Sage Journals: Discover world-class research

Abstract

Background

Definitions of medication overuse headache have changed over time.

Objective

To evaluate the clinical characteristics of medication overuse headache patients admitted for inpatient withdrawal therapy over a period of 32 years.

Methods

We included all patients with medication overuse headache treated from 1 January 1984 to 31 December 2015. We obtained all data from the medical reports and defined three periods, P1 (1984–1993), P2 (1994–2003), and P3 (2004–2015). The p-value adjusted for multiple comparisons was set to 0.005.

Results

Within 32 years, a total of 787 patients accounted for 904 admissions for MOH. From P1 to P3, the proportion of patients with preexisting migraine increased from 44.3% to 53.3% (chi² = 9.0, p = 0.01) and that with preexisting tension-type headache decreased from 47.9% to 34.6% (chi² = 9.3, p < 0.01). The median time since onset of headache and medication overuse headache decreased from 20 to 15 years (p < 0.001) and from 3 to 2 years (p < 0.001). The median cumulative number of single doses decreased from 120 to 90 per month (p = 0.002). Overuse of triptans, non-opioid analgesics, and opioids increased, whereas overuse of ergotamines decreased over time (p < 0.001 for all tests). The use of prophylactic medication before admission increased from 8.3% to 29.9% (chi² = 89.5, p < 0.001).

Conclusion

This retrospective study in a large number of patients with medication overuse headache admitted for inpatient withdrawal therapy over a period of 32 years shows a trend towards changes in the preexisting headache type, a decrease in the time since onset of headache and medication overuse headache, a decrease in the number of drug doses used per month, changes in the type of drugs overused, and an increase in, but still low rate, of prophylactic medication prior to admission.

Keywords

Inpatient withdrawal therapy migraine tension-type headache triptans non-opioid analgesics opioids

Introduction

Medication overuse headache (MOH) affects about 2% of the general population and is most commonly related to chronic migraine (CM) or chronic tension-type headache (TTH) (1). The first classification of headache disorders published in 1988 included “headache induced by chronic substance use or exposure” and differentiated “ergotamine-induced headache” and “analgesics abuse headache” (2). The diagnosis required daily intake of a substance for three or more months, a minimum monthly dose, headache on 15 or more days per month, and headache resolution within 1 month after withdrawal. The second edition of the International Classification of Headache Disorders (ICHD-2) published in 2004 (3) introduced the term “medication overuse headache”, did not define any minimum doses, set the cut-off for overuse at 15 days per month for simple analgesics and at 10 days for any medication over a period of 3 months, and required that headache both worsens during overuse and reverts to its previous pattern within 2 months after discontinuation of the overused medication. ICHD-2 (3) differentiated the overuse of ergotamines, triptans, analgesics, opioids, combination-medication overuse, other medication overuse, and probable MOH. Revised criteria published in 2006 discarded the requirement that headache improves after withdrawal (4) and allowed diagnosis of MOH immediately, not only after follow-up. In ICHD-3 beta (5), the requirement that headache worsens during overuse was also discarded, “analgesic-overuse headache” was replaced by “simple analgesic-overuse headache”, two further types of overuse; that is, overuse of multiple drug classes not individually overused and unverified overuse of multiple drug classes, were added, and probable MOH was discarded. In ICHD-3 (6), the criteria remained unchanged apart from replacing “simple analgesic-overuse headache” by “non-opioid-analgesic-overuse headache” and specifying that “all non-opioid analgesics are regarded as a single class”. Consequently, the use of more than one non-opioid analgesic with every single drug being used on less than 15 days per month is classified as non-opioid analgesic-overuse headache (6).

This change in definitions and the long tradition in the management of MOH at our clinic dating back to 1989 (7) prompted us to examine changes in the characteristics of patients admitted for inpatient treatment of MOH over a period of 32 years.

Methods

The study was approved by the ethics committee of the Medical University of Vienna. We screened the diagnoses at the discharge of all patients admitted to the Department of Neurology of the Medical University of Vienna from 1 January 1984 to 31 December 2015 and identified all subjects with a diagnosis of MOH or corresponding previously used terms such as “drug-abuse headache” or “drug-induced headache”. For assessing the change in the patients' characteristics over time, we defined three periods ranging from 1984–1993 (P1), 1994–2003 (P2), and 2004–2015 (P3). These periods were commensurate with each other and reflected the years before and just after the publication of ICHD-1 (2), the subsequent era of ICHD-1 (2), and the time of ICHD-2 (3) and ICHD-3 beta (5).

We gathered biographic data, smoking status, body mass index (BMI), the time since headache onset, the headache diagnoses, and the details of the use of acute and prophylactic medication from the medical reports. We recorded the use of dihydroergotamine, ergotamine mono- and combination- preparations, almotriptan, eletriptan, frovatriptan, naratriptan, rizatriptan, sumatriptan, zolmitriptan, acetylsalicylic acid, diclofenac, ibuprofen, dexibuprofen, ketoprofen, mefenaminic acid, metamizole, naproxen, paracetamol, other non-opioid mono-preparations, tramadol, other opioids, and combination analgesics. Following ICHD-3 (6), we categorized these drugs as ergotamines, triptans, non-opioid analgesics, opioids, and combination analgesics. Also, we recorded the use of benzodiazepines and other tranquilizers. For each drug, we recorded the number of single doses used per month and the number of days per month the drug was taken. Besides, we calculated cumulative doses for all drugs and in each drug category. We defined use and overuse for every single drug and every drug category according to ICHD-3 (6).

Statistics

We analyzed all data with SPSS for Windows 24.0. Categorial data are shown as numbers and percentages and continuous variables as means and standard deviations (SD) or medians, lower and upper quartiles (Q1, Q3). For comparing P1, P2, and P3, we used chi-square tests for categorical variables, t-tests for normally and Kruskal-Wallis tests for non-normally distributed variables. To adjust for multiple comparisons, Bonferroni correction was performed online using SISA online statistical analysis following the suggestions of Sankoh et al. (8). Considering correlations between the variables, the adjusted p-value was set to 0.005.

Results

Admissions

Screening the diagnoses at discharge revealed 1028 admissions for possible MOH. Reviewing the medical reports did not confirm a diagnosis of MOH in 124 admissions. Accordingly, we included 904 hospital admissions of 787 patients. The number of admissions was 192 in P1, 348 in P2, and 364 in P3. The vast majority of patients (n = 699, 88.8%) was admitted once, 68 patients (8.6%) were admitted twice, and 20 (2.6%) three to five times.

Biographic data

At the time of the first admission, the mean age was 45.8 ± 12.1 years and 77.5% were female. Patients with one versus two or more admissions did not differ with respect to age (45.8 ± 12.2 vs. 45.6 ± 10.5; t = 0.2, p = 0.8) and the proportion of women (76.5% vs. 85.2%, chi² = 3.4, p = 0.07). The proportion of patients with just one admission between 1984 and 2015 was similar in the three study periods and ranged between 86.3% and 91.3%. For detailed biographic data see Table 1. The proportion of women increased from 69.9% in P1 to 79.2% in P3 (chi² = 7.8; p = 0.02). However, this difference did not remain statistically significant after correction for multiple testing.

Table 1.

Biographic characteristics of 787 patients with MOH in P1 (1984–1993), P2 (1994–2003), and P3 (2004–2015). Data presented refer to the first admission.

	Total		P1		P2		P3
	n = 787		n = 183		n = 292		n = 312
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Age (years)	45.8	12.1	45.8	11.8	47.5	12.2	45.5	12.1
	n	%	n	%	n	%	n	%
Female gender	610	77.5	128	69.9	235	80.5	277	79.2
Current relationship status
Married or living with partner	409	52.0	82	44.8	162	55.5	165	52.9
Single	180	22.9	37	20.2	67	22.9	76	24.4
Partner, but living alone	14	1.8	1	0.5	6	2.1	7	2.2
Unknown	184	23.4	63	34.4	57	19.5	64	20.5
Occupation
White-collar worker/official	283	36.0	71	38.8	107	36.6	105	33.7
Retiree	130	16.6	24	13.1	62	21.2	44	14.1
Unemployed	115	14.6	10	5.5	40	13.7	65	20.8
Housewife/househusband	50	6.4	7	3.8	18	6.2	25	8.0
Blue-collar worker	26	3.3	11	6.0	8	2.7	7	2.2
Self-employed	20	2.2	4	1.7	5	1.7	11	3.5
Apprentice/student	12	1.5	1	0.5	3	1.0	8	2.5
Other	8	1.0	1	0.5	2	0.7	5	1.6
Unknown	143	18.2	54	29.5	47	16.1	42	13.5

BMI and smoking

Among all admissions, the BMI was 26.0 ± 5.4 (women: 25.7 ± 5.6, men: 27.0 ± 4.5) and the proportion of smokers was 41.1% (women: 39.3%, men: 47.8%). In P1, P2, and P3, there was neither difference for the BMI (25.1 ± 5.1, 26.2 ± 5.6, and 26.1 ± 5.2), nor for the proportion of smokers (45.1%, 39.7%, and 40.9%, chi² = 1.2; p = 0.6).

Headache

MOH was the diagnosis in all of the medical reports and 95.2% of the medical reports included at least one additional headache diagnosis (Table 2). Besides MOH and the primary pre-existing headache, a third and fourth headache diagnosis was given in 18.7% and 1.4% of the medical reports.

Table 2.

Headache diagnoses in 904 admissions for MOH in P1 (1984–1993), P2 (1994–2003), and P3 (2004–2015).

	Total		P1		P2		P3
	n = 904		n = 192		n = 348		n = 364
Main diagnosis	n	%	n	%	n	%	n	%
MOH	904	100.0	192	100.0	348	100.0	364	100.0
Second diagnosis
Migraine	438	48.5	85	44.3	159	45.7	194	53.3
Tension-type headache	373	41.3	92	47.9	155	44.5	126	34.6
Other headaches	49	5.4	9	4.7	21	6.0	19	5.2
None	44	4.9	6	3.1	13	3.7	25	6.9
Third diagnosis
Migraine	62	6.9	9	4.7	32	9.2	21	5.8
Tension-type headache	90	10.0	10	5.2	34	9.8	46	12.6
Other headaches	19	2.1	4	2.1	6	1.7	9	2.5
None	733	81.1	169	88.0	276	79.3	288	79.1

Among all admissions, the primary pre-existing headache was migraine, TTH, and other headaches in 48.2%, 41.7%, and 5.3%. From P1 to P3, the proportion of migraine increased from 44.3% to 53.3% (chi² = 9.0, p = 0.01) and that of TTH decreased from 47.9% to 34.6% (chi² = 9.3, p < 0.01). Both results did not remain statistically significant after correction for multiple testing.

The median time since the first manifestation of the primary pre-existing headache disorder changed from 20 years in P1 to 15 years in P3 (p < 0.001). The median time since the first manifestation of MOH decreased from 3 years in P1 to 2 years in P3 (p < 0.001). Median time since the first manifestation of the primary pre-existing headache (p < 0.001) and since the first manifestation of MOH (p = 0.005) decreased also significantly when analyzed separately for non-triptan over-users. For further details see Table 3.

Table 3.

Years since onset of the main pre-existing headache (i.e. second diagnosis in Table 2) and years since onset of MOH in 904 admissions for MOH in P1 (1984–1993), P2 (1994–2003), and P3 (2004–2015).

	Total		P1		P2		P3		P1 vs P2 vs P3
	Median	Q1–Q3	Median	Q1–Q3	Median	Q1–Q3	Median	Q1–Q3	p-value¹
Pre-existing headache	20	10–30	20	10–30	20	11–33	15.5	8.3–27.8	<0.001
Migraine	20	15–32	23.5	15–30.5	25	16–35	20	10–30
Tension-type headache	16	8–30	20	10–32	20	9–31	10	5–20
Other headaches	11	5–25	7.5	4.3–17.3	13	10–30	11	4–26.5
MOH	3	1–6	3	1–8	3	1–7.3	2	1–5	<0.001

Kruskal Wallis Test.

Acute medication

Overuse of ergotamines, triptans, non-opioid analgesics, combination analgesics, and opioids was found in 27.9, 18.9, 45.6, 37.5, and 10.6% of the admissions. Overuse of two or more drug classes was present in 47%, whereas overuse of two or more drug classes not individually overused was rare (0.7%). The overuse of more than one non-opioid analgesic was mentioned in 13.5% of all medical reports and 28.2% of the reports referring to the use or overuse of non-opioid analgesics.

The drug classes overused changed markedly in the three study periods (Table 4). Ergotamine overuse decreased, whereas overuse of triptans, non-opioid analgesics, and opioids increased. Furthermore, there was a trend towards an increase in the overuse of multiple drug classes.

Table 4.

Medication overuse in 904 admissions for MOH in P1 (1984–1993), P2 (1994–2003), and P3 (2004–2015). Data are given for all drug classes and for the most common compounds within each class.

	Total		P1		P2		P3		P1 vs. P2 vs.
	n = 904		n = 192		n = 348		n = 364
	n	%	n	%	n	%	n	%	Chi²	p-value
Ergotamines	252	27.9	100	52.1	122	35.1	30	8.2	134.7	<0.001
Ergotamine combinations	174	19.2	63	32.8	91	26.1	20	5.5	77.7	<0.001
DHE combinations	78	8.6	44	22.9	26	7.5	8	2.2	69.4	<0.001
Triptans	171	18.9	5	2.6	50	14.4	116	31.9	77.8	<0.001
Eletriptan	68	7.5	N.A	0.0	11	3.2	57	15.7	59.8	<0.001
Sumatriptan	54	6.0	4	2.1	33	9.5	17	4.7	13.9	0.001
Zolmitriptan	48	5.3	N.A	0.0	4	1.1	44	12.1	56.0	<0.001
Non-opioid analgesics	412	45.6	74	38.5	140	40.2	198	54.4	19.3	<0.001
Mefenaminic acid	122	13.5	4	2.1	28	10.9	80	22.0	45.8	<0.001
Ibuprofen	90	10.0	19	9.9	36	10.3	35	9.6	0.1	>0.05
Metamizol	89	9.8	21	10.9	30	8.6	38	10.4	0.9	>0.05
Paracetamol	88	9.7	25	13.0	18	5.2	45	12.4	13.5	0.001
Combination analgesics	339	37.5	74	38.5	143	43.1	122	33.5	4.5	>0.05
ASA+Pct+Caffeine¹	240	26.5	45	23.4	101	29.0	94	25.8	2.1	>0.05
Opioids	96	10.6	5	2.6	41	11.8	50	13.7	17.2	<0.001
Tramadol	60	6.6	2	1.0	20	5.7	38	10.4	18.6	<0.001
Multiple drug classes	425	47.0	76	39.6	156	44.8	193	53.0	10.2	(0.006)²

ASA: acetylsalicylic acid; Pct: paracetamol.

Not statistically significant after correction for multiple testing.

N.A.: not available.

Comparing medication overuse in patients with and without other pain disorders at the time of the first admission showed more frequent overuse of non-opioid analgesics (58.4% vs. 43.7%; chi² = 8.6, p = 0.002) and a trend towards more frequent overuse of opioids (17.7% vs. 9.6%; chi² = 86.8, p = 0.01), whereas overuse of triptans, ergotamines, and combination analgesics did not differ in the two groups.

Considering all types of acute medication, the median cumulative number decreased over time from 120 single doses per month in P1 to 90 doses in P3 (p = 0.002). The median cumulative number of ergotamine doses decreased from 60 per month in P1 to 30 in P3 (p < 0.001). Median cumulative monthly doses of triptans, non-opioid analgesics, opioids, and combination analgesics did not differ statistically significantly in the three study periods. For more details see Table 5.

Table 5.

Cumulative monthly single doses of 24 compounds or groups of compounds in five drug classes in 904 admissions for MOH in P1 (1984–1993), P2 (1994–2003), and P3 (2004–2015).

	Total		P1		P2		P3		P1 vs. P2 vs.
Cumulative doses	Median	Q1–Q3	Median	Q1–Q3	Median	Q1–Q3	Median	Q1–Q3	p-value¹
Ergotamines	60	30–90	60	35–120	45	30–60	30	30–60	<0.001
Triptans	30	15–45	30	15–30	30	13.5–45	30	15–45	>0.05
Non-opioid analgesics	90	45–150	65	30–180	90	47.5–150	90	45–140	>0.05
Opioids	60	30–100	60	30–90	60	30–120	40	30–90	>0.05
Combination analgesics	100	60–180	120	60–180	120	60–210	90	40–150	(0.017)²
All drug classes	105	60–180	120	60–180	120	60–190	90	45–172.5	0.002

Kruskal Wallis Test.

Not statistically significant after correction for multiple testing.

Regular use of benzodiazepines and other tranquilizers was recorded in 7.2% of all admissions and was similar in the three study periods.

Prophylactic medication

Intake of prophylactic headache medication before or at the time of admission was given in 21.2% of the medical reports. The vast majority (77%) had only one prophylactic medication. The use increased from 8.3% in P1 to 29.9% in P3 (chi² = 89.5, p < 0.001). In detail, the use of amitriptyline and beta-blockers increased, the use of flunarizine remained unchanged, and small numbers did not allow comparisons for botulinum toxin, topiramate, and valproic acid (Table 6).

Table 6.

Use of prophylactic headache medication prior to or at the time of admission in P1, 1984–1993), P2 (1994–2003), and P3 (2004–2015).

	Total		P1		P2		P3		P1 vs. P2 vs. P3
	n = 904		n = 192		n = 348		n = 364
	n	%	n	%	n	%	n	%	Chi²	p-value
Amitriptyline	78	8.6	4	2.1	27	7.8	47	12.9	19.2	<0.001
Betablockers	92	10.2	6	3.1	38	10.9	48	13.2	14.2	0.001
Botulinum toxin	3	0.3	0	0.0	1	0.3	2	0.5	–	–
Flunarizine	34	3.8	8	4.2	13	3.7	13	3.6	0.1	>0.05
Topiramate	22	2.4	0	0.0	2	0.6	20	5.5	–	–
Valproic acid	23	2.5	0	0.0	7	2.0	16	4.4	–	–
Any of these drugs	191	21.2	16	8.3	139	18.4	109	29.9	89.5	<0.001

Besides, there was a marked increase in the use of other antidepressants. The prescription rate was 3.6% in P1, 13.2% in P2, and 34.9% in P3. In contrast, a diagnosis of depression or anxiety was similar in the three study periods and ranged between 31.3% and 33.5%. Antidepressants apart from amitriptyline were used more often in patients with depression/anxiety (36.9%) than in those without (11.6%), whereas the use of amitriptyline was similar in the two groups (10.8% vs. 7.8%).

Finally, the number of patients who had used more than one class of medication (including other antidepressants) increased markedly from 1.6% in P1, to 10.6% in P2, and 26.1% in P3.

Discussion

We retrospectively collected data of 787 patients with MOH, who accounted for a total of 904 admissions for inpatient detoxification over a period of 32 years. Thus, this represents one of the largest retrospective samples of MOH patients ever published covering a period of more than three decades. In 2004, Bigal et al. reported the characteristics of 456 patients with transformed migraine and compared patients treated in 1995 or before to those treated thereafter (9). Mesukunas et al. reported on 1200 patients with probable MOH from a tertiary headache center, but, unlike our study, diagnoses of MOH were based on ICHD-2 (10).

Biographic characteristics

The biographic characteristics were stable over a period of 32 years and similar to previously published data (10,11). Basically, MOH always was and still is a disorder mainly affecting middle-aged women with a long headache history.

BMI and smoking

While in the general population of Austria overweight (BMI 25–30) decreased over the last decades (12), the mean BMI was within the overweight range and did not change over time in our population. Several studies already showed an association between MOH and obesity (13 –15). This association could be explained by central and peripheral pathophysiological mechanisms shared by migraine and obesity, whose emerging players are adipocytokines and a generally pro-inflammatory state (16). Notably, obesity was not the only unhealthy condition reported to be common in MOH. Obesity and daily smoking were associated with MOH independently and in combination (14).

The proportion of smokers in our study was higher than in the general population of Austria (12). Epidemiological studies have repeatedly shown an association between MOH and smoking (13 –15,17). The question arises whether the high proportion of smokers among patients with MOH simply reflects a population prone to addictive behaviors, or if the association goes beyond that. Experimental data on capsaicin-induced pain highlight the role of nicotine deprivation resulting in increased pain sensitivity (18). Pain could, therefore, be a withdrawal-related symptom with deprivation-induced amplification of pain, thus leading patients to retain their smoking habits. However, these considerations are speculative, we could not find any data on the role of nicotine or nicotine withdrawal in MOH. Overall, harmful health behaviors seem to cumulate in MOH patients as well as in those with CM frequently associated with MOH (14).

Headache

The diagnostic category of the primary pre-existing headache tended to shift from TTH to migraine. This development may result from a growing awareness of migraine whereas TTH, even though the most common headache, fell behind. Previously published data already underlined the importance of migraine followed by TTH in MOH aetiology (1,19), but in our population, 20.1% of patients had a third or even fourth pre-existing additional headache type. We did not further classify these additional headache types and could not find any information in the literature regarding this MOH subclass with multiple pre-existing headache types.

The time since the first manifestation of both pre-existing headache and MOH decreased over time both in all patients and in non-triptan over-users, suggesting more awareness towards MOH and improved education of health care providers. However, considering that only a small minority of patients was ever advised to limit the intake frequency of acute headache medications (20), every effort has to be made to enhance educational work on MOH. Even though the debate is still going on about whether or not chronic headache can truly be induced by medication overuse, the perceived importance of MOH in Europe overcomes the importance observed in Northern America (21). Anyway, the European emphasis on MOH may not be completely wrong, considering that the uncritical use of opioids caused an epidemic in the US.

Treatment strategies shifted from an inpatient detoxification program as a standard- setting to outpatient programs. Nowadays, we offer inpatient detoxification preferably to those patients presenting with complicating factors such as co-morbid psychiatric disorders, opioid-overuse, or with difficult social conditions (22).

Acute medication

Interestingly, the median cumulative single doses continuously decreased over the past three decades. Usage of multiple over-the-counter medications, better information on possible side effects and complications, or the change in medication overuse pattern could partly explain this phenomenon. When analysed separately for substance classes, only the intake of median ergotamine single doses decreased.

While ergotamines were major players in migraine treatment in the past, they gradually lost their importance (9,13,23,24) owing to changes in European treatment guidelines (25). In parallel, triptan overuse significantly increased, reaching a proportion of up to 31% in our analysis. Several studies confirmed the importance of triptans in contemporary medication-overuse patterns (11,13,24). Triptan overuse headache typically develops faster and the amount of triptans needed to induce MOH is smaller than for analgesics (26). A survey based on nationwide healthcare claims data estimated that the proportion of migraine patients using a triptan was less than 6% in Austria and 5.9% of them were triptan over-users (27).

In a recent longitudinal study, triptan users showed high treatment efficacy and had better outcomes at one-year follow-up compared to those with poor acute treatment efficacy (as seen with non-opioid analgesics), who had an increased risk for new-onset CM (28). However, it remains speculative whether the early prescription of triptans to migraine patients could prevent chronification and overuse at least in those with high treatment efficacy. As triptan overuse generally becomes more frequent, the goal should be to identify patients with frequent migraine attacks, poor triptan response, and rebound headache to offer them timely prophylactic treatment, before overuse develops.

Non-opioid analgesics and combination analgesics were the top two drug classes overused in our study population, accounting for 45.6 and 37.5% of MOH, respectively. Both are main representatives of over-the-counter medications, often purchased without the need for prescription and thus easily available. Caffeine-containing substances in particular were blamed for inducing MOH (24), but in a population sample of adults with episodic and chronic daily headache their use was at least not associated with an increased risk of chronification in the adjusted analysis (29). According to epidemiological studies, almost all MOH patients purchased non-opioid analgesics and among patients from Europe and Latin America combination analgesics were frequently overused (11,23). The substance most frequently overused in our population, although not recommended in any guidelines, was mefenaminic acid followed by paracetamol, metamizole, and ibuprofen. The widespread use of mefenaminic acid seems to be specific for Austria, but we cannot provide an adequate explanation for this continuing popularity. In contrast, indomethacin and ibuprofen were most frequently overused in Italy (24). In Denmark, paracetamol and ibuprofen were among the most commonly purchased analgesics and the amounts purchased for MOH patients were four to 11-fold higher than the amounts dispensed to chronic headache patients without MOH (23,24). Differences in prescription habits, divergent insurance systems, and methodological issues could lead to this marked variation.

Overuse of multiple substance classes

More than half of the patients in our study reported the overuse of more than one substance class. Compared to this result the percentages reported in other studies were lower, ranging from 1.5–36.8% (9,11). This difference could relate to the fact that our study was conducted at a tertiary headache centre, whose patients were certainly severely affected. A possible relation between the MOH severity and overuse of multiple substance classes is supported by another study reporting this type of overuse in 50% of patients with previous withdrawal experience (24).

Notably, ICHD-3 beta (and ICHD-3) include criteria for overuse of multiple drug classes not individually overused, which was found in only 0.7% in our and 2.2% in another study, respectively (11). More information on this type of newly defined MOH is needed.

Opioids

In Europe, opioids are generally not recommended for headache treatment owing to their limited potential for relief from headache and their dependency potential (25,30). Therefore, it is surprising that the frequency of opioid overuse, expected to be no issue, varies from 1.5–33.3% (9,11,13,23), data not deriving solely from American studies, but rather also from Europe observing opioid overuse in up to 32% (23). Our current data show opioid overuse in 10.6% and, contradicting European treatment guidelines, it even increased over time and therefore well reflects the trend of non-critical opioid prescription for non-cancer pain (31). The question arises of why doctors would prescribe opioids to their headache patients. Maybe they are prescribed for other pain disorders in the first place, and then patients start self-administration to treat their headaches. Supporting this hypothesis, the overuse of opioids in our study tended to be more common in patients with other pain disorders than in those without. In a US-based survey, 17.7% of patients with chronic headache used opioids to treat not only a headache but also other pain conditions (29). An association of back pain, osteoarthritis, and rheumatoid arthritis with chronic migraine was also described in a Danish study (23), and in another recently published study MOH patients used analgesics for other pain conditions, muscle/joint pain and back pain in 23.9%, 22.6% and 21.5%, respectively (32). Besides, lacking familiarity with current treatment guidelines could contribute to generous opioid prescription policy. Also, the prescription of opioids together with antidepressants for co-morbid psychiatric disorders may be associated with the risk of serotonin toxicity (33). Educational campaigns warning of opioid-induced side effects may raise the awareness for uncritical opioid use.

Prophylactic treatment

A striking result of our study is the rare use of prophylactic medication prior to or at the time of admission. Prophylactic treatment was recorded in not more than 21% of the patients during the entire study period, though the proportion increased from 8% in P1 to (still low) 30% in P3. We wonder why the majority of our patients with a long headache history and alarming medication overuse did not receive prophylactic medication. Previously published data already described the under-treatment of CM and showed that prophylactic medication was rather the exception than the norm (9,20,34), raising concerns about insufficient treatment in primary care and neurological offices. Probably, physicians indeed do not recommend or inform patients about prophylactic medication or they do, but patients are not willing to take it. The monoclonal antibodies against calcitonin gene-related peptide could revolutionize prophylactic migraine treatment not only by enhancing acceptance and adherence but also by reducing the risk of medication overuse or by converting overuse to non-overuse (35).

Antidepressants

Many MOH patients suffer from co-morbid affective disorders whose presence is associated with poor treatment response and frequent relapse into MOH after detoxification (1,13,17,35). In the current analysis, the rate of affective disorders was similar in all three study periods. However, the usage of antidepressive treatment increased significantly from 4% in P1 to more than one third in P3. Considering this, affective disorders in patients with MOH obviously often remained untreated in earlier years.

Strengths and limitations

To the best of our knowledge, this is one of the largest clinic-based samples of consecutive MOH patients ever published. The study covers a period of more than three decades and reflects the evolution of MOH characteristics over time. The main limitations are the retrospective study design and the inclusion of inpatients from a tertiary headache centre, bearing the risk of a bias towards more severely affected patients. Accordingly, the results cannot be transferred to the general population.

Conclusion

This retrospective study in a large number of patients with MOH admitted for inpatient withdrawal therapy over a period of 32 years shows a trend towards changes in the preexisting headache type, a decrease in the time since onset of headache and MOH, a decrease in the number of drug doses used per month, changes in the type of the drugs overused and an increase in, but still low rate, of prophylactic medication prior to admission.

Footnotes

Public health relevance

This study highlights the evolution of medication overuse headache over a period of 32 years and represents one of the largest retrospective samples of MOH patients admitted for inpatient detoxification ever published.

The article clearly demonstrates the important role of triptans in contemporary medication overuse patterns.

Contradicting European treatment guidelines, opioid overuse in MOH patients increased over time. Therefore, every effort has to be made to reduce the intake of opioids in patients with chronic headache and non-cancer pain.

The low percentage of patients receiving prophylactic medication shows a still unmet need for more educational work on chronic migraine and MOH.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Diener

Limmroth

. Medication-overuse headache: A worldwide problem. Lancet Neurol 2004; 3: 475–483.

Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988; 8: 1–96.

Olesen

Steiner

. The International Classification of Headache Disorders, 2nd edition. J Neurol Neurosurg Psychiatry 2004; 75: 808–811.

Olesen

Bousser

, et al. Headache Classification Committee of the International Headache Society. New appendix criteria open for a broader concept of chronic migraine. Cephalalgia 2006; 26: 742–746.

Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: 629–808.

Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition. Cephalalgia 2018; 38: 1–211.

Baumgartner

Wessely

Bingol

, et al. Longterm prognosis of analgesic withdrawal in patients with drug-induced headaches. Headache 1989; 29: 510–514.

Sankoh

Huque

Dubey

. Some comments on frequently used multiple endpoint adjustment methods in clinical trials. Stat Med 1997; 16: 2529–2542.

Bigal

Rapoport

Sheftell

, et al. Transformed migraine and medication overuse in a tertiary headache centre – clinical characteristics and treatment outcomes. Cephalalgia 2004; 24: 483–490.

10.

Meskunas

Tepper

Rapoport

, et al. Medications associated with probable medication overuse headache reported in a tertiary care headache center over a 15-year period. Headache 2006; 46: 766–772.

11.

Find

Terlizzi

Munksgaard

, et al. Medication overuse headache in Europe and Latin America: General demographic and clinical characteristics, referral pathways and national distribution of painkillers in a descriptive, multinational, multicenter study. J Headache Pain 2016; 17: 20.

12.

Austria S. Gesundheitsbefragung 2014. Bevölkerung in Privathaushalten im Alter von 15 und mehr Jahren [Health survey 2014. Population in private households aged 15 and over]. http://www.statistik.at (2014, accessed 29 February 2020).

13.

Schwedt

Alam

Reed

, et al. Factors associated with acute medication overuse in people with migraine: Results from the 2017 Migraine in America Symptoms and Treatment (MAST) study. J Headache Pain 2018; 19: 38.

14.

Westergaard

Glumer

Hansen

, et al. Medication overuse, healthy lifestyle behaviour and stress in chronic headache: Results from a population-based representative survey. Cephalalgia 2016; 36: 15–28.

15.

Straube

Pfaffenrath

Ladwig

, et al. Prevalence of chronic migraine and medication overuse headache in Germany – the German DMKG headache study. Cephalalgia 2010; 30: 207–213.

16.

Dominguez

Vieites-Prado

Perez-Mato

, et al. Role of adipocytokines in the pathophysiology of migraine: A cross-sectional study. Cephalalgia 2018; 38: 1005–1006.

17.

Hagen

Linde

Steiner

, et al. Risk factors for medication-overuse headache: An 11-year follow-up study. The Nord-Trondelag Health Studies. Pain 2012; 153: 56–61.

18.

Ditre

Zale

LaRowe

, et al. Nicotine deprivation increases pain intensity, neurogenic inflammation, and mechanical hyperalgesia among daily tobacco smokers. J Abnorm Psychol 2018; 127: 578–589.

19.

Colas

Munoz

Temprano

, et al. Chronic daily headache with analgesic overuse: Epidemiology and impact on quality of life. Neurology 2004; 62: 1338–1342.

20.

Mehuys

Paemeleire

Van Hees

, et al. Self-medication of regular headache: A community pharmacy based survey. Eur J Neurol 2012; 19: 1093–1099.

21.

Loder

Scher

. Medication overuse headache: The trouble with prevalence estimates. Cephalalgia 2020; 40: 3–5.

22.

Rossi

Faroni

Tassorelli

, et al. Advice alone versus structured detoxification programmes for complicated medication overuse headache (MOH): A prospective, randomized, open-label trial. J Headache Pain 2013; 14: 10.

23.

Westergaard

Hansen

Glumer

, et al. Prescription pain medications and chronic headache in Denmark: Implications for preventing medication overuse. Eur J Clin Pharmacol 2015; 71: 851–860.

24.

Grazzi

Grignani

D'Amico

, et al. Is medication overuse drug specific or not? Data from a review of published literature and from an original study on Italian MOH patients. Curr Pain Headache Rep 2018; 22: 71.

25.

Evers

Afra

Frese

, et al. EFNS guideline on the drug treatment of migraine – revised report of an EFNS task force. Eur J Neurol 2009; 16: 968–981.

26.

Limmroth

Katsarava

Fritsche

, et al. Features of medication overuse headache following overuse of different acute headache drugs. Neurology 2002; 59: 1011–1014.

27.

Zebenholzer

Gall

Wöber

. Use and overuse of triptans in Austria – a survey based on nationwide healthcare claims data. J Headache Pain 2018; 19: 34.

28.

Lipton

Fanning

Serrano

, et al. Ineffective acute treatment of episodic migraine is associated with new-onset chronic migraine. Neurology 2015; 84: 688–695.

29.

Scher

Lipton

Stewart

, et al. Patterns of medication use by chronic and episodic headache sufferers in the general population: Results from the frequent headache epidemiology study. Cephalalgia 2010; 30: 321–328.

30.

Thorlund

Sun-Edelstein

Druyts

, et al. Risk of medication overuse headache across classes of treatments for acute migraine. J Headache Pain 2016; 17: 107.

31.

Franklin

. American Academy of Neurology. Opioids for chronic noncancer pain: A position paper of the American Academy of Neurology. Neurology 2014; 83: 1277–1284.

32.

Westergaard

Lau

Allesoe

, et al. Monitoring chronic headache and medication-overuse headache prevalence in Denmark. Cephalalgia 2020; 40: 6–18.

33.

Ansari

Kouti

. Drug interaction and serotonin toxicity with opioid use: Another reason to avoid opioids in headache and migraine treatment. Curr Pain Headache Rep 2016; 20: 50.

34.

Zebenholzer

Andree

Lechner

, et al. Prevalence, management and burden of episodic and chronic headaches – a cross-sectional multicentre study in eight Austrian headache centres. J Headache Pain 2015; 16: 531.

35.

Tepper

. CGRP and headache: A brief review. Neurol Sci 2019; 40: 99–105.

Medication overuse headache in 787 patients admitted for inpatient treatment over a period of 32 years

Abstract

Background

Objective

Methods

Results

Conclusion

Keywords

Introduction

Methods

Statistics

Results

Admissions

Biographic data

BMI and smoking

Headache

Acute medication

Prophylactic medication

Discussion

Biographic characteristics

BMI and smoking

Headache

Acute medication

Overuse of multiple substance classes

Opioids

Prophylactic treatment

Antidepressants

Strengths and limitations

Conclusion

Footnotes

Public health relevance

Declaration of conflicting interests

Funding

References