EPILEPSY RESEARCH, cilt.111, ss.54-60, 2015 (SCI İndekslerine Giren Dergi)
Recurrence of infantile spasms (ISs) is common subsequent to treatment with adrenocorticotropic hormone (ACTH) for West syndrome, and prolonged hypsarrhythmia results in psychomotor deterioration. The evolution to hypsarrhythmia involves conversion of prehypsarrhythmic EEG findings to sporadic hypsarrhythmia paroxysms (HPs), and when paroxysms reach a certain frequency, ISs begin to occur. This retrospective chart study aimed to determine the HP threshold frequency after which ISs begin. Recorded either prior (Group A) or subsequent (Group B) to IS relapse, 248 EEGs were examined in 42 patients. The number of HPs in non-rapid eye movement (NREM) sleep divided by NREM duration constituted the countable hypsarrhythmia paroxysms index (cHPI). After reaching a rate of approximately 10/min, the cHPI lost its feasibility due to the merging of HPs. The durational HPI (dHPI) was also calculated (total duration of HPs during NREM/NREM sleep time x 100). ACTH treatment was administered if cHPI was >= 2/min, with the aim of preventing relapse. The mean cHPI value without a concomitant spasm relapse (in Group A) was 1.20/min. Following relapse, this value rose to 4.10/min. EEGs performed subsequent to relapse (in Group B) were classified into three subgroups (B1, B2, and B3) according to the duration of the time interval between IS relapse and the succeeding EEG recording. One-way analysis of variance (ANOVA) indicated that cHPI values differed significantly between the Group B subgroups. In subgroups B2 and B3, a higher number of EEGs were evaluated via dHPI. Linear regression analysis established that the interval between recurrence and the succeeding EEG recording significantly predicted cHPI values and accounted for 54.2% of the explained variability in cHPI values. Therefore, use of the cHPI for early recognition and intervention may aid in preventing the onset and recurrence of ISs and further deterioration of psychomotor development. (C) 2015 Elsevier B.V. All rights reserved.