Disorders of glucose control or blood glucose disorders

Introduction

Alterations in glucose metabolism are common in the intensive care unit (ICU). The adverse effects of uncontrolled hyperglycemia, especially in a critically ill patient, are fairly well described. Management strategies for hyperglycemia in the critically ill patient have gone through substantial changes over the past two decades. Although hyperglycemia is associated with adverse clinical outcomes, trying to achieve euglycemia in all patients has not proven to be the solution. Although the initial landmark randomized controlled trial showed remarkable benefits to tight glycemic management in the ICU, attempts by other investigators to replicate those findings were futile. On the contrary, it resulted in an increase in hypoglycemic episodes and consequently mortality, as evidenced by the largest multicenter randomized controlled trial to date. ^, It is worth noting that there is an exceedingly large body of evidence from observational and retrospective studies shedding light on the various glycemic domains that clinicians should keep in mind while managing hyperglycemia in both patients with and without diabetes. The presence of a history of diabetes and the extent of a patient’s chronic glycemic state would probably play a role in determining the glycemic target. ^, Furthermore, glycemic variability, regardless of the chosen target, may also have an impact on glycemic management outcomes.

Hyperglycemia is associated with harm in the ICU, as is hypoglycemia, which is equally, if not even more, serious than hyperglycemia in its detrimental impact. ^, Hypoglycemia thus deserves much attention concerning its etiology, risk factors, and interventions aimed at prevention and timely treatment.

This chapter gives an overview of hyperglycemia, hypoglycemia, and their management strategies in the critically ill patient.

Hyperglycemia

Definition of hyperglycemia

When a critically ill patient is hyperglycemic, it is important to delineate the clinical group to which the patient belongs. These include the following:

A known diabetic
Undiagnosed diabetic
Nondiabetic with new-onset hyperglycemia (stress-induced)

The American Diabetes Association (ADA) has published criteria for the diagnosis of diabetes mellitus. A diagnosis of diabetes is established when any one of the following criteria is met:

Fasting plasma glucose >126 mg/dL (7 mmol/L)
Postprandial plasma glucose >200 mg/dL (11.1 mmol/L) after 2 hours of a 75-g oral glucose tolerance test
HbA _1c >6.5%
Classic symptoms of hyperglycemic crisis with a random glucose >200 mg/dL (11.1 mmol/L)

Unfortunately, there is no set definition for hyperglycemia in the critically ill patient, as it is confounded by the presence of multiple variables, including the severity of illness, concurrent administration of glucose-containing medications such as antibiotics, feeding, and hyperglycemia-inducing medication, such as catecholamine infusions or glucocorticoids. When a patient presents with new-onset hyperglycemia (NOH), whether critically ill or not, it is prudent to rule out the presence of undiagnosed diabetes mellitus. The presence of an HbA _1c >6.5% might aid in making this diagnosis. The ADA/American Association of Clinical Endocrinologists (ACCE) in their consensus statement recommend that any blood glucose >180 mg/dL (10 mmol/L) be treated in patients who are critically ill in an attempt to reduce the adverse outcomes associated with hyperglycemia.

Stress-induced hyperglycemia

Hyperglycemia was initially considered to be a normal adaptive response in a critically ill patient to survive a period of acute stress. An increase in the levels of catecholamines, growth hormone, exogenous and endogenous glucocorticoids, and glucagon, along with an increase in circulating cytokines and peripheral insulin resistance, may play an important role in the genesis of stress-induced hyperglycemia. ^, However, stress hyperglycemia in patients with previously normal glucose homeostasis has been associated with an increase in adverse outcomes in patients who are critically ill. These adverse outcomes were attributed to either hyperglycemia per se or free radical injury and its related adverse coronary and intracranial events. Although hyperglycemia by itself during acute illness may not be the actual cause of the increase in mortality, it may indicate and correlate with the degree of underlying disease severity.

Diabetic versus nondiabetic blood glucose management in the ICU

Recent evidence indicates that glycemic targets in the ICU should probably vary based on the presence of a history of diabetes and the chronic glycemic state before ICU admission. ^, ^, ^, ^, Furthermore, Abdelmalak and colleagues reported that 11% of patients presenting for noncardiac surgery have undiagnosed diabetes, and there is no reason not to believe that many patients are probably admitted to ICU who are also undiagnosed diabetics. Diabetic patients who are chronically hyperglycemic can probably tolerate, or might even do better with, a higher-than-normal glycemic target and reduced glycemic variability. A retrospective study of 450 critically ill diabetic patients with a total of 9946 glucose measurements in the study cohort found that patients who had higher (>7%) preadmission HbA _1c levels had a lower mortality when the ICU time-weighted glucose concentration was higher (>180 mg/dL [10 mmol/L]), as compared with patients who had lower HbA _1c (<7%). Similar findings were reported in larger patient cohorts. ^, ^, A meta-analysis of nine studies was conducted in 2012 with regard to hyperglycemia and in-hospital mortality associated with sepsis, measuring unadjusted ICU mortality in five of the nine studies. They observed an increase in ICU mortality and a 2.7-fold increase in hospital mortality in patients with NOH as opposed to diabetic patients with hyperglycemia. This difference may be the result of the fact that in a previously chronically hyperglycemic diabetic, the organs and immune system may have been accustomed to higher blood glucose concentrations, and an acute attempt at euglycemia presents a state of relative hypoglycemia for them. Furthermore, if such an attempt at euglycemia is associated with moderate or severe hypoglycemia, as is the case with many tight glycemic control algorithms, such dangerously low concentrations are harmful by themselves ^, and contribute to a significant increase in glycemic variability. One of the largest database studies evaluating the effect of diabetes on outcomes in critically ill patients found that hyperglycemia was more detrimental in nondiabetics and was potentially protective in diabetics. There was an increased risk of mortality exclusively in diabetic patients in the normoglycemic/hypoglycemic range, suggesting a relative intolerance to relative and absolute hypoglycemia in this subset of patients. In light of these findings, we probably should seek different glycemic targets (and more robust insulin protocols to achieve such targets) that vary based on the presence of a diagnosis of diabetes and previous chronic glycemic states. Marik and Egi have proposed an example of such variable glycemic targets based on the patient’s clinical condition, history of diabetes, and HbA _1c , signifying the extent of the long-term status of glycemic control.

The glycemic domains and the influence of preexisting diabetes diagnosis on outcomes in the ICU

The three domains of glycemic control are hyperglycemia, hypoglycemia, and glucose variability. Each of these has independently been shown to increase mortality in critically ill patients. ^, ^, ^, ^, Two large studies evaluated these three domains and the approach to glycemic management in the ICU, in addition to the role of diabetes in this complex milieu. The results of both studies are strikingly similar. Hyperglycemia and an increase in glucose variability were associated with an increase in mortality in nondiabetic but not in diabetic patients. On the other hand, hypoglycemia was associated with an increase in mortality in both groups. Furthermore, in diabetic patients, premorbid glycemic control has been shown to play an important role in outcomes based on the glycemic goals. In patients with poor preadmission glycemic control, mortality was higher with lower ICU blood glucose levels than their own “normal” (i.e., relative hypoglycemia). Conversely, in patients with good preadmission glycemic control, survival was higher when ICU blood glucose levels were maintained closer to the normal range (which is also their own normal as well). In 2014 Plummer and colleagues further substantiated previous observations by Egi and colleagues, who observed an 18% increase in the risk of mortality for every 20 mg/dL (1.1 mmol/L) increase in maximum blood glucose in nondiabetic and diabetic patients with good preadmission glycemic control. The data from these studies further emphasize that having one glycemic goal for all critically ill patients is probably suboptimal. ^, ^, ^,

Glycemic goal in the ICU

Based on the NICE-SUGAR trial, most centers target a blood glucose level of <180 mg/dL (10 mmol/L) for critically ill patients in an attempt to treat hyperglycemia while reducing the incidence of hypoglycemia noticed with the application of tight glycemic control. The ADA also recommends a goal of 140–180 mg/dL (7.7–10 mmol/L) in ICU patients.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here