Abstracts

Rationale: To identify premorbid medical risk factors for the development of pediatric non-epileptic seizures (NES), this study examined if these children have a medical history of more prior illness, hospitalizations, emergency room(ER) visits, and medication use than sibling controls. Within the NES group, it ascertained if earlier age of NES onset was associated with the predicted increased premorbid medical history. Methods: Medical history was obtained from the parents of 22 children with NES (7 males, 15 females) and 22 siblings (11 males, 11 females). Probands and siblings were compared on continuous measures using mixed models with family as a random effect. For categorical measures, chi-square analyses were used to examine differences in distribution. Fisher s exact test was used for those measures with frequencies less than 5 in each cell. Within NES subjects, Spearman correlations were computed between age of NES onset and medical history measures. Results: The mean age of the NES children,15.2(SD 2.4), was significantly older (p=.005) than their siblings, 13.3(SD 2.4). Between group analyses demonstrated a significantly higher percentage of NES children with a history of epilepsy compared to siblings (68.2% vs. 4.6%, X2 = 19.3, p<.0001). The NES children were also more likely to have a history of CNS medical problems other than epilepsy (meanNES 1.5(1.1) vs. meanSIB 0.3(0.7), p=.0002), and non-CNS medical problems (meanNES 1.8(1.8) vs. meanSIB 1.0(1.5), p=.05). Children with NES have also made significantly more lifetime ER visits (meanNES 5.9(3.8) vs. meanSIB 1.8(2.0), p=.0006) and hospitalizations (meanNES 3.1(2.6) vs. meanSIB 0.7(1.2), p=.001) as well as ER visits (meanNES 3.0(1.6) vs. meanSIB 0.2(0.5), p<.0001 and hospitalizations in the year prior to study entry (meanNES 1.6(0.8) vs. meanSIB 0.1(0.2), p<.0001.) than their siblings. As a result, children with NES have missed significantly more days of school in the month prior to the study than their siblings (mean NES 7.8(7.7) vs. meanSIB 1.7(1.9), p=.004). Children with NES were also taking significantly more prescription medications (mean NES 1.8(2.2) vs. meanSIB 0.4(0.9), p=.002) and total medications (mean NES 3.0(2.9) vs. meanSIB 1.0(1.4),p=.005) than their siblings. Within the NES group, the mean age of onset of NES, 14.3 (2.6) years, correlated negatively with number of ER visits (r=-.55, p=.01), times hospitalized (r=-0.50, p=.02), and CNS medical problems (r=-.50,p=.03). Conclusions: These findings of increased medical risk factors demonstrate that children with NES have a history of lifetime medical problems and treatment. Given the difficulties involved in diagnosing NES which often co-occurs with epileptic seizures, physicians should rule out NES in the differential diagnosis of children with new onset or difficult to control seizures who have a significant lifetime history of somatic problems.