Effectiveness of Tight Glycemic Control in the Medical Intensive Care Unit: A Systematic Review

Abstract

Review question/objective The objective is to conduct a systematic review to determine the best available evidence regarding the effectiveness of tight glycemic control in medical intensive care patients. The review questions are: What is the effectiveness of tight glycemic control on adverse outcomes in nonsurgical critically ill patients? More specifically, what is the rate of hypoglycemia and mortality in medical intensive care patients who are treated with tight glycemic control regimens? What should be the recommended blood glucose range for patients in a nonsurgical, medical ICU? Background Diabetes mellitus (DM) is the seventh leading cause of death in the United States affecting nearly 24 million people with a total direct and indirect cost of $218 billion8. The largest percentage of these dollars is spent on inpatient hospital care which equates to 50% of total expenditures1. The landmark study in 2001 by Van den Berghe et al.25 established improved outcomes for the surgical patient population and escalated a domino effect around the world as hospitals sought to implement tight glycemic control (TGC) in their intensive care areas. The benefits of TGC include decreased morbidity and mortality, lower lengths of stay, and reduced infection rates7,25. Tight glycemic control is defined as the utilization of intensive insulin therapy (IIT) to achieve and maintain blood glucose levels between 80-110 mg/dL (4.4-6.1 mmol/L) as opposed to conventional insulin therapy with blood glucose values equal to or greater than 110-180 mg/dL (6.1-10 mmol/L)3,11,21,24,25. Intensive Insulin Therapy (IIT) as a means to achieve TGC has become controversial in the medical intensive care unit (ICU) or mixed medical-surgical ICU. The benefits of tight control and support for IIT come from earlier published studies involving surgical, trauma or burn patients with comparison to conventional insulin therapy5,10,13,15,24,26. Recent studies have begun to reveal that this control may be too tight; patients in medical intensive care units may benefit from higher blood glucose values and reduced episodes of hypoglycemia4,11,14,17,18,21,24,26. Studies reviewed have determined that TGC is associated with an increased risk of hypoglycemia, with one study identifying a 50-fold increase over conventional insulin therapy3,4,6,9,11,14,18,20,21,22,23,26. Studies also indicate an increase in morbidity and/or mortality associated with the increased episodes of hypoglycemia4,6,18,21,23. Several authors agree that TGC does not establish adequate improvement for recommendation among all ICU patient populations3,6,9,14,18,21,23. The benefits of TGC in patients with severe sepsis and brain injury are inconclusive; studies reveal an increased risk of adverse events from hypoglycemia at previously recommended treatment goals suggesting benefit from looser control6, 16, 19, 22. It is also unclear why certain medical ICU patients demonstrate a higher mortality rate when treated with IIT for less than 72 hours but show a reduction beyond 3 days16,24. Adoption of intermediate blood glucose goals prevails in later studies although no consistent blood glucose range has yet been proposed or consistently implemented18,19,20,21,22. The American Association of Clinical Endocrinologists (AACE) and the American Diabetes Association (ADA) united to produce a consensus statement regarding inpatient glycemic control; they concluded that insulin infusions are still the superlative choice for management of hyperglycemia in most critically ill patients17. Both the AACE and the ADA recommend glucose targets no higher than 180 mg/dL and no less than 110 mg/dL2,17. Hyperglycemia is rampant among critically ill patients and poses great risks if left untreated. Insulin resistance resulting from increased cortisol levels, increased glucose production, catecholamine secretion, and cytokine release contributes to poor outcomes 12. Tight glycemic control (TGC) by means of intensive insulin therapy (IIT) has been shown to be beneficial in the adult surgical patient population7,25, but remains unconvincing for medical intensive care patients due to increased rates of hypoglycemia3,4,6,9,11,14,18,20,21,22,23,26. Disparities among patients may be attributable to severity of illness, difference in protocols, patient characteristics, and nursing workload4,11,18,22. The question still remains as to the benefit of tight glycemic control in all patient populations and the optimal target blood glucose range. Lack of a published or unpublished systematic review surrounding the controversies identified in the literature warrants the completion of this systematic review in order to determine the best available evidence regarding the effectiveness of tight glycemic control in this at-risk patient population. Inclusion criteria Types of participants The review will consider studies that include all patients 18 years of age and older, females and males, all types of conditions or diseases, and all stages of severity admitted to a critical care or intensive care unit who require hyperglycemia management and have not had a surgical procedure. Types of intervention(s)/phenomena of interest The review will consider studies that evaluate the effectiveness of tight glycemic control defined as blood glucose levels between 80-110 mg/dL (4.4-6.1 mmol/L) as compared to conventional insulin therapy with blood glucose values equal to or greater than 110-180 mg/dL (6.1-10 mmol/L). Types of outcomes This review will consider studies that include the following outcome measures: frequency and severity of hypoglycemia (defined as blood glucose < or = 2.2 mmol/L or 40 mg/dL), and frequency of intensive care unit mortality, hospital mortality, and/or 28 day mortality. Types of studies The review will consider any randomised controlled trials; in the absence of RCTs other research designs, such as non-randomised controlled trials and before and after studies, will be considered for inclusion to enable the identification of current best evidence regarding the effects of tight glycemic control in medical ICU patients. Search strategy The search strategy aims to find both published and unpublished studies in English language only. A three-step search strategy will be utilized. An initial limited search of MEDLINE and CINAHL will be undertaken followed by analysis of the text words contained in the title and abstract, and of the index terms used to describe article. A second search using all identified keywords and index terms will then be undertaken across all included databases. Thirdly, the reference list of all identified reports and articles will be searched for additional studies. The databases to be searched include: MEDLINE (1950 to present), AMED, OVID All EBM Reviews (collection includes Cochrane DSR, ACP Journal Club, DARE, CCTR, CMR, HTA, & NHSEED), CINAHL, EMBASE, TRIP, NGC, EBSCO, PubMed, Current Contents, ProQuest, Science Direct, and Expanded Academic ASAP. The search for unpublished studies will include: AHRQ (Agency for Healthcare Research and Quality), Evidence Reports, Digital Dissertations - DAI, Google Scholar, SIGLE (System for Information on Grey Literature in Europe), The New York Academy of Medicine, Gray Literature Report Networked Digital Library of Theses and Dissertations (NDLTD), Scirus, and Mednar. Initial keywords to be used will be: tight glyc$ control, intensive insulin therapy intensive insulin, critical care, and intensive care. Assessment of methodological quality Quantitative papers selected for retrieval will be assessed by two independent reviewers for methodological validity prior to inclusion in the review using standardised critical appraisal instruments from the Joanna Briggs Institute Meta Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI) (Appendix). Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer. Data collection Quantitative data will be extracted from papers included in the review using the standardised data extraction tool from JBI-MAStARI (Appendix). The data extracted will include specific details about the interventions, populations, study methods and outcomes of significance to the review question and specific objectives. Data synthesis Quantitative papers will, where possible be pooled in statistical meta-analysis using the Joanna Briggs Institute Meta Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI). All results will be subject to double data entry. Odds ratio (for categorical data) and weighted mean differences (for continuous data) and their 95% confidence intervals will be calculated for analysis. Heterogeneity will be assessed using the standard Chi-square. Where statistical pooling is not possible the findings will be presented in narrative form. Conflicts of interest none