Journal of Affective Disorders 120 (2010) 62–66

Contents lists available at ScienceDirect

Journal of Affective Disorders

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j a d

Research report

The effectiveness of continuation-maintenance ECT in reducing depressed

older patients' hospital re-admissions
Daniel W. O'Connor a,b,⁎, Betina Gardner a, Ian Presnell a,b, Dhiren Singh a,b,
Maria Tsanglis c, Erica White d
a
Department of Psychological Medicine, Monash University, Melbourne, Australia
b
Southern Health, Melbourne, Australia
c
Caulfield General Medical Centre, Melbourne, Australia
d
Barwon Health, Geelong, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Background: We report on the outcomes in aged patients with severe, treatment-resistant
Received 16 December 2008 depression or psychosis who were given ongoing outpatient continuation-maintenance ECT of
Received in revised form 7 April 2009 varying duration to prevent remission and relapse following a successful course of acute ECT.
Accepted 7 April 2009 Methods: A retrospective chart review of 58 consecutive patients of three Australian aged
Available online 2 May 2009
psychiatry services comparing the number and length of psychiatric admissions before and
after the start of continuation-maintenance ECT.
Keywords:
Results: Four patients had only one treatment and two received over 50 (mean 14.7). Five were
ECT
Aged
still enrolled in a maintenance program two years later. In the two years after continuation-
Depression maintenance ECT started, admissions fell by 53% in number and 79% in duration compared with
Treatment outcome the previous two years. Within the actual treatment period which varied from one patient to
Hospitalization another, admissions fell by 90% in number and 97% in duration compared with the same period
beforehand.
Conclusion: A treatment effect cannot be proven but the severity and chronicity of patients'
conditions make placebo effects and spontaneous remission unlikely. Randomised, controlled
trials are almost impossible in this setting and so carefully conducted reviews and case–control
studies are still of value. Our findings suggest that continuation-maintenance ECT is effective in
carefully selected patients at high risk of relapse.
© 2009 Elsevier B.V. All rights reserved.

Electroconvulsive therapy (ECT) is a safe, effective treat-
ment of depression, mania and catatonia (O'Connor, 2008).
Relapse rates are high, however, once treatment stops. In a
recent study, over 80% of patients whose major depression
remitted successfully with ECT had relapsed within six
months when left without active therapy (Sackeim et al.,
2001). Some form of post-acute, prophylactic treatment is
essential, therefore, if patients are to remain well. Psycho-
tropic medications are simpler and cheaper to administer
than ECT on a ongoing basis but patients who respond
⁎ Corresponding author. Kingston Centre, Warrigal Road, Cheltenham
3192, Victoria, Australia. Tel.: +61 3 9265 1700; fax: +61 3 9265 1711.
E-mail address: Daniel.OConnor@med.monash.edu.au (D.W. O'Connor).
especially well to ECT can be offered further treatments on an
outpatient basis, as can those who fail to respond to
pharmacotherapy or cannot tolerate it (American Psychiatric
Association, 2001).
One approach is to reduce the frequency of ECT after an
acute course, lengthening the intervals between treatments
from weekly to fortnightly then monthly before stopping (so-
called “continuation” ECT). Patients who remain at very high
risk might then be offered “maintenance”, typically monthly,
treatments for an indefinite period, extending to years in
some cases. Continuation ECT aims to prevent relapse.
Maintenance ECT, defined arbitrarily as treatment extending
beyond six months, is intended to prevent recurrence
(American Psychiatric Association, 2001).

0165-0327/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jad.2009.04.005

Downloaded from ClinicalKey.com at Hospital de la Paz December 22, 2016.

For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
 D.W. O'Connor et al. / Journal of Affective Disorders 120 (2010) 62–66
ECT has been deployed in this way virtually from its
inception. Case reports spanning six decades give compelling
accounts of profoundly ill, medication-resistant patients who
were restored to health by maintenance ECT when other
treatments had failed, permitting discharge from long-stay
wards and a return to normal function (Moore, 1943; Karliner
and Wehrheim, 1965; Stiebel, 1995; Fox, 2001). Case reports,
while persuasive, must be set in context. Clinicians are keener
to report successes than failures, leaving open the possibility
that the patients described in published accounts were
atypical and that outcomes were often less positive. Prospec-
tive trials comparing ongoing ECT with other treatments are
less subject to bias but patients who genuinely warrant such
treatment and agree to receive it are few in number, even in
specialist centres (Sackeim, 1994).
The three trials conducted to date have varying relevance
to clinicians. In one, 50 patients treated successfully with ECT
for melancholia were randomly assigned to “sufficient”
treatments, as judged by the treating psychiatrist, or two
extra (continuation) ones. Those in the “extra” treatment
group were no less likely than others to have relapsed
12 weeks later (Barton et al., 1973). In the second study, 210
patients with major depression were assigned randomly post-
ECT either to 10 continuing treatments or to a combination of
nortriptyline and lithium carbonate. Relapse rates over a six-
month period were high in both groups at 37% and 32%
respectively, with no significant difference between them
(Kellner et al., 2006). Finally, 33 aged patients whose
depressive psychoses had remitted with ECT were rando-
mised to nortriptyline with or without weekly, then monthly,
ECT. Over a two-year period, those given ECT had a
significantly longer mean time to relapse of 23 months
compared with 16 months in the medication-only group
(Navarro et al., 2008).
These findings suggest that two extra “continuation”
treatments are too few to impact on relapse rates, and that
longer-term maintenance ECT performs no better than
pharmacotherapy, in unselected clinical series (Barton et al.,
1973; Kellner et al., 2006). It is difficult to know what to make
of these observations, though, since continuation, and
especially maintenance, ECT are never employed routinely,
being held in reserve for high risk patients when other,
simpler options have proved unsuccessful or cannot be
tolerated. By contrast, maintenance ECT did perform better
than medication for psychotically depressed older patients
whose risk of relapse is known to be higher than average
(Navarro et al., 2008).
Chart reviews, while not as rigorous as controlled trials,
can shed light on the costs and benefits of ongoing ECT in this
small, special group. One type of review contrasts the
progress of patients before and after treatment. The second
compares patients given ongoing ECT with others who were
not. Neither methodology is perfect. Patients change with
time; other treatments cannot always be held constant, and
groups are rarely comparable in every respect.
These concerns notwithstanding, the findings reported to
date are impressive. In four audits of maintenance ECT in
which pre-post periods were roughly equal, admissions fell
from an average of 2.0 per year to 0.2 in the first; from
22.7 weeks per year to 3.5 weeks in the second, and from
18.9 days to 3.2 days in the third (Thornton et al., 1990;
Downloaded from ClinicalKey.com at Hospital de la Paz December 22, 2016.
For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
63
Vanelle et al., 1994; Russell et al., 2003). The days spent in
hospital were halved in the fourth (Lim, 2006). Patient
numbers were small, ranging from 10 to 43, and diagnoses
were mixed.
Two other audits employed a retrospective case–control
design. In the first, 29 chronically depressed patients treated
with maintenance ECT for an average of 3.9 years were
compared with patients treated with long-term antidepressant
medication alone. Groups were matched later for age, gender,
diagnosis, age of onset, psychosis and year of treatment. All had
been chronically depressed and responded well to acute ECT. At
five-year follow-up, the cumulative probability of surviving
without relapse or recurrence was 73% for maintenance ECTand
18% for medication (Gagné et al., 2000). In the second study, 21
patients given maintenance ECT were compared with 21 who
refused it. Groups were matched for age, sex, diagnosis and
duration of illness. One year later, 37% of those given ECT had
been re-admitted to hospital compared with 67% of controls
(Swoboda et al., 2001).
The naturalistic, retrospective file review outlined below
adds to this limited literature. It is the largest review reported
to date and will therefore be of use in meta-analyses. It is also
just the second to focus on aged patients. ECT application
rates rise steeply with age in Australia, making analyses of
age-specific outcomes of interest to clinicians (Wood and
Burgess, 2003). In addition, most earlier reports are one to
two decades old and the patients selected for ongoing ECT
may have changed with the advent of better tolerated
antidepressant and antipsychotic medications (American
Psychiatric Association, 2001). Based on the current litera-
ture, we hypothesised that the number and length of
admissions to hospital for treatment of depression would be
significantly lower in the two years after ongoing ECT was
commenced than during the two years prior and, more
specifically, that community-resident, physically robust
patients would benefit more than those who were physically
frailer and living in aged residential facilities (O'Connor,
2008).
1. Methods
1.1. Data collection and analysis
We searched for the records of all patients given post-
acute ECT between 2000 and 2006 in a convenience sample of
three public aged psychiatry services in Victoria, Australia.
The two metropolitan services had dedicated inpatient beds.
The single regional service had limited access to beds but
could “admit” patients when required to the care of a small
dedicated, intensive community treatment team (as distinct
from its larger, general community team) for clearly desig-
nated periods prior to “discharge” to normal community care.
These admissions typically entailed daily visits from a mental
health nurse and weekly contact with a psychiatrist. The term
“admission” here encompasses both inpatient and specialist
community episodes of care. The 2000–2006 timeframe was
selected to ensure access to files in the two years before and
after this period.
ECT records were scrutinized, looking for staggered and
especially outpatient treatments. For audit purposes, con-
tinuation-maintenance ECT was defined as treatment that
64 D.W. O'Connor et al. / Journal of Affective Disorders 120 (2010) 62–66

followed an acute (index) course with the object of prevent- insurance. As a result, the depressed patients given ECT are
ing relapse and/or recurrence. In Australia, psychiatrists tend typically profoundly disabled with high rates of psychosis,
to refer to all ongoing, typically ambulatory, treatments as psychomotor retardation, refusal of food and fluids, and high
“maintenance ECT”. The term “continuation ECT”, referring to suicidality. Many of them have long, complex psychiatric
post-acute treatment not extending beyond six months, is not histories with large numbers of hospital files.
in common usage. We use the term “continuation-main- Treatments were administered by credentialed consultant
tenance” ECT to cover both approaches. No minimum number psychiatrists using Thymatron-IV square wave, brief pulse
of treatments was stipulated. machines with ictal EEG recording. Decisions regarding dosing,
The audit record covered patients' personal details, psychia- laterality, spacing and duration were made on a case-by-case
tric and medical history, ECT treatments, adverse events, the basis by psychiatrists, all of whom had attended an identical
reasons for stopping ECT, patient and carer attitudes, and three-day training program with clearly articulated, published
psychiatric admissions for the two years before and after the protocols (Tiller and Lyndon, 2003). Treatment standards were
start of continuation-maintenance ECT. Admission and discharge in line with current best practice with an emphasis on dose
dates were taken from a state-wide computerised database. Since titration, supra-threshold stimulation and regular clinical
histories were invariably long and complex, we focussed on data review with adjustment of energy levels as required. Continua-
that were certain to be recorded (e.g. diagnosis, current tion ECT regimes typically started with weekly treatments,
medications), avoiding those that might not be present in every moving later to fortnightly then monthly intervals as deter-
instance (e.g. detailed reasons for prescribing continuation- mined by patients' symptoms. Most longer-term maintenance
maintenance ECT), or were difficult to capture reliably given the treatments varied from fortnightly to monthly. All treatments
complexity of patients' histories (e.g. numbers of previous were prescribed, monitored and often administered by the
medication trials), or might have changed from day to day (e.g. patient's usual psychiatrists. There was no specialist ECT team
doses of medications). Data were abstracted by psychiatrists who or program and so patients receiving ongoing ECT accessed no
rated the benefits of acute ECT and patients' and carers' attitudes more time and care than that described here.
to ECT based on written observations. Since these and other
ratings were subject to bias, six records were co-rated by another 2. Results
psychiatrist to gauge inter-rater reliability.
The choice of outcome measure was limited. At the time of Continuation-maintenance ECT was administered on one or
audit, none of the participating services routinely measured more occasions to 54 depressed, aged patients over a six-year
mood, cognition or function before and after treatment, leaving period, representing about one fifth of all patients given acute ECT
admissions to psychiatric wards or intensive community in that period. All continuation-maintenance treatments fol-
treatment teams as the sole marker of treatment success or lowed an acute course (mean 11.8 treatments, range 3 to 33) that
failure. was rated as beneficial in every instance. One patient who lived in
We present data for the two years before and after the start a specialist psychogeriatric nursing home was excluded from the
of continuation-maintenance ECT. This lengthy timeframe following analyses on the grounds that her care needs were so
might prove meaningless for patients given just a few well addressed by a consultant psychiatrist and mental health
treatments and so we also compare admissions for the actual nurses that re-admission to hospital was unlikely, irrespective of
period over which ongoing ECT was administered with exactly ongoing ECT. The remaining 53 patients' characteristics are listed
the same number of days beforehand. We label this (usually in Table 1. Most were community-resident older women with
shorter) period the “treatment comparison period”. In the case chronic, relapsing unipolar depression. Nearly all had prior
of a patient given continuation ECT over a span of 120 days, for experience of acute ECT and had multiple medical co-morbidities.
example, the “comparison period” refers to the 120 days One fifth of patients were given continuation-maintenance ECT
immediately prior. on an involuntary basis (in contrast to 45% of acute courses).
Data analysis was compromised by a steep skew in the
number and length of admissions that failed to respond to
Table 1
trimming and transformation. We relied, therefore, on non- Patient characteristics at start of continuation-maintenance ECT.
parametric statistical tests including Mann–Whitney and Wil-
Gender Female 74%
coxon Signed Ranked tests, using exact significance levels in view
Age (mean, range) Start C/M-ECT 75.9
of the small sample size. Documented variables that we thought (65–92)
might impact on admissions included age (65–74, ≥75), Accommodation Own home 70%
accommodation (community; aged care facility), number of Residential facility 30%
medical diagnoses (0–3, 4+) and number of daily medical Diagnosis Recurrent major depression 93%
Bipolar disorder, 7%
medications (0–2, 3+). Inter-rater agreement was measured
currently depressed
using Cohen's kappa which ranges from one (complete agree- Additional diagnoses Dementia 8%
ment) to zero (agreement no better than chance). The mean Psychiatric history Age first psychiatric admission 59.1
kappa level across 23 audited items was 0.79. (mean, range) (20–83)
Acute ECT, prior to index course 79%
Age first ECT (mean, range) 63.8
1.2. Patients and treatment (23–88)
Previous C/M-ECT 9%
Patients are admitted to Victorian public aged psychiatry Legal status Involuntary 21%
units, either voluntarily or under the state's Mental Health Act, Physical health Medical diagnoses (mean, range) 2.8 (0–9)
Medical medications (mean, range) 3.8 (0–14)
if they have complex care needs and/or lack private health

Downloaded from ClinicalKey.com at Hospital de la Paz December 22, 2016.

For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
 D.W. O'Connor et al. / Journal of Affective Disorders 120 (2010) 62–66
At the outset of continuation-maintenance treatment, 34 (64%)
also took a single antidepressant medication and 9 (17%) took
two. Twenty-six (49%) took an antipsychotic medication with
or without an antidepressant. When this treatment (or the
study period) ended, prescription rates were similar: 32 (60%)
took a single antidepressant and 12 (23%) took two. Twenty-
three (43%) took an antipsychotic with or without an
antidepressant. The antidepressants were almost exclusively
serotonin or noradrenergic reuptake inhibitors or novel mixed
action agents. A tricyclic antidepressant and an irreversible
monoamine oxidase inhibitor were prescribed just to one
patient each. Similarly, most antipsychotics (85%) were
“atypicals”. Medications were sometimes adjusted as ECT
proceeded. With respect to antidepressants, a medication or
combination of medications was added in seven (13%) cases,
stopped in two (4%) and changed from one to another, or from
one combination to another, in six (11%). With respect to
antipsychotics, a medication or combination of medications
was added in five (9%) cases, stopped in nine (17%) and changed
from one to another, or from one combination to another, in
three (6%). Exact doses were not recorded but auditors
commented that most were standard and none were heroic.
2.1. Admissions
Four of the 53 patients had only one post-acute treatment
and two received more than 50 (mean 14.1) in the two-year
observation period. Four were still engaged in a maintenance
program two years later. For the remainder, reasons for
stopping ECT included sustained clinical improvement
(n = 25), lack of benefit (n = 9), patient refusal (n = 5),
intercurrent medical illness (n = 2), an adverse event (n = 3),
death unrelated to ECT (n = 1) and “other” factors (n = 4).
The numbers of psychiatric admissions and occupied bed
days in the two years before and after the start of continuation-
maintenance ECT, as well as the actual treatment comparison
periods, are shown in Table 2. The differences were striking and
statistically significant. In the two years after continuation-
maintenance ECT started, admissions were halved in number
and quartered in duration. Within the treatment comparison
periods, admissions were almost abolished. From the raters'
perspective, 34% of patients derived great benefit from
continuation-maintenance treatment, 43% derived some ben-
efit, 10% gained little or no benefit and 13% could not be
classified.
The two-year reduction in occupied bed days was greatest
for patients aged less than 75 years (72.9 days versus 64.8 for
those aged ≥75); the residents of aged care facilities (110.4 days
Table 2
Mean numbers (± SD) of psychiatric admissions and bed days before and
after continuation-maintenance ECT.
Period Before After Wilcoxon signed
starting rank Z value
Two year Admissions 1.9 (1.6) 0.9 (1.3) − 3.778***
period Admission days 87.8 (91.9) 19.9 (30.8) − 5.138***
Treatment Admissions 1.0 (0.8) 0.1 (0.3) − 5.760***
comparison Admission days 35.9 (36.1) 1.4 (8.0) − 5.401***
period a
***p b 0.0005.
a
Excluding 4 cases with only 1 C-ECT.
Downloaded from ClinicalKey.com at Hospital de la Paz December 22, 2016.
For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
65
versus 49.5 for community residents); those with three or more
medical diagnoses (83.9 days versus 49.8 for those with fewer
diagnoses), and those taking four of more medical medications
(78.1 days versus 58.7 days for those on fewer medications) but
none of these differences proved statistically significant.
2.2. Adverse events
Raters counted the adverse consequences of ECT that
warranted action by medical or nursing staff. Most patients
had no recorded adverse event; 11 had one and two had 2. Of
these 15 adverse events, there were eight instances of
memory loss or confusion; three of changed heart speed or
rhythm; three of hypo- or hyper-tension, and one of dental
injury. All resolved safely.
2.3. Patient and carer attitudes
Patients' attitudes to continuation-maintenance ECT were
often not documented. Ten of 25 (40%) patients whose
opinions were documented in the clinical file were rated as
“mostly positive” on a simple, three-point scale at treatment
outset and nine of 28 (32%) some time later. Seven (28%)
were rated as “mostly negative” at treatment outset and nine
(32%) later. Among carers, 11 of 14 (79%) were mostly positive
at outset and 10 of 15 (66%) felt similarly later.
3. Discussion
Continuation-maintenance ECT appears to be highly
effective in reducing and shortening psychiatric admissions
in older people with severe affective illness. No records were
found of serious, persistent adverse effects. We cannot prove
causation, however, as the study was neither blinded nor
controlled and so alternative explanations must be consid-
ered. These include placebo response, spontaneous remission,
certainty of treatment and concomitant therapy.
Placebo response refers to improvement in mood due to the
non-specific effects of participating in a treatment program
through contact with trusted clinicians, faith in medical
paraphernalia or reliance on externally-mediated therapies.
Substantial placebo responses are the norm in controlled trials
of antidepressant medications, especially in people with mild
depression of recent onset. Those with severe depression,
psychomotor retardation and psychotic symptoms are much
less likely to respond in this way (Dworkin et al., 2005). Nearly a
quarter of the patients in our review whose views were
documented were not well disposed to ECT, and a quarter were
unable or unwilling to consent to it, reducing further but not
excluding the possibility of non-specific positive benefits.
Depression is often chronic in older people, many of whom
have had previous episodes and face chronic stressors, most
notably loneliness, pain and physical disability (Baldwin, 2008).
Mood disorders fluctuate nonetheless and improvement might
still occur spontaneously, even after lengthy periods of incapacity.
Since this is also the time when ECT is most likely to be deployed
as an intervention of last resort, treatment and recovery might
coincide by chance. If this is true, and ECT offers no real additional
protection, relapse rates and re-admissions to hospital are
unlikely to fall significantly over a period as long as two years
as happened in our sample.
66 D.W. O'Connor et al. / Journal of Affective Disorders 120 (2010) 62–66
Ongoing ECT might also have worked, not because it is
better than medication, but simply because it was delivered
reliably to patients whose capacity to take tablets reliably was
compromised by depression, psychosis or cognitive impair-
ment. This is not likely to have been a major factor, however,
since checks of medication compliance were part of normal
clinical practice. Another factor might be that clinicians
continued to test new antidepressant, antipsychotic or adjunc-
tive medications with eventual success but this in itself is
an unlikely explanation for such a dramatic decline in re-
admissions since three-quarters of medication regimes
remained constant.
Randomised, controlled trials help address all these issues
but, for the reasons outlined already, a trial of continuation-
maintenance ECT in patients as disabled as the ones in our study
is almost inconceivable. We cannot prove beyond doubt that
continuation-maintenance ECT worked better than carefully
tailored combinations of psychological and pharmacological
therapies but there is good evidence that acute ECT is an effective
treatment of severe depression, especially when accompanied by
psychomotor retardation and psychosis, and there is no reason to
believe that ongoing treatment is any less helpful (O'Connor,
2008).
We had anticipated that younger, physically fit, community-
resident patients would derive most benefit from ongoing ECT on
the grounds that physical and cognitive frailty might limit
responses to any antidepressant treatment. In fact, the greatest
reductions in occupied bed days were seen in residents of aged
residential facilities and in patients taking larger than average
numbers of medical medications. These differences, while not
statistically significant, were striking nonetheless. Our findings
suggest at the least that frail older people are still candidates for
continuation-maintenance ECT provided that basic safety criteria
are satisfied.
Our positive observations are entirely consistent with earlier
studies, suggesting that continuation-maintenance ECT still has a
useful role to play in modern mental health services. It is an
expensive, time-consuming treatment that will, of necessity, be
reserved for a select minority of patients but it offers clinicians an
extra tool in challenging circumstances. It is not innocuous and
some patients dislike it. It is important, therefore, to ensure that
patients are chosen thoughtfully, that outcomes are monitored
diligently and that treatment is stopped when it looks safe to do
so.
Role of funding source
Funding for this study was provided by the John Cockayne Research
Fund. The Fund played no further role in the study design; in the collection,
analysis and interpretation of data; in the writing of the report, and in the
decision to submit the paper for publication.
Downloaded from ClinicalKey.com at Hospital de la Paz December 22, 2016.
For personal use only. No other uses without permission. Copyright ©2016. Elsevier Inc. All rights reserved.
Conflict of interest
No conflict declared.
References
American Psychiatric Association, 2001. Treatment following the completion
of the index electroconvulsive therapy course. The Practice of Electro-
convulsive Therapy. APA, Arlington, pp. 205–216.
Baldwin, R., 2008. Mood disorders: depressive disorders. Oxford Textbook of
Old Age Psychiatry. Oxford University Press, Oxford, pp. 529–556.
Barton, J.L., Mehta, S., Snaith, R.P., 1973. The prophylactic value of extra ECT in
depressive illness. Acta Psychiatr. Scand. 49, 386–392.
Dworkin, R.H., Katz, J., Gitlin, M.J., 2005. Placebo response in clinical trials of
depression and its implications for research on chronic neuropathic pain.
Neurology 65 (Suppl 4), 7–19.
Fox, H.A., 2001. Extended continuation and maintenance ECT for long-lasting
episodes of major depression. J. ECT 17, 60–64.
Gagné, G.G., Furman, M.J., Carpenter, L.L., Price, L.H., 2000. Efficacy of continuation
ECT and antidepressant drugs compared to long-term antidepressants alone
in depressed patients. Am. J. Psychiatry 157, 1960–1965.
Karliner, W., Wehrheim, H.K., 1965. Maintenance convulsive treatments. Am.
J. Psychiatry 121, 1113–1115.
Kellner, C.H., Knapp, R.G., Petrides, G., Rummans, T.A., Husain, M.M., Rasmussen, K.,
Mueller, M., Bernstein, H.J., O'Connor, K., Smith, G., Biggs, M., Bailine, S.H.,
Malur, C., Yim, E., McClintock, S., Sampson, S., Fink, M., 2006. Continuation
electroconvulsive therapy vs pharmacotherapy for relapse prevention in
major depression. Arch. Gen. Psychiatry 63, 1337–1344.
Lim, L.M., 2006. A practice audit of maintenance electroconvulsive therapy in
the elderly. Int. Psychogeriatr. 18, 751–754.
Moore, N.P., 1943. The maintenance treatment of chronic psychotics by
electrically induced convulsions. J. Ment. Sci. 89, 257–269.
Navarro, V., Gastó, C., Torres, X., Masana, G., Penadés, R., Guarch, J., Vázquez,
M., Serra, M., Pujol, N., Pintor, L., Catalá, R., 2008. Continuation/
maintenance treatment with nortriptyline versus combined nortripty-
line and ECT in late-life psychotic depression: a two-year randomized
study. Am. J. Geriatr. Psychiatry 16, 498–505.
O'Connor, D.W., 2008. Electroconvulsive therapy, In: Jacoby, R., Oppenheimer, C.,
Dening, T. (Eds.), Psychiatry in the elderly, 4th edition. Oxford University Press,
Oxford, pp. 201–214.
Russell, J.C., Rasmussen, K.G., O'Connor, M.K., Copeman, C.A., Ryan, D.A.,
Rummans, T.A., 2003. Long-term maintenance ECT: a retrospective
review of efficacy and cognitive outcome. J. ECT 19, 4–9.
Sackeim, H.A., 1994. Continuation therapy following ECT: directions for
future research. Psychopharmacol. Bull. 30, 501–521.
Sackeim, H.A., Haskett, R.F., Mulsant, B.H., Thase, M.E., Mann, J.J., Pettinati,
H.M., Greenberg, R.M., Crowe, R.R., Cooper, T.B., Prudic, J., 2001.
Continuation pharmacotherapy in the prevention of relapse following
electroconvulsive therapy: a randomized controlled trial. JAMA 285,
1299–1307.
Stiebel, V.G., 1995. Maintenance electroconvulsive therapy for chronically
mentally ill patients: a case series. Psychiatr. Serv. 46, 265–268.
Swoboda, E., Conca, A., König, P., Waanders, R., Hansen, M., 2001.
Maintenance electroconvulsive therapy in affective and schizoaffective
disorder. Pharmacopsychiatry 43, 23–28.
Thornton, J.E., Mulsant, B.H., Dealy, R., Reynolds, C.F., 1990. A retrospective
study of maintenance electroconvulsive therapy in a university-based
psychiatric practice. Convulsive Ther. 6, 121–129.
Tiller, J.W.G., Lyndon, R.W., 2003. Electroconvulsive Therapy: An Australasian
Guide. Postgraduate Foundation, Melbourne.
Vanelle, J.M., Loo, H., Galinowski, A., 1994. Maintenance ECT in intractable
manic–depressive disorders. Convulsive Ther. 10, 195–205.
Wood, D.A., Burgess, P.M., 2003. Epidemiological analysis of electroconvulsive
therapy in Victoria, Australia. Aust. N.Z. J. Psychiatry 37, 307–311.