S Tupola1, P Saar, J Rajantie. 1. Aurora Hospital, Helsinki, Finland; Childreńs Hospital, University of Helsinki, Helsinki, Finland; and Jorvi Hospital, Espoo, Finland.
Abstract
OBJECTIVES: To evaluate the predictive value of the electroencephalogram (EEG) at the time of diagnosis of insulin-dependent diabetes mellitus (IDDM) for subsequent hypoglycemic coma and/or convulsion. To study whether such an episode causes long-term EEG abnormalities. STUDY DESIGN: An EEG was recorded in 36 patients with IDDM 2 to 3 weeks after diagnosis (median age, 7.5 years) and was then repeated after an episode of severe hypoglycemia associated with coma and/or convulsion (median age, 13.3 years). Paired EEGs were also recorded in 36 age-matched and IDDM duration-matched control patients who had never experienced severe hypoglycemia. A single EEG was recorded in 36 healthy children, matched with patients' ages at the time of IDDM diagnosis. RESULTS: Patients with severe hypoglycemia had an abnormal initial EEG recording more often than did control patients with IDDM (22.2% vs 2.8%, P =.03). Each of the healthy children had a normal EEG recording. The odds ratio for risk of subsequent coma and/or convulsion during hypoglycemia in patients with abnormal initial EEG recordings was 8 (95% CI, 1.1-354.7). After such an episode, the frequency of the abnormal EEG recordings was not elevated. CONCLUSIONS: EEG at the time of diagnosis of IDDM may be useful in identifying those patients at increased risk for coma and/or convulsion as a result of hypoglycemia. However, a single such episode did not appear to have a deleterious effect on the subsequent EEG.
OBJECTIVES: To evaluate the predictive value of the electroencephalogram (EEG) at the time of diagnosis of insulin-dependent diabetes mellitus (IDDM) for subsequent hypoglycemic coma and/or convulsion. To study whether such an episode causes long-term EEG abnormalities. STUDY DESIGN: An EEG was recorded in 36 patients with IDDM 2 to 3 weeks after diagnosis (median age, 7.5 years) and was then repeated after an episode of severe hypoglycemia associated with coma and/or convulsion (median age, 13.3 years). Paired EEGs were also recorded in 36 age-matched and IDDM duration-matched control patients who had never experienced severe hypoglycemia. A single EEG was recorded in 36 healthy children, matched with patients' ages at the time of IDDM diagnosis. RESULTS:Patients with severe hypoglycemia had an abnormal initial EEG recording more often than did control patients with IDDM (22.2% vs 2.8%, P =.03). Each of the healthy children had a normal EEG recording. The odds ratio for risk of subsequent coma and/or convulsion during hypoglycemia in patients with abnormal initial EEG recordings was 8 (95% CI, 1.1-354.7). After such an episode, the frequency of the abnormal EEG recordings was not elevated. CONCLUSIONS: EEG at the time of diagnosis of IDDM may be useful in identifying those patients at increased risk for coma and/or convulsion as a result of hypoglycemia. However, a single such episode did not appear to have a deleterious effect on the subsequent EEG.