Can J Diabetes 42 (2018) S115–S123

Contents lists available at ScienceDirect

Canadian Journal of Diabetes

journal homepage:
w w w. c a n a d i a n j o u r n a l o f d i a b e t e s . c o m

2018 Clinical Practice Guidelines

In-Hospital Management of Diabetes

Diabetes Canada Clinical Practice Guidelines Expert Committee
Janine Malcolm MD, FRCPC, Ilana Halperin MD, FRCPC, David B. Miller MD, FRCPC,
Sarah Moore RN(EC), BScN, MN, Kara A. Nerenberg MD, FRCPC, Vincent Woo MD, FRCPC,
Catherine H. Yu MD, FRCPC

Introduction
KEY MESSAGES
Diabetes increases the risk for hospitalization for several reasons,
• Hyperglycemia is common in hospitalized people, even among those without including: cardiovascular (CV) disease, nephropathy, infection, cancer
a previous history of diabetes, and is associated with increased in-hospital
and lower-extremity amputations. In-hospital hyperglycemia is
complications, longer length of stay and mortality.
• Insulin is the most appropriate pharmacologic agent for effectively con- common. A review of medical records of over 2,000 adult patients
trolling glycemia in hospital. A proactive approach to glycemic manage- admitted to a community teaching hospital in the United States
ment using scheduled basal, bolus and correction (supplemental) insulin (>85% were nonintensive care unit patients) found that hypergly-
is the preferred method. The use of correction-only (supplemental) insulin, cemia was present in 38% of patients (1). Of these patients, 26% had
which treats hyperglycemia only after it has occurred, should be discour-
aged as the sole modality for treating elevated blood glucose levels.
a known history of diabetes, and 12% had no history of diabetes prior
• For the majority of noncritically ill hospitalized people with diabetes, to admission. Diabetes has been reported to be the fourth most
preprandial blood glucose targets should be 5.0 to 8.0 mmol/L, in conjunc- common comorbid condition listed on all hospital discharges (2).
tion with random blood glucose values <10.0 mmol/L, as long as these targets Acute illness results in a number of physiological changes (e.g.
can be safely achieved. For critically ill hospitalized people with diabetes,
increases in circulating concentrations of stress hormones) or thera-
blood glucose levels should be maintained between 6.0 and 10.0 mmol/L.
• Hypoglycemia is a major barrier to achieving targeted glycemic control in peutic choices (e.g. glucocorticoid use) that can exacerbate hyper-
the hospital setting. Health-care institutions should develop protocols for glycemia. Hyperglycemia, in turn, causes physiological changes that
the assessment and treatment of hypoglycemia. can exacerbate acute illness, such as decreased immune function
and increased oxidative stress. These lead to a complex cycle of wors-
ening illness and poor glucose control (3). Although a growing body
of literature supports the need for targeted glycemic control in the
KEY MESSAGES FOR PEOPLE WITH DIABETES hospital setting, blood glucose (BG) continues to be poorly con-
trolled and is frequently overlooked in general medicine and surgery
• If your admission to hospital is planned, talk with your health-care pro- services. This is largely explained by the fact that the majority of
viders (e.g. surgeon, anesthetist, primary care provider, diabetes health pro-
hospitalizations for patients with diabetes are not directly related
vider, etc.) before you are admitted in order to develop an in-hospital
diabetes care plan that addresses such issues as: to their metabolic state, thus diabetes management is rarely the
◦ Who will manage your diabetes in the hospital? primary focus of care. Therefore, glycemic control and other dia-
◦ Will you be able to self-manage your diabetes? betes care issues are often not specifically addressed (4).
◦ What adjustments to your diabetes medications or insulin doses may
be necessary before and after medical procedures or surgery?
◦ If you use an insulin pump, are hospital staff familiar with pump
therapy? Screening for and Diagnosis of Diabetes and Hyperglycemia in
• Your blood glucose levels may be higher in hospital than your usual target the Hospital Setting
range due to a variety of factors, including the stress of your illness, medi-
cations, medical procedures and infections.
• Your diabetes medications may need to be changed during your hospital A history of diabetes should be elicited in all patients admitted
stay to manage the changes in blood glucose, or if medical conditions to hospital and, if present, should be clearly identified on the medical
develop that make some medications no longer safe to use. record. In view of the high prevalence of inpatient hyperglycemia
• When you are discharged, make sure that you have written instructions
with associated poor outcomes, an admission BG measurement of
about:
◦ Changes in your dosage of medications or insulin injections or any all patients would help identify people with diabetes, even in the
new medications or treatments absence of a prior diagnosis (1,5). In-hospital hyperglycemia is
◦ How often to check your blood glucose defined as any glucose value >7.8 mmol/L. For hospitalized people
◦ Who to contact if you have difficulty managing your blood glucose
with known diabetes, the glycated hemoglobin (A1C) identifies
levels.
people who may benefit from efforts to improve glycemic control
and tailor therapy upon discharge (6,7). In hospitalized people with
newly recognized hyperglycemia, an A1C among those with diabetes
Conflict of interest statements can be found on page S121. risk factors or associated comorbidities (e.g. cardiovascular disease
1499-2671 © 2018 Canadian Diabetes Association.
The Canadian Diabetes Association is the registered owner of the name Diabetes Canada.
https://doi.org/10.1016/j.jcjd.2017.10.014
S116 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123

[CVD]) (8,9) may help differentiate people with previously undi- Table 1
agnosed diabetes and dysglycemia from those with stress-induced Recommended glycemic targets for hospitalized people with diabetes*

hyperglycemia and provides an opportunity to diagnose and initi- Hospitalized population with diabetes Blood glucose targets (mmol/L)
ate diabetes therapies (10–13). Among people admitted to an inten- Noncritically ill Preprandial: 5.0–8.0
sive care unit (ICU), an A1C drawn at admission allows identification Random: <10.0
of people with previously unknown diabetes, people at risk of gly- Critically ill 6.0–10.0
cemic management challenges and people at an increased risk of CABG intraoperatively 5.5–11.1
Perioperatively for other surgeries 5.0–10.0
mortality (14,15). A1C has been found to be specific for diagnosis Acute coronary syndrome† 7.0–10.0
of diabetes in the hospital setting, although not as sensitive as in Labour and delivery‡ 4.0–7.0
the outpatient setting (13,16). While the threshold for diagnosis of CABG, coronary artery bypass grafting.
diabetes has not been established for hospitalized people, an A1C * Less stringent targets may be appropriate in terminally ill patients or in people
criteria of >6.0% has been found to be highly specific for the diag- with severe comorbidities (see Targets for Glycemic Control chapter, p. S42).
†
nosis of dysglycemia post-hospitalization (13,17). See Management of Acute Coronary Syndromes chapter, p. S190.
‡ See Diabetes and Pregnancy chapter, p. S255.

Glucose Monitoring in the Hospital Setting

the presence of acute illness. For the majority of noncritically ill hos-
Bedside blood glucose monitoring pitalized people, recommended preprandial BG targets are 5.0 to
8.0 mmol/L, in conjunction with random BG values <10.0 mmol/L,
Currently, there are no studies that have examined the effect of as long as these targets can be safely achieved (Table 1). Lower targets
the frequency of bedside BG monitoring on the incidence of hyper- may be considered in clinically stable hospitalized people with a
or hypoglycemia in the hospital setting. The frequency and timing prior history of successful tight glycemic control in the outpatient
of bedside BG monitoring can be individualized; however, moni- setting, while higher targets may be acceptable in terminally ill
toring is typically performed before meals and at bedtime in people people or in those with severe comorbidities. If BG values are
who are eating; every 4 to 6 hours in people who are NPO (nothing ≤3.9 mmol/L, modification of antihyperglycemic therapy is sug-
by mouth) or receiving continuous enteral feeding; and every 1 to gested, unless the event is easily explained by other factors (e.g. a
2 hours for people on continuous intravenous insulin or those who missed meal) (18,30).
are critically ill. Some bedside BG monitoring is indicated in indi-
viduals without known diabetes but receiving treatments known
to be associated with hyperglycemia (e.g. glucocorticoids, octreotide, Glycemic Control in the Critically Ill Patient
parenteral nutrition and enteral nutrition) (18). The implementa-
tion and maintenance of quality control programs by health-care Acute hyperglycemia in the intensive care setting is not unusual
institutions helps to ensure the accuracy of bedside BG monitor- and results from a number of factors, including stress-induced
ing (19,20). The use of glucose meters with bar coding capability counter-regulatory hormone secretion and the effects of medica-
has been shown to reduce data entry errors in medical records (21). tions administered in the ICU (31). Glycemic targets for people with
Data management programs that transfer bedside BG monitoring pre-existing diabetes who are in the critical care setting have not
results into electronic records allow evaluation of hospital-wide gly- been firmly established. Early trials showed that achieving
cemic control (22). normoglycemia (4.4 to 6.1 mmol/L) in cardiac surgery patients or
Capillary blood glucose (CBG) point of care testing (POCT) should patients in postoperative surgical ICU settings reduced mortality
be interpreted with caution in the critically ill patient population. (32). However, subsequent trials in mixed populations of critically
Poor perfusion indices may yield conflicting capillary, arterial and ill patients did not show a benefit of targeting BG levels of 4.4 to
whole BG values using POCT glucose meters (23–25). Venous or arte- 8.3 mmol/L. A meta-analysis of trials of intensive insulin therapy
rial samples are preferred when using a POCT meter for this patient in the ICU setting suggested benefit of intensive insulin therapy in
population. surgical patients, but not in medical patients (33). Conversely, the
Clinical decision support system software integrating CBG POCT Normoglycemia in Intensive Care Evaluation—Survival Using Glucose
can aid in trend analysis, medication dosing, reduce prescription Algorithm Regulation (NICE-SUGAR) study, the largest trial to date
error and reduce length of stay (26). Electronic glucose metric data of intensive glucose control in critically ill medical and surgical
and web-based reporting systems may pose utility for monitoring patients, found an increase in 90-day all-cause mortality (hazard
glycemic management performance within an organization and ratio [HR] 1.14; 95% confidence interval [CI] 1.02–1.28; p=0.02)
enhance opportunities for external benchmarking (27). amongst participants randomized to the intensive glycemic control
arm that targeted BG levels of 4.5 to 6.0 mmol/L (34). Further-
more, intensive insulin therapy has been associated with an
Glycemic Control in the Non-Critically Ill Patient increased risk of hypoglycemia in the ICU setting (33). Therefore,
maintaining a BG level <10.0 mmol/L in critically ill hospitalized
A number of studies have demonstrated that inpatient hyper- people with diabetes is considered a safe target (Table 1). The lower
glycemia is associated with increased morbidity and mortality in limit for the BG target is less well established but generally should
noncritically ill hospitalized people (1,28,29). However, due to a remain >6.0 mmol/L in order to minimize the risks of both hypo-
paucity of randomized controlled trials on the benefits and risks glycemia and mortality. The use of insulin infusion protocols with
of “conventional” vs. “tight” glycemic control in noncritically ill hos- proven efficacy and safety minimizes the risk of hypoglycemia
pitalized people, glycemic targets for this population remain unde- (35–38).
fined. Current recommendations are based mostly on retrospective
studies, clinical experience and judgement. Glycemic targets for
hospitalized people with diabetes are modestly higher than those Role of Intravenous Insulin
routinely advised for outpatients with diabetes given that the hos-
pital setting presents unique challenges for the management of There are few occasions when intravenous insulin is required,
hyperglycemia, such as variations in patient nutritional status and as most people with type 1 or type 2 diabetes admitted to general
 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123 S117

medical wards can be treated with subcutaneous insulin. Intrave- rates in individuals with and without diabetes (51–53). A system-
nous insulin, however, may be appropriate for people who are criti- atic review of randomized controlled trials supports the use of intra-
cally ill (with appropriate BG targets), people who are not eating venous insulin infusion targeting a blood glucose of 5.5 to
and in those with hyperglycemia and metabolic decompensation 11.1 mmol/L over correction (supplemental) subcutaneous insulin
(e.g. diabetic ketoacidosis [DKA] and hyperosmolar hyperglyce- for perioperative glycemic control in CV surgery patients (Table 1).
mic state [HHS]) (see Hyperglycemic Emergencies in Adults chapter, This was demonstrated by a marked reduction in surgical site infec-
p. S109). The evidence to date suggests there is no benefit to intra- tions (odds ratio 0.13) (54).
venous insulin over subcutaneous insulin post-acute stroke (3,39).
Health-care staff education is a critical component of the imple- Minor and moderate surgery
mentation of an intravenous insulin infusion protocol. Intrave-
nous insulin protocols should take into account the patient’s current The perioperative glycemic targets for minor or moderate sur-
and previous BG levels (as well as the rate of change in BG), and geries are less clear. Older studies comparing different methods of
the patient’s usual insulin dose. Several published insulin infu- achieving glycemic control during minor and moderate surgeries
sion protocols appear to be both safe and effective, with low rates did not demonstrate any adverse effects of maintaining perioperative
of hypoglycemia; however, most of these protocols have only been BG levels between 5.0 to 11.0 mmol/L (55–57). Attention has been
validated in the ICU setting, where the nurse-to-patient ratio is placed on the relationship between postoperative hyperglycemia
higher than on medical and surgical wards (3,36). BG determina- and surgical site infections. While the association was well docu-
tions can be performed every 1 to 2 hours until BG has stabilized. mented, the impact and risks of intensive management was less
With the exception of the treatment of hyperglycemic emergen- clear. A recent meta-analysis of 15 randomized controlled trials dem-
cies (e.g. DKA and HHS), consideration should be given to concur- onstrated that intensive perioperative glycemic control (BG target
rently providing people receiving intravenous insulin with some form of <8.3 mmol/L) resulted in decreased odds of surgical site infec-
of glucose (e.g. intravenous glucose or through parenteral or enteral tions when compared to conventional control (BG target of
feeding). <12 mmol/L). The risk of hypoglycemia was increased but there was
no increased risk of stroke or death. The included studies looked
Transition from IV insulin to SC insulin therapy at the intraoperative and immediate postoperative period and used
intravenous insulin to achieve intensive targets. The included studies
Hospitalized people with type 1 and type 2 diabetes may be were mostly cardiac and gastrointestinal and were found to have
transitioned to scheduled subcutaneous insulin therapy from intra- a moderate risk of bias (58).
venous insulin. Short- or rapid- or fast-acting insulin can be admin- Rapid institution of perioperative glucose control must be care-
istered 1 to 2 hours before discontinuation of the intravenous insulin fully considered in patients with poorly controlled type 2 diabe-
to maintain effective blood levels of insulin. If intermediate- or long- tes undergoing monocular phacoemulsification cataract surgery with
acting insulin is used, it can be given 2 to 3 hours prior to intra- moderate to severe nonproliferative diabetic retinopathy because
venous insulin discontinuation. People without a history of diabetes, of the possible increased risk of postoperative progression of reti-
who have hyperglycemia requiring more than 2 units of intrave- nopathy and maculopathy (59). The outcome of vitrectomy, however,
nous insulin per hour, likely require insulin therapy and can be con- does not appear to be influenced by perioperative control (60).
sidered for transition to scheduled subcutaneous insulin therapy. Given the data supporting tighter perioperative glycemic control
The initial dose and distribution of subcutaneous insulin at the during major surgeries and the compelling data showing the adverse
time of transition can be determined by extrapolating the intrave- effects of hyperglycemia, it is reasonable to target glycemic levels
nous insulin requirement over the preceding 6- to 8-hour period between 5.0 to 10.0 mmol/L for minor and moderate surgeries in
to a 24-hour period. Administering 60% to 80% of the total daily cal- patients with known diabetes (Table 1). The best way to achieve
culated dose as basal insulin has been demonstrated to be safe and these targets in the postoperative patient is with a basal bolus insulin
efficacious in surgical patients (40). Dividing the total daily dose regimen (61,62). This approach has been shown to reduce postop-
as a combination of basal and bolus insulin has been demon- erative complications, including wound infections. Despite this
strated to be safe and efficacious in medically ill patients (40,41). knowledge, surgical patients are often treated with correction
(supplemental) rapid-acting insulin alone (63) which may not
Perioperative glycemic control adequately control BG.
The benefits of improved perioperative glycemic control must
The management of individuals with diabetes at the time of be weighed against the risk of perioperative hypoglycemia. Anes-
surgery poses a number of challenges. Acute hyperglycemia thetic agents and postoperative analgesia may alter the patient’s
is common secondary to the physiological stress associated level of consciousness and awareness of hypoglycemia. The risk of
with surgery. Pre-existing diabetes-related complications and hypoglycemia can be reduced by frequent BG monitoring and care-
comorbidities may also influence clinical outcomes. Acute hyper- fully designed management protocols.
glycemia has been shown to adversely affect immune function (42)
and wound healing (43) in animal models. Observational studies
have shown that hyperglycemia increases the risk of postopera- Role of Subcutaneous Insulin
tive infections (44,45), renal allograft rejection (46), and is associ-
ated with increased health-care resource utilization (47). In general, insulin is the preferred treatment for hyperglyce-
mia in hospitalized people with diabetes (35). People with type 1
Cardiovascular surgery diabetes must be maintained on insulin therapy at all times to
prevent DKA. Scheduled subcutaneous insulin administration that
In people undergoing coronary artery bypass grafting (CABG), consists of basal, bolus (prandial) and correction (supplemental)
a pre-existing diagnosis of diabetes has been identified as a risk factor insulin components is the preferred method for achieving and main-
for postoperative sternal wound infections, delirium, renal dys- taining glucose control in noncritically ill hospitalized people with
function, respiratory insufficiency and prolonged hospital stays diabetes or stress hyperglycemia who are eating (35,64). Bolus insulin
(48–50). Intraoperative hyperglycemia during cardiopulmonary can be withheld or reduced in people who are not eating regu-
bypass has been associated with increased morbidity and mortality larly; however, basal insulin should not be withheld. Stable people
S118 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123

can usually be maintained on their home insulin regimen with diabetes (77). The glycemic outcomes were similar between the 2
adjustments made to accommodate for differences in meals and groups; however, the basal-bolus-correctional group had a higher
activity levels, the effects of illness and the effects of other medi- mean glucose than similarly insulin-treated subjects in other studies
cations. In the hospital setting, rapid-acting insulin analogues are (61,66). This less-aggressive treatment may explain the lack of dif-
the preferred subcutaneous bolus and correction insulins (65). Insulin ference between the sitagliptin and the bolus insulin groups.
programs that only react to, or correct for, hyperglycemia have been
demonstrated to be associated with higher rates of hyperglyce-
mia (61,66–69). Insulin is often required temporarily in hospital, even Role of Medical Nutrition Therapy
in people with type 2 diabetes not previously treated with insulin.
In these insulin-naive people, there is evidence demonstrating the Medical nutrition therapy including nutritional assessment and
superiority of basal-bolus-correction insulin regimens (61,66). individualized meal planning is an essential component of inpa-
A number of protocols have been published as part of studies tient glycemic management programs. A consistent carbohydrate
(61,66,69–72). These studies have typically started insulin-naive meal planning system may facilitate glycemic control in hospital-
people on 0.4 to 0.5 units of insulin per kilogram of body weight ized people and facilitate matching prandial insulin doses to the
per day, with 40% to 50% of the total daily dose (TDD) given as basal amount of carbohydrate consumed (61,66,75,78–80).
insulin (detemir, glargine, neutral protamine Hagedorn [NPH]) and
the balance given as bolus (rapid- or short-acting) insulin divided
equally before each meal (i.e. breakfast, lunch and dinner); correc- Special Clinical Situations
tion doses of the bolus insulin are provided if BG values are above
target. Daily review of the person’s BG measurements and modi- Hospitalized people with diabetes receiving enteral or parenteral
fication of insulin doses, as required, facilitates the achievement of feedings
target blood glucose measurements.
When comparing effective protocols, the following was observed. In hospitalized people with diabetes receiving parenteral nutri-
One study compared basal-bolus (plus correction) insulin with tion, insulin can be administered in the following ways: as sched-
glargine and glulisine vs. premixed insulin (30/70) (73). The study, uled regular insulin dosing added directly to the parenteral solution;
although small (a total of 72 patients), had to be stopped early or as scheduled intermediate- or long-acting subcutaneous insulin
because of a tripling of the rate of hypoglycemia, BG <3.8 mmol/L, doses (81). A separate intravenous infusion of regular insulin may
in the premixed insulin group. Average BG levels were not differ- be an alternative method to achieve glycemic control in critical care
ent, but rates of hypoglycemia were. Another study (74) found no (82). For scheduled subcutaneous insulin dosing or regular insulin
difference in BG levels or rates of hypoglycemia when comparing added directly to parenteral solutions, the selected starting insulin
insulin glargine vs. detemir, when used as the basal insulin in a basal- dose may be based on the current estimated TDD of insulin, the com-
bolus program. Yet another study (71) found that using a weight- position of the parenteral nutrition solution and the patient’s weight
based algorithm to titrate insulin glargine resulted in obtaining target (81). Considering the patient’s individual clinical situation is impor-
BG levels faster than a glucose-based algorithm, with no differ- tant when determining insulin dosing. Subcutaneous correction
ence in the rates of hypoglycemia. (supplemental) insulin may be used in addition to scheduled insulin
More recently, a study compared a basal-bolus (plus correc- dosing and dose adjustments made to scheduled insulin should be
tion) insulin regimen with a program that was basal plus correc- adjusted based on the BG pattern.
tion (69). The basal-bolus group had slightly lower BG through the For hospitalized people with diabetes on enteral feeding regi-
day, which was not statistically significant, with no difference in FBG mens, there are few prospective studies examining insulin man-
or in rates of hypoglycemia. Taken together with the earlier studies agement. In 1 randomized controlled trial, low-dose basal glargine
from this group (61,66), it would appear that successful manage- insulin with regular insulin correction dosing was compared against
ment of in-hospital diabetes requires early and aggressive admin- regular insulin correction (supplemental) insulin dosing with the
istration of basal insulin combined with bolus insulin, typically in addition of NPH in the presence of persistent hyperglycemia and
the form of rapid-acting insulin analogue, similar to the approach demonstrated similar efficacy for glycemic control (83). The type
used in the outpatient setting. of feed solution and duration of feed (cyclical vs. continuous) should
be considered. People with diabetes receiving bolus enteral feeds
may be treated in the same manner as people who are eating meals.
Role of Noninsulin Antihyperglycemic Agents Approximately 50% of the TDD can be provided as basal insulin and
50% as bolus insulin, which is administered in divided doses to match
To date, no large studies have investigated the use of non- feed times (75). Correction (supplemental) insulin can be admin-
insulin antihyperglycemic agents on outcomes in hospitalized istered, as needed; added to the same bolus insulin. An insulin with
people with diabetes. There are often short- and/or long-term a shorter half-life, such as NPH, may be preferred for intermedi-
contraindications to the use of noninsulin antihyperglycemic agents ate duration feeding schedules (i.e. overnight), while regular or rapid-
in the hospital setting, such as irregular eating, acute or chronic renal acting insulin may be more appropriate to manage hyperglycemia
failure, and exposure to intravenous contrast dye (75). Stable hos- induced by bolus feeding schedules.
pitalized people with diabetes without these contraindications can In the event that the parenteral or enteral nutrition is unex-
often have their home antihyperglycemic medications continued pectedly interrupted, intravenous dextrose may be required to
while in the hospital. However, if contraindications develop or if prevent hypoglycemia depending on the last dose and type of insulin
glycemic control is inadequate, these drugs should be discontin- administered. When parenteral or enteral feeding schedules are
ued and consideration given to starting the patient on a basal- adjusted in terms of carbohydrate content or duration, the insulin
bolus-supplemental insulin regimen. The advantages and type and dose will need to be re-assessed.
disadvantages of various noninsulin antihyperglycemic therapies
in hospital are discussed in detail in a recent review article (76). Hospitalized people with diabetes receiving corticosteroid therapy
A recent randomized but unblinded study compared sitagliptin
plus basal (and correctional) insulin with a more traditional Hyperglycemia is a common complication of corticosteroid
basal-bolus-correctional insulin program in hospitalized people with therapy, with a prevalence between 20% and 50% among people
 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123 S119

without a previous history of diabetes (84). Although the optimal for continued CSII, procedures to guide medical management of CSII
management of hyperglycemia in people receiving high-dose oral and a consent form outlining the inpatient terms of use (92) support
corticosteroids has not been clearly defined, glycemic monitoring the safe use of CSII use in hospital. Specific algorithms and order
for 48 hours after initiation of steroids may be considered for people sets for management of CSII peri-operatively and during labour and
with or without a history of diabetes (35,84). For management of delivery have been published (93,94).
hyperglycemia, treatment with a basal-bolus with correction insulin
regimen was more effective and safer than a correction (supple-
mental) insulin-only regimen (85), although addition of NPH (dosed Organization of Care
variably from once a day at time of glucocorticoid administration
to every 6 hours depending on glucocorticoid used) was not dem- Institution-wide programs to improve glycemic control in the
onstrated to improve glycemic outcomes (86,87). inpatient setting include the formation of a multidisciplinary steer-
ing committee, professional development programs focused on inpa-
tient diabetes management (95,96), policies to assess and monitor
Self-management of diabetes in hospital the quality of glycemic management, interprofessional team-
based care (including comprehensive patient education and dis-
Although data for self-management in the hospitalized setting charge planning) as well as standardized order sets, protocols and
is limited, self-management in hospital may be appropriate for algorithms for diabetes care within the institution. Implementa-
people who are mentally competent and desire more autonomy over tion of such a program can result in improvements in in-hospital
their diabetes. The majority of evidence pertains to continuous sub- glycemic control (97,98).
cutaneous insulin infusion (CSII) therapy, where continuation of
patient-managed insulin delivery has been associated with reduced Algorithms, order sets and decision support
episodes of severe hyperglycemia and hypoglycemia (88) and high
levels of patient satisfaction (89). In general, any person requiring Order sets for basal-bolus-correction insulin regimens, insulin
insulin therapy who is self-managing diabetes in the hospital setting management algorithms (70,96,99–102), and computerized order
should be able to physically self-administer insulin and perform self- entry systems (101,103) have been shown to improve glycemic
monitoring of blood glucose (SMBG) independently, be familiar with control and/or reduce adverse outcomes in hospitalized people with
the recommended insulin routine, understand sick-day manage- diabetes. Computerized and mobile decision support systems (that
ment guidelines and utilize a flowsheet to facilitate communica- provide suggestions for insulin dosing) have also been used and have
tion of BG results and insulin dosing between the patient and health- been associated with lower mean BG levels (26,104–106); hypo-
care providers. The person with diabetes and the health-care glycemia can be an unintended consequence of tighter glycemic
provider, in consultation with nursing staff, must agree that patient control (70,105).
self-management is an appropriate strategy while hospitalized. Hos-
pitals should have policies and procedures for the assessment of Interprofessional team-based approach
suitability for self-management.
The timely consultation of glycemic management teams has also
been found to improve the quality of care provided, reduce the length
Hospitalized people with diabetes using CSII of hospital stay and lower costs (107,108), although differences in
glycemic control were minimal (109). Deployment of nurses
Although the data are limited, it appears that CSII can be safely (110,111), nurse practitioners and physician assistants (112) with
continued in the hospital setting under certain circumstances (90). specialty training has been associated with greater use of basal-
People maintained on CSII may have decreased length of stay (90); bolus insulin therapy and lower mean BG levels. A provincial survey
however, this may reflect the severity of illness rather than a gly- of over 2,000 people with diabetes admitted to hospital found that
cemic control advantage. People maintained on CSII may have less people were more likely to be satisfied with their diabetes care in
hypoglycemia than those managed by the admitting clinician. People hospital if they had confidence that the team was knowledgeable
on CSII are encouraged to continue this form of therapy whenever about diabetes, presented a consistent message and acknowl-
safe and feasible in hospital. Successful published inpatient proto- edged them in their diabetes care (113).
cols include assessment of pump specific self-management skills
(i.e. how to adjust their basal rate, administer a bolus dose, insert Comprehensive patient education
an infusion set, fill a reservoir, suspend the pump and correct a CBG
result outside their target range), pre-printed orders, flow sheets Programs that include self-management education, such as
and patient consents (88,91,92). If the patient cannot demon- assessment of barriers and goal setting, have also been associated
strate and/or describe the above-mentioned actions and desires to with improvements in glycemic control (97,111).
continue CSII, appropriate education and supports can be pro-
vided. If appropriate supports are not available, CSII may be dis- Metrics for evaluating inpatient glycemic management programs
continued and a basal-bolus-subcutaneous insulin regimen or
intravenous insulin infusion may be initiated. Institutional implementation of hospital glycemic manage-
An increasing number of people are being maintained on CSII ment programs require metrics to monitor progress, assess safety,
during short elective surgical procedures without any reported length of stay and identify opportunities for improvement (27).
adverse events (93), necessitating close collaboration between anes- Implementation of inpatient hyperglycemia quality improvement
thesia and diabetes management teams. Different pump manufac- programs evaluated with real-time metrics have been shown to
turers will recommend discontinuing pumps for certain hospital- improve glycemic control and safety of insulin ordering (97,114).
based procedures (e.g. radiology, cautery, external beam radiation). To date, metrics for monitoring glycemic control programs in
To promote a collaborative relationship between the hospital staff hospitals have not been established (115). This lack of standard-
and the patient, and to ensure patient safety, hospitals must have ization limits the ability for benchmarking and comparison of dif-
clear policies and procedures in place to guide the use of CSII in ferent quality-improvement programs and protocols. Further study
the inpatient setting (92). Documents that stipulate contraindications into the development and implementation of appropriate
S120 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123

standardized metrics for hospital glycemic management pro-

grams is warranted. 2. The frequency and timing of bedside CBG monitoring should be individu-
alized for all in-hospital people with diabetes. Monitoring should typi-
cally be performed:
a. Before meals and at bedtime in people who are eating [Grade D,
Transition from hospital to home
Consensus]
b. Every 4 to 6 hours in people who are NPO or receiving continuous
Interventions that ensure continuity of care, such as arranging enteral feeding [Grade D, Consensus]
continuation of care after discharge (97), telephone follow up and c. Every 1 to 2 hours for people on continuous intravenous insulin or
communication with primary providers at discharge (111), have been those who are critically ill [Grade D, Consensus].

associated with a post-discharge reduction in A1C (111). Provid-

3. Provided that their medical conditions, dietary intake and glycemic control
ing people with diabetes and their family or caregivers with written are stable, people with diabetes should be maintained on their pre-
and oral instructions regarding their diabetes management at the hospitalization noninsulin antihyperglycemic agents or insulin regimens
time of hospital discharge will facilitate transition to community [Grade D, Consensus].
care. Comprehensive instructions may include recommendations
4. For hospitalized people with diabetes treated with insulin, a proactive
for timing and frequency of home glucose monitoring; identifica- approach that includes basal, bolus and correction (supplemental) insulin,
tion and management of hypoglycemia; a reconciled medication along with pattern management, should be used to reduce adverse events
list, including insulin and other antihyperglycemic medications; and and improve glycemic control, instead of only correcting high BG with short-
identification and contact information for health-care providers or rapid-acting insulin [Grade A, Level 1A (61,66,102)].

responsible for ongoing diabetes care and adjustment of glucose-

5. For the majority of noncritically ill hospitalized people with diabetes,
lowering medications. Communication of the need for potential preprandial BG targets should be 5.0 to 8.0 mmol/L in conjunction with
adjustments in insulin therapy that may accompany adjustments random BG values <10.0 mmol/L, as long as these targets can be safely
of other medications prescribed at the time of discharge, such as achieved [Grade D, Consensus].
corticosteroids or octreotide, to people with diabetes and their
6. For most medical/surgical critically ill hospitalized people with diabetes
primary care providers is important. with hyperglycemia, a continuous intravenous insulin infusion should be
used to maintain BG <10.0 mmol/L [Grade B, Level 2 (34)] and >6.0 mmol/L
[Grade D, Consensus].

7. For people with diabetes undergoing CABG, a continuous intravenous insulin

Safety infusion protocol targeting intraoperative glycemic levels between 5.5 and
11.1 mmol/L should be used, rather than subcutaneous insulin, to prevent
Hypoglycemia postoperative infections [Grade A, Level 1A (54)].

8. In hospitalized people with diabetes requiring insulin therapy, protocols

Hypoglycemia remains a major barrier to achieving optimal using basal insulin with/without bolus insulin should be used for post-
glycemic control in hospitalized people with diabetes. Standard- operative glycemic management [Grade B, Level 2 (61)].
ized treatment protocols that address mild, moderate and severe
hypoglycemia may help mitigate this risk. Education of health- 9. In hospitalized people with diabetes, hypoglycemia should be mini-
mized. Protocols for hypoglycemia avoidance, recognition and manage-
care workers about factors that increase the risk of hypoglycemia,
ment should be implemented with nurse-initiated treatment, including
such as sudden reduction in oral intake, discontinuation of paren- glucagon for severe hypoglycemia when intravenous access is not readily
teral or enteral nutrition, unexpected transfer from the nursing available [Grade D, Consensus]. Hospitalized people with diabetes at risk
unit after rapid-acting insulin administration or a reduction in of hypoglycemia should have ready access to an appropriate source of
corticosteroid dose (78) are important steps to reduce the risk of glucose (oral or IV) at all times, particularly when NPO or during diag-
nostic procedures [Grade D, Consensus].
hypoglycemia.
10. Programs consisting of the following elements should be implemented
for optimal inpatient diabetes care:
Insulin administration errors a. Interprofessional team-based approach [Grade B, Level 2 (107,108,112)]
b. Health-care professional development regarding in-hospital diabe-
Insulin is considered a high-alert medication and can be tes management [Grade D, Level 4 (95)]
c. Algorithms, order sets and decision support [Grade C, Level 3
associated with risk of harm and severe adverse events. A systems (26,99,105)].
approach that includes pre-printed, approved, unambiguous d. Comprehensive quality assurance initiatives, including institution-
standard orders for insulin administration and/or a computerized wide BG monitoring systems, inpatient education, and transition/
order entry system may help reduce errors in insulin ordering continuity of care and discharge planning [Grade D, Consensus].
(22).
Abbreviations:
BG, blood glucose; CBG, capillary blood glucose; CABG, coronary artery
bypass grafting; CSII, continuous subcutaneous insulin infusion; ICU, inten-
sive care unit; NPH, neutral protamine Hagedorn; POC, point of care; TDD,
RECOMMENDATIONS total daily dose.

1. An A1C should be measured if not done in the 3 months prior to

admission on:
a. All hospitalized people with a history of diabetes to identify indi-
viduals that would benefit from glycemic optimization [Grade D,
Consensus] Other Relevant Guidelines
b. All hospitalized people with newly diagnosed hyperglycemia or those
with diabetes risk factors to identify individuals at risk for ongoing Glycemic Management in Adults With Type 1 Diabetes, p. S80
dysglycemia [Grade C, Level 3 (16)] Pharmacologic Glycemic Management of Type 2 Diabetes in
c. Repeat screening should be performed 6 to 8 weeks post-hospital
discharge for individuals with an A1C 6.0–6.4% [Grade D, Consensus]
Adults, p. S88
d. In-hospital CBG monitoring should be initiated for individuals with Hyperglycemic Emergencies in Adults, p. S109
an A1C ≥6.5% [Grade D, Consensus]. Management of Acute Coronary Syndromes, p. S190
Treatment of Diabetes in People With Heart Failure, p. S196
 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123 S121

Author Disclosures 26. Nirantharakumar K, Chen YF, Marshall T, et al. Clinical decision support systems
in the care of inpatients with diabetes in non-critical care setting: System-
atic review. Diabet Med 2012;29:698–708.
Dr. Halperin reports personal fees from Dexcom, Novo Nordisk, 27. Maynard G, Schnipper JL, Messler J, et al. Design and implementation of a web-
and QHR technologies, outside the submitted work. Dr. Miller reports based reporting and benchmarking center for inpatient glucometrics. J Dia-
personal fees from Eli Lilly, Novo Nordisk, Sanofi, and AstraZeneca; betes Sci Technol 2014;8:630–40.
28. Baker EH, Janaway CH, Philips BJ, et al. Hyperglycaemia is associated with
and grants and personal fees from Boehringer Ingelheim, Janssen, poor outcomes in patients admitted to hospital with acute exacerbations of
Merck, outside the submitted work. Sarah Moore reports personal chronic obstructive pulmonary disease. Thorax 2006;61:284–9.
fees from Diabetes Care Alliance (Boehringer Ingelheim Eli Lilly Alli- 29. McAlister FA, Majumdar SR, Blitz S, et al. The relation between hyperglyce-
mia and outcomes in 2,471 patients admitted to the hospital with community-
ance), and Merck Canada, outside the submitted work. No other acquired pneumonia. Diabetes Care 2005;28:810–15.
authors have anything to disclose. 30. American Diabetes Association. 13. Diabetes care in the hospital. Diabetes Care
2016;39:S99–104.
31. Lewis KS, Kane-Gill SL, Bobek MB, et al. Intensive insulin therapy
for critically ill patients. Ann Pharmacother 2004;38:1243–51.
References 32. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in criti-
cally ill patients. N Engl J Med 2001;345:1359–67.
1. Umpierrez GE, Isaacs SD, Bazargan N, et al. Hyperglycemia: An independent 33. Griesdale DE, de Souza RJ, van Dam RM, et al. Intensive insulin therapy and
marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin mortality among critically ill patients: A meta-analysis including NICE-SUGAR
Endocrinol Metab 2002;87:978–82. study data. CMAJ 2009;180:821–7.
2. Vasa F. Systematic strategies for improved outcomes for the hyperglycemic hos- 34. NICE-SUGAR Study Investigators, Finfer S, Chittock DR, et al. Intensive
pitalized patient with diabetes mellitus. Am J Cardiol 2005;96:41e–6e. versus conventional glucose control in critically ill patients. N Engl J Med
3. Inzucchi SE. Clinical practice. Management of hyperglycemia in the hospital 2009;360:1283–97.
setting. N Engl J Med 2006;355:1903–11. 35. Moghissi ES, Korytkowski MT, DiNardo M, et al. American Association of Clini-
4. Roman SH, Chassin MR. Windows of opportunity to improve diabetes care when cal Endocrinologists and American Diabetes Association consensus state-
patients with diabetes are hospitalized for other conditions. Diabetes Care ment on inpatient glycemic control. Endocr Pract 2009;15:353–69.
2001;24:1371–6. 36. Goldberg PA, Siegel MD, Sherwin RS, et al. Implementation of a safe and effec-
5. Sud M, Wang X, Austin PC, et al. Presentation blood glucose and death, hos- tive insulin infusion protocol in a medical intensive care unit. Diabetes Care
pitalization, and future diabetes risk in patients with acute heart failure syn- 2004;27:461–7.
dromes. Eur Heart J 2015;36:924–31. 37. Rea RS, Donihi AC, Bobeck M, et al. Implementing an intravenous insulin infusion
6. Umpierrez GE, Reyes D, Smiley D, et al. Hospital discharge algorithm based on protocol in the intensive care unit. Am J Health Syst Pharm 2007;64:385–95.
admission HbA1c for the management of patients with type 2 diabetes. Dia- 38. Nazer LH, Chow SL, Moghissi ES. Insulin infusion protocols for critically ill
betes Care 2014;37:2934–9. patients: A highlight of differences and similarities. Endocr Pract 2007;13:137–
7. Perez A, Reales P, Barahona MJ, et al. Efficacy and feasibility of basal-bolus 46.
insulin regimens and a discharge-strategy in hospitalised patients with 39. Ntaios G, Papavasileiou V, Bargiota A, et al. Intravenous insulin treatment in
type 2 diabetes-the HOSMIDIA study. Int J Clin Pract 2014;68:1264–71. acute stroke: A systematic review and meta-analysis of randomized con-
8. Ochoa PS, Terrell BT, Vega JA, et al. Identification of previously undiagnosed trolled trials. Int J Stroke 2014;9:489–93.
diabetes and prediabetes in the inpatient setting using risk factor and 40. Schmeltz LR, DeSantis AJ, Schmidt K, et al. Conversion of intravenous insulin
hemoglobin A1C screening. Ann Pharmacother 2014;48:1434–9. infusions to subcutaneously administered insulin glargine in patients with hyper-
9. Simpson AJ, Krowka R, Kerrigan JL, et al. Opportunistic pathology-based screen- glycemia. Endocr Pract 2006;12:641–50.
ing for diabetes. BMJ Open 2013 (in press). 41. Bode BW, Braithwaite SS, Steed RD, et al. Intravenous insulin infusion therapy:
10. Norhammar A, Tenerz A, Nilsson G, et al. Glucose metabolism in patients with Indications, methods, and transition to subcutaneous insulin therapy. Endocr
acute myocardial infarction and no previous diagnosis of diabetes mellitus: Pract 2004;10(Suppl. 2):71–80.
A prospective study. Lancet 2002;359:2140–4. 42. Kwoun MO, Ling PR, Lydon E, et al. Immunologic effects of acute hyperglyce-
11. O’Sullivan EP, Duignan J, O’Shea P, et al. Evaluating hyperglycaemia in the mia in nondiabetic rats. JPEN J Parenter Enteral Nutr 1997;21:91–5.
hospitalised patient: Towards an improved system for classification and treat- 43. Verhofstad MH, Hendriks T. Complete prevention of impaired anastomotic
ment. Ir J Med Sci 2014;183:65–9. healing in diabetic rats requires preoperative blood glucose control. Br J Surg
12. Miller DB. Glycemic targets in hospital and barriers to attaining them. Can J 1996;83:1717–21.
Diabetes 2014;38:74–8. 44. Golden SH, Peart-Vigilance C, Kao WH, et al. Perioperative glycemic control
13. Greci LS, Kailasam M, Malkani S, et al. Utility of HbA(1c) levels for diabetes and the risk of infectious complications in a cohort of adults with diabetes.
case finding in hospitalized patients with hyperglycemia. Diabetes Care Diabetes Care 1999;22:1408–14.
2003;26:1064–8. 45. McAlister FA, Man J, Bistritz L, et al. Diabetes and coronary artery bypass surgery:
14. Carpenter DL, Gregg SR, Xu K, et al. Prevalence and impact of unknown dia- An examination of perioperative glycemic control and outcomes. Diabetes Care
betes in the ICU. Crit Care Med 2015;43:e541–50. 2003;26:1518–24.
15. Kompoti M, Michalia M, Salma V, et al. Glycated hemoglobin at admission in 46. Thomas MC, Mathew TH, Russ GR, et al. Early peri-operative glycaemic control
the intensive care unit: Clinical implications and prognostic relevance. J Crit and allograft rejection in patients with diabetes mellitus: A pilot study.
Care 2015;30:150–5. Transplantation 2001;72:1321–4.
16. Manley SE, O’Brien KT, Quinlan D, et al. Can HbA1c detect undiagnosed dia- 47. Estrada CA, Young JA, Nifong LW, et al. Outcomes and perioperative hypergly-
betes in acute medical hospital admissions? Diabetes Res Clin Pract cemia in patients with or without diabetes mellitus undergoing coronary artery
2016;115:106–14. bypass grafting. Ann Thorac Surg 2003;75:1392–9.
17. Malcolm JC, Kocourek J, Keely E, et al. Implementation of a screening program 48. Brandt M, Harder K, Walluscheck KP, et al. Coronary artery bypass surgery in
to detect previously undiagnosed dysglycemia in hospitalized patients. Can J diabetic patients. J Card Surg 2004;19:36–40.
Diabetes 2014;38:79–84. 49. Bucerius J, Gummert JF, Walther T, et al. Diabetes in patients undergoing coro-
18. Umpierrez GE, Hellman R, Korytkowski MT, et al. Management of hypergly- nary artery bypass grafting. Impact on perioperative outcome. Z Kardiol
cemia in hospitalized patients in non-critical care setting: An endocrine society 2005;94:575–82.
clinical practice guideline. J Clin Endocrinol Metab 2012;97:16–38. 50. Bucerius J, Gummert JF, Walther T, et al. Impact of diabetes mellitus on cardiac
19. Lewandrowski K, Cheek R, Nathan DM, et al. Implementation of capillary blood surgery outcome. Thorac Cardiovasc Surg 2003;51:11–16.
glucose monitoring in a teaching hospital and determination of program require- 51. Doenst T, Wijeysundera D, Karkouti K, et al. Hyperglycemia during cardiopul-
ments to maintain quality testing. Am J Med 1992;93:419–26. monary bypass is an independent risk factor for mortality in patients under-
20. Rumley AG. Improving the quality of near-patient blood glucose measure- going cardiac surgery. J Thorac Cardiovasc Surg 2005;130:1144.
ment. Ann Clin Biochem 1997;34(Pt 3):281–6. 52. Gandhi GY, Nuttall GA, Abel MD, et al. Intraoperative hyperglycemia and
21. Boyd JC, Bruns DE. Quality specifications for glucose meters: Assessment by perioperative outcomes in cardiac surgery patients. Mayo Clin Proc
simulation modeling of errors in insulin dose. Clin Chem 2001;47:209–14. 2005;80:862–6.
22. Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order 53. Ouattara A, Lecomte P, Le Manach Y, et al. Poor intraoperative blood glucose
entry and a team intervention on prevention of serious medication errors. JAMA control is associated with a worsened hospital outcome after cardiac surgery
1998;280:1311–16. in diabetic patients. Anesthesiology 2005;103:687–94.
23. Desachy A, Vuagnat AC, Ghazali AD, et al. Accuracy of bedside glucometry in 54. Boreland L, Scott-Hudson M, Hetherington K, et al. The effectiveness of tight
critically ill patients: Influence of clinical characteristics and perfusion index. glycemic control on decreasing surgical site infections and readmission rates
Mayo Clin Proc 2008;83:400–5. in adult patients with diabetes undergoing cardiac surgery: A systematic review.
24. Critchell CD, Savarese V, Callahan A, et al. Accuracy of bedside capillary blood Heart Lung 2015;44:430–40.
glucose measurements in critically ill patients. Intensive Care Med 55. Raucoules-Aime M, Lugrin D, Boussofara M, et al. Intraoperative glycaemic
2007;33:2079–84. control in non-insulin-dependent and insulin-dependent diabetes. Br J Anaesth
25. Petersen JR, Graves DF, Tacker DH, et al. Comparison of POCT and central labo- 1994;73:443–9.
ratory blood glucose results using arterial, capillary, and venous samples from 56. Hemmerling TM, Schmid MC, Schmidt J, et al. Comparison of a continuous
MICU patients on a tight glycemic protocol. Clin Chim Acta 2008;396:10–13. glucose-insulin-potassium infusion versus intermittent bolus application of
S122 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123

insulin on perioperative glucose control and hormone status in insulin- 84. Donihi AC, Raval D, Saul M, et al. Prevalence and predictors of corticosteroid-
treated type 2 diabetics. J Clin Anesth 2001;13:293–300. related hyperglycemia in hospitalized patients. Endocr Pract 2006;12:358–
57. Christiansen CL, Schurizek BA, Malling B, et al. Insulin treatment of the insulin- 62.
dependent diabetic patient undergoing minor surgery. Continuous intrave- 85. Gosmanov AR, Goorha S, Stelts S, et al. Management of hyperglycemia in dia-
nous infusion compared with subcutaneous administration. Anaesthesia betic patients with hematologic malignancies during dexamethasone therapy.
1988;43:533–7. Endocr Pract 2013;19:231–5.
58. de Vries FE, Gans SL, Solomkin JS, et al. Meta-analysis of lower perioperative 86. Ruiz de Adana MS, Colomo N, Maldonado-Araque C, et al. Randomized clini-
blood glucose target levels for reduction of surgical-site infection. Br J Surg cal trial of the efficacy and safety of insulin glargine vs. NPH insulin as basal
2017;104:e95–105. insulin for the treatment of glucocorticoid induced hyperglycemia using con-
59. Suto C, Hori S, Kato S, et al. Effect of perioperative glycemic control in pro- tinuous glucose monitoring in hospitalized patients with type 2 diabetes and
gression of diabetic retinopathy and maculopathy. Arch Ophthalmol respiratory disease. Diabetes Res Clin Pract 2015;110:158–65.
2006;124:38–45. 87. Grommesh B, Lausch MJ, Vannelli AJ, et al. Hospital insulin protocol aims for
60. Kamio S, Kawasaki R, Yamashita H. Influence of systemic conditions and gly- glucose control in glucocorticoid-induced hyperglycemia. Endocr Pract
cemic control on complications of vitrectomy for diabetic retinopathy. Folia 2016;22:180–9.
Ophthalmologica Japonica 2004;55:105–9, (Japanese). 88. Cook CB, Beer KA, Seifert KM, et al. Transitioning insulin pump therapy from
61. Umpierrez GE, Smiley D, Jacobs S, et al. Randomized study of basal-bolus insulin the outpatient to the inpatient setting: A review of 6 years’ experience with
therapy in the inpatient management of patients with type 2 diabetes 253 cases. J Diabetes Sci Technol 2012;6:995–1002.
undergoing general surgery (RABBIT 2 Surgery). Diabetes Care 2011;34:256– 89. Noschese ML, DiNardo MM, Donihi AC, et al. Patient outcomes after imple-
61. mentation of a protocol for inpatient insulin pump therapy. Endocr Pract
62. Huang QX, Lou FC, Wang P, et al. Basal insulin therapy strategy is superior to 2009;15:415–24.
premixed insulin therapy in the perioperative period blood glucose manage- 90. Anstey J, Yassaee A, Solomon A. Clinical outcomes of adult inpatients treated
ment. Chin Med J 2013;126:4030–6. with continuous subcutaneous insulin infusion for diabetes mellitus: A sys-
63. Coan KE, Schlinkert AB, Beck BR, et al. Clinical inertia during postoperative man- tematic review. Diabet Med 2015;32:1279–88.
agement of diabetes mellitus: Relationship between hyperglycemia and insulin 91. Leonhardi BJ, Boyle ME, Beer KA, et al. Use of continuous subcutaneous
therapy intensification. J Diabetes Sci Technol 2013;7:880–7. insulin infusion (insulin pump) therapy in the hospital: A review of one
64. Yogi-Morren D, Lansang MC. Management of patients with type 1 diabetes in institution’s experience. J Diabetes Sci Technol 2008;2:948–62.
the hospital topical collection on hospital management of diabetes. Curr Diab 92. Bailon RM, Partlow BJ, Miller-Cage V, et al. Continuous subcutaneous insulin
Rep 2014;14:458. infusion (insulin pump) therapy can be safely used in the hospital in select
65. Meyer C, Boron A, Plummer E, et al. Glulisine versus human regular insulin patients. Endocr Pract 2009;15:24–9.
in combination with glargine in noncritically ill hospitalized patients with 93. Corney SM, Dukatz T, Rosenblatt S, et al. Comparison of insulin pump therapy
type 2 diabetes: A randomized double-blind study. Diabetes Care 2010;33:2496– (continuous subcutaneous insulin infusion) to alternative methods for
501. perioperative glycemic management in patients with planned postoperative
66. Umpierrez GE, Smiley D, Zisman A, et al. Randomized study of basal-bolus admissions. J Diabetes Sci Technol 2012;6:1003–15.
insulin therapy in the inpatient management of patients with type 2 diabe- 94. Fresa R, Visalli N, Di Blasi V, et al. Experiences of continuous subcutaneous
tes (RABBIT 2 Trial). Diabetes Care 2007;3:2181–6. insulin infusion in pregnant women with type 1 diabetes during delivery from
67. Lee YY, Lin YM, Leu WJ, et al. Sliding-scale insulin used for blood glucose control: four Italian centers: A retrospective observational study. Diabetes Technol Ther
A meta-analysis of randomized controlled trials. Metabolism 2015;64:1183– 2013;15:328–34.
92. 95. Moghissi ES, Inzucchi SE, Mann KV, et al. Hyperglycemia grand rounds: Descrip-
68. Thomann R, Schütz P, Muller B, et al. Evaluation of an algorithm for intensive tive findings of outcomes from a continuing education intervention to improve
subcutaneous insulin therapy in noncritically ill hospitalised patients with glycemic control and prevent hypoglycemia in the hospital setting. Hosp Pract
hyperglycaemia in a randomised controlled trial. Swiss Med Wkly 2013;143: (1995) 2015;43:270–6.
69. Umpierrez GE, Smiley D, Hermayer K, et al. Randomized study comparing a 96. Schnipper JL, Ndumele CD, Liang CL, et al. Effects of a subcutaneous insulin
Basal-bolus with a basal plus correction insulin regimen for the hospital man- protocol, clinical education, and computerized order set on the quality of inpa-
agement of medical and surgical patients with type 2 diabetes: Basal plus trial. tient management of hyperglycemia: Results of a clinical trial. J Hosp Med
Diabetes Care 2013;36:2169–74. 2009;4:16–27.
70. Mader JK, Neubauer KM, Schaupp L, et al. Efficacy, usability and sequence of 97. Bar-Dayan Y, Landau Z, Boaz M, et al. Inpatient hyperglycaemia improve-
operations of a workflow-integrated algorithm for basal-bolus insulin therapy ment quality program. Int J Clin Pract 2014;68:495–502.
in hospitalized type 2 diabetes patients. Diabetes Obes Metab 2014;16:137– 98. Munoz M, Pronovost P, Dintzis J, et al. Implementing and evaluating a mul-
46. ticomponent inpatient diabetes management program: Putting research into
71. Li X, Du T, Li W, et al. Efficacy and safety of weight-based insulin glargine dose practice. Jt Comm J Qual Patient Saf 2012;38:195–206.
titration regimen compared with glucose level- and current dose-based regi- 99. Maynard G, Lee J, Phillips G, et al. Improved inpatient use of basal insulin,
mens in hospitalized patients with type 2 diabetes: A randomized, con- reduced hypoglycemia, and improved glycemic control: Effect of structured
trolled study. Clin Ther 2014;36:1269–75. subcutaneous insulin orders and an insulin management algorithm. J Hosp Med
72. Inagaki N, Goda M, Yokota S, et al. Effects of baseline blood pressure and low- 2009;4:3–15.
density lipoprotein cholesterol on safety and efficacy of canagliflozin in Japa- 100. Noschese M, Donihi AC, Koerbel G, et al. Effect of a diabetes order set on
nese patients with type 2 diabetes mellitus. Adv Ther 2015;32:1085–103. glycaemic management and control in the hospital. Qual Saf Health Care
73. Bellido V, Suarez L, Rodriguez MG, et al. Comparison of basal-bolus and pre- 2008;17:464–8.
mixed insulin regimens in hospitalized patients with type 2 diabetes. 101. Wexler DJ, Shrader P, Burns SM, et al. Effectiveness of a computerized insulin
Diabetes Care 2015;38:2211–16. order template in general medical inpatients with type 2 diabetes: A cluster
74. Zhang T, Lin M, Li W, et al. Comparison of the efficacy and safety of insulin randomized trial. Diabetes Care 2010;33:2181–3.
detemir and insulin glargine in hospitalized patients with type 2 diabetes: 102. Christensen MB, Gotfredsen A, Nørgaard K. Efficacy of basal-bolus insulin regi-
A randomized crossover trial. Adv Ther 2016;33:178–85. mens in the inpatient management of non-critically ill patients with type 2
75. Wesorick D, O’Malley C, Rushakoff R, et al. Management of diabetes and diabetes: A systematic review and meta-analysis. Diabetes Metab Res Rev
hyperglycemia in the hospital: A practical guide to subcutaneous insulin use 2017;33.
in the non-critically ill, adult patient. J Hosp Med 2008;3:17–28. 103. Schnipper JL, Liang CL, Ndumele CD, et al. Effects of a computerized order set
76. Mendez CE, Umpierrez GE. Pharmacotherapy for hyperglycemia in noncritically on the inpatient management of hyperglycemia: A cluster-randomized
ill hospitalized patients. Diabetes Spectr 2014;27:180–8. controlled trial. Endocr Pract 2010;16:209–18.
77. Pasquel FJ, Gianchandani R, Rubin DJ, et al. Efficacy of sitagliptin for the hos- 104. Neubauer KM, Mader JK, Holl B, et al. Standardized glycemic management with
pital management of general medicine and surgery patients with type 2 dia- a computerized workflow and decision support system for hospitalized patients
betes (Sita-Hospital): A multicentre, prospective, open-label, non-inferiority with type 2 diabetes on different wards. Diabetes Technol Ther 2015;17:685–
randomised trial. Lancet Diabetes Endocrinol 2017;5:125–33. 92.
78. Clement S, Braithwaite SS, Magee MF, et al. Management of diabetes and hyper- 105. Lin SD, Tu ST, Lin MJ, et al. A workable model for the management of hyper-
glycemia in hospitals. Diabetes Care 2004;27:553–91. glycemia in non-critically ill patients in an Asian population. Postgrad Med
79. Umpierrez GE, Palacio A, Smiley D. Sliding scale insulin use: Myth or insan- 2015;127:796–800.
ity? Am J Med 2007;120:563–7. 106. Aloi J, Bode BW, Ullal J, et al. Comparison of an electronic glycemic manage-
80. Curll M, Dinardo M, Noschese M, et al. Menu selection, glycaemic control and ment system versus provider-managed subcutaneous basal bolus insulin therapy
satisfaction with standard and patient-controlled consistent carbohydrate meal in the hospital setting. J Diabetes Sci Technol 2016;11:12–16.
plans in hospitalised patients with diabetes. Qual Saf Health Care 2010;19:355–9. 107. Levetan CS, Salas JR, Wilets IF, et al. Impact of endocrine and diabetes team
81. Jakoby MG, Nannapaneni N. An insulin protocol for management of hyper- consultation on hospital length of stay for patients with diabetes. Am J Med
glycemia in patients receiving parenteral nutrition is superior to ad hoc man- 1995;99:22–8.
agement. JPEN J Parenter Enteral Nutr 2012;36:183–8. 108. Koproski J, Pretto Z, Poretsky L. Effects of an intervention by a diabetes
82. Sajbel TA, Dutro MP, Radway PR. Use of separate insulin infusions with total team in hospitalized patients with diabetes. Diabetes Care 1997;20:
parenteral nutrition. JPEN J Parenter Enteral Nutr 1987;11:97–9. 1553–5.
83. Korytkowski MT, Salata RJ, Koerbel GL, et al. Insulin therapy and glycemic control 109. Moraes MA, Rodrigues J, Cremonesi M, et al. Management of diabetes by a
in hospitalized patients with diabetes during enteral nutrition therapy: A ran- healthcare team in a cardiology unit: A randomized controlled trial. Clinics
domized controlled clinical trial. Diabetes Care 2009;32:594–6. 2013;68:1400–7.
 J. Malcolm et al. / Can J Diabetes 42 (2018) S115–S123 S123

110. Sampson MJ, Crowle T, Dhatariya K, et al. Trends in bed occupancy for inpa- Literature Review Flow Diagram for Chapter 16: In-Hospital
tients with diabetes before and after the introduction of a diabetes inpatient
Management of Diabetes
specialist nurse service. Diabet Med 2006;23:1008–15.
111. Dungan K, Lyons S, Manu K, et al. An individualized inpatient diabetes edu-
cation and hospital transition program for poorly controlled hospitalized patients
with diabetes. Endocr Pract 2014;20:1265–73. Citations identified through Additional citations identified
database searches through other sources
112. Mackey PA, Boyle ME, Walo PM, et al. Care directed by a specialty-trained nurse
N=18,354 N=48
practioner or physician assistant can overcome clinical inertia in manage-
ment of inpatient diabetes. Endocr Pract 2014;20:112–19.
113. Rodger ED. Diabetic patients survey of in-hospital experience. Edmonton:
Albertal Health Services, 2015. http://www.albertahealthservices.ca/assets/ Citations after duplicates removed
about/scn/ahs-scn-don-inpatient-diabetes-survey-results.pdf. N=12,677
114. Thompson R, Schreuder AB, Wisse B, et al. Improving insulin ordering safely:
The development of an inpatient glycemic control program. J Hosp Med
2009;4:E30–5.
115. Cook CB, Wellik KE, Kongable GL, et al. Assessing inpatient glycemic control: Title & abstract screening Citations excluded*
What are the next steps? J Diabetes Sci Technol 2012;6:421–7. N=4,954 N=4,448
116. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic
reviews and meta-analyses: The PRISMA statement. PLoS Med 2009;6:e1000097.

Full-text screening Citations excluded*

for eligibility
N=294
N=506

Full-text reviewed Citations excluded*

by chapter authors N=201
N=212

Studies requiring
new or revised
recommendations
N=11

*Excluded based on: population, intervention/exposure, comparator/

control or study design.

From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group

(2009). Preferred Reporting Items for Systematic Reviews and Meta-
Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097.
doi:10.1371/journal.pmed1000097 (116).

For more information, visit www.prisma-statement.org.