Summary:Purpose: Some children with rolandic epilepsy have associated neuropsychiatric deficits resembling symptoms of attention deficit-hyperactivity disorder (ADHD), the most common neurobehavioral disorder of childhood. The clinical overlap between both syndromes has received relatively little attention. The study examines the frequency of rolandic spikes in nonepileptic children with ADHD and compares it with a historic control group of 3,726 normal school-aged children. ADHD patients with and without discharges are compared regarding age at admission, sex, global functioning, and distribution of ADHD subtypes.Methods: The EEGs of 483 ADHD outpatients between 2 and 16 years meeting diagnostic criteria for ADHD according to Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) were evaluated prospectively. If rolandic spikes were present, separate sleep EEGs were performed to exclude a bioelectrical status epilepticus during slow-wave sleep.Results: Rolandic spikes were detected in the EEGs of 27 children (5.6%; 22 boys and five girls). Seizure rate during follow-up tended to be larger in children with rolandic spikes. No significant differences were found between ADHD patients with and without spikes regarding sex and global functioning. ADHD children with rolandic spikes came to our attention significantly earlier than did children without discharges and tended to exhibit more hyperactive-impulsive symptoms, evidenced in a larger proportion of the diagnosis of ADHD combined type than ADHD inattentive type.Conclusions: The frequency of rolandic spikes in children with ADHD is significantly higher than expected from epidemiologic studies. The question arises how ADHD symptoms are related to rolandic spikes in this ADHD subgroup. Possibly rolandic discharges or underlying, not fully understood mechanisms of epileptogenesis decrease the vulnerability threshold, advance the onset, or aggravate the course of ADHD.

Alcoholism is a major contributor to the physical ill-health of homeless people. Treatment or rehabilitation of addictive behaviour among the homeless should be of major concern for adequate service planning or provision.

The results suggest that prevention strategies should be multidimensional and interdisciplinary.

The molecular basis of febrile convulsions (FCs) is poorly understood when compared with recent advances in knowledge of the underlying mechanism for the idiopathic epilepsies. FCs is the commonest seizure syndrome and is usually regarded as having a polygenic basis (1). Chou et al. (2) suggest an insight into FC genetic susceptibility based on the premise that polymorphism within known epilepsy genes might constitute part of the genetic background affecting seizure threshold in the common multifactorial forms of seizure (3).Chou et al. reported a significant association between the α4 subunit of the neuronal nicotinic acetylcholine receptor (CHRNA4) and FCs. The T allele of the (Ser543Ser)-C/T polymorphism predisposes their CT heterozygotes to FCs almost fourfold relative to CC homozygotes, although linkage disequilibrium with an unknown functional variant must be invoked as part of the explanation. The extent to which this result can be extended to other populations remains to be assessed because associations have a poor record for repetition and may be restricted to certain ethnic groups with unique interacting genetic backgrounds.CfoI digestion (cleaving the same site as HhaI) of a similar polymerase chain reaction (PCR) product containing the same polymorphism as studied by Chou et al. shows that their result is not translatable to our Caucasian study population. Comparison of 49 unrelated patients with simple FCs (12 CC, 25 CT, 12 TT; C = 0.5, T = 0.5) with controls (17 CC, 44 CT, 32 TT; C = 0.42, T = 0.58) showed no significant difference in allele frequency (χ 2 1 = 1.38; p ∼ 0.25). The T allele was marginally less frequent in our subjects compared with controls, rather than significantly more frequent, as found by Chou et al. Close fit of our combined FC and control population to HardyWeinberg equilibrium shows no indication for population stratification (χ 2 1 = 0.06; p ∼ 0.8). Observed CHRNA4 genotypes (with expected numbers in parentheses) were 29 (28.4) CC, 69 (70.2) CT, and 44 (43.4) TT.We caution against the acceptance of associations before results have been replicated. However, given that FCs are part of the clinical spectrum in generalized epilepsy with febrile seizures plus and severe myoclonic epilepsy of infancy, considerable scope remains for further association studies between FCs and naturally occurring polymorphisms within the SCN1A, SCN1B, and GABRG2 genes in various populations. The appropriate safeguards must be implemented to minimize spurious positive associations from population stratification, multiple-hypothesis testing, and bias for preferential publication of positive results.