Abstracts

Rationale: Lamotrigine (LTG) has been shown to inhibit the cardiac potassium channel Ikr (Danielsson et al., 2005) and Ikr blocking drugs have been associated with increased risk for the long QT syndrome and sudden cardiac death. Recently we reported four consecutive cases of SUDEP in women with idiopathic epilepsy that were all treated with LTG in monotherapy (Aurlien et al., 2007). An increased risk connected to LTG has, however, never been shown in a clinical study. Here we present a retrospective population based study on the incidence of SUDEP in Rogaland County, Norway, (375.000 inhabitants) in the ten years period 01.08.1995 - 31.07.2005. The incidence of SUDEP associated with each antiepileptic drug (AED) was estimated. Methods: The SUDEP victims were identified by review of hospital records and post mortem reports of deceased individuals with a diagnosis of epilepsy and data from the National Causes of Death Registry. The victims were classified according to the definitions of definite , probable and possible SUDEP (Tomson et al., 2008). With a prevalence of epilepsy of 0.7 per 1000 (Forsgren L., 2004) the number at risk was estimated to 2612 individuals. Based on the market share in defined daily doses (DDD) the incidence of SUDEP was estimated for each AED. Statistical analysis was performed for AEDs with at least 5 SUDEP victims. Results: 26 cases (15 females, 11 males) were identified. 16 were definite, 3 were probable and 7 possible SUDEP. The incidence of SUDEP was 1.0 per 1000 patient-years when all 26 cases were included and 0.7 per 1000 patient-years for definite and probable SUDEP. 10 (9 females, 1 male) of the 26 (38.5 %) were treated with LTG; 5 in monotherapy and 5 in polytherapy. 2 /10 were classified as possible SUDEP, both were on polytherapy. 8 of the 19 cases (42 %) classified as definite and probable SUDEP were on LTG. The incidence of SUDEP for patients treated with LTG (all cases included) was 4.9 per 1000 patient-years compared to 0.7 per 1000 patient-years for patients that were not treated with LTG (p < 0.001, odds ratio 7,7 with 95 % confidence interval 3,5 - 16,8). The incidence of definite and probable SUDEP for patients treated with LTG was 3.9 per 1000 patient-years and 0.46 per 1000 patient-years for patients that were not treated with LTG (p < 0.001, odds ratio 8,8 with 95 % confidence interval 3,6 - 21,6). Among the 26 cases 7 were treated with carbamazepine (CBZ) (4 in monotherapy), and 8 with valproate (VPA) (4 in monotherapy). The SUDEP incidence for CBZ and VPA was not significantly different from the incidence among those not treated with these AEDs. In addition, 3 were treated with phenytoin, 3 with vigabatrin, 3 with oxcarbazepine, 2 with topiramate, 1 with phenobarbital and 1 was untreated. Conclusions: The incidence of SUDEP in patients treated with LTG was significantly increased compared to other AEDs. The findings may suggest a gender difference with a higher incidence in women. The total incidence of SUDEP in Rogaland County was similar to that of previous population based studies.