This article reviews key concepts of cerebral glucose metabolism, neurologic outcomes in clinical trials, the biology of the neurovascular unit and its involvement in secondary brain injury after traumatic brain insults, and current scientific and clinical data that demonstrate a better understanding of the biology of metabolic dysfunction in the brain, a concept now known as cerebral metabolic energy crisis. The use of neuromonitoring techniques to better understand the pathophysiology of the metabolic crisis is reviewed and a model that summarizes the triphasic view of cerebral metabolic disturbance supported by existing scientific data is outlined. The evidence is summarized and a template for future research provided.
Key points
- •
Hyperglycemia is often observed in critical illness and severe TBIs, indicating systemic physiologic stress and severity of injury.
- •
Acute hyperglycemia has been found to be associated significantly with poor functional outcome and high mortality in severe TBI.
- •
Randomized clinical trials addressing hyperglycemia so far have failed to demonstrate improvement in neurologic outcomes after severe TBI, prompting further research to understand the disease process.
- •
Recent advancements in preclinical and clinical research shift the focus to the physiology of glucose use at the neurovascular unit level in the brain, where glucose metabolism is altered.
- •
Future research will shed light into the promise of alternative energy delivery methods.
Introduction
The human brain consumes about 25% of cardiac output, reflecting the high energetic demand that brain cells depend on to function at physiologic conditions. Energy demand and use are dramatically altered following severe traumatic brain injury (TBI) creating a biologic dilemma for neuronal survival and functional preservation.
Despite significant advances in the understanding of brain physiology and evidence demonstrating that blood flow regulation and oxygen delivery optimization improve the chances of survival after TBI, the most up-to-date Brain Trauma Foundation management guidelines do not include a formal recommendation in regards to systemic glucose control or brain glucose optimization. Yet, mounting scientific experimental and clinical evidence demonstrate that systemic glucose derangements and deviation from a physiologic cerebral glucose metabolism further exert a negative impact in recovery from TBI, by exacerbating secondary tissue injury, hindering functional outcomes, and increasing the chance of mortality.
According to the Centers for Disease Control and Prevention, 2.2 million patients visit emergency rooms each year for TBI in the United States alone. Of those, about 250,000 are hospitalized and about 50,000 die as a result of their injury ( http://www.cdc.gov/traumaticbraininjury/data/ ). Improved protocols exist for management of intracranial pressure (ICP), cerebral perfusion pressure, and brain oxygenation. However, despite increased knowledge and understanding about glucose metabolism at the systemic level and in the brain, clinical trials aimed at controlling systemic hyperglycemia have failed to improve neurologic outcomes and survival after TBI. In a general critical care patient population, results have shown higher mortality with the intensive insulin therapy (IIT) strategy to control elevated serum glucose. In the neurologic population that suffers from severe TBI the results have been equally disappointing.
A thorough review of the pathophysiologic mechanisms behind cerebral metabolic failure supported by current scientific evidence and an outline toward future directions in management and research are discussed.
Introduction
The human brain consumes about 25% of cardiac output, reflecting the high energetic demand that brain cells depend on to function at physiologic conditions. Energy demand and use are dramatically altered following severe traumatic brain injury (TBI) creating a biologic dilemma for neuronal survival and functional preservation.
Despite significant advances in the understanding of brain physiology and evidence demonstrating that blood flow regulation and oxygen delivery optimization improve the chances of survival after TBI, the most up-to-date Brain Trauma Foundation management guidelines do not include a formal recommendation in regards to systemic glucose control or brain glucose optimization. Yet, mounting scientific experimental and clinical evidence demonstrate that systemic glucose derangements and deviation from a physiologic cerebral glucose metabolism further exert a negative impact in recovery from TBI, by exacerbating secondary tissue injury, hindering functional outcomes, and increasing the chance of mortality.
According to the Centers for Disease Control and Prevention, 2.2 million patients visit emergency rooms each year for TBI in the United States alone. Of those, about 250,000 are hospitalized and about 50,000 die as a result of their injury ( http://www.cdc.gov/traumaticbraininjury/data/ ). Improved protocols exist for management of intracranial pressure (ICP), cerebral perfusion pressure, and brain oxygenation. However, despite increased knowledge and understanding about glucose metabolism at the systemic level and in the brain, clinical trials aimed at controlling systemic hyperglycemia have failed to improve neurologic outcomes and survival after TBI. In a general critical care patient population, results have shown higher mortality with the intensive insulin therapy (IIT) strategy to control elevated serum glucose. In the neurologic population that suffers from severe TBI the results have been equally disappointing.
A thorough review of the pathophysiologic mechanisms behind cerebral metabolic failure supported by current scientific evidence and an outline toward future directions in management and research are discussed.
Brain glucose metabolism principles
Glycolysis is perhaps one the most preserved biologic processes from prokaryotes to mammals consisting of the biochemical steps that allow glucose use as a source of energy. It occurs in the cytoplasm and results in production of pyruvate, lactate, and ATP. Pyruvate then diffuses across cellular compartments to reach the mitochondria where it is prepared to enter the citric acid cycle (Krebs cycle) in the form of acetyl-CoA. The end result is further generation of ATP, CO 2 , and nicotinamide adenine dinucleotide. Nicotinamide adenine dinucleotide then enters the electron transport chain and through oxidative phosphorylation results in the production of large amounts of ATP. Oxygen is consumed as the electron acceptor allowing restoration of NAD + to maintain the cycle. This well-defined pathway constitutes aerobic respiration ( Fig. 1 ).
The Krebs cycle is not only crucial for generation of chemical energy, but also to provide the cell with the necessary precursor materials to synthesize some amino acids, and in the case of neurons, neurotransmitters. Not all pyruvate generated by glycolysis enters the Krebs cycle; about 15% of pyruvate is converted to lactate, which in turn can be used to generate energy. Under anaerobic conditions, the amount of pyruvate converted to lactate increases as a biologic mechanism to partially supply energy demand. Under the same conditions, glycerophosphate, an intermediate element of glycolysis, also dramatically accumulates.
Glucose is used at high rates by neurons and other brain cells. Strong evidence now shows that glycolysis in astrocytes and oxidative metabolism in neurons is increased as a function of neuronal activity. The transport of glucose from the bloodstream across the blood-brain barrier is driven by a concentration gradient and mediated actively via the endothelial glucose transporter 1. Because neurons are not able to synthesize glucose its transport mechanisms must be highly efficient to meet activity-driven demands. Shuttling from the extracellular space into astrocytes and neurons occurs through glucose transporter 1 and glucose transporter 3, respectively. Efficient transport of fuel provides an essential element for generation of ATP and precursor molecules necessary for the synthesis of neurotransmitters essential for neuronal function.
Glucose control in critical illness
Energy demands and expenditure in critical illness differ significantly from normal physiologic states. States of organ dysfunction, trauma, infection, or a combination pose a challenge for the mechanisms responsible for glucose control. Hyperglycemia observed in acute illness may be caused by insulin insufficiency, insulin resistance, impaired glucose use, stress caused by hormonal dysregulation (eg, cortisol), and increased catecholamines. Hyperglycemia is an independent predictor of adverse outcomes and increased mortality in survivors of myocardial infarction admitted to the intensive care unit (ICU) at 180 days, 1 year, and 2 years, even in the absence of preexisting diabetes. Similar observations were confirmed in patients with critical illness of all causes, who had higher mortality rates at 180 days when hyperglycemia was observed in ICU. Glucose elevation at any time throughout the ICU course has been shown to have a higher independent positive predictive value for mortality than the Acute Physiology and Chronic Health Evaluation II score. The recognition of this relationship and the quest for glucose optimization in critical illness has been an area of intense investigation over the last decades, leading to large clinical trials aimed at testing whether glucose control could lead to improved survival and better outcomes. The first clinical trial in this area aimed at maintaining serum glucose levels in the range of 80 to 100 mg/dL using IIT. Mortality during intensive care was improved by about 50% (4.6% intensive therapy vs 8.0% standard therapy) highlighting the importance of stress-induced hyperglycemia and sparking further research interest in this area. The indication for ICU stay in two-thirds of the patients included in this study was elective cardiac surgery. A large follow-up clinical trial by the same team failed to demonstrate a mortality benefit of using IIT in a medical ICU population with a more wide range of critical illness conditions. Subsequent more inclusive clinical trials and a meta-analysis failed to support this observation in more heterogeneous groups of patients. Adverse events were also more frequent in the patients with IIT.
An effort to address this important question was put forward with the NICE-SUGAR study, a large multicenter, international clinical trial in which 6104 patients were randomized to undergo tight glucose control (80–108 mg/dL) or standard therapy (goal 180 mg/dL). This study found that intensive glucose control among adults in the ICU increased mortality at 90 days. Severe hypoglycemia (glucose <40 mg/dL) was significantly more frequent in the intensive control arm (6.8% vs 0.5%).
A salient aspect from these trials was the limited number of patients with a primary neurologic diagnosis or TBI, indicating that observations in the critically ill patient still remained widely inapplicable in neurologic patients. Recent evidence indicates that functional outcomes and mortality after TBIs are affected by systemic glucose derangements and that systemic glucose levels are not accurate indicators of brain glucose delivery or use. This has prompted active preclinical and clinical research to address these questions.
Glucose metabolism in traumatic brain injury
Pathophysiology of Hyperglycemia in Traumatic Brain Injury
Similar to other critical illnesses, hyperglycemia in TBIs is a manifestation of severity of the disease and the mechanisms behind it have been under investigation for several decades. As a direct consequence of acute brain injury, an early surge in sympathetic activity leads to an increase in systemic circulating catecholamines. In experimental models of isolated brain injury, the degree of sympathoadrenal response seems to be graded in a linear relationship with severity of the brain injury. The catecholamine surge occurs within minutes of the insult and may be transient, whereas the circulating glucose surge follows soon after in a sustained steady fashion.
In addition to the observation of spontaneous early hyperglycemia after TBI, now there is mounting evidence indicating this early anomaly may further worsen secondary injury.
Preclinical research in animal models of TBI has shown that induced hyperglycemia at the time of injury increased the accumulation of neutrophils in contusional areas potentially exacerbating inflammation and abnormalities in blood flow. In this study, delayed hyperglycemia did not have the same effect, raising the question of an early time window of higher susceptibility to the negative effects from a glucose surge.
In experimental models of intracerebral hemorrhage hyperglycemia has been shown to worsen cerebral edema surrounding the hematoma and exacerbate neuronal death in these same brain regions. In models of ischemic brain injury, early hyperglycemia resulted in a dramatic exacerbation of neocortical neuronal necrosis. In humans with severe TBI hyperglycemia is associated with cerebral tissue acidosis; however, it is not clear whether there is a causal relationship.
The observation that hyperglycemia exacerbates neuronal injury and death across several types of neurologic insults suggests a possible common underlying mechanism in secondary brain injury. One plausible mechanism that has been proposed for several decades now involves disruption of the blood-brain barrier leading to dysregulation in blood flow, glucose transport and use and excitotocity. The modern concept of neurovascular unit dysfunction offers a more global and integrated physiologic explanation for the derangements observed in a wide range of neurologic insults including TBI.
Association of Hyperglycemia and Brain Injury Outcomes
Systemic hyperglycemia has been associated with worse outcomes in a wide variety of acute neurologic insults, including intracranial hemorrhage, ischemic stroke, and aneurysmal subarachnoid hemorrhage. In ischemic stroke this observation led to the design of pilot clinical trials aimed at testing safety of insulin infusions for targeted blood glucose control in the hours subsequent to the clinical event. A more recent phase 2 trial has been completed demonstrating that tight glucose control is safe in acute ischemic stroke, yet a definitive phase 3 trial is still ongoing.
The association of spontaneous systemic hyperglycemia early after TBI and poor outcomes has been reported. In a study involving 59 subjects with moderate and severe brain injuries (Glasgow Coma Scale [GCS], 3–10) serum glucose levels greater than 200 mg/dL in the first 24 hours from hospital admission were associated with worse neurologic outcome at hospital discharge, 3 months, and 1 year.
In a cohort of 169 patients with severe TBI (craniotomy for evacuation of mass occupying hematoma, ICP monitor placed), those with initial GCS less than 8 had significantly higher serum glucose than patients with GCS 12 to 15 on admission (192 vs 130 mg/dL). Furthermore, hyperglycemia was significantly worse in patients who died or remained in vegetative state when compared with those with good outcome or moderate disability (217 mg/dL vs 167 mg/dL). These studies, however, were conducted at centers where administration of dextrose solutions and steroids were routine at the time, leaving open the possibility of a biologic effect of those interventions, despite statistical correction. Subsequent studies conducted rigorously avoided that issue. In a series of 267 patients with isolated moderate and severe TBI most of whom underwent craniotomy for evacuation of mass-occupying lesion, admission serum glucose was significantly higher in those with GCS less than or equal to 8 compared with moderate injury patients with GCS 9 to 12 (203 mg/dL vs 164 mg/dL). Furthermore, systemic glucose was significantly higher in patients with unfavorable outcome as measured by glasgow outcome scale (GOS) (GOS 3,2,1 = 204 mg/dL vs GOS 4, 5 = 179 mg/dL). In a study involving 77 patients with severe TBI (admission GCS, ≤8), hyperglycemia within the first 5 ICU days, defined as two or more episodes of serum glucose greater than 170 mg/dL, was independently associated to increased hospital mortality at 21 days (survival rate 51% in patients with hyperglycemia vs 83% in patients without hyperglycemia).
In a more recent study describing 380 patients admitted with TBI, peak glucose levels in the first 24 hours of hospital admission was an independent predictor of in-hospital mortality (cutoff glucose value for increased mortality, 160 mg/dL). Across brain injury severity groups, peak-glucose within 24 hours was significantly higher in nonsurvivors. Similar observations have been made in the pediatric and adolescent population.
Clinical Trials of Glucose Control in Traumatic Brain Injury
The observed association between hyperglycemia and worsened neurologic outcomes has led to clinical trials testing the neurologic effect of using of IIT in the neurologic ICU. In a clinical trial patients with TBIs were randomized to glucose goal less than 200 mg/dL or IIT (80–120 mg/dL). In this single center trial, the rate of hypoglycemic episodes per patient was nearly as twice as much on the IIT group; however, mortality rates were similar at hospital discharge and 6 months. In a before/after study, designed to evaluate the effects of IIT implementation as part of routine practice (serum glucose goal, 80–120 mg/dL), 1957 patients in the “before” arm were compared with 1888 patients in the targeted “implementation arm.” Use of the IIT protocol was associated to higher rates of hypoglycemia events and higher rate of mortality, specifically related to hypoglycemic events.
In the first prospective clinical trial aimed at addressing the question of glucose control in a neurologically ill population, patients admitted to the neurologic ICU at a single center were randomized to IIT with serum glucose goal 80 to 110 mg/dL or conventional therapy with serum glucose goal less than 151 mg/dL. No benefit in mortality or functional outcome at 3 months was observed. Instead a trend toward increased mortality in the IIT group was found.
Cerebral Metabolic Energy Crisis: A Window Toward Goal-Targeted Therapies
The failure of clinical trials aimed at impacting outcomes by controlling hyperglycemia highlights the need to advance research efforts to elucidate the fate of glucose in the injured brain with focus on the neurovascular unit. Traditionally, the cornerstone in physiologic management of patients with severe TBI has been to focus on optimization of key physiologic brain variables: ICP, cerebral perfusion pressure, and brain oxygenation. There is no evidence-based clinical guideline, at present, as to how glucose or other sources of energy should be delivered to the brain to sustain metabolic demands after TBI.
The neurovascular unit is the functional building block in the brain that ensures a proper match between energy supply and demand ( Fig. 2 ). The cascade of pathophysiologic changes that occur as the result of traumatic injury to the brain includes cytotoxic edema, macrovascular dysfunction, microvascular dysfunction, and cellular energetic failure. These alterations, indicative of neurovascular unit dysfunction, are manifested at the organ level as increased ICP, cerebral vasospasm, cerebral blood flow autoregulatory failure, hypoxemia, cerebral tissue ischemia, and cerebral glucose metabolic dysfunction. These physiologic derangements converge in a common pathway that leads to energetic failure, ultimately explained by either suboptimal delivery of oxygen and nutrients and/or cellular dissociation from normal glycolysis and oxidative metabolism.
