The Brain Trauma Foundation has an extensive set of guidelines for the management of traumatic brain injury. One of these involves the routine administration of anti-seizure prophylaxis. The risk of seizure after TBI is higher than in the general population, and a high risk group has been identified (GCS<10, depressed skull fracture, cortical contusion, subdural / epidural / intracerebral hematoma, penetrating injury, or seizure within 24 hours.
This guideline showed that there was no data to support a recommendation with the highest degree of evidence, but it did suggest that anti-seizure meds decreased the incidence of early seizures (within 7 days), but these seizures had no effect on outcome. This meant that you might consider giving phenytoin, but there was always the risk associated with giving any new medication, and no real benefit to the outcome. This is one reason why the guideline is applied inconsistently by trauma professionals.
Researchers at the University of Florida – Jacksonville will be presenting a paper at AAST 2013 further questioning the advisability of this guideline. They retrospectively reviewed 93 patients who were admitted to their hospital for at least 7 days with severe blunt TBI. They split the patients into two groups, a control group (43 patients) who did not receive phenytoin and a study group (50 patients) who had therapeutic levels.
Here are the factoids:
- More seizures occurred in the phenytoin group (4% vs 0%)
- Mortality was the same (8% vs 7%)
- Discharge disposition (home vs rehab) was the same
- Hospital length of stay was longer in the prophylaxis patients (36 vs 25 days)
- Glasgow Outcome Scale was worse in the phenytoin group (2.9 vs 3.4)
Bottom line: Don’t pay much attention to this study. It’s very small, making the quality of the results pretty shaky. There is no comment on matching other injuries that might account for a difference in length of stay. And is this the proper way to find a difference in Glasgow Outcome Scale, an integer value? What does 2.9 or 3.4 mean? Nothing. It has to be a 3 or a 4, no decimal point.
Although the data supporting the original Brain Trauma Foundation guideline is a bit weak, this paper does not support changing current practice. Needs more work (and more numbers).
Related post:
Reference: More Harm Than Good: Anti-seizure prophylaxis after traumatic brain injury does not decrease seizure rates but may inhibit functional recovery. AAST 2013 Paper 10.
