Should We Re-estimate the Seizure Threshold during Maintenance ECT?

To the Editor,

Eliciting generalized seizure activity in the brain by using an electrical stimulus that exceeds the seizure threshold (ST) is the fundamental dogma of effective electroconvulsive therapy (ECT). On the one hand, the magnitude of the electric stimulus above the ST has been linked with adverse cognitive effects. On the other hand, the administration of sub-convulsive electric stimulation has been associated with the risk of cardiac arrhythmias. ¹ Hence, determining the appropriate ST becomes the cornerstone of effective ECT administration. ST increases in most patients receiving an acute course of ECT, and it has been hypothesized to be due to the anticonvulsant properties of ECT. ² However, these changes in ST are not permanent. The effect of an acute course of ECT on the quantitative electroencephalogram (EEG) had lasted up to 2 months in an earlier study. ³ Evidence on the effect of continuation and maintenance regimens of ECT on ST is scant. The current report describes the reduction of ST within 2 weeks in a patient receiving maintenance ECT (M-ECT).

Ms. A, a 32-year-old single lady diagnosed with schizophrenia, had the onset of the illness at 17 years. She was on a stable dose of 700 mg/day of clozapine, 30 mg/day of olanzapine, and 1500 mg/day of sodium valproate. For the past 7 years, she has received M-ECT for treatment-resistant auditory hallucinations associated with significant distress. ECT was administered with bi-temporal electrode placement using the Niviqure-VR device (Niviqure Meditech Pvt. Ltd., Bangalore; pulse width of 0.3–2 ms and peak pulse amplitude of 1000 mA). Biphasic, brief-pulse stimuli with a frequency of 125 pulses per second were used. Anesthesia was induced with intravenous thiopentone sodium (usually 4–5 mg/kg). Succinylcholine (0.5–1 mg/kg) was used as a muscle relaxant, and she was premedicated with glycopyrrolate (0.2 mg). Seizures were monitored using the “cuff method,” and a motor seizure duration of 15 s was considered adequate. After the index ECT course, the duration between each ECT was gradually spaced out from three times a week to once a week in the first 2 months. Later, over 2 months, it was spaced out to a frequency of once every fortnights. Valproate was withheld on the nights before the M-ECT sessions. She has been receiving M-ECT for 7 years, with 201 sessions so far. The median stimulus dose was 480 millicoulombs (mC) (IQR 90), with a median seizure duration of 37 s (IQR 21). As auditory hallucinations and distress worsened, the frequency of M-ECT was changed to once a week for the last 2 months. With this regimen, there was an improvement in symptoms and no subsequent worsening. Two weeks after the last M-ECT session, the ST, re-estimated using the titration method, was 90 mC, with a seizure duration of 16 seconds. She continued to have stable symptom reduction even at the re-titrated lower dose of 90 mC. She received the next session of M-ECT after 2 weeks (90 mC and 31 s of motor seizures). She was subsequently discharged. She did not have worsening symptoms for the next 7 months and was continued on the above dose of medications.

In this case, we found a reduction in the ST within 2 weeks in a patient on long-term M-ECT with stable symptom reduction. In usual clinical practice, a constant stimulus dose is delivered during M-ECT, and re-titration of ST is not carried out even when the interval between two successive sessions is longer. Few studies have explored the change in ST in ECT sessions with longer inter-session intervals. In an earlier study, the changes in seizure duration, which indirectly indicate the change in ST, were significant for ECT sessions spaced for 2 months or longer. ⁴ Our report contrasts with an earlier study that did not find an increase in seizure duration, indirectly reflecting the unchanged ST, in patients receiving M-ECT with a mean treatment interval of 12.1 days (range, 2–47 days). ⁵ Another study that investigated the changes in seizure duration as a function of ST in M-ECT concluded that an early increase in the seizure duration at a fixed stimulus dose could indicate relapse. ⁶ This contrasts with our report, where our patient had stable symptom reduction. Lowering the stimulus dose in M-ECT has been shown to improve seizure quality. ⁷ The non-linearity in the relationship between stimulus intensity with respect to ST and duration should be kept in mind while interpreting such results. Seizure duration alone is not a good guide to determine clinical efficacy. ⁸ Lack of ictal EEG monitoring and formal assessment of symptoms are the limitations in our case. Although there are no clear guidelines on the duration after which ST has to be recalculated in M-ECT, periodic re-titration, at least once in a few months, would prevent unwarranted high stimulus dosing, and thereby reduce the risk of cognitive adverse effects.