A follow-up of an incident case-referent study of febrile convulsions seven years after the onset

Abstract

A cohort of 92 children with an initial febrile convulsion (FC), prospectively identified in a community-based study 1985-1987, was compared with a cohort of 185 age- and sex-matched referents from the same study area. The median time of follow-up was 6.7 years (range: 5.7-7.7) and the median age at follow-up was 8.3 years (range 6.5-14). The cases had their first FC at a median age of 18 months (range 5-67 months), their second FC at 24 months (11-108 months) and their third FC at 26 months (13-92 months). FC recurred in 42% of the FC cohort, and 3.8% of the children in the referent cohort experienced FC. Single or recurrent afebrile seizures occurred in 4.3% and epilepsy in 3.3% of the FC cohort, while no afebrile seizures occurred in the referent cohort. The risk of having a sibling with FC was three times (95% confidence interval 1.3-6.2) higher in the FC cohort, while there was no difference between the cohorts in the risk of siblings developing afebrile seizures. There was no difference between the cohorts in the utilization of health services during the follow-up period. Two children in the FC cohort went to a school for the mentally retarded. There was no aetiological relationship between the FC and the mental retardation in these cases. All other children attended normal schools and none needed remedial instruction.

Febrile Convulsions in 220 Children – Neurological Sequelae at 12 Years Follow-Up

“…We identified 17 cohort studies (including the NGPSE) via Medline search and referenced articles [2, 3, 6, 8, 15, 25, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40], but have excluded 4 [6, 38, 39, 40]because of uncertainty about inclusion, recruitment or duration of follow-up. For the remaining 13 studies, the percentage developing epilepsy plotted against mean (or median) age at the last available follow-up is shown (fig.…”

“…Studies of FC, using significance testing [8, 15], have ascribed differing levels of risk of epilepsy and other neurological sequelae to the presence of complex FCs, numbers of FCs, family history of FC or epilepsy, prior or subsequent neurological deficit. Odds ratios have been used in 2 studies, a small cohort had non-significant findings due to wide confidence intervals [36], the other [25]has as yet limited follow-up, but demonstrates an increased rate of unprovoked seizures at 2 years in those who had experienced more FCs: OR 4.2 (95% CL 1.9, 6.6), 20.4 (95% CL 4.4, 36.4) for children with 1 or 4 FCs, respectively. These children were analysed after assignment to high- and low-risk (of FC recurrence) groups (stratified by family history, age at first FC and temperature at first FC) in a multivariate model for risk of unprovoked seizures after FC.…”

“…A further overview pooling data from 6 studies [3, 8, 15, 35, 36]which report the type of first FC is shown in figure 2; the NGPSE is the only study with a point estimate of the odds ratio for epilepsy following a complex as opposed to simple first FC of less than one (figures greater than one imply a tendency for the outcome measured to occur and conversely for less than one). However, there is no evidence of heterogeneity between studies (χ 2 = 5.15, d.f.…”

“…In this cohort, 11% of the children were treated with long-term AEDs including 8 patients with simple FCs; modern protocols might regard this as overtreatment. Two other community-based studies mention treatment rates of 0.5% in Sweden [36]and 13.7% in the UK [8]. As there is little evidence of benefit in AED therapy in FCs and clear evidence of harm, it can only be advocated under exceptional circumstances.…”

Risk factors for subsequent febrile seizures in the FEBSTAT study

“…Population-based and birth cohort studies have examined the risk for subsequent febrile seizures (FS) and find that the risk of a first recurrence ranges from 23%–42% 1; 2 with more than 90% of recurrence occurring during the first two years. In pooled studies, the risk for a first recurrence during a three year period ranges from 28% to 32% in population-based studies and 33% to 43% in clinic-based studies 3 .…”