Bacteriochlorophyll aggregates in positively charged micelles.

Micellar complexes were prepared from bacteriochlorophyll a and bacteriopheophytin a with the cationic detergents, cetyltrimethyl ammonium bromide and cetylpyridinium chloride. These complexes have spectroscopic properties (absorption, circular dichroism) which are very different from the ones formed with non-ionic detergents like Triton X-100, and also with anionic detergents. Bacteriochlorophyll a forms two complexes: One is blue-shifted and has excitonically coupled Qy transitions. The second one is extremely red-shifted. The unusual properties are suggested to result from interactions of the positively charged head-group of the detergent with the tetrapyrrole.

Role of the liver in the disposition of intravenous nitroglycerin in the rat.

Previous studies have indicated that the liver is the main site of nitroglycerin (NTG) elimination when the drug is systematically infused. To examine this hypothesis, we measured the apparent systemic clearance (Cls) of nitroglycerin in anesthesized rats receiving a constant intravenous infusion at a dose of 100 micrograms per kg per min. Animals were divided into shunt and sham groups; the former had undergone a portal vein ligation 10 days prior to the study, while the latter was subjected to a sham operation. On the study day, half of the animals of each group also received probenecid at 200 mg/kg, i.v., a drug previously reported to inhibit organic nitrate ester reductase (ONER) activity in rat liver. Arterial NTG samples were obtained at 41, 43 and 45 min of infusion in all four experimental groups; Cls was 439 +/- 32 ml per kg per min (mean +/- S.E.) in sham, 460 +/- 44 in sham and probenecid, 477 +/- 39 in shunt, and 461 +/- 34 in shunt and probenecid animals. During NTG infusion, hepatic blood flow (measured with a constant infusion of indocyanine green) was decreased markedly in shunted rats as was liver/body weight, indicating hepatic atrophy. The specific activity of hepatic ONER was similar in all four groups. In spite of marked differences in hepatic blood flow and hepatic mass, the Cls was similar in all four groups. The liver does not appear to be a major site for the elimination of systemic nitroglycerin as hitherto assumed.

Long-wavelength-absorbing forms of bacteriochlorophyll a in solutions of Triton X-100.

At least three forms of Triton X-100-solubilized bacteriochlorophyll a (BChl a) have been characterized by UV/visible/near-IR absorption and CD spectra. One, absorbing at 770 nm, is similar to a monomeric solution in methanol. The two others have strongly red-shifted absorption peaks (860 nm and 930, 835 nm) and intense and complex CD bands in this region, indicative of strong interaction of at least two and three molecules of BChl a, respectively.

Isosorbide dinitrate in experimental portal hypertension: a study of factors that modulate the hemodynamic response.

Isosorbide dinitrate, a long-acting vasodilator, has been tested in human portal hypertension with conflicting results. In order to determine some of the factors that could affect the individual response to this drug, we infused isosorbide dinitrate at a low dose (10 to 25 micrograms per kg per min) and a high dose (100 micrograms per kg per min) to rats with portal vein stenosis. Under pentobarbital anesthesia, portal pressure was measured with an ileocolic vein catheter while cardiac output and regional blood flows were measured with the microsphere technique. At a dose that decreased arterial pressure by approximately 10%, cardiac output remained unchanged while portal vein inflow decreased significantly; portal pressure was not reduced (10.7 +/- 0.2 vs. 10.0 +/- 0.3 mm Hg), indicating a rise in portal vascular resistance. At a high dose of isosorbide dinitrate, arterial pressure and cardiac output fell markedly; portal pressure decreased only modestly (11.3 +/- 0.3 vs. 9.8 +/- 0.6 mm Hg, p less than 0.05), but portal flow was unchanged, indicating a reduction in portal vascular resistance. In addition, portal hypertensive rats received a constant i.v. infusion of N-acetyl-cysteine; the combination of the latter and isosorbide dinitrate markedly potentiated the effects on arterial pressure. Thus, the dose of the drug and the presence of cysteine-containing compounds appear to modulate the hemodynamic response to isosorbide dinitrate. Clinical testing with this drug should be undertaken with consideration of these factors.

Hemodynamic and renal effects of atrial natriuretic factor in portal hypertensive rats. Potentiation by Phe-Ile-Orn-vasopressin.

The effects of atrial natriuretic factor (ANF) on splanchnic hemodynamics and renal function in portal hypertensive models are described incompletely. Furthermore, ANF-induced vasodilatation and hypotension may limit the assessment of its own renal physiological effects. We infused ANF (human ANF 102-126) to anesthetized portal vein-ligated rats, a model with prehepatic portal hypertension. Arterial pressure was reduced by 17%, but portal pressure was unaffected. Diuresis and natriuresis were explained in part by an increase in glomerular filtration rate; in addition, renal vascular resistance was significantly decreased. The natriuretic response to ANF was slightly, but significantly, decreased in portal hypertensive rats as compared to controls (fractional excretion of sodium, 1.8 +/- 0.4 vs. 2.9 +/- 0.3; P less than .05). The addition of Phe-Ile-Orn-vasopressin, a V1 receptor agonist, normalized arterial pressure but induced a significant decrease in portal pressure (15 +/- 0.9 mm Hg base line vs. 12.8 +/- 0.7 combination group; P less than .01). Furthermore, the combination of both drugs markedly potentiated the natriuretic effects (0.4 +/- 0.1 microEq/min of control vs. 10.0 +/- 2.3 ANF vs. 32.2 +/- 3.3 combination group; P less than .001). The natriuretic potentiation resulted from increments in glomerular filtration rate and renal blood flow. Normalization of arterial pressure may enhance the renal physiological effects of ANF, in this portal hypertensive model.

Stenosis of a portacaval anastomosis affects circadian locomotor activity in the rat: a multivariable analysis.

The study of hepatic encephalopathy is limited by the lack of standardized experimental models to assess behavior. We have shown that rats continuously monitored while running on a wheel show abnormalities of the circadian rhythm of locomotor activity after portacaval anastomosis (PCA), such that entrainment of running activity to the light-dark cycle is severely impaired. To identify factors that affect postoperative circadian behavior, we have performed a multivariable analysis of 69 sham-operated controls and 107 rats after PCA. Our results indicate that shunt stenosis, as determined by the pressure gradient from the splenic pulp to the inferior vena cava, ameliorated the postoperative deterioration of the circadian rhythm. In addition, postoperative behavior was affected by preoperative performance, diet, and gender. Postoperative body weight gain, spleen weight, and liver atrophy did not impact this model. Because shunt stenosis is known to ameliorate hepatic encephalopathy in humans, our findings support the validity of this behavioral end point as a correlate of hepatic encephalopathy. Measurement of the pressure gradient across the anastomosis and achievement of sufficient preoperative entrainment appear critical for the standardization of the model.

Portal-systemic shunting and the hemodynamic effects of nitroglycerin in the rat.

A hemodynamic response to oral nitroglycerin has been used recently to estimate the extent of portal systemic shunts in humans. We studied the hemodynamic effects of this drug in the rat with partial portal vein ligation, a model in which portal hypertension develops and the extent of portal-systemic collaterals can be measured. Two to ten days after the initial procedure, shunted and sham-operated controls received a continuous infusion of nitroglycerin into a jugular vein, the ileocolic vein, or intrajejunally. The response of systolic arterial pressure to intravenous nitroglycerin was similar in shunted and sham animals. The cumulative dose-response curve showed an earlier and more pronounced response in shunted rats receiving nitroglycerin intraportally, suggesting a marked increase in bioavailability. When the arterial response to nitroglycerin, infused intraportally at 1 microgram/kg X min, was compared to the extent of ileocolic shunting, measured with radioactive microspheres, a close correlation (r = 0.84) was seen; shunting values ranged from 0% (sham) to 97%. Additionally, in shunted rats a more pronounced response to intraportal, compared to intrajejunal, infusion suggested a contribution of the intestine to the first-pass elimination of the drug. The bioavailability of intraportal and intrajejunal nitroglycerin is influenced by the extent of portal systemic collaterals. A significant reduction in portal vein pressure in response to nitroglycerin seen in this animal model supports its testing as a therapeutic agent in portal hypertension.

Function of the arachidonate-deficient spontaneously hypertensive rat kidney.

In order to assess the effect of renal prostaglandins on the glomerular filtration rate (GFR) and electrolyte excretion in the spontaneously hypertensive rat (SHR), we perfused isolated SHR and normotensive Wistar-Kyoto (WKY) kidneys after pretreatment with either a control diet or a diet deficient in arachidonate, the precursor of prostaglandins. When perfusion pressures were increased from 100 to 160 mm Hg, renal vascular resistances (RVR) increased by 32-41%. RVR of SHR kidneys always exceeded that of WKY kidneys by a nearly constant amount, and arachidonate deficiency had little effect on this relationship. In contrast to RVR, the GFR increased severalfold. GFR and urine flow were greater in WKY than in SHR kidneys, a relationship unaffected by arachidonate deficiency. Changes in sodium and chloride excretion occurred in parallel with GFR. Although arachidonate-deficient SHR and WKY kidneys manifested potassium wastage compared to controls, arachidonate deficiency did not result in altered sodium or chloride excretion. The results suggest that direct actions of arachidonate or renal prostaglandins are not responsible for most functional differences between SHR and WKY kidneys.

Brain edema and intracranial hypertension in rats after total hepatectomy.

BACKGROUND/AIMS: Glutamine, generated from ammonia in astrocytes, may account for brain edema in acute liver failure. Recent studies showing decreased intracranial pressure after hepatectomy in humans suggest that factors released by the necrotic liver could play a pathogenic role in brain swelling. The aim of this study was to examine whether brain edema and intracranial hypertension develop in hepatectomized rats. METHODS: Rats underwent a portacaval anastomosis or a sham operation. At 24 hours, animals underwent a second sham operation or a total hepatectomy. Intracranial pressure was continuously monitored, and cortical water and glutamine contents were measured after the rats were killed. In a second experiment, hepatectomized and devascularized (portacaval anastomosis plus hepatic artery ligation) rats were killed every 2 hours and at the time of intracranial hypertension. RESULTS: Although brain edema developed in both groups with liver failure, devascularization resulted in a higher brain water content in spite of an equivalent increase in glutamine concentration. Intracranial pressure increased to a similar degree in both groups, but all parameters increased earlier in anhepatic rats. CONCLUSIONS: Hepatectomized rats develop brain edema and intracranial hypertension. The temporal sequence in this model supports the role of glutamine as an organic osmolyte. In addition, other factors (e.g., brain volume) may contribute to intracranial hypertension in hepatectomized rats.

Results of endoprosthetic hip joint replacement with the aluminum ceramic-metal composite prosthesis "Lindenhof".

The first clinical results of the Lindenhof ceramic-metal composite prosthesis implanted in our hospital in Freiburg are presented. We observed that same favorable early results as the conventional prostheses in a correct position. The implants are incorporated into the bone within 8-12 weeks. The radiographic films show the adaptation of the supporting bone around the ceramic socket. We explain the failures due to our initial lack of technical experience and/or anatomical deformation of the pelvic bone. complications caused by the post-operative treatment during the 12 weeks following surgery did not occur. The combination of a cemented metal femoral component with a ceramic head seems to be a reasonable compromise to use the favorable physical and biochemical properties of the bioceramic material as long as there is no satisfactory solution for a stable cementless fixation of the femoral stem in to the bone. The advantages of the Lindenhof prosthesis predominate the disadvantages: expensive instruments and a post-operative treatment of several months.

Pressure-dependent action of angiotensin II on glomerular filtration in the isolated rat kidney.

We examined the action of angiotensin II (AII) on isolated rat kidney perfused with a recirculating cell-free solution at either 12 pKa (90 mmHg) or 17 kPa (128 mmHg). The renal perfusion pressure was maintained constant while sufficient AII was added to increase the renal vascular resistance by 50%. In the low-pressure kidneys. AII increased the glomerular filtration rate (GFR) by 155%, increased sodium reabsorption by 157%, and decreased the urine sodium concentration by 34% without affecting sodium excretion. In the high-pressure kidneys, GFR initially was significantly greater but was not affected by AII. In these experiments, AII had no effect on sodium reabsorption or excretion and decreased the urine sodium concentration by 7%. The data suggest that AII could be involved in autoregulation of the GFR without producing large changes in sodium excretion.

Disruption of circadian locomotor activity in rats after portacaval anastomosis is not gender dependent.

A recent study suggested that female rats are less affected by a portacaval anastomosis (PCA) than their male counterparts, as measured by body weight and changes in locomotor activity. In this study, we evaluated the entrainment of locomotor activity to the light/dark (LD) cycle, a consistent abnormality in the portacaval shunted rat. The degree of entrainment was measured in male and female rats before and after PCA or sham operation. All four groups of animals showed strong entrainment to an LD cycle before surgery. After portacaval anastomosis, male and female rats exhibited a highly significant decrease in overall motor activity as compared with the preoperative period and as compared with sham-operated animals of the same gender. The percentage of total activity during daytime was significantly increased after portacaval anastomosis. The reduction in parameters of entrainment indicates a disruption of circadian function in both portacaval-shunted groups. Portal pressure measurements confirmed the patency of the shunts. Cortical brain glutamine levels were similarly increased in male and female shunted rats. The loss of body weight was slightly, but not significantly, more pronounced in male animals after shunt surgery. In conclusion, our results do not support a role for gender in the disruption of circadian function in rats after PCA.

Intracranial pressure waves and intracranial hypertension in rats with ischemic fulminant hepatic failure.

Brain edema and intracranial hypertension are a major cause of death in fulminant hepatic failure. We have shown that brain water measured in rats after hepatic devascularization (portacaval anastomosis followed in 24 to 48 hr by ligation of the hepatic artery) increases with the progression of encephalopathy. In this study, we examined whether intracranial hypertension develops in this model of fulminant hepatic failure. Using a fiberoptic pressure transducer, intracranial pressure rose from 3.3 +/- 1.1 mm Hg to 23.7 +/- 2.7 mm Hg (mean +/- S.E.M.) by the time the corneal reflex was lost; intracranial pressure was unchanged in control rats. Immediately after ligation of the hepatic artery, intracranial pressure was normal and remained stable until the last hours of the experiment, when it progressively rose, suggesting a loss of intracranial compliance. In addition, sudden and short episodes of marked increases in intracranial pressure (greater than 50 mm Hg) not related to seizure activity markedly decreased cerebral perfusion pressure. Internal carotid artery blood flow, an indirect measure of cerebral perfusion, decreased 29% +/- 12% by the end of the experiment. The time elapsed from ligation of the hepatic artery until loss of the corneal reflex (range 340 to 940 min) was related to the change in cerebral perfusion pressure, suggesting that an increase in systemic arterial pressure at the time of the initial rise in intracranial pressure may result in an increased length of survival. In this animal model, widely used to study the pathogenesis of hepatic encephalopathy, intracranial hypertension invariably appears in the terminal phase of the course. The development of intracranial pressure waves may be an indication that brain herniation is imminent.

Chronic hyponatremia exacerbates ammonia-induced brain edema in rats after portacaval anastomosis.

BACKGROUND/AIM: Abnormalities in brain organic osmolytes are associated with hepatic encephalopathy and with chronic hyponatremia. In spite of the high frequency of hyponatremia in acute and chronic hepatic failure, its role in the development of neurological complications in liver disease is poorly understood. We aimed to study the effect of prior hyponatremia on the development of ammonia-induced brain edema in rats after portacaval anastomosis. In this model, brain swelling is mediated in part through an increase in brain glutamine, an organic osmolyte. METHODS: Hyponatremia was induced in rats with 1-desamino-8-D-arginine vasopressin (DDAVP) administered through an osmotic minipump for 1 week. This was followed by performance of a portacaval anastomosis and ammonia infusion. At the end of the infusion, brain water (density gradient) and key brain organic osmolytes (HPLC) were measured. RESULTS: Rats with hyponatremia showed a decrease in all three brain organic osmolytes measured: glutamine, myo-inositol and taurine. Hyperammonemia resulted in the expected rise in glutamine, with a reduction of myo-inositol and taurine. In the combined group (hyponatremia plus hyperammonemia), the rise in brain glutamine induced by ammonia infusion was attenuated (10.6+/-0.9 mM/kg vs. 15.5+/-0.8 mM/kg hyperammonemia alone; p<0.05). In spite of this limited rise in brain glutamine, ammonia infusion to hyponatremic rats exacerbated brain swelling (82.3+/-0.3% vs. 80.6+/-0.1%; p<0.05). CONCLUSIONS: Hyponatremia worsens brain swelling in a model of ammonia-induced brain edema. The decrease in the concentration of brain organic osmolytes induced by hyponatremia does not protect the brain from the development of ammonia-induced brain edema.

Function of the isolated spontaneously hypertensive rat kidney after blood pressure reduction.

We treated 20-week-old spontaneously hypertensive rats (SHR) with either placebo or hydralazine, reserpine and hydrochlorothiazide for 1 month. Mean arterial pressure in treated SHR averaged 113 +/- 7 mm Hg (mean +/- SE), compared to 162 +/- 12 mm Hg in animals receiving placebo (p less than 0.01). Glomerular filtration rate (GFR) and sodium excretion were similar in both groups. In isolated perfused kidneys, the GFR and sodium excretion were significantly greater in the treatment group than in the placebo group at a perfusion pressure of 140 mm Hg (p less than 0.01). Renal vascular resistance (RVR) of kidney from treated SHR was no different from RVR of kidney from placebo SHR. Hydralazine (6 mM) and diazoxide (4 mM) increased the GFR and sodium excretion of isolated SHR kidney perfused at 140 mm Hg (p less than 0.05), but decreased RVR significantly (p less than 0.05). We conclude that prolonged antihypertensive treatment renders higher GFR values to isolated SHR kidneys perfused at 140 mm Hg, with sodium excretion varying in proportion to the GFR. The addition of vasodilators to the perfusate of isolated SHR kidneys partially reproduced these changes, but only at extremely high concentrations unlike to be attained in vivo.

Glutamine, myo-inositol, and organic brain osmolytes after portocaval anastomosis in the rat: implications for ammonia-induced brain edema.

Brain myo-inositol, an organic osmolyte, is decreased in cirrhotic patients with hepatic encephalopathy but appears unchanged in fulminant hepatic failure. An osmoregulatory response to the increase in brain glutamine may explain the decrease in brain myo-inositol; if this is the case, organic osmolytes may account for differences in the development of brain edema seen in acute or chronic liver failure. The response of myo-inositol and nine other organic osmolytes to the increase in brain glutamine at different time intervals after portacaval anastomosis (PCA) in the rat was studied. Organic osmolytes were measured in brain tissue and cerebrospinal fluid. Water in cerebral cortex was measured after ammonia infusion with the gravimetric method. Six weeks after PCA, despite an increase in brain glutamine (PCA, 16.4 +/- 2 mmol.kg wt-1.kg wt-1; sham, 5 +/- 1 mmol.L-1.kg wt-1), the content of total organic osmolytes did not increase (PCA, 44.1 +/- 3; sham, 43 +/- 4) because of a decrease of other osmolytes (myo-inositol, 54%; urea, 39%; taurine, 33%; and glutamate, 8%). Brain myo-inositol was lower at 3 weeks (3.4 +/- 0.5 kg wt-1) than at 1 day after PCA (4.7 +/- 0.5 kg wt-1). An ammonia infusion resulted in brain edema at both time points. In conclusion, the reduction in brain myo-inositol in PCA rats is accompanied by the decrease of other organic osmolytes, supporting the view that changes in myo-inositol reflect an osmoregulatory response. The decrease in brain myo-inositol is more marked as time elapses after PCA. In a model in which short-term and large doses of ammonia were infused, the decrease in brain myo-inositol did not prevent the development of brain swelling. Understanding brain osmoregulatory mechanisms may provide new insights into hepatic encephalopathy and brain edema in fulminant hepatic failure.

Cerebral hyperemia and nitric oxide synthase in rats with ammonia-induced brain edema.

BACKGROUND/AIM: Brain edema is a common fatal complication in acute liver failure. It is related to an acute change in brain osmolarity secondary to the glial accumulation of glutamine. Since high cerebral blood flow (CBF) precedes cerebral herniation in fulminant hepatic failure we first determined if an increase in brain water and glutamine are prerequisite to a rise in CBF in a model of ammonia-induced brain edema. Secondly, we determined if such a cerebral hyperperfusion is mediated by nitric oxide synthase (NOS). METHODS: Male rats received an end-to-side portacaval anastomosis (PCA). At 24 h, they were anesthetized with ketamine and infused with ammonium acetate (55 microM/kg per min). Studies were performed at 60, 90, 120, 150 and 180 min after starting the ammonia infusion and once the intracranial pressure had risen three-fold (mean 210'). Brain water (BW) was measured using the gravimetry method and CBF with the radioactive microsphere technique. Glutamine (GLN) in the CSF was sampled via a cisterna magna catheter. The neuronal NOS was specifically inhibited by 1-2-trifluoromethylphenyl imidazole (TRIM, 50 mg/kg intraperitoneally) and in separate studies nonspecifically by N-omega-nitro-L-arginine (L-NNA, 2 microg/kg per min intravenously) RESULTS: At 90', brain water was significantly increased (P < 0.015) as compared to the 60' group while CBF was significantly different at 150'. A significant correlation was observed between values of CBF and brain water (r = 0.88, n = 36, P < 0.001). Administration of either TRIM or L-NNA did not prevent the development of cerebral hyperperfu. sion and edema. CONCLUSION: We observed that cerebral hyperemia follows an initial rise in brain water content, rather than in the cerebrospinal fluid concentration of glutamine. The rise in CBF further correlated with brain water accumulation and was of critical importance for the development of intracranial hypertension. The unique mechanism for the rise in CBF in hyperammonemia was not prevented by NOS inhibition indicating that NO is not the mediator of high CBF and intracranial hypertension.

Cerebral blood flow and the hyperdynamic circulation of rats after portacaval anastomosis.

Both increased and decreased values of cerebral blood flow have been reported in liver disease. Furthermore, the relation between the cerebral circulation and the generalized hemodynamic disturbance seen in chronic liver disease with portal-systemic shunting has not been fully characterized. We studied this problem in a well defined model of the hyperdynamic circulation, the rat after portacaval anastomosis (PCA). Using the radioactive microsphere technique, cardiac output and regional blood flows were measured; regional vascular resistances were then calculated. While the fraction of cardiac output perfusing the splanchnic bed was significantly increased, the corresponding brain fraction was reduced. Blood flow to the cerebral hemispheres and midbrain was significantly decreased. Arterial vasodilatation was demonstrated by the fall in arterial pressure, systemic vascular resistance as well as splanchnic and renal resistances; cerebrovascular resistance, however, was unchanged. No relation between values of arterial pressure and cerebral blood flow was seen, making a failure of cerebrovascular autoregulation unlikely. The decrease in hemispheric and midbrain perfusion without changes in vascular resistance suggests that a drop in blood flow is appropriately coupled to a reduction in brain metabolism. The cerebral circulation does not participate in the hyperdynamic state that is seen in this model.

Reduction of cerebral perfusion precedes rise of intracranial pressure in rats with ischemic fulminant liver failure.

In fulminant liver failure, brain edema may progress to intracranial hypertension. However, the rise in intracranial pressure is a late event in this sequence. We investigated the relationship between cerebral perfusion and development of intracranial hypertension in a well-characterized model of fulminant liver failure, the rat subjected to hepatic devascularization (n = 11). In addition, we examined the effects of hyperglycemia on the development of brain edema because high blood glucose level can exacerbate other forms of brain edema, as seen in stroke. Intracranial pressure was continuously monitored with a cisterna magna catheter; relative changes in blood flow were continuously assessed with a Doppler flow probe on the internal carotid artery. Cerebral perfusion decreased by 62%, with the greatest reduction before the onset of increased intracranial pressure. Intracranial pressure did not change until 2 hr before death, at which time it increased exponentially. Brain water in fulminant liver failure rats was significantly increased compared with that in controls. Hyperglycemia (200 to 220 mg/dl) had no effect on time elapsed until loss of corneal reflex, percentage of brain water, maximal intracranial pressure or pattern of change in cerebral perfusion compared with euglycemia (80 to 100 mg/dl). Sham-operated animals showed no changes in measured parameters. We conclude that a linear reduction in cerebral perfusion precedes the rise of intracranial pressure in this model, a decrease that may reflect changes in brain metabolic activity at the time that brain edema develops. Carotid blood flow monitoring may be a useful noninvasive tool for the detection of cerebral events in fulminant liver failure.

Indomethacin prevents the development of experimental ammonia-induced brain edema in rats after portacaval anastomosis.

Patients with fulminant hepatic failure (FHF) die with brain edema, exhibiting an increased cerebral blood flow (CBF) at the time of cerebral swelling. Mild hypothermia prevents brain edema in experimental models and in humans with FHF, an effect associated with normalization of CBF. To study the effects of alterations of CBF on the development of brain edema, we administered intravenous (IV) indomethacin to rats receiving an ammonia infusion after portacaval anastomosis. This model predictably develops brain edema and a marked increase in CBF at 3 hours of infusion. Brain water was measured with the gravimetry technique; CBF was monitored with both laser Doppler flowmetry and radioactive microspheres, whereas intracranial pressure (ICP) was monitored with a cisterna magna catheter. Coadministration of indomethacin prevented the increase in CBF seen with ammonia alone (110 +/- 19% vs. -2 +/- 9%) as well as the increase in brain water (80.86 +/- 0.12% vs. 80.18 +/- 0.06%) and the increase in ICP. Plasma ammonia and brain glutamine levels were markedly elevated in the ammonia-infused group and unaffected by indomethacin. However, ammonia uptake by the brain was significantly reduced by indomethacin. Levels of 6-keto-PGF(1alpha), a stable metabolite of prostacyclin, were reduced in the cerebrospinal fluid (CSF) of indomethacin-treated animals. As with mild hypothermia, avoiding cerebral vasodilatation with indomethacin will prevent the development of brain edema in this hyperammonemic model. Cerebral vasoconstriction reduces cerebral ammonia uptake and, if selective to the brain, may be of benefit in FHF.