Sage Journals: Discover world-class research

Abstract

The mechanism for headache in patients with acute ischaemic stroke are not completely understood. We analysed the relationship between headache and the early worsening of neurological symptoms in patients with acute ischaemic stroke, and we studied the possible biochemical mechanisms implicated. Headache at the onset of ischaemic stroke predicted progression with a sensitivity, specificity, and positive predictive value of 56%, 99%, and 98%, respectively. CSF concentrations of glutamate, Interleukin-6, and NO-m were significantly greater in patients with progressing stroke than in patients with nonprogressing stroke, and these biochemical markers were also significantly higher in patients with headache than in those without headache. Results of this study suggest that headache at the onset of ischaemic stroke is an independent predictor of neurological worsening and we hypothesize that headache might be a surrogate marker of the molecular mechanisms involved in neurological worsening after acute stroke.

Keywords

Headache ischaemic stroke cytokines glutamate

Introduction

Headache is present in 10% to 39% of patients during the acute phase of cerebral infarction (1–5). The mechanisms for headache in patients with ischaemic stroke are not completely understood (5). Prevailing views include dilation of collaterals, focal distension of the artery, local ischaemia of the arterial muscle, and irritation of the pain-sensitive arterial wall by atheroma (6). Serotonin and other vasoactive peptides released from the junctional elements of the trigeminal vascular system may also play a role (7). Recently, excitatory amino acids (EAAs) have been related to the pathophysiology of headache that occurs at the onset of stroke (8).

In acute stroke, excitotoxicity (9, 10), inflammatory reactions (11), and nitric oxide generation (NO) (12) are the biochemical mechanisms underlying early clinical worsening that occurs in 26–34% of patients with acute ischaemic stroke (10, 13–19). The rise of neuroexcitatory amino acids in plasma and CSF in patients with headache during the acute phase of cerebrovascular diseases (8), supports the hypothesis that headache at the onset of ischaemic stroke could be a good predictor of neurological deterioration.

In this study we analysed the relationship between headache and the early worsening of neurological symptoms in patients with acute cerebral ischaemia, and we studied the possible biochemical mechanisms implicated.

Patients and methods

We studied 241 patients with a first-ever acute ischaemic stroke. The patients were selected from a larger cohort of 271 consecutive patients with hemispheric infarction admitted within the first 24 h duration from symptoms onset. The characteristics of the group and inclusion criteria for the whole series of patients have already been published (9, 10). In summary all patients had a persistent focal neurological deficit and absence of mass effect or cerebral haemorrhage on the cranial CT carried out before inclusion. The presence of headache at stroke onset was particularly evaluated in these patients. For the purposes of this study we excluded 30 patients in whom headache was not recorded due to communication difficulties. The total number of patients were included in a prospective clinical investigation conducted between October 1992 and December 1996 to analyse clinical, biochemical, and radiological factors related to progressing stroke. The protocol was approved by the Ethics Committee of Hospital Clínico Universitario de Santiago de Compostela, and informed consent was given by the patients themselves, or by their relatives.

On admission, blood pressure and body temperature were recorded and blood chemistry studies, chest X-ray, ECG and nonenhanced cranial CT were performed. Stroke subtype was classified according to the TOAST definitions (20). We did not include patients with transient ischaemic attack. Stroke severity was assessed by a neurologist immediately after these tests with the Canadian Stroke Scale (CSS). The CSS measures level of consciousness (alert, 3; drowsy, 1.5), speech (normal, 1; expressive deficit, 0.5; receptive deficit, 0), orientation (orientated, 1; disoriented or non applicable, 0), facial paresis (none, 0.5; present, 0), weakness in arm, hand, and leg (none, 1.5; mild, 1; significant, 0.5; total, 0; scored individually for each item), on a total score from 1.5 (maximal deficit) to 10 (absence of deficit) (21). Following already published criteria (22), progressing stroke was diagnosed when the CSS score decreased 1 or more points between admission and 48 h. This difference represents the change with the highest sensitivity, retaining good specificity (23).

All the CT examinations were carried out with a CT System 3000 plus (GEC) scanner with a 512×512 display. Early signs of infarction were carefully evaluated during the first examination; these signs included hypodensity consistent with the clinical picture, and the indirect signs of cerebral infarction like obscuration of the lenticular nucleus, obscuration of the cortex, and mass effect on the cortical sulci and structures of the median line. On days 4–7 after clinical onset, a second nonenhanced brain CT was performed to asses the volume (in cubic centimetres) of the infarction, according to the formula 0.5×a×b×c, where a and b represented the largest perpendicular diameters measured on CT scan and c the slice thickness. Based on the extension and topography of the infarct on CT (24), the patients were grouped as follows: total anterior circulation infarcts (TACI), partial anterior circulation infarctions (PACI), lacunar infarcts (LACI) and posterior circulation infarcts (POCI).

Blood and CSF samples were drawn during the painful period in those patients with headache, and in all patiens without headache; the interval between inclusion and taking the CSF was 2.6±3.7 h. Blood was obtained by venipuncture and collected in crystal tubes containing EDTA-K3. Plasma was prepared by centrifugation (3000×g for 15 min) and stored at −70°C. A lumbar puncture was performed if there were no signs of mass effect on CT. CSF samples were prepared by centrifugation (2000×g for 10 min) and immediately stored at −80°C.

Quantification of glutamate was performed by HPLC following the method described elsewhere (8–10). NO generation was calculated by quantifying the final products of its reactions, nitrates and nitrites, by a colourimetric assay (Cayman Chemical Co, Ann Arbor, MI), following the method described in a previous work (12). IL-6 concentrations in CSF samples were measured with commercially available quantitative sandwich enzyme-linked immunoadsorbent assay (Quantikine) kits obtained from R & D System (Minneapolis, MN, USA), as previously described (11). These measurements were made by technicians from an independent laboratory who were unaware of the clinical outcome and neuroimaging findings. As our original purpose was to determine EAAs, further biochemical markers were only investigated in patients from whom we had obtained and stored (frozen) CSF samples for possible future laboratory determinations. Availability of samples relied exclusively on the amount of CSF obtained at the time of inclusion and not on other clinical and biochemical factors.

Statistical analysis

Continuous variables were expressed as mean±one standard deviation, or as median [quartiles], when they were not normally distributed. Categorical data were analysed with the χ² test with the Fisher correction when necessary. Comparison between two groups of continuous variables was performed by using the Mann–Whitney test.

The importance of headache for subsequent early neurological deterioration (0=no, 1=yes), was assessed by logistic regression analysis based on the maximum likelihood ratio. A model was fitted to assess the adjusted OR of progressing stroke for those baseline clinical and radiological variables which reached a P-value <0.15 in the bivariate analysis. The resulting model was further adjusted by the ultimate infarct volume. In order to test the interaction between headache and the CSF biochemical markers on neurological deterioration, we calculated the odds ratios for headache in three separate logistic models with a further adjustment for glutamate, IL-6, and NO-m, respectively.

Results

Of the 241 patients studied, 73 (30.6%) complained of headache at onset of stroke. The main characteristics of the studied population are shown in Table 1. Patients who experienced headache had longer delay to hospital admission, higher stroke severity, systolic blood pressure, frequency of early CT signs of infarction on admission, and higher infarct volume than those who did not have headache. The frequency of posterior territory infarctions was similar between the two groups. Concentrations of glutamate, IL-6, and NO metabolites (NO-m) in CSF were significantly higher in patients with headache than in those without headache.

Table 1

Clinical characteristics and biochemical markers in cerebrospinal fluid of patients with and without headache at stroke onset

	Stroke without headache n=168	Stroke with headache n=73	P-value
Age, year	67.4±10.9	68.6±9.6	0.274
Sex, male (%)	61.3%	83.6%	0.460
Admission delay (hours)	7.6±5.5	9.5±6.4	0.017
Body temperature, °C	36.9±0.7	37.6±0.6	0.314
Serum glucose, mg/dl	160.9±58.7	208.1±56.3	0.642
Systolic blood pressure, mmHg	154±33.6	144.8±23.7	0.004
Diastolic blood pressure, mmHg	87.3±16.9	82.9±14.9	0.804
Admission CSS score	6 [1.5–10]	5 [1.5–10]	0.0452
Early CT signs (%)	55.3	88.3	<0.0001
Stroke aetiology (%):
Atherosclerotic	31.8	43.8	0.004
Cardioembolic	32.7	41.1
Lacunar	20.8	2.7
Undetermined	8.3	12.3
Stroke location, n (%):
TACI (total anterior cerebral infarction)	0 (0%)	18 (24.6%)	<0.001
PACI (partial anterior cerebral infarction)	102 (60.7%)	44 (60.4%)
LACI (lacunar cerebral infarction)	35 (20.8%)	2 (2.7%)
POCI (posterior cerebral infarction)	31 (18.5%)	9 (12.3%)
Progressing stroke (%)	19.6	83.6	<0.0001
Ultimate infarct volume, cc	18.3±19.8	65.2±55.7	<0.0001
Biochemical markers in CSF∗
Glutamate (µm/l)	6.9±4.7	12.4±3.4	<0.0001
IL-6 (pg/ml)	7.7±12.9	38.4±22.0	<0.0001
NO-m (µm/l)	3.3±5.1	6.6±7.0	<0.0001

CSS, Canadian Stroke Scale; CSF, cerebrospinal fluid.

^∗Glutamate was determined in 67 patients with headache and in 145 patients without headache; IL-6 was determined in 30 patients with headache and in 50 patients without headache; NO-m was determined in 37 patients with headache and in 62 patients without headache.

Progressing stroke occurred in 84% and 20% of patients with and without headache, respectively. Patients who showed stroke progression had longer delay from stroke onset to hospital admission, higher stroke severity, systolic and diastolic blood pressure, stronger acute-phase response (increased serum glucose and fibrinogen concentrations, and elevated body temperature), and ultimate infarct volume in comparison with those who had not stroke progression (Table 2). The frequency of headache on admission was significantly higher in the group with subsequent neurological worsening. Headache at the onset of ischaemic stroke predicted progression with a sensitivity, specificity, and positive predictive value of 56%, 99% and 98%, respectively.

Table 2

Clinical characteristics and biochemical markers in cerebrospinal fluid of patients with and without progressing stroke

	No progressing stroke n=147	Progressing stroke n=94	P-value
Age (years)	67.4±11.1	68.3±9.7	0.715
Male (%)	59.9	64.9	0.497
Admision delay (hours)	7.2±5.4	9.7±6.3	0.004
Body temperarture (°C)	36.8±0.6	37.8±0.5	<0.0001
Serum glucose (mg/dl)	146.2±50.5	220.4±49.6	<0.0001
Blood pressure (mmHg)
Systolic	156.1±33.7	144.4±25.4	0.033
Diastolic	89.0±15.9	81.3±16.2	<0.0001
Admission CSS score	6.5 (1.5–10)	4.8 (1.5–8.5)	0.021
Early CT signs (%)	48.3	88.3	<0.0001
Headache (%)	8.2	64.9	<0.0001
Stroke aetiology (%):			0.123
Atherosclerotic	36.7	44.7
Cardioembolic	34.1	37.2
Lacunar	19.7	8.5
Undetermined	9.5	9.6
Ultimate infarct volume (cc)	14.0±14.1	62.7±51.8	<0.0001
Biochemical markers in CSF
Glutamate (µm/l) (n=238)	6.1±4.2	12.9±3.5	<0.0001
IL-6 (pg/ml) (n=83)	3.7±12.9	37.4±21.2	<0.0001
NO-m (µm/l) (n=102)	2.3±5.1	7.36±8.8	<0.0001

CSS, Canadian Stroke Scale; CSF, cerebrospinal fluid.

CSF concentrations of glutamate, IL-6, and NO-m were significantly greater in patients with progressing stroke than in patients with non-progressing stroke (Table 2).

In the logistic regression analysis, headache was an independent factor (OR 16.01, 95% CI 5.40–47.48, P<0.0001) associated with early neurological deterioration, after adjustment for the baseline related factors in the bivariate analyses (Table 3). When ultimate infarct volume was included as a covariate in a further analysis, it was not selected by the model as an independent predictor of progressing stroke (data not shown). The effect of headache remained significant after further adjustment for glutamate and NO-m, but was markedly reduced by the IL-6 effect (Table 4).

Table 3

Adjusted odds ratio of early neurological deterioration for baseline clinical and CT variables

	OR	IC 95%	P-value
Admission delay	1.04	0.91–1.17	0.552
Body temperature	15.21	5.42–42.66	<0.0001
Serum glucose	1.01	1.01–1.03	0.0003
Blood pressure
Systolic	0.97	0.94–0.99	0.031
Diastolic	0.97	0.94–1.008	0.110
CSS on admission	1.32	0.98–179	0.063
Early CT signs of infarction	0.55	0.13–2.21	0.402
Headache	15.43	5.18–45.99	<0.0001

Table 4

Adjusted odds ratio of early neurological deterioration for headache without and with a further adjustment for biochemical markers

Covariates (∗)	Adjusted OR for headache (without biochemical marker)	Adjusted OR for headache (include biochemical marker)	Adjusted OR for the biochemical marker
Model 1
CSF Glutamate	23, 48 (7, 06–78, 13)	21, 67 (3, 18–147, 30)	1, 69 (1, 2–11, 36)
Model 2
CSF IL-6	10, 54 (1, 63–67, 82)	5, 30 (0, 50–56, 34)	1, 10 (1, 01–1, 20)
Model 3
CSF NO-m	15, 13 (2, 37–96, 33)	9, 99 (1, 37–72, 38)	1, 15 (0, 93–1, 42)

^(∗)Covariates include body temperature, serum glucose, systolic blood pressure, and early CT signs of infarction, plus glutamate, IL-6 or NO-m in models 1, 2 and 3, respectively.

Discussion

In the last few years, several risk factors of neurological deterioration in acute cerebral infarcts have been described, including clinical, neuroimaging, ultrasonographic and biochemical parameters; some are associated with an increased risk of early neurological deterioration (progressing infarct), while others are mainly related to the development of delayed worsening (25). Clinical and neuroradiological factors predict worsening in less than 65% of cases (19). In contrast, glutamate concentration in plasma higher than 200 µmol/l is the more powerful marker of progressing stroke, able to correctly predict 97% of cases (10), but is time consuming to determine. Given the known association between headache and elevated glutamate levels in CSF (8) we studied the prognostic value of this symptom in stroke progression. The present findings demonstrate that headache at the onset of ischaemic stroke is an independent predictor of neurological worsening, with high specificity and positive predictive value.

The cause of headache in ischaemic cerebrovascular disease is poorly understood. A simple explanation could be a direct effect of the embolus or thrombus on the vascular wall, with or without reflex dilation. However, this theory does not explain differences in the incidence of headache between stroke subtypes or stroke location (26). In some persons the dura in occipital part is supplied by the posterior cerebral artery and not by the external carotid artery. This could explain, why patients with posterior cerebral artery strokes suffered more often from headache. However, in the current study, the frequency of posterior territory infarctions was similar between the two groups of patients with and without headache.

Headache in cerebrovascular disorders has been mostly related to electrochemical stimulation of the trigeminovascular afferent system, since the pial nerve fibers are of trigeminal origin, and the perivascular fibers contain vasoactive neuropeptides (e.g. calcitonin gene-related peptide, substance P) which, upon release into the vessel wall, increase blood vascular permeability (27, 28). What triggers the trigeminovascular system in ischaemic stroke remains speculative, but a disturbance in blood flow or dilation of the pain-sensitive collateral vasculature has been proposed (29). Depolarization of the trigeminal system results in an afferent discharge perceived as pain, and in a release of vasoactive neuropeptides producing subsequent vasodilation (30). Platelet aggregation at the ischaemic site, and possibly at systemic circulation, provides another potential explanation for triggering the trigeminovascular pathway (31). The products of this release reaction include serotonin and prostaglandins.

Biochemical changes have also been implicated in the pathogenesis of headache in ischaemic stroke. Glutamate concentrations in the CSF of patients with headache at onset of cerebral infarction are significantly higher than those of patients without headache, suggesting glutamate participation in the genesis of pain (8). The release of EAAs might not be the cause but rather an epiphenomenom of pain. Although symptomatic spreading depression, and consequently glutamate release may be related to the cortical lesion (32), glutamate increase is more likely associated with the hypoperfusion that develops in the core of infarction and in the surrounding ischaemic penumbral area (33). The present study replicates preliminary results in a larger series of patients (8) and, interestingly, describes new molecular markers, such as IL-6 and NO metabolites, associated with headache in acute ischaemic stroke.

Increased levels of all these biochemical compounds have been associated with progressing stroke in clinical investigations (10–12), therefore we hypothesize that headache might be a surrogate marker of the molecular mechanisms involved in neurological worsening after acute stroke. This fact could be particularly evident with respect to the inflammatory response that accounts for neurological deterioration (11), because the effect of headache on progressing stroke was markedly reduced after adjustment for IL-6 concentrations.

Although serum cytokines studies in headache patients have yielded conflicting results, increased plasma levels of IL-5 and IL-4 have been described in 84.3% and 37.5% of patients with migraine without aura (34). These results support the growing arguments of an immunoallergic mechanism in the pathophysiology of migraine. Other studies have found a significant decrease in the expression of ICAM-1 during migraine crisis experimentally induced by the administration of isosorbide dinitrate (a nitric oxide donor) and in spontaneous migraine attacks (34). It has been hypothesized that ICAM-1 down-regulation may be a ‘defence’ mechanism to inhibit during a migraine attack the critical step of transendotelial migration into the cerebral tissues of activated leucocytes, as proposed in the ‘sterile inflammation’ hypothesis (35).

Although our results suggest that the association of headache with subsequent progressing stroke may also be related to high glutamate and NO generation, other potential mechanisms may be involved since the odds of neurological deterioration for headache were not substantially modified after adjustment for glutamate and NO-m concentrations (Table 4). Headache could be an epiphenomenon of brain oedema and intracranial hypertension leading to neurological deterioration, however, its prognostic effect was independent of the early CT signs of infarction and ultimate infarct volume.

In conclusion, headache is a strong predictor of early neurological deterioration in acute ischaemic stroke. Excitotoxic and inflammatory mechanisms may play a important role in the pathophysiology of headache that accompany cerebral infarction.

References

Vestergaard

Andersen

Nielsen

Jensen

. Headache in stroke. Stroke 1993; 24: 1621–4.

Arboix

Massons

Oliveres

Arribas

Titus

. Headache in acute cerebrovascular disease: a prospective study in 240 patients. Cephalalgia 1994; 14: 37–40.

Castillo

Leira

Aldrey

Martínez

Noya

. La cefalea en el infarto isquémico de la arteria cerebral media: estudio prospectivo de 100 pacientes. Rev Neurol (Barc) 1995; 23: 737–40.

Portenoy

Abissi

Lipton

Berger

Mebler

Baglino

Headache in cerebrovascualr disease. Stroke 1984; 15: 1009–12.

Olesen

Friberg

Skyhoj Olsen

. Ischaemia-induced (symptomatic) migraine may be more frequent than migraine-induced ischaemic insults. Brain 1993; 116: 187–202.

Nichols

Mawad

Mohr

. Focal headache during balloon inflation in the internal carotid middle cerebral arteries. Stroke 1990; 21: 555–9.

Joseph

D'Andrea

Oster

Welch

KMA

. Whole blood platelets function in acute ischemic stroke. Importance of dense body secretion and effects of antithrombotic agents. Stroke 1989; 20: 38–44.

Castillo

Martínez

Corredera

Aldrey

Noya

. Amino acid transmitters in patients with headache during the acute phase of cerebrovascular ischemic disease. Stroke 1995; 26: 2035–9.

Castillo

Dávalos

Naveiro

Noya

. Neuroexcitatory amino acids and their relation to infarct size and neurological déficit in ischemic stroke. Stroke 1996; 27: 1060–5.

10.

Castillo

Dávalos

Noya

. Progresion of ischemic stroke and excitotoxic amino acids. Lancet 1997; 349: 79–83.

11.

Vila

Castillo

Dávalos

Chamorro

. Proinflammatory cytokines and early neurological worsening in ischemic stroke. Stroke 2000; 31: 2325–9.

12.

Castillo

Rama

Dávalos

. Nitric oxide-related brain damage in acute ischemic stroke. Stroke 2000; 31: 852–7.

13.

Britton

Röden

. Progression of stroke after arrival at hospital. Stroke 1985; 16: 629–32.

14.

Dávalos

Cendra

Teruel

Martínez

Genís

. Deteriorating ischemic stroke. Risk factor and prognosis. Neurology 1990; 40: 1865–9.

15.

Jorgensen

Nakayama

Raaschou

Olsen

. Effect of blood pressure and diabetes on stroke and progression. Lancet 1994; 344: 1865–9.

16.

Toni

Fiorelli

Gentile

Bastianello

Sacchetti

Argentino

Progressing neurological deficit secondary to acute ischemic stroke. Arch Neurol 1995; 52: 670–5.

17.

Dávalos

Castillo

Pumar

Noya

. Body temperature and fibrinogen are related to early deterioration in acute ischemic stroke. Cerebrovasc Dis 1997; 7: 64–9.

18.

Yamamoto

Bogousslavsky

van Melle

. Predictors of neurological worsening in acute ischemic stroke. Stroke 1997; 28: 251.

19.

Dávalos

Toni

Iweins

Lesaffre

Bastianello

Castillo

Neurological deterioration in acute ischemic stroke. Potential predictors and associated factors in the European Cooperative Acute Stroke Study. Stroke 1999; 30: 2631–6.

20.

Adams

Bendixen

Kappelle

Biller

Love

Gordon

Marsh

. Classification of subtype of acute ischemic stroke: definitions for use in a multicenter clinical trial, TOAST. Trial for Org 10172 in acute ischemic stroke. Stroke 1993; 24: 35–41.

21.

Côté

Battista

Shurvell

Norris

Wolfson

. The Canadian Neurological Scale: a preliminary study in acute stroke. Stroke 1986; 17: 731–7.

22.

Dávalos

Crendra

Teruel

Martínez

Genís

. Deteriorating ischemic stroke. Risk factors and prognosis. Neurology 1990; 40: 1865–9.

23.

Côté

Battista

Wolfson

Boucher

Adam

Hachinski

. The Canadian Neurological Scale. Validation and reliability assesment. Neurology 1989; 39: 638–43.

24.

Bamford

Sandercock

Dennis

Burn

Warlow

. Classification and natural history of clinically identifiable subtypes of cerebral infarct. Lancet 1991; 337: 1521–6.

25.

Castillo

Leira

. Predictors of deteriorating cerebral infarct: role of inflammatory mechanisms. Would its early treatment be useful? Cerebrovasc Dis 2001; 11 (Suppl. 1): 40–8.

26.

Mitsias

Ramadan

. Headache in ischemic cerebrovascular disease. Part II. Mechanism and predictive value. Cephalalgia 1992; 12: 341–4.

27.

Moskowitz

. The neurobiology of vascular head pain. Ann Neurol 1984; 16: 157–68.

28.

Moskowitz

Nozaki

Kraig

. Neocortical spreading depression provokes the expression of c-fos protein-like immunoreactivity within trigeminal caudalis nucleus via trigeminovascular mechanisms. J Neurosci 1993; 13: 1167–77.

29.

Moskowitz

. The visceral organ brain. Implications for the pathophysiology of vascular head pain. Neurology 1991; 41: 182–6.

30.

Goadsby

Edvinsson

Ekman

. Vasoactive peptide release in the extracerebral circulation of humans during migraine headache. Ann Neurol 1990; 28: 183–7.

31.

Shah

Beamer

Coull

. Enhanced in vivo platelet activation in subtypes of ischemic stroke. Stroke 1985; 16: 10–3.

32.

Van Harreveld

. The nature of the chick's magnesium-sensitive retinal spreading depression. J Neurobiol 1984; 15: 333–44.

33.

Olsen

Larsen

Herning

Skriver

Lassen

. Blood flow and vascular reactivity in collaterally perfused brain tissue: evidence of an ischemic penumbra in patients with acute stroke. Stroke 1983; 14: 332–41.

34.

Munno

Centonze

Marinaro

Bassi

Lacedra

Causarano

Cytokines and migraine: increase of IL-5 and IL-4 plasma levels. Headache 1998; 38: 465–7.

35.

Martelletti

Granata

Giacovazzo

. Serum IL-1 beta is increased in cluster headache. Cephalalgia 1993; 13: 343–5.

Headache as A Surrogate Marker of the Molecular Mechanisms Implicated in Progressing Stroke

Abstract

Keywords

Introduction

Patients and methods

Statistical analysis

Results

Discussion

References