Are geomagnetic disturbances a risk factor for the sudden unexplained death of epileptic children?

From SUDEP Wiki
Jump to navigation Jump to search

Schnabel R, May TW, Beblo M, and Gruhn LF (2003) Are geomagnetic disturbances a risk factor for the sudden unexplained death of epileptic children? Int J Biometeorol 48:1 20–4.

Link to Article

Abstract: The relationship between geomagnetic disturbances (with amplitudes above 50 nT) and the sudden unexplained death (SUD) of epileptic adults has been controversial up to now. In an autopsy-based cohort study (from 1964 to 1992) 15 epileptic children with SUD aged 1-14 years were compared to 63 epileptic children whose cause of death was known (KCD). The time of death as well as two international geomagnetic indices were assessed: Bartels' planetary 3-h signs (Kp) for the time of death and 2 h before, and the mean planetary daily amplitudes (Ap) on the day of death and on the 4 critical days before (and on the 4 post-mortem days as a control). Among the 15 SUD patients 9 died between 6 a.m. and 9 a.m. (universal time) whereas the KCD group showed an approximately uniform distribution of the time of death ( P = 0.026, Kolmogorov-Smirnov test). The Kp signs at death and 2 h before as well as the Ap indices at death and for the 4 pre-mortem days or after were not different in comparisons between the SUD and the KCD group ( P > 0.2, Mann-Whitney test). Furthermore the frequency of sudden storm commencement at death and for the 4 days before revealed no deviation of the SUD (26.7%) from the KCD group (23.8%) ( P > 0.2, Fisher test). The results do not support the hypothesis that geomagnetic disturbances are a risk factor for the sudden unexplained death of epileptic children.

Keywords: Geomagnetic activity, Sudden unexplained death, Children, Epilepsy


  • Retrospective cohort study of SUDEP cases from 28 year period that compared 15 cases of SUDEP in decedents 1-14 years old to 63 deaths of known causes among children with epilepsy. Comparing time of death with two indices of geomagnetic activity, no evidence of a correlation was found. The small sample sizes would have the effect of decreasing power to detect difference. Additionally it could be argued that this question might not best be answered through a case-control design; for example, is there a positive correlation between local geomagnetic activity and rate of SUDEP, or do SUDEP deaths tend to occur at times where likely geomagnetic activity is atypically high? Negative answers to these questions with sufficiently large sample sizes might more definitely refute the hypothesis. Because most SUDEP victims have suffered hundreds or thousands of seizures in their lives, an explanation will continue to be sought for the fatality of a later seizure.


Network Graph

Retrieving data for the network graph...