1979 Volume 19 Issue 1 Pages 61-71
To investigate the spasmogenicity of hemoglobins, we have analyzed the amount of total hemoglobin, oxyhemoglobin and methemoglobin in bloody or xanthochromic cerebrospinal fluid (CSF) and in intracranial hematoma. Morphological changes of the endothelial surface of spastic arteries and their vasorelaxative responses to papaverine were also studied using rat basilar artery.
Total hemoglobin was determined in 85 CSF samples obtained from patients with ruptured aneurysm. Vasospasm was seen most frequently in patients with bloody or xanthochromic CSF. However, it was also true in many cases that bloody or xanthochromic CSF was not accompanied by vasospasm. No significant correlation could be observed between vasospasm and the amount of hemoglobin. In analysis of 90 CSF samples, vasospasm was seen in 39% of those containing oxyhemoglobin and in 64% of those containing both oxyhemoglobin and methemoglobin. The relatively high frequency of vasospasm in the presence of methemoglobin in the CSF is likely attributable to the large amount of hemoglobin or blood clots in the subarachnoid space.
In comparative studies of hemoglobin and barium chloride, vasorelaxative response to papaverine was found to be highly correlated to the severity and duration of vasospasm. Scanning electron microscopic studies demonstrated that endothelial cells were damaged after vasospasm induced by both hemoglobin and barium chloride. Endothelial alterations were observed initially as an increase of craters, blebs, and microvilli and poorly defined marginal folds, followed by separation of cell junctions and by detachment of endothelial cells. Adherence and aggregation of platelets were located at the denuded regions of the endothelium. These observations suggest that the organic changes of the arterial wall might be implicated in the pathogensis and pathophysiology of vasospasm.
