Cerebrospinal fluid factors following subarachnoid haemorrhage accelerate collagen lattice contraction by fibroblasts.

Proliferation in the intimal layer and medial necrosis are the most consistent findings in the cerebral artery following subarachnoid haemorrhage (SAH) in man. Recently, SEM studies from our laboratory have also shown marked endothelial injury as demonstrated by a profuse platelet carpet. Myofibroblasts proliferate in response to the platelet derived growth factor (PDGF), and abundant collagen is present in the vessel wall. We have employed experiments using fibroblast-populated collagen lattices to study cerebrospinal fluid (CSF) from patients with recent SAH. Isolated rat tail collagen and cultured human dermal fibroblasts are mixed together, placed in 35 mm Petri dishes, and allowed to gel. CSF samples are placed on the surface of the collagen lattice, using 0.2 ml saline for control. The collagen lattices are then incubated and daily measurements recorded. We found that CSF samples from patients with recently ruptured aneurysms significantly accelerate contraction of the collagen lattice. The factor in CSF is heat stable and has a molecular weight of less than 6000.

Vasoconstriction of the basilar artery in rabbits: the vasoconstrictor activity of blood, CSF and blood-CSF mixtures incubated for varying periods of time.

After subarachnoid haemorrhage, the amount of blood in the subarachnoid space and the time elapsed since the haemorrhage seem important factors in predicting the occurrence of vasospasm. In this experiment we studied the responses of the rabbit basilar artery to acute topical application of samples of human cerebrospinal fluid (CSF), mixtures of blood and CSF incubated at 37 degrees C for 0 to 12 days. The degree and the duration of the vasoconstriction depended on the time of incubation and on the concentration of blood in the samples. Samples incubated for five and seven days caused severe and long-lasting vasoconstrictions. Samples incubated for 0, 1, 3, 8, 10 and 12 days caused slight or moderate and short-lasting vasoconstriction. Samples incubated for 5, 7, 8, and 10 days showed a linear relationship between the concentration of blood and the degree of constriction up to the blood-CSF ratio of 0.6 to 1 (blood content of 37.5%); further increases of blood concentration did not cause further vasoconstriction. This phenomenon may be related to the saturation of arterial receptors. The course of the vasoconstriction using samples with different blood concentrations suggests the existence of more vasoactive substances. This work seems to confirm the importance of blood concentration in the CSF and of time as factors governing the development of vasospasm.

CSF smooth-muscle constrictor activity associated with cerebral vasospasm and mortality in SAH patients.

Cerebrospinal fluid (CSF) was collected preoperatively (by lumbar puncture) or perioperatively (by lumbar or ventricular drain) from 32 patients with subarachnoid hemorrhage (SAH) from ruptured intracranial aneurysms. Samples were also obtained from six control patients without evidence of subarachnoid blood. Smooth-muscle constrictor activity in the CSF was measured by bioassay using the isolated rat stomach fundus preparation. Concentrations of unidentified smooth-muscle constrictor substances were considerably greater in CSF from a group of seven patients with evidence of severe angiographic vasospasm and/or delayed ischemic deficits who died (73.8 +/- 39.7 nmol/liter prostaglandin E2 (PGE2) equivalents), as compared to 25 other SAH patients who survived (6.5 +/- 1.4 nmol/liter PGE2 equivalents), and six control patients (1.17 +/- 0.34 nmol/liter PGE2 equivalents). The data suggest that there is a relationship between smooth-muscle constrictor substances in the CSF after SAH and both the degree of angiographic vasospasm and the outcome. It is possible that this relationship might be exploited clinically.

Vasoactivity of cerebrospinal fluid from patients with brain swelling.

The triggering mechanisms of diffuse cerebral swelling are poorly understood. The findings in this study are in keeping with the hypothesis that chemicals released by brain trauma into the CSF promote vasodilatation. Other proposals have included brainstem stimulation with increased cerebral blood flow and defective autoregulation. A number of factors may be operative.

Vasospasmogenic substance produced following subarachnoid haemorrhage, and its fate.

Fresh blood and supernatants of blood-CSF mixtures incubated for 1 to 15 days were applied to the basilar artery of adult cats, and the degree of constriction was measured with a surgical microscope. The constriction due to fresh blood was weak and transient. It seems possible to assume that serotonin isolated from platelets participates greatly in the transient vasoconstriction induced by fresh blood. Supernatants of blood-CSF mixtures incubated for three days had weak activity in comparison with the powerful and long-lasting activity of those incubated for seven days. Furthermore, mixtures incubated for 15 days had little or no activity. This change in the vasoconstrictive activity was similar to, and coincides chronologically with clinical late spasm following subarachnoid haemorrhage 34. We investigated the vasospasmogenic substance in the seventh day mixture. Heat coagulation, ultrafiltration, sephadex G-100 gel-chromatography, disc-electrophoresis, and Spectrophotography show that extracellular oxyHb has a strong spasmogenic activity. In the 15th day mixture, oxyHb is spontaneously converted to metHb. Experimentally, oxyHb has a strong vasoconstrictive activity, and metHb has no vasoconstrictive activity. We have had success in oxidizing oxyHb into metHb with sodium nitrite, thus preventing experimental vasospasm.

[Role of serum fibrinogen penetrating the subarachnoid space in the pathogenesis of brief cerebral arteriospasm]. / Die Rolle des in den Subarachnoidalraum eingedrungenen Blutfibrinogens in der Pathogenese des kurzzeitigen Hirnarterienspasmus.

A mixture of fibrinogen (4 ml in a 1 per cent solutions) and thrombin (10 I.U.) were instilled into the cisterma magna of 18 dogs which led to the formation of a fibrin clot in the subarachnoid space. This fibrin clot produced a spasm of the large cerebral arteries, which resulted in a narrowing to 84.8 +/- 1.8 per cent of the initial value. The subsequent administration of 400 units of fibrinolysin into the cisterna magna led to the dissolution of the fibrin clot and the dilatation of the arterie to 116.3 +/- 2.5 per cent of the initial value. Experiments carried out in 15 cats showed that, after being brought in the subarachnoid space, the spasmogenic factor of the clot, which consisted of all blood components, passes through the arachnoid into the subdural space. The penetration of the spasmogenic factor from the blood clot located on the surface of the arachnoid from the subdural to the subrarachnoid space was not observed. The arachnoid proves to be a membrane which can only be passed in one direction by the spasmogenic blood factor.

A pressor substance in the cerebrospinal fluid of normotensive and hypertensive patients.

A pressor polypeptide was isolated from the cerebrospinal fluid of normotensive and hypertensive patients. Pharmacologically it behaves like angiotensin I. A very significant correlation (r = 0.83, P < 0.001) was found between the concentration of this polypeptide and the blood pressure of essential hypertensive patients.