Role of adenosine 5'-triphosphate in vasospasm after subarachnoid hemorrhage: human investigations.

OBJECTIVE: Adenosine 5'-triphosphate (ATP) is a vasoactive compound found in high concentrations inside erythrocytes. This compound may contribute to vasospasm after subarachnoid hemorrhage (SAH). We assessed the hypothesis that ATP contributes to vasospasm in humans. METHODS: ATP and hemoglobin concentrations were measured in cerebrospinal fluid (CSF) from humans with SAH and in blood incubated in vitro. The vasoactivity of the human CSF samples and of fractionated (fractions with molecular weight greater than or less than 10 kDa) and unfractionated blood incubated in vitro was assessed by application of samples to canine basilar artery segments under isometric tension. RESULTS: ATP in human CSF declined within 72 hours of SAH to concentrations too low to contract cerebral arteries. Vasoactivity of human CSF correlated with the concentration of hemoglobin. The vasoactivity of incubated erythrocyte hemolysates remained high despite a decline in ATP concentrations. Fractionation of incubated erythrocyte hemolysates showed that for incubation periods up to 7 days, all vasoactivity was in a fraction of molecular weight greater than 10 kDa. CONCLUSION: ATP is unlikely to contribute to vasospasm because the concentrations in CSF after SAH in humans are not high enough to cause vasospasm after 72 hours. The vasoactivity of erythrocyte hemolysate is not related to the ATP or ferrous hemoglobin content but may be related to the total hemoglobin content. Therefore, ATP is unlikely to be a major cause of clinically significant delayed vasospasm.

U74389G prevents vasospasm after subarachnoid hemorrhage in dogs.

OBJECTIVE: Oxygen-derived free radicals may contribute to vasospasm after the rupture of an intracranial aneurysm through direct vasoconstricting effects occurring within the arterial wall or, secondarily, by causing lipid peroxidation in the subarachnoid erythrocytes with secondary induction of vasoconstriction. U74389G is a potent inhibitor of lipid peroxidation and a scavenger of oxygen-derived free radicals. This study determined the relative contributions of oxygen-derived free radicals and lipid peroxidation to vasospasm in the double-hemorrhage dog model. METHODS: Sixteen dogs underwent baseline (Day 0) cerebral angiography and induction of subarachnoid hemorrhage by two injections of blood into the cisterna magna 2 days apart. They were randomized to receive drug vehicle (n=8) or U74389G (n=8, 3 mg/kg of body weight/d) intravenously. Drug administration and end point analysis were blinded. The end points were angiographic vasospasm, as assessed by comparison of angiograms obtained before and 7 days after subarachnoid hemorrhage, and the levels of malondialdehyde and salicylate hydroxylation products (dihydroxybenzoic acids) in cerebrospinal fluid and of malondialdehyde in subarachnoid blood clots and basilar arteries 7 days after hemorrhage. RESULTS: Comparisons within groups of Day 0 and Day 7 angiograms and between groups of angiograms obtained at Day 7, showed significant vasospasm in animals in the vehicle group (mean+/-standard error, 51%+/-4) but not in the U74389G group (25%+/-11, P < 0.05, unpaired t test). High-pressure liquid chromatographic assays of malondialdehyde and dihydroxybenzoic acids in cerebrospinal fluid, subarachnoid blood clots, and basilar arteries showed no significant differences between groups. CONCLUSION: The significant prevention of vasospasm by U74389G without change in levels of indicators of free radical reactions suggests that the effect of the drug is related to other processes occurring in the arterial wall and that cerebrospinal fluid levels of oxygen radicals and lipid peroxides are not useful markers of vasospasm.

Adenosine triphosphate and hemoglobin in vasospastic monkeys.

Adenosine triphosphate (ATP) is a vasoactive compound found in high levels inside erythrocytes that may contribute to vasospasm occurring after subarachnoid hemorrhage (SAH). This study was instituted to test whether ATP causes vasospasm in a monkey model. Thirty-two monkeys were randomized to four groups of eight monkeys each to undergo cerebral angiography at baseline (Day 0) and then at Day 7 after subarachnoid placement of: 1) agarose, 2) ATP in agarose, 3) autologous hemolysate in agarose, or 4) purified human hemoglobin A(0) in agarose. Vasospasm was assessed by comparison of Day 0 and Day 7 angiograms between and within groups and by pathological examination of a subset of perfusion-fixed monkeys. Levels of adenine nucleotides were measured on Day 7 in subarachnoid agarose by high-pressure liquid chromatography. There was significant vasospasm of the right middle cerebral artery in groups given ATP (-28 +/- 7% reduction, paired t-test, p < 0.05), hemolysate (-23 +/- 7%, p < 0.05), or pure hemoglobin (-15 +/- 2%, p < 0.005). Analysis of variance revealed no significant differences between groups in diameters of cerebral arteries on Day 7. Pathological examination showed mild inflammation in the subarachnoid spaces of animals exposed to hemolysate or hemoglobin and less inflammation in those given ATP or agarose. There were no pathological changes in the cerebral arteries of animals in any group. Most of the ATP diffused out of the subarachnoid agarose by Day 7, and levels of adenine nucleotides in subarachnoid agarose were higher on Day 7 in animals exposed to hemoglobin or hemolysate. It is concluded that ATP could contribute to vasospasm occurring after SAH but that further investigations are necessary to determine if levels of ATP adjacent to vasospastic arteries are sufficient to contribute to vasospasm. In addition, no observation was made of severe vasospasm with histopathological changes in the arteries equivalent to that produced by whole blood clot in the subarachnoid space of monkeys. It should be determined whether this is because a single compound, such as ATP or hemoglobin, causes vasospasm, but that placing the compound in agarose alters its delivery and decreases the amount of vasospasm produced, or whether vasospasm is a more complex, multifactorial process.