Keywords
Cerebral Angiography - Headache - Migraine Disorders - Headache Disorders, Secondary
- Headache Disorders
Palavras-chave
Angiografia Cerebral - Cefaleia - Transtornos de Enxaqueca - Cefaleias Secundárias
- Transtornos da Cefaleia
INTRODUCTION
Angiography headache (AH) is defined as a headache that is temporally associated with
the performance of cerebral digital subtraction angiography (DSA), which meets the
diagnostic criteria established by the third edition of the International Classification
of Headache Disorders (ICHD-3),[1] shown in [Table 1]. In the literature, reports of its frequency are highly variable (0.3 to 55.6%).[2]
[3]
[4]
[5]
[6] It is usually underestimated by neuroradiologists, possibly due to its usually benign
and transient character. Nonetheless, in our clinical practice, we see that patients
with a previous history of AH tend to be resistant to performing new cerebral DSAs.
The recommendation to perform subsequent cerebral DSAs is especially frequent in patients
who need follow-up after endovascular treatment, as noninvasive imaging methods may
not provide as accurate information as cerebral DSAs. In such cases, the expectation
to perform a new cerebral DSA may cause patients great suffering and psychological
stress, making AH a real clinical issue.
Table 1
The ICHD-3 diagnostic criteria for angiography headache
|
A. Any new headache fulfilling criterion C;
|
|
B. Intra-arterial carotid or vertebral angiography has been performed;
|
|
C. Evidence of causation demonstrated by at least two of the following:
1. Headache has developed during or within 24 hours of angiography;
2. Headache has resolved within 72 hours after the angiography;
3. Headache has one of the following sets of characteristics: (a) developing during
contrast injection and lasting < 1 hour, (b) developing a few hours after the angiography
and lasting > 24 hours, (c) occurring in a patient with migraine and having the features
of migraine with or without aura.
|
|
D. Not better accounted for by another ICHD-3 diagnosis.
|
Abbreviation: ICHD-3, International Classification of Headache Disorders.
The literature has relatively few investigations of AH and its risk factors. Female
sex was associated with AH in the study of Gil-Gouveia et al.,[2] Aktan et al.,[7] and Demir et al.[6] Among the studies that found an association between AH and previous history of headaches,
we name Ramadan et al.,[5] Aktan et al.,[7] and Demir et al.[6] Regarding demographic factors, a higher education level was associated with AH in
the publications of Kwon et al.[4] and Demir et al.[6]
Therefore, even though the evidence indicates a moderate/high incidence of AH, there
is still limited data about the risk factors associated with its occurrence. The present
study aimed to identify the extent to which demographic and clinical factors and the
technical characteristics of cerebral DSA are associated with the occurrence of AH.
METHODS
This is an observational, analytical, cross-sectional study approved by the Federal
University of São Paulo ethics committee and all participating hospitals (Research
Ethics Committee 93292918.0.0000.5505). Data was acquired from August 2018 to October
2019 in the Department of Interventional Neuroradiology of 5 hospitals in the metropolitan
region of Vitória, state of Espírito Santo, Brazil.
The sample comprised patients whose cerebral DSA had been indicated by the principal
investigator and patients that other specialists referred to the hospitals participating
in the study. [Table 2] shows both inclusion and exclusion criteria. The main investigator informed all
participants about the nature of the study on the day of their procedure. They were
invited to participate upon a signed consent form.
Table 2
Inclusion and exclusion criteria of the study
|
Inclusion criteria
|
Age between 18 and 80 years old
|
|
Men or women
|
|
Cognition level that allows the analysis of the occurrence of AH
|
|
Consent to the participation in the study, granted by signing an informed consent
form
|
|
Exclusion criteria
|
Unstable clinical condition
|
|
Need to use sedative medication to perform the DSA, regardless of type and dose
|
|
Presence of disease and/or neurological sequelae that impair(s) the assessment of
the occurrence of AH
|
|
Patients who have presented a change in their clinical and/or neurological condition
between the angiography and the post procedure interview with a current condition
that makes it impossible to evaluate the occurrence of AH
|
|
Indication for concomitant coronary and/or peripheral vascular angiography
|
|
Patients who have not stopped using metformin for at least 48 hours prior the DSA
|
|
Hemorrhagic or ischemic stroke in the last 90 days
|
Abbreviations: AH, angiography headache; DSA, digital subtraction angiography.
The principal investigator personally interviewed the participants before the cerebral
DSA using forms developed by the authors based on their clinical experience. Information
on the patient's sociodemographic profile, comorbidities, and medications in use were
collected. Migraine, chronic headache, hypertension, diabetes, overweight, obesity,
kidney disease, depression, and anxiety were the comorbidities evaluated in our study.
We followed ICHD-3 for diagnostic criteria of migraine and chronic headaches. To measure
the other conditions, we considered whether the patient was taking any antihypertensive
or antidiabetic medication, had a glucose level ≥ 126mg/dl or a creatinine level ≥
1.6mg/dl on laboratory tests and had a body mass index (BMI) revealing overweight
or obesity on height and weight anthropometry. Depression and anxiety were considered
whether the patient had reached the ≥ 9 cutoff point on the translated Hospital Anxiety
and Depression Scale (HAD).[8]
Immediately after the procedure, the patients were personally interviewed to assess
the occurrence of AH. Between the 3rd and 7th days post-DSA, a new interview was performed by telephone to confirm or exclude the
occurrence of AH. All procedures and interviews were conducted exclusively by the
main investigator.
All procedures were performed through the femoral route using the iodinated contrast
media agent Omnipaque (Iohexol) 300. The number of injected vessels and the complementary
performance of aortography and/or 3D angiography varied between subjects, according
to the recommendation for the cerebral DSA. All aortographies and 3D angiographies
were performed using contrast media pressure injectors. All data related to cerebral
DSA was collected immediately after the procedure.
Statistical analyses were performed using IBM SPSS Statistics for Windows, version
24.0 (IBM Corp., Armonk, NY, United States). Descriptive statistics were reported
using frequencies and percentages for categorical variables, means and standard deviations
(SDs) were reported for normally distributed continuous variables, and medians and
interquartile ranges were reported for asymmetrical continuous variables. Categorical
and continuous variables were compared using the chi-squared test and the independent
Student t-test or the Mann–Whitney U test, respectively.
Multivariable binary logistic regression was analyzed to identify the association
of the sociodemographic, clinical, and technical risk factors associated with AH.
The goodness-of-fit of the regression model was tested using Cox and Snell R-square,
Nagelkerke R-square, and Hosmer and Lemeshow tests. The variables considered for inclusion
using the stepwise logistic regression approach were based on statistical significance
from univariate statistics (p ≤ 0.10), factors that are known to be associated with
the presence of AH in previous studies, or the researcheŕs own clinical experience.
The covariates included were age, sex, previous history of migraine, depression and
anxiety comorbidity (patients who have both diagnosis), hypertension, diabetes, aortography,
and 3D angiography. The outcome was described with odds ratios (ORs) and 95% confidence
intervals (Cis). There was no missing data. Values of p < 0.05 were considered statistically significant.
RESULTS
A total of 129 subjects were subject to elective DSA over 15 months. Fifteen subjects
were excluded (2 for not having discontinued the use of metformin for a period equal
to or greater than 48 hours, 6 due to the need for sedation to perform the DSA, and
7 because they had cerebrovascular disease with an ictus date of < 90 days), leaving
114 subjects evaluated in the present study. None of the subjects refused to participate.
The overall frequency of AH was 45.6% (52/114).
[Table 3] shows the demographic, clinical, and technical data. The mean (SD) age of the whole
group was 52.8 (±13.8) years old. The group mainly comprised women (86/114, 75.4%).
The main diagnosis observed on angiography was aneurysm (76/114, 66.7%), followed
by normal angiography (13/114, 11.4%) and arteriovenous shunt (9/114, 7.9%). Regarding
clinical characteristics, 29.8% (34/114) of the participants had a history of migraine,
and 10.5% (12/114) had chronic headaches.
Table 3
Demographic, clinical, and technical factors reaching statistical significance
|
Variables
|
Total
|
Angiography headache
|
p-value
|
|
No
|
Yes
|
|
Participants
|
|
114 (100.0%)
|
62 (54.4%)
|
52 (45.6%)
|
–
|
|
Sex
|
Female
|
86 (75.4%)
|
40 (46.5%)
|
46 (53.5%)
|
0.003
|
|
Male
|
28 (24.6%)
|
22 (78.6%)
|
6 (21.4%)
|
|
|
Migraine
|
No
|
80 (70.2%)
|
52 (65.0%)
|
28 (35.0%)
|
< 0.0001
|
|
Yes
|
34 (29.8%)
|
10 (29.4%)
|
24 (70.6%)
|
|
Depression
|
No
|
89 (78.1%)
|
54 (60.7%)
|
35 (39.3%)
|
0.011
|
|
Yes
|
25 (21.9%)
|
8 (32.0%)
|
17 (68.0%)
|
|
Anxiety
|
No
|
84 (73.7%)
|
51 (60.7%)
|
33 (39.3%)
|
0.023
|
|
Yes
|
30 (26.3%)
|
11 (36.7%)
|
19 (63.3%)
|
|
Hypertension
|
No
|
48 (42.1%)
|
32 (66.7%)
|
16 (33.3%)
|
0.025
|
|
Yes
|
66 (57.9%)
|
30 (45.5%)
|
36 (54.5%)
|
|
Mean contrast media volume ± SD (mL)
|
133.40 ± 40.72
|
145.38 ± 42.68
|
124.35 ± 37.13
|
0.022
|
|
|
n
= 109
|
n
= 58
|
n
= 51
|
|
|
Pressure injector
|
No
|
23 (21.1%)
|
20 (87.0%)
|
3 (13.0%)
|
< 0.0001
|
|
Yes
|
86 (78.9%)
|
38 (44.2%)
|
48 (55.8%)
|
|
3D Angiography
|
No
|
30 (27.5%)
|
24 (80.0%)
|
6 (20.0%)
|
0.001
|
|
Yes
|
84 (73.7%)
|
38 (45.2%)
|
46 (54.8%)
|
Abbreviation: SD, standard deviation.
In our study, 71.2% (37/52) of the subjects had unilateral headaches, and 28.8% (15/52)
had bilateral headaches; for 51.9% (27/52), the headache was stabbing, and 48.1% (25/52)
experienced the nonstabbing type of headache. The average maximum pain intensity was
7.8/10.
Regarding the procedure, the mean (SD) contrast media volume used was 133.4ml (±40.7).
In our sample, 73.7% (84/114) and 10.5% (12/114) underwent 3D angiography and aortography,
respectively. In five patients, both procedures were performed, and they were excluded
from the analysis.
A higher proportion of AH than expected was observed in subjects who underwent 3D
angiography (46/84; 54.8%; p = 0.001), suggesting that the 3D angiography might be associated with AH. It was
also interesting to note a significantly smaller mean contrast media volume in the
AH group compared with subjects without AH (p = 0.022).
Due to sample size restrictions, the best multivariable regression model was employed
with six predictors and is represented in [Table 4]. A previous history of migraine (OR: 4.9; 95%CI 1.62–14.7; p = 0.005) and 3D angiography (OR: 6.6; 95%CI 2.04–21.5; p = 0.002) were significantly associated with the occurrence of AH. Covariate depression
and anxiety (OR: 3.74; 95%CI = 0.84–16.5; p = 0.082), female sex (OR: 3.02; 95%CI: 0.99–9.2; p = 0.052), hypertension (OR: 2.26; 95%CI: 0.89–5.7; p = 0.085), and diabetes mellitus (OR: 5.08; 95% CI 0.98–26.2; p = 0.052) were not significant in the adjusted model.
Table 4
Multivariable binary logistic regression analysis of demographic, clinical, and procedural
predictors of angiography headache
|
Variables
|
Adjusted OR (95%CI)
|
p-value
|
|
Female sex
|
3.02 (0.99–9.2)
|
0.052
|
|
Hypertension
|
2.26 (0.89–5.7)
|
0.085
|
|
Diabetes mellitus
|
5.08 (0.98–26.2)
|
0.052
|
|
Depression and anxiety
|
3.74 (0.84–16.5)
|
0.082
|
|
Migraine
|
4.9 (1.62–14.7)
|
0.005
|
|
3D Angiography
|
6.62 (2.04–21.5)
|
0.002
|
Abbreviations: CI, confidence interval; OD, odds ratio.
Notes: Cox and Snell R-square; 0.32 Nagelkerke R-square; 0.43 Hosmer and Lemeshow
test Χ2 = 4.79, df = 8, p = 0.779. The numbers in bold are statistically significant.
DISCUSSION
According to the ICHD-3, the temporal relationship between the supposed triggering
factor and the onset of the headache is important in acute secondary headaches. Following
this guideline, a patient with a headache starting 24 hours after cerebral DSA could
fulfill the criteria for AH (e.g., criteria A, B, C2, C3c, and D). We believe it is
inappropriate to establish this diagnosis with a headache starting 24 hours after
the procedure. Therefore, we adapted the ICHD-3 diagnostic criteria by excluding subjects
whose headaches started > 24 hours after the cerebral DSA.
Significant variation in the frequency of AH can be explained by the heterogeneity
of the inclusion criteria used in different studies and the lack of uniformity in
their definitions of AH.[3]
[4]
[5]
[6]
[7]
Women predominated in the sample (75.84%), which is consistent with the fact that
intracranial aneurysms represented the main recommendation for performing cerebral
DSA, as this cerebrovascular disease is more prevalent in women.[9] Despite some studies having found association between AH and female sex,[2]
[6]
[7] in the multivariate analysis this factor did not reach statistical significance.
As far as we know, this is the first study that systematically assesses the relationship
between psychiatric disorders and AH. The covariate depression and anxiety was not
statistically significant in the multivariate analysis. The association between psychiatric
disorders and pain could be explained by the fact that these diseases share biological
pathways and neurotransmitters.[10]
Interestingly, our study found an association between hypertension and the occurrence
of AH, which should be analyzed cautiously. Blood pressure was not monitored before,
during, or after the procedure. These facts lead to the hypothesis that some cases
attributed to AH may represent cases of secondary headaches due to a significant rise
in blood pressure (a headache attributed to a hypertensive crisis without hypertensive
encephalopathy). In fact, when adjusted for other variables, hypertension was not
a significant predictor of AH.
This is the first study showing a strong association between AH and 3D angiography
(OR: 6.6). This technological resource allows a highly detailed analysis of intracranial
vessels and is essential for analyzing cerebrovascular diseases. The 3D acquisition
requires the infusion of contrast media in a greater volume and pressure than those
commonly used for conventional acquisitions. Therefore, the observed association may
be explained by the fact that the contrast media injected under greater pressure determines
an acute mechanical stimulus in the wall of the cervical and cranial arteries, causing
pain at the time of the cerebral DSA. This hypothesis had already been raised by Martins
et al. when they reported a series of 11 patients who developed headaches during intracranial
endovascular procedures. For those authors, the most similar pathogenesis of the pain
is the direct stimulation of the arterial wall of the craniocervical arteries.[11] Another evidence that the stimulation of the arterial wall may play a role in the
pathophysiology of AH is the study by Aktan et al.[7] These authors assessed 139 subjects who underwent cerebral DSA and 30 control subjects
(who underwent peripheral DSA). The frequency of AH was 30.2% in the group subject
to cerebral DSAs and 10% in the control group.
The results suggest that a previous history of migraine was associated with a higher
chance of AH. This association can be explained by the fact that migraine and AH share
similar pathophysiology: the activation of the trigeminovascular system.[2]
[6]
[7]
[12]
It is credible to assume that the total contrast media volume used in the cerebral
DSA was directly associated with the occurrence of AH. It is interesting to note that
there was an inverse relationship between the amount of contrast and the occurrence
of AH. In our series, many patients who underwent angiography using a small total
amount of contrast usually required 3D acquisition protocols. Thus, we consider that
the AH in these patients was caused by the 3D acquisition and not by the use of a
small amount of contrast.
The absence of a standardized assessment of the cognitive function of the participants
and the fact that the interviews were not blinded represent important limitations
of our study. The convenience sample due to budget constraints for the present study
was also a significant limitation.
In conclusion, our study found that 3D angiography using a pressure injector is a
strong predictor for the occurrence of AH, which has never been reported before. The
association between migraine and AH confirms the results of previous studies.
As this is a cross-sectional study, true causality may not be defined yet.
Knowledge of these factors may allow future clinical researchers to assess prophylactic
measures in patients at greater risk for developing AH.
