Transcript p<0.001 - Worcester Research and Publications

Comorbid migraine in major depressive disorder suggests
a subgroup related to bipolar disorder
K. Gordon-Smith 1, P. Ridley 3, A. Perry 1 S. Knott 1, L. Forty 2, N. Craddock 2, I. Jones 1,2, L.A. Jones 1
1Department
of Psychological Medicine, University of Worcester, UK
2Division of Psychological Medicine and Clinical Neurosciences, Cardiff University, UK
3College of Medical and Dental Sciences, University of Birmingham, UK
Background
• It has been hypothesised that patients with major depressive disorder (MDD) and migraine
may represent a subgroup of individuals due to differences in their clinical characteristics
when compared to those with MDD without migraine.
• Female gender, longer depressive episodes, greater severity of depressive symptoms,
greater seasonal variation in symptoms, greater number of depressive episodes, higher
prevalence of cyclothymic and irritable temperaments and agoraphobia have been found to
be associated with migraine in MDD (Hung et al., 2005; Oedegaard and Fasmer, 2005).
• Previous small-scale studies suggest the specific clinical characteristics associated with
presence of migraine in MDD may overlap with those of bipolar disorder.
Aims
In a large UK sample with broad and rich clinical descriptions:
1. Compare a broad range of clinical characteristics in participants who have MDD with and
without migraine.
Results
• The presence of migraine in MDD was associated with female gender (77% vs 57%,
p<0.001), younger age of onset (23 vs 27 years, p=0.002), history of attempted suicide
(38% vs 23%, p=0.002), and more panic/agoraphobia symptomatology (6 vs 4, p<0.001)
(Table1).
• Female gender (OR=2.44, p=0.006) and younger age of onset (OR=0.97, p=0.013)
remained significant in a multivariate model.
• There were no significant differences between the scores of participants with and
without migraine on any scale of the TEMPS-A.
• For two of the three clinical characteristics found to be significantly different between
the MDD+m and MDD-m groups, the characteristics of the MDD+m group were
significantly more similar to one or both of the BPI and BPII groups compared to the
MDD-m group (age at onset and history of suicide attempt) (Table 2).
Table 1. Lifetime demographic and clinical characteristics of participants with major depressive disorder
with and without migraine
MDDRmigraine
(n=218)
MDDR+
migraine
(n=134)
p-value
51 (40-59)
49 (40-56)
p=0.070
Sex (Female), % (n)
56.9% (124)
76.9% (103)
p<0.001
Family history of affective disorder,
% (n)
82.1% (138)
92.1% (105)
p=0.017
Age of illness onset, median (IQR)
27 (20-35.5)
23 (18-30.3)
p=0.002
Episodes per year of illness,
median (IQR)
0.227
(0.143-0.413)
0.226
(0.146-0.333)
p=0.415
Attempted suicide, % (n)
22.7% (49)
38.3% (51)
p=0.002
Ever a tobacco smoker, % (n)
52.7% (68)
47.9% (101)
p=0.386
Heaviest alcohol consumption over recommended
limit, % (n)
47.2% (101)
41.9% (54)
p=0.336
GSM-V: panic/agoraphobia, median (IQR)
4 (2-6)
6 (5-7)
p<0.001
GSM-V: social anxiety,
median (IQR)
7 (4-9)
8 (5-10)
p=0.119
GSM-V: Obsessive compulsive, median (IQR)
4 (2-6)
4 (2.25-6)
p=0.960
17 (13-20)
17 (11-21)
p=0.662
2. Explore possible similarities between those characteristics associated with the presence of
migraine in MDD and those in bipolar disorder.
Age at interview, median (IQR)
Methods
Participants
• Subset of participants with a lifetime DSM-IV diagnosis of recurrent MDD (MDDR) assessed
for the presence of migraine (n= 352) who were recruited to an ongoing programme of
research into the genetic and non-genetic determinants of bipolar disorder (BD) and
related mood disorders (Bipolar Disorder Research Network, BDRN).
Clinical Assessment
• Lifetime psychiatric history – details of illness course were collected via semi-structured
interview (Schedules for Clinical Assessment in Neuropsychiatry) and review of medical
case notes.
• Modified General Spectrum Measure 5 (GSM-V) –self-report measure of
panic/agoraphobia, social anxiety and obsessive compulsive symptomatology.
• Hypomania Checklist (HCL-32) – self report measure of hypomanic/manic symptoms.
Temperament Assessment
• Temperament Evaluation of Memphis, Pisa, Paris and San Diego Autoquestionnaire
(TEMPS-A) –self-report questionnaire measuring cyclothymic, depressive, irritable,
hyperthymic and anxious temperaments.
HCL-32 total, median (IQR)
Migraine Assessment
• An 11 item self-report headache questionnaire based on the International Headache
Society’s diagnostic criteria was used to make a lifetime diagnosis of migraine.
MDDR without
migraine
(MDDR-m) n=218
MDDR with
migraine
(MDDR+m) n=134
Table 2: Characteristics associated with the presence of migraine in major depressive disorder: relationship
with bipolar disorder
Statistical Analysis
• Chi-squared and Mann-Whitney U tests were used to compare demographic, clinical and
temperament measures between the MDDR-m and MDDR+m groups. Due to the
number of comparisons, a stringent significance level of p<0.01 was employed.
• All demographic and lifetime clinical course variables significant at p<0.05 were included
as explanatory variables in a binary logistic regression model, with presence of migraine
as the outcome variable.
• All clinical characteristics found to be significantly associated (p<0.01) with the presence
of migraine in MDDR were then compared between MDDR+m and MDDR-m individuals
and two further groups of individuals with bipolar I disorder (BPI; n=364) and bipolar II
disorder (BPII; n=43).
• All categorical variables were analysed by chi-squared tests, with each pair of groups
being analysed independently by post-hoc chi-squared tests. Continuous variables were
compared by Kruskal-Wallis tests, with differences between each pair of groups being
assessed by post-hoc Mann-Whitney U tests. A significance level of 1% (p<0.01) was
employed for all variables.
www.bdrn.org
[email protected]
MDDR= recurrent major depressive disorder; IQR = inter-quartile range; GSM-V = General 5-Spectrum Measure;
HCL-32 = Hypomania Checklist. Significant results at p<0.01 are shown in bold. Numbers vary for each variable
due to missing data.
MDDR-
MDDR+
migraine
migraine
(n=218)
(n=134)
BPI (n=364)
BPII (n=43)
p-value
Post-hoc
analysis
MDDR+m, BPI
Median age of
27
23
23
21
and BPII <
illness onset, years
p<0.001
(20-35.5)
(18-30.3)
(18-31)
(15-31)
MDDR-m
(IQR)
(p<0.005)
MDDR+m and
Attempted suicide,
22.7%
38.3%
39.8% (135)
32.6% (14)
p<0.001
BPI > MDDR-m
% (n)
(49)
(51)
(p<0.003)
MDDR+m>
GSM-V: Median
4
6
4
5
BPI and
panic/agoraphobia
p<0.001
(2-6)
(5-7)
(2-7)
(4-6)
MDDR-m
score (IQR)
(p<0.001)
MDDR= recurrent major depressive disorder; BPI=bipolar I disorder; BPII=bipolar II disorder; IQR = Inter-quartile
range; GSM-V = General 5-Spectrum Measure. Significant results at p<0.01 are shown in bold.
Conclusions
• The presence of migraine in MDD delineates a subgroup of individuals with a more severe
illness course.
• The clinical presentation of this subgroup more closely resembles that of bipolar disorder
than that of MDD without migraine.
• The presence of migraine in major depression may be a marker of a specific subgroup that
could be useful in future research.
Limitations
• Relied on participant self-report of migraine, temperament and anxiety symptoms.
• Sample size was not large enough to stratify the MDDR+m group by presence/absence of
aura.
Acknowledgements
We would like to thank all of the participants who have kindly
given their time to participate in our research and all the mental
health professionals who were involved in the recruitment of
participants.