C. Edward Coffey

Seizure threshold in electroconvulsive therapy: I. Initial seizure threshold

We measured initial seizure threshold by means of a structured stimulus dosage titration procedure in a clinical sample of 111 depressed patients undergoing brief-pulse, constant-current electroconvulsive therapy (ECT). Initial seizure threshold was approximately 60 millicoumbs (mc) (10 Joules) on average, but varied widely (6-fold) across patients. Initial seizure threshold was predicted by four variables: electrode placement (higher with bilateral), gender (higher in men), age (higher with increasing age), and dynamic impedance (inverse relationship). Use of neuroleptic medication was associated with a lower seizure threshold. EEG seizure duration was inversely related to initial seizure threshold, but no other relations with seizure duration were found. These findings may have important clinical implications for stimulus dosing strategies in ECT.

Myocardial infarction and stroke

We used a computer data bank to evaluate 740 consecutive patients admitted to a cardiac care unit with myocardial infarction. Stroke occurred in 18 (2.4%) patients in the hospital; the anterior circulation was involved in 76% of strokes. Hospital mortality was 61% in patients with stroke and 13% in patients without stroke. Atrial arrhythmia was a significant (p 5 0.03) risk factor for stroke, but peak creatine kinase and ventricular arrhythmia were not. Cardiac pump failure, apical or anterior-lateral myocardial infarction, and history of previous stroke were associated with an increased risk of stroke. Clinical and pathologic data suggested an embolic etiology for most strokes that complicate acute myocardial infarction.

Seizure threshold in electroconvulsive therapy (ECT) II. The anticonvulsant effect of ECT

To measure the anticonvulsant effects of a course of electroconvulsive therapy (ECT), we used a flexible stimulus dosage titration procedure to estimate seizure threshold at the first and sixth ECT treatments in 62 patients with depression who were undergoing a course of brief pulse, constant current ECT given at moderately suprathreshold stimulus intensity. Seizure threshold increased by approximately 47% on average, but only 35 (56%) of the 62 patients showed a rise in seizure threshold. The rise in seizure threshold was associated with increasing age, but not with gender, stimulus electrode placement, or initial seizure threshold. Dynamic impedance decreased by approximately 5% from the first to the sixth ECT treatment, but there was no correlation between the change in dynamic impedance and the rise in seizure threshold. No relation was found between the rise in seizure threshold and either therapeutic response status or speed of response to the ECT treatment course. These findings confirm the anticonvulsant effect of ECT but suggest that such effects are not tightly coupled to the therapeutic efficacy of moderately suprathreshold ECT.

Cerebral laterality and emotion: The neurology of depression

Abstract Differential specialization of the cerebral hemispheres for cognitive functioning is well established. A converging body of clinical and experimental studies conducted with brain-damaged patients, normal subjects, and psychiatric patients is reviewed which indicates that there is also hemispheric asymmetry in the regulation of emotional behavior. These data indicate that the right hemisphere may be uniquely specialized for the perception, experience, and expression of emotion. Theories to account for this hemispheric asymmetry are reviewed and a right-hemisphere dominance model is proposed. Future studies are outlined to further our understanding of the neuroanatomic basis of emotion. The recognition that the right hemisphere may be specialized for mediating emotion may have clinical implications.

Electroconvulsive therapy of depression in patients with white matter hyperintensity

Magnetic resonance imaging (MRI) uses radiofrequency radiation in the presence of a magnetic field to create cross-sectional images of the body. Among its many clinical applications, brain MRI has been shown to be an especially sensitive technique for diseases of the white matter, e.g. multiple sclerosis (MS), subcortical arteriosclerotic encephalopathy (SAE), radiation necrosis, etc. (Bradley et al. 1984; Zimmerman et al. 1986). In these diseases, the lesions typically appear as areas of high signal intensity on spinecho (n-weighted) images (Young et al. 1983; Bradley et al. 1984; Zimmerman et al. 1986). The lesions may be focal and/or diffuse and are commonly observed next to the ventricular lining. Some degree of white matter hyperintensity (WMH) may also be seen in patients who have no evidence of active cerebral dysfunction (Bradley et al. 1984). The clinical and pathophysiological significance of WMH in these patients is unclear. Recently, four patients were referred for electroconvulsive therapy (ECT) of major depression that had been refractory to drug therapy. Each was found to have WMH on brain MRI scans, raising the question of how such patients should be managed. We believe this to be the first report of ECT in a series of depressed patients with WMH.

The largest Lyapunov exponent of the EEG during ECT seizures as a measure of ECT seizure adequacy

Attributes of the electroencephalogram (EEG) recorded during electroconvulsive therapy (ECT) seizures appear promising for decreasing the uncertainty that exists about how to define a therapeutically adequate seizure. In the present report we study whether one promising and not yet tested ictal EEG measure, the largest Lyapunov exponent (lambda1), is useful in this regard. We calculated lambda1 from 2 channel ictal EEG data recorded in 25 depressed subjects who received right unilateral ECT. We studied the relationship of lambda1 to treatment therapeutic outcome and to an indirect measure of treatment therapeutic potency, the extent to which the stimulus intensity exceeds the seizure threshold. We found lambda1 could be reliably calculated from ictal EEG data and that the global mean, maximum, and standard deviation of lambda1 were smaller in the more therapeutically potent moderately suprathreshold ECT and in therapeutic responders. These results imply a more predictable or consistent pattern of EEG seizure activity over time in more therapeutically effective ECT seizures. These findings also suggest the promise of lambda1 as a marker of ECT seizure therapeutic adequacy and build on our previous work suggesting that lambda1 may be useful for classifying seizures and for reflecting the relative physiologic impact of seizure activity.

The dexamethasone suppresion test and quantitative cerebral anatomy in depression

To determine whether structural brain abnormalities in patients with depression are related to cortisol state, we examined the relationship between the dexamethasone suppression test (DST) and brain magnetic resonance imaging (MRI) in 40 inpatients with severe depression referred for electroconvulsive therapy (ECT). Prior to ECT, 27 (68%) of the patients exhibited nonsuppression on the DST. Frontal lobe volume was negatively correlated with peak post-dexamethasone cortisone (r = -0.37) and was 13% smaller in DST nonsuppressors than suppressors; these findings were no longer significant after adjustments for age, gender, and cranial size. Lateral and third ventricular volumes were also correlated with peak postdexamethasone cortisol (r = 0.34 and 0.33, respectively), but not after adjustments for age, gender, and cranial size. Subcortical hyperintensity was associated with peak postdexamethasone cortisol and was more common in DST nonsuppressors than suppressors. Again these findings were no longer significant after adjustments for age. Finally, longitudinal DST and brain MRI studies in 11 of these patients revealed no changes in regional brain volumes nor in postdexamethasone cortisol up to six months after ECT. However, within individual patients, postdexamethasone cortisol was positively (and significantly) correlated with frontal lobe volume.

Prediction of the utility of a switch from unilateral to bilateral ECT in the elderly using treatment 2 ictal EEG indices.

Background The choice of whether to administer nondominant unilateral (UL) or bilateral (BL) ECT remains controversial. Methods A study in which moderately suprathreshold UL nonresponders at treatment 6 were randomized to UL or BL ECT offered the opportunity to explore whether ictal EEG indices at treatment 2 might predict response to UL ECT, and also which UL ECT nonresponders are likely to respond to BL ECT. Results We found that less postictal suppression in response to the second UL ECT stimulus was predictive of a poorer subsequent therapeutic response to UL ECT, but of a better therapeutic response if switched to BL ECT. A multivariate ictal EEG model was developed that had a significant capacity to differentiate those who will respond to UL ECT versus those who will not respond to UL ECT, but who will be therapeutic responders when switched to BL ECT. Conclusions This study raises the possibility that ictal EEG indices at treatment 2 may identify situations when UL ECT is physiologically and therapeutically inadequate, and when BL ECT is likely to be more effective. The determination of whether such predictive physiologic models are of clinical utility for the prediction of outcome awaits further study.

Augmentation of ECT seizures with caffeine

Over a course of electroconvulsive therapy (ECT) there is frequently a progressive rise in seizure threshold and an associated reduction in the duration of the electrically induced seizure (Sackeim et al. 1986). The recognition of brief ECT-induced seizures is of clinical importance because they may not be as effective as those of moderte duration and because they may herald the occurrence of missed seizures at subsequent ECT treatments (Ottosson 1960; Milstein and Small 1984; Sackeim et al. 1986). Techniques to maintain adequate seizure duration over a course of ECT are therefore of considerable clinical relevance. Because of the paucity of applicable clinical data, the definition of “adequate seizure duration” remains unclear. The proposal of a rather arbitrary cutoff of 25 set (Weiner 1979; Fink and Johnson 1982) has received wide clinical acceptance. The typical clinical approach to maintaining seizures of 25 set or greater during a course of ECT is to increase the intensity of the ECT stimulus when seizure duration becomes too brief (Fink 1979; Weiner 1979). A higher stimulus energy, however, may be associated with greater encephalopathic side effects (confusion, amnesia, EEG slowing) and obviously cannot be accomplished when the ECT device is already at maximal settings (Fink et al. 1958; Ottosson 1960). Recently, we (Hinkle et al. 1987) described six depressed inpatients receiving ECT whose seizure durations were declining despite maximal settings on three different commercially available ECT instruments. In each case, we found that the use of caffeine administered intravenously prior to the delivery of the ECT stimulus resulted in a marked lengthening of the seizure duration (mean increase of 107%), and all patients subsequently experienced a clinical remission of their depression (Hinkle et al. 1987). These findings raised the possibility that caffeine pretreatment could be used routinely to limit increases in electrical energy delivered during a course of ECT. We now describe a new series of depressed patients receiving ECT in whom pretreatment with caffeine reversed a pattern of declining seizure duration and reduced the need for frequent increases in ECT stimulus over the course of therapy.

Effect of ECT treatment number on the ictal EEG

Recent evidence suggests that attributes of the ictal electroencephalogram (EEG) may be clinically useful for estimating the extent to which the electroconvulsive therapy (ECT) stimulus exceeds the seizure threshold (relative stimulus intensity). Such a tool could allow a practitioner, who chose, on the basis of expected therapeutic response and side effect rates, to implement stimulus dosing to maintain relative stimulus intensity over the treatment course, despite the uncertain rise in seizure threshold that occurs. One potential confounding factor is a possible systematic change in the ictal EEG over the treatment course that is not due to changes in seizure threshold. We explored the effect of treatment number by comparing ictal EEG data obtained at treatments across the ECT course that were delivered at the identical relative stimulus intensity. We found that the ictal EEG at treatment 1 was characterized by a greater mid-ictal amplitude and post-ictal suppression (trend) than subsequent treatments for barely suprathreshold unilateral ECT, but not for barely suprathreshold bilateral or moderately suprathreshold unilateral ECT, and that this change may affect therapeutic effectiveness. These findings suggest the importance of treatment-number effects for the clinical application of the ictal EEG and point to possible physiological differences between unilateral and bilateral ECT.