Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Treatment of traumatic brain injury with moderate hypothermia

Marion DW, Penrod LE, Kelsey SF, Obrist WD, Kochanek PM, Palmer AM, Wisniewski SR, DeKosky ST.

N Engl J Med 1997; 336: 540-546. [abstract]

Reviewed by Barry Markovitz, MD

Review posted May 15, 1997

---

I. What is being studied?:

The study objective:

To assess the effects of induced hypothermia on post-traumatic cerebral physiology, inflammatory mediators found in the CSF, and functional outcome.

The study design:

Randomized, prospective, controlled trial.

The patients included:

82 patients with closed head injury, Glascow Coma Scores (GCS) of 3 to 7, aged 16 to 75 years, admitted within 6 hours of injury, who were unable to follow commands.

The patients excluded:

Patients were excluded with clinical brain death, prolonged hypoxia or hypotension, a gunshot wound, pregnancy, an undetermined time of injury, or normal CT scan findings.

The interventions compared:

Intentional moderate hypothermia (induced by cooling blankets and iced saline gastic lavage) to 32 to 33 degrees C (rectal temperature) for 24 hours following enrollment versus maintenance of normothermia.

The outcomes evaluated:

Glascow Outcome Score (GOS) at 3, 6, and 12 months following injury. Poor outcome, or a GOS of 1, 2, or 3 (death, vegetative state, or severe disability) was compared with good outcome, or a GOS of 4 or 5 (moderate to mild or no disability).

Secondary measures included physiologic monitoring variables, mortality, length of hospital stay, rates of complications (e.g., delayed intracranial hematomas, deep venous thrombosis, other organ dysfunction), and CSF neurotransmitter and mediator levels.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. In addition, a block randomization scheme was employed to attempt equal assignment of patients in the treated and control groups with both low GCS (3-4) and higher GCS (5-7).

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

Virtually. Only one patient (hypothermia group) was lost to follow-up at 12 months.

Were patients analyzed in the groups to which they were randomized?

Yes. There was no cross-over of patients between groups.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

The patients were comatose (making their knowledge of group assignment unlikely to cause a bias). The managing physicians were apparently not blinded to treatment, but the physical medicine and rehabilitation specialists who evaluated the patients' functional status in follow-up were unaware of patient group assignment.

4. Were the groups similar at the start of the trial?

No. Age, cause of injury, presence of pupillary abnormality, and associated injuries were evaluated and found not to differ between the two groups. However, the normothermia patients tended to have worse initial Glasgow Coma Scores and CT scan findings.

5. Aside from the experimental intervention, were the groups treated equally?

Apparently. Patients were treated according to the "Guidelines for the Management of Severe Head Injury" (1) and included intraventricular catheter measurement and treatment of intracranial pressure elevations. Goals for cerebral perfusion pressure and intracranial pressure were set and maintained with (in order of escalating therapy indicated): CSF drainage, mannitol, repeat CT scan (to evaluate for new mass lesions), pentobarbital, and dopamine. Patients in both groups received neuromuscular blockade and fentanyl infusions.

III. What were the results?

1. How large was the treatment effect?

Hypothermia treated patients had significantly better outcomes at 3, 6, and 12 months after injury than normothermia patients. At 12 months, 24 (62%) of hypothermia vs. 16 (38%) of normothermia patients had good (GOS 4 or 5) outcomes (p= 0.05). Subgroup analysis revealed that only patients with GCS of 5-7 (73% of patients) benefited from hypothermia.

However, these results were confounded by CT class and initial GCS. After adjustment for these factors (using logistic regression) there was a strong, but nonsignificant, trend towards benefit for hypothermia across all patients at 3 months (RR 0.3; 95% CI 0.1-1.0), 6 months (RR 0.5; 95% CI 0.2-1.2), and 12 months (RR 0.5; 95% CI 0.2-1.2). In patients with initial GCS of 5-7, the adjusted analysis showed significant benefit for hypothermia at 3 and 6 months (RR 0.2; 95% CI 0.1-0.9) and a strong trend at 12 months (RR 0.3; 95% CI 0.1-1.0).

In the first 36 hours of treatment (during which time the hypothermia group was cooled and then allowed to slowly rewarm), treated patients had lower intracranial pressure (15.4 vs. 19.7, p=0.01) and higher cerebral perfusion pressure (82.4 vs. 77.4, p=0.05). Between 37 and 60 hours of treatment, hypothermia patients had lower cerebral perfusion pressure (73.5 vs. 84.1, p<0.001) due to a lower mean arterial pressure (92.6 vs. 101.4, p<0.001).

In patients with an initial GCS of 5-7, CSF glutamate and interleukin-1ß concentrations were significantly lower in the hypothermia patients.

2. How precise was the estimate of the treatment effect?

In those patients with a higher initial GCS, at 6 months post-injury we can be 95% certain that the chance of a bad outcome was reduced with hypothermia treatment by 10 to 90% (after accounting for initial CT class). At 12 months, however, the risk reduction ranges from 90% to no difference in outcome.

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

Uncertain. This investigation was limited to adult patients, and the results could reasonably be extrapolated to teenagers. However, anatomic and metabolic differences in the infant and young child, as well as the likelihood of a different incidence of mass lesions (2) in pediatric patients, would argue against unreserved acceptance of the applicability of these data to very young patients.

It is noteworthy as well that hypothermia did not seem to be beneficial to patients with an initial GCS of 3-4. However, these patients only comprised 27% of the sample; a larger sample size of such patients might prove hypothermia of benefit to them as well.

2. Were all clinically important outcomes considered?

Yes. Functional status after a sufficient time period is the most important clinically relevant outcome.

3. Are the likely treatment benefits worth the potential harms and costs?

Yes (for adults and teenagers). No difference in complication rates or duration of hospital stay was detectable between the groups. Using the unadjusted data for patients with initial GCS of 5-7, only 2.9 (± 0.3) patients would need to be treated with hypothermia to prevent a negative outcome. This is a very favorable "number needed to treat." [Read more about NNT: Number Needed to Treat from NHS Centre for EBM.]

References

1. Bullock R, Chesnut RM, Clifton G, et al. Guidelines for the management of severe head injury. J Neurotrauma 1996; 13:639-734. [no abstract available]

2. Bruce DA, Raphaely, RC, Goldberg AI, et al. Pathophysiology, treatment, and outcome following severe head injury in children. Child's Brain 1979; 5:174-191. [abstract]

Comments

Back to the EB Journal Club Index