Chronic migraine (CM) is a kind of repeated headache disorder, with a high disability rate among the population [
Repeated migraine episodes can increase the risk of cerebrovascular diseases, such as stroke and increasing white matter lesion [
CM-MOH is the most common form of chronic migraine and is usually accompanied with increased clinically silent lesions [
This cross-sectional study was conducted between December 2015 and January 2017. Patients were recruited from the neurology outpatient clinic, the First Affiliated Hospital of Sun Yat-sen University. They were divided into three groups according to the criteria of the 3rd edition beta version of International Classification of Headache Disorders (ICHD-III beta) [
We collected demographical data of all participants at the first visit, including age, sex, education, job, weight, height, relevant medical history, and family history. A detailed questionnaire was used to record their headache characteristics, including headache years, location, headache nature, headache duration, comprised symptoms, headache frequency, monthly headache days, and monthly headache attacks. Analgesics used to control headache and their doses were recorded. For CM-MOH patients, monthly analgesic pills doses were calculated.
Several scales evaluating life quality and psychiatric status were conducted in patients. The Migraine Disability Assessment Test (MIDAS) was used to assess the headache frequency in three months and how often it limited their participation in daily activities. The Hamilton Anxiety Scale and Hamilton Depression Scale were used to display their mood state. The Short Form (36) Health Survey (SF-36) was used to evaluate the participants’ health statement. Results of SF-36 were divided into eight sections, including vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, and mental health. The Pittsburgh Sleep Quality Index (PSQI) was used to assess the sleep quality of participants.
Addenbrooke’s Cognitive Examination (ACE-R) is a set of tests for cognitive dysfunction screening. With a total score of 100 points, it is composed of five elements, including attention/orientation (18 points), memory (26 points), verbal fluency (14 points), language (26 points), and visuospatial abilities (16 points) [
To reduce the likelihood of fatigue in participants, after collecting basic information, we performed the cognitive evaluation at the sequence of ACE-R, TMT A + B, and DST, followed by neuropsychology assessment.
All statistical analysis was performed on IBM SPSS 24.0 for Windows (SPSS Inc., Chicago, IL, USA). Categorical variables such as sex, education, smoking history, and alcohol history were presented with frequency. Continuous variables followed the normal distribution and were presented with mean ± standard deviation; otherwise, presented with median (interquartile range). One-way analysis of variance (ANOVA), Kruskal–Wallis test, and chi-squared tests were used to evaluate the differences among groups, respectively. For the pairwise comparison between groups, the Tukey–Kramer test after ANOVA and Mann–Whitney test after Kruskal–Wallis test were performed, following with Bonferroni correction of the
In this study, there were 116 participants, including 21 in CM-MOH, 20 in CMwoMOH, 35 in MO, and 40 in control. The baseline characteristics are shown in Table
Baseline characteristic of the participants.
CM | MO | Control | |||
---|---|---|---|---|---|
CM-MOH | CMwoMOH | ||||
No. | 21 | 20 | 35 | 40 | |
Age, year, mean ± SDa | 48.90 ± 13.51 | 48.40 ± 10.33 | 45.89 ± 7.10 | 47.10 ± 7.04 | |
Female, %b | 80.95 | 80.00 | 77.14 | 77.50 | |
Education, year, %b | 0–6 | 19.05 | 20.00 | 17.14 | 17.50 |
7–9 | 52.38 | 50.00 | 51.43 | 52.50 | |
10–12 | 14.28 | 20.00 | 20.00 | 17.50 | |
>12 | 14.28 | 10.00 | 11.43 | 12.50 | |
Hypertension, %b | 14.28 | 10.00 | 0 | 0 | |
Diabetes, %b | 0 | 0 | 0 | 0 | |
High LDL, %b | 9.52 | 10.00 | 11.43 | 0 | |
Smoke, %b | 9.52 | 5.00 | 11.43 | 7.50 | |
Alcohol, %b | 0 | 0 | 2.86 | 2.50 | |
Family history, %b | 52.38 | 25.00 | 28.57 | NA |
aANOVA,
All the cases in our study took NSAIDs, such as aminopyrine, phenacetin, aspirin, ibuprofen, and acetaminophen, to relieve headache. Most of them were compound preparation. The VAS score, drug dosages, and frequency were higher in CM-MOH, compared with CMwoMOH and MO, respectively (
Headache characteristic of cases.
CM | MO | ||
---|---|---|---|
CM-MOH | CMwoMOH | ||
Headache years, mean ± SDa | 22.67 ± 12.27 | 17.40 ± 10.68 | 12.54 ± 9.08 |
VAS, mean ± SDb | 9.05 ± 1.24 | 7.30 ± 1.95 | 7.71 ± 1.51 |
Duration, day, mean ± SDc | 1.05 ± 1.15 | 1.51 ± 0.78 | 1.63 ± 1.01 |
Headache frequency, days/month, median (IQR)d | 30 (20–30) | 20 (15–30) | 3 (1–4) |
Dosage, pills/attack, median (IQR)e | 3 (2–6) | 1 (0–1) | 1 (0–1) |
Analgesic frequency, days/month, median (IQR)e | 30 (20–30) | 1.5 (0–4.5) | 1 (0–3) |
aANOVA, CM-MOH vs. MO,
As shown in Table
Neuropsychological assessment of casesa.
CM | MO | |||
---|---|---|---|---|
CM-MOH | CMwoMOH | |||
Anxietyb,1,3 | 12 (5–16.5) | 6 (4.3–8) | 4 (3–6) | |
Depressionb,1,2 | 4 (2.5–13.5) | 2 (2–3) | 1 (0–2) | |
MIDAS | 0 (0–180) | 12 (0–47.3) | 6 (3–18) | |
SF-36 | Physical functioningb | 90 (85–95) | 90 (75–93.8) | 95 (90–100) |
Physical role | 75 (12.5–100) | 50 (25–75) | 50 (25–100) | |
Body pain | 40 (22–68) | 51 (42–54) | 51 (41–74) | |
General healthb,1,3 | 45 (20–50) | 30 (20–48.8) | 52 (40–70) | |
Vitalityb,1,3 | 55 (40–75) | 50 (45–60) | 80 (60–80) | |
Social roleb | 67 (44–89) | 72.5 (56–78) | 78 (78–89) | |
Emotional roleb,1 | 33 (16.5–33) | 49.5 (33–66) | 100 (33–100) | |
Mental healthb,3 | 60 (48–76) | 50 (48–61) | 72 (56–80) | |
|
||||
PSQI | Overall sleep quality | 2 (1–2) | 1 (1–1.8) | 1 (1–2) |
Sleep latency | 2 (1–3) | 1 (0–2.8) | 1 (0–2) | |
Duration of sleep | 1 (0–2) | 1 (0–1) | 0 (0–1) | |
Sleep efficiencyb,1,2 | 1 (0–1) | 0 (0–0) | 0 (0–0) | |
Sleep disturbance | 1 (1–2) | 1 (1–1) | 1 (1–1) | |
Need meds to sleep | 0 (0–1.5) | 0 (0–0) | 0 (0–0) | |
Day dysfunction due to sleepiness | 1 (0–1) | 0 (0–1) | 0 (0–1) | |
Totalb,1 | 7 (5–9.5) | 4.5 (3–7) | 5 (3–7) |
aKruskal–Wallis tests, median (interquartile range); b
The score of memory and TMT B were significantly reduced in CM-MOH, compared with MO and control (
Scores of cognitive function assessment in different groupsa.
CM | MO | Control | ||
---|---|---|---|---|
CM-MOH | CMwoMOH | |||
MMSE | 29 (27–29.5) | 28.5 (28–29) | 28 (28–29.5) | 29 (28–30) |
ACE-R | 83 (74.5–88) | 83 (76.3–88.5) | 86 (78–92) | 86 (82.3–89.8) |
Attention/orientation | 18 (17–18) | 17 (17–18) | 17 (17–18) | 18 (17–18) |
Memoryb,1,2 | 21 (18–23) | 21.5 (19.3–23) | 24 (21–25) | 23 (22–24) |
Language fluency | 8 (7–9) | 8 (7–9) | 9 (7–11) | 9 (8–10) |
Language | 23 (18.5–24.5) | 20 (18–22) | 20 (18–23) | 21 (19–24) |
Visuospace | 15 (13.5–16) | 15.5 (14.3–16) | 16 (15–16) | 16 (15–16) |
TMT A | 50.3 (41.2–76.7) | 50.3 (35.1–75) | 45.1 (34.9–59.4) | 48.2 (39.7–59.3) |
TMT Bb,1,2 | 145.3 (119.2–198.9) | 111.2 (87.9–151.7) | 116.3 (97.5–129.4) | 119.6 (98.5–126.5) |
DST | 31 (23.5–45.5) | 32 (20–41.3) | 37 (30–45) | 35.5 (30–42.8) |
aKruskal–Wallis tests, median (interquartile range); b
The low 20% performance of each cognitive score was defined as cognitive decline. Due to the narrow score width of MMSE and attention/orientation, they did not implement the partition of cognitive decline. The threshold of cognitive decline in other values was as follows: ACE-R ≤77 points, memory ≤20 points, language fluency ≤7 points, language ≤18 points, visuospace ≤14 points, TMT A ≥35.29 s, TMT B ≥95.86 s, and DST ≤25 points. In cognitive decline, the morbidity rate was higher in CM-MOH and CMwoMOH than in control, especially in ACE-R total score, language fluency, and executive function (
Morbidity of cognitive decline in different groupsa.
CM | MO | Control | ||
---|---|---|---|---|
CM-MOH | CMwoMOH | |||
ACE-R1 | 8 (38.1) | 7 (35.0) | 6 (17.1) | 3 (7.5) |
Memory1 | 10 (47.6) | 6 (30.0) | 7 (20.0) | 5 (12.5) |
Language fluency1 | 8 (38.1) | 8 (40.0) | 9 (25.7) | 3 (7.5) |
Language | 5 (23.8) | 6 (30.0) | 14 (40.0) | 8 (20.0) |
Visuospace | 8 (38.1) | 5 (25.0) | 6 (17.1) | 6 (15.0) |
TMT A | 8 (38.1) | 5 (25.0) | 6 (17.1) | 4 (10.0) |
TMT B1 | 11 (52.4) | 5 (25.0) | 6 (17.1) | 1 (2.5) |
DST1 | 7 (33.3) | 8 (40.0) | 3 (8.6) | 5 (12.5) |
The low 20% performance of each cognitive evaluation was defined as cognitive decline. aChi-squared tests, cases of cognitive decline (%). 1Chi-squared tests,
Univariate regression analysis revealed that age and education were the independent risk factors of cognitive decline (
Risk factor analysis of cognitive decline after adjustment.
|
Cognitive decline | Adjusted OR (95% CI) |
|
|
---|---|---|---|---|
|
||||
Control (reference) | 40 | 3 | 1.00 | — |
CM-MOH | 21 | 8 | 8.52 (1.83–39.81) |
|
CMwoMOH | 20 | 7 | 7.14 (1.50–34.04) |
|
MO | 35 | 6 | 2.72 (0.60–12.28) | 0.194 |
MO (reference) | 35 | 6 | 1.00 | — |
CMwoMOH | 20 | 7 | 2.63 (0.68–10.14) | 0.161 |
CMwoMOH (reference) | 20 | 7 | 1.00 | — |
CM-MOH | 21 | 8 | 1.19 (0.31–4.67) | 0.799 |
|
||||
|
||||
Control (reference) | 40 | 5 | 1.00 | — |
CM-MOH | 21 | 10 | 6.92 (1.86–25.71) |
|
CMwoMOH | 20 | 6 | 3.05 (0.77–12.00) | 0.112 |
MO | 35 | 7 | 1.80 (0.50–6.41) | 0.370 |
MO (reference) | 35 | 7 | 1.00 | — |
CMwoMOH | 20 | 6 | 1.70 (0.46–6.23) | 0.427 |
CMwoMOH (reference) | 20 | 6 | 1.00 | — |
CM-MOH | 21 | 10 | 2.27 (0.60–8.57) | 0.226 |
|
||||
|
||||
Control (reference) | 40 | 3 | 1.00 | — |
CM-MOH | 21 | 8 | 7.67 (1.74–33.88) |
|
CMwoMOH | 20 | 8 | 8.24 (1.85–36.67) |
|
MO | 35 | 9 | 4.39 (1.07–17.96) | 0.040 |
MO (reference) | 35 | 9 | 1.00 | — |
CMwoMOH | 20 | 8 | 1.88 (0.57–6.19) | 0.301 |
CMwoMOH (reference) | 20 | 8 | 1.00 | — |
CM-MOH | 21 | 8 | 0.93 (0.26–3.33) | 0.912 |
|
||||
|
||||
Control (reference) | 40 | 8 | 1.00 | — |
CM-MOH | 21 | 5 | 1.25 (0.34–4.64) | 0.742 |
CMwoMOH | 20 | 6 | 1.72 (0.48–6.20) | 0.405 |
MO | 35 | 14 | 2.91 (1.00–8.51) | 0.051 |
MO (reference) | 35 | 14 | 1.00 | — |
CMwoMOH | 20 | 6 | 0.59 (0.17–2.03) | 0.592 |
CMwoMOH (reference) | 20 | 6 | 1.00 | — |
CM-MOH | 21 | 5 | 0.72 (0.17–3.05) | 0.660 |
|
||||
|
||||
Control (reference) | 40 | 6 | 1.00 | — |
CM-MOH | 21 | 8 | 3.94 (1.04–14.99) | 0.044 |
CMwoMOH | 20 | 5 | 1.86 (0.45–7.75) | 0.393 |
MO | 35 | 6 | 1.21 (0.33–4.41) | 0.776 |
MO (reference) | 35 | 6 | 1.00 | — |
CMwoMOH | 20 | 5 | 1.54 (0.37–6.49) | 0.554 |
CMwoMOH (reference) | 20 | 5 | 1.00 | — |
CM-MOH | 21 | 8 | 2.12 (0.49–9.10) | 0.314 |
|
||||
|
||||
Control (reference) | 40 | 4 | 1.00 | — |
CM-MOH | 21 | 8 | 3.25 (0.64–16.54) | 0.156 |
CMwoMOH | 20 | 5 | 1.63 (0.31–8.61) | 0.564 |
MO | 35 | 6 | 2.18 (0.52–9.11) | 0.287 |
MO (reference) | 35 | 6 | 1.00 | — |
CMwoMOH | 20 | 5 | 0.99 (0.22–4.50) | 0.987 |
CMwoMOH (reference) | 20 | 5 | 1.00 | — |
CM-MOH | 21 | 8 | 1.83 (0.36–9.22) | 0.466 |
|
||||
|
||||
Control (reference) | 40 | 1 | 1.00 | — |
CM-MOH | 21 | 11 | 50.80 (5.35–482.31) |
|
CMwoMOH | 20 | 5 | 11.30 (1.14–111.95) | 0.038 |
MO | 35 | 6 | 8.82 (0.97–79.82) | 0.053 |
MO (reference) | 35 | 6 | 1.00 | — |
CMwoMOH | 20 | 5 | 1.28 (0.30–5.51) | 0.739 |
CMwoMOH (reference) | 20 | 5 | 1.00 | — |
CM-MOH | 21 | 11 | 4.50 (0.98–20.54) | 0.052 |
|
||||
|
||||
Control (reference) | 40 | 5 | 1.00 | — |
CM-MOH | 21 | 7 | 2.83 (0.67–11.91) | 0.157 |
CMwoMOH | 20 | 8 | 4.25 (1.06–17.09) | 0.041 |
MO | 35 | 3 | 0.65 (0.14–3.04) | 0.580 |
MO (reference) | 35 | 3 | 1.00 | — |
CMwoMOH | 20 | 8 | 6.59 (1.35–32.08) | 0.020 |
CMwoMOH (reference) | 20 | 8 | 1.00 | — |
CM-MOH | 21 | 7 | 0.67 (0.16–2.84) | 0.582 |
All the data were adjusted by age and education. The low 20% performance of each cognitive evaluation was defined as cognitive decline.
Our study indicated that CM patients had increased risk of cognitive decline, especially in language fluency. Besides, CM-MOH sufferers were in higher risk of memory and executive dysfunction. In addition, the cognitive function had no obvious differences between CMwoMOH and CM-MOH.
It has been illustrated that the cognitive function of migraine sufferers declined during acute attacks. This process may be due to the decreased regional blood flow during migraine episodes [
There were overlaps between pain conduction pathway and cognitive regions in the brain. For example, the anterior cingulate cortex could regulate selective attention, working memory, and ability of identifying mistakes [
In our study, we found that the performance of language fluency was poor in CM. Our examination of language fluency consisted of phonemic and semantic elements [
Long-term exposure in inflammation could impair cognitive function. The mechanism of it may be the direct damage from prostaglandin and the prostaglandin-induced suppression of amyloid-
In addition, our study displayed that the estimation of anxiety and depression was severe in CM-MOH sufferers, as well as worse in life quality and sleep quality, compared with MO. Previous studies had found that constant suffering of anxiety, depression, or lack of sleep could influence cognition [
Our study had some limitations. Firstly, more objective index of cognitive assessment, such as fMRI, was not obtained. Secondly, the sample size in our study was small. Last, as a cross-sectional design study, we could not evaluate the progress of cognitive decline in these patients. Further study with large sample size to assess the changes of cognitive performance with disease progress, especially after withdrawal of pain killers, is needed.
There was cognitive function decline in CM patients, both in CM-MOH and CMwoMOH. NSAIDs had no influence on the cognition of CM sufferers. Further studies are needed to trace the cognitive function in other types of CM-MOH.
The datasets supporting the conclusions of this article are included within the article.
The authors declare that they have no conflicts of interest.
This study was supported by grants from the National Natural Science Foundation of China (Grant no. 81100830) and Science and Technology Program of Guangzhou (Grant no. 201508020026).
Supplementary table: risk factor analysis of cognitive decline (univariate regression analysis).