Use of this content is subject to the Terms and Conditions of the MD Consult web site.
Psychiatric Clinics of North America
Volume 25 • Number 1 • March 2002
Copyright © 2002 W. B. Saunders Company
Electroconvulsive therapy in the medically ill
Keith G. Rasmussen, MD*
Teresa A. Rummans, MD
Jarrett W. Richardson, MD
Departments of Psychiatry and Psychology
Mayo Medical School
Rochester, Minnesota USA
E-mail address:
This work was supported by National Institutes of Health grant no. MH55484-05.
PII S0193-953X(03)00057-1
Electroconvulsive therapy (ECT) often is given to depressed patients with a variety of comorbid medical diagnoses. With attention to pretreatment medical and neurological evaluation, ECT can be safe even in highly medically or neurologically complicated patients. This article reviews the most commonly encountered medical and neurological illnesses in ECT patients and provides suggestions for safe and effective use of ECT in such populations.
Three questions face the practitioner who is treating a patient with combined medical and psychiatric illness for whom electroconvulsive therapy (ECT) is being considered:
- Is ECT effective in such patients?
- Does ECT cause deterioration of the underlying medical condition or significant side effects?
- Is modification of ECT technique indicated to lessen the risk of adverse outcomes?
To answer these questions definitively, prospective trials are needed in which the severity of both the medical and the psychiatric illness are blindly rated in a systematic manner before, during, and after treatment. Such trials are generally lacking. Furthermore, in relying on case reports, one must consider that these publications tend to be biased either in favor of excellent results or unexpectedly poor outcomes; however, a large body of case reports and case series published over several decades do inform the ECT clinician about a reasonable approach to patients with a wide array of medical disorders. This section focuses on medical conditions in which ECT has a direct physiologic effect and therefore places the patient at risk for complications. These include cardiovascular, neurologic, respiratory, and other disorders. In addition, the reader is strongly encouraged to apply the principles contained in the APA Committee on ECT report,[4]which has excellent sections on ECT in medically ill patients.
CARDIOVASCULAR DISEASES
Cardiac complications represent the most common source of significant morbidity and mortality during ECT. Even so, deaths are extremely rare, and the majority of cardiac patients can be given ECT with a low risk of complications.[2]There are several fundamental points in providing safe ECT to such patients. The first is accurate pre-ECT identification, through history, physical examination, and screening laboratory tests, of those patients with cardiovascular disease. The next step is pre-ECT cardiology consultation. The purpose of this is to determine the severity of cardiac disease and may include functional or structural cardiac imaging studies. The cardiology consultant can also recommend any modification of the patient's current treatment regimen to provide further stabilization of the cardiac condition (e.g., an increase in diuretic dosage in a patient with congestive heart failure that is not optimally controlled). Finally, the medical consultant can recommend strategies that may reduce cardiac risk during the ECT treatments, such as the administration of β-blockers to lessen the ECT-induced decrease in cardiac rate-pressure product. A third important strategy for safe administration of ECT to cardiac patients is careful vigilance throughout the ECT course, both at the times of treatment and in the intertreatment period, for any treatment-emergent complications so that prompt intervention can prevent deterioration of a minor complication into a major one.
An elegant demonstration of the importance of these points is provided by two reports from the Payne Whitney Clinic. Gerring and Shields[27]reported on a series of 17 cardiac patients given ECT in 1975–1976. None of the patients had pre-ECT medical consultations, electrolyte studies, or digitalis levels, even though 13 of the patients were on this medication. Most patients did not get continuous ECG monitoring during their treatments, nor were pretreatment antihypertensive medications given. There were four life-threatening cardiac complications and one death in this series. In contrast, Rice et al[75]reported on their experience at the same institution treating 26 cardiac patients with ECT in 1990–1991. All of the technical disadvantages described earlier for the earlier cohort had been reversed, so that routine pre-ECT laboratory and ECG studies were performed, most patients had pre-ECT medical consultations, and ECT technique followed modern standards, including continuous ECG monitoring and the availability of cardiac medications at the time of treatment. No major cardiac complications occurred in this cohort of patients. The following sections consider the aspects of ECT technique in patients with specific types of cardiovascular disorders.
Coronary Artery Disease
Cardiac complications during ECT are more common in those with coronary artery disease (CAD) than with other cardiac conditions.[102]ECT induces increases in pulse and blood pressure that result in a 200% to 300% increase in myocardial oxygen consumption during and for a few minutes after the seizure[24]; however, ECT can be safely administered to the majority of patients with CAD. For patients with known CAD, pre-ECT cardiology consultation is recommended to assess the stability of the patient's cardiac status and to assist with risk-reduction strategies.[4][5]The most important of such strategies is optimal stabilization of cardiac status before treatment.[2]Pretreatment cardiac imaging is commonly requested to assess the stability of the patient's myocardial capacity during stress. Any cardiac medications the patient is taking, with the exception of diuretics or lidocaine-type agents, should be given with a small sip of water the mornings of treatment. A further strategy is the intravenous administration of β-blockers immediately pretreatment to dampen the increase in myocardial oxygen consumption.[8]If pretreatment β-blockade is used, then an anticholinergic agent, such as glycopyrrolate or atropine, should also be given to minimize the risk of vagally mediated prolonged asystole or bradycardia, which sometimes occur early in the seizure or if no seizure is elicited by the electrical stimulus.[43]Depression commonly accompanies the postmyocardial infarction period—hence, the question of how long to wait until ECT is safe. Consideration should be given to the severity of the patient's depression and cardiac stability. A patient a few weeks out of a mild myocardial infarction with stable myocardial function may be safer to treat than a patient many months or years out of myocardial infarction with a very low ejection fraction or with exercise-induced ischemic changes on cardiac imaging. Close communication between psychiatrist and cardiologist is essential in these circumstances.
Dysrhythmias, Pacemakers, and Implantable Defibrillators
Patients with asymptomatic, uncomplicated cardiac conduction delays, such as first degree heart block or bundle branch block, can be safely treated with ECT without special precautions. Patients with atrial fibrillation also usually can be treated safely[59]; however, pretreatment cardiologic consultation can ensure optimal stabilization and treatment of the dysrhythmia. In particular, ventricular rate should be normalized. Occasionally, conversion to normal sinus rhythm occurs during ECT, so pretreatment anticoagulation would be considered optimal.[59]Patients with any type of malignant ventricular arrhythmia are at higher risk for hemodynamic instability during ECT, so pretreatment cardiology consultation is mandatory for risk assessment and risk-reduction strategies. Lidocaine in antiarrhythmic doses has potent anticonvulsant properties, so it should not be given before an ECT treatment unless truly necessary and no alternative antiarrhythmic is available.
Patients with pacemakers usually do not need any special adjustment or attention aside from ensuring that monitoring and other equipment are well grounded.[2]Consultation with a cardiologist is indicated, and some older types of pacemakers may need to be turned from demand mode to fixed mode before treatment.[2]The use of transesophageal atrial pacing during ECT for patients at significant risk for asystole has been described.[89]Patients with implantable cardiodefibrillators usually can be safely treated with ECT, but pretreatment consultation with a cardiac electrophysiologist is recommended.[4][61]
Hypertension
Despite the transient, peri-ictal increase in blood pressure, ECT does not result in sustained increases in blood pressure and in fact may result in a decrease in blood pressure.[88]Common sense dictates that optimum blood pressure control be attained in hypertensive patients pre-ECT. Antihypertensive medications (except for diuretics, which could result in a full bladder that may rupture during the seizure) should be given with a small sip of water in the mornings of treatments. If there are significant risks to the patient for even brief periods of hypertension (e.g., vascular aneurysms, cardiac aneurysm, severe left ventricular or valvular compromise), premedication with antihypertensive agents is indicated, selecting the drug that is most appropriate to the patient's clinical situation. Short-acting drugs (e.g., esmolol) administered at the time of anesthesia induction may be preferable to avoid delayed hypotension that can occur with longer-acting drugs (e.g., labetolol). Although β-blockade shortens seizure length during ECT,[92]no direct evidence shows that this interferes with therapeutic efficacy.
Vascular Malformations and Anticoagulation
ECT has been safely administered in the presence of known cardiac[14]and aortic aneurysms,[12]although a case of cardiac rupture during ECT has been reported.[3]Pretreatment consultation with a vascular surgeon in cases of known aneurysms or other vascular malformations (e.g., arteriovenous fistula) is important in assessing the risk of rupture during ECT. Risk-reduction strategies may include pretreatment with antihypertensive agents to reduce shear force against the aneurysm.
For patients who are on anticoagulation for deep venous thrombosis, valvular heart disease, atrial fibrillation, or another indication, it is prudent and safe to continue it during ECT with close attention to maintaining the proper PT or PTT.[4]
Congestive Heart Failure and Valvular Heart Disease
As ECT causes a temporary but sharp increase in cardiac rate-pressure product, patients with congestive heart failure are at risk for decompensation during a course of ECT[72]; however, successful administration of ECT to patients with stable low cardiac output has been described.[86]ECT has been successfully administered to patients with stable valvular heart disease.[32][68][72]As with the other cardiac conditions discussed, the most important aspect of management in such cases is careful pre-ECT cardiologic evaluation and risk assessment. In particular, patients with congestive heart failure should have their medical management optimized, with very careful ongoing assessment during the course of ECT looking for signs of congestive decompensation.
NEUROLOGIC DISORDERS
A large body of case reports and case series published over several decades do inform the ECT clinician about a reasonable approach to patients with a wide array of neurologic disorders. This section considers such patients grouped by type of neurologic illness. A thorough consideration of all published papers is beyond the scope of this article, but the interested reader is referred to various reviews published over the years [2][18][36][81][84][103].
Dementia
A vexing problem regarding the literature on ECT in demented patients is the lack of standardized criteria for the establishment of a dementia syndrome independent of depression-related cognitive dysfunction.[87]In fact, in routine clinical circumstances, it is often difficult to determine whether the patient with coexisting severe depression and cognitive dysfunction has a separate dementing illness or depression-related cognitive dysfunction.[87]A further problem with this literature is the lack of long-term cognitive follow-up in most reports. Nevertheless, the extant literature does provide guidance on safe, effective ECT in the depressed, demented patient.
Numerous case reports indicate that ECT can be efficacious in depressed and demented patients for the depressive symptoms without causing undue or long-lasting increases in memory disturbance.[67]Several case series provide more substantial information. In a prospective study, Reynolds et al[74]systematically followed depression ratings in a group of three demented depressives. Two of these patients had an excellent antidepressant response to ECT without cognitive worsening, whereas one had no antidepressant response and a sharp reduction in mini-mental state exam (MMSE) score. Gaspar and Samarasinghe[26]reported data on three patients with a diagnosis of primary dementia and depression. Similarly, two showed a good antidepressant response, and one did not. None of the patients were believed to have sustained a long-lasting (i.e., more than a few days) worsening of their baseline cognitive status, although no systematic cognitive evaluations were done.
In a larger series, Nelson and Rosenberg[55]reported their experience treating 21 depressed and demented patients with ECT. Global antidepressant efficacy ratings based on chart review were similar to those of a depressed but nondemented ECT cohort. Post-ECT confusion was believed to be greater in the depressed and demented group. No efficacy or cognitive follow-up data were presented.
In the largest series to date of depressed and demented patients given ECT, Rao and Lyketsos[67]described 31 patients, 16 diagnosed with vascular dementia, 4 with Alzheimer's dementia, and 11 with dementia of uncertain cause. Half the patients were believed to suffer delirium (criteria not specified) with ECT. Two thirds were believed to have a “good response” in terms of depression. MMSE score actually increased an average of 1.62 points by treatment end. This indicates the beneficial effect of ECT on attention and concentration after clearing of the acute post-treatment confusion, even in demented patients. Of note, one patient had tardive seizures in the recovery room after ECT. No efficacy or cognitive data were presented separately for the different demented groups.
To summarize, several dozen depressed ECT patients believed to have a separate dementing illness have been described in the literature. Approximately two thirds obtained noticeable antidepressant benefit, with approximately half showing greater than usual confusional states. Sorely needed is a trial with systematic cognitive evaluations in patients diagnosed with dementia by standardized criteria, and subclassified by dementia type, compared with nondemented ECT controls and with depressed and demented patients treated with medications. Finally, in an effort to spare excess memory disturbance, consideration should be given to treating depressed and demented patients with right unilateral or bifrontal electrode placement and twice-weekly treatment frequency.
Parkinson's Disease
A variety of case reports, small series, and one sham-ECT-controlled trial document that ECT is highly effective for depression in Parkinson's disease and may improve motor function as well[52][69]; however, delirium is common during ECT,[22]as is treatment-emergent dyskinesia,[15]but post-treatment cognitive function may be improved.[63]The duration of ECT-related antiparkinsonian effects has been variable, but long-term follow-up data have been presented. For example, Fall et al[19]systematically followed up 16 nondemented, nondepressed patients with advanced Parkinson's disease before, immediately after, and for up to 1.5 years after courses of mostly unilateral ECT. Fifteen subjects experienced some improvement in the motor symptoms of PD, which was dramatic in some cases. Half of the patients had sustained improvement for 3 to 18 months after ECT, without maintenance ECT. Approximately one third developed severe confusional states during ECT, the duration of which was a few days to 2 weeks. Interestingly, 4 of these 5 patients went on to have long-lasting (i.e., at least several months) motor improvement. Baseline increases in cerebrospinal fluid (CSF)-to-serum albumin ratios, which were thought to reflect violation of the blood-CSF barrier, accurately predicted the 5 subjects who developed interictal delirium. None of the 11 subjects who did not develop interictal delirium had pre-ECT increases in this ratio, indicating that violation of the blood-CSF barrier before ECT predisposes to ECT-induced confusion.
In another series, Pridmore et al[65]treated seven advanced, nondepressed PD patients with only four unilateral ECT treatments. Immediate post-ECT and 2-week post-ECT ratings in all spheres of motor function indicated improvements. Four patients developed severe, although temporary, confusion. One patient developed a dyskinesia responding to lowering of the levodopa dose. In a follow-up report on these and several other such patients given index (but not maintenance) ECT, three of nine initial ECT responders had improvements lasting 2 to 10 weeks, whereas the other six responders had improvements lasting 10 to 35 months.[64]
Attempts have been made to prolong the acute ECT-induced improvement in motor symptoms of Parkinson's disease using maintenance ECT, in which treatments are given every few weeks for variable lengths of time.[1][20][34][83][98][101]In these reports, a total of 13 patients with Parkinson's disease given maintenance ECT are described. Several of the patients had had several courses of index ECT with good, although relatively short-lived, motoric improvement. In each case, maintenance ECT seemed to extend this period of improvement, in some cases for up to 1 year. The frequency of such treatments was highly variable, depending on the length of each patient's improvement, and was balanced against cognitive impairment.
Several summary points can be made regarding the use of ECT in Parkinson's disease patients:
- It is highly effective for the severe depression so often seen in such patients.
- In a substantial proportion of patients, a variety of extrapyramidal signs (EPS) phenomena also may improve, often faster than the depression.
- The duration of this antiparkinsonian benefit may be from a few days to years.
- Maintenance ECT may extend the period of improvement in patients who can tolerate it from a cognitive standpoint.
- Treatment emergent dyskinesias and delirium are relatively common[15][22]and may be helped by careful reduction of dopaminergic doses.
- Twice-weekly treatment frequency and right unilateral or bifrontal electrode placement should be considered as methods to reduce memory impairment.
Neuroleptic-Induced Movement Disorders