Electroconvulsive Therapy as an Anticonvulsant

Electroconvulsive Therapy as an Anticonvulsant

Implications for Its Mechanism of Action in Affective Illness

ROBERT M. POST, FRANK PUTNAM, THOMAS W. UHDE, AND SUSAN R. B. WEISS Biological Psychiatry Branch

National Institute of Mental Health Building 10, Room 3N212

9000 Rockville Pike Bethesda, Maryland 20205

INTRODUCI‘ION

The rationale for the introduction of the use of electroconvulsive therapy (ECT) in the treatment of major psychiatric illness contained the latent idea that is the major focus of this manuscript-that ECT may paradoxically be an effective anticonvulsant. Based on the observations of many investigators that there was an inverse relationship between seizures and psychosis, von Meduna introduced convulsive therapy as a potential treatment strategy for major psychiatric illnesses.’ The use of major motor seizures themselves as an effective treatment modality, particularly for the major affective disorders, obscured the possibility that the repeated seizures might be effective not because of the convulsions but because of their anticonvulsant effects. It has largely been forgotten that ECT has in fact been used to treat seizure disorders and some of their associated behavioral problems”’ (also S. Sato, personal communication, 1982). More recently, Sackeim et al. have again reported success in the use of ECT as an anticonvulsant for epileptic patients resistant to pharmacotherapy.6

In light of our experience that anticonvulsant treatments, particularly carbamaze- pine, are effective in the treatment of both manic and depressive episodes, we became increasingly interested in the pers ective that electroconvulsive shock (ECS) in animals is a potent anticonvu1sant.R Examining the anticonvulsant effects of ECS provides a new conceptual approach to the study of its possible mechanism of action. One can explore similarities and differences between the effects of ECS and psycho- tropic and anticonvulsant drugs effective in manic-depressive i l lne~s”~ on biochemis- try and behavior. In addition, the anticonvulsant effects of ECS are an easily measurable index of the efficacy of this treatment, so that the time course and mechanisms of effects can be delineated in relationship to biochemistry. Finally, one can use appropriate biochemical and pharmacological dissection to elucidate the possible factors that account for the anticonvulsant effects of ECS as a first step to assessing whether these factors then are important to its effects in affective disorders.

Thus, in this manuscript we review the available data on the anticonvulsant effects of ECS and begin a discussion of these findings and their mechanistic implications. We also briefly review our experience with the effects of more traditional anticonvulsants in manic-depressive illness and the comparative ability of these anticonvulsants to inhibit kindled seizures.

376

POST et al.: Em AS AN ANTICONVULSANT 377

THE ANTICONVULSANT EFFECTS OF ECS ON COMPLETED AMYGDALA-KINDLED SEIZURES

As summarized in TABLE 1, a substantial body of evidence suggests that single or multiple ECS in rats exerts substantial anticonvulsant effects against amygdala- kindled seizures. Babington reported that a single ECS administered up to 120 minutes prior to an amygdala-kindled seizure reduced seizure duration by more than 50% of control.’5 With longer intervals, the effect was less robust. Handforth indicated that multiple ECS administered 12 to 24 hours prior to amygdala-kindled seizures exerted long-lasting anticonvulsant effects for a period of 24-48 hours.I6 In our study, we observed that a single ECS followed by a six-day delay did not significantly affect amygdala-kindled seizures. However, a series of seven once-daily ECS significantly reduced amygdala-kindled seizure durations for up to five days following the termina- tion of ECS. These data are illustrated in FIGURE l . The top part of the figure indicates that afterdischarge duration in established amygdala-kindled seizures was not signifi- cantly affected by either real or sham ECS administered once or for seven consecutive days. In contrast, seizure stage and seizure duration were substantially reduced for five

TABLE 1. Effects of ECS on Completed Amygdala-Kindled Seizures Interval Type Seizures Duration

Babington & 15 minutes Single 1 1 1 WedekingI5 30 minutes Single 1 1 1

60 minutes Single 1 1 120 minutes Single 1 1 180 minutes Single (1) 240 minutes Single (1,

Urea & FrenkZ6+’ - - Handforth16 12-24 hours Multiple 1 1 1 24-48 hours Post et aI.’ 24 hours Single (6-day delay)

“Eighteen or twenty days of prior ECS had no effect on development of kindling.

1-5 days 1 - i 1 Multiple ( x 7)

days by the series of seven ECS. The fact that afterdischarge was not significantly affected indicates that the animals were not refractory to further excitatory effects of the amgydala-kindled stimulation, but that the prior chronic ECS treatment was significantly affecting the mechanisms and pathways involved in converting this excitatory process into a full-blown seizure.

The paradigm of one ECS followed by a six-day delay in comparison to seven chronic ECS was employed in order to assess the possible effects of dopamine autoreceptor desensitization as a potential mechanism for the effects of ECS on amygdala kindling. Chiodo and Antelman had reported that a single ECS followed by a seven-day delay was as effective as six chronic ECS with a two-day delay in producing dopamine autoreceptor subsensitivity,” a mechanism they postulated was important to the antidepressant effects of ECS and tricyclic antidepressants. The fact that a single ECS with a six-day delay in our paradigm was not sufficient to inhibit amygdala kindling suggests that dopamine autoreceptor desensitization is not impor- tant, at least to the anticonvulsant mechanisms of ECS.

Anticonvulsants such as carbamazepine, valproic acid, and diazepam are also

378

120-

110

loo z c

3 D

u-

I-

0 90-,

2 00-

w 70-

2: 60-•

4 40-

0 50- v,

y1 t 30- a

20

10

ANNALS NEW YORK ACADEMY OF SCIENCES

- -

CI

-

-

O L 0

AFTER-DISCHARGE DURATION

--SHAM X 1 *-oECS X 1 *-SHAM X 7 +-.ECS X 7

1 I I I I I I

1 2 3 4 5 6 7

SEIZURE STAGE

I I I I 1 I I I

1 2 3 4 5 6 7 8 9 10

0 . O h I l I I I 1 I I l l I I I I I I 1 I I

0 1 2 3 4 5 6 7 1 1 2 3 4 5 6 7 8 9 10 PRETREATMENT DAYS RE S U M PT ION 0 F, ,ApvLy D A LA KINDLING

FIGURE 1. Electroconvulsive seizures inhibit amygdala-kindled seizures without affecting afterdischarges. Animals were amygdala kindled until the first stage 3 or 4 seizure was observed. During the next seven days rats received only real or sham electroconvulsive seizures (ECS). Compared with sham ECT controls, seven daily ECS, but not a single ECS followed by a six-day delay, markedly suppressed kindled seizures when amygdala stimulation was resumed (p < 0,0001; F = 3.59; df = 19,92). Afterdischarge duration was not significantly affected. The suppressive effects of repeated ECS persisted for five days. On day five, kindled seizure stage was significantly reduced (p -= 0.04; t - -2.40; df = 9) as was seizure duration (p < 0.005; t =

-3.65).

POST er al.: ECT AS AN ANTICONVULSANT 379

highly effective in inhibiting seizure stage and seizure duration, while the afterdis- charge component of amygdala-kindled seizures is more difficult to suppress. As revealed in the studies of Albright and Burnham, carbamazepine is the most potent of the anticonvulsants in decreasing amygdala-kindled afterdischarge duration compared to that kindled from the cortex,’* suggesting that carbamazepine has the best ratio of effects on ability to suppress limbic compared to neocortical afterdischarge excitatory foci. Carbamazepine was able to decrease amygdala afterdischarge duration to 55% of that seen in the cortex, and was followed by the other anticonvulsants in decreasing order of efficacy: valproate (37%), phenobarbital (33%), phenytoin (30%), methsux- imide (25%), clonazepam (22%), ethosuximide ( 1 4%), and surprisingly last, diazepam (0%). While Ashton and Waugq~ier ’~ and many other studies reviewed elsewhere” also found potent anticonvulsant effects of carbamazepine (EDso of 5 mg/kg, intraperitoneally), Albertson et al. reported a different rank order of anticonvulsant efficacy to that reviewed above.20

In addition to the typical anticonvulsants, Horovitz2’ and Babington2’ reported that the tricyclic antidepressants nortriptyline, amitriptyline, and imipramine significantly suppressed amygdala-kindled seizures. Moreover, a dose of amitriptyline that was not effective acutely, after 14 days of treatment did reduce seizure duration, suggesting some parallelism to the lag in onset of antidepressant effects. Other investigators have replicated some aspects of these findings but not others. For example, Knobloch et al. found that the antidepressant amitriptyline acutely inhibited amygdala-kindled sei- zures better than those kindled from the cortex:’ neither amitriptyline nor imipramine showed this selective effect following subacute administration. Moreover, as observed by Babington, the antidepressant iprindole did not alter seizures kindled from either site.22 High doses of mianserin, when given acutely or subacutely, did show selective attenuation of amygdaloid seizures.

In contrast to the dose-related effects of imipramine on arngydala-kindled seizures, Lange er al. reported a biphasic effect of imipramine against ECS and pentylene tetrazol seizures in mice.“ This anticonvulsant effect at low doses and proconvulsant effect at high doses (like that of lidocaine) on “classical” motor seizure models could account for the clinical observations of seizures during tricyclic treatment, even in the face of dose-related anticonvulsant effects on the amygdala-kindled type.

It is noteworthy that Babington and Wedeking found that ECS in the rat produced a greater magnitude and duration of anticonvulsant effects on amygdala-kindled seizures compared to those produced in the septum or sensorimotor cortex?’ However, the ultimate utility of the amygdala kindling model for assessing potential antidepres- sant and antimanic effects of somatic treatments requires further study.

EFFECT OF ECS ON THE DEVELOPMENT OF KINDLING

The effects of ECS on the development of kindling are summarized in TABLE 2. Babington found profound effects of ECS administered ‘/2 hour or 20 hours prior to amygdala stimulation. In his series, he also found that ECS administered 24 hours prior to kindling was moderately effective in suppressing the development of seizures; afterdischarges were not measured. We found that ECS administered 6 hours prior to each once-daily amygdala-kindled stimulation using suprathreshold stimulation with 50 Hz biphasic square waves for one second duration completely inhibited the development of amygdala kindling, as illustrated in FIGURE 2. In contrast, ECS administered immediately after amygdala kindling (i.e., 24 hours prior to the next stimulation) was not effective under these conditions. These data are similar to those of Urca and Frenk, who found that ECS five minutes prior to kindling potently

380 ANNALS NEW YORK ACADEMY OF SCIENCES

suppressed the development of amygdala seizures, while a 24-hour interval was without effect.% However, Tsuru et aL2’ and Handforth16 did find that ECS adminis- tered 24 hours prior to kindling was effective in suppressing the development of seizures, at least to a mild to moderate extent.

As illustrated in FIGURE 3, when ECS treatment was terminated, amygdala kindling proceeded at a relatively normal rate. These data are consistent with those of Babington,22 indicating that the ECS appeared to suppress the development of amygdala-kindled seizures but that the process would proceed normally once the ECS was discontinued.

The highly potent effects of ECS in inhibiting the development of amygdala kindling raise the question of the relative efficacy of this approach to that achieved by more traditional anticonvulsant treatments. In fact, there are considerable data that ECS may be a more potent anticonvulsant for amygdala-kindled seizure development

TABLE 2. Effects of ECS on Development of Amygdala Kindling Interval between

ECS and Kindling Babington &

Wedekingl’

Post el a/.’

Post et a/.’ and

Tsuru et a1.l

Urca & FrenkZ6

0.5 hours 20 hours 24 hours

6 hours

24 hours (- 1 minute post)

24 hours (5 minutes post)

ECS alone (20 days) 5 minutes

(0.3 second post) 24 hours ECS alone ( 18 days) 22 hours 24 hours

(immediately post)

Etrect on After Discharges Seizures

Effect on

NA” 1 1 1 N A 1 1 1 N A 1 1 - 1 1 1

(1) 5 1 1 1 -

“NA means not ascertained.

in the rat than is the drug carbamazepine. While carbamazepine is extremely effective in inhibiting completed kindled seizures, and does inhibit the process or development of kindling in the cat and data from our laboratory3” and those of Wada et uI.,’~ Baltzer et al.,” and Albertson et ~ 1 . ’ ~ indicate that carbamazepine is not effective in inhibiting the development of kindling in the rat. These findings are consistent with others indicating that differential biochemical and pharmacological processes may underlie the development of kindled seizures compared to those maintaining completed kindled seizures. Moreover, there may be important species differences in the responsivity to different anticonvulsant agents. Nonetheless, at least in the rat, ECS would appear to be a more potent anticonvulsant than carbamazepine in the develop- ment phase of amygdala kindling. It is noteworthy that another agent used in the treatment of manic-depressive illness-lithium carbonate-is not effective in inhibit- ing either completed kindled seizures or their development?”

POST et al.: ECT AS AN ANTICONVULSANT

25

20-

15

10

5 -

0 -

381

-

- -

B

4’0

:I 70

- SEIZURE STAGE

AFKR-DISCHARGE DURATION u--a Shsm ECS 6 Hours Rior to Kindling

CI ECS Irnmdmety After Kindling P ._

6o *--+ ECS 6 Hours Rar to KKldliW

35[ SEIZURE DURATION 30

382 ANNALS NEW YORK ACADEMY OF SCIENCES

CLINICAL AND MECHANISTIC IMPLICATIONS OF ANTICONVULSANT EFFECTS OF ECS

Sangdee et al. reported that a kindling-like effect (i.e., proconvulsant) of ECS itself could be induced in mice.’3 Ramer and Pine1 indicated that the interval between ECS (one hour vs. three days) was important to whether seizure thresholds increased or decreased following ECS.3‘ Whether ECS treatment of patients can, in some instances, facilitate subsequent seizure development remains c~ntroversial.”~~ If seizure facilita- tion or sensitization does occur, it would appear to be a very uncommon outcome and of

w Sham ECS 6 hours prior to kindling

prior to kindling *--+ ECS 6 hours

30

25

10 r i

ECS Pretreatment n R i I \ I n

I

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 STIMULATION NUMBER

FIGURE 3. Electroconvulsive seizures (ECS) inhibit the development of amygdala kindling and suppress completed kindled seizures. ECS administered six hours prior to each amygdala stimulation blocks the development of kindling. However, when ECT treatment was discontinued on day 19, the remaining animals showed a relatively normal rate of kindling development. When active ECT was given to the previously sham-treated group on day 24, acute anticonvulsant effects of ECS were again evident. Data in the first two weeks parallel those in FIGURE 2 but include additional animals in each group which did not have histologically confirmed electrode placements.

unknown relationship to the clear anticonvulsant effects of ECS, at least on amygdala-kindled seizures.

Taken together, the studies summarized in TABLES 1 and 2 and FIGURFS 1-3 suggest that ECS exerts potent acute and long-term effects on amygdala-kindled seizures and their development. The anticonvulsant effects of electroconvulsive shock appear to be of a magnitude comparable to that produced by more typical anticonvul- sants. Albright and Burnham found that carbamazepine and valproic acid were the most effective in inhibiting the amygdala-kindled compared to cortical-kindled afterdischarge.” These two anticonvulsants appear to have the best efficacy in

POST et al.: ECT AS AN ANTICONVUISANT 383

providing acute and prophylactic treatment for bipolar di~order.’~.’~ These data raise many questions for further discussion. Are the anticonvulsant effects of these drugs and of ECS related to their clinical efficacy in the mood disorders? Elsewhere we have discussed whether or not the limbic mechanisms of action of carbamazepine are related to its clinical efficacy in manic-depressive illness.” Carbamazepine has multiple biochemical and potently suppresses excitability in many seizure models and in many areas of brain outside the limbic system that could also be important to its clinical effects.“ The potent anticonvulsant effects of ECS against amygdala kindling and possibly other seizures are provocative and suggest the importance of exploring possible mechanisms of this action of ECS in its own right and in relationship to its efficacy in affective illness.

There are numerous classical and peptide neurotransmitter candidates for the anticonvulsant effects of ECS.u945 As reviewed by Sackeim et aL6 and Green and collaborators in this volume, there is evidence that ECS, like many pharmacological anticonvulsants, may also increase y-aminobutyric acid (GABA) effects and decrease GABA turnover. Are these common effects also related to the psychotropic properties of these treatments? We suggest that this is a particularly important area of investigation. Based on the data that treatment with a variety of agents effective in manic-depressive illness (carbamazepine, lithium, valproic acid, progabide, and ECT) appears to increase GABA and the mounting evidence of GABA defects in affectively ill patients, we would propose this system as a candidate for the psychotropic effects of ECT. In a similar vein, one might ask whether the effects of ECS on catecholamine release and turnover are important to either its anticonvulsant or psychotropic properties, especially since noradrenergic mechanisms have been linked to the anticonvulsant effects of carbamazepine” and noradrenergic alternatives have long been hypothesized in the affective disorders.

Recent evidence suggests that ECS may transiently increase benzodiazepine receptors,M a phenomenon that could be important to its anticonvulsant and/or psychotropic properties, although other data suggest that these changes in benzodiaze- pine receptor binding may not persist for 24 hours after a series of ECS and thus may not be clinically rele~ant.4~ Moreover, the anticonvulsant effects of carbamazeprine, in contrast to those of diazepam, do not appear to involve the classical “central” benzodiazepine receptor site, but may involve the so-called peripheral benzodiazepine receptor site in brain, as is evidenced by the reversal of the anticonvulsant effects of carbamazepine by Ro5-4864.”

As reviewed in this volume (Fink; Holaday; Cowen) and elsewhere, there is evidence that ECS influences a variety of neural peptides including endogenous opiate substances,“*49 thyrotropin releasing hormone (TRH), adrenmrticotropin releasing hormone (ACTH), a-melanotropic stimulating hormone (cxMSH),~ prolactin?’ vaso- pressin, and, potentially, corticotropin releasing factor (CRF) and somatostatin. Each one of these peptide systems requires exploration as a possible mediator or modulator of anticonvulsant and/or psychotropic effects. The endogenous opiates are also candidates at least for the anticonvulsant effects of ECS.5’ Altered response to CRF has been closely linked to both affective illnesss3 and kindled s e i z u r ~ , ’ ~ further suggesting the possible importance of this peptide in the modulation of the endocrine and/or therapeutic responsivity following ECT.

A novel approach to the study of the mechanisms underlying the long-lasting anticonvulsant effects of ECS has been explored by Isaac and S ~ a g e r . 5 ~ These investigators observed that daily ECS for 22 days in the cat resulted in a progressive elevation of electroconvulsive threshold. This phenomenon lasted for approximately 20 days. Cerebrospinal fluid (CSF) collected from cats with these elevated thresholds, when transferred into the ventricular space of untreated animals, conferred an elevated

384 ANNAIS NEW YORK ACADEMY OF SCIENCES

threshold to the recipient animal. These data suggest that a substance present in CSF of cats receiving repeated ECS may confer the altered threshold to untreated animals. Long and Tortella have made similar observations in the rat that following ECS a substance is released into CSF that has anticonvulsant effects$* the effects are naloxone reversible, suggesting that it is an opiate peptide, possibly beta-endorphin.

Somatostatin may also deserve study as a potential mediator of the effects of ECS. The anticonvulsant carbamazepine decreases somatostatin in the CSF of affectively ill patient^,'^ and the depletion or inactivation of somatostatin (with cysteamine or antibodies, respectively) is anticonvulsant to kindled seizures.” Moreover, seizures induced by amygdala kindling produce long-lasting (at least two months) increases in brain somatostatin in the rat.’* These data suggest a possible role of somatostatin in anticonvulsant mechanisms, and six studies reporting decreases in somatostatin in the CSF of depressed patientssg suggest a possible role in affective illness.

In addition to the common effects on GABA turnover of three treatments (ECT, carbamazepine, and lithium) that are effective in both phases of manic-depressive illness, these agents also share common effects on thyroid. Surprisingly, ECT, carbamazepine, and lithium all decrease plasma thyroxine,a and this hormone thus remains a possible candidate for the psychotropic (but not anticonvulsant) effects of these agents.

Engel et al. reported that electroconvulsive treatment in patients produced diffuse increases in cerebral glucose metabolism and postictal decreases!’ These findings have been replicated in animal studies, although additional ECT patients have only shown the pattern of postictal decreases in glucose consumption!* This pattern of ECT in man and that reported by Blackwood et al. in the rat3’ are very different from the regional selective changes in glucose utilization produced by am~gdala-kindled~~ or lidocaine-kindledM seizures and could be important to the interpretation of the anticonvulsant effects of ECS.

As noted above, it is possible that the anticonvulsant and psychotropic effects of ECS and the anticonvulsants may be mediated through different mechanisms. This would considerably complicate the elucidation of the important mechanisms for the psychotropic effects of ECS, particularly since there is at present no clear index or end point measurable in animals for the psychotropic effect. The possibility of dissociating anticonvulsant, psychotropic, and side effects is highlighted by the finding that the anticonvulsant carbamazepine may reverse the amnesic effects of electroconvulsive shock in mice. Mondadori and Classen found that carbamazepine was the only anticonvulsant capable of reversing ECS-induced amnesia in mice.65 Indirect evidence suggests that carbamazepine may be acting as an agonist at vasopressin recep- t o r ~ . ~ * ~ ’ * ” It is possible that this unique action of carbamazepine among the anticonvul- sants accounts for its ability to counter ECS-induced amnesia. If this were borne out in further experiments it might suggest that ECS and carbamazepine have opposing effects on vasopressin function that are not directly related to their psychotropic properties, but may be closely related to their effects on cognitive processes.

CONCLUSIONS

Our conceptualization of ECS as an anticonvulsant treatment may facilitate studies of the mechanism of action of this treatment, especially since this provides an easily measurable effect of ECS. It may also help focus on the questions of why seizures stop and what endogenous anticonvulsant mechanisms are ECS enhancing. However, at this time it remains only a working hypothesis for further exploration that the anticonvulsant effects of ECS and the anticonvulsant agents are related to their

POST et al.: ECT AS AN ANTICONWLSANT 385

psychotropic efficacy in the treatment of manic and depressed patients. T h e biochemi- cal and physiological mechanisms underlying their anticonvulsant effects may be differentiable from those related to their therapeutic efficacy in affective illness. Moreover, even the demonstration that common mechanisms of ECS and other anticonvulsant agents were closely related to their psychotropic efficacy in affective illness would not imply that a seizure disorder was an underlying pathophysiological process in affective illness. Carbamazepine and other anticonvulsants are effective in the treatment of pain syndromes and a variety of other paroxysmal syndrome^'^ that do not appear to involve an ictal process. Nevertheless, it remains an intriguing proposi- tion that mechanisms common to ECS and anticonvulsants such as carbamazepine and valproic acid (and either similar to or different from the prototypic treatment for manic-depressive illness, i.e., lithium carbonate) are related to their efficacy in the treatment of both the manic and depressive phases of affective disorders.

REFERENCES

1. VON MEDUNA, L. J. 1935. Die konvulsionstherapie der schizophrenie. Psychiatr. Neurol.

2. KALINOWSKI, L. B. & F. KENNEDY. 1943. Observations in electric shock therapy applied to

3. CAPLAN, G. 1946. Electrical convulsion therapy in the treatment of epilepsy. J. Ment. Sci.

4. TAYLOR, J. H. 1946. Control of grand ma1 epilepsy with electreshock. Dis. New. Syst.

5. WOLFF, G. E. 1956. Electro-convulsive therapy: a help for epileptics. Am. Prac. Dig. Treat.

Wochenschr. 37: 317-319.

problems of epilepsy. J. Nerv. Ment. Dis. 9 8 56.

92: 784-793.

7: 284-285.

7: 1791-1793. 6.

7.

8.

9.

10.

11.

12.

13.

SACKEIM, H. A., P. DECINA, S. MALITZ, S. R. RESOR & I. PROHOVNIK. 1983. Anticonvul- sant and antidepressant properties of electroconvulsive therapy: a proposed mechanism of action. Biol. Psychiatry 18 1301-1310.

POST, R. M., F. W. PUTNAM & N. R. CONTEL. 1981. Electroconvulsive shock inhibits amygdala kindling. In Proceedings of Society for Neuroscience, 1 1 th Annual Meeting, Los Angeles, Calif. (Abstr. No. 187.13, page 587).

POST, R. M., F. PUTNAM, N. R. CONTEL & B. GOLDMAN. 1984. Electroconvulsive seizures inhibit amygdala kindling: implications for mechanisms of action in affective illness. Epilepsia 2 5 234-239.

BALLENGER, J. C. & R. M. POST. 1980. Carbamazepine (Tegretol) in manic-depressive illness: a new treatment. Am. J. Psychiatry 137: 782-790.

BALLENGER, J. C. & R. M. POST. 1978. Therapeutic effects of carbamazepine in affective illness: a preliminary report. Commun. Psychopharmacol. 2: 159-178.

POST, R. M., T. W. UHDE, J. C. BALLENGER, D. C. CHATTERJI, R. F. GREEN & W. E. BUNNEY, JR. 1983. Carbamazepine and its -10.11-epoxide metabolite in plasma and CSF relationship to antidepressant response. Arch. Gen. Psychiatry 40: 673-676.

POST, R. M., T. W. UHDE, J. C. BALLENCER & K. M. SQUILLACE. 1983. Prophylactic efficacy of carbamazepine in manic-depressive illness. Am. J. Psychiatry 140: 1602- 1604.

POST, R. M., J. C. BALLENGER, T. W. UHDE & W. E. BUNNEY, JR. 1984. Efficacy of carbamazepine in manic-depressive illness: implications for underlying mechanisms. In Neurobiology of Mood Disorders. R. M. Post & J. C. Ballenger, Eds.: 777-816. Williams & Wilkins. Baltimore, Md.

POST, R. M., T. W. UHDE, W. BERRE~TINI & C. KELLNER. 1984. Selective response to the anticonvulsant carbamazepine in manic-depressive illness: a case study. J. Clin. Psycho- pharmacol. 4 178-185.

BABINGTON, R. G. & P. W. WEDEKING. 1975. Blockade of tardive seizures in rats by electroconvulsive shock. Brain Res. 8 8 141-144.

HANDFORTH, A. 1982. Postseizure inhibition of kindled seizures by electroconvulsive shock. Exp. Neurol. 7 8 483-491.

14.

15.

16.

386 ANNALS NEW YORK ACADEMY OF SCIENCES

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

CHIOM), L. A. & S. M. ANTELMAN. 1980. Electroconvulsive shock: progressive dopamine autoreceptor subsensitivity independent of repeated treatment. Science 210 799-801.

ALBRIGHT, P. S. & W. M. BURNHAM. 1980. Development of a new pharmacological seizure model: effects of anticonvulsants on cortical- and amygdala-kindled seizures in the rat. Epilepsia 21: 68 1-689.

ASHTON, D. & A. WAUQUIER. 1979. Behavioral analysis of the effects of 15 anticonvulsants in the amygdaloid kindled rat. Psychopharmacology 6 5 7-1 3.

ALBERTSON, T. E.. S. L. PETERSON 8c L. G. STARK. 1980. Anticonvulsant drugs and their antagonism of kindled amygdaloid seizures in rats. Neuropharmacology 19 643-652.

HOROVITZ, Z. P. 1966. The amygdala and depression. I n Anti-Depressant Drugs. S. Garattini, M. Milan & N. G. Dukes, Eds.: 121-129. Excerpta Medica. Amsterdam, Holland.

BABINGTON, R. G. 1977. The pharmacology of kindling. I n Animals Models of Psychiatry and Neurology. I. Hanin & E. Usdin, Eds.: 141-149. Pergamon Press. Oxford & New York.

KNOBLOCH, L. C., J. M. GOLDSTEIN & J. B. MALICK. 1982. Effects of acute and subacute antidepressant treatment on kindled seizures in rats. Pharmacol. Biochem. Behav.

LANGE, S. C., R. M. JULIEN & G. W. FOWLER. 1976. Biphasic effects of imipramine in experimental models of epilepsy. Epilepsia 17: 183-196.

BABINGTON, R. G. & P. W. WEDEKING. 1973. The pharmacology of seizures induced by sensitization with low intensity brain stimulation. Pharmacol. Biochem. Behav. 1: 461- 467.

URCA, G. & H. FRENK. 1982. Electroconvulsive shock disrupts amygdaloid kindling: dissociation between behavioral and electrographic events. Exp. Neurol. 7 8 492-502.

TSURU, N., H. NINOMIYA, H. FUKUOKA & D. NAKAHARA. 1981. Alterationsof amygdaloid kindling phenomenon following repeated electroconvulsive shocks in rats. Folia Psychiatr. Neurol. Jpn. 3 5 167-174.

WADA, J. A., D. P. CAIN & E. JUNG. 1977. Pharmacological prophylaxis of amygdaloid kindling in Senegalese baboons. Electroencephalogr. Clin. Neurophysiol. 43(E108).

WADA, J. A., M. SATO, A. WAKE, J. R. GREEN & A. S. TROUPIN. 1976. Prophylactic effects of phenytoin, phenobarbital, and carbamazepine examined in kindled cat preparations. Arch. Neurol. 33: 426-434.

POST, R. M., S. R. B. WEISS & A. PERT. 1984. Differential effects of carbamazepine and lithium on sensitization and kindling. Prog. Neuropsychopharmacol. Biol. Psychiatry 8: 425-434.

BALTZER, V., K. KLEBS & M. SCHMUTZ. 1981. Effects of oxcarbazepine, a compound related to carbamazepine, and of GP47 779. its main metabolite in man, on the evolution of amygdaloid-kindled seizures in the rat. Paper presented at the Epilepsy International Congress, Kyoto, Japan, Sept. 17-21.

ALBERTSON, T. E., R. M. JOY & L. G. STARK. 1984. Carbamazepine. A pharmacological study in the kindling model of epilepsy. Neuropharmacology 23: 1 1 17-1 124.

SANGDEE, P., S. A. TURKANIS & R. KARLER. 1982. Kindling-like effect induced by repeated corneal electroshock in mice. Epilepsia 23 471-479.

RAMER, D. & J. P. J. PINEL. 1976. Progressive intensification of motor seizures produced by periodic electroconvulsive shock. Exp. Neurol. 51: 421-433.

DEVINSKY, 0. & M. S. DUCHOWNY. 1983. Seizures after convulsive therapy: a retrospective case survey. Neurology 33: 921-925.

SMALL, J. G., V. MILSTEIN, I. F. SMALL & P. H. SHARPLEY. 1981. Does ECT produce kindling? Biol. Psychiatry 16: 773-778.

BLACKWOOD, D. H. R., R. E. CULL, C. P. L. FREEMAN, J. I. EVANS & C. MAWDSLEY. 1980. A study of the incidence of epilepsy following ECT. J. Neurol. Neurosurg. Psychiatry

FINK, M. 1978. Efficacy and safety of induced seizures (EST) in man. Compr. Psychiatry

POST, R. M., T. W. UHDE, R. T. JOFFE, P. P. ROY-BYRNE & C. KELLNER. 1985. Anticonvulsants in psychiatric illness: new treatment alternatives and theoretical impli-

17: 461-465.

43: 1098-1 102.

19 1-18.

POST et al.: ECT AS AN ANTICONVULSANT 387

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50. 51.

52.

53.

54.

55.

56.

57.

58.

59.

cations. In The Neuropharmacology of Epilepsy. M. R. Trimble, Ed.: 141-171. John Wiley & Sons, Ltd. Chichester, England.

POST, R. M. & T. W. UHDE. Are the psychotropic effects of carbamazepine in manic- depressive illness mediated through the limbic system? Psychiatr. J. Univ. Ottawa lO(4). (In press.)

POST, R. M., T. W. UHDE, J. C. BALLENGER & W. E. BUNNEY, JR. 1982. Carbamazepine, temporal lobe epilepsy, and manic-depressive illness. Adv. Biol. Psychiatry 8 117-156.

POST, R. M., T. W. UHDE & J. C. BALLENGER. 1984. Efficacy of carbamazepine in affective disorders: implications for underlying physiological and biochemical substrates. In Anticonvulsants in Affective Disorders. H. M. Emrich, T. Okuma & A. A. Muller, Eds.: 93-1 15. Elsevier Science Publishers. Amsterdam, Holland.

POST, R. M., D. R. RUBINOW & T. W. UHDE. 1984. Biochemical mechanisms of action of carbamazepine in affective illness and epilepsy. Psychopharmacol. Bull. 20 585-589.

GRAHAME-SMITH, D. G. 1984. The neuropharmacological effects of electroconvulsive shock and their relationship to the therapeutic effect of electroconvulsive therapy in depression. I n Frontiers in Biochemical and Pharmacological Research in Depression. E. Usdin et a/., Eds.: 327-343. Raven Press. New York, N.Y.

LERER, B. 1984. Electroconvulsive shock and neurotransmitter receptors: implications for mechanism of action and adverse effects of electroconvulsive therapy. Biol. Psychiatry

PAUL, S. M. & P. J. SKOLNICK. 1978. Rapid changes in brain benzodiazepine receptors after experimental seizures. Science 202 892-894.

WEISS, S. R. B., R. M. POST, J. PATEL & P. S. MARANGOS. 1985. Differential mediation of the anticonvulsant effects of carbamazepine and diazepam. Life Sci. 36 2413-2419.

BELENKY, G. L. 1983. Electroconvulsive shock and endogenous opioids. Abstr. Am. Psychiatr. Assoc. No. 103.

HOLADAY, J. W., F. C. TORTELLA, J. B. LONG, G. L. BELENKY & R. J. HITZEMAN. 1986. Endogenous opioids and their receptors: evidence for involvement in the postictal effects of electroconvulsive shock. Ann. N.Y. Acad. Sci. (This volume.)

1 9 361-383.

VARGAS, L. 1983. Hormone patterns after electroconvulsive therapy. Lancet 1: 365. WHALLEY, L. J., H. DICK, A. G. WATTS, J. E. CHRISTIE, R. ROUE, G. LEVY, W. J.

SHEWARD & G. FINK. 1982. Immediate increases in plasma prolactin and neurophysin but not other hormones after electroconvulsive therapy. Lancet 2: 1065-1068.

LONG, J. B. & F. C. TORTELLA. 1984. Cerebrospinal fluid from convulsed rats causes a naloxone-reversible increase in the seizure threshold of recipient animals: regulation of postseizure inhibition by endogenous opioid systems. SOC. Neurosci. Abstr. (145th Annual Meeting, Anaheim, Calif., Oct. 10-15, Abstract No. 273.6, page 929).

GOLD, P. W., G. P. CHROUSOS. C. K. KELLNER, R. M. POST, A. ROY, P. AVGERINOS, H. M. SCHULTE, E. OLDFIELD & D. L. LORIAUX. 1984. Psychiatric implications of basic and clinical studies with corticotropin-releasing factor. Am. J. Psychiatry 141: 619-227.

WEISS, S. R. B., R. M. POST, P. W. GOLD, G. CHROUSOS, T. L. SULLIVAN, D. WALKER & A. PERT. 1984. CRF-induced aggressive behavior and seizures: facilitation of amygdala kindling. Soc. Neurosci. Abstr. (145th Meeting, Anaheim, Calif., Oct. 1CL15, Abstract No. 55.9, page 178).

ISAAC, L. & J. SWANGER. 1983. Alteration of electroconvulsive threshold by cerebrospinal fluid from cats tolerant to electroconvulsive shock. Life Sci. 33 2301-2304.

S. REICHLIN. 1983. Somatostatin in affective illness. Arch. Gen. Psychiatry 4 0 403- 412.

HIGUCHI, T., G. S . SIKAND, N. KATO, J. A. WADA & H. G. FRIESEN. 1983. Profound suppression of kindled seizures by cysteamine: possible role of somatostatin to kindled seizures. Brain Res. 288 359-362.

KATO, N., T. HIGUCHI, H. G. FRIESEN & J. A. WADA. 1983. Changes of immunoreactive somatostatin and beta-endorphin content in rat brain after amygdala kindling. Life Sci.

POST, R. M., P. W. GOLD & D. R. RUBINOW. Neuropeptides in manic-depressive illness. I n

RUBINOW, D. R., P. W. GOLD, R. M. POST, J. C. BALLENGER, R. COWDRY, J. BOLLINGER &

32 2415-2422.

%a ANNALS NEW YORK ACADEMY OF SCIENCES

60.

61.

62.

63.

64.

65.

66.

61.

Neuropeptides in Psychiatric and Neurological Disease. C. B. Nemeroff, Ed. Johns Hopkins University Press. Baltimore, Md. (In press.)

JOFFE, R. T., P. P. ROY-BYRNE, T. W. UHDE & R. M. POST. 1984. Thyroid function and affective illness: a reappraisal. Biol. Psychiatry 19 1685-1691.

ENGEL, J., D. E. KUHL & M. E. PHELFS. 1982. Patterns of human local cerebral glucose metabolism during epileptic seizures. Science 21 8: 64-66.

ACKERMANN, R. F., J . ENGEL, JR. & L. BAXTER. 1986. Positron emission tomography and autoradiographic studies of glucose utilization following electroconvulsive seizures in humans and rats. Ann. N.Y. Acad. Sci. (This volume.)

ENGEL, J., JR., L. WOLFSON & L. BROWN. 1978. Anatomical correlates of electrical and behavioral events related to amygdaloid kindling. Ann. Neurol. 3 538-544.

POST, R. M., C. KENNEDY, M. SHINOHARA, K. SQUILLACE, M. M~YAOKA, S . SUDA, D. H. INGVAR & L. SOKOLOFF. 1984. Metabolic and behavioral consequences of lidocaine- kindled seizures. Brain Res. 3U: 295-304.

MONDADORI, C. & W. CLASSEN. 1984. The effects of various antiepileptic drugs on E-shock-induced amnesia in mice: dissociability of effects on convulsions and effects on memory. Acta Neurol. Scand. 9 9 125-129.

BERRETTINI, W., R. M. POST, E. K. WORTHINGTON & J. B. CASPER. 1982. Human platelet vasopressin receptors. Life Sci. 30: 425432.

GOLD, P. W., G. L. ROBERTSON, J. C. BALLENGER, W. KAYE, J. CHEN, D. R. RUBINOW, F. K. GOODWIN & R. M. POST. 1983. Carbamazepine diminishes the sensitivity of the plasma arginine vasopressin response to osmotic stimulation. J. Clin. Endocrinol. Metab. 57: 952-951.