The response of the regional cerebral blood flow (CBF) to brain topical superfusion of an increased K+ concentration in the artificial cerebrospinal fluid (ACSF) ([K+ ]ACSF ) at 20 mM was characterised in a closed cranial window preparation in barbiturate anaesthetised and ventilated rats. It was concluded that the vasodilator nitric oxide (NO) is a modulator of the rise in CBF following increased [K+ ]ACSF .
As a by-product of this study, we accidentally discovered spontaneous ischaemic responses of an unknown mechanism in response to the co-application of an inhibitor of the NO-synthase (NOS) with increased [K+ ]ACSF . These spontaneous ischaemic responses were characterised in the next study.
It turned out that the spontaneous ischaemic responses to the co-application of a NOS inhibitor with increased [K+ ]ACSF were caused by an inverted coupling between neuronal/astroglial metabolism and CBF: The increased [K+ ]ACSF triggered a neuronal/astroglial depolarisation wave which, under physiological conditions, leads to a cortical spreading hyperaemia. However, under the conditon of NOS inhibitor with increased [K+ ]ACSF , the coupling was inverted so that vasoconstriction was induced instead of vasodilatation resulting in energy compromise. Under this condition, the neuronal/astroglial [Seite 15↓] network was unable to repolarise since the process of repolarisation requires energy. Because the vasoconstrictive stimulus was coupled to the neuronal/astroglial depolarisation, a vicious circle of prolonged ischaemia was established. That neuronal activation can induce ischaemia, was an unexpected finding. The most characteristic feature of this new ischaemic variant was its propagation in the cerebral cortex together with the neuronal/astroglial depolarisation wave. This caused us to name it ‘cortical spreading ischaemia’.
The most potent natural NO-lowering agent is haemoglobin when it is released into the extracellular space. Haemoglobin binds NO with an affinity 1500 times higher than its affinity for oxygen at its haeme iron and reactive sulfhydril groups at cysteineβ 93. Therefore, we investigated whether cortical spreading ischaemia also occurred in response to haemoglobin and increased [K+ ]ACSF . We showed that this protocol was at least similarly effective as NOS inhibition with increased [K+ ]ACSF (Fig. 2).
Haemoglobin (21.03 ± 0.75 mM) and K+ (102.4 ± 3.9 mM) are the protein and ion, respectively, with the highest concentration in the red blood cell. As described in PART I, DINDs occur in a close temporal correlation with haemolysis of the subarachnoid blood. The subarachnoid level of haemoglobin reaches its maximum on the seventh day after SAH (Pluta et al. 1998). Values of up to 500 µM have been reported from intracranial haematomas in humans (Ohta et al. 1980). Similarly, extracellular K+ concentrations of up to 50 mM were measured in intracranial haematomas in neurosurgical patients (Ohta et al. 1983). This led to the hypothesis that DINDs may be caused by a mechanism related to cortical spreading ischaemia.
Under physiological conditions, a cortical spreading hyperaemia (upper trace, left part) (CBF = cerebral blood flow) is coupled to a spreading neuronal depolarisation wave (lower trace, left part) (DC = slow direct current potential) propagating from cranial window 1 to cranial window 2. However, when red blood cell products, haemoglobin and increased K+, are in the subarachnoid space, the coupling between neuronal activation and cerebral blood flow is inverted so that a cortical spreading ischaemia (upper trace, right part) is coupled to the spreading neuronal depolarisation wave (lower trace, right part). Due to a vicious circle, both depolarisation wave and its CBF response are prolonged under this condition (see text).
In this study, we showed that cortical spreading ischaemia induced by haemoglobin and increased [K+ ]ACSF in the rat led to bell-shaped or laminar infarcts in the cortex very similar to the pathoanatomical pattern of DINDs in man (Fig. 1 ). That the lesions are essentially restricted to the brain cortex is explained by the fact that spreading neuronal depolarisation waves do not run in the white matter. Deeper structures may only be damaged [Seite 17↓] by the waves when long penetrating arteries are constricted at the cortical level.
In this study, we investigated the effect of an NO-dependent and NO-independent vasodilator, S -nitroso-N -acetylpenicillamine and papaverine, respectively, on cortical spreading ischaemia produced by NOS inhibition with increased [K+ ]ACSF . We found that particularly the NO-donor but also papaverine was able to convert cortical spreading ischaemia to cortical spreading hyperaemia.
The Brazilian physiologist Leão first described neuronal/astroglial depolarisation waves. He coined the name ‘cortical spreading depression’ in 1944. One year later, together with his colleague Morison, he came up with the hypothesis that cortical spreading depression may be the correlate of migrainous aura based on the striking resemblance of its electrophysiological features and the clinical presentation of migrainous aura. Cortical spreading depression is a short, regenerative depression of spontaneous neuronal activity in the grey matter that propagates at a rate of approximately 3 mm/min (Lauritzen 1994). The neuronal/astroglial depolarisation wave underlying cortical spreading ischaemia is not the same as but closely related to cortical spreading depression. Therefore, it was obvious to investigate whether migrainous aura-like symptoms may occur during the clinical course after SAH. Interestingly, within a short period of time, we found two patients with migraine who [Seite 18↓] experienced migrainous-aura like symptoms several minutes after the onset of acute headache induced by SAH. Both patients developed a DIND later on. Unfortunately, at the time of DIND, the patients were unable to adequately describe their symptoms because of a profoundly altered mental status. However, also the initially gained information on a migrainous aura-like attack as a symptom of SAH was remarkable since literature suggests the occurrence of spreading depression-like depolarisations at the acute stage of SAH in animal models (Hubschmann and Kornhauser 1980; Busch et al. 1998). Our findings were also of clinical value since migraine is the most important misdiagnosis of SAH, which can lead to a delay of aneurysm surgery (Edlow and Caplan 2000).
The 21-residue peptide endothelin-1 has gained much attention in neurological and neurosurgical research being both a powerful vasoconstrictor and a neuronal and astroglial modulator. Numerous studies have investigated whether endothelin-1 is involved in the pathogenesis of arterial spasm after SAH (reviewed by Zimmermann and Seifert 1998). Endothelin-1 was also suggested to play a role in stroke and migraine. In our animal study, we demonstrated that this endothelium-derived factor is the most potent inductor of cortical spreading depression currently known. This may have implications particularly for migraine research since clinical observations strongly suggest that endothelial irritation may somehow initiate one of the pathways leading to cortical spreading depression. In addition, it may be of significance for DINDs since preliminary results indicate that endothelin-1 can replace increased [K+ ]ACSF in the protocol used to initiate cortical spreading ischaemia.
Over the last 20 years, the risk to develop a DIND has significantly decreased. This has been related, at least partially, to the prophylactic use of nimodipine and improved fluid management (see above). In this study, we investigated whether nimodipine or moderate hypervolaemic haemodilution convert cortical spreading ischaemia produced by haemoglobin with increased [K+ ]ACSF to cortical spreading hyperaemia (compare II 4). The positive result of our study supported a link between DINDs and cortical spreading ischaemia (Fig. 3 ).
It is often erroneously believed that the typical substrate of DINDs after SAH are territorial infarcts. This is not supported by the autopsy studies, which showed a large predominance of triangular, round or laminar cortical ischaemic lesions and, in addition, ischaemic patches in deeper cerebral structures (Robertson 1949; Falconer 1954; Birse and Tom 1960; Crompton 1964; Stoltenburg-Didinger and Schwarz 1987; Neil-Dwyer et al. 1994). In non-operated patients, the relation between autopsy cases with cortical to those with territorial infarcts was 13:1, while in operated patients, it was 3:1 due to a relative increase in the frequency of territorial infarcts (Stoltenburg-Didinger and Schwarz 1987). The significance of the cortical lesions is clinically underestimated because they are typically undetected by CT (Stoltenburg-Didinger and Schwarz 1987; Neil-Dwyer et al. 1994). We presented two cases demonstrating the suitability of magnetic resonance (MR) imaging to visualize such cortical lesions (Fig. 1 ).
Comparison of cerebral blood flow (CBF)- and direct current (DC)-responses to the spreading neuronal/astroglial depolarisation wave (SND). While the upper two traces represent CBF at two different laser-Doppler positions, the lower trace gives the DC-potential at one cranial window.
A Physiological artificial cerebrospinal fluid (ACSF) topically, 0.9% saline intravenously (control): The depolarisation wave is associated with a cortical spreading hyperaemia (CSH) under physiological conditions (= cortical spreading depression (CSD) of Leão).
B ACSF containing haemoglobin ([Hb]ACSF ) and elevated K+ ([K+ ]ACSF ) topically, 0.9% saline intravenously: Cortical spreading ischaemia (CSI) in response to the depolarisation wave.
C ACSF containing the red blood cell products topically, nimodipine is given intravenously at a dose of 2 µg/kg bodyweight/min (i.v.): Cortical spreading ischaemia reverts to a cortical spreading hyperaemia in presence of nimodipine. Only a short initial hypoperfusion (*) preceding cortical spreading hyperaemia indicates the presence of the red blood cell products in the subarachnoid space. The effect of moderate hypervolaemic haemodilution was similar to that of nimodipine.
|© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.|
|DiML DTD Version 3.0||Zertifizierter Dokumentenserver|
der Humboldt-Universität zu Berlin
|HTML-Version erstellt am: