Research investigation of glycemic control in critically ill patients has become prominent since a landmark study by van den Berghe and colleagues1 in 2001, and more recently with the release of the 2012 Surviving Sepsis Campaign guidelines.2 Early research on cardiac surgery patients suggested that maintaining serum glucose levels between 80 and 100 mg/dL reduced mortality1; however, multiple studies since have not replicated these data and have even found the opposite result. Current international guidelines for nondiabetic, critically ill patients recommend a conventional serum glucose level target of less than 180 mg/dL when spontaneous food intake is not possible.3 The newest guidelines for the management of sepsis also recommend a target glucose level less than 180 mg/dL in septic patients.2 This article discusses the current research on glucose control in medical and surgical critically ill patients and presents the rationale for the proposed guidelines. Specifically, this article reviews studies comparing intensive tight glucose control to more conventional glucose management. Glucose control is crucial in health and disease management and applies to both advanced practice nurses prescribing therapy, as well as critical care nurses who monitor glucose and manage complications at the bedside.
HYPERGLYCEMIA: NOT SO SWEET
Hyperglycemia is common among patients experiencing acute and critical illness. Normally, humans maintain a blood glucose level of 80 to 125 mg/dL. Nearly 80% of glucose is taken into the cell by 1 of 5 glucose transporter channel proteins. Glucose is also taken up into the muscle, with the help of insulin, and can be stored in the liver in the form of glycogen.4
During the “stress response” in critical illness (Figure 1), patients experience natural hyperglycemia regardless of diabetic history. During the first few days after injury, the body responds with hypermetabolism, a hyperdynamic cardiovascular state, and inflammation. When hypermetabolism occurs, hyperglycemia is a by-product, in conjunction with increased lactate production, increased oxygen consumption, and protein catabolism.5 During critical illness, glucose production is increased while insulin resistance occurs simultaneously, especially in conditions such as sepsis.5,6 Unlike in healthy individuals, the pancreas loses the ability to release insulin and block hepatic glucose production. In addition, the neuroendocrine axis pathways release hormones that worsen hyperglycemia. Hormones such as corticotrophin, growth hormone, and glucagon are elevated in critical illness and inhibit hepatic glycogenesis and peripheral glycolysis while promoting gluconeogenesis and the breakdown of glycogen in the liver and muscle (Figure 2).4
While mild hyperglycemia is initially helpful in potentiating the body’s inflammatory response, it can become detrimental, especially with the addition of an exogenous glucose supply as happens in many critically ill patients.5,6 Persistent hyperglycemia has been shown to cause abnormal immune function, increased infection rates, hemodynamic disturbances, and electromyocardial dysfunction.4 Controlling hyperglycemia is critical in improving patient outcomes and reducing mortality.
The issue of glycemic control in both diabetic and nondiabetic critically ill patients is becoming increasingly important to nurses. As research shows that hyperglycemia is detrimental to healing, advanced practice and bedside nurses must be aware of patient glucose levels and work to keep them within an appropriate range. Examining current literature and guidelines assists nurses in knowing what appropriate glucose management consists of. Critical illness is a vulnerable time for patients, and nurse practitioners are frequently responsible for day-to-day care of patients in the intensive care unit (ICU). The advanced practice nurse is an asset to the health care team and has the ability to educate nurses and physicians about glycemic control, in addition to monitoring and treating hyperglycemia.
ANALYSIS OF THE LITERATURE
Four studies examining the effect of tight serum glucose control versus conventional serum glucose control and the impact on mortality and morbidity were included. All 4 studies examined included a mix of men and women, and all subjects were required to be older than 18 years. In addition, all studies had patient mortality as the primary end point; however, studies differed on whether the study stopped with ICU mortality, or if they followed up patients in the hospital and after discharge for a period. The details of the articles are listed in Table 1.
Each study utilized different blood glucose target ranges (Table 2). Although the glucose ranges used in these studies are not dramatically different, the differences could impact the amount of insulin administered and frequency of titration needed in order to maintain blood glucoses in those specific ranges.
The protocols for checking blood glucose and administering insulin were similar among the studies, but not identical. Blood glucose levels were checked via capillary or arterial blood, hourly initially and increased intervals once the blood glucose level was stable. Blood glucose was checked more often in all studies if a patient was hypoglycemic, defined as a blood glucose less than 40 mg/dL.7-10 Intravenous regular insulin was administered in all studies when subjects had a glucose level greater than the upper limit of the target glucose range.7-10 Although protocols were relatively similar, using even slightly different methods and frequencies of glucose measurement could potentially affect the comparison of results.
IMPACT ON OUTCOMES
The 4 studies examined patient mortality as the primary end point of the study. All studies assessed patient mortality in the ICU, hospital, or both. In addition to mortality, the impact of insulin therapy on hypoglycemia, days of mechanical ventilation, incidence of infection, sepsis and organ failure, and length of stay in the ICU and hospital was measured.7-10
Two studies were designed to look at an absolute reduction in ICU mortality, with hospital mortality as a secondary end point.7,9 Conversely, the other 2 studies looked at death from any cause in the hospital.8,10 Patient mortality rates between the intensive insulin therapy (IIT) and conventional insulin therapy (CIT) groups in the ICU and the hospital are listed in Figure 3.
Intensive Care Unit
The 4 studies found differing results when comparing mortality in the ICU alone after implementing IIT or CIT. No study found significant reductions in mortality with the use of IIT in the ICU.7-10 Whereas Preiser and colleagues9 found slightly increased mortality in the IIT group, both Arabi and colleagues7 and van den Berghe and colleagues10 found the opposite trend. None of the results were statistically significant in these studies.
Among the 3 studies that looked specifically at ICU mortality, specific subgroups of patients were found to have differing mortality rates. Arabi and colleagues7 found a mild increase in mortality among patients who were admitted with a Glasgow Coma Scale score less than 9 and were randomized to the IIT group. The authors also noted that IIT appeared to decrease mortality in patients who had a baseline Acute Physiology and Chronic Health Evaluation II score less than 22 and in patients who had a baseline body mass index less than 26.2 kg/m2.7 These data hint that healthier patients may fare better with lower glucose levels than sicker patients. van den Berghe and colleagues10 found that patients treated with IIT and who stayed in the ICU for more than 3 days had significantly lower mortality than did patients treated with CIT. Although these results suggest a benefit of IIT in small groups of patients, the data are not strong enough to be generalizable and, at best, suggest areas for further research.
In addition to examining mortality in the ICU alone, 3 of the studies examined mortality rates during the entire hospital stay. The NICE-SUGAR study8 found higher mortality in the IIT group compared with CIT. Among those who died in the IIT group, cardiovascular causes were the most common cause of death. This study was the only one to find significant overall changes in hospital mortality when comparing insulin regimens
Conversely, the 2 other studies found the opposite trend. Arabi and colleagues7 and van den Berghe and colleagues10 noted higher mortality in the CIT group, indicating a trend toward improved outcomes with tighter glycemic control. In examining patients who stayed in the ICU for more than 3 days, van den Berghe and colleagues10 found a significant reduction in mortality for patients treated with IIT. These data were mimicked at the 90-day follow-up for patients treated for more than 3 days in the ICU.10 This information suggests benefit in 1 subgroup; however, predicting length of stay is many times impossible, so this information has limited clinical benefit.
Although the studies were designed to examine mortality, they also took note of the effect of glucose control on morbidity. All 4 studies used the criterion for hypoglycemia as a blood sugar less than 40 mg/dL; all studies showed that the group treated with IIT experienced significantly more hypoglycemia than the CIT group.7-10 Arabi and colleagues7 found that 28.6% of IIT patients experienced at least 1 episode of hypoglycemia versus only 3.6% of CIT patients. The other 3 studies showed much lower rates of hypoglycemia; however, the trend remained the same.8-10
In some studies, hypoglycemia appeared to have an impact on outcomes, which has great relevance to practitioners when considering IIT versus CIT. Arabi and colleagues7 found that patients who experienced hypoglycemia had significantly higher mortality. In this study, 25% of IIT patients who experienced hypoglycemia died, compared with 12.5% of CIT patients.7 Similarly, Preiser and colleagues9 found significantly higher mortality in IIT subjects. In addition, the patients experiencing hypoglycemia had higher Sequential Organ Failure Assessment scores, indicating more organ dysfunction. Preiser and colleagues9 found that each 1-point increase in Acute Physiology and Chronic Health Evaluation II score was a risk factor for experiencing hypoglycemia. Similarly, van den Berghe and colleagues10 found that patients who stayed in the ICU for more than 3 days underwent renal replacement therapy (RRT) or experienced liver failure were at higher risk for hypoglycemic events. Because all studies had a higher incidence of hypoglycemia with IIT, targeting a tight glucose level, especially in patients who have extended ICU stays or renal or hepatic involvement, may increase the risk for hypoglycemia-related mortality.
In addition to mortality and hypoglycemia, studies also looked at the impact of tight glucose control on mechanical ventilation needs. Prolonged mechanical ventilation can lead to increased morbidity and infection rates. Because hyperglycemia is also tied to worsened immune function and increased infection rates, weaning mechanical ventilation as early as possible is vital when treating critically ill patients.
Three studies found that patients in the IIT group experienced fewer ventilator days.7,9,10 The NICE-SUGAR Study8 found no difference between IIT and CIT in regard to mechanical ventilation as both groups had an average of 6.6 ventilator days. van den Berghe and colleagues10 were the only ones to find a significant reduction in ventilator days with the use of IIT. In this study, the significant results applied only to patients who stayed in the ICU for more than 3 days.10 Clinically, this information is important because fewer ventilator days may lead to less risk of contracting ventilator-associated pneumonia, which can greatly impact patient outcomes.
Infection and Sepsis
Hyperglycemia has been found to increase the risk of infection by decreasing immune function. As a result, patients without controlled glucose levels are at risk for further morbidity from infection and sepsis. The 4 studies did not find significant trends in infection and sepsis risk when comparing patients receiving IIT or CIT. Arabi and colleagues7 found a slight trend toward more incidence of sepsis in CIT patients. Both the NICE-SUGAR Study8 and the study by van den Berghe and colleagues10 found similar rates of positive blood cultures in both the CIT and IIT groups. While hyperglycemia is associated with abnormal immune function, keeping blood sugar within a tighter range does not appear to improve infection rates of critically ill patients.
Organ Failure and Involvement
Critical illness places patients at risk for multiple organ involvement and failure. Poor glucose control can potentiate failure because of the effect on immune function. All studies examined the effect of IIT and CIT on organ failure. Most commonly, renal function was examined and failure was determined by the requirement of RRT. Two studies found a trend among CIT patients toward increased organ failure and an increased need for RRT; however, the results were not significant in either study.7,9 The NICE-SUGAR Study8 found similar rates of renal impairment in the IIT and CIT groups.
As with mechanical ventilation, van den Berghe and colleagues10 were the only ones to find some significant differences when it came to renal failure. In this study, patients with IIT experienced fewer cases of newly acquired kidney injury, defined as creatinine that doubled from baseline or creatinine greater than 2.5 mg/dL.10 Similarly, patients who stayed in the ICU for more than 3 days and received IIT had significantly less kidney injury than did patients in the CIT group. Whereas the incidence of kidney injury was significantly lowered by IIT, requiring RRT was not significantly different.10 These data suggest that IIT may have a protective effect on kidney function; however, more research is needed to validate this theory.
Length of Stay
One of the minor subcategories examined was the impact of IIT on length of stay in the ICU and the hospital. Although this factor is not as impactful as an effect on mortality, length of stay impacts overall cost. Two studies found trends toward a shorter length of stay in patients treated with IIT.7,10 Arabi and colleagues7 found that patients treated with IIT stayed a mean of 9.6 days in the ICU versus 10.8 days with CIT subjects. Regarding total hospital stay, these researchers7 noted IIT patients stayed a mean of 54.1 days compared with 57.5 days with patients treated with CIT. van den Berghe and colleagues10 found that patients treated with IIT had significantly earlier discharges from both the ICU and the hospital.
When comparing these results to those discussed earlier related to reduced mechanical ventilation days, reduced infection, and less incidence of organ failure, it appears IIT may decrease length of stay by reducing some common morbidities associated with critical illness. Reducing length of stay is significant to providers and patients because even 1 less day in the hospital can result in thousands of dollars saved. Although potentially promising, the results presented in these 4 studies are not statistically significant and only weakly suggest a benefit in the length of stay. The impact on length of stay is an area of potential future research.
The literature analyzed in this article indicates trends toward similar or reduced length of stay when comparing IIT and CIT, thus suggesting a potential reduction in cost with lower glucose levels. Sadhu and colleagues11 examined the economic impact of IIT. In 1 study, each 50 mg/dL increase in glucose led to 0.76 more postoperative days and $2824 more in inpatient hospital charges.11 Another study found hyperglycemia on admission was associated with a longer length of stay and $1349 higher inpatient hospital charges.11 Although maintaining glucose less than 110 mg/dL is not recommended, it is possible that a parameter between that and the recommended 180 mg/dL would be both safe and cost-saving. It is important to ensure practitioners are providing not only evidence-based practice that improves patient outcomes, but also cost-effective care to avoid the rising costs of health care.
Although the results of the 4 studies presented are not statistically significant, there was a high tendency toward episodes of hypoglycemia leading to increased mortality or morbidity. The lack of statistically significant and definitive outcomes prevents us from recommending changing current practice. At this time, the American Association of Clinical Endocrinologists and American Diabetes Association consensus statement recommends a target glucose range of 140 to 180 mg/dL in most critically ill patients.12 This literature suggests that there may be some benefit in preventing morbidity with a lower glucose level, while still greater than 110 mg/dL. The current recommended upper range of 180 mg/dL is up for discussion and the topic for future research.
The studies performed examining IIT and CIT for glucose control raise topics of future research in this area. There remains a lack of research on the specific effects of glucose control on factors such as sepsis, mechanical ventilation, and length of stay. In addition, research is needed to determine what insulin protocol leads to the best outcomes. In the 4 studies analyzed, each author used a different protocol for checking blood glucose and titrating insulin. By comparing protocols and looking specifically at which one leads to the best outcomes, researchers can look at the topic of glycemic control using the same tool. A recently published article used a computerized insulin dose calculator in an attempt to reduce the incidence of human error and hypoglycemia. The study found better glycemic management and improved nursing satisfaction with the computerized tool.13 Larger studies based on this type of approach can help determine the best method of dosing insulin in an attempt to avoid adverse effects, specifically hypoglycemia, which can cause poor patient outcomes.
Although more research is needed in determining the best way to manage hyperglycemia in critical illness, the research presently supports close monitoring and avoiding episodes of hypoglycemia. Nurse practitioners and other providers have critical roles in providing evidence-based care. Managing hyperglycemia is currently conducted according to hospital policy; however, nurse leaders are instrumental in changing policies and updating them based on current evidence. As more research is created, nurses will be able to provide quality, evidence-based care for patients experiencing hyperglycemia, while reducing adverse outcomes that may increase the risk of mortality.
1. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med. 2001; 345 (19): 1359–1367.
2. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock 2012. Crit Care Med. 2012; 41 (2): 580–637.
3. Ichai C, Preiser JC. International recommendations for glucose control in adult non diabetic critically ill patients. Crit Care (London, England). 2010; 14 (5): R166.
4. Bochicchio GV, Scalea TM. Glycemic control in the ICU. Adv Surg. 2008; 42: 261–275.
5. van Cromphaut SJ. Hyperglycaemia as part of the stress response: the underlying mechanisms. Best Pract Res Clin Anaesthesiol. 2009; 23 (4): 375–386.
6. Losser MR, Damoisel C, Payen D. Bench-to-bedside review: glucose and stress conditions in the intensive care unit. Crit Care (London, England). 2010; 14 (4): 231.
7. Arabi YM, Dabbagh OC, Tamim HM, et al. Intensive versus conventional insulin therapy: a randomized controlled trial in medical and surgical critically ill patients. Crit Care Med. 2008; 36 (12): 3190–3197.
8. NICE-SUGAR Study, Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009; 360 (13): 1283–1297.
9. Preiser JC, Devos P, Ruiz-Santana S, 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 Medicine. 2009; 35 (10): 1738–1748.
10. van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin therapy in the medical ICU. N Engl J Med. 2006; 354 (5): 449–461.
11. Sadhu AR, Ang AC, Ingram-Drake LA, Martinez DS, Hsueh WA, Ettner SL. Economic benefits of intensive insulin therapy in critically ill patients: the Targeted Insulin Therapy to Improve Hospital Outcomes (TRIUMPH) project. Diabetes Care. 2008; 31 (8): 1556–1561.
12. Moghissi ES, Korytkowski MT, DiNardo M, et al. American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Endocr Pract. 2009; 15 (4): 353–369.
13. Dumont C, Bourguignon C. Effect of a computerized insulin dose calculator on the process of glycemic control. Am J Crit Care. 2012; 21 (2): 106–115.
1KEY: RRT- renal replacement therapy; IIT- intensive insulin therapy; CIT- conventional insulin therapy; RBC- red blood cell; ICU- intensive care unit; BMI- body mass index; APACHE II- Acute Physiology and Chronic Health Evaluation II; SOFA- Sequential Organ Failure Assessment; LOS- length of stay.