Vasopressin accelerates experimental ammonia-induced brain edema in rats after portacaval anastomosis

Journal of Hepatology 39 (2003) 193–199 www.elsevier.com/locate/jhep

Vasopressin accelerates experimental ammonia-induced brain edema in rats after portacaval anastomosis Chuhan Chung, Javier Vaquero, Jeanne Gottstein, Andres T. Blei* Hepatology Section and Department of Medicine, VA Chicago Health Care System, Lakeside Division and Northwestern University, 400 E. Ontario St., Chicago, IL 60611, USA

Background/Aims: Cerebral hyperemia is an important contributor to the development of brain edema in fulminant hepatic failure. Rats receiving an ammonia infusion after portacaval anastomosis (PCA) demonstrate a rise in cerebral blood flow (CBF) with brain edema at 180 min. Vasopressin (VP), a systemic vasoconstrictor which in the rat dilates cerebral vessels through V2 receptors, was used to ascertain the effects of increasing CBF. Methods: Changes in CBF were measured with Laser Doppler flowmetry (LDF). Absolute CBF was measured with radioactive microspheres to calculate oxygen and ammonia uptake. Results: Compared to the NH3 1 Vehicle group, VP 1 NH3 infusion accelerated the rise in CBF (117 6 21 vs. 2 6 6 12%, P < 0.01), and the development of brain edema (81.09 6 0.17 vs. 80.29 6 0.06%, P < 0.01). Radioactive microspheres confirmed these results (254 6 44 vs. 106 6 9.5 ml/min/100 g, P < 0.01). Oxygen uptake was similar. Ammonia uptake was more than twofold higher in the VP 1 NH3 group. A V1 antagonist negated the higher mean arterial pressure (MAP) that occurs with VP but cerebral hyperemia still occurred. A V2 antagonist resulted in similar systemic pressures, CBF and brain water compared to the VP 1 NH3 group. Conclusions: In this model, an increase in CBF with VP hastens the development of brain edema while increasing ammonia delivery to the brain. q 2003 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved. Keywords: Cerebral blood flow; Portacaval anastomosis; Vasopressin

1. Introduction Recent observations in fulminant hepatic failure (FHF) have noted an increased cerebral blood flow (CBF) in patients with brain edema and intracranial hypertension. In particular, hypothermia [1], hyperventilation [2], and pharmacological vasoconstrictors [3], measures that reduce CBF, have been shown to ameliorate brain swelling and elevated intracranial pressure (ICP). The clinical benefits of controlling CBF have provided insight into the pathogenesis of brain edema in FHF. Current views on the pathogenesis of brain edema in FHF emphasize a ‘two-hit’ phenomenon with an osmolar alteration in the brain with a subsequent alteration in cerebrovascular regulation [4]. Although controversy remains over the precise role of CBF in the Received 2 November 2002; received in revised form 18 March 2003; accepted 1 April 2003 * Corresponding author. Tel.: þ 1-312-503-3453; fax: þ1-312-908-6192. E-mail address: [email protected] (A.T. Blei).

development of brain edema, a general consensus exists that higher values of CBF are associated with worse clinical outcomes [5,6]. Larsen, moreover, has proposed the notion of ‘luxury perfusion’ in FHF with even normal values of CBF exceeding the metabolic needs of the encephalopathic brain [6]. CBF has also been investigated in an experimental model of selective cerebral hyperemia and brain edema: the rat after portacaval anastomosis (PCA) receiving an ammonia infusion [7]. In the setting of stable systemic hemodynamics, a predictable rise in CBF with resulting elevation in ICP and brain edema within 3 h of ammonia infusion occurs. Cerebral hyperemia is necessary for the appearance of brain swelling in this model. Diverse modalities to control the rise in CBF, such as methionine-sulfoximine [7] and mild hypothermia [8], prevent ammonia-induced brain swelling. Indomethacin, a cerebral vasoconstrictor, blunts the rise in CBF thereby preventing brain edema and elevated ICP compared to PCA animals receiving ammonia alone [9]. A disproportionate reduction of ammonia uptake by the

0168-8278/03/$30.00 q 2003 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-8278(03)00185-5

194

C. Chung et al. / Journal of Hepatology 39 (2003) 193–199

brain accompanies the reduction of CBF and represents another potential mechanism by which manipulation of blood flow may alter brain edema [9]. In the current study, we examine the effects of vasopressin on the development of brain edema in the PCA model. Vasopressin, a potent and widely used systemic vasoconstrictor, has been shown experimentally in the rat to affect CBF by dilating cerebral arteries [10,11]. Its systemic pressor effect occurs through peripheral V1 receptors, while it has been postulated that it preferentially dilates cerebral blood vessels through V2 receptors [12]. This effect may explain results from experimental [13] and clinical studies [14] that show vasopressin to be superior to epinephrine with respect to neurological outcomes after cardiac arrest. We hypothesized that vasopressin administration would potentiate the ammonia-induced increase in CBF in the PCA model and hasten the development of brain edema and elevated ICP. A role for potentially enhanced uptake of ammonia by the brain with this increase in CBF was also explored.

2. Materials and methods 2.1. Performance of PCA This study was performed in male Sprague–Dawley rats (Charles River, Wilmington, DE) weighing 300–400 g. The animals were housed in cages, fed a regular chow diet with free access to water, and kept on a 12:12 light/dark cycle. After 1 week, they were anesthetized with isoflurane and an end-to-side PCA was constructed using a continuous suture technique. The anastomosis was completed and the animals were returned to individual cages where feeding and water ingestion promptly resumed.

2.2. Experimental preparation After 24 h, animals received pentobarbital anesthesia intraperitonealy (50 mg/kg). Catheters (PE-50, Intramedic; Parsippany, NJ) were placed in the femoral artery and both femoral veins for intravenous infusions. Three burr holes, 2 –3 mm in diameter were drilled through the skull. One hole was used for placement of a PE-10 catheter into the cisterna magna, and one for identification of the sagittal sinus. A third hole over the parietal cortex allowed for placement of a laser Doppler probe (Probe 407; Perimed; Stockholm, Sweden). Laser Doppler flowmetry (LDF) was performed using a Periflux Laser Doppler System 5000 monitor (Perimed; Stockholm) allowing for continuous measurement of relative changes in CBF. A tracheostomy was performed and the rat was artificially ventilated (Harvard, model 683; South Natick, MA). The arterial and intracisternal catheters were connected to a pressure transducer (Ohmeda; Oxnard, CA) and values of mean arterial pressure (MAP) and ICP were continuously inscribed in a computer (Power Lab Software; Castle Hill, Australia). Body temperature was monitored with an intraabdominal thermistor and maintained at approximately 378C with the aid of a heating blanket. Recordings of LDF were continuously monitored and stored in a computer using specific software (Perisoft, Perimed; Stockholm, Sweden) and analyzed for relative percentage changes.

2.3. Experimental procedure Once stable pressure and flow tracings were noted, ammonium acetate (55 mmol/kg/min) or a control infusion of saline was infused intravenously at a rate of 0.038 ml/min. Ten minutes after ammonia infusion, half of the animals received a continuous infusion of Arg8-Vasopressin (CalBiochem, LaJolla, CA) at a rate of 10 ng/kg/min. The remaining half received a continuous infusion of saline. After 105 min, the animals receiving

ammonia and vasopressin infusion exhibited a significant increase in CBF and ICP, a point where the study was terminated. Before sacrifice, arterial blood was drawn (0.5 ml) to determine arterial blood gases and oxygen content (Instrumentation Laboratories; Lexington, MA) and plasma was stored at 2708C for later measurement of plasma ammonia via an enzymatic assay. A similar volume was obtained from the sagittal sinus for measurement of blood gases, oxygen, and ammonia content. Once the experiments were completed, the animals were decapitated, and the brain removed. One cerebral hemisphere was kept at 48C and processed within 30 min. The hemisphere was cut into coronal slices and 10 mg punch biopsies were obtained from the gray matter of the cortex. Brain water content of each specimen was measured gravimetrically using a density gradient of bromobenzene–kerosene pre-calibrated with K2SO4, as previously described [15]. The equilibration point of the samples was read at 1 min and averaged from three samples from the same coronal slice; conversion from specific gravity to brain water was calculated and results expressed as percentage of water content. In a subsequent experiment, CBF, cardiac output, right and left renal blood flow (RBF) were calculated using the microsphere technique at the time of intracranial hypertension and at matched times in the other three groups. Prior to sacrifice, 15-mm microspheres labeled with 85Sr (3M Nuclear, Boston, MA) were injected into the left ventricle. For the experiment, 85,000 microspheres were ultrasonicated and aspirated into a syringe, vortexed for 1 min and injected over 20 s into the left ventricle. After the animals were sacrificed, the brain and kidneys were obtained. CBF, cardiac output, right and left RBFs were calculated from the radioactivity measured in those tissues by a gamma counter. For all tissues, organ blood flow ¼ 0.786 £ [cpm organ]/[cpm blood]; Cardiac output ¼ 0.786 £ [cpm injected]/[cpm blood]; Cardiac index (CI) ¼ cardiac ouput/body weight. Plasma ammonia levels were measured enzymatically using a commercially available kit (Sigma; St. Louis, MO).

2.4. Experimental design Three sets of experiments were performed.

2.4.1. Experiment A In this experiment, we evaluated the effect of Vasopressin on relative changes of cerebral perfusion and brain water in PCA rats receiving an ammonia infusion using LDF. Eighteen rats were divided into four groups: Control þ Veh ðn ¼ 3Þ; Ammonia (NH3) þ Veh ðn ¼ 6Þ; Control þ VP ðn ¼ 3Þ; NH3 þ VP ðn ¼ 6Þ. The animals receiving NH3 þ VP demonstrated an approximate doubling of CBF and ICP at 105 min. This time frame was the reference time for the other groups and all animals were sacrificed at 105 min.

2.4.2. Experiment B To confirm the results from the LDF experiments, we measured CBF using the microsphere technique as described above. RBF, CI, cerebral metabolic rate for oxygen (CMRO2), and cerebral uptake of ammonia were calculated from these measurements. An additional 18 animals were used: Control þ Veh ðn ¼ 3Þ; NH3 þ Veh ðn ¼ 6Þ; Control þ VP ðn ¼ 3Þ; NH3 þ VP ðn ¼ 6Þ. The animals receiving ammonia infusion received a dose of 55 mmol/kg/min. The vasopressin dose used in Experiment A was also used in this set of experiments.

2.4.3. Experiment C To assess the contribution of V1 and V2 receptors on CBF for the vasopressin-infused animals ðn ¼ 3Þ, we co-infused ammonia with a selective V1 antagonist, (b-Mercapto-b, b-cyclopentamethylenepropionyl1, O-Et-Tyr2, Val4, Arg8-Vasopressin) (Sigma V4253), at a bolus dose of 5 nmol/kg IV, followed by an infusion of 5 nmol/kg/h infusion. To assess the contribution of V1 receptors on the cerebral vasculature, we co-infused ammonia with a selective V2 antagonist ðn ¼ 3Þ, (Adamantaneacetyl1, O-Et-D -Tyr2, Val4, Arg8,9-Vasopressin) (Sigma V2381) at a bolus dose of 1.2 nmol/kg IV, followed by an infusion of 1.2 nmol/kg/h infusion. This V2 antagonist represents a potent and selective peptide antagonist for V2 receptors, and in the dose used has been shown to block the renal tubular effects of vasopressin [16].

C. Chung et al. / Journal of Hepatology 39 (2003) 193–199

195

2.5. Statistical analysis One-way analysis of variance (ANOVA) was used to analyze differences between the four groups in Experiment A. Differences between group means were calculated using the Tukey’s test. Student’s unpaired t-test was used to compare differences between two groups. The Animal Care and Use Committee at VA Chicago Health Care System Lakeside Division approved this study.

3. Results 3.1. Experiment A 3.1.1. General aspects Body weights, baseline MAP, and heart rates were similar in all groups. MAP was significantly higher (P , 0:01) in the two groups that received Vasopressin infusion at the time of sacrifice. Arterial blood gases were within normal limits at the time of sacrifice (Table 1). 3.1.2. LDF, brain water, and ICP The NH 3 þ VP group demonstrated a significant 112 ^ 28% increase ðP , 0:01Þ in LDF compared to the other three groups (Fig. 1). The NH3 þ VP group had significantly higher brain water content than the animals given ammonia only, 81.09 ^ 0.17 vs. 80.29 ^ 0.06% ðP , 0:01Þ at 105 min. This greater brain water translated into significantly higher ICP than the other three groups, ðP , 0:05Þ (Fig. 2, Table 2). 3.1.3. Vasopressin raises arterial ammonia levels and LDF As expected, both groups of animals that received ammonia had significantly ðP , 0:01Þ higher levels of plasma ammonia than the two groups that received the vehicle infusion. However, the NH3 þ VP animals had unexpectedly higher arterial ammonia levels compared to the group that received ammonia alone (1307 ^ 144 vs. 1028 ^ 53 mmol/l, P , 0:05). To control for the higher arterial levels of ammonia found in the NH3 þ VP group, we infused another set ðn ¼ 3Þ of animals with a higher dose of ammonium acetate, 62 mmol/kg/min, to obtain comparable arterial ammonia levels. In the group given the higher ammonia infusion, we obtained arterial ammonia levels

Fig. 1. Increase in LDF in the NH3 1 VP group. This increase in flow occurs at 105 min compared to the 180 min typically needed to develop cerebral hyperemia in the NH3 1 Vehicle group (*P < 0.01).

comparable to the NH 3 þ VP group (1349 ^ 47 vs. 1307 ^ 144 mmol/l) but LDF did not increase above baseline levels (Table 3). 3.2. Experiment B 3.2.1. General aspects All baseline variables measured in the two groups including body weight, baseline heart rate, and blood pressure were similar. 3.2.2. CBF, cardiac output, RBF A significant increase in absolute CBF as measured by cerebral microspheres was seen in the NH3 þ VP group compared to the NH3 þ Veh group (254 ^ 44 vs. 106 ^ 9.5 ml/min/100 g, P , 0:01), and were consistent with the LDF results (Table 2, Fig. 3). CI and RBF were similar in all four groups although a trend toward decreased CI and RBF was seen in the groups given vasopressin.

Table 1 General characteristics Group

Veh þ Veh NH3 þ Veh Veh þ VP NH3 þ VP

Weight (g)

316 ^ 17 320 ^ 9 318 ^ 10 329 ^ 6

MAP (mmHg)

Heart rate (beats/min)

Final arterial blood gas

Initial

Finals

Initial

Final

pH

pCO2

pO2

90 ^ 2 90 ^ 2 91 ^ 3 86 ^ 1

84 ^ 2 76 ^ 2 109 ^ 8* 106 ^ 4*

365 ^ 17 394 ^ 7 369 ^ 12 383 ^ 10

370 ^ 10 326 ^ 11 319 ^ 11 326 ^ 16

7.40 ^ 0.01 7.42 ^ 0.01 7.43 ^ 0.02 7.41 ^ 0.01

44 ^ 1 46 ^ 1 42 ^ 2 45 ^ 0.4

92 ^ 5 97 ^ 2 98 ^ 4 100 ^ 5

Values expressed as mean ^ SE; one-way ANOVA with Tukey’s test. *P , 0.05.

196

C. Chung et al. / Journal of Hepatology 39 (2003) 193–199

Fig. 2. The NH3 1 Vehicle group demonstrates significantly higher brain water content than the animals that do not receive ammonia infusion (#P < 0.05). The NH3 1 VP group reveals greater brain water at 105 min than the other three groups (*P < 0.01). This greater brain water translates into significantly higher ICP at 105 min (*P < 0.01).

3.2.3. Oxygen and ammonia uptake by the brain As noted in Experiment A, arterial ammonia levels in the NH3 þ VP group were significantly higher than the NH3 þ Vehicle group. CMRO2 was not significantly different between the NH 3 þ VP and NH 3 þ Vehicle groups (7.5 ^ 2.1 vs. 5.7 ^ 0.7 ml O2/min/100 g). There was a significant difference seen in total ammonia uptake by the brain between the NH3 þ VP vs. NH3 þ Vehicle groups (68.3 ^ 20 vs. 26.1 ^ 6.5 mmol/min/100 g, P , 0:05). The

Fig. 3. CBF as measured by radioactive microsphere technique corroborates the changes found by LDF. A greater than twofold increase in CBF is seen in the NH3 1 VP animals compared to the other three groups (*P < 0.01). CI and RBF were not significantly different among the four groups.

arterial –sagittal sinus ammonia difference was not significantly different between these two groups (276 ^ 61 vs. 256 ^ 68 mmol/l) indicating that the greater ammonia uptake by the NH3 þ VP group primarily reflects the increase in CBF (Table 4). 3.3. Experiment C 3.3.1. V1 and V2 antagonists: effects on MAP, serum NH3, LDF, and brain water (Fig. 4) Adding the selective V1 antagonist, (b-Mercapto-b, b-cyclopentamethylenepropionyl 1 ,O-Et-Tyr 2, Val4,Arg8-Vasopressin) negated the rise in MAP that occurred with the NH 3 þ VP group (81 ^ 3 vs. 106 ^ 4 mmHg, P , 0:05). Serum ammonia levels were comparable to the NH3 þ Vehicle group (1018 ^ 109 vs. 1028 ^ 53 mmol/l). Despite a non-elevated MAP, the V1 antagonist group demonstrated significantly higher LDF than the NH3 þ Vehicle group (86 ^ 24% vs. 2 6 ^ 12, P , 0:05). The elevated brain water content of 80.82 ^ 0.07% was not significantly different from the NH3 þ VP group and was significantly higher ðP , 0:05Þ than the NH3 þ Vehicle group, 80.29 ^ 0.05%. The animals given the V2 antagonist, (Adamantaneacetyl1, O-Et-D -Tyr2, Val4, Arg8,9-Vasopressin) demonstrated a significantly higher increase in MAP than the V1 antagonist group (97 ^ 2 vs. 81 ^ 3 mmHg, P , 0:05). Serum ammonia levels in this group were comparable to the NH3 þ VP group (1307 ^ 249 vs. 1307 ^ 146 mmol/l), reflecting the importance of V1 activation on raising serum ammonia in this model. These animals showed evidence of brain edema and cerebral hyperemia when compared to the

C. Chung et al. / Journal of Hepatology 39 (2003) 193–199

197

Table 2 CBF by radioactive microspheres, corresponding LDF, brain H2O, and final ICP Group

Final CBF (ml/min/100 g)

Laser Doppler flow (%)

Final brain H2O (%)

Final ICP (mmHg)

Veh þ Vehicle Veh þ VP NH3 þ Vehicle NH3 þ VP

55 ^ 4.8 94 ^ 12 106 ^ 10 254 ^ 44*

21.9 ^ 7.6 19 ^ 9.5 26 ^ 12 117 ^ 21*

79.93 ^ 0.04 80.00 ^ 0.06 80.29 ^ 0.06* 81.09 ^ 0.17**

3.4 ^ 0.3 2.8 ^ 0.7 3.7 ^ 0.3 6.9 ^ 0.7*

Values expressed as mean ^ SE; one-way ANOVA with Tukey’s test. *P , 0.01 for CBF; *P , 0.01 for LDF; **P , 0.01 and *P , 0.05 for brain water and *P , 0.05 for ICP.

control group. The change in CBF as measured by LDF (94 ^ 53 vs. 86 ^ 24%) and brain water content (80.54 ^ 0.20 vs. 80.82 ^ 0.07%) were not significantly different between the V2 and V1 antagonist groups.

4. Discussion In this study, we demonstrate that vasopressin administration to an established experimental model of ammoniainduced brain edema leads to an accelerated rise in brain swelling and ICP. The mechanism by which vasopressin exerts this effect appears to be through an earlier increase in cerebral perfusion. This study complements our previous report on the effects of indomethacin, a cerebral vasoconstrictor, whose effects ameliorate the development of brain edema [9]. The mechanisms by which vasopressin results in an increase in CBF are of interest. Our findings with the co-infusion of vasopressin with V1 and V2 antagonists suggest that both receptor systems may be involved. Indeed, a selective rise in CBF occurs in the setting of normal MAP after V1 receptor blockade pointing to a V2-mediated effect on the cerebral vasculature. However, cerebral hyperemia also occurs when vasopressin is combined with V2 receptor blockade, which results in elevated MAP. The increase in CBF in this setting suggests a disruption of cerebral autoregulation with cerebral hyperemia occurring in conjunction with elevated systemic arterial pressures. This finding has been reported in human FHF, with CBF rising in parallel with MAP [17]. Hyperammonemia, a prerequisite for the development of brain edema in this model, became more pronounced in the

presence of vasopressin. The elevation in arterial ammonia after vasopressin infusion in the PCA model has not been reported but we speculate that this effect may reflect reduced blood flow to different organs, including the liver and muscle. The liver after PCA is only supplied by the hepatic artery and vasoconstrictors will reduce hepatic blood flow. In our experimental paradigm, a higher arterial ammonia level by itself does not increase CBF and brain swelling. Other mechanisms for vasopressin-induced cerebral hyperemia and brain edema have been postulated. An increased CBF in response to vasopressin has been linked to an increase in CMRO2 [12]. This effect was thought to be related to an excitatory effect of vasopressin on noradrenergic pathways within the CNS [18]. In our experiments, we found no significant difference in CMRO2 for the vasopressin-infused animals compared to the group that received ammonia alone. Whether the discrepancy between our values and those of earlier studies reflects the inability of the brain exposed to hyperammonemia to increase CMRO2 [19] is unclear. At the cellular level, vasopressin has been thought to facilitate the flux of water and regulate cell volume in the CNS, possibly through specialized water channels [20]. Cultured astrocytes exposed to vasopressin, for instance, remain swollen in the presence of a hypotonic medium compared to a control group [21]. We did not examine expression of aquaporin channels in brain, but vasopressin infusion alone, while causing a modest increase in CBF, does not lead to brain swelling. Brain edema only occurs with exposure to vasopressin and ammonia. Previous experimental studies with vasopressin and its effects on CBF have revealed conflicting data. Faraci et al. demonstrated in cats that vasopressin selectively dilates

Table 3 Vasopressin increases arterial ammonia Group

Final arterial NH3 (mmol/l)

Final percentage change in LDF

Final percentage brain water (%)

NH3 þ Veh (n ¼ 6) NH3 þ VP (n ¼ 6) " NH3 þ Veh (n ¼ 3)

1028 ^ 53 1307 ^ 146* 1349 ^ 47*

26 ^ 12 117 ^ 21* 211.5 ^ 2

80.29 ^ 0.06 81.09 ^ 0.17* 80.38 ^ 0.01

Higher ammonia infusion does not lead to increased LDF. Values expressed as mean ^ SE; one-way ANOVA with Tukey’s test. *P , 0.05 for arterial ammonia; *P , 0.01 for LDF and *P , 0.01 for brain water.

198

C. Chung et al. / Journal of Hepatology 39 (2003) 193–199

Fig. 4. V1 and V2 antagonist effects on NH3 1 VP-infused animals on MAP, percentage change in LDF, and percentage brain water. Addition of a V1 antagonist to the NH3 1 VP group results in a normalized MAP (*P < 0.05) but leads to a percentage change in LDF and percentage brain water that is similar to the NH3 1 VP group. These results suggest a V2 receptor-mediated effect on CBF. The V2 antagonist with NH3 1 VP infusion leads to a MAP that is similar to the NH3 1 VP group. However, these animals also demonstrate comparable increase in LDF and brain water. These results imply that autoregulation of CBF is lost in the PCA animals receiving ammonia infusion.

large cerebral arteries with constriction of pial arteries resulting in unchanged overall CBF [22]. Research in a rat model, however, revealed elevated CBF that is blocked with V2 receptor antagonists [12]. In a canine model, CBF doubled after vasopressin administration, with suppression of vasodilatation by pretreatment with the nitric oxide inhibitor, N G-monomethyl-L -arginine [10]. These different results may reflect species-specific effects of vasopressin. The increased CBF to all regions of the brain in experimental pigs 90 s after vasopressin bolus compared to epinephrine after cardiopulmonary arrest suggests a specific effect of vasopressin on the cerebral circulation [23]. This effect was independent of any differences in MAP between the two groups.

A loss of cerebral vascular autoregulation is an unexpected observation in our model. Occurring within 24 h of performance of the PCA, it is uncertain if such changes are maintained over time. Such autoregulatory loss has been described in the setting of FHF but not in chronic liver disease [24]. In FHF, noradrenaline challenge resulted in increased CBF while restoration of cerebral autoregulation with mild hypothermia [1] or hyperventilation [2] negated this effect of noradrenaline. Our results point to potential deleterious effects of vasopressin in conditions of ‘acute’ hyperammonemia and highlight possible dangers of using this agent to treat the hepatorenal syndrome in FHF. Vasopressin infusion more than doubled ammonia uptake by the brain, which was nearly identical to the values seen for ammonia-infused animals at 180 min in a previous study [9]. In both conditions, a higher absolute CBF is seen suggesting that cerebral hyperemia is a factor that increases brain ammonia uptake. Whether the acute increase in CBF has effects on the permeability of the brain for ammonia is unclear but cerebral ammonia uptake appears increased in patients with chronic liver disease. [25] The finding that patients with FHF who died of cerebral herniation had higher cerebral ammonia uptake than survivors [26] points to the potential benefits of ameliorating CBF and the delivery of ammonia to the brain. An ammonia infusion into the PCA rat provides a predictable model to study the role of blood flow in the pathogenesis of brain edema in FHF. Previously, we espoused the concept of an intracerebral signal mediating the rise in CBF after the generation of glutamine within astrocytes [4]. The current studies extend this concept by showing that an external stimulus can accelerate the process of cerebral hyperemia. The sum of our studies as well as ample clinical experience [1 – 5] underscores the importance of CBF in the pathogenesis of brain edema and the potential for improving or worsening brain edema.

Acknowledgements This work was supported by a Merit Review from the Veterans Administration Research Service and the Stephen B. Tips Memorial Fund at Northwestern Memorial Foundation.

Table 4 Ammonia and oxygen uptake by the brains Group

Final CBF (ml/min/100 g)

Cerebral O2 uptake (ml O2/min/100 g)

Cerebral NH3 uptake (mmol/min/100 g)

Delta NH3 (art– sag sinus) (mmol/l)

NH3 þ Veh (n ¼ 6) NH3 þ VP (n ¼ 6)

106 ^ 10 254 ^ 44*

5.7 ^ 0.7 7.5 ^ 2.1

26.1 ^ 6.5 68.3 ^ 20.4*

276 ^ 60.9 256 ^ 66.7

Values expressed as mean ^ SE; Student’s t-test. *P , 0.01 vs. other group for CBF; *P , 0.05 vs. other group for cerebral NH3 uptake.

C. Chung et al. / Journal of Hepatology 39 (2003) 193–199

References [1] Jalan R, Olde Damink SW, Deutz NE, Hayes PC, Lee A. Restoration of cerebral blood flow autoregulation and reactivity to CO2 in acute liver failure by moderate hypothermia. Hepatology 2001;34:50–4. [2] Strauss G, Hansen BA, Knudsen GM, Larsen FS. Hyperventilation restores cerebral blood flow autoregulation in patients with acute liver failure. J Hepatol 1998;28:199– 203. [3] Clemmesen JO, Hansen BA, Larsen FS. Indomethacin normalizes intracranial pressure in acute liver failure: a twenty-three-year-old woman treated with indomethacin. Hepatology 1997;26:1423 –5. [4] Blei AT, Larsen FS. Pathophysiology of cerebral edema in fulminant hepatic failure. J Hepatol 1999;31:771–6. [5] Aggarwal S, Kramer D, Yonas H, Obrist W, Kang Y, Martin M, et al. Cerebral hemodynamic and metabolic changes in fulminant hepatic failure: a retrospective study. Hepatology 1994;19:80–7. [6] Larsen FS. Cerebral circulation in liver failure: Ohm’s law in force. Semin Liver Dis 1996;16:281 –92. [7] Master S, Gottstein J, Blei AT. Cerebral blood flow and the development of ammonia-induced brain edema in rats after portacaval anastomosis. Hepatology 1999;30:876–80. [8] Cordoba J, Crespin J, Gottstein J, Blei AT. Mild hypothermia modifies ammonia-induced brain edema in rats after portacaval anastomosis. Gastroenterology 1999;116:686–93. [9] Chung C, Gottstein J, Blei AT. Indomethacin prevents the development of experimental ammonia-induced brain edema in rats after portacaval anastomosis. Hepatology 2001;34:249–54. [10] Suzuki Y, Satoh S, Oyama H, Takayasu M, Shibuya M. Regional differences in the vasodilator response to vasopressin in canine cerebral arteries in vivo. Stroke 1993;24:1049 – 53. (discussion p. 1053– 4). [11] Takayasu M, Kajita Y, Suzuki Y, Shibuya M, Sugita K, Ishikawa T, et al. Triphasic response of rat intracerebral arterioles to increasing concentrations of vasopressin in vitro. J Cereb Blood Flow Metab 1993;13:304– 9. [12] Kozniewska E, Szczepanska-Sadowska E. V2-like receptors mediate cerebral blood flow increase following vasopressin administration in rats. J Cardiovasc Pharmacol 1990;15:579–85. [13] Wenzel V, Lindner KH, Krismer AC, Voelckel WG, Schocke MF, Hund W, et al. Survival with full neurologic recovery and no cerebral

[14]

[15]

[16]

[17]

[18] [19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

199

pathology after prolonged cardiopulmonary resuscitation with vasopressin in pigs. J Am Coll Cardiol 2000;35:527–33. Wenzel V, Lindner KH. Arginine vasopressin during cardiopulmonary resuscitation: laboratory evidence, clinical experience and recommendations, and a view to the future. Crit Care Med 2002; 30(4 Suppl):S157–61. Traber P, DalCanto M, Ganger D, Blei AT. Effect of body temperature on brain edema and encephalopathy in the rat after hepatic devascularization. Gastroenterology 1989;96:885–91. Manning M, Przybylski JP, Olma A, Klis WA, Kruszynski M, Wo NC, et al. No requirements of cyclic conformation of antagonists in binding to vasopressin receptors. Nature 1987;329:839 –40. Larsen FS, Ejlersen E, Hansen BA, Knudsen GM, Tygstrup N, Secher NH. Functional loss of cerebral blood flow autoregulation in patients with fulminant hepatic failure. J Hepatol 1995;23:212–7. Olpe HR, Steinmann M. Responses of locus coeruleus neurons to neuropeptides. Prog Brain Res 1991;88:241–8. Wendon JA, Harrison PM, Keays R, Williams R. Cerebral blood flow and metabolism in fulminant liver failure. Hepatology 1994;19: 1407– 13. Niermann H, Amiry-Moghaddam M, Holthoff K, Witte OW, Ottersen OP. A novel role of vasopressin in the brain: modulation of activitydependent water flux in the neocortex. J Neurosci 2001;21:3045– 51. Sarfaraz D, Fraser CL. Effects of arginine vasopressin on cell volume regulation in brain astrocyte in culture. Am J Physiol 1999;276: E596– E601. Faraci FM, Mayhan WG, Schmid PG, Heistad DD. Effects of arginine vasopressin on cerebral microvascular pressure. Am J Physiol 1988; 255:H70–6. Prengel AW, Lindner KH, Keller A. Cerebral oxygenation during cardiopulmonary resuscitation with epinephrine and vasopressin in pigs. Stroke 1996;27:1241–8. Guevara M, Bru C, Gines P, Fernandez-Esparrach G, Sort P, Bataller R, et al. Increased cerebrovascular resistance in cirrhotic patients with ascites. Hepatology 1998;28:39–44. Lockwood AH, Yap EW, Wong WH. Cerebral ammonia metabolism in patients with severe liver disease and minimal hepatic encephalopathy. J Cereb Blood Flow Metab 1991;11:337–41. Strauss GI, Knudsen GM, Kondrup J, Moller K, Larsen FS. Cerebral metabolism of ammonia and amino acids in patients with fulminant hepatic failure. Gastroenterology 2001;121:1109–19.