To the Editor.—
The difficulty in getting statistically significant correlations between limbic epilepsy and episodic aggression was emphasized in the article "Intensive Monitoring of Epileptic Patients With a History of Episodic Aggression" by Drs Ramani and Gumnit in the Archives (1981;38:570-571).The significance of their data was restricted by the selection of their patient population. They specifically excluded extremely violent or dangerous patients from admission to their hospital unit. Even if the number of their patients was increased tenfold, prevalence or incidence studies of episodic rage behavior in their patients with limbic epilepsy would be unreliable. One wonders if a similar bias was involved in the study by King and Ajmone Marsan1 (cited by Drs Ramani and Gumnit as an example of a negative relationship between temporal lobe epilepsy [TLE] and abnormal aggressivity), since a patient with ictal rage2 had been rejected by their clinic because of her