0
Clinical Guidelines |

Intensive Insulin Therapy in Hospitalized Patients: A Systematic Review FREE

Devan Kansagara, MD, MCR; Rongwei Fu, PhD; Michele Freeman, MPH; Fawn Wolf, MD; and Mark Helfand, MD, MPH
[+] Article and Author Information

From Oregon Health & Science University, Portland Veterans Affairs Medical Center, and Portland Diabetes & Endocrinology Center, Portland, Oregon.


Acknowledgment: The authors thank Andrew Hamilton, MS, MLS, for designing the search strategy and Linda Humphrey, MD, and Roger Chou, MD, for their expert advice and continuous support.

Financial Support: By the Department of Veterans Affairs, Veterans Health Administration (VHA Project ESP 05-225, VA #01-0206). The American College of Physicians provided funding for the preparation of this manuscript.

Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M10-1583.

Requests for Single Reprints: Devan Kansagara, MD, MCR, Portland Veterans Affairs Medical Center, 3710 SW U.S. Veterans Hospital Road, Mailcode R&D 71, Portland, OR 97239; e-mail, kansagar@ohsu.edu.

Current Author Addresses: Drs. Kansagara and Helfand and Ms. Freeman: Portland Veterans Affairs Medical Center, 3710 SW U.S. Veterans Hospital Road, Mailcode R&D 71, Portland, OR 97239.

Dr. Fu: Oregon Health & Science University, 3181 SW Sam Jackson Park Road, Mailcode CB669, Portland, OR 97239.

Dr. Wolf: Portland Diabetes & Endocrinology Center, 1130 NW 22nd Avenue, Suite 400, Portland, OR, 97210.

Author Contributions: Conception and design: D. Kansagara, F. Wolf.

Analysis and interpretation of the data: D. Kansagara, M. Helfand, R. Fu, M. Freeman, F. Wolf.

Drafting of the article: D. Kansagara.

Critical revision of the article for important intellectual content: D. Kansagara, M. Helfand.

Final approval of the article: D. Kansagara, M. Helfand, F. Wolf.

Statistical expertise: R. Fu.

Obtaining of funding: D. Kansagara.

Administrative, technical, or logistic support: M. Helfand, M. Freeman.

Collection and assembly of data: D. Kansagara, M. Freeman, F. Wolf.


Ann Intern Med. 2011;154(4):268-282. doi:10.7326/0003-4819-154-4-201102150-00008
Text Size: A A A

Background: The benefits and harms of intensive insulin therapy (IIT) titrated to strict glycemic targets in hospitalized patients remain uncertain.

Purpose: To evaluate the benefits and harms of IIT in hospitalized patients.

Data Sources: MEDLINE and Cochrane Database of Systematic Reviews from 1950 to January 2010, reference lists, experts, and unpublished sources.

Study Selection: English-language randomized, controlled trials comparing protocols titrated to strict or less strict glycemic targets.

Data Extraction: Two reviewers independently abstracted data from each study on sample, setting, glycemic control interventions, glycemic targets, mean glucose levels achieved, and outcomes. Results were grouped by patient population or setting. A random-effects model was used to combine trial data on short-term mortality (≤28 days), long-term mortality (90 or 180 days), infection, length of stay, and hypoglycemia. The Grading of Recommendations Assessment, Development, and Evaluation system was used to rate the overall body of evidence for each outcome.

Data Synthesis: In a meta-analysis of 21 trials in intensive care unit, perioperative care, myocardial infarction, and stroke or brain injury settings, IIT did not affect short-term mortality (relative risk, 1.00 [95% CI, 0.94 to 1.07]). No consistent evidence showed that IIT reduced long-term mortality, infection rates, length of stay, or the need for renal replacement therapy. No evidence of benefit from IIT was reported in any hospital setting, although the best evidence for lack of benefit was in intensive care unit settings. Data combined from 10 trials showed that IIT was associated with a high risk for severe hypoglycemia (relative risk, 6.00 [CI, 4.06 to 8.87]; P < 0.001). Risk for IIT-associated hypoglycemia was increased in all hospital settings.

Limitations: Methodological shortcomings and inconsistencies limit the data in perioperative care, myocardial infarction, and stroke or brain injury settings. Differences in insulin protocols and patient and hospital characteristics may affect generalizability across treatment settings.

Conclusion: No consistent evidence demonstrates that IIT targeted to strict glycemic control compared with less strict glycemic control improves health outcomes in hospitalized patients. Furthermore, IIT is associated with an increased risk for severe hypoglycemia.

Primary Funding Source: U.S. Department of Veterans Affairs Health Services Research and Development Service.

Hyperglycemia is common among medical and surgical inpatients with and without known diabetes (12) and is associated with poor outcomes across various inpatient subpopulations (1, 37). Hyperglycemia may be a marker of severe, acute illness or may worsen outcomes by contributing to inflammation, oxidative stress, poor immune function, and endothelial dysfunction (89). Initial studies of adjustable insulin infusions to decrease blood glucose levels raised interest in inpatient glycemic control strategies (1011), and several organizations called for implementing intensive insulin therapy (IIT) strategies using adjustable insulin infusions titrated to strict glycemic targets in the intensive care unit (ICU) (9, 12). Despite early evidence of benefit from IIT (1316), many subsequent trials, including the largest IIT trial to date (17), have not found a consistent benefit.

We conducted a systematic review of studies evaluating the use of IIT to achieve glycemic control in hospitalized patients. The objectives of this review are to evaluate the benefits and harms of IIT and to discuss reasons for discrepancies in the literature. The American College of Physicians will use this review to guide recommendations for management of inpatient hyperglycemia.

The U.S. Department of Veterans Affairs' Evidence-based Synthesis Program commissioned the full report on which this review is based (18). This review updates the previous review and addresses 2 key questions: 1) Does the use of IIT to achieve strict glycemic control compared with less strict glycemic control improve health outcomes in inpatients in surgical intensive care, medical intensive care, general medicine ward, and perioperative settings or in inpatients with acute myocardial infarction (MI) or acute stroke? 2) What are the harms of strict glycemic control in these subpopulations?

Data Sources and Searches

We searched MEDLINE and the Cochrane Database of Systematic Reviews for literature published from database inception in 1950 through January 2010. We obtained additional articles from consultation with experts and from reference lists of pertinent studies, reviews, and editorials. Table 1 in the Supplement describes the search strategies in detail. We searched ClinicalTrials.gov for information about unpublished studies. All citations were imported into the electronic database EndNote X2 (Thomson Reuters, New York, New York).

Study Selection

Three investigators reviewed the abstracts of citations identified from literature searches. We retrieved full-text articles of potentially relevant abstracts for further review. Each article was reviewed by using the eligibility criteria shown in Table 2 in the Supplement. Eligible articles were published in English and provided primary data on the use of IIT in hospitalized patients. We excluded studies that evaluated fixed-dose insulin and glucose–insulin–potassium infusions.

To evaluate the efficacy of and risk for hypoglycemia associated with IIT in hospitalized patients, we included randomized, controlled trials that reported at least one of the following prespecified outcomes: death, cardiovascular events, congestive heart failure, disability, wound infection, sepsis, or renal failure requiring hemodialysis. We defined “perioperative trials” as those in which IIT was begun before, during, or immediately after surgery and was discontinued less than 24 hours after surgery. We included studies of patients with MI during the postthromblytic era (that is, 1995 or later).

Because the safety of IIT may vary on the basis of intervention and implementation characteristics, we evaluated hypoglycemia rates in controlled and uncontrolled studies of IIT protocols, even if the studies did not report other health outcomes (see the Appendix for study selection details).

Data Extraction and Quality Assessment

From each study, we abstracted the following characteristics: study design, objectives, setting, demographics (sex, age, baseline illness), participant eligibility and exclusion criteria, number of participants, years of enrollment, duration of follow-up, study and comparator interventions, method used to monitor blood glucose levels, target range for blood glucose level control, outcomes measured, analytic method used, variables adjusted in the analysis, results of the study and mean blood glucose levels achieved in each group, information on concomitant therapy or nutrition, occurrence of hypoglycemia in each group, and any other adverse events.

The quality of each study was rated as good, fair, or poor on the basis of U.S. Preventive Services Task Force criteria (see Table 3 in the Supplement) (19). When reviewers disagreed about quality rating, consensus was reached through discussion with all authors.

Data Synthesis and Analysis

The primary outcome of interest was short-term mortality, defined as death occurring within 28 days of or during the ICU or hospital stay. If studies reported more than 1 of these outcomes, we preferentially used 28-day mortality for the analysis, followed by hospital or ICU mortality. We conducted a sensitivity analysis based on the definition of short-term mortality. Secondary outcomes included 90- or 180-day mortality, infection, length of stay, and hypoglycemia. For each outcome, we abstracted the number of events and total participants from each treatment group and obtained a pooled estimate of relative risk (RR) by using a random-effects model (20). Statistical heterogeneity was assessed by using the Cochran Q test and the I2 statistic (21). All analyses were done by using Stata software, version 10.0 (StataCorp, College Station, Texas).

We conducted prespecified subgroup analyses comparing ICU studies with non-ICU studies and did sensitivity analyses on the following aspects: 1) the proportion of diabetic patients included, which we calculated by using 25% as a cut point based on a natural division in the included studies; 2) the mean blood glucose level achieved in the intervention group, which we calculated by using a blood glucose level of 6.7 mmol/L (120 mg/dL) as the cut point because a lower threshold (≤6.1 mmol/L [≤110 mg/dL]) would have yielded only 1 study; and 3) study quality.

Rating the Body of Evidence

We assessed the overall quality of evidence for outcomes by using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) system (22). The overall quality of evidence for each outcome is rated as high, moderate, low, or very low on the basis of the risk for bias, consistency, precision, directness, and other characteristics of the body of evidence.

Role of the Funding Source

The U.S. Department of Veterans Affairs Health Services Research and Development Service supported this review but had no role in the design, conduct, analysis, or submission of the manuscript for publication.

Literature Flow

Our literature search (Appendix Figure 1) identified 3055 publications, including 3 unpublished or ongoing trials (2325). Of the 461 articles that we selected for full-text review, 31 trials conducted among perioperative or critically ill patients or patients with acute MI or stroke were included. We also found 29 insulin protocol studies not reporting health outcomes (Appendix).

Grahic Jump Location
Appendix Figure 1.
Literature search and selection.

MI = myocardial infarction; MICU = medical intensive care unit; SICU = surgical intensive care unit.

Grahic Jump Location
Study Limitations

Because IIT requires intensive nursing care and monitoring, none of the trials was blinded. Therefore, unintended co-interventions that could influence patient outcomes, such as wound care, catheter care, or ventilator weaning, were possible. Because no trial reported sufficient nursing care information to decrease this risk for bias, no study was assigned a quality rating of good. Ten trials had additional methodological flaws that further increased the risk for bias (Table 1 and Tables 4 to 6 in the Supplement). For example, 3 ICU trials had important differences in baseline patient characteristics and incompletely reported patient selection, allocation, blinding, or description of outcome assessment, which suggested that the groups were not comparable and that outcomes may have been differentially assessed (2628). We also considered whether blood glucose levels achieved or rates of hypoglycemia were not reported, because these data were important in understanding the balance of benefits and harms (2934).

Table Jump PlaceholderTable 1.  Trials in Intensive Care Units
Effects of IIT on Mortality

Across and within subgroups, IIT had an overall neutral effect on mortality, although the strength of the evidence varied among subgroups. Twenty-one randomized, controlled trials that comprised 14 768 inpatients reported at least 1 short-term mortality event in the treatment or control group. A meta-analysis of these trials (Figure) shows that IIT did not affect short-term mortality (RR, 1.00 [95% CI, 0.94 to 1.07]), with no statistical heterogeneity among studies (I2 = 0.0%; P = 0.46). Stratifying trials according to the blood glucose level achieved in the treatment group (<6.7 mmol/L [<120 mg/dL]) or whether the percentage of diabetic patients was less than 25% (RR, 1.02 [CI, 0.95 to 1.10]) or more than 25% (RR, 0.90 [CI, 0.77 to 1.06]) did not significantly change results (Appendix Figure 2). Sensitivity analyses showed no effect of short-term mortality or study quality on the results, after exclusion of studies rated as poor quality (RR, 1.00 [CI, 0.94 to 1.07]; I2 = 20.2%).

Grahic Jump Location
Figure.
Short-term mortality in studies of intensive insulin therapy, by inpatient setting.

Short-term mortality includes death occurring within 28 d of or during the ICU or hospital stay; we used 28-d mortality in the meta-analysis when a study reported >1 outcome. Events is the number of deaths among participants in the treatment and control groups. CABG = coronary artery bypass graft; CVA = cerebrovascular accident; ICU = intensive care unit; MI = myocardial infarction; MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 2.
Short-term mortality in studies of intensive insulin therapy, by the mean glucose level achieved in the intervention group.

Short-term mortality includes death occurring within 28 d of or during the ICU or hospital stay; we used 28-d mortality in the meta-analysis when a study reported >1 outcome. Events is the number of deaths among participants in the treatment and control groups. NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study.

* During the intraoperative period in Butterworth and colleagues' study (45), the treatment group achieved mean blood glucose levels of approximately 6.1 to 10.2 mmol/L (110 to 185 mg/dL). Glucose levels were not reported in Yang and colleagues' study (34).

Grahic Jump Location

Thirteen studies found that IIT did not reduce 90- or 180-day mortality (RR, 1.06 [CI, 0.99 to 1.12), with no statistical heterogeneity (I2 = 0.0%; P = 0.57) (Appendix Figure 3). Subgroup results are discussed below.

Grahic Jump Location
Appendix Figure 3.
Mortality at 90 or 180 d in studies of intensive insulin therapy, by inpatient setting.

Events is the number of deaths among participants in the treatment and control groups. CABG = coronary artery bypass graft; CVA = cerebrovascular accident; ICU = intensive care unit; MI = myocardial infarction; MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

Grahic Jump Location
ICU Settings

Of the 13 trials conducted in ICUs (Table 1), 6 reported data on critically ill patients in the surgical ICU (SICU) (1617, 2627, 3536), 6 reported data on critically ill patients in the medical ICU (MICU) (17, 28, 35, 3739), and 3 included a mixed SICU and MICU population for whom specific ICU subgroup data were not available (4042). A meta-analysis of all 12 ICU trials (Figure) found that IIT did not affect short-term mortality (RR, 0.98 [CI, 0.89 to 1.09]; I2 = 36.0%; P = 0.10).

Patients in the SICU.

Four trials rated as fair quality conducted in patients in the SICU showed no mortality benefit of using IIT to achieve normal blood glucose levels (5.1 to 6.9 mmol/L [92 to 125 mg/dL]) (17, 26, 3536). In the largest of these trials, the multicenter NICE-SUGAR (Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation) study (17), IIT was associated with excess 90-day mortality in patients in the SICU (RR, 1.31 [CI, 1.07 to 1.61]).

Conversely, Van den Berghe and colleagues' trial involving patients in the SICU (16) was discontinued early after all-cause ICU mortality was found to be significantly lower in the IIT treatment group (4.6% vs. 8%; RR, 0.58 [CI, 0.38 to 0.78]) (Table 1). The short-term mortality benefit was limited to the subgroup of patients who required 5 or more days of ICU care (10.6% vs. 20.2%; P = 0.005), and long-term mortality, defined as death that occurs within 90 or 180 days of the ICU or hospital stay, did not differ between the 2 groups (43). A recent small, poor-quality trial (27) found a mortality benefit (2% vs. 5%; P = 0.044) associated with a very modest decrease in blood glucose level.

Patients in the MICU.

Among 5 fair-quality trials and 1 poor-quality trial involving patients in the MICU (17, 28, 35, 3739), only 1 study (28) showed a benefit from achieving strict glycemic control (6.2 to 6.4 mmol/L [111 to 115 mg/dL]). Five trials (17, 28, 35, 3738) excluded patients who were not expected to require prolonged intensive care. In 1 trial (37), in-hospital mortality decreased in the subgroup of patients who remained in the ICU for at least 3 days (RR, 0.82 [CI not given]; P = 0.009), whereas a trend toward increased mortality was seen in patients with shorter ICU stays (RR, 1.09 [CI, 0.9 to 1.32]).

Mixed ICU Populations.

Five fair-quality trials (17, 35, 4042) that included mixed MICU and SICU populations also demonstrated no overall mortality benefit by using IIT to achieve normal blood glucose levels of 6.3 to 6.7 mmol/L (113 to 120 mg/dL). Two of these trials had MICU and SICU subgroup–specific information reported earlier (17, 35). The large multicenter NICE-SUGAR trial found that IIT was associated with an increase in 90-day mortality (RR, 1.14 [CI, 1.02 to 1.28]), with approximately 1 excess death for 39 patients treated with IIT.

Patients Receiving Perioperative Care

Four fair-quality trials (4447) and 3 poor-quality trials (2931) evaluating perioperative IIT in patients mainly undergoing cardiac surgery (Table 1 in the Supplement) found no short-term mortality benefit, but mortality rates were low.

Patients With MI

In 5 of 6 trials including patients with acute MI (32, 4851), IIT was not associated with a mortality benefit (Table 5 in the Supplement). A frequently cited older trial that may be only minimally applicable to current patients with MI (13) compared insulin infusion plus long-term postdischarge insulin therapy with usual care and found a mortality reduction at 1 year (18.6% vs. 26.1%; RR, 0.69 [CI, 0.49 to 0.96]; P = 0.027). Whether the benefit was related to the acute intervention or the longer-term insulin therapy remains uncertain. A trial designed to address this uncertainty (51) found no mortality benefit, but important baseline characteristics were not comparable among groups, resulting in a poor quality rating.

Patients With Stroke and Acute Brain Injury

Among the few relevant studies involving patients with stroke and acute brain injury (3334, 5254), use of IIT showed no mortality benefit (Table 6 in the Supplement). The largest trial of patients with stroke (33) reported blood glucose data for less than 50% of the participants, and 1 trial (34) reported no data on blood glucose levels achieved using ITT.

Patients in General Medicine Wards

No trials evaluated the effects of IIT on health outcomes in patients in general medical wards.

Effects of IIT on Infection

Sixteen trials evaluated the effects of IIT on the incidence of infection (Appendix Figure 4). The definition of infection varied across studies. Nine studies reported sepsis as an outcome and found a marginally significant 21% decrease in the risk for sepsis with IIT (RR, 0.79 [CI, 0.62 to 1.00]), although heterogeneity among trials was significant (I2 = 53.1%; P = 0.029). A sensitivity analysis suggested that the benefit and heterogeneity were derived solely from Van den Berghe and colleagues' trial of patients in the SICU (16) (RR, 0.89 [CI, 0.74 to 1.09]; I2 = 29.0%; P = 0.20). Most other trials, including the large NICE-SUGAR study (17), found no effect of IIT on sepsis. A pooled analysis of the other 7 studies reporting the occurrence of wound infections, urinary tract infections, pneumonia, or a combination of these events found a neutral effect of IIT on infection (RR, 0.68 [CI, 0.36 to 1.30]), although heterogeneity among these studies was also significant (I2 = 56.3%; P = 0.033) (Appendix Figure 4).

Grahic Jump Location
Appendix Figure 4.
Effects of intensive insulin therapy on rates of infection in various inpatient settings.

We included inpatients in the MICU, SICU, and perioperative settings as well as patients with stroke or acute brain injury. Events is the number of participants with 1 or more infections in the treatment and control groups. MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

* Includes wound infections, urinary tract infections, pneumonia, or a combination of these conditions.

† Data include only cardiopulmonary bypass subgroup.

‡ Data include only off-pump cardiopulmonary bypass subgroup.

Grahic Jump Location
Effects of IIT on Inpatient Length of Stay

The effects of IIT on hospital and ICU length of stay are uncertain. Trial results were heterogeneous (I2 = 95.3% and P < 0.001 in 13 trials; I2 = 69.9% and P < 0.001 in 17 trials), and results within subgroups were conflicting. Trials in mixed MICU and SICU settings, including the NICE-SUGAR study (17, 35, 4042), found that IIT did not affect hospital length of stay (combined effect size, 0.008 day [CI, −0.836 to 0.853 day]; I2 = 0.0%; P = 0.93) or ICU length of stay (combined effect size, −0.039 day [CI, −0.335 to 0.257 day]; I2 = 0.0%; P = 0.76). However, 4 trials of patients in the SICU (2627, 36, 55) found a decrease in ICU length of stay with IIT (combined effect size, −1.484 days [CI, −2.233 to −0.734 day]; I2 = 50.0%; P = 0.112).

Effects of IIT on Other Outcomes

The effects of IIT on other outcomes were largely neutral (Table 1 and Tables 4 to 6 in the Supplement). Eight of 9 trials (17, 27, 35, 3738, 4042) showed that IIT had no effect on the need for new renal replacement therapy. However, Van den Berghe and colleagues' trial of patients in the SICU (16) showed that IIT was associated with a reduced need for new renal replacement (4.8% vs. 8.2%; P < 0.007). One trial in patients with stroke (34) found significantly fewer patients in the IIT group who were severely disabled 6 months later, but no effects on long-term disability were shown in 3 other trials in patients with stroke or acute brain injury (33, 5253) or in 2 trials in neurosurgical (36) or perioperative (45) patients.

The effect of IIT on cardiovascular events was mixed. Four trials (13, 27, 5051) reported no differences between treatment groups in cardiovascular events. Four trials (2829, 4748) reported a decrease in cardiovascular events in the IIT intervention group, but the risk reduction in 1 of the trials seemed disproportionate compared with the modest and brief blood glucose level differential achieved (47). Conversely, more strokes occurred in the IIT treatment group in 1 perioperative trial (46).

Effects of IIT on Hypoglycemia

Intensive insulin therapy was associated with an increased risk for hypoglycemia in all settings (Table 2). A meta-analysis of 10 trials (Appendix Figure 5) found that IIT was associated with a 6-fold increased risk for severe hypoglycemia, defined as a blood glucose level less than 2.2 mmol/L (<40 mg/dL). However, the absolute risk varied across studies (RR, 6.00 [CI, 4.06 to 8.87]; I2 = 57.9%; P < 0.001). The addition of 11 trials with higher cut points defining hypoglycemia produced similar results (RR, 4.43 [CI, 2.30 to 8.53]; I2 = 94.5%; P < 0.001). Risk for hypoglycemia did not significantly differ between trials achieving treatment group blood glucose levels less than 6.7 mmol/L (<120 mg/dL) compared with those achieving levels of 6.7 mmol/L or greater (≥120 mg/dL) (RR, 5.99 vs. 4.28; P = 0.39 for the comparison between groups).

Table Jump PlaceholderTable 2.  Hypoglycemia Risk in Trials of Intensive Insulin Therapy
Grahic Jump Location
Appendix Figure 5.
Risk for hypoglycemia in studies of intensive insulin therapy in various inpatient settings.

We included inpatients in the MICU, SICU, and perioperative settings as well as patients with traumatic brain injury. We defined hypoglycemia as a blood glucose level less than 2.2 mmol/L (<40 mg/dL). Events is the number of participants with 1 or more hypoglycemic episodes in the treatment and control groups. MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

Grahic Jump Location

To examine the safety characteristics of various approaches to IIT, we reviewed an additional 29 studies not evaluating health outcomes that reported the safety of insulin infusions (Appendix). These studies were mostly small, single-center studies of iteratively developed protocols in carefully selected patient populations. Almost all of the centers that achieved blood glucose levels greater than 6.7 mmol/L (>120 mg/dL) also had low rates of hypoglycemia. Centers achieving levels less than 6.7 mmol/L (<120 mg/dL) had mixed results: Several centers safely achieved these levels by using sophisticated insulin protocols incorporating multiple variables or computerized algorithms, whereas other centers had very high rates of hypoglycemia.

A small number of in-hospital adverse effects of hypoglycemia were reported during IIT (Table 2). However, many critically ill patients included in these studies were sedated, which limited event detection. Three trials involving patients in the MICU (35, 38, 56) found excess mortality or extended length of inpatient stay among patients treated with IIT who experienced at least 1 episode of severe hypoglycemia. In 1 trial (35), the mortality risk in patients treated with IIT who had hypoglycemia was double that in control patients with hypoglycemia (25.0% vs. 12.5%; P < 0.05). Another trial (38) found that patients treated with IIT who had hypoglycemic events were significantly more likely to require prolonged hospitalization (2.4% vs. 0.3%; P = 0.05). One study (40) reported a case of cardiac asystole related to treatment-induced hypoglycemia.

In a synthesis of evidence from randomized, controlled trials, we found that use of IIT to achieve strict glucose control compared with less strict control did not reduce mortality or length of hospital stay but did substantially increase the risk for severe hypoglycemia in various hospital settings. Although strict glucose control was associated with a marginally significant reduction in septicemia, our sensitivity analysis found that the marginal benefit was entirely caused by Van den Berghe and colleagues' trial of patients in the SICU (16). Furthermore, a larger, more recent trial including patients in the SICU (17) found no association between glucose control and septicemia. Using the GRADE system (Table 3), we summarized the strength of evidence supporting these findings in each of the hospital settings that we evaluated and found the strongest body of evidence in ICU settings.

Table Jump PlaceholderTable 3.  Summary of the Evidence for the Effects of Intensive Insulin Therapy, by Outcome and Inpatient Setting

Subsequent trials (5758) have not replicated the initial encouraging findings from Van den Berghe and colleagues' trial of patients in the SICU (16). Several factors may account for this discrepancy. Parenteral nutrition was used routinely only in Van den Berghe and colleagues' trials of patients in the SICU and MICU (16, 37) and may be associated with adverse effects in critically ill patients, including an increased risk for infection (5960). Therefore, the observed benefits of IIT in these patients may actually reflect a reduction in harm from aggressive nutrition practices.

Capillary blood sampling is the more commonly used method for monitoring blood glucose levels but is less dependable than arterial blood sampling (61), which was routinely used in Van den Berghe and colleagues' trial of patients in the SICU (16). Most ICU trials achieved a slightly smaller differential in blood glucose levels between the intervention and control groups than that in Van den Berghe and colleagues' trial. Lower blood glucose level targets for control groups in recent trials (17, 56), and different blood glucose reporting methodologies, such as measuring mean glucose versus morning glucose levels, may partly explain this discrepancy. However, despite the slightly higher blood glucose levels achieved with IIT in recent trials, the risk for severe hypoglycemia remained consistently and substantially elevated.

Several trials that showed no mortality benefit from IIT in critically ill patients were discontinued early because of an excess risk for hypoglycemia in the intervention groups, which suggested that the lack of observed benefit may reflect inadequate power (35, 38, 56). However, these trials did not demonstrate a consistent trend toward benefit. Furthermore, the inability to implement the insulin infusion protocol in various clinical trial settings without causing high rates of hypoglycemia may underscore the complexity of IIT and problems with generalizability across institutions.

Although many experts acknowledge the lack of convincing evidence showing benefit from IIT targeted to very strict ranges of blood glucose level, they are reluctant to discontinue glycemic control because of the potential complications of hyperglycemia (9, 62). However, the benefits of achieving more moderate blood glucose targets have not been established; therefore, it is imperative that any glycemic control strategy also minimizes harms.

Glucose targets are a key determinant of safety, although other factors also may be important. Previous reviews of insulin protocols stressed that, given the various factors influencing the safety of IIT implementation, each institution should individualize its protocol on the basis of its patient population as well as its institutional and provider resources (6365). These reviews speculate that safer protocols should incorporate bolus insulin doses, account for the direction and rate of changes in blood glucose levels, and allow “off-protocol” adjustments (6465).

Our review of insulin protocol studies also suggests that protocol characteristics are important but perhaps less so than the blood glucose level target itself (Appendix). We found that use of sophisticated protocols did not consistently decrease the risk for hypoglycemia when glucose levels less than 6.7 mmol/L (<120 mg/dL) were achieved. Observational studies also show an increased risk for hypoglycemia when institutions implement stricter blood glucose level targets over time (6667). In addition, nearly all institutions that achieved blood glucose levels greater than 6.7 mmol/L (>120 mg/dL) also had low rates of hypoglycemia. Furthermore, the relatively low rates of hypoglycemia in the IIT trial control groups, which generally used target blood glucose levels ranging from 7.8 to 11.1 mmol/L (140 to 200 mg/dL), suggest that higher target blood glucose levels are safer.

The consequences of inpatient hypoglycemia are unclear. In several trials, hypoglycemia was associated with excess mortality; however, whether hypoglycemia is a causative factor or simply a marker of more severe disease remains uncertain, and recent observational studies attempting to clarify this issue have yielded conflicting results (6869). One recent cohort study of diabetic patients found that severe hypoglycemia was associated with an increased risk for incident dementia, and another study found that inpatient hypoglycemia was associated with increased long-term mortality in diabetic patients with the acute coronary syndrome (7071).

Two previous reviews focused on ICU settings, whereas we examined the effects of IIT in various hospital settings (5758). We reached conclusions similar to those reached in previous studies of the risk for hypoglycemia and lack of benefit associated with IIT in ICU settings. We also found that this lack of benefit and possible harm was less well studied but extended to other hospital settings. Our broader evaluation strengthens the finding that IIT is difficult to implement safely and that initial benefits seen in Van den Berghe and colleagues' trial of patients in the SICU (16) subsequently have not been realized despite attempts to do so in various settings. Finally, because of continued interest in glucose control strategies to reduce the perceived harm of hyperglycemia, our review also adds to findings from previous reviews by examining the safety of various insulin infusion programs.

Our review has several potential limitations. Studies were grouped according to hospital setting, but overlap may occur in some of the subgroups. For instance, patients with MI may receive care in an MICU setting. However, results were consistent across subgroups. No study could isolate the effects of the IIT intervention itself. The increased intensity of nursing care associated with IIT implementation in the intervention groups may have caused unintended co-interventions, such as improved catheter care or earlier attempts at ventilator weaning. The beneficial effects of these co-interventions therefore could potentially overshadow the harmful effects of the IIT intervention itself. The generalizability of results from the included studies may be limited by patient characteristics, event rates, concomitant therapy, institution characteristics, and monitoring methodology. Most studies were conducted in a single center, yet the relative success or failure of an insulin-based intervention may depend, in part, on characteristics that are unique to a particular institution or health system. Finally, the included trials usually relied on blood glucose measurements from a defined reference time rather than from 24-hour blood glucose levels achieved and therefore could obscure the inherent variability in glucose control through the course of intensive care (7273).

Many institutions are likely to pursue less aggressive glycemic control, but the health benefits of achieving moderate blood glucose level targets, such as 7.8 to 11.1 mmol/L (140 to 200 mg/dL), should be examined. Future studies also should evaluate the cost and patient, nurse, and physician acceptance of implementing insulin infusion protocols in hospitalized patients. Individual institutions describing their experience implementing intensive insulin protocols suggest that the increase in nursing workload and fear of hypoglycemia were significant although potentially surmountable barriers to implementation (7476). Moreover, evaluating the feasibility and safety of transitioning patients from insulin infusion to subcutaneous insulin and ultimately to a safe outpatient regimen is warranted.

In summary, no consistent evidence shows that IIT improves health outcomes in hospitalized patients. However, this intervention may be associated with a high risk for severe hypoglycemia, especially when the blood glucose level target is less than 6.7 mmol/L (<120 mg/dL). The consequences of severe hypoglycemia in hospitalized patients have not been well studied. However, given the lack of compelling evidence for benefit, the potential for serious harm should forestall efforts to routinely implement very strict targets for blood glucose control in hospitalized patients.

Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE.  Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002; 87:978-82.
PubMed
CrossRef
 
Levetan CS, Passaro M, Jablonski K, Kass M, Ratner RE.  Unrecognized diabetes among hospitalized patients. Diabetes Care. 1998; 21:246-9.
PubMed
 
Furnary AP, Gao G, Grunkemeier GL, Wu Y, Zerr KJ, Bookin SO. et al.  Continuous insulin infusion reduces mortality in patients with diabetes undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2003; 125:1007-21.
PubMed
 
Pomposelli JJ, Baxter JK 3rd, Babineau TJ, Pomfret EA, Driscoll DF, Forse RA. et al.  Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. JPEN J Parenter Enteral Nutr. 1998; 22:77-81.
PubMed
 
Bochicchio GV, Salzano L, Joshi M, Bochicchio K, Scalea TM.  Admission preoperative glucose is predictive of morbidity and mortality in trauma patients who require immediate operative intervention. Am Surg. 2005; 71:171-4.
PubMed
 
Capes SE, Hunt D, Malmberg K, Gerstein HC.  Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet. 2000; 355:773-8.
PubMed
 
Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC.  Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke. 2001; 32:2426-32.
PubMed
 
McCowen KC, Malhotra A, Bistrian BR.  Stress-induced hyperglycemia. Crit Care Clin. 2001; 17:107-24.
PubMed
 
Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG, et al. American Diabetes Association Diabetes in Hospitals Writing Committee.  Management of diabetes and hyperglycemia in hospitals. Diabetes Care. 2004; 27:553-91.
PubMed
 
Furnary AP, Zerr KJ, Grunkemeier GL, Starr A.  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.
PubMed
 
Hirshberg E, Lacroix J, Sward K, Willson D, Morris AH.  Blood glucose control in critically ill adults and children: a survey on stated practice. Chest. 2008; 133:1328-35.
PubMed
 
Institute for Health Care Improvement.  Implement effective glucose control: establish a glycemic control policy in your ICU. 2008. Accessed atwww.ihi.org/IHI/Topics/CriticalCare/IntensiveCare/Changes/IndividualChanges/EstablishaGlycemicControlPolicyinYourICU.htmon 28 December 2010.
 
Malmberg K, Rydén L, Efendic S, Herlitz J, Nicol P, Waldenström A. et al.  Randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year. J Am Coll Cardiol. 1995; 26:57-65.
PubMed
 
Furnary AP, Cheek DB, Holmes SC, Howell WL, Kelly SP.  Achieving tight glycemic control in the operating room: lessons learned from 12 years in the trenches of a paradigm shift in anesthetic care. Semin Thorac Cardiovasc Surg. 2006; 18:339-45.
PubMed
 
Zerr KJ, Furnary AP, Grunkemeier GL, Bookin S, Kanhere V, Starr A.  Glucose control lowers the risk of wound infection in diabetics after open heart operations. Ann Thorac Surg. 1997; 63:356-61.
PubMed
 
Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M. et al.  Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001; 345:1359-67.
PubMed
 
Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, et al. NICE-SUGAR Study Investigators.  Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009; 360:1283-97.
PubMed
 
www.hsrd.research.va.gov/publications/esp/Hyperglycemia-2008.pdfKansagara D, Wolf F, Freeman M, Helfand M.  Management of inpatient hyperglycemia: a systematic review. U.S. Department of Veterans Affair Health Services Research and Development Service. Evidence-based Synthesis Program Report. Washington, DC: U.S. Department of Veterans Affairs; 2008.
 
Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM, et al. Methods Work Group, Third US Preventive Services Task Force.  Current methods of the US Preventive Services Task Force: a review of the process. Am J Prev Med. 2001; 20:21-35.
PubMed
 
DerSimonian R, Laird N.  Meta-analysis in clinical trials. Control Clin Trials. 1986; 7:177-88.
PubMed
 
Higgins JP, Thompson SG, Deeks JJ, Altman DG.  Measuring inconsistency in meta-analyses. BMJ. 2003; 327:557-60.
PubMed
 
Brozek JL, Akl EA, Alonso-Coello P, Lang D, Jaeschke R, Williams JW, et al. GRADE Working Group.  Grading quality of evidence and strength of recommendations in clinical practice guidelines. Part 1 of 3. An overview of the GRADE approach and grading quality of evidence about interventions. Allergy. 2009; 64:669-77.
PubMed
 
Iwamoto G.  Comparative trial between computer-guided intravenous infusion protocol versus a standard insulin infusion algorithm versus a simple calculated infusion protocol in medical and surgical ICU. Accessed athttp://clinicaltrials.gov/show/NCT00582309on 28 December 2010.
 
Umpierrez GE.  Comparative trial between a computer-guided intravenous infusion protocol versus a standard insulin infusion algorithm in medical ICU. Accessed athttp://clinicaltrials.gov/show/NCT00394524on 28 December 2010.
 
Lacherade J-C.  Multicentre randomized trial assessing the impact of maintaining 2 blood glucose levels on hospital mortality in patients admitted to the ICU (INSUREA study). Accessed athttp://clinicaltrials.gov/show/NCT00591071on 28 December 2010.
 
Grey NJ, Perdrizet GA.  Reduction of nosocomial infections in the surgical intensive-care unit by strict glycemic control. Endocr Pract. 2004; 10:Suppl 246-52.
PubMed
 
Kirdemir P, Yildirim V, Kiris I, Gulmen S, Kuralay E, Ibrisim E. et al.  Does continuous insulin therapy reduce postoperative supraventricular tachycardia incidence after coronary artery bypass operations in diabetic patients? J Cardiothorac Vasc Anesth. 2008; 22:383-7.
PubMed
 
Farah R, Samokhvalov A, Zviebel F, Makhoul N.  Insulin therapy of hyperglycemia in intensive care. Isr Med Assoc J. 2007; 9:140-2.
PubMed
 
Lazar HL, Chipkin SR, Fitzgerald CA, Bao Y, Cabral H, Apstein CS.  Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation. 2004; 109:1497-502.
PubMed
 
Li JY, Sun S, Wu SJ.  Continuous insulin infusion improves postoperative glucose control in patients with diabetes mellitus undergoing coronary artery bypass surgery. Tex Heart Inst J. 2006; 33:445-51.
PubMed
 
Smith A, Grattan A, Harper M, Royston D, Riedel BJ.  Coronary revascularization: a procedure in transition from on-pump to off-pump? The role of glucose-insulin-potassium revisited in a randomized, placebo-controlled study. J Cardiothorac Vasc Anesth. 2002; 16:413-20.
PubMed
 
Rasoul S, Ottervanger JP, Timmer JR, Svilaas T, Henriques JP, Dambrink JH. et al.  One year outcomes after glucose-insulin-potassium in ST elevation myocardial infarction. The Glucose-insulin-potassium study II. Int J Cardiol. 2007; 122:52-5.
PubMed
 
Gray CS, Hildreth AJ, Sandercock PA, O'Connell JE, Johnston DE, Cartlidge NE, et al. GIST Trialists Collaboration.  Glucose-potassium-insulin infusions in the management of post-stroke hyperglycaemia: the UK Glucose Insulin in Stroke Trial (GIST-UK). Lancet Neurol. 2007; 6:397-406.
PubMed
 
Yang M, Guo Q, Zhang X, Sun S, Wang Y, Zhao L. et al.  Intensive insulin therapy on infection rate, days in NICU, in-hospital mortality and neurological outcome in severe traumatic brain injury patients: a randomized controlled trial. Int J Nurs Stud. 2009; 46:753-8.
PubMed
 
Arabi YM, Dabbagh OC, Tamim HM, Al-Shimemeri AA, Memish ZA, Haddad SH. et al.  Intensive versus conventional insulin therapy: a randomized controlled trial in medical and surgical critically ill patients. Crit Care Med. 2008; 36:3190-7.
PubMed
 
Bilotta F, Caramia R, Paoloni FP, Delfini R, Rosa G.  Safety and efficacy of intensive insulin therapy in critical neurosurgical patients. Anesthesiology. 2009; 110:611-9.
PubMed
 
Van den Berghe G, Wilmer A, Hermans G, Meersseman W, Wouters PJ, Milants I. et al.  Intensive insulin therapy in the medical ICU. N Engl J Med. 2006; 354:449-61.
PubMed
 
Brunkhorst FM, Engel C, Bloos F, Meier-Hellmann A, Ragaller M, Weiler N, et al. German Competence Network Sepsis (SepNet).  Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med. 2008; 358:125-39.
PubMed
 
Savioli M, Cugno M, Polli F, Taccone P, Bellani G, Spanu P. et al.  Tight glycemic control may favor fibrinolysis in patients with sepsis. Crit Care Med. 2009; 37:424-31.
PubMed
 
Mackenzie IM, Ercole A, Blunt M, Ingle S, Palmer CR.  Glycaemic control and outcome in general intensive care: the East Anglian GLYCOGENIC study. Br J Intensive Care. 2008; 18:121-6.
 
Preiser JC, Devos P, Ruiz-Santana S, Mélot C, Annane D, Groeneveld J. et al.  A prospective randomised multi-centre controlled trial on tight glucose control by intensive insulin therapy in adult intensive care units: the Glucontrol study. Intensive Care Med. 2009; 35:1738-48.
PubMed
 
De La Rosa Gl C, Donado JH, Restrepo AH, Quintero AM, González LG, Saldarriaga NE, et al. Grupo de Investigacion en Cuidado intensivo: GICI-HPTU.  Strict glycaemic control in patients hospitalised in a mixed medical and surgical intensive care unit: a randomised clinical trial. Crit Care. 2008; 12:R120.
PubMed
 
Ingels C, Debaveye Y, Milants I, Buelens E, Peeraer A, Devriendt Y. et al.  Strict blood glucose control with insulin during intensive care after cardiac surgery: impact on 4-years survival, dependency on medical care, and quality-of-life. Eur Heart J. 2006; 27:2716-24.
PubMed
 
Barcellos Cda S, Wender OC, Azambuja PC.  Clinical and hemodynamic outcome following coronary artery bypass surgery in diabetic patients using glucose-insulin-potassium (GIK) solution: a randomized clinical trial. Rev Bras Cir Cardiovasc. 2007; 22:275-84.
PubMed
 
Butterworth J, Wagenknecht LE, Legault C, Zaccaro DJ, Kon ND, Hammon JW Jr. et al.  Attempted control of hyperglycemia during cardiopulmonary bypass fails to improve neurologic or neurobehavioral outcomes in patients without diabetes mellitus undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2005; 130:1319.
PubMed
 
Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, O'Brien PC. et al.  Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Ann Intern Med. 2007; 146:233-43.
PubMed
 
Subramaniam B, Panzica PJ, Novack V, Mahmood F, Matyal R, Mitchell JD. et al.  Continuous perioperative insulin infusion decreases major cardiovascular events in patients undergoing vascular surgery: a prospective, randomized trial. Anesthesiology. 2009; 110:970-7.
PubMed
 
Cheung NW, Wong VW, McLean M.  The Hyperglycemia: Intensive Insulin Infusion in Infarction (HI-5) study: a randomized controlled trial of insulin infusion therapy for myocardial infarction. Diabetes Care. 2006; 29:765-70.
PubMed
 
Oksanen T, Skrifvars MB, Varpula T, Kuitunen A, Pettilä V, Nurmi J. et al.  Strict versus moderate glucose control after resuscitation from ventricular fibrillation. Intensive Care Med. 2007; 33:2093-100.
PubMed
 
van der Horst IC, Zijlstra F, van't Hof AW, Doggen CJ, de Boer MJ, Suryapranata H, et al. Zwolle Infarct Study Group.  Glucose-insulin-potassium infusion inpatients treated with primary angioplasty for acute myocardial infarction: the glucose-insulin-potassium study: a randomized trial. J Am Coll Cardiol. 2003; 42:784-91.
PubMed
 
Malmberg K, Rydén L, Wedel H, Birkeland K, Bootsma A, Dickstein K, et al. DIGAMI 2 Investigators.  Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2): effects on mortality and morbidity. Eur Heart J. 2005; 26:650-61.
PubMed
 
Azevedo JR, Lima ER, Cossetti RJ, Azevedo RP.  Intensive insulin therapy versus conventional glycemic control in patients with acute neurological injury: a prospective controlled trial. Arq Neuropsiquiatr. 2007; 65:733-8.
PubMed
 
Bruno A, Kent TA, Coull BM, Shankar RR, Saha C, Becker KJ. et al.  Treatment of hyperglycemia in ischemic stroke (THIS): a randomized pilot trial. Stroke. 2008; 39:384-9.
PubMed
 
Walters MR, Weir CJ, Lees KR.  A randomised, controlled pilot study to investigate the potential benefit of intervention with insulin in hyperglycaemic acute ischaemic stroke patients. Cerebrovasc Dis. 2006; 22:116-22.
PubMed
 
Van den Berghe G, Wouters PJ, Kesteloot K, Hilleman DE.  Analysis of healthcare resource utilization with intensive insulin therapy in critically ill patients. Crit Care Med. 2006; 34:612-6.
PubMed
 
Devos P, Preiser JC, Melot C.  Impact of tight glucose control by intensive insulin therapy on ICU mortality and the rate of hypoglycaemia: final results of the GLUCONTROL study. Intensive Care Med. 2007; 33:Suppl 2S189.
 
Wiener RS, Wiener DC, Larson RJ.  Benefits and risks of tight glucose control in critically ill adults: a meta-analysis. JAMA. 2008; 300:933-44.
PubMed
 
Griesdale DE, de Souza RJ, van Dam RM, Heyland DK, Cook DJ, Malhotra A. et al.  Intensive insulin therapy and mortality among critically ill patients: a meta-analysis including NICE-SUGAR study data. CMAJ. 2009; 180:821-7.
PubMed
 
Bistrian BR, McCowen KC.  Nutritional and metabolic support in the adult intensive care unit: key controversies. Crit Care Med. 2006; 34:1525-31.
PubMed
 
Heyland DK, Dhaliwal R, Drover JW, Gramlich L, Dodek P, Canadian Critical Care Clinical Practice Guidelines Committee.  Canadian clinical practice guidelines for nutrition support in mechanically ventilated, critically ill adult patients. JPEN J Parenter Enteral Nutr. 2003; 27:355-73.
PubMed
 
Kanji S, Buffie J, Hutton B, Bunting PS, Singh A, McDonald K. et al.  Reliability of point-of-care testing for glucose measurement in critically ill adults. Crit Care Med. 2005; 33:2778-85.
PubMed
 
Inzucchi SE, Siegel MD.  Glucose control in the ICU—how tight is too tight? [Editorial]. N Engl J Med. 2009; 360:1346-9.
PubMed
 
Nazer LH, Chow SL, Moghissi ES.  Insulin infusion protocols for critically ill patients: a highlight of differences and similarities. Endocr Pract. 2007; 13:137-46.
PubMed
 
Meijering S, Corstjens AM, Tulleken JE, Meertens JH, Zijlstra JG, Ligtenberg JJ.  Towards a feasible algorithm for tight glycaemic control in critically ill patients: a systematic review of the literature. Crit Care. 2006; 10:R19.
PubMed
 
Wilson M, Weinreb J, Hoo GW.  Intensive insulin therapy in critical care: a review of 12 protocols. Diabetes Care. 2007; 30:1005-11.
PubMed
 
Treggiari MM, Karir V, Yanez ND, Weiss NS, Daniel S, Deem SA.  Intensive insulin therapy and mortality in critically ill patients. Crit Care. 2008; 12:R29.
PubMed
 
Taylor BE, Schallom ME, Sona CS, Buchman TG, Boyle WA, Mazuski JE. et al.  Efficacy and safety of an insulin infusion protocol in a surgical ICU. J Am Coll Surg. 2006; 202:1-9.
PubMed
 
Kosiborod M, Inzucchi SE, Goyal A, Krumholz HM, Masoudi FA, Xiao L. et al.  Relationship between spontaneous and iatrogenic hypoglycemia and mortality in patients hospitalized with acute myocardial infarction. JAMA. 2009; 301:1556-64.
PubMed
 
Turchin A, Matheny ME, Shubina M, Scanlon JV, Greenwood B, Pendergrass ML.  Hypoglycemia and clinical outcomes in patients with diabetes hospitalized in the general ward. Diabetes Care. 2009; 32:1153-7.
PubMed
 
Whitmer RA, Karter AJ, Yaffe K, Quesenberry CP Jr, Selby JV.  Hypoglycemic episodes and risk of dementia in older patients with type 2 diabetes mellitus. JAMA. 2009; 301:1565-72.
PubMed
 
Svensson AM, McGuire DK, Abrahamsson P, Dellborg M.  Association between hyper- and hypoglycaemia and 2 year all-cause mortality risk in diabetic patients with acute coronary events. Eur Heart J. 2005; 26:1255-61.
PubMed
 
Finney SJ, Zekveld C, Elia A, Evans TW.  Glucose control and mortality in critically ill patients. JAMA. 2003; 290:2041-7.
PubMed
 
Smith SM, Oveson KE, Strauss W, Raven K, Lefevre MC, Ahmann AJ. et al.  Ultradian variation of blood glucose in intensive care unit patients receiving insulin infusions. Diabetes Care. 2007; 30:2503-5.
PubMed
 
Goldberg PA, Inzucchi SE.  Selling root canals: lessons learned from implementing a hospital insulin infusion protocol. Diabetes Spectrum. 2005; 18:28-33.
 
Rea RS, Donihi AC, Bobeck M, Herout P, McKaveney TP, Kane-Gill SL. et al.  Implementing an intravenous insulin infusion protocol in the intensive care unit. Am J Health Syst Pharm. 2007; 64:385-95.
PubMed
 
Aragon D.  Evaluation of nursing work effort and perceptions about blood glucose testing in tight glycemic control. Am J Crit Care. 2006; 15:370-7.
PubMed
 
Scalea TM, Bochicchio GV, Bochicchio KM, Johnson SB, Joshi M, Pyle A.  Tight glycemic control in critically injured trauma patients. Ann Surg. 2007; 246:605-10.
PubMed
 
Beilman GJ, Joseph JI.  Practical considerations for glucose control in hospitalized patients. Diabetes Technol Ther. 2005; 7:823-30.
PubMed
 
Button E, Keaton P.  Glycemic control after coronary bypass graft: using intravenous insulin regulated by a computerized system. Crit Care Nurs Clin North Am. 2006; 18.
PubMed
 
Davis ED, Harwood K, Midgett L, Mabrey M, Lien LF.  Implementation of a new intravenous insulin method on intermediate-care units in hospitalized patients. Diabetes Educ. 2005; 31.
PubMed
 
González-Michaca L, Ahumada M, Ponce-de-León S.  Insulin subcutaneous application vs. continuous infusion for postoperative blood glucose control in patients with non-insulin-dependent diabetes mellitus. Arch Med Res. 2002; 33:48-52.
PubMed
 
Markovitz LJ, Wiechmann RJ, Harris N, Hayden V, Cooper J, Johnson G. et al.  Description and evaluation of a glycemic management protocol for patients with diabetes undergoing heart surgery. Endocr Pract. 2002; 8:10-8.
PubMed
 
Raucoules-Aimé M, Labib Y, Levraut J, Gastaud P, Dolisi C, Grimaud D.  Use of i.v. insulin in well-controlled non-insulin-dependent diabetics undergoing major surgery. Br J Anaesth. 1996; 76:198-202.
PubMed
 
Rood E, Bosman RJ, van der Spoel JI, Taylor P, Zandstra DF.  Use of a computerized guideline for glucose regulation in the intensive care unit improved both guideline adherence and glucose regulation. J Am Med Inform Assoc. 2005; 12:172-80.
PubMed
 
Simmons D, Morton K, Laughton SJ, Scott DJ.  A comparison of two intravenous insulin regimens among surgical patients with insulin-dependent diabetes mellitus. Diabetes Educ. 1994; 20:422-7.
PubMed
 
Eigsti J, Henke K.  Innovative solutions: development and implementation of a tight blood glucose management protocol: one community hospital's experience. Dimens Crit Care Nurs. 2006; 25:62-5.
PubMed
 
Reed CC, Stewart RM, Sherman M, Myers JG, Corneille MG, Larson N. et al.  Intensive insulin protocol improves glucose control and is associated with a reduction in intensive care unit mortality. J Am Coll Surg. 2007; 204:1048-54.
PubMed
 
Chaney MA, Nikolov MP, Blakeman BP, Bakhos M.  Attempting to maintain normoglycemia during cardiopulmonary bypass with insulin may initiate postoperative hypoglycemia. Anesth Analg. 1999; 89:1091-5.
PubMed
 
Carvalho G, Moore A, Qizilbash B, Lachapelle K, Schricker T.  Maintenance of normoglycemia during cardiac surgery. Anesth Analg. 2004;99:319-24, table of contents. Anesth Analg. 2004; 99:319-24.
PubMed
 
Chee F, Fernando TL, Savkin AV, van Heeden V.  Expert PID control system for blood glucose control in critically ill patients. IEEE Trans Inf Technol Biomed. 2003; 7:419-25.
PubMed
 
Wong XW, Chase JG, Shaw GM, Hann CE, Lotz T, Lin J. et al.  Model predictive glycaemic regulation in critical illness using insulin and nutrition input: a pilot study. Med Eng Phys. 2006; 28:665-81.
PubMed
 
Hemmerling TM, Schmid MC, Schmidt J, Kern S, Jacobi KE.  Comparison of a continuous glucose-insulin-potassium infusion versus intermittent bolus application of insulin on perioperative glucose control and hormone status in insulin-treated type 2 diabetics. J Clin Anesth. 2001; 13:293-300.
PubMed
 
Quinn JA, Snyder SL, Berghoff JL, Colombo CS, Jacobi J.  A practical approach to hyperglycemia management in the intensive care unit: evaluation of an intensive insulin infusion protocol. Pharmacotherapy. 2006; 26:1410-20.
PubMed
 
Shaw GM, Chase JG, Wong J, Lin J, Lotz T, Le Compte AJ. et al.  Rethinking glycaemic control in critical illness—from concept to clinical practice change. Crit Care Resusc. 2006; 8:90-9.
PubMed
 
Rowen M, Schneider DJ, Pratley RE, Sobel BE.  On rendering continuous glucose monitoring ready for prime time in the cardiac care unit. Coron Artery Dis. 2007; 18:405-9.
PubMed
 
Osburne RC, Cook CB, Stockton L, Baird M, Harmon V, Keddo A. et al.  Improving hyperglycemia management in the intensive care unit: preliminary report of a nurse-driven quality improvement project using a redesigned insulin infusion algorithm. Diabetes Educ. 2006; 32:394-403.
PubMed
 
Elinav H, Wolf Z, Szalat A, Bdolah-Abram T, Glaser B, Raz I. et al.  In-hospital treatment of hyperglycemia: effects of intensified subcutaneous insulin treatment. Curr Med Res Opin. 2007; 23:757-65.
PubMed
 
Ku SY, Sayre CA, Hirsch IB, Kelly JL.  New insulin infusion protocol Improves blood glucose control in hospitalized patients without increasing hypoglycemia. Jt Comm J Qual Patient Saf. 2005; 31:141-7.
PubMed
 
Lien LF, Spratt SE, Woods Z, Osborne KK, Feinglos MN.  Optimizing hospital use of intravenous insulin therapy: improved management of hyperglycemia and error reduction with a new nomogram. Endocr Pract. 2005; 11:240-53.
PubMed
 
Vogelzang M, Loef BG, Regtien JG, van der Horst IC, van Assen H, Zijlstra F. et al.  Computer-assisted glucose control in critically ill patients. Intensive Care Med. 2008; 34:1421-7.
PubMed
 
Kanji S, Singh A, Tierney M, Meggison H, McIntyre L, Hebert PC.  Standardization of intravenous insulin therapy improves the efficiency and safety of blood glucose control in critically ill adults. Intensive Care Med. 2004; 30:804-10.
PubMed
 
Bland DK, Fankhanel Y, Langford E, Lee M, Lee SW, Maloney C. et al.  Intensive versus modified conventional control of blood glucose level in medical intensive care patients: a pilot study. Am J Crit Care. 2005; 14:370-6.
PubMed
 
Pachler C, Plank J, Weinhandl H, Chassin LJ, Wilinska ME, Kulnik R. et al.  Tight glycaemic control by an automated algorithm with time-variant sampling in medical ICU patients. Intensive Care Med. 2008; 34:1224-30.
PubMed
 
Juneja R, Roudebush C, Kumar N, Macy A, Golas A, Wall D. et al.  Utilization of a computerized intravenous insulin infusion program to control blood glucose in the intensive care unit. Diabetes Technol Ther. 2007; 9:232-40.
PubMed
 
Hovorka R, Kremen J, Blaha J, Matias M, Anderlova K, Bosanska L. et al.  Blood glucose control by a model predictive control algorithm with variable sampling rate versus a routine glucose management protocol in cardiac surgery patients: a randomized controlled trial. J Clin Endocrinol Metab. 2007; 92:2960-4.
PubMed
 
Braithwaite SS, Edkins R, Macgregor KL, Sredzienski ES, Houston M, Zarzaur B. et al.  Performance of a dose-defining insulin infusion protocol among trauma service intensive care unit admissions. Diabetes Technol Ther. 2006; 8:476-88.
PubMed
 
Dortch MJ, Mowery NT, Ozdas A, Dossett L, Cao H, Collier B. et al.  A computerized insulin infusion titration protocol improves glucose control with less hypoglycemia compared to a manual titration protocol in a trauma intensive care unit. JPEN J Parenter Enteral Nutr. 2008; 32:18-27.
PubMed
 
Toschlog EA, Newton C, Allen N, Newell MA, Goettler CE, Schenarts PJ. et al.  Morbidity reduction in critically ill trauma patients through use of a computerized insulin infusion protocol: a preliminary study. J Trauma. 2007; 62:1370-5.
PubMed
 
Iapichino G, Albicini M, Umbrello M, Sacconi F, Fermo I, Pavlovich R. et al.  Tight glycemic control does not affect asymmetric-dimethylarginine in septic patients. Intensive Care Med. 2008; 34:1843-50.
PubMed
 
Mitchell I, Knight E, Gissane J, Tamhane R, Kolli R, Leditschke IA, et al. Australian and New Zealand Intensive Care Society Clinical Trials Group.  A phase II randomised controlled trial of intensive insulin therapy in general intensive care patients. Crit Care Resusc. 2006; 8:289-93.
PubMed
 
Saager L, Collins GL, Burnside B, Tymkew H, Zhang L, Jacobsohn E. et al.  A randomized study in diabetic patients undergoing cardiac surgery comparing computer-guided glucose management with a standard sliding scale protocol. J Cardiothorac Vasc Anesth. 2008; 22:377-82.
PubMed
 
McMullin J, Brozek J, McDonald E, Clarke F, Jaeschke R, Heels-Ansdell D. et al.  Lowering of glucose in critical care: a randomized pilot trial. J Crit Care. 2007; 22:112-8.
PubMed
 
Metchick LN, Petit WA Jr, Inzucchi SE.  Inpatient management of diabetes mellitus. Am J Med. 2002; 113:317-23.
PubMed
 
Umpierrez GE, Smiley D, Zisman A, Prieto LM, Palacio A, Ceron M. et al.  Randomized study of basal-bolus insulin therapy in the inpatient management of patients with type 2 diabetes (RABBIT 2 trial). Diabetes Care. 2007; 30:2181-6.
PubMed
 
Dickerson LM, Ye X, Sack JL, Hueston WJ.  Glycemic control in medical inpatients with type 2 diabetes mellitus receiving sliding scale insulin regimens versus routine diabetes medications: a multicenter randomized controlled trial. Ann Fam Med. 2003; 1:29-35.
PubMed
 
Datta S, Qaadir A, Villanueva G, Baldwin D.  Once-daily insulin glargine versus 6-hour sliding scale regular insulin for control of hyperglycemia after a bariatric surgical procedure: a randomized clinical trial. Endocr Pract. 2007; 13:225-31.
PubMed
 
Korytkowski MT, Salata RJ, Koerbel GL, Selzer F, Karslioglu E, Idriss AM. et al.  Insulin therapy and glycemic control in hospitalized patients with diabetes during enteral nutrition therapy: a randomized controlled clinical trial. Diabetes Care. 2009; 32:594-6.
PubMed
 
Ahmann AJ, Maynard G.  Designing and implementing insulin infusion protocols and order sets. J Hosp Med. 2008; 3:42-54.
PubMed
 
Fischer KF, Lees JA, Newman JH.  Hypoglycemia in hospitalized patients. Causes and outcomes. N Engl J Med. 1986; 315:1245-50.
PubMed
 
Shilo S, Berezovsky S, Friedlander Y, Sonnenblick M.  Hypoglycemia in hospitalized nondiabetic older patients. J Am Geriatr Soc. 1998; 46:978-82.
PubMed
 
Vriesendorp TM, van Santen S, DeVries JH, de Jonge E, Rosendaal FR, Schultz MJ. et al.  Predisposing factors for hypoglycemia in the intensive care unit. Crit Care Med. 2006; 34:96-101.
PubMed
 
Krinsley J.  Glycemic control in critically ill patients: Leuven and beyond [Editorial]. Chest. 2007; 132:1-2.
PubMed
 
Cochran A, Davis L, Morris SE, Saffle JR.  Safety and efficacy of an intensive insulin protocol in a burn-trauma intensive care unit. J Burn Care Res. 2008; 29:187-91.
PubMed
 
Kreisel SH, Berschin UM, Hammes HP, Leweling H, Bertsch T, Hennerici MG. et al.  Pragmatic management of hyperglycaemia in acute ischaemic stroke: safety and feasibility of intensive intravenous insulin treatment. Cerebrovasc Dis. 2009; 27:167-75.
PubMed
 
DeSantis AJ, Schmeltz LR, Schmidt K, O'Shea-Mahler E, Rhee C, Wells A. et al.  Inpatient management of hyperglycemia: the Northwestern experience. Endocr Pract. 2006; 12:491-505.
PubMed
 
Goldberg PA, Sakharova OV, Barrett PW, Falko LN, Roussel MG, Bak L. et al.  Improving glycemic control in the cardiothoracic intensive care unit: clinical experience in two hospital settings. J Cardiothorac Vasc Anesth. 2004; 18:690-7.
PubMed
 
Zimmerman CR, Mlynarek ME, Jordan JA, Rajda CA, Horst HM.  An insulin infusion protocol in critically ill cardiothoracic surgery patients. Ann Pharmacother. 2004; 38:1123-9.
PubMed
 
Plank J, Blaha J, Cordingley J, Wilinska ME, Chassin LJ, Morgan C. et al.  Multicentric, randomized, controlled trial to evaluate blood glucose control by the model predictive control algorithm versus routine glucose management protocols in intensive care unit patients. Diabetes Care. 2006; 29:271-6.
PubMed
 
Dilkhush D, Lannigan J, Pedroff T, Riddle A, Tittle M.  Insulin infusion protocol for critical care units. Am J Health Syst Pharm. 2005; 62:2260-4.
PubMed
 
Vogelzang M, Zijlstra F, Nijsten MW.  Design and implementation of GRIP: a computerized glucose control system at a surgical intensive care unit. BMC Med Inform Decis Mak. 2005; 5:38.
PubMed
 
Tamaki M, Shimizu T, Kanazawa A, Tamura Y, Hanzawa A, Ebato C. et al.  Efficacy and safety of modified Yale insulin infusion protocol in Japanese diabetic patients after open-heart surgery. Diabetes Res Clin Pract. 2008; 81:296-302.
PubMed
 
Clayton SB, Mazur JE, Condren S, Hermayer KL, Strange C.  Evaluation of an intensive insulin protocol for septic patients in a medical intensive care unit. Crit Care Med. 2006; 34:2974-8.
PubMed
 
Chant C, Wilson G, Friedrich JO.  Validation of an insulin infusion nomogram for intensive glucose control in critically ill patients. Pharmacotherapy. 2005; 25:352-9.
PubMed
 
Balkin M, Mascioli C, Smith V, Alnachawati H, Mehrishi S, Saydain G. et al.  Achieving durable glucose control in the intensive care unit without hypoglycaemia: a new practical IV insulin protocol. Diabetes Metab Res Rev. 2007; 23:49-55.
PubMed
 
Krinsley JS.  Glycemic control, diabetic status, and mortality in a heterogeneous population of critically ill patients before and during the era of intensive glycemic management: six and one-half years experience at a university-affiliated community hospital. Semin Thorac Cardiovasc Surg. 2006; 18:317-25.
PubMed
 
Theilen BM, Gritzke KA, Knutsen PG, Riek AE, McGill JB, Sicard GA. et al.  Inpatient glycemic control on the vascular surgery service. Endocr Pract. 2008; 14:185-91.
PubMed
 
Bell DA, Strong AJ.  Glucose/insulin infusions in the treatment of subarachnoid haemorrhage: a feasibility study. Br J Neurosurg. 2005; 19:21-4.
PubMed
 
van Wezel HB, Zuurbier CJ, de Jonge E, van Dam EW, van Dijk J, Endert E. et al.  Differential effects of a perioperative hyperinsulinemic normoglycemic clamp on the neurohumoral stress response during coronary artery surgery. J Clin Endocrinol Metab. 2006; 91:4144-53.
PubMed
 
Miriam A, Korula G.  A simple glucose insulin regimen for perioperative blood glucose control: the Vellore regimen. Anesth Analg. 2004; 99:598-602.
PubMed
 
Appendix
Background

Over the past decade, numerous trials have evaluated how decreasing blood glucose levels by using IIT has affected health outcomes in various inpatient settings. Despite initial encouraging evidence from a study of patients in the SICU (16), subsequent trials have shown no consistent evidence that IIT decreases mortality, length of stay, or infection rates. Moreover, IIT interventions in these trials were associated with a nearly 6-fold increased risk for hypoglycemia.

Although the health outcome benefits of more moderate glucose control (for example, blood glucose levels targeted to 7.8 to 10.0 mmol/L [140 to 180 mg/dL]) have not been studied, the association of increased blood glucose levels with an increased risk for infection, dehydration, and poor hospital outcomes has caused many centers to pursue moderate degrees of blood glucose control. Given the lack of evidence showing that moderate glycemic control provides health outcome benefits, it is imperative that this intervention minimizes the risk for hypoglycemia. To evaluate the relative safety of various IIT approaches, we reviewed controlled and uncontrolled clinical trials of IIT, including trials reporting only intermediate outcomes (such as glucose control) as well as observational studies.

Methods

We searched MEDLINE and the Cochrane Database of Systematic Reviews for literature published from database inception through January 2010. Table 1 in the Supplement provides the search strategies in detail. We obtained additional articles from systematic reviews, reference lists of pertinent studies, reviews, and editorials and by consulting experts. We also searched for information about unpublished studies on ClinicalTrials.gov. All citations were imported into an electronic database (EndNote 9.0).

To assess the risk for hypoglycemia associated with IIT, we included controlled and uncontrolled studies that evaluated IIT protocols in hospitalized patients, even if the studies did not report health outcomes. We excluded IIT studies that did not report rates of hypoglycemia (10, 7787). To avoid studies with potential selection bias, we excluded prospective cohort studies in which patients were not consecutively enrolled or that had excessive loss to follow-up (8796). We also excluded studies in which the intervention was evaluated over a short period (defined as ≤6 months), because tight glycemic control strategies require personnel training and institutional acceptance and we believed that these studies were less likely to provide externally valid results (97102).

Results

We reviewed 3054 abstracts and retrieved 460 articles for full-text review. We also included 31 trials that reported at least 1 prespecified health outcome (mortality, infection, cardiovascular events, disability, or need for renal replacement therapy) in our accompanying meta-analysis on the benefits and harms of IIT in hospitalized patients. In the paragraphs below, we report the results of studies not designed to evaluate a prespecified health outcome other than hypoglycemia: specifically 29 studies reporting on the safety of insulin infusions and 4 studies reporting on the safety of subcutaneous insulin regimens.

Insulin Infusions

In 10 studies, the intervention groups achieved a mean blood glucose level of 6.7 mmol/L or less (≤120 mg/dL) (Table 7 in the Supplement). The definition of hypoglycemia differed across studies. Four studies reported relatively low rates of severe hypoglycemia; 3 of these studies used a complex computerized algorithm (103105), and 1 study used a sophisticated insulin infusion protocol that incorporated insulin sensitivity estimates (106). However, 4 studies reported high rates (defined as affecting >5% of participants) of severe hypoglycemia; 2 of these studies used a computerized algorithm (107108), and 2 studies used a relatively simple infusion protocol (109110).

Participants in 10 studies achieved more modest glucose control, defined as a mean blood glucose level of 6.7 to 7.8 mmol/L (120 mg/dL to 140 mg/dL), and almost all of these studies reported low rates of severe hypoglycemia. Most of these studies used relatively simple infusion protocols, and 1 study used a computerized algorithm (111). Of note, 1 small study (112) (the pilot study used to prepare for the NICE-SUGAR trial [17]) used a very simple infusion protocol and found a very high rate of severe hypoglycemia, whereas the phase 2 pilot study reported improved safety (110).

Two observational studies evaluated the safety of transitioning to progressively stricter blood glucose level targets over time. One of these trials was a very large single-center retrospective study evaluating the effects of an increasingly aggressive IIT policy in the ICU. The investigators found a nearly 4-fold increase in the incidence of hypoglycemia as the institution transitioned from no insulin protocols to using IIT to achieve a target blood glucose level of 4.4 to 7.2 mmol/L (80 to 130 mg/dL) and finally using this therapy to achieve a target blood glucose level of 4.4 to 6.1 mmol/L (80 to 110 mg/dL) (66). The infusion protocol details were not available. A second study of a relatively simple infusion protocol reported that the rate of severe hypoglycemia doubled as the blood glucose level target increased from 6.7 to 8.3 mmol/L (120 to 150 mg/dL) to 4.4 to 7.2 mmol/L (80 to 110 mg/dL); however, the overall rate of severe hypoglycemia remained at less than 5% (67).

We had excluded studies in which the intervention was evaluated for 6 months or less because we believed that they were less likely to provide externally valid results than longer studies. In these shorter-term studies, the definition of hypoglycemia varied and rates of hypoglycemia ranged from 4% to 14%.

Subcutaneous Insulin

Most trials evaluating health outcomes have used insulin infusions to achieve blood glucose control. However, subcutaneous insulin is used more often in clinical settings, especially in the general medicine ward. Subcutaneous sliding-scale insulin (SSI) regimens have numerous theoretical disadvantages when used as the sole method of achieving inpatient glycemic control, and researchers have called for a reduction in the widespread use of subcutaneous SSI (113).

Very few controlled trials have compared SSI with more intensive insulin regimens, and none of these trials has evaluated health outcomes other than hypoglycemia. Several small, single-center trials in general medical populations (114115) and gastric bypass recipients (116) found that SSI was less effective in lowering blood glucose levels than basal-bolus insulin regimens, although safety was similar between groups. A recent small randomized, controlled trial in general medical ward patients receiving enteral nutrition found that SSI and basal-bolus insulin treatment groups achieved similar blood glucose levels and had similar rates of hypoglycemia (117).

Discussion

The safety of IIT implementation probably depends on multiple factors. We reviewed dozens of insulin protocols and found that they differed in patient characteristics, target blood glucose ranges, the time required to achieve the target blood glucose levels, the incidence and definition of hypoglycemia, the rationale or algorithm used for adjusting the insulin rates, the methods used to assess effectiveness, and the methods of glucose monitoring. We found that IIT protocols using higher blood glucose targets generally were associated with lower rates of hypoglycemia. Several studies reported blood glucose levels less than 6.7 mmol/L (<120 mg/dL) with relatively low rates of hypoglycemia. In general, protocols that safely achieved this degree of glucose control were iteratively developed at single institutions and used complex protocols incorporating insulin sensitivity information and computerized algorithms. However, we did find exceptions, such as centers that reported high hypoglycemia rates despite using relatively sophisticated protocols. Moreover, observational studies suggest that adopting more aggressive IIT protocols over time is associated with a concomitant increase in rates of hypoglycemia.

Previous reviews speculated that better protocols would incorporate bolus insulin doses, account for the direction and rate of glucose change, and allow for “off-protocol” adjustments; however, this conclusion is not based on direct comparisons of protocols (6465). Given the various factors involved in insulin protocol implementation, experts have suggested that each institution should individualize its approach to protocol implementation on the basis of institutional and provider resources and its patient population (118). For example, patients in the MICU seem to have the highest risk for hypoglycemia (119122).

Three recent fair-quality systematic reviews (6466) have attempted to clarify characteristics of insulin infusion protocols that are able to decrease blood glucose levels without increasing rates of hypoglycemia, but no completed studies directly compared insulin infusion protocols. The reviews stressed that, given these various factors, each institution should individualize its approach to protocol implementation on the basis of its patient population and its institutional and provider resources (64).

Limitations

Several factors limit this body of evidence. Most of these trials were small, single-center studies conducted in ICU or surgical settings. This fact may limit the generalizability of findings, especially to general medical ward settings. Patient selection criteria were not always clear. Researchers conducting the included studies consecutively enrolled patients in the insulin protocol being investigated, but the method for choosing patients for the protocol were unclear (104, 106, 123). Studies also varied greatly in populations studied, glycemic targets, definitions of hypoglycemia, and the actual IIT protocols. However, no studies have directly compared differing insulin infusion approaches.

Finally, we reviewed the safety of subcutaneous SSI. The very limited body of evidence on this factor suggests that SSI and more complex basal-bolus regimens have similar safety profiles, although SSI may be less effective in controlling blood glucose levels.

Figures

Grahic Jump Location
Appendix Figure 1.
Literature search and selection.

MI = myocardial infarction; MICU = medical intensive care unit; SICU = surgical intensive care unit.

Grahic Jump Location
Grahic Jump Location
Figure.
Short-term mortality in studies of intensive insulin therapy, by inpatient setting.

Short-term mortality includes death occurring within 28 d of or during the ICU or hospital stay; we used 28-d mortality in the meta-analysis when a study reported >1 outcome. Events is the number of deaths among participants in the treatment and control groups. CABG = coronary artery bypass graft; CVA = cerebrovascular accident; ICU = intensive care unit; MI = myocardial infarction; MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 2.
Short-term mortality in studies of intensive insulin therapy, by the mean glucose level achieved in the intervention group.

Short-term mortality includes death occurring within 28 d of or during the ICU or hospital stay; we used 28-d mortality in the meta-analysis when a study reported >1 outcome. Events is the number of deaths among participants in the treatment and control groups. NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study.

* During the intraoperative period in Butterworth and colleagues' study (45), the treatment group achieved mean blood glucose levels of approximately 6.1 to 10.2 mmol/L (110 to 185 mg/dL). Glucose levels were not reported in Yang and colleagues' study (34).

Grahic Jump Location
Grahic Jump Location
Appendix Figure 3.
Mortality at 90 or 180 d in studies of intensive insulin therapy, by inpatient setting.

Events is the number of deaths among participants in the treatment and control groups. CABG = coronary artery bypass graft; CVA = cerebrovascular accident; ICU = intensive care unit; MI = myocardial infarction; MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 4.
Effects of intensive insulin therapy on rates of infection in various inpatient settings.

We included inpatients in the MICU, SICU, and perioperative settings as well as patients with stroke or acute brain injury. Events is the number of participants with 1 or more infections in the treatment and control groups. MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

* Includes wound infections, urinary tract infections, pneumonia, or a combination of these conditions.

† Data include only cardiopulmonary bypass subgroup.

‡ Data include only off-pump cardiopulmonary bypass subgroup.

Grahic Jump Location
Grahic Jump Location
Appendix Figure 5.
Risk for hypoglycemia in studies of intensive insulin therapy in various inpatient settings.

We included inpatients in the MICU, SICU, and perioperative settings as well as patients with traumatic brain injury. We defined hypoglycemia as a blood glucose level less than 2.2 mmol/L (<40 mg/dL). Events is the number of participants with 1 or more hypoglycemic episodes in the treatment and control groups. MICU = medical intensive care unit; NICE-SUGAR = Normoglycemia in Intensive Care Evaluation–Survival Using Glucose Algorithm Regulation study; SICU = surgical intensive care unit.

Grahic Jump Location

Tables

Table Jump PlaceholderTable 1.  Trials in Intensive Care Units
Table Jump PlaceholderTable 2.  Hypoglycemia Risk in Trials of Intensive Insulin Therapy
Table Jump PlaceholderTable 3.  Summary of the Evidence for the Effects of Intensive Insulin Therapy, by Outcome and Inpatient Setting

References

Umpierrez GE, Isaacs SD, Bazargan N, You X, Thaler LM, Kitabchi AE.  Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab. 2002; 87:978-82.
PubMed
CrossRef
 
Levetan CS, Passaro M, Jablonski K, Kass M, Ratner RE.  Unrecognized diabetes among hospitalized patients. Diabetes Care. 1998; 21:246-9.
PubMed
 
Furnary AP, Gao G, Grunkemeier GL, Wu Y, Zerr KJ, Bookin SO. et al.  Continuous insulin infusion reduces mortality in patients with diabetes undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2003; 125:1007-21.
PubMed
 
Pomposelli JJ, Baxter JK 3rd, Babineau TJ, Pomfret EA, Driscoll DF, Forse RA. et al.  Early postoperative glucose control predicts nosocomial infection rate in diabetic patients. JPEN J Parenter Enteral Nutr. 1998; 22:77-81.
PubMed
 
Bochicchio GV, Salzano L, Joshi M, Bochicchio K, Scalea TM.  Admission preoperative glucose is predictive of morbidity and mortality in trauma patients who require immediate operative intervention. Am Surg. 2005; 71:171-4.
PubMed
 
Capes SE, Hunt D, Malmberg K, Gerstein HC.  Stress hyperglycaemia and increased risk of death after myocardial infarction in patients with and without diabetes: a systematic overview. Lancet. 2000; 355:773-8.
PubMed
 
Capes SE, Hunt D, Malmberg K, Pathak P, Gerstein HC.  Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke. 2001; 32:2426-32.
PubMed
 
McCowen KC, Malhotra A, Bistrian BR.  Stress-induced hyperglycemia. Crit Care Clin. 2001; 17:107-24.
PubMed
 
Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer RG, et al. American Diabetes Association Diabetes in Hospitals Writing Committee.  Management of diabetes and hyperglycemia in hospitals. Diabetes Care. 2004; 27:553-91.
PubMed
 
Furnary AP, Zerr KJ, Grunkemeier GL, Starr A.  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.
PubMed
 
Hirshberg E, Lacroix J, Sward K, Willson D, Morris AH.  Blood glucose control in critically ill adults and children: a survey on stated practice. Chest. 2008; 133:1328-35.
PubMed
 
Institute for Health Care Improvement.  Implement effective glucose control: establish a glycemic control policy in your ICU. 2008. Accessed atwww.ihi.org/IHI/Topics/CriticalCare/IntensiveCare/Changes/IndividualChanges/EstablishaGlycemicControlPolicyinYourICU.htmon 28 December 2010.
 
Malmberg K, Rydén L, Efendic S, Herlitz J, Nicol P, Waldenström A. et al.  Randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year. J Am Coll Cardiol. 1995; 26:57-65.
PubMed
 
Furnary AP, Cheek DB, Holmes SC, Howell WL, Kelly SP.  Achieving tight glycemic control in the operating room: lessons learned from 12 years in the trenches of a paradigm shift in anesthetic care. Semin Thorac Cardiovasc Surg. 2006; 18:339-45.
PubMed
 
Zerr KJ, Furnary AP, Grunkemeier GL, Bookin S, Kanhere V, Starr A.  Glucose control lowers the risk of wound infection in diabetics after open heart operations. Ann Thorac Surg. 1997; 63:356-61.
PubMed
 
Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M. et al.  Intensive insulin therapy in the critically ill patients. N Engl J Med. 2001; 345:1359-67.
PubMed
 
Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, et al. NICE-SUGAR Study Investigators.  Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009; 360:1283-97.
PubMed
 
www.hsrd.research.va.gov/publications/esp/Hyperglycemia-2008.pdfKansagara D, Wolf F, Freeman M, Helfand M.  Management of inpatient hyperglycemia: a systematic review. U.S. Department of Veterans Affair Health Services Research and Development Service. Evidence-based Synthesis Program Report. Washington, DC: U.S. Department of Veterans Affairs; 2008.
 
Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM, et al. Methods Work Group, Third US Preventive Services Task Force.  Current methods of the US Preventive Services Task Force: a review of the process. Am J Prev Med. 2001; 20:21-35.
PubMed
 
DerSimonian R, Laird N.  Meta-analysis in clinical trials. Control Clin Trials. 1986; 7:177-88.
PubMed
 
Higgins JP, Thompson SG, Deeks JJ, Altman DG.  Measuring inconsistency in meta-analyses. BMJ. 2003; 327:557-60.
PubMed
 
Brozek JL, Akl EA, Alonso-Coello P, Lang D, Jaeschke R, Williams JW, et al. GRADE Working Group.  Grading quality of evidence and strength of recommendations in clinical practice guidelines. Part 1 of 3. An overview of the GRADE approach and grading quality of evidence about interventions. Allergy. 2009; 64:669-77.
PubMed
 
Iwamoto G.  Comparative trial between computer-guided intravenous infusion protocol versus a standard insulin infusion algorithm versus a simple calculated infusion protocol in medical and surgical ICU. Accessed athttp://clinicaltrials.gov/show/NCT00582309on 28 December 2010.
 
Umpierrez GE.  Comparative trial between a computer-guided intravenous infusion protocol versus a standard insulin infusion algorithm in medical ICU. Accessed athttp://clinicaltrials.gov/show/NCT00394524on 28 December 2010.
 
Lacherade J-C.  Multicentre randomized trial assessing the impact of maintaining 2 blood glucose levels on hospital mortality in patients admitted to the ICU (INSUREA study). Accessed athttp://clinicaltrials.gov/show/NCT00591071on 28 December 2010.
 
Grey NJ, Perdrizet GA.  Reduction of nosocomial infections in the surgical intensive-care unit by strict glycemic control. Endocr Pract. 2004; 10:Suppl 246-52.
PubMed
 
Kirdemir P, Yildirim V, Kiris I, Gulmen S, Kuralay E, Ibrisim E. et al.  Does continuous insulin therapy reduce postoperative supraventricular tachycardia incidence after coronary artery bypass operations in diabetic patients? J Cardiothorac Vasc Anesth. 2008; 22:383-7.
PubMed
 
Farah R, Samokhvalov A, Zviebel F, Makhoul N.  Insulin therapy of hyperglycemia in intensive care. Isr Med Assoc J. 2007; 9:140-2.
PubMed
 
Lazar HL, Chipkin SR, Fitzgerald CA, Bao Y, Cabral H, Apstein CS.  Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation. 2004; 109:1497-502.
PubMed
 
Li JY, Sun S, Wu SJ.  Continuous insulin infusion improves postoperative glucose control in patients with diabetes mellitus undergoing coronary artery bypass surgery. Tex Heart Inst J. 2006; 33:445-51.
PubMed
 
Smith A, Grattan A, Harper M, Royston D, Riedel BJ.  Coronary revascularization: a procedure in transition from on-pump to off-pump? The role of glucose-insulin-potassium revisited in a randomized, placebo-controlled study. J Cardiothorac Vasc Anesth. 2002; 16:413-20.
PubMed
 
Rasoul S, Ottervanger JP, Timmer JR, Svilaas T, Henriques JP, Dambrink JH. et al.  One year outcomes after glucose-insulin-potassium in ST elevation myocardial infarction. The Glucose-insulin-potassium study II. Int J Cardiol. 2007; 122:52-5.
PubMed
 
Gray CS, Hildreth AJ, Sandercock PA, O'Connell JE, Johnston DE, Cartlidge NE, et al. GIST Trialists Collaboration.  Glucose-potassium-insulin infusions in the management of post-stroke hyperglycaemia: the UK Glucose Insulin in Stroke Trial (GIST-UK). Lancet Neurol. 2007; 6:397-406.
PubMed
 
Yang M, Guo Q, Zhang X, Sun S, Wang Y, Zhao L. et al.  Intensive insulin therapy on infection rate, days in NICU, in-hospital mortality and neurological outcome in severe traumatic brain injury patients: a randomized controlled trial. Int J Nurs Stud. 2009; 46:753-8.
PubMed
 
Arabi YM, Dabbagh OC, Tamim HM, Al-Shimemeri AA, Memish ZA, Haddad SH. et al.  Intensive versus conventional insulin therapy: a randomized controlled trial in medical and surgical critically ill patients. Crit Care Med. 2008; 36:3190-7.
PubMed
 
Bilotta F, Caramia R, Paoloni FP, Delfini R, Rosa G.  Safety and efficacy of intensive insulin therapy in critical neurosurgical patients. Anesthesiology. 2009; 110:611-9.
PubMed
 
Van den Berghe G, Wilmer A, Hermans G, Meersseman W, Wouters PJ, Milants I. et al.  Intensive insulin therapy in the medical ICU. N Engl J Med. 2006; 354:449-61.
PubMed
 
Brunkhorst FM, Engel C, Bloos F, Meier-Hellmann A, Ragaller M, Weiler N, et al. German Competence Network Sepsis (SepNet).  Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med. 2008; 358:125-39.
PubMed
 
Savioli M, Cugno M, Polli F, Taccone P, Bellani G, Spanu P. et al.  Tight glycemic control may favor fibrinolysis in patients with sepsis. Crit Care Med. 2009; 37:424-31.
PubMed
 
Mackenzie IM, Ercole A, Blunt M, Ingle S, Palmer CR.  Glycaemic control and outcome in general intensive care: the East Anglian GLYCOGENIC study. Br J Intensive Care. 2008; 18:121-6.
 
Preiser JC, Devos P, Ruiz-Santana S, Mélot C, Annane D, Groeneveld J. et al.  A prospective randomised multi-centre controlled trial on tight glucose control by intensive insulin therapy in adult intensive care units: the Glucontrol study. Intensive Care Med. 2009; 35:1738-48.
PubMed
 
De La Rosa Gl C, Donado JH, Restrepo AH, Quintero AM, González LG, Saldarriaga NE, et al. Grupo de Investigacion en Cuidado intensivo: GICI-HPTU.  Strict glycaemic control in patients hospitalised in a mixed medical and surgical intensive care unit: a randomised clinical trial. Crit Care. 2008; 12:R120.
PubMed
 
Ingels C, Debaveye Y, Milants I, Buelens E, Peeraer A, Devriendt Y. et al.  Strict blood glucose control with insulin during intensive care after cardiac surgery: impact on 4-years survival, dependency on medical care, and quality-of-life. Eur Heart J. 2006; 27:2716-24.
PubMed
 
Barcellos Cda S, Wender OC, Azambuja PC.  Clinical and hemodynamic outcome following coronary artery bypass surgery in diabetic patients using glucose-insulin-potassium (GIK) solution: a randomized clinical trial. Rev Bras Cir Cardiovasc. 2007; 22:275-84.
PubMed
 
Butterworth J, Wagenknecht LE, Legault C, Zaccaro DJ, Kon ND, Hammon JW Jr. et al.  Attempted control of hyperglycemia during cardiopulmonary bypass fails to improve neurologic or neurobehavioral outcomes in patients without diabetes mellitus undergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2005; 130:1319.
PubMed
 
Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, O'Brien PC. et al.  Intensive intraoperative insulin therapy versus conventional glucose management during cardiac surgery: a randomized trial. Ann Intern Med. 2007; 146:233-43.
PubMed
 
Subramaniam B, Panzica PJ, Novack V, Mahmood F, Matyal R, Mitchell JD. et al.  Continuous perioperative insulin infusion decreases major cardiovascular events in patients undergoing vascular surgery: a prospective, randomized trial. Anesthesiology. 2009; 110:970-7.
PubMed
 
Cheung NW, Wong VW, McLean M.  The Hyperglycemia: Intensive Insulin Infusion in Infarction (HI-5) study: a randomized controlled trial of insulin infusion therapy for myocardial infarction. Diabetes Care. 2006; 29:765-70.
PubMed
 
Oksanen T, Skrifvars MB, Varpula T, Kuitunen A, Pettilä V, Nurmi J. et al.  Strict versus moderate glucose control after resuscitation from ventricular fibrillation. Intensive Care Med. 2007; 33:2093-100.
PubMed
 
van der Horst IC, Zijlstra F, van't Hof AW, Doggen CJ, de Boer MJ, Suryapranata H, et al. Zwolle Infarct Study Group.  Glucose-insulin-potassium infusion inpatients treated with primary angioplasty for acute myocardial infarction: the glucose-insulin-potassium study: a randomized trial. J Am Coll Cardiol. 2003; 42:784-91.
PubMed
 
Malmberg K, Rydén L, Wedel H, Birkeland K, Bootsma A, Dickstein K, et al. DIGAMI 2 Investigators.  Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2): effects on mortality and morbidity. Eur Heart J. 2005; 26:650-61.
PubMed
 
Azevedo JR, Lima ER, Cossetti RJ, Azevedo RP.  Intensive insulin therapy versus conventional glycemic control in patients with acute neurological injury: a prospective controlled trial. Arq Neuropsiquiatr. 2007; 65:733-8.
PubMed
 
Bruno A, Kent TA, Coull BM, Shankar RR, Saha C, Becker KJ. et al.  Treatment of hyperglycemia in ischemic stroke (THIS): a randomized pilot trial. Stroke. 2008; 39:384-9.
PubMed
 
Walters MR, Weir CJ, Lees KR.  A randomised, controlled pilot study to investigate the potential benefit of intervention with insulin in hyperglycaemic acute ischaemic stroke patients. Cerebrovasc Dis. 2006; 22:116-22.
PubMed
 
Van den Berghe G, Wouters PJ, Kesteloot K, Hilleman DE.  Analysis of healthcare resource utilization with intensive insulin therapy in critically ill patients. Crit Care Med. 2006; 34:612-6.
PubMed
 
Devos P, Preiser JC, Melot C.  Impact of tight glucose control by intensive insulin therapy on ICU mortality and the rate of hypoglycaemia: final results of the GLUCONTROL study. Intensive Care Med. 2007; 33:Suppl 2S189.
 
Wiener RS, Wiener DC, Larson RJ.  Benefits and risks of tight glucose control in critically ill adults: a meta-analysis. JAMA. 2008; 300:933-44.
PubMed
 
Griesdale DE, de Souza RJ, van Dam RM, Heyland DK, Cook DJ, Malhotra A. et al.  Intensive insulin therapy and mortality among critically ill patients: a meta-analysis including NICE-SUGAR study data. CMAJ. 2009; 180:821-7.
PubMed
 
Bistrian BR, McCowen KC.  Nutritional and metabolic support in the adult intensive care unit: key controversies. Crit Care Med. 2006; 34:1525-31.
PubMed
 
Heyland DK, Dhaliwal R, Drover JW, Gramlich L, Dodek P, Canadian Critical Care Clinical Practice Guidelines Committee.  Canadian clinical practice guidelines for nutrition support in mechanically ventilated, critically ill adult patients. JPEN J Parenter Enteral Nutr. 2003; 27:355-73.
PubMed
 
Kanji S, Buffie J, Hutton B, Bunting PS, Singh A, McDonald K. et al.  Reliability of point-of-care testing for glucose measurement in critically ill adults. Crit Care Med. 2005; 33:2778-85.
PubMed
 
Inzucchi SE, Siegel MD.  Glucose control in the ICU—how tight is too tight? [Editorial]. N Engl J Med. 2009; 360:1346-9.
PubMed
 
Nazer LH, Chow SL, Moghissi ES.  Insulin infusion protocols for critically ill patients: a highlight of differences and similarities. Endocr Pract. 2007; 13:137-46.
PubMed
 
Meijering S, Corstjens AM, Tulleken JE, Meertens JH, Zijlstra JG, Ligtenberg JJ.  Towards a feasible algorithm for tight glycaemic control in critically ill patients: a systematic review of the literature. Crit Care. 2006; 10:R19.
PubMed
 
Wilson M, Weinreb J, Hoo GW.  Intensive insulin therapy in critical care: a review of 12 protocols. Diabetes Care. 2007; 30:1005-11.
PubMed
 
Treggiari MM, Karir V, Yanez ND, Weiss NS, Daniel S, Deem SA.  Intensive insulin therapy and mortality in critically ill patients. Crit Care. 2008; 12:R29.
PubMed
 
Taylor BE, Schallom ME, Sona CS, Buchman TG, Boyle WA, Mazuski JE. et al.  Efficacy and safety of an insulin infusion protocol in a surgical ICU. J Am Coll Surg. 2006; 202:1-9.
PubMed
 
Kosiborod M, Inzucchi SE, Goyal A, Krumholz HM, Masoudi FA, Xiao L. et al.  Relationship between spontaneous and iatrogenic hypoglycemia and mortality in patients hospitalized with acute myocardial infarction. JAMA. 2009; 301:1556-64.
PubMed
 
Turchin A, Matheny ME, Shubina M, Scanlon JV, Greenwood B, Pendergrass ML.  Hypoglycemia and clinical outcomes in patients with diabetes hospitalized in the general ward. Diabetes Care. 2009; 32:1153-7.
PubMed
 
Whitmer RA, Karter AJ, Yaffe K, Quesenberry CP Jr, Selby JV.  Hypoglycemic episodes and risk of dementia in older patients with type 2 diabetes mellitus. JAMA. 2009; 301:1565-72.
PubMed
 
Svensson AM, McGuire DK, Abrahamsson P, Dellborg M.  Association between hyper- and hypoglycaemia and 2 year all-cause mortality risk in diabetic patients with acute coronary events. Eur Heart J. 2005; 26:1255-61.
PubMed
 
Finney SJ, Zekveld C, Elia A, Evans TW.  Glucose control and mortality in critically ill patients. JAMA. 2003; 290:2041-7.
PubMed
 
Smith SM, Oveson KE, Strauss W, Raven K, Lefevre MC, Ahmann AJ. et al.  Ultradian variation of blood glucose in intensive care unit patients receiving insulin infusions. Diabetes Care. 2007; 30:2503-5.
PubMed
 
Goldberg PA, Inzucchi SE.  Selling root canals: lessons learned from implementing a hospital insulin infusion protocol. Diabetes Spectrum. 2005; 18:28-33.
 
Rea RS, Donihi AC, Bobeck M, Herout P, McKaveney TP, Kane-Gill SL. et al.  Implementing an intravenous insulin infusion protocol in the intensive care unit. Am J Health Syst Pharm. 2007; 64:385-95.
PubMed
 
Aragon D.  Evaluation of nursing work effort and perceptions about blood glucose testing in tight glycemic control. Am J Crit Care. 2006; 15:370-7.
PubMed
 
Scalea TM, Bochicchio GV, Bochicchio KM, Johnson SB, Joshi M, Pyle A.  Tight glycemic control in critically injured trauma patients. Ann Surg. 2007; 246:605-10.
PubMed
 
Beilman GJ, Joseph JI.  Practical considerations for glucose control in hospitalized patients. Diabetes Technol Ther. 2005; 7:823-30.
PubMed
 
Button E, Keaton P.  Glycemic control after coronary bypass graft: using intravenous insulin regulated by a computerized system. Crit Care Nurs Clin North Am. 2006; 18.
PubMed
 
Davis ED, Harwood K, Midgett L, Mabrey M, Lien LF.  Implementation of a new intravenous insulin method on intermediate-care units in hospitalized patients. Diabetes Educ. 2005; 31.
PubMed
 
González-Michaca L, Ahumada M, Ponce-de-León S.  Insulin subcutaneous application vs. continuous infusion for postoperative blood glucose control in patients with non-insulin-dependent diabetes mellitus. Arch Med Res. 2002; 33:48-52.
PubMed
 
Markovitz LJ, Wiechmann RJ, Harris N, Hayden V, Cooper J, Johnson G. et al.  Description and evaluation of a glycemic management protocol for patients with diabetes undergoing heart surgery. Endocr Pract. 2002; 8:10-8.
PubMed
 
Raucoules-Aimé M, Labib Y, Levraut J, Gastaud P, Dolisi C, Grimaud D.  Use of i.v. insulin in well-controlled non-insulin-dependent diabetics undergoing major surgery. Br J Anaesth. 1996; 76:198-202.
PubMed
 
Rood E, Bosman RJ, van der Spoel JI, Taylor P, Zandstra DF.  Use of a computerized guideline for glucose regulation in the intensive care unit improved both guideline adherence and glucose regulation. J Am Med Inform Assoc. 2005; 12:172-80.
PubMed
 
Simmons D, Morton K, Laughton SJ, Scott DJ.  A comparison of two intravenous insulin regimens among surgical patients with insulin-dependent diabetes mellitus. Diabetes Educ. 1994; 20:422-7.
PubMed
 
Eigsti J, Henke K.  Innovative solutions: development and implementation of a tight blood glucose management protocol: one community hospital's experience. Dimens Crit Care Nurs. 2006; 25:62-5.
PubMed
 
Reed CC, Stewart RM, Sherman M, Myers JG, Corneille MG, Larson N. et al.  Intensive insulin protocol improves glucose control and is associated with a reduction in intensive care unit mortality. J Am Coll Surg. 2007; 204:1048-54.
PubMed
 
Chaney MA, Nikolov MP, Blakeman BP, Bakhos M.  Attempting to maintain normoglycemia during cardiopulmonary bypass with insulin may initiate postoperative hypoglycemia. Anesth Analg. 1999; 89:1091-5.
PubMed
 
Carvalho G, Moore A, Qizilbash B, Lachapelle K, Schricker T.  Maintenance of normoglycemia during cardiac surgery. Anesth Analg. 2004;99:319-24, table of contents. Anesth Analg. 2004; 99:319-24.
PubMed
 
Chee F, Fernando TL, Savkin AV, van Heeden V.  Expert PID control system for blood glucose control in critically ill patients. IEEE Trans Inf Technol Biomed. 2003; 7:419-25.
PubMed
 
Wong XW, Chase JG, Shaw GM, Hann CE, Lotz T, Lin J. et al.  Model predictive glycaemic regulation in critical illness using insulin and nutrition input: a pilot study. Med Eng Phys. 2006; 28:665-81.
PubMed
 
Hemmerling TM, Schmid MC, Schmidt J, Kern S, Jacobi KE.  Comparison of a continuous glucose-insulin-potassium infusion versus intermittent bolus application of insulin on perioperative glucose control and hormone status in insulin-treated type 2 diabetics. J Clin Anesth. 2001; 13:293-300.
PubMed
 
Quinn JA, Snyder SL, Berghoff JL, Colombo CS, Jacobi J.  A practical approach to hyperglycemia management in the intensive care unit: evaluation of an intensive insulin infusion protocol. Pharmacotherapy. 2006; 26:1410-20.
PubMed
 
Shaw GM, Chase JG, Wong J, Lin J, Lotz T, Le Compte AJ. et al.  Rethinking glycaemic control in critical illness—from concept to clinical practice change. Crit Care Resusc. 2006; 8:90-9.
PubMed
 
Rowen M, Schneider DJ, Pratley RE, Sobel BE.  On rendering continuous glucose monitoring ready for prime time in the cardiac care unit. Coron Artery Dis. 2007; 18:405-9.
PubMed
 
Osburne RC, Cook CB, Stockton L, Baird M, Harmon V, Keddo A. et al.  Improving hyperglycemia management in the intensive care unit: preliminary report of a nurse-driven quality improvement project using a redesigned insulin infusion algorithm. Diabetes Educ. 2006; 32:394-403.
PubMed
 
Elinav H, Wolf Z, Szalat A, Bdolah-Abram T, Glaser B, Raz I. et al.  In-hospital treatment of hyperglycemia: effects of intensified subcutaneous insulin treatment. Curr Med Res Opin. 2007; 23:757-65.
PubMed
 
Ku SY, Sayre CA, Hirsch IB, Kelly JL.  New insulin infusion protocol Improves blood glucose control in hospitalized patients without increasing hypoglycemia. Jt Comm J Qual Patient Saf. 2005; 31:141-7.
PubMed
 
Lien LF, Spratt SE, Woods Z, Osborne KK, Feinglos MN.  Optimizing hospital use of intravenous insulin therapy: improved management of hyperglycemia and error reduction with a new nomogram. Endocr Pract. 2005; 11:240-53.
PubMed
 
Vogelzang M, Loef BG, Regtien JG, van der Horst IC, van Assen H, Zijlstra F. et al.  Computer-assisted glucose control in critically ill patients. Intensive Care Med. 2008; 34:1421-7.
PubMed
 
Kanji S, Singh A, Tierney M, Meggison H, McIntyre L, Hebert PC.  Standardization of intravenous insulin therapy improves the efficiency and safety of blood glucose control in critically ill adults. Intensive Care Med. 2004; 30:804-10.
PubMed
 
Bland DK, Fankhanel Y, Langford E, Lee M, Lee SW, Maloney C. et al.  Intensive versus modified conventional control of blood glucose level in medical intensive care patients: a pilot study. Am J Crit Care. 2005; 14:370-6.
PubMed
 
Pachler C, Plank J, Weinhandl H, Chassin LJ, Wilinska ME, Kulnik R. et al.  Tight glycaemic control by an automated algorithm with time-variant sampling in medical ICU patients. Intensive Care Med. 2008; 34:1224-30.
PubMed
 
Juneja R, Roudebush C, Kumar N, Macy A, Golas A, Wall D. et al.  Utilization of a computerized intravenous insulin infusion program to control blood glucose in the intensive care unit. Diabetes Technol Ther. 2007; 9:232-40.
PubMed
 
Hovorka R, Kremen J, Blaha J, Matias M, Anderlova K, Bosanska L. et al.  Blood glucose control by a model predictive control algorithm with variable sampling rate versus a routine glucose management protocol in cardiac surgery patients: a randomized controlled trial. J Clin Endocrinol Metab. 2007; 92:2960-4.
PubMed
 
Braithwaite SS, Edkins R, Macgregor KL, Sredzienski ES, Houston M, Zarzaur B. et al.  Performance of a dose-defining insulin infusion protocol among trauma service intensive care unit admissions. Diabetes Technol Ther. 2006; 8:476-88.
PubMed
 
Dortch MJ, Mowery NT, Ozdas A, Dossett L, Cao H, Collier B. et al.  A computerized insulin infusion titration protocol improves glucose control with less hypoglycemia compared to a manual titration protocol in a trauma intensive care unit. JPEN J Parenter Enteral Nutr. 2008; 32:18-27.
PubMed
 
Toschlog EA, Newton C, Allen N, Newell MA, Goettler CE, Schenarts PJ. et al.  Morbidity reduction in critically ill trauma patients through use of a computerized insulin infusion protocol: a preliminary study. J Trauma. 2007; 62:1370-5.
PubMed
 
Iapichino G, Albicini M, Umbrello M, Sacconi F, Fermo I, Pavlovich R. et al.  Tight glycemic control does not affect asymmetric-dimethylarginine in septic patients. Intensive Care Med. 2008; 34:1843-50.
PubMed
 
Mitchell I, Knight E, Gissane J, Tamhane R, Kolli R, Leditschke IA, et al. Australian and New Zealand Intensive Care Society Clinical Trials Group.  A phase II randomised controlled trial of intensive insulin therapy in general intensive care patients. Crit Care Resusc. 2006; 8:289-93.
PubMed
 
Saager L, Collins GL, Burnside B, Tymkew H, Zhang L, Jacobsohn E. et al.  A randomized study in diabetic patients undergoing cardiac surgery comparing computer-guided glucose management with a standard sliding scale protocol. J Cardiothorac Vasc Anesth. 2008; 22:377-82.
PubMed
 
McMullin J, Brozek J, McDonald E, Clarke F, Jaeschke R, Heels-Ansdell D. et al.  Lowering of glucose in critical care: a randomized pilot trial. J Crit Care. 2007; 22:112-8.
PubMed
 
Metchick LN, Petit WA Jr, Inzucchi SE.  Inpatient management of diabetes mellitus. Am J Med. 2002; 113:317-23.
PubMed
 
Umpierrez GE, Smiley D, Zisman A, Prieto LM, Palacio A, Ceron M. et al.  Randomized study of basal-bolus insulin therapy in the inpatient management of patients with type 2 diabetes (RABBIT 2 trial). Diabetes Care. 2007; 30:2181-6.
PubMed
 
Dickerson LM, Ye X, Sack JL, Hueston WJ.  Glycemic control in medical inpatients with type 2 diabetes mellitus receiving sliding scale insulin regimens versus routine diabetes medications: a multicenter randomized controlled trial. Ann Fam Med. 2003; 1:29-35.
PubMed
 
Datta S, Qaadir A, Villanueva G, Baldwin D.  Once-daily insulin glargine versus 6-hour sliding scale regular insulin for control of hyperglycemia after a bariatric surgical procedure: a randomized clinical trial. Endocr Pract. 2007; 13:225-31.
PubMed
 
Korytkowski MT, Salata RJ, Koerbel GL, Selzer F, Karslioglu E, Idriss AM. et al.  Insulin therapy and glycemic control in hospitalized patients with diabetes during enteral nutrition therapy: a randomized controlled clinical trial. Diabetes Care. 2009; 32:594-6.
PubMed
 
Ahmann AJ, Maynard G.  Designing and implementing insulin infusion protocols and order sets. J Hosp Med. 2008; 3:42-54.
PubMed
 
Fischer KF, Lees JA, Newman JH.  Hypoglycemia in hospitalized patients. Causes and outcomes. N Engl J Med. 1986; 315:1245-50.
PubMed
 
Shilo S, Berezovsky S, Friedlander Y, Sonnenblick M.  Hypoglycemia in hospitalized nondiabetic older patients. J Am Geriatr Soc. 1998; 46:978-82.
PubMed
 
Vriesendorp TM, van Santen S, DeVries JH, de Jonge E, Rosendaal FR, Schultz MJ. et al.  Predisposing factors for hypoglycemia in the intensive care unit. Crit Care Med. 2006; 34:96-101.
PubMed
 
Krinsley J.  Glycemic control in critically ill patients: Leuven and beyond [Editorial]. Chest. 2007; 132:1-2.
PubMed
 
Cochran A, Davis L, Morris SE, Saffle JR.  Safety and efficacy of an intensive insulin protocol in a burn-trauma intensive care unit. J Burn Care Res. 2008; 29:187-91.
PubMed
 
Kreisel SH, Berschin UM, Hammes HP, Leweling H, Bertsch T, Hennerici MG. et al.  Pragmatic management of hyperglycaemia in acute ischaemic stroke: safety and feasibility of intensive intravenous insulin treatment. Cerebrovasc Dis. 2009; 27:167-75.
PubMed
 
DeSantis AJ, Schmeltz LR, Schmidt K, O'Shea-Mahler E, Rhee C, Wells A. et al.  Inpatient management of hyperglycemia: the Northwestern experience. Endocr Pract. 2006; 12:491-505.
PubMed
 
Goldberg PA, Sakharova OV, Barrett PW, Falko LN, Roussel MG, Bak L. et al.  Improving glycemic control in the cardiothoracic intensive care unit: clinical experience in two hospital settings. J Cardiothorac Vasc Anesth. 2004; 18:690-7.
PubMed
 
Zimmerman CR, Mlynarek ME, Jordan JA, Rajda CA, Horst HM.  An insulin infusion protocol in critically ill cardiothoracic surgery patients. Ann Pharmacother. 2004; 38:1123-9.
PubMed
 
Plank J, Blaha J, Cordingley J, Wilinska ME, Chassin LJ, Morgan C. et al.  Multicentric, randomized, controlled trial to evaluate blood glucose control by the model predictive control algorithm versus routine glucose management protocols in intensive care unit patients. Diabetes Care. 2006; 29:271-6.
PubMed
 
Dilkhush D, Lannigan J, Pedroff T, Riddle A, Tittle M.  Insulin infusion protocol for critical care units. Am J Health Syst Pharm. 2005; 62:2260-4.
PubMed
 
Vogelzang M, Zijlstra F, Nijsten MW.  Design and implementation of GRIP: a computerized glucose control system at a surgical intensive care unit. BMC Med Inform Decis Mak. 2005; 5:38.
PubMed
 
Tamaki M, Shimizu T, Kanazawa A, Tamura Y, Hanzawa A, Ebato C. et al.  Efficacy and safety of modified Yale insulin infusion protocol in Japanese diabetic patients after open-heart surgery. Diabetes Res Clin Pract. 2008; 81:296-302.
PubMed
 
Clayton SB, Mazur JE, Condren S, Hermayer KL, Strange C.  Evaluation of an intensive insulin protocol for septic patients in a medical intensive care unit. Crit Care Med. 2006; 34:2974-8.
PubMed
 
Chant C, Wilson G, Friedrich JO.  Validation of an insulin infusion nomogram for intensive glucose control in critically ill patients. Pharmacotherapy. 2005; 25:352-9.
PubMed
 
Balkin M, Mascioli C, Smith V, Alnachawati H, Mehrishi S, Saydain G. et al.  Achieving durable glucose control in the intensive care unit without hypoglycaemia: a new practical IV insulin protocol. Diabetes Metab Res Rev. 2007; 23:49-55.
PubMed
 
Krinsley JS.  Glycemic control, diabetic status, and mortality in a heterogeneous population of critically ill patients before and during the era of intensive glycemic management: six and one-half years experience at a university-affiliated community hospital. Semin Thorac Cardiovasc Surg. 2006; 18:317-25.
PubMed
 
Theilen BM, Gritzke KA, Knutsen PG, Riek AE, McGill JB, Sicard GA. et al.  Inpatient glycemic control on the vascular surgery service. Endocr Pract. 2008; 14:185-91.
PubMed
 
Bell DA, Strong AJ.  Glucose/insulin infusions in the treatment of subarachnoid haemorrhage: a feasibility study. Br J Neurosurg. 2005; 19:21-4.
PubMed
 
van Wezel HB, Zuurbier CJ, de Jonge E, van Dam EW, van Dijk J, Endert E. et al.  Differential effects of a perioperative hyperinsulinemic normoglycemic clamp on the neurohumoral stress response during coronary artery surgery. J Clin Endocrinol Metab. 2006; 91:4144-53.
PubMed
 
Miriam A, Korula G.  A simple glucose insulin regimen for perioperative blood glucose control: the Vellore regimen. Anesth Analg. 2004; 99:598-602.
PubMed
 

Letters

NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Comments

Submit a Comment
Submit a Comment

Supplements

Summary for Patients

Clinical Slide Sets

Terms of Use

The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.

Toolkit

Want to Subscribe?

Learn more about subscription options

Advertisement
Related Articles
Related Point of Care
Topic Collections
PubMed Articles

Want to Subscribe?

Learn more about subscription options

Forgot your password?
Enter your username and email address. We'll send you a reminder to the email address on record.
(Required)
(Required)