Blood glucose target levels for critically ill adult patients

  • Protocol
  • Intervention

Authors

  • Stephan C Kettner,

    Corresponding author
    1. Medical University of Vienna, General Hospital Vienna, Department of Anaesthesia, General Intensive Care and Pain Control, Vienna, Austria
    • Stephan C Kettner, Department of Anaesthesia, General Intensive Care and Pain Control, Medical University of Vienna, General Hospital Vienna, Waehringer Guertel 18-20, Vienna, A-1090, Austria. stephan.kettner@meduniwien.ac.at.

    Search for more papers by this author
  • Reinhard Hahn,

    1. Medical University of Vienna, General Hospital Vienna, Department of Anaesthesia, General Intensive Care and Pain Control, Vienna, Austria
    Search for more papers by this author
  • Alois Gessl,

    1. Medical University of Vienna, Vienna General Hospital, Department of Internal Medicine III, Division of Endocrinology, Vienna, Austria
    Search for more papers by this author
  • Harald Herkner

    1. Medical University of Vienna, Department of Emergency Medicine, Vienna, Austria
    Search for more papers by this author

Abstract

This is the protocol for a review and there is no abstract. The objectives are as follows:

The primary objective of this systematic review is to assess the effects of maintenance of blood glucose within defined target levels, relative to usual care or defined higher target levels, on mortality and morbidity in adult patients receiving intensive care.

As secondary objectives, we will examine whether these effects are modified by target glucose levels; underlying conditions; dose and route of insulin; glucose level, nutrition, co-medication; and structure of the protocol. We will also investigate the robustness of our estimates by performing sensitivity analyses according to risk of bias and methods of glucose measurement.

Background

Hyperglycaemia is usually defined as a blood glucose value > 140 mg/dL (> 7.8 mmol/L) in the fasting unstressed patient (Moghissi 2009). Hypoglycaemia is defined as a blood glucose value < 70 mg/dL (< 3.9 mmol/L), moderate hypoglycaemia as < 55 mg/dL (< 3.0 mmol/L), and severe hypoglycaemia as a blood glucose level < 40 mg/dL (< 2.2 mmol/L) (Moghissi 2009). In the critically ill patient, hyperglycaemia commonly occurs and is associated with an increased mortality (Dungan 2009). Traditional management of blood glucose in the critically ill patient has focused only on preventing hyperglycaemia and hypoglycaemia (van den Berghe 2001), where the target glucose levels for hyperglycaemia have been variously defined. In the typical management strategy substantial delay may occur in the correction of blood glucose abnormalities, cut-off limits for hyperglycaemia and hypoglycaemia are variable, and methods for blood glucose determination are not standardized. In 2001, a large randomized controlled trial (RCT) in critically ill surgical patients showed a substantially reduced mortality with an insulin dosing protocol designed to maintain blood glucose within narrow limits (target levels between 80 and 110 mg/dL) compared to standard treatment (van den Berghe 2001). Subsequently maintenance of glucose within defined target limits in critically ill patients was investigated in a number of studies, showing conflicting results (Wiener 2008). A potentially harmful side effect of normoglycaemia achieved by insulin infusion is the high rate of hypoglycaemia reported in many studies (Wiener 2008). The difficulties in evaluating the findings of the published studies include problems in achieving normoglycaemia in critically ill patients; the controversy over the appropriate target glucose range; the different intensive care unit (ICU) settings (medical versus surgical); and concerns about their external validity (Finfer 2009). Besides these controversies, a variable terminology has been used to describe normoglycaemia for critically ill patients, including as examples tight glucose control (Wiener 2008), strict glycaemic control (De La Rosa 2008), strict glucose control (Oksanen 2007), intensive glucose control (Finfer 2009), and intensive insulin therapy (van den Berghe 2001). Accordingly, systematic reviews and meta-analyses have inconsistent conclusions (Griesdale 2009; Marik 2010; Wiener 2008).

Description of the condition

Stress induced hyperglycaemia, irrespective of diabetes, is very common in critically ill patients in the ICU (Dungan 2009). Numerous responsible cytokines and counter-regulatory hormones have been implicated. These include glucagon, epinephrine (adrenaline), glucocorticoids, growth hormone, and insulin-like growth factor. Overall these factors can result in decreased insulin secretion from the pancreas and decreased insulin sensitivity in peripheral tissues (Lee 1987). Insulin sensitivity of the liver is, however, only mildly affected (Lang 1992). Consequently excessive hepatic gluconeogenesis results in acute hyperglycaemia and insulin resistance syndrome.

Description of the intervention

The intervention is defined as the treatments applied to achieve a defined lower target level for blood glucose in critically ill patients, yielding normoglycaemia as defined by the authors. This intervention is compared to target levels for blood glucose higher than in the intervention group. These treatments may include insulin administration or a reduction of the daily administered amount of non-protein calories, or both.

How the intervention might work

In critically ill patients hyperglycaemia may increase complications such as polyneuropathy, severe infections, or multiple organ failure and thus increase mortality. The scientific concept is that any degree of hyperglycaemia above normal (110 mg/dL) in a condition of ischaemia and reperfusion additionally damages those cells that take up glucose passively. Normoglycaemia may prevent the deleterious effects of hyperglycaemia.

Why it is important to do this review

The effects of normoglycaemia on the outcomes of intensive care patients are still unclear. This review will evaluate RCTs and quasi-randomized controlled clinical trials that investigate the effects of blood glucose management strategies on clinically relevant outcomes. This review will consider studies that enrolled critically ill patients. We define critically ill patients as adult patients in intensive care units due to medical disease or surgical consequences.

Objectives

The primary objective of this systematic review is to assess the effects of maintenance of blood glucose within defined target levels, relative to usual care or defined higher target levels, on mortality and morbidity in adult patients receiving intensive care.

As secondary objectives, we will examine whether these effects are modified by target glucose levels; underlying conditions; dose and route of insulin; glucose level, nutrition, co-medication; and structure of the protocol. We will also investigate the robustness of our estimates by performing sensitivity analyses according to risk of bias and methods of glucose measurement.

Methods

Criteria for considering studies for this review

Types of studies

We will consider parallel group RCTs and quasi-randomized controlled clinical trials that evaluate the effect of normoglycaemia versus conventional strategies on clinically relevant outcomes.

Types of participants

We will include adult patients (defined as 16 years of age or older) in ICUs due to medical disease or surgical consequences.

We will exclude trials that solely investigate patients with acute myocardial infarction. This is because the pathophysiologic rationale for the benefit in cardiac patients (improved myocardial energy delivery) is different than that in other critically ill patients. The proposed mechanism of efficacy of normoglycaemia in critically ill patients is through the reduction of infection, improvement of coagulation abnormalities, and anti-inflammatory activity.

Types of interventions

The intervention is defined as the application of a defined target level for blood glucose in critically ill patients to yield normoglycaemia. We will accept the authors' definitions of conventional and experimental controls. To avoid overlap of the intervention groups and control groups due to expectedly varying blood glucose target levels across studies, we will not consider studies that used target levels for blood glucose above 150 mg/dL in the intervention group.

This intervention is compared to standard care or target levels for blood glucose higher than in the intervention group.

Blood glucose levels may be measured from arterial, capillary, or venous blood or plasma.

We do not intend to primarily investigate the effect of intensive insulin therapy versus less intensive insulin therapy, and likewise we will not investigate the effect of tight glucose control versus less frequent glucose measurements, although that might implicitly be included in some studies.

Types of outcome measures

We will evaluate the effectiveness of normoglycaemia by considering the following.

Primary outcomes

Overall mortality: assuming that mortality is variably reported, we will initially report the mortality closest to 28 days to reduce heterogeneity in reporting. Where available, we will also report long-term mortality (90 days or later).

Secondary outcomes
  1. Incidence of hypoglycaemia: moderate (less than 55 mg/dL (3.0 mmol/L)), severe (less than 40 mg/dL (2.2 mmol/L)), or as defined by study authors.

  2. Bloodstream infections, as defined by study authors.

  3. Ventilator associated pneumonia, as defined by study authors.

  4. ICU length of stay, assessed as a proxy of cost.

  5. Quality of life: objective measurements of physiologic function, psychological well-being, or return to work by participants as reported by study authors.

Search methods for identification of studies

Electronic searches

We will identify trials by searching the current Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (Appendix 1); MEDLINE (1966 to date) (Appendix 2); EMBASE (1980 to date) (Appendix 3); CINAHL (1982 to date) (Appendix 4); Database of Abstracts of Reviews of Effects (DARE) (1990 to date) and SCOPUS (1996 to date), using the Ovid platform. 

The highly sensitive search strategies described in Appendix 5b of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) will be combined with our subject search terms for searching for RCTs in MEDLINE and EMBASE. We will use the free text and the associated MeSH terms in combination with the RCT sensitive search strategy to identify studies for inclusion in the review. We will assess retrieved studies for any free text terms or MeSH terms for intensive insulin therapy that have not been used. We will incorporate any new terms that we identify into the search strategy. We will report the modified search strategy in full in the final review. We will adapt the MEDLINE (Appendix 1) search strategy to reflect the subject headings found in the thesauri used by EMBASE and CINAHL.

We will not apply language restrictions.

Searching other resources

We will search for trials by checking the reference lists of retrieved studies. We will search abstracts books of critical care medicine conferences in the Journal of Critical Care Medicine, Intensive Care Medicine, American Journal of Respiratory and Critical Care Medicine, and Anaesthesiology from 2000 to date.

We will search trial registries to identify unpublished studies: www.clinicaltrials.gov and www.controlled-trials.com/.

Data collection and analysis

Selection of studies

Two authors (SCK and RH) will independently assess studies on the basis of the title and abstract. A third author (HH) will assess an article for inclusion if there is disagreement between the first two authors and a majority decision will be made regarding inclusion. We will retrieve eligible studies in full. Two authors will independently assess full trial reports. A third author will assess the article for inclusion if there is disagreement between the first two authors and a majority decision will be made regarding inclusion. We will report the selection of studies in a flow chart.

Data extraction and management

We will extract data on type of participants, interventions, and outcomes in each trial using a standardized extraction form. We will attempt to contact the first author for clarification for published reports with insufficient information. Two authors (AG and RH) will independently and in duplicate perform data extraction. We will include in our extraction the following specific data.

(a) Participants (sex, age, setting, geographical regions, socioeconomic status, co-morbid conditions).

(b) Methods of glucose management (intervention versus control):

- blood glucose target levels (intervention group and control group);

- insulin (how delivered, doses and timing, and who delivered therapy);

- amount of intravenous dextrose that they received;

- amount and type of corticosteroids used, route (enteral versus parenteral);

- type of nutrition, and the caloric intake of the patients.

(c) Outcomes, as reported above.

(d) Items of bias risk, as reported below.

Assessment of risk of bias in included studies

We will systematically assess methods of study design and conduct. We will present risk of bias tables and a risk of bias graph for the main outcome. A non-numerical assessment will be used (Higgins 2011). Two authors will independently assess all studies (SCK and RH). A third author (HH) will resolve disagreements by vote. Items of bias risk will include:

  1. random sequence generation,

  2. allocation concealment,

  3. blinding of the participants and personal,

  4. blinding of outcome assessment,

  5. incomplete outcome data addressed (attrition bias),

  6. selective reporting,

  7. intention-to-treat (ITT) analysis,

  8. sponsoring by industry as potential source of bias, and

  9. other biases.

We will display the results by creating a 'risk of bias' graph and a 'risk of bias' summary figure using RevMan 5.1 software, if appropriate. We will present the risk of bias in the results section. We will provide summary assessments of the risk of bias for each outcome within and across studies.

Unit of analysis issues

The unit of analysis is to be based on the individual patient. We will include parallel study designs only, hence we do not expect design related unit of analysis issues such as in cross-over or cluster randomized trials. From parallel group studies that compare more than two intervention arms (for example more than two glucose target levels) we will use only one comparison per meta-analysis to avoid unit of analysis problems from this source.

If repeated outcomes are reported we will use the most distant outcome only to avoid such unit of analysis errors. Multiple treatment attempts per participant can cause a unit of analysis error. We will use data from patients as randomized and not from treatment attempts in order to calculate the estimates.

Dealing with missing data

We will contact authors to obtain information in the case of missing data for the primary outcome. If this information cannot be obtained, and for all other outcomes, we will use available data. We will not use algorithms to estimate missing data. If any study did not report withdrawals, then we plan to assume that there were no withdrawals.

Assessment of heterogeneity

We will assess clinical heterogeneity by inspection of the clinical variables across studies. We will conduct meta-analysis only if clinical heterogeneity is assumedly negligible. Otherwise we will present the data in a forest plot without a summary effect. Factors potentially introducing clinical or methodological heterogeneity will be tabulated and examined in detail (see Subgroup analysis and investigation of heterogeneity).

We will assess statistical heterogeneity by the I2 statistic. If I2 is larger than 40% we will assume relevant statistical heterogeneity and use random-effects models, otherwise we will use a fixed-effect model to estimate the summary effect across studies.

Assessment of reporting biases

We will assess the potential for reporting bias graphically using standard funnel plots and a contour enhanced funnel plot (Peters 2008) with the confunnel command in Stata (Stata Corp, College Station, TX).

Data synthesis

We will use ITT data where available. Effects will be assembled graphically in forest plots and initially explored graphically.

The effects from variables with frequency data will generally be presented as risk ratios. As we assume adverse events will be rare, we will use the Peto odds ratio to estimate the effect. For length of stay we will use weighted mean differences. Quality of life will assumedly be measured with different instruments, therefore we will use standardized mean differences.

We will produce a summary of findings table using the GRADEpro software (GRADEpro 2008).

Subgroup analysis and investigation of heterogeneity

We plan subgroup analyses to investigate potential sources of heterogeneity due to:

  1. medical versus surgical critical care,

  2. amount and route of insulin administration,

  3. amount of glucose infusion,

  4. amount and route of nutrition,

  5. amount of corticoids,

  6. potassium repletion,

  7. management of control groups, and

  8. the difference between target glucose levels and actual glucose levels in the intervention groups and control groups.

To assess differences between subgroups we will initially use the Q-partitioning method for categorised factors followed by meta-regression if sufficient data are available.

Sensitivity analysis

For the main outcome we will investigate the influence of within-study bias risk on the overall effect estimate. Therefore, we will repeat the analysis for studies at low risk of bias and for studies at undetermined or higher risk of bias. We plan a sensitivity analysis of the influence of different glucose measurement methods on the main effect.

Acknowledgements

We would like to thank Jane Cracknell (Managing Editor Cochrane Anaesthesia Review Group (CARG)) and Mark Neuman (Content Editor CARG) for their great support. We are grateful for helpful comments by the peer reviewers (Simon Finfer, Paul E Marik, Michael Nurok, Greet Van den Berghe, and Hannah Wunsch), our statistical editor (Marialena Trivella), and Karen Hovhannisyan (CARG TSC).

Appendices

Appendix 1. Search strategy for CENTRAL, The Cochrane Library

#1 MeSH descriptor Insulin explode all trees
#2 MeSH descriptor Hypoglycemic Agents explode all trees
#3 MeSH descriptor Blood Glucose, this term only
#4 (insulin or (hypoglyc?emic agent*)):ti,ab
#5 ((glyc?emic or glucose) near control)
#6 (#1 OR #2 OR #3 OR #4 OR #5)
#7 MeSH descriptor Critical Illness, this term only
#8 MeSH descriptor Critical Care, this term only
#9 MeSH descriptor Intensive Care, this term only
#10 MeSH descriptor Postoperative Care, this term only
#11 MeSH descriptor Intensive Care Units, this term only
#12 MeSH descriptor Burn Units explode all trees
#13 MeSH descriptor Coronary Care Units, this term only
#14 MeSH descriptor Postoperative Complications, this term only
#15 MeSH descriptor Postoperative Period, this term only
#16 ((critical* near ill*) or critical care or ICU or intensive care or burn unit* or coronary care):ti,ab or (postoperative* or post-operative*):ti,ab
#17 (#7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16)
#18 (#6 AND #17)

Appendix 2. Search strategy for MEDLINE (OvidSP)

1. exp Insulin/ or exp Hypoglycemic Agents/ or Blood Glucose/ or (insulin or hypoglyc?emic agent*).ti,ab. or ((glyc?emic or glucose) adj3 control).mp.
2. Critical Illness/ or Critical Care/ or Intensive Care/ or Postoperative Care/ or Intensive Care Units/ or Burn Units/ or Coronary Care/ or Postoperative Complications/ or exp Postoperative Period/ or ((critical* adj3 ill*) or critical care or ICU or intensive care or burn unit* or coronary care).ti,ab. or (postoperative* or post-operative*).ti,ab.
3. 1 and 2
4. ((randomized controlled trial or controlled clinical trial).pt. or randomized.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti.) not (animals not (humans and animals)).sh.
5. 3 and 4

Appendix 3. Search strategy for EMBASE (OvidSP)

1. exp insulin/ or antidiabetic agent/ or glucose blood level/ or (insulin or hypoglyc?emic agent*).ti,ab. or ((glyc?emic or glucose) adj3 control).ti,ab.
2. critical illness/ or intensive care/ or postoperative care/ or intensive care unit/ or coronary care unit/ or postoperative complication/ or postoperative period/ or ((critical* adj3 ill*) or critical care or ICU or intensive care or burn unit* or coronary care).ti,ab. or (postoperative* or post-operative*).ti,ab.
3. 1 and 2
4. (placebo.sh. or controlled study.ab. or random*.ti,ab. or trial*.ti,ab.) not (animals not (humans and animals)).sh.
5. 3 and 4

Appendix 4. Search strategy for CINAHL (EBSCOhost)

S1 (MM "Insulin+")
S2 (MM "Hypoglycemic Agents")
S3 (MH "Blood Glucose")
S4 AB insulin or AB hypoglyc?emic agent*
S5 AB ( glyc?emic or glucose ) and AB control
S6 S1 or S2 or S3 or S4 or S5
S7 (MH "Critical Illness")
S8 (MH "Critical Care")
S9 (MH "Intensive Care Units")
S10 (MH "Postoperative Care")
S11 (MH "Burn Units")
S12 (MH "Coronary Care Units")
S13 (MH "Postoperative Complications")
S14 (MH "Postoperative Period")
S15 AB ( critical care or ICU or intensive care or burn unit* or coronary care ) or AB critical* ill*
S16 TI postoperative*
S17 S7 or S8 or S9 or S10 or S11 or S12 or S13 or S14 or S15 or S16
S18 S6 and S17
S19 (MM "Random Assignment") or (MH "Clinical Trials")
S20 (MH "Placebos")
S21 (MM "Multicenter Studies")
S22 (MM "Comparative Studies")
S23 (MM "Double-Blind Studies") or (MM "Single-Blind Studies") or (MM "Triple-Blind Studies")
S24 AB random* or TI random*
S25 S19 or S20 or S21 or S22 or S23 or S24
S26 S18 and S25

Appendix 5. Data extraction form

Name of author extracting data:            ____________________________

 

Date form completed:                          ____________________________

 

Study ID

Title

 

 

 

 

Study ID for RevMan

(Family name of first author and year of publication + letter if more than one per year, e.g. Smith 2001b)

 
Are there other articles of same study? (If Yes, write Study IDs)

 

Yes, Unclear, No

 

Study Eligibility

 (please circle)Source (page no. in report)

Type of study

Can the study be described as randomized?

 

Yes, Unclear, No

 

Participants

1. Were the participants adults (?16a) and in ICUs?

 

Yes, Unclear, No

 

Interventions

1. Was one group treated with strict normoglycaemia?

 

2. Was the other group treated conventional?

 

 

Yes, Unclear, No

 

Yes, Unclear, No

 

 

Outcomes: Did the study report any one of

1. The overall mortality?

 

2. Incidence of hypoglycaemia?

 

3. Ventilator associated pneumonia?

 

4. Bloodstream infections?

 

5. ICU length of stay?

 

6. Quality of life?

 

 

Yes, Unclear, No

 

Yes, Unclear, No

 

Yes, Unclear, No

 

Yes, Unclear, No

 

Yes, Unclear, No

 

Yes, Unclear, No

 

Conclusion: Do not proceed if any of the above answers are ‘No’. If study to be ‘included’ or ‘excluded & listed in excluded table’, record below the information to be inserted into tables. If included continue to page 2.

 

 ï  Included                    Excluded and should be listed in the excluded table

 ï  Excluded and should NOT be listed in the excluded table

 ï  More information needed before inclusion decision (specify):                                                      

Record for tables: 

 

 

 

 

 

Source of key information

Electronic database

(Which one?)

 

Unpublished source

(Where?)          

 

Personal communication

(From whom?) 

 
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)

Low risk

Unclear risk

High risk

 
Allocation concealment (selection bias)

Low risk

Unclear risk

High risk

 
Blinding of participants and personnel (performance bias)

Low risk

Unclear risk

High risk

 
Blinding of outcome assessment (detection bias)

Low risk

Unclear risk

High risk

 
Incomplete outcome data (attrition bias)

Low risk

Unclear risk

High risk

 
Selective reporting (reporting bias)

Low risk

Unclear risk

High risk

 
Intention-to-treat (ITT) analysis

Low risk

Unclear risk

High risk

 
Sponsoring by industry

Low risk

Unclear risk

High risk

 
Other bias

Low risk

Unclear risk

High risk

 

Description of Intervention                             Intervention                             Control

Glucose goal (mg/dL) .......................................................................................................... 
Morning glucose achieved (mg/dL)mean (SD)mean (SD)
Mean glucose achieved (mg/dL)mean (SD)mean (SD)
Other glucose achieved (mg/dL) (specify)mean (SD)mean (SD)
Insulin dose (units/day)mean (SD)mean (SD)
Corticosteroids (no./total no. (%))  
Nonprotein calories (kcal/day)mean (SD)mean (SD)
-       enteral route (kcal/day)mean (SD)mean (SD)
-       parenteral route (kcal/day)mean (SD)mean (SD)
-       intravenous glucose (kcal/day)mean (SD)mean (SD)
Potassium Repletionmean (SD)mean (SD)

 

Setting

Country 
SettingSingle ICU                                           > 1 ICU (specify no.)
Type of ICU (& no.)

Medical                                                           Surgical

Mixed unit                                            Other (specify)

 

Participants                               Intervention        Control

No of randomized participants (n)  

Age (mean/SD)

 

  
Ethnicity (Caucasian/ Asian/ African/ unclear)  

Sex of participants

(M/F numbers or %)

  
APACHE II score (time/ (mean/median))  
Inclusion criteria

 

 

 

 

 

 
Exclusion criteria

 

 

 

 

 

 

 

Measurements/ structure of the protocol

Study protocolA nurse driven  B physician driven        C other (specify)
Who measuredA RN                B physician                 C other (specify)
Type of measurementA glucometer    B unclear                      C other (specify)
Source of glucose measurement

1.Whole blood                                     2.Serum

A arterial           B venous                     C capillary

Glucose control protocol

(brief description)

 

 

 

 

 

 

 

Outcomes Continuous Data

Main Group / Subgroup

(circle)

MedicalSurgical ICUDiabeticNon-diabetic 
OutcomesUnit of measurementIntervention GroupControl GroupP-value95% CI or any further details if outcome only described in text
  nmeanSDmedianIQRnmeanSDmedianIQR 
ICU length of stay             
Quality of life             
                

 

Outcomes - Dichotomous Data

Outcomes

Intervention Group

 

 

Control Group

(n = )

 

P-value

Any further information
 unitn/N

mean

median

SD

SE

IQR

unitn/N

mean

median

SD

SE

IQR

  
Overall Mortality          
Incidence of hypoglycaemia          
Bloodstream infections          
Ventilator associated pneumonia          

 

Other information which you feel is relevant to the results:

Indicate if: any data were obtained from the primary author; if results were estimated from graphs etc; or calculated by you using a formula (this should be stated and the formula given). In general if results not reported in paper(s) are obtained this should be made clear here to be cited in review.

 

Freehand space for writing actions such as contact with study authors and changes

 

 

 

If the study describes subgroups, fill in the last page for every available group.

What's new

DateEventDescription
4 February 2013New citation required and major changes

This protocol is an updated version of a previously published Cochrane protocol (Henderson 2005).

  1. This new version has been taken over by a new review team (Kettner SC, Hahn R, Gessl A, Herkner H), updated and modified.

  2. The protocol's title has been changed from "Intensive insulin therapy and strict glucose control for critically ill patients" to "Blood glucose target levels for critically ill adult patients".

  3. The previous version (Henderson 2005) will be withdrawn from The Cochrane Library when this version is published.

History

Protocol first published: Issue 3, 2005

DateEventDescription
1 September 2008AmendedConverted to new review format.

Contributions of authors

Conceiving the review: SK, RH, AG, HH

Co-ordinating the review: SK

Undertaking manual searches: RH

Screening search results: SK, RH, AG

Organizing retrieval of papers: RH

Screening retrieved papers against inclusion criteria: RH, AG

Appraising quality of papers: RH, AG, HH

Abstracting data from papers: RH, AG, HH

Writing to authors of papers for additional information: RH

Providing additional data about papers: RH, AG

Obtaining and screening data on unpublished studies: AG, HH

Data management for the review: SK, HH

Entering data into Review Manager (RevMan 5.1): HH, SK

Analysis of data: HH, SK

Interpretation of data: SK, RH, AG, HH

Writing the review: SK, RH, AG, HH

Guarantor for the review (one author): SK

Declarations of interest

Stephan C Kettner: none known

Reinhard Hahn: none known

Alois Gessl: speaker for Novartis, Novo Nordisk, Sanofi Aventis, consulting fee for Novartis

Harald Herkner: none known

Sources of support

Internal sources

  • Medical University of Vienna, Vienna, Austria.

  • The Cochrane Anaesthesia Review Group, Rigshospitalet, Copenhagen, Denmark.

External sources

  • No sources of support supplied

Notes

This protocol is an updated version of a previously published Cochrane protocol (Henderson 2005).

  1. This new version has been taken over by a new review team (Kettner SC, Hahn R, Gessl A, Herkner H), updated and modified.

  2. The protocol's title has been changed from "Intensive insulin therapy and strict glucose control for critically ill patients" to "Blood glucose target levels for critically ill adult patients".

  3. The previous version (Henderson 2005) will be withdrawn from The Cochrane Library when this version is published.

Ancillary