Reply

Abstract

We appreciate the comments by Gilbert and Buncher In the abstract of our article,1Pressler R.M. Robinson R.O. Wilson G.A. Treatment of interictal epileptiform discharges can improve behavior in children with behavioral problems and epilepsy.J Pediatr. 2005; 146: 112-117Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar we refer to “ambulatory EEG,” which is normally used for and hence implies long-term monitoring. We used ambulatory rather then routine recording in this drug-trial as a safety measure: subclinical EEG discharges occur at anytime within a 24-hour period but random 20 minutes EEG recordings sample only a small fraction of this activity.2Binnie C.D. Aarts J.H. Houtkooper M.A. Laxminarayan R. Martins de Silva A. Meinardi H. Temporal characteristics of seizures and epileptiform discharges.Electroenceph Clin Neurophysiol. 1984; 58: 498-505Abstract Full Text PDF PubMed Scopus (35) Google Scholar It is known that this pattern is frequently modulated by extrinsic or intrinsic factors.3Kellaway P. Frost J.D. Crawley J.W. Time modulation of spike-and-wave activity in generalized epilepsy.Ann Neurol. 1980; 8: 491-500Crossref PubMed Scopus (82) Google Scholar Assessments of the therapeutic response based on random routine EEGs are therefore likely to be inaccurate4Milligan N. Richens A. Methods of assessment of antiepileptic drugs.Br J Clin Pharmacol. 1981; 11: 443-456Crossref PubMed Scopus (27) Google Scholar and in a drug trial long-term monitoring bares a smaller risk of false positive or negative results. This does not imply that in a clinical setting the same methods are necessary, but in some patients it may be helpful to assess the true burden of interictal discharges with long-term monitoring before considering treatment. We now present a Table that the Editors deleted from the paper in the interest of brevity. It shows observed differences in global and parental behavioural subscale T-scores between placebo and lamotrigine phases stratified for EEG changes (with or without reduction of discharges) and statistical results. Values shown are calculated by subtracting the score during placebo phase from the corresponding score during lamotrigine phase so that a negative score reflects an improvement in behaviour and a positive score a worsening. By comparing “active drug phase” with “placebo phase” rather with “baseline” in combination with a randomised, cross-over design we reduced the risk of placebo and order effects.TableChanges of behavioral scores during active drug treatment stratified for EEG changesInterictal dischargesReduction of frequency Mean (SD)No reduction of frequency Mean (SD)FdfReduction of duration Mean (SD)No reduction of duration Mean (SD)FdfGlobal score−1.3 (4.6)0.5 (3.5)2.2⁎P < .05.14,24−1.2 (4.5)0.4 (3.5)2.5⁎P < .05.14,24Parental sub-scores Antisocial−2.7 (5.0)−0.2 (4.2)0.314,37−2.6 (4.9)−0.2 (4.3)014,37 Anxious−1.4 (7.4)0.1 (4.7)014,37−1.5 (7.2)0.1 (4.8)0.114,37 Conduct disorder−3.4 (6.4)−1.4 (4.2)5.7⁎P < .05.14,37−2.7 (6.8)−1.9 (3.6)4.4⁎P < .05.14,37 Hyperactive−2.0 (6.2)−0.8 (5.6)1.014,37−1.7 (6.2)−1.1 (5.6)0.714,37 Learning problem−1.4 (7.2)−0.2 (3.7)2.914,37−0.9 (7.3)−0.6 (3.2)2.414,37 Obsessive0 (6.4)0.5 (5.5)2.814,370.5 (6.6)0 (5.2)2.814,37 Psychosomatic−2.8 (10.3)−0.5 (6.6)2.914,37−3.5 (10.5)0.3 (5.6)4.4⁎P < .05.14,37 Restless−2.6 (4.1)−1.8 (6.9)014,37−2.6 (4.0)−1.8 (7.1)014,37 P < .05. Open table in a new tab Only patients with a reduction of interictal discharges showed an improvement of the global behavioural score and the improvement of parental subscale scores was bigger in all sub-tests even though this reached statistical significance in only 2 subscales. We agree that this table is necessary to understand and accurately interpret the data. As we stressed in our article, the magnitude of change in behavioral scores is small, but these scores are group results, and do not reflect changes in single patients which may be much bigger and are likely to be of clinical significance in some patients. This problem is inherent to group analyses of any behavioral scores. We would like to thank Dr. Nora Donaldson, Biostatistics Department, Institute of Psychiatry, King’s College London, for statistical advice. Epileptiform discharges and the behavior of children with epilepsyThe Journal of PediatricsVol. 149Issue 2PreviewPressler et al in a randomized controlled crossover trial addressed the question of whether reducing epileptiform discharges improves behavior in children with epilepsy.1 This clinical trial addresses an important topic but has key deficiencies with regard to reporting of clinical trials. The electroencephalograms (EEGs) used for this research study are not standard of care for epilepsy (or behavior). The 12- to 24-hour EEG duration should have been disclosed in the abstract. The authors fail to report effect sizes on behavior outcomes. Full-Text PDF