new horizons Copyright © 1995 by the Society of Critical Care Medicine

Steroids in Head Injury

Daniel F. Kelly, MD

Failure of high-dose corticosteroid therapy to ameliorate intracranial hypertension or im­prove long-term neurologic outcome in patients with traumatic brain injury has been shown in several prospective, randomized clinical tri­als. Additionally, the risk of complications, in­cluding elevations in serum glucose levels and gastrointestinal hemorrhage, although rela­tively low, make routine use of glucocorticoids further unwarranted in head-injured patients. {New Horizons 1995; 3:453-455)

key words: corticosteroid; cerebral edema;

gastrointestinal hemorrhage; head injury; in­tracranial pressure; methylprednisolone; spi­nal cord injury; steroid; traumatic brain injury

Since the 1950s, synthetic corticosteroids have been an important component of the pharmacologic armamentarium for treating certain neurologic dis­orders. High-dose glucocorticoid therapy is effective in reducing vasogenic edema associated with pri­mary and metastatic brain tumors, brain abscesses, and spinal cord tumors (1). Corticosteroid adminis­tration also ameliorates the symptoms associated with the acute inflammatory response of multiple sclerosis (2). More recently, megadose methylpred­nisolone has been shown to modestly improve recov­ery of function following acute traumatic spinal cord injury if treatment is begun within 8 hrs of injury (3). Use of steroids in managing patients with trau­matic brain injury was popular in the 1970s and

From the Division ofNeurosurgery, University of California at Los Angeles School of Medicine, Los Angeles, CA.

Address requests for reprints to: Dr. Daniel Kelly, UCLA Center for Health Sciences, Division of Neurosurgery, 74-140 CHS, 10833 Leconte Ave, Los Angeles, CA 90025. 1063-7389/95/0303-453$03.00/0

1980s, and is still used by some practitioners. Sev­eral prospective, randomized trials in patients with severe head injury (SHI), however, have failed to demonstrate the efficacy of high-dose steroids in reducing intracranial pressure (ICP) or in improv­ing long-term neurologic recovery. This article pro­vides a brief review of corticosteroid therapy in head injury.


In a nonrandomized trial of head-injured patients, Gobiet et al. (4) compared a no-steroid group of 35 patients to two groups who received either low-dose (24 patients) or high-dose (34 patients) dexa-methasone; the three groups were treated over 3 subsequent years, respectively. The high-dose group received a 48-mg bolus followed by a total of 96 mg on day 1, 48 mg on day 2, 96 mg on day 3, 48 mg on day 4, then 24 mg on days 5 through 8. It was reported that the high-dose group had a signifi­cantly lower mortality (23%) than the low-dose dexamethasone group (41.5%) or the no-steroid group (45.5%). However, a re-analysis of these figures failed to demonstrate statistical significance among the different treatment groups (5).

In five subsequent prospective, randomized trials published from 1979 to 1986 (5-9) totaling 555 SHI patients, no improvement in 6-month outcome, as defined by the Glasgow Outcome Scale, was demon­strated in the steroid-treated groups compared to those not receiving steroids (Table 1). All were placebo-controlled, double-blind studies except the one by Saul et al (9). In three of these studies (5-7), dexamethasone was used, and in two reports (8, 9), methylprednisolone was administered. The three reports which assessed control of intracranial hy­pertension (5, 7, 9) demonstrated no beneficial effect from steroid use. In fact, the study by Dearden et al. (5) suggested that glucocorticoids may be contra-indicated in head-injured patients with raised ICP.

A study by Gudeman et al. (10) looked specifically at the effect of high-dose methylprednisolone in con­trolling ICP in 20 consecutive SHI patients. During the first 12 hrs after admission, patients received 40

mg at 6-hr intervals, followed by a 2000-mg bolus, then 500 mg every 6 hrs for the next 24 hrs. The dose was then rapidly tapered over the next 48 hrs. Methylprednisolone failed to reduce ICP in these patients, nor was there evidence that the incidence of pressure waves or the peak ICP was reduced. Steroid administration also had no effect on the volume-pressure response, which is a measure of brain elastance. Compared to a historical control group of 262 patients, the course of ICP and long-term outcome were not improved by high-dose me-thylprednisolone.


Known complications of high-dose glucocorticoid use include gastrointestinal hemorrhage, poor wound healing, elevations in serum glucose levels, and an increased risk of infection. The incidence of major complications in head-injured patients attributable to steroid use, however, appears to be relatively low.

In three reports (5, 7, 8), major gastrointestinal bleeding occurred in 1 of 216 steroid-treated pa­tients for an incidence <0.5%. Gudeman et al. (10)

Table 1. Outcome with high-dose steroid therapy in severely head-injured patients

No. of Author Patients

Steroid Regimen

Good Outcome (%)

Cooper et al. 76


Low-dose Dex (16 mg/day x 6) 44

High-dose Dex (96 mg/day Placebo

x 6) 29 37

Saul et al.» 100 (1981)

MP (5 mg/kg/day) No steroid

52 62

Braakman 161

etal. (1983)

Dex (100 mg/day x 4 days) Placebo

30 32

Giannotta 88 etal. (1984)"

Low-dose MP (260 mg x 1 High-dose MP (4700 mg x Placebo

day) 18 1 day) 26 31

Dearden 130

et al. (1986)

Dex (100 mg/day x 3) Placebo

44 56


Good outcome includes patients with good recovery or moderate disability as defined by the Glasgow Outcome Scale (15). In all regimens, a steroid taper which lasted from 3 to 8 days followed the initial dosing.

Dex, dexamethasone; MP, methylprednisolone. °In the series by Saul et al., good outcome also included severe disability; the duration of steroid treatment and the type of taper were not stated in this report; 'the initial dosing regimens in this report are estimates based on a 70-kg adult.

reported a 50% incidence of gastric hemorrhage in their series of 20 patients. Given that erosive gas­trointestinal lesions may occur in over 90% of SHI patients within 24 hrs of injury, it is unlikely that this high rate of gastrointestinal bleeding is solely due to steroids (11). The incidence of intracranial or extracranial infections also has not been shown to be increased with the use of high-dose steroids in head-injured patients (5, 7, 8). The study by Dearden et al. (5) did demonstrate significantly higher serum glucose levels in patients receiving glucocorticoids compared to the placebo group. Gudeman et al. (10) also documented hyperglycemia with glycosuria in 85% of patients receiving high-dose methylpredniso­lone. Cooper et al. (7), however, noted no difference in insulin requirements in steroid-treated patients compared to those given placebo.

In a study by Marshall et al. (12) of 121 neurosurgical patients receiving high-dose dexa­methasone with antacids, three major complications (2.4%) occurred that were directly attributable to steroid use, including a duodenal perforation result­ing in death, a gastric hemorrhage requiring trans­fusion, and a Candida esophagitis which resolved. Two of these patients had previous gastrointestinal disease. In these three patients, the daily dose of dexamethasone did not exceed a dose of 40 mg/day. Five patients (4%) also developed hyperglycemia. The complications attributable to steroids in this study did not appear to be dose-related; in 26 pa­tients with brain tumors treated with a daily steroid dose of 80 mg/day, no steroid-related complications were reported. Overall, this study concluded that high-dose steroids were relatively safe in neuro­surgical patients, but that caution was necessary in patients with pre-existing gastrointestinal disease. The number of patients in the study with head inju­ries was not reported.


Five randomized, prospective clinical trials con-vincingly demonstrate that high-dose glucocorticoid therapy is ineffective in altering the course of intrac­ranial hypertension or in improving long-term neu-rologic outcome after SHI. Even though steroids have been proven to be relatively safe in neuro­surgical patients, the multitude of potential serious side effects further warrant against their routine use in the head-injured patient. It remains rather puzzling that a statistically significant, albeit small, therapeutic effect was seen with megadose methyl­prednisolone therapy in the Second National Acute

Spinal Cord Injury Study, but no benefit has been demonstrated in head-injured patients (3). The im­proved outcome in the spinal cord injury patients has been attributed to the antioxidant, free-radical scavenging effect of methylprednisolone which is seen at very high doses (13). It is hoped that the antioxidant non-glucocorticoid 21-aminosteroids cur­rently being evaluated in clinical trials of moder­ately and severely head-injured patients will prove efficacious in improving outcome in this patient popu­lation (13,14).


1. Galicich JH, French LA: Use of dexamethasone in the treat­ment of cerebral edema resulting from brain tumors and brain surgery. Am Pract 1961; 12:169-174

2. Griffiths TD, Newman PK: Steroids in multiple sclerosis. J Clin Pharm Ther 1994; 19:219-222

3. Bracken MB, Shepard MJ, Collins WF, et al: A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury. Results of the Second National Acute Spinal Cord Injury Study. N Engi J Med 1990; 322:1405-1411

4. Gobiet W, Bock WJ, Liesegang J, et al: Treatment of acute cerebral edema with high dose of dexamethasone. In: Intrac-ranial Pressure III. Beks JWF, Bosch DA, Brock M (Eds). Berlin, Springer-Verlag, 1976, pp 231-235

5. Dearden NM, Gibson JS, McDowall DG, et al: Effect of high-dose dexamethasone on outcome from severe head injury. J Neurosurg 1986; 64:81-88

6. Braakman R, Schouten HJA, van Dishoeck MB, et al:

Megadose steroids in severe head injury: Results of a pro­spective double-blind clinical trial. J Neurosurg 1983;


7. Cooper PR, Moody S, Clark WK, et al: Dexamethasone and severe head injury. A prospective double-blind study. J Neurosurg 1979; 51:307-316

8. Giannotta SL, Weiss MH, Apuzzo MLJ, et al: High-dose glucocorticoids in the management of severe head injury. Neurosurgery 1984; 15:497-501

9. Saul TG, Ducker TB, Salcman M, et al: Steroids in severe head injury. A prospective randomized clinical trial. J Neurosurg 1981; 54:596-600

10. Gudeman SK, Miller JD, Becker DP: Failure of high-dose steroid therapy to influence intracranial pressure in pa­tients with severe head injury. J Neurosurg 1979; 51:301—306

11. Brown TH, Davidson PF, Larson GM: Acute gastritis occur­ring within 24 hours of severe head injury. Gastrointest Endosc 1989; 35:37-40

12. Marshall LF, King J, Langfitt TW: The complications of high-dose corticosteroid therapy in neurosurgical patients:

A prospective study. Ann Neural 1977; 1:201-203

13. Hall ED: The neuroprotective pharmacology of methylpred­nisolone. J Neurosurg 1992; 76:13-22

14. Hall ED, Yonkers PA, McCall JM, et al: Effects of the 21-aminosteroid U74006F on experimental head injury in mice. J Neurosurg 1988; 68:456-461

15. Jennett B, Bond M: Assessment of outcome after severe brain damage. A practical scale. Lancet 1975; i:480-484