Phenytoin versus Placebo

Dikmen et al. (1991)

USA

RCT

PEDro=6

N_initial=244, N_Final=124

Population: Head Injury. Phenytoin Group (n=104): Mean Age=30.9 yr; Gender: Male=82, Female=22; Median GCS=11.  Placebo Group (n=101): Mean Age=32.9 yr; Gender: Male=70, Female=31; Median GCS=9.

Treatment: Patients were randomized to receive phenytoin (prophylactic medications) or a placebo for 1 yr. Patients then observed for another 1 yr while unmedicated.

Outcome Measure: Halstead –Reitan Neuropsychological Test Battery, Katz Adjustment Scale, Sickness Impact Profile.

1.        From 1 to 12 mo, more participants in the treatment group stopped receiving their assigned drug (p<0.01) due to idiosyncratic reactions and requests.

2.        Those severely injured (GCS≤8) and receiving phenytoin did more poorly on most neuropsychological measures than controls determined by the overall rank-sum type test at 1 mo (p<0.05).  No significant differences found at 1yr.

3.        No significant differences in neuropsychological performance were found between groups for patients with moderate injuries (GCS≥9) at 1 mo or 1 yr.

4.        Changes in neuropsychological measures from 12 to 24 mo showed that phenytoin had a small but negative widespread cognitive effect as evidenced by the overall rank-sum type test (p<0.05).

Temkin et al. (1990)

USA

RCT

Population: TBI; Mean Age=34 yr; Gender: Male=309, Female=95; GCS≤10=256.

Treatment: Participants were randomized to either the phenytoin (n=208) or placebo group (n=196). Phenytoin group received an initial dose of 20 mg/kg intravenously, then serum levels were maintained at 3–6 µmol/l. Treatment started within 24 hr of injury and continued for 1 yr. Follow up at 2 yr.

Outcome Measure: Occurrence of early (<1 wk) and late (>8 days) seizures.

1.        Cumulative early seizure rates were 3.6% in the phenytoin group and 14.2% in the control group (p<0.001); Phenytoin was associated with a decrease of 73% in the risk of early seizures.

2.        Late seizure occurrence (day 8 to 2 yr) did not differ significantly between the treatment and control group (27.5% vs 21.2%, p>0.2).

3.        More participants in the phenytoin group stopped taking the drug between day 8 and 1 yr, mainly due to idiosyncratic reactions or requests (103 vs 67).

Young et al. (1983)

USA

RCT

PEDro=6

N=244

Population: TBI; Phenytoin Group (n=136): Mean Age=24.4 yr; Gender: Male=110, Female=26. Placebo Group (n=108): Mean Age=25.8 yr; Gender: Male=91, Female=71.

Treatment: Patients were administered phenytoin (concentration between 10 and 20 µg/ml) or placebo, starting within 24 hr of injury.

Outcome Measure: Occurrence of early seizures (≤1 wk of injury).

1.        5 in the phenytoin group and 4 in the control group had early seizures (p=0.75).

2.        Mean time from injury to early seizure in the treatment and control group was 3.2 and 4.5 days, respectively (p=0.41).

Young et al. (1983)

USA

RCT

PEDro=6

N_initial=214, N_final=179

Population: TBI; Mean Age=25.2 yr; Gender: Male=178, Female=36.

Treatment: Participants treated with Phenytoin (n=105; concentration between 10 and 20 µg/ml) or placebo (n=74) starting within 24 hr of injury. Treated for 18 mo, switched to phenobarbital if there was a hypersensitivity to phenytoin (n=20).

Outcome Measure: Occurrence of late (>7 days post injury) seizures.

1.        Late seizures occurred in 11 (12.9%) of the phenytoin group, 2 (10%) of the phenobarbital group, and 8 (10.8%) of controls.

2.        There were no significant differences between groups in the percentage of late seizures (p=0.75).

McQueen et al. (1983)

UK

RCT

PEDro=7

N=164

Population: TBI; Age: 5-15 yr=43, 16-65 yr=121; Gender: Male=130, Female=34.

Treatment: Patients received either phenytoin (n=84) or placebo (n=80) for 1 yr. Phenytoin administration for adults was 300mg and for children 5 mg/kg. Follow-up continued for 2 yr.

Outcome Measure: Occurrence of seizures.

1.        Only 48% of the treatment group had plasma levels greater than 40µmol/l.

2.        9.1% of participants developed post-traumatic epilepsy with first 2 yr.

3.        At 1 yr, 6 participants in the treatment group and 5 in the control group developed post-traumatic epilepsy.

4.        8 participants in the treatment group and 7 in the control group developed seizures by 2 yr.

Bhullar et al. (2014)

USA

Case Control

N=93

Population: TBI, GCS=3-8; No Prophylaxis Group (n=43): Gender: Male=28, Female=15. Phenytoin Prophylaxis Group (n=50): Gender: Male=42, Female=8.

Treatment: Medical records were reviewed and patients were divided into two groups: no prophylaxis and phenytoin prophylaxis.

Outcome Measure: Occurrence of early (<7d post injury) seizures, length of stay (LOS), Glasgow Outcome Scale (GOS), modified Rankin Scale (mRS).

1.        There was no significant difference in the occurrence of early seizures between the no prophylaxis and phenytoin groups (2.3% versus 4.0%, p=1.0).

2.        The phenytoin group, compared to no prophylaxis, had longer hospital stays (36± 31 vs 25± 16d, p=0.03), and worse functional outcome at discharge (GOS, 2.9± 1.0 versus 3.4±1.1, p=0.01; mRS, 3.1± 1.5 versus 2.3±1.7, p=0.02).

Phenytoin versus Levetiracetam

Younus et al.

(2018)

Pakistan

RCT

PEDro= 6

N = 140

Population: Phenytoin Group (N=69): Mean GCS= 11.23. Levetiracetam group (N=73): Mean GCS=11.17. Overall: 117 males, 23 females; Mean Age= 29.48±16.24yr.

Intervention: TBI patients admitted to the hospital were randomized into the Phenytoin medication group, or the Levetiracetam group. Both groups received medication for 7 days. No statistical differences between groups at baseline.

Outcomes:  Abnormal EEG, Seizure activity (7-10 days), Glasgow Coma Scale (GCS).

1.        The number of abnormal EEGs was found to be significantly different between the two groups (p=0.002) showing the Levetiracetam group had fewer individuals with abnormal EEG.

2.        The amount of seizure activity at follow-up was significantly different between groups (p=0.014), showing the Levetiracetam group had fewer instances of seizures.

3.        There was no significant difference between GCS scores at follow-up between the two groups.

Szaflarski et al. (2010)

USA

RCT

PEDro=8

N=52

Steinbaugh et al. (2012)

USA

Addition to Szaflarski et al. 2010 RCT

Population: TBI=46; SAH=6; Phenytoin group (PHT; n=18): Mean Age=35yr; Gender: Male=13, Female=5; Mean GCS=4. Levetiracetam group (LEV; n=34): Mean Age=44yr; Gender: Male=26, Female=8; Mean GCS=5.

Treatment: Patients were randomized within 24h of injury. Patients received either a loading dose of intravenous PHT 20mg/kg, then 5mg/kg/d or intravenous LEV at 20mg/kg, and then 1000mg every 12hr/7d.

Outcome Measure: Occurrence of early seizures, Glasgow Outcome Scale (GOS), GOS-Extended (GOSE), Disability Rating Scale (DRS), Resource Utilization Questionnaire.

Addition: Patients received continuous video EEG (cEEG) for up to 72h which was compared to the outcomes collected.

1.        There were no significant differences in the occurrence of early seizures between the PHT and LEV groups (3 versus 5, p=1.0)

2.        There were no significant between-group differences in GOS at discharge (p=0.33) and 6mo post discharge (p=0.89).

3.        There were no significant differences in the occurrence of fever, increased intracranial pressure, stroke, hypotension, arrhythmia, renal/ liver abnormalities or death between the two groups (p>0.15 for all).

4.        Compared to the LEV group, those in the PHT group experienced a significant worsening of their neurological status more often (p=0.024), and experienced anemia less often (p=0.076).

5.        Compared to PHT group, the LEV group showed significantly lower DRS at 3 and 6mo (p=0.006 and p=0.037), and higher GOSE at 6mo (p=0.016) in patients who survived.

1. The presence of focal slowing, epileptiform discharges, and seizures were not predictive of outcome (GOS-E, DRS). More severe slowing was positively associated with DRS at discharge, 3 and 6mo (p=0.084) and negatively associated with GCS at discharge.

Khan et al. (2016)

Pakistan

Cohort

N=154

Population: Mean Age=24.15yr; Gender: Males=115, Females=29; Mean GCS: 59.1% (8-13), 40.9% (3-7).

Intervention: Group A received Phenytoin (5 mg/kg/day), group B received Levetiracetam (10-20 mg/kg/day).

Outcome Measure: Incidence of post-traumatic seizures, efficacy of drug on moderate vs severe TBIs.

1.        There were no significant differences between groups in terms of the drug efficacy of Phenytoin vs Levetiracetam.

2.        There was no significant difference in how each drug impacted moderate vs severe TBI and seizure rates.

Javed et al. (2016)

Pakistan

Cohort

N=100

Population: Group 1 (n=50): Mean Age=31.16yr. Group 2 (n=50): Mean Age=34.96.

Intervention: Group 1: received IV phenytoin and Levetiracetam (35 mg/kg three times daily). Group 2: EEG monitoring.

Outcome Measure: Incidence of post-traumatic seizures.

1.        There were no significant differences between the number patients in each group which had post-traumatic seizures.

Radic et al. (2014)

USA

Case Control

N=288

Population: Subdural Hematoma; Levetiracetam group (LEV; n=164): Mean Age=65.96yr; Gender: Male=98, Female=66; Mean GCS=13.5.  Phenytoin group (PHT; n=124): Mean Age=62yr; Gender: Male=85, Female=39; Mean GCS=12.7.

Treatment: Patients were retrospectively analyzed. Those who received LEV were compared to those who received PHT for seizure prophylaxis.

Outcome Measure: Seizure rate and adverse drug events.

1.        There was no significant difference between LEV and PHT in clinical or electrographic seizure risk for patients without a midline shift.

2.        In subjects with midline shift >0 mm, LEV was associated with an increased risk of electrographic seizures during hospitalization (p=0.028) and a decreased risk of adverse drug effects (p=0.001), compared with PHT use.

Gabriel & Rowe (2014)

USA

Cohort

N=19

Population: TBI; Phenytoin Group (PHT, n=14): Mean Age=46.8yr; Gender: Male=10, Female=4; Mean GCS=3. Levetiracetam Group (LEV, n=5): Mean Age=48.8yr; Gender: Male=3, Female=2; Mean GCS=14.

Treatment: Participants were divided based on prophylactic treatment: PHT or LEV. Follow-up interview conducted.

Outcome Measure: Glasgow Outcome Scale- Extended (GOS-E), occurrence of seizures, medication-related complications.

1.        Groups were not similar at baseline in terms of median GCS at presentation (p=0.016) and ICU discharge (p=0.044). The PHT group, compared to the LEV group, also had a longer period of time between injury and GOS-E assessment (808.8 versus 484.4d, p=0.001).

2.        There was no significant difference in the mean GOS-E scores at follow-up (PHT 5.07 versus LEV 5.60, p=0.58).

3.        There was no significant difference between groups for occurrence of early or late seizures (both p=0.53).

4. Compared to the PHT group, the LEV group was significantly less likely to experience mediation-related complications (p=0.038); the PHT group had a significantly higher rate of days with fever (p=0.014).

Inaba et al. (2013)

USA

Prospective Controlled Trial

N=813

Population: TBI; Levetiracetam Group (LEV, n=406): Mean Age=51.7yr; Gender: Male=300, Female=106; Mean GCS=12.1. Phenytoin Group (PHT, n=407): Mean Age=53.6yr; Gender: Male=280, Female=127; Mean GCS=12.6.

Treatment: Participants were administered either LEV at 1000mg every 12h or PHT. In the PHT group the loading dose was 20mg/kg then 5mg/kg/d every 8h. Treatment lasted 7d.

Outcome Measure: Seizure occurrence.

1.        There was no significant difference in seizure rates between groups (1.5% versus 1.5%, p=0.997).

2. There was no significant differences between groups (LEV versus. PHT) in terms of adverse drug reactions (7.9% versus 10.3%, p=0.227), complications (28.3% versus 27.0%, p=0.679) or mortality rates (5.4% versus 3.7%, p=0.236).

Kruer et al. (2013)

USA

Retrospective Cohort

N=109

Population: TBI; Median GCS=5. Phenytoin Group (PHT, n=89): Median Age=43.1yr; Gender: Male=76, Female=13. Levetiracetam Group (LEV, n=20): Median Age=34.1yr; Gender: Male=19, Female=1.

Treatment: Retrospective review of patients administered PHT or LEV.

Outcome Measure: Occurrence of early seizures.

1.        One patient from each group seized in the first 7d (p=0.335).

2. Hospital length of stay did not differ significantly between groups (median days, LEV 26.5 versus PHT 11, p=0.134).

Jones et al. (2008)

USA

Cohort

N=27

Population: Severe TBI; Gender: Male=20, Female=7.

Treatment: Patients received Levetiracetam (n=15; 500mg IV every 12h for 7d) administered within 24hr of injury and were compared to a retrospective cohort of patients who received phenytoin (n=12).

Outcome Measure: Occurrence of early seizures.

1.        There was a significant difference in the occurrence of abnormal EEG findings (seizure or seizure tendency with epileptiform activity) between groups (p= 0.003), with the Levetiracetam group having more abnormal findings.

2. There was no significant difference between groups for actual seizures (p=0.556).

Additional Studies of Phenytoin

Dikmen et al. (2000)

USA

RCT

PEDro=8

N_initial=279, N_final=107

Population: TBI; Gender: Male=228, Female=51. Group 1 (n=94): Mean Age=37.14 yr; Mean GCS=11.3. Group 2 (n=91): Mean Age=36.58 yr; Mean GCS=11.23. Group 3 (n=94): Mean Age=35.85 yr; Mean GCS=12.11.

Treatment: Patients randomized into three groups within 24 hr of injury: 1) valproic acid (VPA) for 1 mo then 5mo of placebo; 2) VPA for 6 mo; and 3) phenytoin (PHT) for 1 wk then placebo until 6 mo post injury.

Outcome Measure: A battery of neuropsychological measures.  

1.        There was a trend towards a higher mortality rate in the VPA groups compared to the PHT group (p=0.07).

2.        There were no significant differences at 1, 6 or 12 mo on the composite measures based on all the neurospsychological measures, or on only the cognitive measures (0.551<p<0.812).

3.        No individual measure showed a significant difference among the treatment groups at 1, 6 or 12 mo post-injury.

Temkin et al. (1999)

USA

RCT

PEDro=7

N­_initial=379, N­_final=283

Population: TBI; Gender: Male=310, Female=69; Phenytoin Group (n=132): Mean Age=36 yr; Mean GCS=11.7.

Valproate (1mo, n=120): Mean Age=40 yr; Mean GCS=11.6. Valproate (6mo, n=127): Mean Age=36 yr; Mean GCS=11.1.

Treatment: Patients were divided into three groups within 24 hr of injury: (1) phenytoin for 1 wk (20 mg/kg then 5 mg/kg/day), placebo until 6 mo post injury; (2) Valproate (20 mg/kg, then 15 mg/kg/day) for 1 mo, placebo for 5 mo; or (3) valproate for 6 mo.  Follow-up continued for 2 yr.

Outcome Measure: Incidence of early and late (>7 day post injury) seizures, mortality rates.

1.        There was no significant difference in the number of early seizures between the combined valproate (4.5%) and phenytoin (1.5%, p=0.14) groups.

2.        There is no significant difference between groups (p=0.19) in the occurrence of late seizures.

3.        Late seizures occurred in 11, 17, and 15 participants in the 1 mo and 6 mo valproate groups and the phenytoin group, respectively.

4.        There was no significant differences in mortality rates between groups (7.2% phenytoin versus 13.4% in the combined valproate group, p=0.07).

5.        In the phenytoin group, a participant had a rash requiring medication at 1 wk and in the valproate (6 mo) group a participant had low neutrophil count at 2-4 wk, both thought to be treatment related.

Bhullar et al. (2014)

USA

Case Control

N=93

Population: TBI; Gender: Male=70, Female=23; GCS=3-8.  

Treatment: Medical records were reviewed and patients were divided into two groups: no prophylaxis (n=43) and Phenytoin prophylaxis (n=50).

Outcome Measure: Occurrence of early (<7 days post injury) seizures, length of stay (LOS), Glasgow Outcome Scale (GOS), modified Rankin Scale (mRS).

1.        No significant difference in early seizures between the no prophylaxis and phenytoin group (2.3% versus 4.0%, p=1.0).

2.        The Phenytoin group, compared to no prophylaxis, had longer hospital stays (36± 31 versus 25± 16 days, p=0.03), worse functional outcome at discharge (GOS, 2.9± 1.0 versus. 3.4±1.1, p=0.01; mRS, 3.1± 1.5 versus 2.3±1.7, p=0.02).

Servit & Musil (1981)

Czechoslovakia

PCT

N=167

Population: TBI; Mean Age=30.6 yr; Gender: Male=128, Female=39.

Treatment: Participants in the treatment group (n=143) were administered Phenytoin (160-240 mg/day) and phenobarbital (20-60 mg/day). The control group (n=24) was treated with conventional methods for 2 yr.

Outcome Measure: Occurrence of late seizures.

1.        Posttraumatic epilepsy occurred in 25% of the control and 2.1% of the treatment group after discontinuing therapy (p<0.001).

2.        One individual (0.7%) had a seizure during prophylactic treatment.