While headache has been described as a side effect in 3% to 48% of patients treated with electroconvulsive therapy (ECT),1,2 true migraine is rarely reported.3,4 Such headaches are often short-lived and successfully treated with acetaminophen, acetylsalicylic acid, or other nonsteroidal anti-inflammatory drugs. For post-ECT headache and migraine that do not respond to typical analgesics and whose severity threatens the continuation of ECT, treatments including propofol,5 sumatriptan,6 and dihydroergotamine3 have been suggested. We report the successful use of a combination of naproxen and propranolol for ECT-induced migraine.
To date, there have been no reports on the use of propranolol in electroconvulsive therapy (ECT)-induced migraine; we describe a 32-year-old woman who was successfully treated with propranolol for this condition. Over a course of ECT, the patient developed increasingly severe migraine which was refractory to treatment with acetaminophen, codeine, and naproxen. Sumatriptan did not relieve the headache and aggravated the nausea. Successful migraine relief was achieved with a combination of propranolol and naproxen, administered before and after ECT. Propranolol reduced blood pressure and decreased the heart rate, measured before and immediately after ECT. Propranolol, possibly in combination with naproxen, may be useful in both acute and prophylactic treatment of post-ECT migraine.
A 32-year-old married mother of two was admitted to hospital for treatment of severe depression. The patient had a history of migraine which began in adolescence, became less frequent in her early 20s (about six episodes per year), and increased again during two pregnancies in her late 20s. During this period, migraine comprised bilateral headache with nausea, visual phenomena (sparkling lights throughout the visual field, kaleidoscopic with bending over), phonosensitivity, and photosensitivity.
During the year and a half prior to ECT, while the patient had been suffering from a treatment-refractory mood disorder, she reported mild migraine occurring every 2 to 3 days and more severe migraine every 7 to 10 days. The less severe headaches were unilateral, throbbing, and accompanied by mild nausea, phonosensitivity, and photosensitivity. The severe migraines were bilateral and throbbing, with the associated features of nausea, phonosensitivity, and photosensitivity, feeling off-balance and dizzy, blurred vision with “white dots,” and required her to go to bed. Migraine could be triggered by fatigue, stress, wine, and menses. Treatment included fixed-dose combinations of codeine, caffeine, acetylsalicylic acid and butalbital, or acetaminophen with codeine, but there was no prophylactic therapy.
The initial bilateral ECT treatment produced a headache and jaw ache that was effectively treated with two 500-mg doses of oral naproxen. A headache, increasing in severity, occurred on the following day and was treated successfully with 650 mg oral acetaminophen. After the second ECT treatment, the patient immediately reported a headache which was treated with acetaminophen 650 mg, followed by three doses of naproxen 500 mg, which resulted in eventual relief. Following the third ECT session, the patient immediately reported a severe bilateral headache which was not relieved by acetaminophen or naproxen, but the pain was reduced by 60 mg of oral codeine. The next day, the patient reported her headache to be less severe but still present. Naproxen 500 mg and codeine 60 mg were given 1 hour before the fourth ECT. Nevertheless, a post-ECT headache occurred which was unsuccessfully treated with 60 mg of codeine and 500 mg of naproxen. The headache was still present on awakening the next day and the patient described the quality of the pain as being similar to that in previous migraine attacks. The naproxen and codeine prophylactic treatment was again administered 1 hour before the fifth ECT, after which a severe headache occurred, accompanied by nausea. After this ECT treatment, the migraine was refractory to a further dose of naproxen 500 mg and codeine 60 mg, and also to oral sumatriptan 100 mg; the sumatriptan worsened the nausea. A severe headache occurred after the sixth ECT treatment, despite administration of naproxen and codeine an hour before ECT. Two days following this treatment, 20 mg of propranolol and 500 mg of naproxen were administered to control the persistent pain, and resulted in migraine relief within 4 hours. The treatment was repeated the next day with good results. Following this, it was decided to use propranolol prophylactically. One hour prior to the seventh ECT session, propranolol 20 mg and naproxen 500 mg were administered; no headache was reported following this ECT treatment. Later that morning, a headache did develop but was effectively treated with a further dose of naproxen and propranolol. From this point on, combined oral treatment with propranolol 20 mg and naproxen 500 mg was administered 1 hour prior to each ECT session and on an as-needed basis. Over the entire course of ECT treatment, from after the 6th ECT to the day after the 14th ECT, 48 doses of propranolol 20 mg and naproxen 500 mg were administered over a period of 28 days, including the doses administered 1 hour before each ECT session and those needed between ECT treatments. The patient received 2.6 (range, 2 to 4) doses of propranolol/naproxen on ECT treatment days (including premedication) and 1.4 (range, 0 to 2) doses on non-ECT treatment days. The patient received more propranolol/naproxen on ECT days (t26 = 4.60, P<.001), with most of the difference between ECT and non-ECT days accounted for by the pre-ECT dose. The propranolol/naproxen treatment was usually helpful in relieving migraine pain: 60% of the doses resulted in substantial relief (either sleeping or rated as “effective” by nursing staff), while 30% resulted in partial relief. Following the completion of the ECT series, treatment with propranolol and naproxen ended. Severe migraine persisted at a rate of approximately once per week and appeared to be more easily triggered than before ECT.
The patient received a total of 14 ECT treatments. The patient's heart rate and blood pressure were taken immediately before ECT, and at intervals after the end of the seizure. The mean heart rate without propranolol was 102 beats per minute before ECT and 103, 98, 91, and 90 beats per minute at 5, 10, 20, and 30 minutes after ECT, respectively; the values decreased with propranolol pretreatment to 88, 93, 90, 88, and 85 beats per minute, respectively. Analysis of variance showed that propranolol resulted in a significant decrease in heart rate (F1,12 = 15.65, P = .002). Without propranolol, the mean systolic blood pressure was 113 mm Hg before ECT and 103, 96, 96, and 94 mm Hg, at 5, 10, 20, and 30 minutes, respectively, after ECT; with propranolol, the values decreased to 93, 98, 95, 91, and 93 mm Hg, respectively. The mean diastolic blood pressure before propranolol was 64 mm Hg before ECT and 62, 56, 54, and 53 mm Hg, at 5, 10, 20, and 30 minutes, respectively, after ECT; with propranolol, the values became 61, 56, 56, 55, and 55 mm Hg, respectively. The blood pressure-lowering effect of propranolol was largely confined to systolic blood pressure (mean decrease in systolic blood pressure, 6.5 mm Hg; mean decrease in diastolic blood pressure, 1.3 mm Hg; F1,12 = 4.584, P = .054). Trend analysis on the five evaluation time intervals (one before and four following ECT) showed that propranolol had the effect of mitigating the high pre-ECT and immediate post-ECT values for systolic blood pressure (P = .03) and pulse rate (P = .03).
In this patient, repeated ECT treatment caused increasingly frequent and severe migraine. Weiner et al4 reported that of those patients who report having headache and migraine prior to ECT, about 7% experience exacerbation of both intensity and frequency of attacks post-ECT. Thus, the experience of our patient was not unusual.
Although the ictal rise in blood pressure was not measured in our patient, and may have been limited by the propranolol, the fact that propranolol lowered systolic blood pressure particularly around the time of seizure induction, might have contributed to its antimigraine effect. Propranolol, however, was first used successfully in our patient as an acute treatment and not before ECT, suggesting that the ictal rise in blood pressure may not have been the most important factor in the causation of our patient's post-ECT migraine. Cortelli et al7 have reported that modifications of supine blood pressure induced by propranolol were not related to its antimigraine effects. Propranolol also decreased the patient's heart rate. Zigelman et al8 have found that migraineurs have an autonomic imbalance, and have shown that propranolol reduces low frequency heart rate fluctuations, thus moderating the vascular instability related to migraine. Such a vascular effect may have contributed to the efficacy of propranolol in the treatment of migraine in our patient.
Serotonergic mechanisms have been implicated both in the antidepressant effect of ECT9 and in the pathogenesis of migraine.10,11 In a review of post-ECT headaches, Weiner et al4 postulated a serotonergic mech-anism for ECT-induced headaches, noting that ECT may increase serotonergic transmission by desensitizing serotonin (5-hydroxytryptamine: 5-HT) 5-HT1A autoreceptors and upregulating 5-HT2 receptors. Some investigators have hypothesized that migraine is caused by a state of 5-HT excess10,12-14 and that successful treatment results from a decrease in 5-HT levels. Weiner et al4 suggest that the upregulation of 5-HT2 receptor sites following ECT treatment, resulting in an increase of 5-HT function, may induce migraine in susceptible individuals. 5-HT2 antagonists have played a significant role in the prophylactic treatment of migraine. The degree of involvement of 5-HT2 receptors in migraine treatment has, however, been disputed,11,15 and currently, a single site of action or unifying hypothesis for the mechanism of action of all antimigraine drugs has not been accepted.
Propranolol is well established for the prophylaxis of migraine and, thus, the successful prophylactic treatment of migraine with propranolol in our patient is not unusual. The fact that propranolol was used effectively as an acute treatment is worthy of note, as investigators have reported a lack of efficacy of this drug in treating acute migraine.16,17 A β-adrenergic antagonist, propranolol is thought to act antagonistically at the 5-HT2 receptor (specifically the 5-HT2C and 5-HT2B subtypes) in migraine prevention.18 Recent reports suggest that antagonism at the 5-HT1A autoreceptor may also play a role in the clinical efficacy of propranolol.13,19 Chugani et al13 found an increase in the capacity to synthesize serotonin in patients with migraine after prophylactic treatment with β-adrenergic antagonists, an increase that may have resulted from 5-HT1A autoreceptor antagonism. Thus, in our patient, prophylaxis may also have been achieved through the combined antagonistic action at both the 5-HT1A and 5-HT2 receptor sites.
Sumatriptan, a 5-HT1B/1D agonist used in the acute treatment of migraine, has also, to some extent, agonistic activity at the 5-HT1A receptor,19 but no affinity at the 5-HT2 and 5-HT3 sites.20,21 Thus, propranolol and sumatriptan have opposite actions at the 5-HT1A receptor. In some patients, sumatriptan has been reported to be an effective treatment for ECT-induced migraine.6,14 However, in our patient, migraine was resistant to sumatriptan, a finding in approximately 20% of patients receiving oral sumatriptan.22 Antagonism at the 5-HT2 receptor by propranolol may have been necessary for migraine relief in this patient. Furthermore, the action of sumatriptan at the 5-HT1A receptor might explain why sumatriptan did not relieve our patient's migraine pain and worsened the nausea.21
The effectiveness of certain antimigraine treatments may be altered by ECT or may depend on the clinical effectiveness of ECT. Visser et al23 found that certain pharmacokinetic factors, notably the slope of plasma level elevation and possibly the degree of brain penetration, may be relevant to the response to sumatriptan. In some patients, ECT might affect the ability of certain antimigraine drugs (eg, sumatriptan) to act effectively at their target receptors, which would then result in reduced clinical efficacy. Whether or not this occurred in our patient is uncertain.
Naproxen was used continually throughout the course of ECT treatment in an attempt to relieve our patient's migraine. A nonsteroidal anti-inflammatory drug (NSAID), naproxen is known to be effective in the prophylaxis of migraine.24 Concurrent use of NSAIDs and β-blockers has been suggested for effective relief of migraine pain.25 It is difficult, however, to know what role naproxen played in the successful relief of migraine in this patient, because propranolol was not administered in the absence of naproxen. Our main goal was to provide effective migaine relief and permit the completion of the course of ECT. We were reluctant to withdraw the naproxen in case it was working synergistically with the propranolol.
Because propranolol has been reported to induce depression in some individuals,26 it is possible that it may have had adverse effects on our patient's mood. Despite the routine use of beta blockers, including propranolol, to control the cardiovascular effects of ECT,27,28 however, there have been no reports to date of these drugs increasing the severity of depression during ECT treatment or prolonging the course of ECT.
We have reported a case in which a patient was effectively treated with valproate for migraine induced by selective serotonin reuptake inhibitors.29 By controlling the side effects of the primary treatment, we were able to continue an effective treatment for severe depression. Similarly, in the case of the patient described in the current report, a course of ECT was completed only because the associated migraine was controlled.
Electroconvulsive therapy has multiple, diverse, and complex effects on the nervous system. Such effects may differ among patients, and therefore migraine treatment following ECT must be individualized. Migraine relief with propranolol is not unusual in itself. We believe, however, that this is the first reported instance of the successful use of propranolol for post-ECT migraine. Propranolol, possibly in combination with naproxen, should be considered for acute or prophylactic treatment of migraine triggered or exacerbated by ECT.