Calorie and protein supplementation improves nutritional status. This support may improve outcome and decrease morbidity and mortality in acutely brain-injured patients. Investigators have observed a poor tolerance to enteral feedings after brain injury and have noted that this persists for approximately 14 days postinjury. This delay has been attributed to increased gastric residuals, prolonged paralytic ileus, abdominal distention, aspiration pneumonitis, and diarrhea. In the present investigation, 23 brain-injured patients with an admission 24-hour peak Glasgow Coma Scale (GCS) score between 4 and 10 were studied for 18 days from hospital admission. The mean duration from injury to initiation of full-strength, full-rate enteral feeding was 11.5 days. Seven of the 23 patients tolerated enteral feedings within the first 7 days following hospital admission (mean 4.3 days), four patients tolerated feedings between 7 and 10 days postadmission (mean 9 days), and 12 patients did not tolerate feedings until after 10 days postinjury (mean 15.9 days). There was a marginally significant relationship between low GCS scores on admission and length of days to enteral feeding tolerance (p = 0.07). A significant inverse relationship was observed between daily peak intracranial pressure (ICP) and time to tolerance of feedings (p = 0.02). There was no significant relationship between feeding tolerance and days to return of bowel sounds (p = 0.12). Serum albumin levels decreased during the investigation (mean +/- standard error to the mean: 3.2 +/- 0.12 gm/dl on Day 1; 2.7 +/- 0.23 gm/dl on Day 16; normal = 3.5 to 5.0 gm/dl), whereas the percentage of patients tolerating feedings increased over the course of the study. The authors conclude that patients with acute severe brain injury do not adequately tolerate feedings via the enteral route in the early postinjury period. Tolerance of enteral feeding is inversely related to increased ICP and severity of brain injury. It is suggested that parenteral nutritional support is required following brain injury until enteral nutrition can be tolerated.

The acute response to injury and infection is manifested by increased synthesis of acute-phase proteins by the liver, an increased white blood cell count, fever, a negative nitrogen balance, and altered serum mineral levels (zinc, iron, and copper). This response is thought to be partially mediated by cytokines such as interleukin-1, but has not been well studied in head-injured patients. In this study, 25 patients were studied for evidence of the acute-phase response extending from hospital admission up to 21 days postinjury. The patients were divided into two groups to determine if severity of injury influenced the response. Group 1 consisted of nine patients with admission peak 24-hour Glasgow Coma Scale (GCS) scores of 4 or less; Group 2 consisted of 16 patients with admission peak 24-hour GCS scores of 8 or greater. All patients demonstrated some evidence of the acute-phase response. Serum alpha-1 acid glycoprotein, ceruloplasmin, and C-reactive protein levels were elevated on admission and throughout the study. Serum albumin and zinc levels were depressed on admission; zinc levels gradually normalized by Day 21 in both groups, but hypoalbuminemia was observed throughout the study period. Serum copper levels were normal on admission but increased to above normal in both groups by Day 11 postinjury. Urinary urea nitrogen excretion was elevated in both groups and peaked on Day 7 for Group 1 and Day 11 for Group 2 patients. The patients with admission GCS scores equal to or less than 4 had overall higher temperatures than were seen in those with GCS scores greater than or equal to 8 (p = 0.009). All patients but one had an elevated white blood cell count on admission. It is concluded that brain-injured patients with admission GCS scores of 3 to 4 and 8 to 14 demonstrate an acute-phase response which lasts for at least 3 weeks postinjury. It is speculated that this response is at least partially mediated by increased intraventricular interleukin-1 activity.

Twenty severely brain-injured patients with Glasgow Coma Scale scores of 4-9 were prospectively randomized to receive one of two standard amino acid formulas, starting with the first day of hospital admission up to day 14 postinjury. Formula 2 (patient group 2) had 54 percent more leucine, 53 percent more isoleucine, 74 percent more valine, 28 percent less phenylalanine, 31 percent less methionine, 111 percent more proline, 38 percent less alanine, and 38 percent less glycine than formula 1 (patient group 1). Groups 1 and 2 received statistically equal overall mean parenteral nutrition calories and protein (2173 +/- 147 vs. 2059 +/- 143 kcal, and 77 +/- 12 vs. 83.1 +/- 6 g, respectively). There was a significant difference in overall mean urinary urea nitrogen excretion (group 1 = 24.6 +/- 1.3 vs. group 2 = 18.3 +/- 1.1, p = 0.02) and nitrogen balance (group 1 = -8.0 +/- 2.1 vs. group 2 = +1.8 +/- 1.2, p = 0.01). Mean overall isoleucine values were significantly higher in group 2 (overall mean 77 mumol/L vs. 62 mumol/L, p = 0.04). Phenylalanine levels were significantly higher in group 1 (107 mumol/L) versus group 2 (82 mumol/L) patients (p = 0.01). Arginine levels were significantly higher in group 1 (78 mumol/L) versus group 2 (49 mumol/L) patients (p = 0.0002). This observation suggests that some standard intravenous amino acid formulas may be more apt to promote positive nitrogen balance than others.