Patients with head injury must overcome central as well as peripheral metabolic insults. In addition to specific tissue damage to the brain, a cellular biochemical cascade occurs that can negatively affect organ function, cause a systemic response to injury, and may cause secondary tissue injury. The metabolites involved in this cascade are numerous and complex. Cytokines are important cell-to-cell communication mediators during injury. It is speculated that cytokines, such as interleukin 1 (IL-1), interleukin 6 (IL-6), tumor necrosis factor (TNF), and interleukin 8 (IL-8), which are found in elevated amounts in both human and basic trials after head injury, play a role in the cellular cascade of injury. Some of the metabolic events produced by small doses of cytokine infusion in animals, as well as humans, include fever, neutrophilia, muscle breakdown, altered amino acid metabolism, depression of serum zinc levels, production of hepatic acute phase reactants, increased endothelial permeability, and expression of endothelial adhesion molecules. These are all known sequelae of severe head injury. Cytokines have also been implicated in organ failure. Infusion of cytokines in basic science trials revealed that organ functions of the gut, liver, and lung are negatively altered by high-dose cytokine infusion. Infusion of certain cytokines has been shown to cause death of brain cells, increase blood-brain barrier permeability, and cause cerebral edema. This suggests that cytokines may also play a role in the sequelae of organ demise. These effects of cytokines have been attenuated in basic trials by blocking the initial signaling system of cytokines or by decreasing serum cytokine activity. We hypothesize that cytokines that are elevated after head injury play a role in the pathology of injury, including altered metabolism and organ demise.

Sixty-eight patients were entered into a randomized, prospective, double-blinded controlled trial of supplemental zinc versus standard zinc therapy to study the effects of zinc supplementation on neurologic recovery and nutritional/metabolic status after severe closed head injury. One month after injury, the mortality rates in the standard zinc group and the zinc-supplemented group were 26 and 12%, respectively. Glasgow Coma Scale (GCS) scores of the zinc-supplemented group exceeded the adjusted mean GCS score of the standard group at day 28 (p = 0.03). Mean motor GCS score levels of the zinc-supplemented group were significantly higher on days 15 and 21 than those of the control group (p = 0.005, p = 0.02). This trend continued on day 28 of the study (p = 0.09). The groups did not differ in serum zinc concentration, weight, energy expenditure, or total urinary nitrogen excretion after hospital admission. Mean 24-h urine zinc levels were significantly higher in the zinc-supplemented group at days 2 (p = 0.0001) and 10 (p = 0.01) after injury. Mean serum prealbumin concentrations were significantly higher in the zinc-supplemented group (p = 0.003) at 3 weeks after injury. A similar pattern was found for mean serum retinol binding protein level (p = 0.01). A significantly larger number of patients in the standard zinc group had craniotomies for evacuation of hematoma; thus a bias may have been present. The results of this study indicate that zinc supplementation during the immediate postinjury period is associated with improved rate of neurologic recovery and visceral protein concentrations for patients with severe closed head injury.

Fifty-one brain-injured patients with peak 24-hour admission Glasgow Coma Scale (GCS) scores of 4 to 10 were prospectively randomly assigned to receive total parenteral (TPN) or enteral (EN) nutrition. Patients were studied from hospital admission to 18 days postinjury. Outcome was assessed by the Glasgow Outcome Scale at 3 months, 6 months, and 1 year postinjury. The TPN group received a significantly higher cumulative mean intake of protein than the EN group (mean +/- standard error of the mean: 1.35 +/- 0.12 vs. 0.91 +/- 0.9 gm/kg/day; p = 0.004). Mean cumulative caloric balance was also significantly higher in the TPN than in the EN group (75.6% +/- 5.13% vs. 59% +/- 4.26%; p = 0.02). Nitrogen balance was significantly more negative in the EN group during the 1st week postinjury (p = 0.002). The incidence of pneumonia, urinary tract infections, septic shock, and infections was not significantly different between groups. Classic nutritional assessment parameters such as anergy screens, total lymphocyte counts, and albumin levels were not significantly different between groups. The 11 patients in the EN group who did not tolerate tube feedings for 1 week postinjury had a significantly higher incidence of septic shock (p = 0.008). The change over time in GCS scores between groups was significantly different, with the TPN group showing a mean four-point increase in GCS score compared with a three-point increase in the EN group (p = 0.02). At 3 months the TPN group had a significantly higher percentage of favorable outcomes (43.5% vs. 17.9%, respectively; p = 0.05). At 6 months, 43.5% of the TPN group had a favorable outcome while 32.1% of the EN group had a favorable outcome (p = 0.29). By 1 year, 47.8% of the TPN group and 32.1% of the EN group had a favorable outcome (p = 0.20). In conclusion, more calories and protein usually can be administered to acute brain injury patients via the TPN route than by EN feedings via nasogastric or nasoduodenal routes. Traditional parameters for nutritional assessment are not useful in studying the efficacy of nutritional support during the first 2 weeks after head injury. Neurological recovery from head injury occurs more rapidly in patients with better early nutritional support.

Severe head injury is associated with a stress response that includes hyperglycemia, which has been shown to worsen outcome before or during cerebral ischemia. To better define the relationship between human head injury and hyperglycemia, glucose levels were followed in 59 consecutive brain-injured patients from hospital admission up to 18 days after injury. The patients who had the highest peak admission 24-hour serum glucose levels had the worse 18-day neurologic outcome (p = 0.01). Patients with peak 24-hour admission glucose levels greater than 200 mg/dL had a two-unit increase in Glasgow Coma Scale score while patients with admission peak 24-hour serum glucose levels less than or equal to 200 mg/dL had a four-unit increase in Glasgow Coma Scale score during the 18-day study period (p = 0.04). There was a significant relationship between 3-month and 1-year outcome and peak admission 24-hour serum glucose level (p = 0.02 and p = 0.02, respectively). Those patients with admission peak 24-hour serum glucose levels less than or equal to 200 mg/dL had a greater percentage of favorable outcome at 18 days, 3 months, and 1 year than those with admission peak 24-hour glucose levels greater than 200 mg/dL (p = 0.0007, p = 0.03, and p = 0.005, respectively). A significant relationship between admission peak 24-hour Glasgow Coma Scale score and 18-day, 3-month, and 1-year outcomes was found (p = 0.0001, p = 0.0002, and p = 0.0002, respectively). Patients with mean admission peak 24-hour Glasgow Coma Scale scores of 3.5, 6, and 10 had mean admission 24-hour peak serum glucose levels of 252 +/- 23.5, 219.1 +/- 19, and 185.8 +/- 21, respectively (p = 0.05). These relationships were not significantly altered when confounding variables such as the amount of glucose given over the initial 24-hour postinjury period, the presence of diabetes or multiple injuries, and whether patients were given steroids, dilantin, or insulin were statistically incorporated. These data suggest that admission hyperglycemia is a frequent component of the stress response to head injury, a significant indicator of severity of injury, and a significant predictor of outcome from head injury.

Most patients with moderate to severe head injury initially do not tolerate enteral feedings postinjury. This intolerance is more prolonged than that found in patients suffering other types of trauma. The authors prospectively evaluated 12 patients with moderate to severe head injury (Glasgow Coma Scale score between 4 and 10) throughout their hospitalization for liquid gastric emptying as a possible mechanism for intolerance to enteral feeding. During Week 1, the majority of patients displayed a delay in gastric emptying. Patients also displayed an abnormal biphasic response (gastric emptying faster than normal during the early stage but prolonged later). By Week 2, many patients still had delayed and abnormal biphasic responses to gastric emptying. By Week 3, an improvement was observed with the majority of patients exhibiting rapid gastric emptying, but delays and abnormal biphasic responses were still seen. Patients who initially had rapid or normal gastric emptying tolerated full-strength full-rate feedings significantly earlier compared with those who experienced delayed gastric emptying (8.5 +/- 0.5 days vs. 13.7 +/- 3.2 days, p less than 0.001). All patients tolerated full-strength full-rate feedings by Day 16 postinjury (range 7 to 16 days) except the two patients who displayed delayed gastric emptying for prolonged periods of time (mean 25 days). This is the first study to longitudinally evaluate gastric emptying following head injury. The authors suggest that patients with moderate to severe head injury often experience alterations in gastric emptying which may affect their ability to tolerate enteral feedings.

Favorable outcome equals good recovery and moderate disability. Unfavorable outcome equals severe disability, vegetative state, and death.

Severe head trauma patients (HT) exhibit markedly elevated energy expenditure and 24-hr urinary urea nitrogen excretion (UUN) values. The objective of this study was to compare seven spinal cord injured patients (SCI) to seven HT for changes in UUN and measured energy expenditure (MEE) over the first 18 days following injury. Energy expenditure was measured by indirect calorimetry and compared to values predicted by the Harris Benedict Equation (PEE). There were six quadriplegics and one paraplegic in the SCI group. HT patients had peak Glasgow Coma Scale scores of 3 to 10 for the first 24 hr postinjury. Patients were studied prospectively and matched for age, sex, and admitting weight Week 1 following the injury, SCI had mean UUN values of 0.18 +/- 0.04 g/kg/day vs 0.18 +/- 0.01 for HT patients. The mean MEE/PEE ratio was 0.56 for the SCI and 1.4 for HT (p less than 0.01). Over the entire study period the mean UUN value for SCI was 0.23 +/- 0.03 g/kg vs 0.21 +/- 0.01 for HT. The mean MEE/PEE ratio for SCI was 0.94 while HT remained elevated at 1.5 (p less than 0.05). Although the UUN was comparable in SCI vs HT, there was a significant difference in MEE/PEE between the groups. The elevation in UUN observed in SCI is not due to a hypermetabolic state. This suggests that different mechanisms promote the increased nitrogen excretion observed in these two populations.

A prospective longitudinal evaluation of serum zinc concentrations was performed in 26 head-trauma patients, and 24-hour urine zinc excretion was determined in 15 of these subjects. Patients had markedly depressed admission serum zinc concentrations (mean +/- standard error of the mean: 40.2 +/- 3.2 micrograms/dl; normal values: 70 to 120 micrograms/dl), which gradually increased during the 16-day study period. All subjects demonstrated increased urinary zinc losses throughout the study period. Urinary zinc excretion was greater in patients with more severe head injuries. Indeed, patients with more severe head trauma had mean peak urinary zinc losses of greater than 7000 micrograms/day (normal less than 500 (micrograms/day). The implications of this altered zinc metabolism for protein metabolism, wound healing, and immune function, and the specific role of zinc in brain function and recovery from injury are discussed.