its short- and long-term therapeutic efficacy

Transcription

1 Lithium carbonate in cluster headache: assessment of its short- and long-term therapeutic efficacy Gian Camillo Manzoni, Giorgio Bono, Marina Lanfranchi, Giuseppe Micieli, Mario Giovanni Terzano and Giuseppe Nappi CEpMGH Manzoni, G. C., Bono, G., Lanfranchi, M., Micieli, G., Terzano, M. G. & Nappi, G : Lithium carbonate in cluster headache: assessment of its short- and long-term therapeutic efficacy. Cephalalgia, Vol. 3, pp Oslo. ISSN The short- and long-term effects of administration of lithium carbonate in cluster headache (CH) have been investigated. Of the 90 patients treated (78 males and 12 females), 8 had episodic CH and 22 had the chronic form of the disease. The doses used were almost always 900 mg/day. Eleven of the 22 patients with chronic CH showed a definite, constant improvement both short and long term. In 7 of the 22 patients, lithium treatment provided excellent results initially but was later followed by some transient worsening; in the remaining 4 only partial benefits were observed initially and treatment proved still less effective after a few months. The effects of cessation of lithium administration after at least five months of continuous treatment were studied in 9 cases. In of them the attacks re-appeared immediately, whereas in 3 the attacks occurred again only after free intervals of four to six months. Of the 8 patients with episodic CH, 2 proved highly responsive to treatment, 2 only partially responsive, and 1 refractory. In 3 cases, after one to three years of continuous treatment, euthyroid goitre developed, which disappeared after the drug was discontinued. 0 Cluster headache, lithium. Gian Camillo Manzoni, Marina Lanfranchi, Mario Giovanni Terzano, Department of Neurology, Headache Centre, University of Parma, Parma, Italy; Giorgio Bono, Giuseppe Micieli, Giuseppe Nappi, Department of Neurology, Headache Centre, University of Pavia, Pavia, Italy; Accepted I Introduction Since the first reports of the high effectiveness of lithium in cluster headache (CH) by Ekbom (l), Graham (2) and MacGregor (the latter being cited by Graham in a headache round published by Rogado (3) in the journal Headache), several authors (4-11) have evaluated the effects of lithium in the chronic and episodic forms of the disease. In the investigations conducted so far, however, the number of patients treated has been rather small. In order to better evaluate the effectiveness of the treatment with lithium carbonate in CH, we have analysed the short- and long-term effects of this drug in a broad series of patients. Patients and methods Our study included 90 patients with CH who had been followed at the Headache Centres in the Departments of Neurology of the universities of Parma and Pavia between 1977 and 1982 (Table 1). Twenty-two of them were suffering from chronic CH and 8 from episodic CH. Of the 22 chronic patients, 1 had a primary form of the disease and a secondary form. Seventy-eight of our patients were males and 12 were females. When treatment was started, their mean age was 38.9 years (min 17, max 7). They had been suffering from CH for, on average, 8.2 years (min 3, max 28). The diagnosis of CH was made in accordance with the recommendations of the Ad Hoc Committee on Classification of Headache (12). For diagnosis of the chronic type, the suggestions of Ekbom and Olivarius (13) were adhered to. The 22 patients with chronic CH were followed for a minimum of 3 to a maximum of 48 months (average 22 months). The doses used were 300 mg three times a day; in a few cases, however, and only for a limited period

2 110 Gian Camillo Manzoni et al. CEPHALALGIA 3 (1983) Table 1. Lithium treatment in cluster headache. Distribution of patients by clinical subtypes and sex. Cluster headache 90 (78 M, 12 F) Episodic 8 (59 M, 9 F) I Primary 1 (14 M, 2 F) Chronic 22 (19 M, 3 F) c (5 M, 1 F) of time, the dosage used was 00 or 1200 mg daily. In 9 of these 22 cases, treatment was interrupted one or more times between 5 and 42 months after it was started in order to assess its consequences on the CH course. In the 8 patients with episodic CH, treatment was started when the subjects had had a cluster period for, on average, 24 days (min 3, max 80); in all cases the dose administered daily was 900mg. Of these 8 patients, 50 were treated with lithium carbonate for one cluster period only, 8 for two periods, and for three. The drug was administered without interruption for over a year in 4 cases; three of these cases had a high frequency of cluster periods and one was also suffering from manic-depressive psychosis. With the exception of these 4 cases, in all other cases treatment was discontinued only when the patients had been free from attacks for at least one week. The plasma and erythrocyte lithium level was checked weekly during the first month of treatment and in each succeeding month. Any clinical improvement was quantified according to the Headache Index Ratio, obtained by dividing the weekly lithiumtreatment Headache Index by the weekly pre-treatment Headache Index. The percent improvement is the reciprocal number of the Headache Index Ratio X 100. In order to calculate the pre-treatment Headache Index, the frequency and intensity of attacks were accurately recorded for four weeks prior to lithium therapy in the patients with chronic CH, and in the single week preceding treatment in those with episodic CH. In chronic CH patients, the short-term results were evaluated within two weeks of the beginning of treatment, while the long-term results were assessed- throughout treatment (22 months on average; min 3, max 48). In episodic CH patients, the results were measured as early as the fifth day of treatment. This was done in order to avoid, as far as possible, any interference from the spontaneous remission of the cluster period. Results In the second week of lithium treatment, over 80% of the patients with chronic CH improved by more than 90%, and only 2 patients improved by less than 0% (Table 2). The weekly Headache Index of the 22 patients averaged 32.8 (min 12, rnax 5) prior to treatment and dropped to an average of.0 (min 0, max 24). At this initial stage of treatment, in no case did side-effects occur requiring the drug to be discontinued; in 7 patients side-effects did appear but were mild and generally did not last long (Table 2). Later on in the course of treatment of the 4 subjects who initially had improved by less than 90%, cluster attacks were still frequent or would re-appear frequently so that lithium treatment had to be discontinued after 3 to months. The other 18 patients, who were followed for 9 to 48 months (average 27), exhibited two different patterns. In 11 patients, the definite improvement observed soon after treatment had started was almost unchanged; the other group (7 patients) worsened periodically, although the initial benefits derived from treatment were still in general quite apparent (Table 2). In some of the latter patients, as little as 00 mg per day was enough to relieve headache over certain periods of time, sometimes as long as a few months; in other periods, however, the doses had to be increased to 1200 mg and even then did not prove entirely effective. No significant differences were found in the distribution of patients with the primary or the secondary form of headache between the groups with or without long-term benefits.

3 CEPHALALGIA 3 (1983) Lithium in cluster headache 111 Table 2. Results of lithium treatment in 22 chronic cluster headache patients. Initial results Long-term results Improvement I '90% 0-90% <0% ' 2 mean improvement 82.9% 11 Improvement constantly maintained 7 Improvement maintained with transient worsenings 4 No benefit Side-effects Side-effects Severe, requiring discontinuation 0 Severe, requiring discontinuation 2 Mild, allowing continuation (goitre 2) (tremor 4, diarrhoea 2, abdominal pain 1, olfactory Mild, allowing continuation 4 hallucination 1, insomnia 1, vertigo 1, increased (tremor 2, nausea 1, diffuse headache 1, lethargy thirst 1) 1) Of the 4 patients in whom lithium treatment was discontinued because of its poor effects, 3 had primary chronic CH and 1 had secondary chronic CH. Of the remaining 18, in whom the initial improvement persisted though not to the same degree for all of them, 13 had primary chronic CH and 5 had a secondary chronic form of the disease. The plasma and erythrocyte lithium levels varied from 0.3 to 0.8 meq/l and from 0.10 to 0.35 meq/l, respectively. Among the long-term side-effects, goitre developed in two cases (a 2-year-old woman in the 18th month of treatment and a 31- year-old man in the 40th month of treatment). This, however, did not impair the normal functioning of the thyroid gland and disappeared completely after lithium was discontinued. The effects of one or more interruptions of treatment on the CH course was evaluated in 9 patients. Two of them (Case nos. and 7) were those who had developed goitre. In the remaining 7 cases, treatment was interrupted on purpose after 5 to 42 months of lithium therapy in order to assess the resulting effects (Table 3). In most cases ( patients), the attacks re-appeared almost immediately, their frequency and intensity being sometimes greater than prior to treatment. All these patients, however, improved again dramatically when lithium administration was resumed. In 3 patients (Nos. 2, 5 and 8), one of whom had primary chronic CH and two secondary chronic CH, long periods of remission (, 4 and 8 months, respectively) were observed after interruption of treatment, as if their'chronic form Table 3. Course of chronic cluster headache after interruption of lithium treatment (9 patients). Patient Age (years) Diagnosis Duration of disease (years) Treatment duration before interruptions (months) Course of headache after interruption Primary Primary Prim a r y Prim a r y Periodic course Periodic course Periodic course

4 112 Gian Carnillo Manzoni et al. CEPHALALGIA 3 (1983) Table 4. Results of lithium treatment in 8 episodic cluster headache patients. >90% 2 2 <0% 1 mean improvement 8.3% Side-effects severe, requiring discontinuation 0 mild, allowing continuation 18 (tremor, increased thirst 5, insomnia 4, diarrhoea 3, lethargy 2, diffuse headache 1, goitre 1) had turned episodic. Neither the duration of the disease nor the timing of drug withdrawal seemed to have any clear correlation with the subsequent course of the headache. Of the 8 patients with episodic CH, about three-fourths improved by more than 0% (Table 4). In these 8 patients the weekly Headache Index, which averaged 30.3 (min 12, max 58) prior to treatment, dropped to an average of 9. (min 0, max 42). None of our patients was forced to have treatment interrupted because of side-effects. Mild side-effects appeared in 18 cases, the most frequent being tremor, thirst and insomnia. Of the 1 patients who improved by less than 0%, 13 had administration of lithium discontinued before their cluster period was over because the drug had proved ineffective. In the remaining 55 patients, the cluster period on average lasted 43 days, which means that it was not significantly shorter than the average previous period (48 days). In no case did interruption of the treatment after patients had been free from attacks for at least one week coincide with a reappearance of the cluster period. The plasma and erythrocyte lithium levels varied from 0.3 to 0.7 me& and from 0.12 to 0.34 me&, respectively. When administered only within a cluster period, lithium turned out to have no effect at all on the CH course; in other words, it could not prevent or delay onset of future cluster periods. Fourteen patients were treated with lithium over two subsequent periods; in three of them the drug was considerably less effective in the second period than it had been in the first, whereas the remaining 11 patients improved as much in the second period as they did in the first. In these 14 cases, however, headache was now reduced by only 0.9% compared to 80.3% after the first treatment period. In six patients lithium was given for a third period, too; in this period headache improved by only 41.% on average. Four subjects with episodic CH were treated with lithium carbonate without interruption for over a year. This therapeutic approach was followed in three cases because of the high frequency of cluster periods, i.e. more than four a year, and in one case because the patient was also suffering from manic-depressive illness. In none of these four patients did attacks occur again as long as treatment was continued. In the three cases with a high frequency of cluster periods, the attacks resumed within 1 to 3 weeks of withdrawal of lithium. Discussion Studies conducted in chronic CH so far (1, 4-,8,10,11) agree on the high effectiveness of lithium therapy (Table 5). Also in episodic CH, almost all studies (5, 7-9) have shown good results (Table ). Only Ekbom (1 1) has reported rarely obtaining any beneficial results in episodic CH. The analysis of our results points to some considerations. As far as chronic CH is concerned, it may be concluded that: (1) Lithium is highly effective not only in the short term but also in the long term in the great majority of the patients. In this respect our findings do not agree with those of Ekbom (ll), who has reported that two of his three patients treated for between 18 and 3 months appeared to become tolerant to the effects of the drug. This discrepancy may be due to the paucity of data in the mentioned study (11). (2) After a long period of administration, i.e. at least one year, it might be advisable to discontinue lithium; in some cases, fairly long periods of remission were observed. (3) The lithium doses should be adjusted in accordance with the spontaneous, fluctuating course of this type of headache.

5 ~ ~~ CEPHALALGIA 3 (1983) Lithium in cluster headache 113 Table 5. Lithium in chronic cluster headache: review of the literature. Length of Percent Number Lithium Dosage Plasma level follow-up improvement Author(s) of patients preparation (mg/day) (me&) (months) < >90 Ekbom (1) Kudrow (4) Mathew (5) Szulc-Kuberska and Klimek () Savoldi et al. (8) Manzoni et al. (10) Ekbom (11) sulphate not stated carbonate carbonate carbonate carbonate carbonate 3OC sulphate not stated Thus, as little as 00mg or maybe even 300 mg will be enough at certain times, whereas at other times the dosage will have to be increased up to 1200 mg. (4) The doses used for lithium to be effective and consequently the plasma and erythrocyte levels of this drug are on average lower than those which proved effective in the treatment of manic depressive psychosis. (5) In general, lithium treatment is very well tolerated; it should be borne in mind, however, that patients treated over a long period of time may develop goitre. As far as episodic CH is concerned, it is much more difficult to determine whether a drug is effective or not. Among other things, in this type of headache it is impossible to predict how long the cluster periods are going to last, since the length of the previous period or periods is only slightly indicative. Nevertheless, the results that we obtained in our patients with episodic CH point to the following conclusions: (1) Lithium treatment has proved to be effective, even though the patients treated seem to be less responsive than in the chronic form of the disease. In a large number of patients, the drug causes a decreased frequency and intensity of attacks over a short period of time. Also in this respect, our findings do not agree with those of Ekbom (ll), who has reported little or no effects in seven episodic CH patients. (2) Lithium does not seem to significantly reduce the duration of the cluster period. This finding, however, is a doubtful one, since it is impossible to predict how long the cluster period is going to last. (3) When lithium is administered over several subsequent periods, its effectiveness appears to be partially reduced, at least in some cases. This finding, however, must certainly be investigated further. (4) Continued lithium treatment seems to be efficacious in episodic CH patients in whom cluster periods are particularly frequent. (5) Lithium, when administered only within cluster periods, does not appear to affect the CH course in any way. Table. Lithium in episodic cluster headache: review of the literature. Length of Percent Number, Lithium Dosage Plasma level follow-up improvement Author( s) of patients preparation (mg/day) (me&) (months) <0 C-90 >90 Ekbom (1) 2 sulphate not stated _ Mathew (5) 14 carbonate Medina (7) 12 not stated not stated not stated _ - Savoldi et al. (8) 2 carbonate Manzoni et al. (9) 14 carbonate Ekbom (11) 7 sulphate not stated

6 114 Gian Carnillo Manzoni et al. CEPHALALGIA 3 (1983) References 1. Ekbom K. Lithium vid kroniska symptom av cluster headache. Opusc Med 1974;19: Graham J R. Treatment of cluster headache (workshop). Chairman; Ottar Sjaastad, Sixteenth Annual Meeting, American Association for the Study of Headache. June, Rogado A Z. Headache rounds. Electrocution. Headache 197;1: Kudrow L. Lithium prophylaxis for chronic cluster headache. Headache 1977;17: Mathew N T. Clinical subtypes of cluster headache and response to lithium therapy. Headache 1978;18:2-30. Szulc-Kuberska J, Klimek A. Lithium treatment of chronic Horton s headaches. Neurol Neurochir Pol 1978; 12: Medina J L, Fareed J, Diamond S. Blood amines and platelet changes during treatment of cluster headache with lithium and other drugs. Headache 1978;18: Savoldi F, Nappi G, Bono G. I sali di litio nel trattamento della cefalea a grappolo. Riv Neurol 1979;49: Manzoni G C, Terzano M G, Trabattoni G. I1 trattamento della cefalea a grappolo con carbonato di litio. Ateneo Parmense [Acta Biomed] 1979;50: Manzoni G C, Terzano M G. Lithium carbonate in chronic cluster headache: assessment of therapeutic efficacy and possible mechanisms of action. Ital J Neurol Sci 1980;3: Ekbom K. Lithium for cluster headache: review of the literature and preliminary results uf long-term treatment. Headache 1981;21: Ad Hoc Committee on Classification of Headache. JAMA 192;179: Ekbom K, Olivarius B de Fine. Chronic migrainous neuralgiadiagnostic and therapeutic aspects. Headache 1971 ;I 1 :97-101