Objective: This study tests the primary hypothesis that secondary generaliz
ation of partial seizures is more likely after anterior temporal lobectomy
(ATL) than before ATL, and the secondary hypothesis that antiepileptic drug
withdrawal accounts for increased generalization of seizures postoperative
ly. Background: The authors observed that some patients had generalized ton
ic-clonic (GTC) seizures after but not before ATL, by using a new classific
ation of outcome that compares preoperative and postoperative seizure frequ
encies by seizure type. Methods: Twenty patients with refractory temporal l
obe epilepsy had postoperative GTC seizures or nongeneralizing complex part
ial (CP) seizures in a consecutive ATL series. All had reduced seizure freq
uency postoperatively and more than 2 years of follow-up on antiepileptic d
rugs. The authors calculated a generalization fraction, as (number of GTC s
eizures)/(number of CP and GTC seizures), for 2 years before and 2 years af
ter surgery. Results: Postoperative generalization fractions were greater t
han preoperative generalization fractions (Wilcoxon signed-rank test, p < 0
.01). Most postoperative GTC seizures were not associated with antiepilepti
c drug withdrawal, and postoperative GTC seizures were not more associated
with drug withdrawal than were postoperative CP seizures. Patients with mor
e than two GTC seizures per year preoperatively were more likely than other
patients to have postoperative GTC seizures. Conclusions: Patients with re
duced seizure frequency after ATL have a greater tendency for partial seizu
res to secondarily generalize postoperatively. This phenomenon is not expla
ined by antiepileptic drug withdrawal.