New Rating Scale Helps Evaluate Refractory Chronic Migraine Patients
Headaches are the most common neurological disorder seen by primary care physicians.1 Migraines, characterized as unilateral, pulsating, severe headaches with or without aura, occur in up to 18% of adult women and 10% of men.2 Up to 4% of adults will develop chronic migraines, defined by the International Classification of Headache Disorders as chronic daily headaches for at least 3 months, occurring greater than 15 days per month, where at least 8 of the days are migrainous.2 However, there is a subset of chronic migraine patients who go on to develop refractory chronic migraines (RCM) and are perhaps the most difficult migraine patients to treat.3
In an effort to better understand and study this disorder, the American Headache Society has proposed a set of 5 diagnostic criteria for RCM, which are outlined in Table 1.4,5 Patients filling these criteria satisfy the diagnosis of chronic migraine but have failed adequate trials of both abortive and preventive migraine treatments. In addition, patients with RCM are more likely to have suffered from migraines over many years, have migraines more days per month, and have medication overuse headaches (MOH). RCM patients frequently suffer from a comorbid medication condition, such as irritable bowel syndrome (IBS), fibromyalgia, temporomandibular disorders (TMD), chronic pelvic pain, painful bladder syndrome, and chronic fatigue, as well as psychiatric comorbidities, such as anxiety, depression, bipolar spectrum, personality disorders, somatization, and post-traumatic stress disorders.3
To better measure the severity of the disorder, we developed a novel rating scale for RCM severity. In this article, we present the results of our 10-year study, as well as the use of the rating scale to evaluate patients with RCM.
Why Categorize Patients
RCM often is a debilitating illness, with an enormous impact on quality of life (QOL). Long-term outcomes for patients with RCM have not been thoroughly studied. In addition, RCM ranges in severity from mild to severe. For clinical and research purposes, it is important to categorize RCM patients according to severity. Knowing the severity of the disease helps clinicians better plan treatments. There are a number of major challenges clinicians face when dealing with RCM, with each diagnostic category requiring a different approach.3 These challenges include:
- Determining the role of disability, and whether disability should help to define RCM
- Determining how resistant to the myriad of treatments a patient has to be to be considered to have RCM
- There is no accepted, identifiable biological marker for RCM
- The degree of refractoriness can change over time, improving or worsening. What role does this varying severity play?
- There are various subsets of RCM—post-traumatic headache, RCM with or without MOH, RCM with or without major psychiatric comorbidities, etc.
From 2000 to 2010 at our headache center, we followed 129 patients (109 women) who were 18 years or older at start of study and who met criteria for RCM. The study design and the results of the study were recently published in the Journal of Headache Pain.6 A refractory scale of this author’s design was used for assessment. The scale ranges from 2 (least severe) to 10 (most severe) (Table 2).5,7-10 The patients were assigned a number between 2 and 10 at the start of the study and then were reassessed in 2010. The severity groupings were as follows: a score of 2, 3, or 4 was defined as mild RCM; a score of 5, 6, 7 was defined as moderate RCM; and a score of 8, 9, 10 was defined as severe RCM.
QOL was measured by adding pain, functioning, and mood scores (each on a 1-10 scale, with 1=best, 10=worst). The QOL rating scale ranged from 3 (best) to 30 (worst). Pain was assessed via a visual analog scale of 1 to 10 (10=worst). The patient’s ability to function was determined by the level of the work and/or home activities. Mood assessment included symptoms of depression, anxiety, and insomnia using the Fourth Edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria.
Because this was a retrospective study, treating physicians collected the data, which we then ‘‘de-identified’’ and collected as anonymous ‘‘batch’’ data. We obtained informed consent and consulted our local Institutional Review Board. To analyze if these pre–post scores differed across the 3 pain severity groups (mild, moderate, severe), we conducted a one-way analysis of variance (ANOVA). To determine if treatment was significantly effective at decreasing the level of pain and improving QOL, we calculated pre–post paired sample t tests for each severity group and then used Cohen’s effect size formula for paired sample t tests.
Of the 129 patients included in the study, 24 (18.5%) were categorized as mild (average score, 3.79), 67 (52%) as moderate (average score 6.04), and 38 (29.5%) as severe (average score, 9.02). Table 3 summarizes the changes in QOL over the 10-year period. Overall QOL improvement was 33%. Broken down in more detail, of the mild RCM patients, 66% improved by 30% or more in QOL during the 10 years. In the moderate severity group, 57% improved by 30% or more, and in the severe group, 61% improved by 30% or more. QOL over 10 years was the same, or worse, in 4% of those in the mild group, 16% of those in the moderate group, and 18% of those in the severe group.6
Improvements in pain scores were even more dramatic (Table 4). Over the 10-year study period, pain levels declined by ≥30% in 80% of the mild RCM group, 72% of the moderate group, and 71% of the severe group. Pain levels were the same, or worse, over the 10 years in 4% of the mild group, 15% of the moderate group, and in 18% of the severe group.6
Overall, the results suggest that the treatment was effective in improving QOL and reducing level of pain for all severity groups. Most of the RCM patients (60%) had at least a 30% improvement in QOL, whereas 73% experienced a ≥30% improvement in pain levels. While the patients categorized as severe also improved over 10 years, they had significantly lower QOL and higher pain scores than patients categorized as mild or moderate.6