Relation of post-stroke headachto cerebrovascular pathology and hemodynamics

Introduction

Cerebrovascular stroke is a medical hazard that is responsible for many morbidities and mortalities. Despite its’ high prevalence, headache attributed to ischemic stroke is often undertreated and overlooked [14]. It is an important symptom that may be severe and persistent after ischemic stroke [8]. The prevalence of headache attributed to ischemic stroke is ranging between 7.4% and 34% [23].
The clinical criteria of headache attributed to ischemic stroke were introduced by the International Classification of Headache Disorders (ICHD), which was finally updated for the third time in 2018 [4]. These new criteria more accurately reflected the close temporal relationship between the onset of headache and the development of focal signs and symptoms of an ischemic stroke, an improvement of headaches, and better classified the entity of persistent post-stroke headaches [17].
The exact pathophysiology of such headache is still unclear [24]. Till now, many theories have been introduced to justify such headache. Activation of the trigemino-vascular pathway is the most accepted theory for headache production [11,24]. However, there were many conflicting data in the literature regarding the triggers that activate this pathway [3,11,24].
Atherosclerosis and collateral opening can be considered as another triggering factors in the pathogenesis of post-stroke headache. Despite that the trigemino-vascular system could be activated by biochemical triggers, opening collaterals could be involved. Due to difficulty in the assessment of opening of collaterals, the literature does not contain enough data to explain its’ role in the production of post-stroke headache till now [14].
Duplex ultrasound is considered an easy, feasible, safe, more accurate, and non-invasive tool for the assessment of cerebrovascular changes, including the collateral system and hemodynamics [6,12].
This study aimed to understand the relation of a post-stroke headache to cerebrovascular pathology and hemodynamic changes by using a high-resolution duplex ultrasound examination.

Material and methods

The current study was an observational retrospective case-control study that was conducted in Kasralainy hospital, Cairo University, and Al-Azhar University hospitals from January 2021 to August 2021. We included 239 patients, with a recent ischemic stroke within three months from the onset. Patients were further sub-divided into two groups: Group I (cases) included patients fulfilling criteria of the ICHD-3 for headache attributed to ischemic stroke, and group II (controls) consisted of headache-free stroke patients. We excluded patients with intra-cranial hemorrhages, transient ischemic attacks, artery dissections, superficial temporal arteritis, unconscious, confused, and/or aphasic patients.
Demographic data collected included age, gender, history of pre-stroke headache, and associated comorbidities (hypertension, diabetes mellitus, ischemic heart disease, atrial fibrillation, dyslipidemia, smoking, presence of old stroke, and using anti-hypertensive medications). Clinical symptoms and signs, initial computed tomography, and/or magnetic resonance imaging of the patient’s brain and cerebrovascular system were obtained. Strokes were classified according to TOAST (Trial of Org 10172 in Acute Stroke Treatment) classification [1]. As regards to vascular distribution of stroke, Oxfordshire Community Stroke Project (OCSP) was applied to classify the patients into partial anterior circulation infarcts (PACI), lacunar circulation infarcts (LACI), and posterior circulation infarcts (POCI) [25]. The presence of headache attributed to ischemic stroke was considered if the patient reported newly developed headache or changing in pre-stroke headache character. Features of post-stroke headache, including character (tightening, pulsatile, or stabbing), intensity (mild to moderate, if not interfere with daily activities; or moderate to severe, if interfere with daily activities), side (ipsilateral to the infarction, contralateral, unilateral alternating, or bilateral), and location [diffuse, anterior (frontal, temporal, or parietal); posterior (occipital)] were collected. In addition, we identified the time of post-stroke headache with the onset of stroke (72 hours before stroke onset, within 7 days post-stroke, or after 7 days post-stroke) and headache-accompanying symptoms (nausea, vomiting, photophobia, and phonophobia).
Ultrasound duplex assessment was carried out for all patients using Phillips iU 22 ultrasound equipment, with a low frequency of 3-9 MHz linear probe for extra-cranial vascular systems (carotid and vertebrobasilar systems), and low frequency of 1-5 MHz phased array for trans-cranial cerebrovascular system. In case of carotid stenosis, data are interpreted according to the Society of Radiologists in Ultrasound Consensus Criteria for Carotid Stenosis [12]. Parameters obtained transcranially were interpreted according to Baumgartner criteria [6].
This study was conducted in agreement with the declarations of Helsinki, and the institutional review board approval was obtained from the ethics unit of Faculty of Medicine, Al-Azhar University. Each patient was requested to sign a written informed consent before enrollment into the study.
The sample size calculation was conducted using MedCalc software for windows after applying the prevalence of carotid atherosclerosis among patients with stroke-associated headache that was reported by Paciaroni et al. [24] to be 59.1%, with alpha error as 5% and beta error 20%. These resulted in recruitment of 238 patients. Statistical analysis was conducted using SPSS platform for Windows, version 26. Data were presented according to their type as mean ± standard deviation (SD), in normally distributed continuous data. Categorical data were presented as numbers and percentages (%). Comparison between the case and control groups was conducted using Student’s t-test and chi-square (c2) test, according to type of data. Multivariate logistic regression analysis was applied to identify the significant predictors of post-stroke headache. P-values of less than 0.05 were considered statistically significant.

Results

Group I included 112 patients and group II consisted of 127 matched patients. Pre-morbid risk factors did not show statistically significant differences between the two groups. Pre-stroke headache and type of stroke showed a statistically significant difference between the two groups (p < 0.001) (Table I).
The tightening character of headache was the most prevalent type (69.6%); however, it was of moderate intensity in most of the cases (73.2%). Around two-thirds (61.6%) of patients started headache within seven days post-stroke, while in 31.3% occurred after seven days. Headache was ipsilateral to infarction in 58% of patients (Table II).
Despite most of the extra-cranial atherosclerotic diseases were mild in most of the examined arteries, there was a considerable fraction (19.6%) reported with severe internal carotid artery (ICA) stenosis in the case group. Similarly, severe stenosis up to total occlusion of the vertebral artery (VA) was reported in 28.6% of patients with post-stroke headache, with significant difference when compared with patients without headache. A statistically significant increase in the proportion of large artery stenosis was detected in group I (Table III).
The frequency of collateral opening was higher in the case group. Headache was more frequent in patients with the leptomeningeal collateral opening (p < 0.001). Group I showed an increased frequency of intra-cranial stenosis (irrespective of the grade of stenosis) in both anterior and posterior circulation (Table IV).
The multi-variable regression analysis revealed that the presence of pre-stroke headache was an independent predictor for post-stroke headache occurrence (OR = 28.187, 95% CI: 6.612-120.158%, p < 0.001). The presence of any intra-cranial cerebrovascular pathological changes, regardless its’ nature, can also predict headache development (OR = 25.071, 95% CI: 6.498-96.722%, p < 0.001). Mild stenosis (< 50%) of either PCA or VA segment 4 and opening collaterals were also strong predictors for post-stroke headache (p < 0.001, for all) (Table V).

Discussion

The intra-cranial cerebrovascular pathological changes, including opening of the collateral channels and variable degrees of stenosis of cerebrovascular systems, were implicated in the production of the post-stroke headache. Headache is a commonly reported symptom in patients with acute ischemic stroke. Despite the literature being crowded with data suggesting that the activation of the trigemino-vascular system due to cortical spreading depression (CSD) triggered by cerebral ischemia, the pathophysiology of headache attributed to ischemic stroke has not been well-established until now [9,23]. The literature suggested headache attributed to ischemic stroke was not related to intra-cranial vasodilation or increased flow through collateral circulation [15]. This could be assigned to the lack of association between post-stroke headache and angiographically demonstrated collateral circulation [11].
Assessment of cerebrovascular structures and hemodynamics in this study using duplex ultrasound has added valuable data in understanding its’ pathophysiology. It was shown that post-stroke headache was more common in patients with pre-morbid headache attacks, yet with a significant change of character following the cerebrovascular event. Large artery atherosclerotic stenosis was strongly related to the headache precisely in the territory of the posterior circulation. Patients with collaterals opening, particularly leptomeningeal one, showed a higher frequency of post-stroke headache. Regarding the history of pre-stroke headache, our results were concomitant with previous literature, which revealed the presence of pre-stroke headache in 22-57% of patients with stroke-attributed headache [26]. This could be attributed to stroke-associated pathophysiological changes, which could affect the pattern of patient’s usual type of headache [22]. In our patient’s sample, tension headache was the most prevalent type in the pre-morbid state coming in the first place before the vascular headache. This could be explained through a release of common headache-associated chemical mediators in both vascular and tension-type headache, vasoactive substances, amino acid neurotransmitters, and platelet activation factor [10,11,16]. Subsequently, ischemic stroke could affect the production of these substances making changes in the character of the pre-stroke headache [10]. There was a significant association between previous ischemic strokes and development of post-stroke headache. The association of old ischemic stroke and development of post-stroke headache could be explained through an opening of more pain-sensitive collaterals’ channels as a consequence of each ischemic event [15]. The time of this collateral opening was reported to be in close temporal relation with the onset of ischemic stroke [14]. That is why our results showed that post-stroke headache was more likely to occur within the first seven days post-stroke. These results were in agreement with a previous study, which revealed that post-stroke headache occurred mostly within the first 3-4 days post-stroke [19]. This could be also explained by the endothelial release of bradykinins, prothrombotic factors, glutamate, interleukin-6, nitric oxide metabolites, calcitonin gene-related peptides, substance P, and excitotoxic substances during the acute phase of ischemic stroke [13,18]. Despite that supratentorial lesions are usually accompanied by headache in ophthalmic distribution of the trigeminal nerve [3], our data revealed that the anteriorly-located headache was more associated with posterior circulation infarcts. This may indicate a relationship between the richly innervated posterior circulation vasculature through the trigeminal system [20].
Activation of the trigemino-vascular pathway in posterior circulation affection could be attributed to ischemia of the trigeminal nucleus, dysfunction of the serotoninergic nuclei of the brainstem, or ischemia of the dura of the posterior fossa [2,7,27]. Furthermore, the enhanced trigemino-vascular pathway, in addition to collaterals opening, occur mainly ipsilaterally to the infarction, explaining the high frequency of ipsilateral post-stroke headache [13]. However, due to large receptive area of the trigeminal system, some intra-cranial vessels have contralateral trigeminal innervation [21]. This gives an acceptable explanation for reporting contralateral headache in some patients. The association of headache with ischemic stroke related to carotid artery disease revealed in our study was reported in previous research. Its’ prevalence ranged between 26% and 35% [5]. Being involved in the development of thrombotic stroke [8], carotid artery disease could explain the significant association between thrombotic, large, and small vessel diseases, stroke, and post-stroke headache, rather than cardioembolic stroke. However, the prevalence of headache associated with large vessel disease was higher than that linked to small vessel disease, indicating an implication of arterial caliber in headache production. Occlusion of a large artery produces large infarction, which in turn produces large inflammation and irritation to pain-sensitive structures, causing headache [5]. Our study could be limited by selection bias, as the patients with disturbed level of consciousness or those with aphasia could not provide information regarding the presence of headache. Therefore, patients with poor prognosis were not included.

Conclusions

Headache attributed to ischemic stroke is a common phenomenon, especially in patients with pre-stroke headache, history of old stroke, posterior circulation infarction, and large artery disease. This headache was of moderate-intensity with clinical characteristics of tension-type, and it was in close temporal relation with clinical symptoms and signs in most of the cases. The intra-cranial cerebrovascular pathological changes, including opening of the collateral channels and variable degrees of stenosis of cerebrovascular systems, were implicated in the production of that headache. Our results could help further studies for better understanding and hence, provide better management for post-stroke headache. Further prospective studies assessing the effect of headache attributed to ischemic stroke on the prognosis and outcome of ischemic stroke and post-stroke disability and quality of life are also needed.

Disclosure

The authors report no conflict of interest.

References