Nimesulide 90 mg orally twice daily does not influence postoperative morphine requirements after major chest surgery.

BACKGROUND: Cyclooxygenase 2 inhibition has proven analgesic efficacy in a variety of surgical procedures. We postulated that perioperative cyclooxygenase 2 inhibition significantly reduces postoperative morphine requirements after major thoracic surgery and investigated the site of this potential analgesic effect. METHODS: Ninety-two patients participated in this single-center, double-blind, randomized, placebo-controlled, parallel-group trial. Patients between the ages of 18 and 80 yr undergoing a thoracotomy or median sternotomy were randomized to receive either nimesulide or placebo in combination with a standard analgesic regimen perioperatively. Nimesulide was administered orally the evening before surgery and at 12-h intervals for 5 days postoperatively. The primary efficacy variables were morphine consumption and pain scores for the first 48 h postoperatively. The secondary efficacy variable was the effect of nimesulide on cyclooxygenase activity in cerebrospinal fluid (CSF). RESULTS: Pain scores at rest or with movement, and total morphine consumption for the first 48 h postoperatively, were not statistically different between the groups. The mean difference in total morphine consumption up to 48 h postoperatively between the nimesulide and placebo group was a 9.0 mg reduction (95% CI: -28.9 to 10.9 mg) (P = 0.37). Adjusted mean (se) CSF 6-keto-PGF1alpha (6-keto-PGF1alpha) concentrations increased by 54.7 (25.7) pg/mL from preoperatively to Day + 2 postoperatively in the placebo group, whereas adjusted mean (se) CSF 6-keto-PGF1alpha concentration decreased by 0.6 pg/mL (18.2 pg/mL) in the nimesulide group. These changes were not statistically different between the groups (P = 0.095). CONCLUSION: Nimesulide, at a dose of 90 mg twice daily in combination with a standard analgesic regimen, does not influence pain scores, morphine requirements, or CSF prostaglandin levels after major thoracic surgery.

Effects of dopamine on pial arteriolar diameter and CSF prostanoid levels in piglets.

We examined effects of topically applied dopamine on pial arteriolar diameter and CSF prostanoid levels in newborn pigs. Vascular responses were determined using the closed cranial window technique and intravital microscopy, and prostanoids were determined by radioimmunoassay. Topical application of dopamine did not change arteriolar diameter at 10(-7)-10(-5) M, but constricted arterioles at 10(-4) (16%) and 10(-3) M (30%). Intravenous administration of indomethacin (5 mg/kg) did not alter this constriction. In addition, CSF prostanoid levels did not increase in response to application of dopamine except for a modest increase of prostaglandin E2 at 10(-3) M. We conclude that dopamine is a constrictor at high doses of piglet pial arterioles and that this response is not modified by endogenous prostanoids.

Alteration of thromboxane and prostacyclin levels in experimental spinal cord injury.

We measured levels of thromboxane B2 and 6-keto-PGF1 alpha in rabbit spinal cord and cat CSF after impact injury to spinal cord. Rabbit tissue thromboxane B2 levels increased more than 6-keto-PGF1 alpha. In cat, CSF thromboxane B2 was higher the first hour postinjury; CSF 6-keto-PGF1 alpha also increased, but less so. These results imply activation of arachidonic acid metabolism. The relatively greater increase of thromboxane B2 suggests that thromboxane-prostacyclin imbalance may contribute to post-traumatic ischemia.

A study on cisternal CSF levels of arachidonic acid metabolites after aneurysmal subarachnoid hemorrhage.

Arachidonic acid (AA) metabolites may play an important role in the pathogenesis of cerebral vasospasm which complicate subarachnoid hemorrhage. Authors have studied levels of 4 major AA metabolites in lumbar CSF samples and in CSF collected from perianeurismatic cisterns of 40 patients admitted with diagnosis of subarachnoid hemorrhage. Lumbar levels of AA metabolites are significantly higher in SAH patients than in control cases; moreover, cisternal CSF levels of PGD2, TxB2 and LTC4 are significantly higher than lumbar levels. Cisternal CSF levels (expressed in pg/ml +/- SEM) are in the "spasm" group: PGD2: 1129.62 +/- 146.33; 6-keto-PGF1 alpha: 214.2 +/- 19.96; TxB2: 4350.25 +/- 656.87; LTC4: 2582.19 +/- 381.83. In the "no spasm" group: PGD2 460.1 +/- 55.89; 6-keto-PGF1 alpha: 306.37 +/- 88.74; TxB2: 5752.5 +/- 899.25; LTC4: 812.92 +/- 142.06. Statistical analysis (paired t-test) shows values significantly higher for cisternal levels of PGD2 (P less than 0.005) and LTC4 (P less than 0.005) in patients presenting vasospasm. This suggests the importance of the subarachnoidal clot as a source of vasoactive compounds. Higher levels of leukotriene C4 in patients presenting vasospasm suggest a role for the compound in the genesis of local inflammatory processes and morphological changes of the arterial wall.

Effect of hydrocephalus on prostaglandins and thromboxane B2 in ventricular cerebrospinal fluid.

The concentrations of prostaglandin F2 alpha, prostaglandin E2, 6-ketoprostaglandin F1 alpha (prostacyclin metabolite), and thromboxane B2 were assayed in ventricular cerebrospinal fluid obtained from 28 patients with hydrocephalus (17 obstructive, 11 communicating). Seven patients received dexamethasone or hydrocortisone on the day of sampling. No patient received nonsteroidal anti-inflammatory compounds for 48 hours before sampling. The median values did not differ significantly between the two types of hydrocephalus or from the concentrations in lumbar cerebrospinal fluid obtained from patients without intracranial pathology during lumbar myelography for possible lumbar disc disease. Hence, there is no evidence that eicosanoids accumulate in the ventricles in hydrocephalus, and it is unlikely that they have a significant role in its symptomatology.

Experimental and clinical studies of eicosanoids in cerebrospinal fluid after spinal cord injury.

OBJECTIVE: In an attempt to elucidate a possible role for eicosanoids in the pathogenesis of spinal cord injury (SCI), we measured the concentration of leukotriene (LT) C4, thromboxane B2, and 6-keto-prostaglandin F1 alpha in cerebrospinal fluid in both a canine experimental model and 11 patients with SCIs. METHODS: The eicosanoid concentration in cerebrospinal fluid was measured by radioimmunoassay. Neurological severity was assessed according to the grading system of Frankel et al.. Control samples were obtained from 20 patients without SCIs. RESULTS: In the canine model, a significant increase in all eicosanoid levels was found on Days 1 to 7, which subsequently returned to the control levels. In the clinical study, the highest mean (+/- standard error of the mean) concentrations of LTC4, thromboxane B2, and 6-keto-prostaglandin F1 alpha in the acute stage of SCI were 95.9 +/- 10.7, 175.2 +/- 38.2, and 167.5 +/- 39.9 pg/ml, respectively. These concentrations were five to nine times higher than control levels. There was a good correlation between cerebrospinal fluid LTC4 levels and the neurological severity. The time-dependent change in LTC4 concentrations in seven patients with SCIs was similar to that observed in the canine model. In addition, the highest mean concentrations of the eicosanoids measured in patients with complete paralysis was also similar to those of the canine model. The eicosanoid concentrations in five patients with SCI were measured more than 6 months after the onset of injury. Although all eicosanoid levels had elevated in the acute stage of injury, they were not elevated and showed the same levels as the controls at the chronic stage. CONCLUSION: The findings suggest that enhanced arachidonate metabolism occurs in humans and support the evidence from animal experiments that emphasizes the importance of eicosanoids in the secondary processes mediating ischemia and edema.

Arachidonate metabolites and vasospasm after subarachnoid haemorrhage.

A wide literature exists about the pathogenesis of cerebral arterial spasm following subarachnoid haemorrhage: several compounds have been identified in human cerebrospinal fluid as possible vasoactive agents involved in the biochemical mechanism of vasospasm onset. Many experimental evidences exist for a major involvement of arachidonate metabolites. The present work represents a review of experimental data supporting the hypothesis of cerebral arterial spasm as a result of an imbalanced vascular regulatory mechanism involving arachidonate metabolites. The authors have also monitored, in 25 cases of aneurysmal subarachnoid haemorrhage, lumbar and cisternal CSF levels of prostacyclin and PGD2, as representative of vasodilating and, respectively, vasoconstrictor compounds. In all cases CSF arachidonate metabolite levels after SAH were significantly higher than in control cases. Ten patients presented with symptomatic vasospasm: lumbar CSF PGD2 levels show fluctuations with superimposed peaks related to the neurological deterioration due to vasospasm, while lumbar CSF prostacyclin concentration-trend suggest a decreasing synthesis. In 15 patients presenting without vasospasm, lumbar CSF concentration of arachidonate metabolites are in a 'steady-state'. These data confirm the existence of an imbalanced biochemical situation promoting vasospasm, markedly in cisterns near to the ruptured aneurysmal wall. The evaluation of cisternal CSF levels of arachidonate metabolites supports the hypothesis of the clotting phenomenon around the ruptured aneurysm as an important predictive pattern of vasospasm, as shown in CT findings.

Levels of 6-ketoprostaglandin F1 alpha in neonatal cerebrospinal fluid.

6-Ketoprostaglandin F1 alpha determinations were made by radioimmunoassay on samples of cerebrospinal fluid from 41 neonates. Levels were below the smallest quantity detectable (50 pg/ml) in 29 and greater than 200 pg/ml in only 3 babies. These results suggest that 6-ketoprostaglandin F1 alpha is not a major prostaglandin in the cerebrospinal fluid of human infants.

Eicosanoid levels in CSF of premature infants with posthemorrhagic hydrocephalus.

The cerebrospinal fluid (CSF) of 11 premature infants suffering from posthemorrhagic hydrocephalus was examined by radioimmunoassay for prostaglandin (PG) E2, PGF2 alpha, PGD2, 6-keto PGF1 alpha, thromboxane B2 (TxB2) and peptidoleukotrienes (LTC4/LTD4). The LTs were detected in the CSF of more of these patients (70%) than any of the other eicosanoids, and usually in the highest concentration. Among the 11 posthemorrhagic patients CSF eicosanoid levels were highest when determined soon after injury. Moreover, the variety of eicosanoids present, as well as concentrations, in these infants decreased with time. The types of eicosanoids most evident in the CSF of patients who required shunting were TxB2 and LTs, being present together in 5 of 6 (83%) of these infants. In contrast, 1 of 5 (20%) of the patients who did not require this neurosurgical intervention contained both TxB2 and LTs, the remaining having only one or neither eicosanoid. The highest average concentration for each eicosanoid studied was (pg/ml): PGE2, 628; PGF2 alpha, 985; PGD2, 1410; 6-keto PGF1 alpha, 544; TxB2, 486 and LTs, 1229. This study is the first to demonstrate that the CSF of preterm infants may contain a wide variety of eicosanoids and indicates that these lipids are a manifestation of neurological assault.

Cisternal and lumbar CSF concentration of arachidonate metabolites in vasospasm following subarachnoid hemorrhage from ruptured aneurysm: biochemical and clinical considerations.

Two representative cases of subarachnoid hemorrhage in which prostaglandin D2 (PGD2) and 6-keto-prostaglandin F1 alpha (6-keto-PGF1 alpha), stable metabolite of prostacyclin (PGI2), were monitored with serial lumbar punctures and detected in cisternal CSF during operations for aneurysm, are reported. In the case with demonstrated arterial vasospasm, prostaglandin D2 has a concentration trend with characteristic peak related to vasospasm; the synthesis of prostacyclin appears inhibited after the hemorrhage. In the patient without radiologic evidence of vasospasm, arachidonate metabolite concentration trend appears in a steady-state. Cisternal prostaglandin D2 concentration in the patient with demonstrated vasospasm is two times the highest lumbar CSF concentration, while 6-keto-prostaglandin F1 alpha concentration is very low. This suggests the role of the clotting phenomenon and likely confirms the importance of arachidonate metabolites in the genesis of cerebral arterial spasm following subarachnoid hemorrhage.

Arachidonic acid metabolism following aneurysm rupture. Evaluation of cerebrospinal fluid and serum concentration of 6-keto-prostaglandin F1 alpha and thromboxane B2 in patients with subarachnoid hemorrhage.

Experimental investigations have suggested an important role of arachidonic acid metabolites in the genesis of cerebral vasospasm following subarachnoid hemorrhage. In this clinical study the cerebrospinal fluid (CSF) and serum levels of the two main arachidonic acid metabolites prostacyclin and thromboxane A2 are evaluated by measuring their stable degradation products 6-keto-prostaglandin F1 alpha (6-keto-PGF1 alpha) and thromboxane B2 (TXB2) using radioimmunoassay methods during the pre- and postoperative course in patients after aneurysm rupture. Although the serum levels of both substances do not seem to be important for the clinical course of the patients, the CSF concentrations of 6-keto-PGF1 alpha and TXB2 provide important data. A close correlation between the initial TXB2 level of the individual patient and the amount of blood in the basal cisterns as detected by computed tomography scan can be demonstrated. The predictive value of this additional information for the occurrence of cerebral angiospasm is discussed. Comparing the CSF levels of both metabolites the slight preoperative elevation of 6-keto-PGF1 alpha is significantly surmounted by an extraordinary rise in TXB2 concentration. Postoperatively, after cleavage of the basal cisterns there is a decline in the CSF levels of both substances. The pre- and postoperative clinical course in comparison to the CSF levels of 6-keto-PGF1 alpha and TXB2 is demonstrated in four patients. A nearly normal course of TXB2 and 6-keto-PGF1 alpha seems to be associated with an uneventful clinical course, whereas a high TXB2 level--whether occurring preoperatively or, even more important, as a secondary postoperative rise--seems to be associated with ischemic complications and neurological deterioration. It is suggested that pre- and postoperative monitoring of CSF levels of 6-keto-PGF1 alpha and especially TXB2 may serve as a possible indicator for the detection of patients at risk of developing cerebral vasospasm.

[Cerebral vasospasm. Clinical and experimental aspects]. / Il vasospasmo cerebrale. Aspetti clinico-sperimentali.

A large number of experimental data suggest a possible biochemical hypothesis for the trigger stimulus of cerebral vasospasm after subarachnoid hemorrhage (SAH). Among several classes of possible spasmogens, arachidonic acid metabolites may play a primary role. Authors have measured with radioimmunoassay technique (R.I.A.) the levels of four arachidonate metabolites (PGD2, TxB2, 6-keto-PGF1 alpha and i-LTC4) in lumbar and cisternal cerebrospinal fluid (CSF) of patients admitted with diagnosis of aneurysmal SAH. In all cases a significant activation of arachidonate metabolism is found, if compared to control cases. Patients with demonstrated vasospasm have significantly higher CSF levels of PGD2 and i-LTC4. Cisternal CSF levels of four metabolites are significantly higher than lumbar CSF levels. This suggests the correlation between subarachnoidal clot extension and the risk for vasospasm. Authors also present an experimental animal model of SAH, which is reliable from a pathological standpoint. This model could be therefore used in the study of neurochemical and neuropharmacological aspects of SAH.

Clinical implication of the changes of cAMP, TXA2 and PGI2 in CSF of asphyxiated newborns.

To evaluate the changes of 3', 5'-cyclic adenosine monophosphate (cAMP), thromboxane A2(TXA2) and prostacyclin (PGI2) in cerebrospinal fluid (CSF) in the asphyxiated newborn and explore their roles in hypoxic-ischemic brain damage (HIBD). Thirty-six full term newborns were divided into 3 groups, including 12 with moderate-severe hypoxic-ischaemic encephalopathy (HIE), 13 with mild HIE, 11 without HIE (control group). The levels of cAMP, TXB2 (TXA2 metabolite) and 6-keto-PGF1 alpha (PGI2 metabolite) in CSF and plasma were measured 36-72 h after birth by RIA, and the concentrations were expressed as nM/L (cAMP), ng/L(TXB2 and 6-keto-PGF1 alpha). The infants were followed-up at 6 and 12 month of age and Mental Development Index (MDI) and Psychomotor Development Index (PDI) were measured using Bayley Scales of Infant Development (BSID). The CSF cAMP level in moderate-severe HIE group was 8.60 +/- 2.40, significantly lower than that of the mild HIE group (14.83 +/- 2.84) and the control group (24.43 +/- 2.39) (for both P < 0.01). The levels of TXB2 and 6-keto-PGF1 alpha in CFS in the moderate-severe HIE group (206.06 +/- 29.74, 168.47 +/- 23.02, respectively) were significantly higher than in the mild HIE group (83.37 +/- 28.57, 131.42 +/- 16.57, respectively, P < 0.01) and the control group (41.77 +/- 21.58, 86.23 +/- 13.05, respectively, P < 0.01). The level changes of cAMP, TXB2 and 6-keto-PGF1 alpha in plasma in all groups were similar to those in CSF, but no significant difference was found between mild HIE group and the control group (P > 0.05). The follow-up results showed that MDI and PDI of the moderate-severe HIE group were the lowest (84.79 +/- 13.34, 83.50 +/- 13.28, respectively), followed by mild HIE group (102.19 +/- 7.02, 99.94 +/- 9.08, respectively), with the control group being the highest (116.63 +/- 12.08, 116.69 +/- 10.87, respectively). Univariate analysis showed some significant difference (the moderate-severe HIE group vs. the mild HIE group or the control group, P < 0.01; the mild HIE group vs. the control group P < 0.05). The results suggested that the concentration of cAMP, TXA2 and T/K ratio in CSF after neonatal asphyxia might be sensitive markers in evaluating the severity of brain damage in early stage and predicting the future outcome.

Gross-total hematoma removal of hypertensive basal ganglia hemorrhages: a long-term follow-up.

BACKGROUND AND PURPOSE: Hypertensive basal ganglia hemorrhage (HBGH) accounts for 35%-44% of cases of hypertensive intracranial hemorrhage (ICH), which is one of the most devastating forms of cerebrovascular disease. In this study, intracerebral hematoma was evacuated with a burr hole craniectomy. The relationships of residue hematoma volume to brain edema, inflammation factors and the long-term prognosis of HBGH patients were studied. METHODS: One hundred and seventy-six patients with HBGH were randomly divided into gross-total removal of hematoma (GTRH) and sub-total removal of hematoma (STRH) groups. The pre-operative and post-operative data of the patients in the two groups were compared. The pre-operative data included age, sex, hematoma volume, time from the ictus to the operation, Glasgow Coma Scale (GCS) scores, and the European Stroke Scale (ESS) scores. The post-operative information included edema grade, level of thromboxane B2 (TXB2), 6-keto-prostaglandin F1a (6-K-PGF1a), tumor necrosis factor-a (TNF-a) and endothelin (ET) in hematoma drainage or cerebral spinal fluid (CSF), ESS and Barthel Index (BI). RESULTS: There was no statistical difference between the two groups (P>0.05) in the pre-operative data. The levels of TXB2, 6-K-PGF1a, TNF-a and ET in the GTRH group were significantly lower than those in the STRH group at different post-operative times. The ESS in the GTRH group increased rapidly after the operation and was higher than that in the STRH group. There was a significant difference between the two groups (P<0.05). The post-operative CT scan at different times showed that the brain edema grades were better in the GTRH group than in the STRH group. The BI was higher in the GTRH group than in the STRH group (P<0.05). CONCLUSIONS: GTRH is an effective method to decrease ICH-induced injury to brain tissue. Such effect is related to decreased perihematomal edema formation and secondary injury by coagulation end products activated inflammatory cascade.

Carbon monoxide-prostaglandin E2 interaction in the hypothalamic circulation.

The heme oxygenase (HO)-carbon monoxide pathway was earlier shown to increase hypothalamic blood flow after inhibition of nitric oxide synthesis in rats. We hypothesized that this effect is mediated by prostaglandin E2 (PGE2). Inhibition of constitutive HO activity decreased cerebral PGE2 production and simultaneously increased hypothalamic nitric oxide synthase (NOS) activity without changing hypothalamic blood flow. Furthermore, HO blockade induced cyclooxygenase-dependent decrease and NOS-mediated increase of the hypothalamic blood flow after inhibition of NOS and cyclooxygenase, respectively. Therefore, constitutive carbon monoxide release seems to have two indirect effects on the hypothalamic circulation: vasodilation mediated by PGE2 and vasoconstriction as a result of NOS inhibition.

Prostanoid concentrations in human CSF following acute ischaemic brain infarction.

Thromboxane A2 and prostacyclin are two compounds which have been implicated as important modulators of local cerebral blood flow. Concentrations of the stable metabolites of these two compounds, thromboxane B2 and 6-keto-PGF1 alpha, were measured in cerebrospinal fluid (CSF) from eight acute ischaemic stroke patients and 14 patients with no evidence of cerebrovascular disease. Concentrations of thromboxane B2 ranged from 0.15 to 4.0 pg/ml and were significantly higher (P = 0.025) in the ischaemic stroke group when compared with the control group (0.1-0.3 pg/ml). Simultaneously acquired concentrations of 6-keto-PGF1 alpha were not elevated in the stroke group when compared to normals. These clinical findings support evidence from animal studies that emphasizes the importance of cerebral prostaglandins in mediating the secondary vascular changes of cerebral infarction. In conclusion there is an aberration in CSF thromboxane B2 concentrations in patients who have had a stroke. This may be an acute or chronic phenomenon.

GC/MS assay of prostaglandins in cerebrospinal fluid from humans and monkeys.

The objective of this project has been to develop a sensitive and specific assay for prostaglandins in human cerebrospinal fluid (CSF) from patients with alcoholism and appropriate controls using gas chromatography/mass spectrometry. This study was initiated because numerous literature reports strongly suggest that a relationship exists between ethanol's central nervous system effects and the central production of prostaglandins. In both human and animal studies, administration of prostaglandin synthesis inhibitors prior to administration of ethanol attenuated central nervous system effects of ethanol. Samples from alcoholics after a three week period of abstinence and normals contained none of the measured prostaglandins (PGE2, PGE1, PGF1a, PGF2a, 6-keto-PGF1a) at a concentration more than twice the limit of quantification (3 pg/mL CSF). Comparison of GC/MS and radioimmunoassay methods provided further validation for these results. Literature reports of much higher levels of prostaglandins in normal controls, i.e., tens to hundreds of pg/mL CSF, appear to be incorrect. Examination of monkey CSF provided a positive control, since several of the prostaglandins were easily quantifiable in these samples.

Subarachnoid blood causes pial arteriolar constriction in newborn pigs.

BACKGROUND AND PURPOSE: The present study was designed to determine in newborn animals the delayed effect of subarachnoid blood on pial arteriolar diameter and eicosanoid concentrations in cortical periarachnoid fluid. METHODS: Forty-eight to 96 hours after subarachnoid blood installation, closed cranial windows were implanted over the cerebral area exposed to blood in anesthetized, artificially ventilated newborn piglets. All pial arterioles greater than 60 microns in diameter were measured, and cortical periarachnoid fluid was collected for the determination of eicosanoids. RESULTS: Subarachnoid blood resulted in a 20% to 30% decrease in the average diameter of pial arterioles exposed to blood for 48 to 96 hours, a decreased number of large pial arterioles (greater than 200 microns), and an increased number of small arterioles (60 to 100 microns). No changes in dilator prostanoids (prostacyclin [as 6-keto-prostaglandin F1 alpha] and prostaglandin E2) were detected. Concentrations of vasoconstrictor prostanoids in cortical cerebrospinal fluid increased. Thromboxane B2 increased to 430 +/- 70 pg/mL, and prostaglandin F2 alpha increased to 1370 +/- 180 pg/mL compared with 250 +/- 20 and 860 +/- 70 pg/mL, respectively, in the control group. The concentration of peptidoleukotrienes increased to 400 to 600 pg/mL 72 to 96 hours after blood installation, while the level in the control group was less than 80 pg/mL. CONCLUSIONS: The altered balance between vasodilator and vasoconstrictor eicosanoids could contribute to cerebral vasoconstriction after subarachnoid blood installation in newborn pigs.

Prevention of chronic cerebral vasospasm in dogs with ibuprofen and high-dose methylprednisolone.

Severe chronic cerebral vasospasm was produced in dog basilar arteries by two injections, 2 days apart, of autologous blood into the cisterna magna of 25 dogs. Treatment with ibuprofen (n = 8) or high-dose methylprednisolone (n = 8) after the first injection of blood prevented or reduced angiographic vasospasm. Cerebrospinal fluid concentrations of prostaglandin E2, prostaglandin F2 alpha, 6-ketoprostaglandin F1 alpha (a metabolite of prostacyclin), and thromboxane B2 (a metabolite of thromboxane A2) were measured in both treated and untreated (n = 7) dogs. In untreated dogs, the level of prostaglandin E2 increased 94-fold by Day 8 after the first injection of blood and was strongly and positively correlated with the degree of angiographic vasospasm. Treatment with ibuprofen and high-dose methylprednisolone prevented or significantly reduced this increase in prostaglandin E2 concentration. Smaller increases in cerebrospinal fluid concentrations of thromboxane B2 and 6-ketoprostaglandin F1 alpha occurred after experimental subarachnoid hemorrhage; the magnitude of these increases was also reduced by ibuprofen or high-dose methylprednisolone treatment. In contrast, prostaglandin F2 alpha levels were not significantly altered during the study. These data show that enhanced prostaglandin E2 synthesis occurs during experimental subarachnoid hemorrhage, and the by-products generated in its synthesis may play a role in the pathogenesis of cerebral vasospasm.