SynopsisFifty-nine patients suffering from a major depressive episode, for whom electroconvulsive therapy (ECT) was clinically indicated, were randomly assigned to one of three electrode placement groups for treatment with brief pulse, threshold-level ECT: bitemporal (BT), right unilateral (RU) or bifrontal (BF). Comparison of these groups in terms of number of treatments, duration of treatment, or incidence of treatment failure, showed that the bilateral placements were superior to the unilateral; comparison of Hamilton, Montgomery–Åsberg, and visual analogue scale scores showed that the bifrontal placement was superior to both bitemporal and unilateral treatment. Bitemporal treatment showed therapeutic results intermediate between BF and RU. Because BF ECT causes fewer cognitive side effects than either RU or BT, and is independently more effective, it should be considered as the first choice of electrode position in ECT.

S U M M A R Y Diabetic patients have longer interpeak latencies in the brainstem auditory evoked responses than age-matched controls. The delay is not related to clinical hearing loss or blood glucose level at time of testing. Since waves I and II are normal in latency, the conduction velocity of the eighth nerve is not involved. The delay occurs between waves II and V, which would reflect altered transmission times in auditory brainstem and midbrain structures, and suggests the presence of a central neuropathy in patients with diabetes.It is well known that diabetic patients develop peripheral and autonomic neuropathy. Recent reviewsl'3 have suggested that they may also suffer from central neuropathy, or degeneration of the higher nervous system; De Jong4 has pointed to clinical and pathological evidence that the brain parenchyma might be affected. Kent5 has argued that diabetic patients show some neurological and psychological symptoms that might signify premature aging. However, physiological evidence of central nervous system dysfunction has not been forthcoming, and relatively little attention has been paid to the possibility of central neuropathy in such patients.In a pilot study6 7 we compared the auditory cortical evoked responses of a group of diabetics with those of an age-matched control group, and found the diabetics to have greater suppression of the cortical auditory evoked response at high rates of stimulation, implying a slowing of the recovery process in the central nervous system. The interpretation of these changes was not clear; similar modifications of the auditory evoked response could occur in clinical hearing loss, which has been reported in diabetics.8-'2 Hypoglycaemia,

SYNOPSIS Forty patients suffering from a major depressive disorder, for whom electroconvulsive therapy (ECT) was clinically indicated, were assigned to one of three electrode placement groups: bitemporal (BT), right unilateral (RU) or bifrontal (BF). Comparisons of these groups in terms of cognitive status showed that the BF placement, which avoided both temporal regions, spared both verbal and nonverbal functions. These differential effects, which were independent of the degree of clinical depression, were not, however, evident three months after the last ECT.

No abstract