OnabotulinumtoxinA effectiveness on chronic migraine, negative emotional states and sleep quality: a single-center prospective cohort study

Background OnabotulinumtoxinA (OnabotA) is considered effective in in patients with chronic migraine (CM) who failed on traditional therapies. This study was designed to evaluate the effect of OnabotA injection series on migraine outcome, negative emotional states and sleep quality in patients with CM. Methods A total of 190 patients with CM (mean (SD) age: 39.3 (10.2) years; 87.9% were female) were included. Data on Pittsburgh sleep quality index (PSQI), headache frequency and severity, number of analgesics used, Migraine Disability Assessment Scale. (MIDAS) scores and Depression, Anxiety and Stress Scale (DASS-21) were evaluated at baseline (visit 1) and 4 consecutive follow up visits, each conducted after OnabotA injection series; at week 12 (visit 2), week 24 (visit 3), week 36 (visit 4) and week 48 (visit 5) to evaluate change from baseline to follow up. Results From baseline to visit 5, significant decrease was noted in least square (LS) mean headache frequency (from 19.5 to 8.4, p = 0.002), headache severity (from 8.1 to 6.1, p = 0.017), number of analgesics (from 26.9 to 10.4, p = 0.023) and MIDAS scores (from 67.3 to 18.5, p < 0.001). No significant change from baseline was noted in global PSOI and DASS-21 scores throughout the study. Conclusions Our findings revealed that OnabotA therapy was associated with significant improvement in migraine outcome leading to decrease in headache frequency and severity, number of analgesics used and MIDAS scores. While no significant change was noted in overall sleep quality and prevalence of negative emotional states, patients without negative emotional states at baseline showed improved sleep quality throughout the study. Electronic supplementary material The online version of this article (doi:10.1186/s10194-017-0723-4) contains supplementary material, which is available to authorized users.


Background
Chronic Migraine (CM) is a complex and severely disabling neurological disorder, characterized by occurrence of headache on ≥15 days per month for >3 months with at least five attacks fulfilling criteria of migraine without aura on ≥8 days per month [1].
Increased comorbidity between migraine and several psychiatric conditions has consistently been reported with higher prevalence of mood and anxiety disorders, personality disorders and post-traumatic stress disorder (PTSD) among migraineurs than in the general population [11][12][13]. Also, increased headache frequency was shown to be correlated with higher rate of depression, anxiety and post-traumatic stress disorder [5,[14][15][16].
Recent studies also indicate a higher prevalence of poor sleep quality in patients with than without migraine and association of frequency of migraine headache with poor sleep quality [17][18][19][20][21][22].
This seems notable given the association of comorbidities with increased burden of CM in terms of productivity loss, impaired HRQoL, healthcare utilization and emotional burden 5 as well as treatment complications and poor clinical outcomes [13].
Despite availability of preventive and abortive treatments as the mainstays of treatment for migraines, onethird of migraine sufferers remain symptomatic due to frequent partial response to treatment that leads to physical disability and high risk of medication overuse [22][23][24]. Therefore, development of new and effective therapeutic alternatives is of particular significance in the management of CM [24,25].
Onabotulinumtoxin A (OnabotA) recently was discovered to be effective in preventing recurrent migraines in patients with CM who failed on traditional therapies [22,26,27]. Hence, based on data from PREEMPT (Phase 3 REsearch Evaluating Migraine Prophylaxis Therapy) trials confirming its efficacy in reduction of number of headache days and migraine days in CM patients with favorable safety and tolerability [28][29][30][31], OnabotA (155-195 U) is specifically indicated as a prophylactic treatment for CM in adults [26,32].
The present study was designed to evaluate the effect of OnabotA injection series on migraine outcome, negative emotional states and sleep quality in patients with CM.

Study population
A total of 190 consecutive patients with CM (mean (SD) age: 39.3 (10.2) years; 87.9% were female) were included in this single-center prospective cohort study conducted between May 2012 and May 2016. After baseline visit at study enrollment (visit 1), patients were followed up for 48 weeks via 4 consecutive follow up visits each conducted after a new OnabotA injection series; at week 12 (visit 2), week 24 (visit 3), week 36 (visit 4) and week 48 (visit 5), respectively. Most of the patients had a history of prophylactic treatment for migraine which was either insufficient or was discontinued due to intolerance.
Written informed consent was obtained from each subject following a detailed explanation of the objectives and protocol of the study which was conducted in accordance with the ethical principles stated in the "Declaration of Helsinki" and approved by the Acibadem University Clinical Research Ethics Committee.

Study parameters
Data on patient demographics (age, gender), educational status, diagnosis (CM, CM + medication overuse), migraine duration (year), family history for migraine, previous migraine treatments, migraine triggers are collected at baseline visit (visit 1). Data on migraine outcome [headache frequency and severity, number of analgesics used and Migraine Disability Assessment Scale (MIDAS) scores], sleep quality [Pittsburgh sleep quality index (PSQI)] and negative emotional states [Depression, Anxiety, and Stress Scale-21 (DASS-21)] were evaluated at baseline and follow up visits. Medication overuse was baseline of simple analgesics on ≥ 15 days. Follow up period was composed of 4 consecutive visits each conducted after a new OnabotA injection series; at week 12 (visit 2), week 24 (visit 3), week 36 (visit 4) and week 48 (visit 5), respectively. Change from baseline to follow up was evaluated based on data from visit 1 to visit 3 (24 weeks) for PSQI, from visit 1 to visit 4 (36 weeks) for DASS-21 scores, while based on data from visit 1 to visit 5 (48 weeks) for headache frequency and severity, number of analgesics used and MIDAS score. Data at visit 2, visit 3, visit 4 and visit 5 refer to changes in parameters observed after 1 st , 2 nd , 3 rd and 4 th OnabotA injection series, respectively.

Diagnosis of CM
CM was diagnosed based on International Classification of Headache Disorders (third revision) (ICHD-3 beta) diagnostic criteria that require headache occurring on ≥15 days per month for >3 months with at least five attacks fulfilling criteria of migraine without aura on ≥8 days per month [1].

OnabotA injection series
Administration of OnabotA was performed as 31 fixedsite, fixed-dose intramuscular injections applied at seven specified head and neck muscle points every 12 weeks for a minimum of 24 weeks (2 treatment cycles) according to injection scheme proposed in the PREEMPT studies [28,29,33]. We additionally administered OnabotA among occipitalis, temporalis or trapezius muscles using a follow-the-pain strategy.

MIDAS
The MIDAS is a 5-question tool to quantitatively evaluate the headache-related disability in terms of the number of days in the past 3 months and activity limitations due to migraine. MDAS was developed by Stewart et al. [34] and validated and checked for reliability for Turkish by Ertas et al. [35] The score obtained can be graded as follows: grade I (0 to 5 days) indicative of little or no disability; grade II (6 to 10 days), mild disability; grade III (11 to 20 days), moderate disability; and grade IV (greater than 21 days), severe disability.

DASS-21
The DASS-21 is a 21-item questionnaire which includes three self-report scales designed to measure the negative emotional states of depression, anxiety and stress. The Depression scale includes items that measure symptoms typically associated with dysphoric mood (e.g., sadness or worthlessness). The Anxiety scale includes items that are primarily related to symptoms of physical arousal, panic attacks, and fear (e.g., trembling or faintness). Finally, the Stress scale includes items that measure symptoms such as tension, irritability, and a tendency to overreact to stressful event. Each item is scored on a 4point scale (0 = Did not apply to me at all, to 3 = Applied to me very much or most of the time) to rate the extent to which they have experienced each state over the past week. Sum of the score of each item reveals the total score with higher scores indicating greater levels of distress. Each state is categorized into normal/mild/moderate/severe/extremely severe based on cut-off scores recommended for depression (0-4/5-6/7-10/11-13/14+), anxiety (0-3/4-5/6-7/8-9/10+) and stress (0-7/8-9/10-12/ 13-16/17+) [36][37][38]. Psychometric properties of the Turkish version of the DASS was studied by Hekimoglu et al. and DASS was shown to be an excellent instrument for measuring features of depression, hyperarousal, and tension in clinical groups [39].

PSQI
The PSQI, used to assess sleep quality, was developed by Buysse et al. [40] and validated and checked for reliability for Turkish by Agargun et al. [41]. The scale consists of 24 items; eighteen items are scored and yield seven component scores. Each component is scored between 0 and 3 and the total of these scores gives the scale score. The total score ranges between 0 and 21 and the higher the score is, the worse the sleep quality. A total score under 5 indicates 'good sleep quality' , while a score above 5 shows 'poor sleep quality'.

Statistical analysis
Statistical analysis was made using IBM SPSS Statistics (IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Version 21.0, Armonk, NY: IBM Corp). Change over time analysis was based on number of patients with available data on both baseline (visit 1) and follow-up visits (visit 2 to 5), and performed via repeated measures variance analysis and McNemar test for continuous and categorical variables, respectively. Since time to follow up visits showed variability among patients, change over time analysis of continuous variables were adjusted for time (visit 2 = 12 weeks; visit 3 = 24 weeks; visit 4 = 36 weeks; visit 5 = 48 weeks). No specific procedure was defined for missing data. Data were expressed as n(%), mean (standard deviation; SD), median (inter-quartile range, IQR) and mean (lower and upper boundaries of 95% confidence interval; CI) where appropriate. p < 0.05 was considered statistically significant.
During the study period overall four injection series were applied per patient within a total treatment time of 48 weeks (median (IQR) 62.4 (53.8-85.2) weeks).
Most of the participants were university graduates (41.1%), diagnosed with CM per se (48.9%) without triggers (76.8%) and for median (IQR) 3.0 (1.0-10.0) years. Family history for migraine was evident in 28.9% of patients; while most common previous medication was non steroid anti-inflammatory drugs (88.4%). Medication overuse was evident in 76.2% of patients (Table 1).
Mean headache severity was significantly decreased from 8.1 at baseline to 6.2 at visit 2 (p < 0.001), to 5.8 at visit 3 and visit 4 (p < 0.001 for each) and to 6.1 at visit 5 (p = 0.017) ( Table 3).
Based on patients with valid data for both baseline and follow up visits, no significant change was noted in percentage of patients categorized to have normal anxiety, depression and stress scores at each follow up visit (Table 5).
Based on patients with valid data for both baseline and follow up visits, no significant difference was noted in global scores and thus overall sleep quality from baseline to visit 2 or visit 3. Considering component scores, mean (95% CI lower bound-upper bound) scores for subjective sleep quality (from baseline 1.  (Table 7).

Sleep quality with respect to DASS-21 scores
Good sleep quality was noted at visit 1, visit 2 and visit 3 34.3, 40.0 and 57.1% of patients with normal depression scores at visit 1, in 38.7, 45.5 and 50.0% of patients with normal anxiety scores at visit 1, and in 33.3, 46.2 and 60.0% of patients with normal stress scores at visit 1, respectively (Table 8).

Adverse events and treatment alterations under OnabotA therapy
OnabotA therapy was associated with minor and temporary side effects (e.g., asymmetry of the position of the eyebrows and neck ache) in some patients. In one patient who has a very thin cervical region, dysphagia and difficulty in swallowing appeared while regressed after the third week of therapy. Treatment was continued in this patient with omission of further injections to the cervical area.

Discussion
Our findings revealed that administration of four Ona-botA injection series over 48 weeks in chronic migraineurs was associated with a significant improvement in all migraine parameters including headache frequency and severity, number of analgesics used and MIDAS scores, while no significant change from baseline was noted in overall sleep quality and prevalence of negative emotional states. Patients without negative emotional states at baseline showed improved sleep quality throughout the study. Significant reduction in monthly headache frequency, headache severity, number of analgesics used and MIDAS scores in our cohort support the efficacy of OnabotA in reduction of the frequency, intensity, and duration of chronic migraines as well as decreasing medication overuse among CM patients who failed on traditional preventive therapies [22, [25][26][27][42][43][44].
Reduction in MIDAS scores, indicating lesser amount of lost work and personal time due to migraine [6], after OnabotA therapy in our cohort seems notable given that a day lived with severe migraine is considered to be as disabling as a day lived with dementia, quadriplegia or acute psychosis and more disabling than blindness, paraplegia, angina or rheumatoid arthritis [4,45,46]. CM has been associated with a high intake of abortive medications with estimated analgesic overuse in 50-80% of patients with CM that may lead to the development of medication overuse headache (MOH) [47]. Given the similar rates of medication overuse (76.2%) in our cohort at baseline, reduction in in number of analgesics used starting from the first injection and consistently throughout the study period emphasizes the potential role of OnabotA therapy in prevention of MOH via headache episode reduction [25,42,[48][49][50].
In terms of monthly headache frequency, headache severity, number of analgesics used and MIDAS scores our findings support the association of OnabotA therapy with rapid improvement in migraine parameters usually after the first session as reported in past studies [26,27,30,[51][52][53].
The rapid improvement of migraine parameters after OnabotA therapy has been linked to the combined pharmacological and the placebo effect at the beginning of the treatment [51,52], while increased benefit offered via repeated OnabotA injections over time is associated with a prophylactic cumulative effect [31,54]. Also, patients who respond well to therapy early have been suggested to maintain and continue these reductions over at least 2 years [27,[55][56][57], while limited data are available on long-term efficacy of OnabotA therapy with evaluation of efficacy only up to1 year in most of clinical trials [42,44,50].
Alike to our findings, use of four OnabotA injections over 48 weeks in CM patients was reported to be associated with significant improvement in monthly headache days, migraine days and medication days that continued throughout the entire study period [42]. Nonetheless, in a longer-term study with seven OnabotA injections in CM patients, based on a decrease in initial efficacy of OnabotA therapy after the third injection, authors concluded that actual improvement and amelioration of daily headache under OnabotA therapy needs several months to be consolidated [52].
Long-term efficacy of OnabotA in CM patients has recently been evaluated in some studies based on administration of OnabotA therapy for 2 years [25], 4 years [43] and 7-9 injections series [52,58]. However, while finding are consistent regarding long-term efficacy of therapy with no serious adverse events in responders, they varied in terms of durability of benefit, non-responder rates and the length and necessity of withdrawal of treatment at scheduled times [25,43,52,58].
Both depression and anxiety have been proposed to be a risk factor for migraine chronification [13,59,60], while migraine headache frequency and headache severity were reported to be associated with higher rate of depression and anxiety disorders [4,5,14,61]. Accordingly, identification of abnormal scores for depression, anxiety and stress in almost half of patients in our cohort is in line with higher rates of selfreported mood and anxiety disorders in patients with CM as compared with general population and EM patients [4,5,13,14] with more than 40.0% of CM patients to meet criteria for moderate to severe anxiety and depression [4,13].
Reduction in headache frequency via 24-week OnabotA therapy was reported to be associated with reduction in depression (via Beck Depression Inventory-II) and anxiety scores (Generalized Anxiety Disorder-7 scale) in a past among CM patients [62]. In fact, even a direct effect of OnabotA injection in amelioration of depressive symptoms has been suggested that probably occur via the facial feedback mechanism with consideration of OnabotA as a  IQR inter-quartile range, n patient count without missing data safe adjunctive treatment to pharmacotherapy for major depressive disorder [63].
In this regard, it seems notable that despite significant reductions obtained via OnabotA therapy in both monthly headache frequency and headache severity in our cohort, no significant change from baseline occurred in depression, anxiety and stress rates throughout the study.
High rates of negative emotional states regardless of the ongoing OnabotA therapy in our patients seems notable given the impact of psychiatric comorbidities on disease prognosis, treatment, and clinical outcomes and higher prevalence of severe headache-related disability, headache impact and poor quality of life in migraineurs with than without psychiatric comorbidities [11,13,64]. Hence our findings emphasize consideration of comorbid psychiatric disorders in diagnostic evaluation and formulation of treatment plan in CM patients, being aware of the likely negative impact of co-morbid psychiatric disorder on treatment outcomes, adherence and quality of life [5,14,65].
Identification of poor sleep quality at baseline in 72.9% of our patients seems consistent with high prevalence (30-79%) of co-morbid poor sleep quality among migraineurs [20,[66][67][68], particularly in those with 8 or more migraine days per month [17]. Higher prevalence of poor sleep quality was also noted in patients with migraine compared to those without migraine [17][18][19][20][21], while high migraine frequency was shown to correlate with poor sleep quality and a higher prevalence of poor sleepers in chronic migraineurs [21].
In a past study on the effects of OnabotA on jaw motor events during sleep in sleep bruxism patients, no significant change from baseline was noted in usual sleep variables such as sleep efficiency, arousal index, sleep stages, or awakenings per hour during follow-up recordings [69].
Similarly, our findings revealed no direct effect of OnabotA injection on overall sleep quality at week 12 and week 24, while significant improvement in subjective sleep quality, sleep latency and sleep disturbance at week 12. This seems notable given the reported association of higher migraine frequency with higher scores for certain domains of PSQI such as "cannot get to sleep within 30 minutes," "wake up in the middle of the night or early morning," and "bad dreams" among migraineurs [21].
Sleep-related migraine is considered to be associated with a more severe and disabling clinical course given the increased mean attack severity and monthly use of   LS mean least square mean, n patient count with valid data at each visit, P p value of repeated measures variance analysis with reference to baseline symptomatic drugs in patients with than without sleep related migraine despite similar monthly headache frequency [70]. Although data on sleep quality are available for only two OnabotA injections in our cohort, given that patients without negative emotional states at baseline showed improved sleep quality throughout the study, maintenance of co-morbid negative emotional states throughout the study seems to be associated with lack of improvement in overall sleep quality, despite significantly decreased migraine frequency.
Nonetheless, it should also be noted that while the presence of negative emotional states such as depression was reported to be reciprocally associated with poor sleep quality [66,71] and migraine history and comorbid anxiety and depression were shown as predictors of sleep quality [20] in some studies, poor sleep quality was shown to be associated uniquely with migraine itself, regardless of comorbid depression, anxiety or sleep disorders in other studies, particularly among EM patients [17,44,72,73].
Certain limitations to this study should be considered. First, due to observational nature, non-randomized allocation and thereby the likelihood of main selection bias and confounding is possible. Second, although provide data on real-life clinical practice, potential lack of generalizability seems another important limitation due to single-center design of the study. Third, lack of intervention considering timing and number of follow up visits in accordance with the observational nature caused Table 6 Overall PSQI scores at study visits  relatively limited follow-up data and the frequency of patient visits to be not uniform with considerable loss to follow up rate challenging analysis of sleep and mood variables. Fourth, analysis of data on negative emotional states and sleep quality were based on self-report rather than objective measures. However, our analysis was based on use of validated questionnaires along with evidence on high concordance between self-report instruments and clinical diagnosis of psychiatric disorders [13]. Lack of follow up data on sleep quality after second OnabotA injection, lack of data on inter-individual variations in headache characteristics with likely impact on therapeutic response as well as lack of data on adverse events are other limitations which otherwise would extend the knowledge achieved in the current study.

Conclusions
In conclusion, our findings in a cohort of chronic migraineurs revealed that OnabotA injection was associated with significant improvement in migraine outcome  leading to decrease in headache frequency and severity, number of analgesics used and MIDAS scores. While no significant change was noted in overall sleep quality and prevalence of negative emotional states with OnabotA injections, patients without negative emotional states at baseline showed improved sleep quality throughout the study. There is a need for future larger-scale long-term longitudinal studies addressing the durability as well as predictors of efficacy of OnabotA in CM patients.

Availability of data and materials
The dataset(s) supporting the conclusions of this article is included within the article and its additional file 1.
Authors' contributions EIA contributed to conception/design of the research; EIA, PYD, ASK and SK contributed to acquisition, analysis, or interpretation of the data; EIA, PYD, ASK and SK drafted the manuscript; EIA, PYD and ASK critically revised the manuscript; and EIA agrees to be fully accountable for ensuring the integrity and accuracy of the work. All authors read and approved the final manuscript.