 70–109 0.5 70–109 1 70–109 1.5
110–119 0.5 110–119 1 110–119 2 110–119 3
120–149 1 120–149 1.5 120–149 3 120–149 5
150–179 1.5 150–179 2 150–179 4 150–179 7
180–209 2 180–209 3 180–209 5 180–209 9
210–239 2 210–239 4 210–239 6 210–239 12
240–269 3 240–269 5 240–269 8 240–269 16
270–299 3 270–299 6 270–299 10 270–299 20
300–329 4 300–329 7 300–329 12 300–329 24
330–359 4 330–359 8 330–359 14 [ 330 28
[ 360 6 [ 360 12 [ 360 16 — —
60 = hypoglycemia (see below for treatment).
Abbreviation: BG, blood glucose.
110 MOGHISSI HIRSCH

When to notify the physician
If there is any blood glucose change greater than 100 mg/dL in 1 hour
If blood glucose level is above 360 mg/dL
For hypoglycemia that has not resolved within 20 minutes of
administering 50 mL of D50W IV and discontinuing the insulin drip
Adapted from Trence DL, Kelly JL, Hirsch IB. The rationale and management of hyper-
glycemia for in-patients with cardiovascular disease: time for change. J Clin Endocrinol
Metab 2003;88:2430–7; with permission.
Appendix 2. Example of standardized subcutaneous insulin orders
Blood glucose monitoring
Before meals and at bedtime
_______ hours after meals
2:00 to 3:00 AM
Suggested lag times for prandial insulin:
Aspart/Lispro: 0 to 15 minutes before eating
Regular: 30 minutes before eating
If blood glucose is 60 mg/dL:
A. If patient can take orally, give 15 g of fast-acting carbohydrate
(4-oz fruit juice/nondiet soda, 8-oz nonfat milk, or 3 to 4 glucose
tablets)
B. If patient cannot take orally, give 25 mL of D50 as IV push
C. Check ﬁnger capillary glucose every 15 minutes and repeat above if
blood glucose is 80 mg/dL
Insulin
orders Breakfast Lunch Dinner Bedtime
Prandial Give____units of: Give____units of: Give____units of: —
Lispro Lispro Lispro
Aspart Aspart Aspart
Regular Regular Regular
Basal Give____units of: — Give____units of: Give____units of:
NPH NPH NPH
Lente Lente Lente
Ultralente Ultralente Ultralente
Glargine Glargine Glargine
Goal premeal blood glucose: 80–150 mg/dL.
111

Premeal ‘‘correction dose’’ algorithm for hyperglycemia to be adminis-
tered in addition to scheduled insulin dose to correct premeal hyperglycemia.
Lispro
Aspart
General insulin dosing recommendations
Patients with type 1 diabetes
This patient must have insulin to prevent ketosis. Even if the patient is
not eating, he/she will need at least basal insulin (NPH/lente/ultralente/
glargine) to prevent ketosis.
1. When admitting a patient with type 1 diabetes, continue the basal insulin
that the patient was taking at home at the same dose. If the patient will not
be eating, use an insulin drip rather than subcutaneous insulin. The
prandial insulin (regular/lispro/aspart) may require adjustment depend-
ing on the patient’s situation. If the patient is eating much less, the prandial
insulin will need to be reduced. Many hospitalized patients are under
signiﬁcant metabolic stress (infection, glucocorticoids, and so forth) and
may require larger doses of prandial insulin despite eating less.
2. If a patient is newly diagnosed, the usual daily insulin requirement is 0.5
to 0.7 U/kg/d. Half should be given as basal insulin and the remainder
as prandial insulin.
Algorithm Premeal BG Additional insulin (units)
Low-dosea
150–199 1
200–249 2
250–299 3
300–349 4
[ 349 5
Medium-doseb
150–199 1
200–249 3
250–299 5
300–349 7
[ 349 8
High-dosec
150–199 2
200–249 4
250–299 7
300–349 10
[ 349 12
Individualized 150–199
200–249
250–299
300–349
[ 349
Abbreviation: BG, blood glucose.
a
For patients requiring 40 U of insulin per day.
b
For patients requiring 40–80 U of insulin per day.
c
For patients requiring [ 80 U of insulin per day.
112 MOGHISSI HIRSCH

Patients with type 2 diabetes
1. If patient is using insulin at home, continue the outpatient regimen and
adjust as needed.
2. If patient has not been using insulin previously, the usual total daily
insulin requirement is 0.4 to 1.0 U/kg/d.
3. Individual insulin doses vary widely and adjustments should be made
based on the bedside and laboratory glucose levels.
Note: Individual insulin doses vary widely, and adjustments should be
based on bedside and laboratory glucose levels.
Adapted from Trence DL, Kelly JL, Hirsch IB. The rationale and management of hyper-
glycemia for in-patients with cardiovascular disease: time for change. J Clin Endocrinol
Metab 2003;88:2430–87; with permission.
References
[1] Centers for Disease Control and Prevention (CDC). Diabetes fact sheet. Available at: http://
diabetes.org/diabetes-statistics/national-diabetes-fact-sheet.jsp. Accessed February 20,
2004.
[2] Narayan KMV, Boyle JP, Thompson TJ, et al. Lifetime risk for diabetes mellitus in the
United States. JAMA 2003;290:1884–90.
[3] Hogan P, Dall T, Nikolov P. American Diabetes Association. Economic costs of diabetes in
the US in 2002. Diabetes Care 2003;26:917–32.
[4] Malmberg K, Norhammar A, Wedel H, et al. Glycometabolic state at admission: important
risk marker of mortality in conventionally treated patients with diabetes mellitus and acute
myocardial infarction: long-term results from the Diabetes and Insulin-Glucose Infusion in
Acute Myocardial Infarction (DIGAMI) study. Circulation 1999;99:2626–32.
[5] Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill
patients. N Engl J Med 2001;345:1359–67.
[6] Furnary AP, Zerr KJ, Grunkemeier GL, et al. Continuous intravenous insulin infusion
reduces the incidence of deep sternal wound infection in diabetic patients after cardiac
surgical procedures. Ann Thorac Surg 1999;67:352–60 [discussion: 360–2].
[7] Lazar HL, Chipkin SR, Fitzgerald CA, et al. Tight glycemic control in diabetic coronary
artery bypass graft patients improves perioperative outcomes and decreases recurrent
ischemic events. Circulation 2004;109:1497–502.
[8] Centers for Disease Control and Prevention. Diabetes Public Health Resource. Available at:
http://www.cdc.gov/diabetes/statistics. Accessed February 20, 2004.
[9] American College of Endocrinology. Position statement on inpatient diabetes and metabolic
control. Endocr Pract 2004;10:77–82.
[10] Webster MW, Scott RS. What cardiologists need to know about diabetes. Lancet 1997;
350(Suppl):123–8.
[11] Tenerz A, Norhammar A, Silveira A, et al. Diabetes, insulin resistance, and the metabolic
syndrome in patients with acute myocardial infarction without previously known diabetes.
Diabetes Care 2003;26:2770–6.
[12] Norhammar A, Tenerz A, Nilsson G, et al. Glucose metabolism in patients with acute
myocardial infarction and no previous diagnosis of diabetes mellitus: a prospective study.
Lancet 2002;359:2140–4.
[13] Abbott R, Kannel W, Wilson P. The impact of diabetes on survival following myocardial
infarction in men and women. JAMA 1988;260:3456–60.
113

[14] Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia and increased risk of death after
myocardial infarction in patients with and without diabetes: a systematic overview. Lancet
2000;355:773–8.
[15] Malmberg K, Ryden L, Efendic S, et al. A randomized trial of insulin-glucose infusion
followed by subcutaneous insulin treatment in diabetic patients with acute myocardial
infarction: effects on one year mortality. J Am Coll Cardiol 1995;26:57–65.
[16] Malmberg K, Ryden L, Hamsten A, et al. Effects of insulin treatment on cause-specific one-
year mortality and morbidity in diabetic patients with acute myocardial infarction. Eur
Heart J 1996;17:1337–44.
[17] Malmberg K. Prospective randomized study of intensive insulin treatment on long-term
survival after acute myocardial infarction in patients with diabetes mellitus. BMJ 1997;314:
1512–5.
[18] Perdomo G, Commerford SR, Richard A-MT, et al. Increased beta-oxidation in muscle cells
enhances insulin-stimulated glucose metabolism and protects against fatty acid induced
insulin resistance despite intramyocellular lipid accumulation. J Biol Chem 2004;279(26):
27177–86.
[19] Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and
arrhythmias. Lancet 1994;343:155–8.
[20] Lopaschuk GD, Wambolt RB, Barr RL. An imbalance between glycolysis and glucose
oxidation is a possible explanation for the detrimental effects of high levels of fatty acids
during aerobic reperfusion of ischemic hearts. J Pharmacol Exp Ther 1993;264:135–44.
[21] Jouven X, Charles MA, Desnos M, et al. Circulating nonesterified fatty acid level as
a predictive risk factor for sudden death in the population. Circulation 2001;104:756–61.
[22] Hirsch IB. Impact of insulin therapy on non-glycemic effects during acute illness. Endo Pract
2004;10(Suppl 2):63–70.
[23] Gabay C, Kushner I. Acute phase proteins and other systemic responses to inflammation.
N Engl J Med 1999;340:448–54.
[24] Zeidler C, Kanz L, Hurkuck KL. In vivo effects of interleukin-6 on thrombopoiesis in
healthy and irradiated primates. Blood 1992;80:2740–5.
[25] Dandona P, Aljada A, Bandyopadhyay A. The potential therapeutic role of insulin in acute
myocardial infarction in patients admitted to intensive care and in those with unspecified
hyperglycemia. Diabetes Care 2003;26:516–9.
[26] Dandona P, Aljzcz Z, Mohanty P, et al. Insulin inhibits intranuclear factor jb and stimulates
jb in mononuclear cells in obese subjects: evidence for an anti-inflammatory effect. J Clin
Endocrinol Metab 2001;86:3257–65.
[27] Aljada A, Ghanim H, Saadeh R, et al. Insulin inhibits NF-jb and MCP-1 expression in
human aortic endothelial cells. J Clin Endocrinol Metab 2001;86:450–3.
[28] Ghanim H, Mohanty P, Aljada A, et al. Insulin reduces the pro-inflammatory transcription
factor, activation protein-1 (AP-1), in mononuclear cells (MNC) and plasma matrix
metalloproteinase-9 (MMP-9) concentration. Diabetes 2001;50(Suppl 1):A408.
[29] Steinberg HO, Brechtel G, Johnson N, et al. Insulin-mediated skeletal muscle vasodilation is
nitric-oxide dependent. A novel action of insulin to increase nitric oxide release. J Clin Invest
1994;94:1172–9.
[30] Edwards FH, Grover FL, Shroyer AL, et al. The Society of Thoracic Surgeons, National
Cardiac Surgery Database: current risk assessment. Ann Thorac Surg 1997;63:903–8.
[31] Golden SH, Pear-Vigilance C, Kao WH, et al. Preoperative glycemic control and the risk of
infectious complications in a cohort of adult with diabetes. Diabetes Care 1999;22:1408–14.
[32] Latham R, Lancaster AD, Covington JF, et al. The association of diabetes and glucose
control with surgical site infection among cardiothoracic surgery patients. Infect Control
Hosp Epidemil 2001;22:607–12.
[33] Zerr KJ, Furnary AP, Grunkemeier GL, et al. Glucose control lowers the risk of wound
infection in diabetics after open heart operations. Ann Thorac Surg 1997;63:356–61.
114 MOGHISSI HIRSCH

[34] Furnary AP, Gao G, Grunkemeier GL, et al. Continuous insulin infusion reduces mortality
in patients with diabetes undergoing coronary artery bypass grafting. J Thorac Cardiovasc
Surg 2003;125:1007–21.
[35] Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia and prognosis of stroke in non
diabetic and diabetic patients: a systematic overview. Stroke 2001;32:2426–32.
[36] Williams LS, Rotich J, Qi R, et al. Effects of admission hyperglycemia on mortality and costs
in acute ischemic stroke. Neurology 2002;59:67–71.
[37] Pulsinelli WA, Levy DE, Sigsbee B, et al. Increased damage after ischemic stroke in
patients with hyperglycemia with or without established diabetes mellitus. Am J Med 1983;
74:540–4.
[38] Pomposelli JJ, Baxter JK III, Babineau TJ, et al. Early postoperative glucose control predicts
nosocomial infection rate in diabetic patients. J Parenter Enteral Nutr 1998;22:77–81.
[39] Golden SH, Peart-Vigilance C, Kao WH, et al. Perioperative glycemic control and the risk
of infectious complications in a cohort of adults with diabetes. Diabetes Care 1999;22:
1408–14.
[40] Joshi N, Caputo G, Weitekamp M, et al. Infections in patients with diabetes mellitus. N Engl
J Med 1999;341:1906–12.
[41] Wheat L. Infection and diabetes mellitus. Diabetes Care 1980;3:187–97.
[42] Mowat A, Baum J. Chemotaxis of polymorphonuclear leukocytes from patients with
diabetes mellitus. N Engl J Med 1971;284:621–7.
[43] Bagdade J, Root R, Bulger R. Impaired leukocyte function in patients with poorly controlled
diabetes. Diabetes 1974;23:9–15.
[44] Bagdade JD, Stewart M, Walters E. Impaired granulocyte adherence. A reversible defect in
host defense in patients with poorly controlled diabetes. Diabetes 1978;27:677–81.
[45] van Oss CJ, Border JR. Influence of intermittent hyperglycemic glucose levels on the
phagocytosis of microorganisms by human granulocytes in vitro. Immunol Commun 1978;7:
669–76.
[46] Davidson N, Sowden J, Fletcher J. Defective phagocytosis in insulin-controlled diabetics:
evidence for a reaction between glucose and opsonizing proteins. J Clin Pathol 1984;37:
783–6.
[47] Alexiewicz J, Kumar D, Smogorzewski M, et al. Polymorphonuclear leukocytes in non-
insulin-dependent diabetes mellitus: abnormalities in metabolism and function. Ann Intern
Med 1995;123:919–24.
[48] Leibovici L, Yehezkelli Y, Porter A, et al. Influence of diabetes mellitus and glycemic control
on the characteristics and outcome of common infections. Diabet Med 1996;13:457–63.
[49] Kwoun M, Ling P, Lydon E, et al. Immunologic effects of acute hyperglycemia in
nondiabetic rats. J Parenter Enteral Nutr 1997;21:91–5.
[50] McManus L, Bloodworth R, Prihoda T, et al. Agonist-dependent failure of neutrophil
function in diabetes correlates with extent of hyperglycemia. J Leukoc Biol 2001;70:
395–404.
[51] Black CT, Hennessey PJ, Andrassy RJ. Short-term hyperglycemia depresses immunity
through nonenzymatic glycosylation of circulating immunoglobulin. J Trauma 1990;30:
830–2 [discussion: 832–3].
[52] Bouter KP, Meyling FH, Hoekstra JB, et al. Influence of blood glucose levels on peripheral
lymphocytes in patients with diabetes mellitus. Diabetes Res 1992;19:77–80.
[53] Selby JV, Ray GT, Zhang D, et al. Excess costs of medical care for patients with diabetes in
a managed care population. Diabetes Care 1997;20:1396–402.
[54] Weir CJ, Murray GD, Dyker AG, et al. Is hyperglycaemia an independent predictor of poor
outcome after acute stroke? Results of a long term follow up study. BMJ 1997;314:1303–6.
[55] Wahab NN, Cowden EA, Pearce NJ, et al. ICONS Investigators. Is blood glucose an
independent predictor of mortality in acute myocardial infarction in the thrombolytic era?
J Am Coll Cardiol 2002;40:1748–54.
115

[56] Umpierrez GE, Isaacs SD, Bazargan N, et al. Hyperglycemia: an independent marker of
in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002;
87:978–82.
[57] Koproski J, Pretto Z, Poretsky L. Effects of an intervention by a diabetes team in
hospitalized patients with diabetes. Diabetes Care 1997;20:1553–5.
[58] Levetan CS, Salas JR, Wilets IF, et al. Impact of endocrine and diabetes team consultation
on hospital length of stay for patients with diabetes. Am J Med 1995;99:22–8.
[59] Levetan CS, Passaro MD, Jablonski KA, et al. Effect of physician specialty on outcomes in
diabetic ketoacidosis. Diabetes Care 1999;22:1790–5.
[60] Almbrand B, Johannesson M, Sjostrand B, et al. Cost-effectiveness of intense insulin
treatment after acute myocardial infarction in patients with diabetes mellitus; results from
the DIGAMI study. Eur Heart J 2000;21:733–9.
[61] Zhan C, Miller MR. Excess length of stay, charges, and mortality attributable to medical
injuries during hospitalization. JAMA 2003;290:1868–74.
[62] Furnary AP, Chaugle H, Zerr KJ, et al. Postoperative hyperglycemia prolongs length of stay
in diabetic CABG patients. Circulation 2000;102:II–556.
[63] Clement S, Braithwait SS, Magee MF, et al. Management of diabetes and hyperglycemia in
hospitals. Diabetes Care 2004;27:553–91.
[64] Queale WS, Seidler AJ, Brancati FL. Glycemic control and sliding scale insulin use in
medical inpatients with diabetes mellitus. Arch Intern Med 1997;157:545–52.
[65] Hirsch IB, Farkas-Hirsch R. Sliding scale or sliding scare: it’s all sliding nonsense. Diabetes
Spectrum 2001;14:79–81.
[66] Hirsch IB, Paauw DS, Brunzell J. Inpatient management of adults with diabetes. Diabetes
Care 1995;18:870–8.
[67] Trence DL, Kelly JL, Hirsch IB. The rationale and management of hyperglycemia for in-
patients with cardiovascular disease: time for change. J Clin Endocrinol Metab 2003;88:
2430–87.
[68] Markovitz L, Wiechmann R, Harris N, et al. Description and evaluation of a glycemic
management protocol for diabetic patients undergoing heart surgery. Endocr Pract 2002;8:
10–8.
[69] Albert Starr Academic Center for Cardiac Surgery. Star Wood Research. Available at:
http://www.starrwood.com/research/insulin.html. Accessed March 14, 2004.
[70] Ku SY, Sayre CA, Hirsch IB, et al. New insulin infusion protocol improves blood control in
hospitalized patients without increasing hypoglycemia. Jt Comm J Qual Improv, in press.
[71] Dewitt DE, Hirsch IB. Outpatient therapy for type 1 and type 2 diabetes: scientific review.
JAMA 2003;289:2254–64.
[72] Tabak AG, Tamas G, Zgibor J, et al. Targets and reality: a comparison of healthcare
indicators in the US (Pittsburgh Epidemiology of Diabetes Complications Study) and
Hungary (DiabCare Hungary). Diabetes Care 2000;23:1284–9.
[73] Haywood RA, Manning WG, Kaplan SH, et al. Starting insulin therapy in patients with type
2 diabetes: effectiveness, complications, and resource utilization. JAMA 1997;278:1663–9.
116 MOGHISSI HIRSCH

Moghissi

More Related Content

What's hot (20)

Viewers also liked (8)

Similar to Moghissi (20)

More from Residentes1hun (20)

Moghissi