Prednisone vs. Placebo in Withdrawal Therapy Following Medication Overuse Headache

Abstract

This proof-of-concept study evaluated the efficacy of prednisone for the treatment of withdrawal symptoms in patients with medication overuse headache (MOH) in a randomized, placebo-controlled, double-blind design. Twenty patients were randomized and underwent in-patient withdrawal therapy. The total number of hours with severe or moderate headache within the first 72 and 120 h was significantly lower in the prednisone group. The results show that prednisone might be effective in the treatment of medication withdrawal headache.

Introduction

Chronic headache is a frequent disorder with a prevalence of 2–5% in the general population. Within the last 5 years epidemiological studies have shown that up to a third of these cases represent patients with medication overuse headache (MOH) or probable MOH (1–4) according to the classification of the International Headache Society, 2nd edn (ICHD-II, Code 8.2). The condition is defined as headache on > 15 days/month due to the frequent intake of antiheadache medication on 10–15 days/month (5). There is general agreement that all drugs used for the treatment of headache can cause MOH in patients with a primary headache disorder (6). Treatment of choice is the complete discontinuation and subsequent withdrawal from the overused headache medication. Discontinuation of overused medication usually leads to severe exacerbation of a headache called ‘withdrawal headache’ (7). In addition, most patients experience typical withdrawal symptoms such as nausea, vomiting, arterial hypotension, tachycardia, sleep disturbances, restlessness, anxiety and nervousness. Depending on the overused drugs, withdrawal symptoms may vary with respect to their intensity and may last up to 14 days, in special cases even longer (8). Due to a lack of treatment recommendations, therapeutical strategies to alleviate withdrawal symptoms vary among centres and countries. Among those agents suggested for withdrawal therapy are analgesics, tranquillizers, neuroleptics, antidepressant drugs (amitriptyline), antiepileptic drugs (valproate), intravenous dihydroergotamine, subcutaneous sumatriptan, oxygen and electric stimulation. None of these treatments has been investigated in a proper randomized, placebo-controlled trial. Few prospective studies have addressed the issue of the treatment of withdrawal symptoms during withdrawal therapy in patients suffering from MOH. An open non-placebo controlled study performing withdrawal on an out-patient basis has suggested that prednisone may be effective in reducing withdrawal symptoms (9). The aim of this study was therefore to evaluate the potential clinical efficacy of prednisone in a randomized, placebo-controlled, double-blind study for the treatment of withdrawal symptoms following withdrawal therapy in patients with MOH.

Methods

The study was approved by the local ethics committee (no. 03–2138) of the University of Essen. Written consent was obtained from all study participants prior to enrolment in the study.

Study population

Twenty patients were enrolled. Inclusion criteria were: men or women aged 18–65 years suffering from MOH according to the diagnostic criteria of the ICHD-II (5) caused by 5-HT1b/d-agonists, ergots, combined preparations of analgesics with caffeine or codeine, or analgesics. Exclusion criteria were: patients suffering from other conditions than migraine or episodic tension-type headache as primary headache, coexisting psychiatric conditions such as major depression or psychosis in the medical history, patients with diabetes and pregnancy.

Study design and protocol

This was a double-blind, placebo-controlled, randomized, single-centre pilot study. Patients with MOH underwent withdrawal therapy on an in-patient basis in the Department of Neurology, University Hospital of Essen. Placebo or prednisone 100 mg were given once daily for the first 5 days of withdrawal. The medication was blinded, randomized by standardized computer-based randomization software and delivered by the pharmacy of the University Hospital of Essen.

Severity of headache was documented every hour in a diary using a four-point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe headache). Intravenous 1000 mg lysine acetylsalicylic acid (LAS) was given as rescue medication in case of severe withdrawal headaches. In Germany, LAS is the substance of choice for the treatment of severe migraine attacks and withdrawal headaches if there are contraindications for triptans or other oral antiheadache medication.

Outcome criteria

The primary end-point was the number of hours with severe or moderate headache within the first 72 h of withdrawal. Secondary end-points were the number of hours with severe or moderate headache within the first 120 h of withdrawal and the amount of rescue medication used in each group.

Statistical analysis

A Wilcoxon rank sum test was use to compare interval scaled (age, number of headache days per month and number of hours with severe or moderate headache and number of rescue medications during 120 h of withdrawal), and χ² test for ordinal scaled data (gender) variables. The non-parametric test was chosen because of the relatively small number of patients. All statistics were calculated with the SPSS program (9.0.1; SPSS Inc., Chicago, IL, USA). The level of significance was set at 0.05.

Results

Patient population

Twenty patients were enrolled into the study (10 in each group). Complete sets of data were available for 18 patients (each group n = 9). One patient in the prednisone group was not suffering from any headaches in the days before starting treatment, the treatment or follow-up period. One patient in the placebo group did not record his headache severity exactly for every hour of the day. Because of missing data these two subjects could not be evaluated. Individual data of each patient are shown in Table 1.

Table 1

Data of each individual patient

No.	Gender	Age, years	Headache		Headache last 3 months days/month	Overused medication		Prophylaxis last 3 months	Hours per day with severeor moderate headache					Number of rescue medications	Study medication
Primary headache	Since (years)	Dose/ week	Preparations	Day 1	Day 2	Day 3	Day 4	Day 5
1	Female	40	MA	26	28	17	Rizatriptan and ibuprofen	None	10	8	9	12	14	5	Placebo
2	Female	54	MA	23	30	6	Ergots	None	19	19	19	18	18	3	Prednisone
3	Female	44	MoA	30	16	20	Zolmitriptan and ASA	None	19	19	16	3	0	5	Placebo
4	Female	48	MA	17	20	8	Sumatriptan and metamizol	None	18	1	22	1	0	2	Placebo
5	Male	49	TTH	20	30	18	ASA and paracetamol	None	3	0	0	7	0	0	Prednisone
6	Female	50	MoA	20	30	14	Rizatriptan and ergots	None	0	0	0	0	0	0	Prednisone
7	Female	62	MA	54	30	7	Ergots	None	20	16	0	0	0	3	Placebo
8	Female	42	MA	27	30	6	Rizatriptan	None	17	18	18	13	10	7	Prednisone
9	Female	31	MoA	16	30	20	ASA/paracetamol/ caffeine	None	1	3	1	6	3	0	Prednisone
10	Female	43	TTH	5,5	30	15	Ibuprofen and flupirtine	Amitriptyline	NA	NA	NA	NA	NA	0	Placebo
11	Female	29	MoA	6	30	15	Paracetamol	None	13	12	0	0	0	0	Placebo
12	Male	49	MoA	27	15	3	Zolmitriptan	None	14	0	0	0	0	0	Prednisone
13	Female	53	MoA	29	19	5	Zolmitriptan	Doxepine	18	17	1	0	0	0	Placebo
14	Male	27	MoA	11	30	5	Zolmitriptan	None	0	0	0	0	0	0	Prednisone
15	Female	36	MoA	23	30	35	Paracetamol/ caffeine	None	2	8	2	1	1	0	Prednisone
16	Female	53	MoA	41	30	5	Rizatriptan	β-Blocker	1	0	0	0	0	0	Prednisone
17	Female	39	MoA	25	30	10	Ergots	None	13	15	9	3	2	2	Placebo
18	Female	46	MA	30	30	6	Zolmi- and rizatriptan	Flunarizine/ clomipramine	6	1	0	0	0	0	Prednisone
19	Female	51	MA	44	30	14	Zolmi- and almotriptan	Topiramate/ doxepine	5	12	0	2	1	4	Placebo
20	Female	56	MoA	30	24	6	Rizatriptan and ASA	None	17	9	4	1	0	0	Placebo

MA, Migraine with aura; MoA, migraine without aura; NA, not available; TTH, tension-type headache; ASA, acetylsalicylic acid.

No significant demographic and clinical differences were found between the two treatment groups (Table 2).

Table 2

Clinical and demographic data of the patient population

	Prednisone	Placebo
Men/women	1/8	0/9
Mean age (years)	45	47
Migraine	n = 8	n = 9
Tension-type headache	n = 1	n = 0
Headache days/month	28	25
Patients	n = 9	n = 9

Outcome criteria

The total number of hours with severe or moderate headache within the first 72 (primary end-point) and 120 h (Fig. 1) was significantly lower in the prednisone group than in the placebo group (18.1 vs. 36.7 h, P = 0.031, and 27.22 vs. 42.67 h, P = 0.05). Patients in the prednisone group requested rescue medication less frequently than those of placebo group, which however, did not reach significance (11 dosages vs. 21 dosages, P = 0.3). Adverse events were not reported.

Figure 1

Hours with severe or moderate headache within the first 72 and 120 h after withdrawal of headache medication.

Discussion

The main finding of this study was that 100 mg prednisone given daily for 5 days decreases the duration of withdrawal headache significantly compared with placebo-treated patients and is well tolerated. Significant adverse events were not reported by the participants. The duration and severity of withdrawal headache in the placebo group were in line with the results of former studies investigating withdrawal symptoms in patients with MOH undergoing withdrawal therapy. Since this trial was planned as a proof-of-concept study, only 20 patients were enrolled. Although the results are clearly in favour of prednisone therapy, this is a shortcoming of the study because the small number of patients does not allow subgroup analysis regarding the influence of the primary headache type, accompanying withdrawal symptoms, substance use, or gender. The issue of headache recurrence following sudden prednisone discontinuation (no tapering) was not addressed in this pilot trial and will be studied in a larger, future trial.

Our results suggest that prednisone could be effective in treatment of medication withdrawal headache and that the drug may facilitate withdrawal therapy. In contrast to our results, a recent Norwegian placebo-controlled study has shown that 60 mg prednisone given for 2 days and tapering the dose the following 4 days had no effect on withdrawal headache in patients with chronic daily headache and medication overuse (10). Typically, the intensity of withdrawal headache increases between days 2 and 4 after withdrawal of the previously used antiheadache medication and wanes again by days 6 and 8. Hence, it is tempting to speculate that a dosage of 60 mg for the first 2 days was simply not enough to achieve significant effects during withdrawal therapy (10).

Mechanisms leading to MOH are still little known. Down-regulation of serotonin-receptors (5-HT1B/D) or prostaglandin-synthesizing enzymes in anatomical structures that are involved in the transmission and modulation of nociceptive signals may lead to an impairment of antinociceptive activity and subsequently result in a permanent feeling of headache (11). The possible action of steroids in the treatment of MOH and other headaches such as cluster headache or status migrainosus still needs to be determined. The anti-inflammatory effect and the inhibition of cyclo-oxygenase, cyclo-oxygenase-II in particular, are possible mechanisms responsible for the positive effect of prednisone. As a consequence of this trial, a larger, multicentre, Phase III study has been initiated at various centres in Europe.