Hyperoxic ventilation at the critical hematocrit: effects on myocardial perfusion and function.

BACKGROUND: Hemodilution reduces hematocrit (Hct) and blood oxygen content. Tissue oxygenation is mainly preserved by increased cardiac output. As myocardial O2-demands increase, coronary vasodilatation becomes necessary to increase myocardial blood flow. Myocardial ischemia occurs at a critical Hct-value (Hctcrit), with accompanying exhaustion of coronary reserve. Hyperoxic ventilation is known to both reverse peripheral tissue hypoxia at Hctcrit and also to induce coronary vasoconstriction. This study aimed to determine whether hyperoxic ventilation at Hctcrit further exacerbates myocardial ischemia and dysfunction. METHODS: Nine anesthetized pigs ventilated on room air were hemodiluted by 1:1 exchange of blood with pentastarch (6%HES) to Hctcrit, defined as onset of myocardial ischemia (ECG changes). At Hctcrit, hyperoxic ventilation was started. Measurements were performed at baseline, at Hctcrit, and after 15 min of hyperoxic ventilation. We determined myocardial blood flow (microsphere method), arterial O2-content, subendocardial O2-delivery and myocardial function (left ventricular pressure increase). RESULTS: At Hctcrit 7 (6;8)%, O2-content was reduced [3.7 (3.1;3.9) ml dl(-1)]. Despite a compensatory increase of myocardial blood flow [531 (449;573), ml min(-1)100 g(-1)], all pigs displayed myocardial ischemia and compromised myocardial function (P < 0.05). Hyperoxic ventilation produced increased coronary vascular resistance secondary to vasoconstriction, and reduced myocardial blood flow [426 (404;464), ml min(-1)100 g(-1); P < 0.05]. Myocardial oxygenation was found to be maintained by increased O2-content [4.4 (4.2;4.8), ml dl(-1); P < 0.05], the contribution of dissolved O2 to subendocardial O2-delivery increased (32 vs. 8%; P < 0.05), which preserved myocardial function. CONCLUSION: Hyperoxic ventilation at Hctcrit is followed by coronary vasoconstriction and reduction of coronary blood flow. However, myocardial oxygenation and function is maintained, as increased O2-content (in particular dissolved O2) preserves myocardial oxygenation.

Correlation of lesion volume and brain swelling from a focal brain trauma.

Brain edema and secondary growth of a traumatic brain tissue necrosis are important manifestations of secondary brain damage and of prognostic significance in severe head injury. Aim of the current study was to analyze the interdependency of the resulting brain swelling from the size of the focal traumatic lesion. Male Sprague-Dawley rats were intubated and mechanically ventilated. A trephination was made over the left parietal cortex for induction of a cold lesion. Different injury severities were achieved by varying the contact time of the cooled copper-cylinder and the exposed cortex. Animals were randomized into 12 experimental groups. Hemispheric brain swelling was measured in groups A1-A6 (n = 4-8) by gravimetry 24 hrs after lesions of six increasing severity levels. Correspondingly, in animals of groups B1-B6 (n = 5-7) the volume of necrosis was planimetrically assessed in histological serial sections of the brain obtained 24 hrs after trauma of different severity. In groups A1-A6. hemispheric brain swelling (increase in weight) was growing with increasing contact duration of the cold probe with the exposed cerebral cortex, i.e. from 7.7 +/- 0.4% (5 s) to a maximum of 9.9 +/- 0.5% (25 s). Longer contact periodes (30 s) were not further effective to increase hemispheric brain swelling. The contact times and extent of swelling were linearly correlated between 5 s and 25 s (r = 0.47; p < 0.01). The volume of necrosis in groups B1-B6 increased from 35.7 +/- 3.7 mm3 (5 s) to 106.3 +/- 10.3 mm3 (30 s). There was again a linear correlation between the duration of contact of the cold probe (i.e. injury severity) with the brain cortex and the volume of necrosis (r = 0.77; p < 0.01). Accordingly. the lesion volume could be increased in a reproducible manner from 35.7 up to 106.3 mm3 by extending the contact times of the cooling device and cerebral cortex. Hemispheric swelling, predominantly due to vasogenic brain edema, was expanding in relationship with the volume of necrosis.

Role of bradykinin B2 receptors in the formation of vasogenic brain edema in rats.

Bradykinin is a mediator of brain edema acting through B2 receptors. However, it is not known if bradykinin mediates the formation of cytotoxic or vasogenic brain swelling. To investigate this question we subjected rats to a cryogenic brain lesion over the left parietal cortex, a model well known to produce predominantly vasogenic brain edema. We inhibited bradykinin B2 receptors with the recently characterized nonpeptide B2 receptor antagonist, LF 16-0687. The animals were assigned to three groups (n = 10, each) receiving 10, or 100 microg/kg/min LF 16-0687 or vehicle (0.9% NaCl). Treatment started 15 min before trauma and was continued for 24 h. Another three groups of animals (n = 10, each) received 10 microg/kg/min LF 16-0687 starting 30 or 60 min after trauma or vehicle (0.9% NaCl) for 24 h. Animals were then sacrificed and swelling and water content of the brain were determined. In the vehicle treated group the traumatized hemisphere swelled by 9.3 +/- 1.1% as compared to the untraumatized contralateral side. Pretreatment with 10 microg/kg/min LF 16-0687 decreased brain swelling significantly to 6.4 +/- 1.3% (p < 0.05). Pre-treatment with 100 microg/kg/min was found to be less effective and did not result in a significant reduction of brain swelling (7.4 + 1.3%). Treatment with LF 16-0687 for 24 h (10 microg/kg/min) started 30 or 60 min after trauma did not reduce brain water content or hemispheric swelling. These results demonstrate that brain injury-mediated bradykinin production induces vasogenic brain edema by B2 receptor stimulation. Our findings further clarify the role of bradykinin in the pathophysiology of brain edema formation and confirm the therapeutic potency of bradykinin B2 receptor inhibition.

Role of nitric oxide in the secondary expansion of a cortical brain lesion from cold injury.

We have investigated the role of nitric oxide (NO) as mediator of the secondary growth of a traumatic cortical necrosis. For this purpose, a highly standardized focal lesion of the brain was induced in 46 Sprague-Dawley rats by cold injury. Twenty-four hours later--the timepoint of maximal lesion spread--the animals were sacrificed and brains were removed for histomorphometry of the maximal necrosis area and volume. The animals were divided into five experimental groups. Group I received the NO donor L-arginine as i.v. bolus 10 min prior to trauma (300 mg/kg body weight; n = 10) and a second bolus of the same dosage intraperitoneally 1 h after trauma. Group II (n = 10)--serving as control of group I--was infused with an i.v. bolus of 1 mL/kg isotonic saline 10 min prior to and a subsequent bolus i.p. 1 h after trauma. Group III (n = 8) received 100 mg/kg b.w. of the inducible NOS (iNOS) inhibitor aminoguanidine (AG) 1 h before and 8 h after trauma by intraperitoneal route. Group IV was administered with the nitric oxide synthase (NOS) inhibitor N(G)-nitro-L-arginine (L-NNA; 100 mg/kg b.w., i.p.; n = 8); group V--the controls of group III and IV--was administered with isotonic saline (1 mL/kg b.w. i.p.; n = 10) 1 h before and 8 h after trauma. In the control group with i.v./i.p. sham treatment (II), the focal lesion led to a cortical necrosis with a maximum area of 3.1 +/- 0.3 mm2 and a lesion volume of 5.7 +/- 0.5 mm3 at 24 h after trauma. In animals with administration of L-arginine, the focal lesion had a maximum area of 3.1 +/- 0.3 mm2 and a volume of 5.3 +/- 0.5 mm3. Hence, the NO donor did not affect the secondary growth of necrosis. Animals with i.p. sham treatment (group V) had a maximal lesion area of 3.6 +/- 0.2 mm2 and lesion volume of 6.2 +/- 0.4 mm3. Administration of aminoguanidine afforded significant attenuation of the lesion growth. Accordingly, the maximal area of necrosis spread only to 2.8 +/- 0.2 mm2 with a volume of 4.5 +/- 0.5 mm3, respectively, at 24 h after trauma (p < 0.01 vs group V). On the other hand, administration of L-NNA did not influence the maximal lesion area (3.7 +/- 0.2 mm2) or lesion volume (6.5 +/- 0.5 mm3) evolving at 24 h after trauma. Thus, neither the enhancement of the formation of NO by L-arginine nor gross inhibition of the synthesis of NO by L-NNA did affect the secondary spread of the necrosis from a focal trauma. The marked attenuation of the posttraumatic necrosis growth by the iNOS inhibitor aminoguanidine strongly indicates an important role of iNOS product in this phenomenon. These findings, thus, demonstrate that the expansion of a primary necrotic focal lesion is a secondary process which can be therapeutically inhibited. Thereby, the growth of a focal tissue necrosis from trauma is clearly identified as a manifestation of secondary brain damage. This information is deemed important for the better understanding of the pathophysiology of traumatic brain injury and for the targeted development of specific treatment modalities.

Interstitial lactate in the peritraumatic penumbra of rat brain.

A traumatic brain tissue necrosis is expanding to approximately 150% within 24 h after lesion. This process is accompanied by marked reduction of the perifocal cerebral blood flow likely to activate anaerobic glycolysis from a reduced O2-supply leading to an accumulation of lactic acid. The current study was carried out to assess the interstitial levels of lactic acid as a potential factor of secondary brain damage. A microdialysis probe was stereotactically implanted approximately 2 mm below the brain surface of the parietal cortex in Sprague Dawley rats (250-300 g bw; n = 6) in chloralhydrate anaesthesia. The position of the probe was controlled by histology. 24 h later a standardised cortical cold injury was induced above the probe in halothane/N2O anaesthesia. Dialysate (2 microliters/min) was collected in 15 min intervals, starting 1 h prior to and continuing until 4 hrs after trauma. The lactate concentration in the dialysate was fluorometrically determined by an enzymatic assay. Under baseline conditions dialysate concentrations of 324 +/- 48 microM were observed. A release of lactate was not found initially after trauma. Between 70 and 105 min later, however, the interstitial lactate levels briefly increased to 416 +/- 34 microM (n.s.), while reaching baseline levels again thereafter. Thus, the current results do not confirm an increased accumulation of lactate in the interstitial compartment of the penumbra despite a marked perifocal hypoperfusion of the brain after focal injury. The transitory increase in lactate at 90 min after trauma is unlikely to have caused a severe tissue acidosis which might be held liable for the secondary growth of the brain lesion induced by the focal injury.

Attenuation of secondary lesion growth in the brain after trauma by selective inhibition of the inducible NO-synthase.

In previous studies we have demonstrate that aminoguanidine pretreatment attenuates the secondary necrosis growth after focal brain trauma. Purpose of the present investigation was to elucidate the therapeutic potential of this iNOS-inhibitor when administered post lesion. Sprague-Dawley rats were subjected to a highly standardized cortical freezing lesion and administered with aminoguanidine (100 mg/kg i.p.) 15 min and 8 hrs after trauma or with isotonic saline, respectively. Animals were assigned to one of three experimental groups. The animals of group I--which served as reference for the histomorphometric determination of the spread of the primary lesion--were sacrificed 5 min after trauma. Group II, receiving isotonic saline and group III with aminoguanidine were subjected to perfusion fixation 24 hrs after trauma for evaluation of the necrosis growth. In controls with saline, the volume of the cortical necrosis increased from 6.07 +/- 1.04 mm3 (5 min) to 8.39 +/- 1.57 mm3 at 24 hrs (group II) after trauma. Treatment with aminoguanidine (group III) led to significant attenuation of the expansion of the necrosis to 6.77 +/- 0.87 mm3 at 24 hrs. Thus, the pathological role of activation of the inducible NO-synthase in the phenomenon of secondary lesion growth is confirmed by the present data on iNOS-inhibition. Attenuation of expansion of the lesion is achieved even when initiating therapy after trauma.

LF16-0687 a novel non-peptide bradykinin B2 receptor antagonist reduces vasogenic brain edema from a focal lesion in rats.

Head injury world wide is still the most frequent cause of morbidity and mortality among the population under 45 years. Approximately 50% of patients dying from severe head injury have a therapy refractory intracranial pressure rise (Baethmann 1998). Traumatic brain edema, e.g. resulting from disruption of the blood-brain barrier is viewed as an important factor of the increased intracranial pressure. Bradykinin, an active peptide of the kallikrein-kinin system is considered to enhance brain edema formation which is attributed to its permeabilizing effect on the blood-brain barrier and on dilation of arterial blood vessels in the brain mediated by B2-receptors facilitating extravasation. Currently, LF16-0687, a novel non-peptide bradykinin B2 receptor antagonist was experimentally tested as to its therapeutical potential on vasogenic brain edema from a cortical focal lesion. Following trephination of the skull in anaesthesia, male Sprague-Dawley rats were subjected to a focal cold injury of the left parietal cortex. Animals of two experimental groups were receiving either LF16-0687 as high or low dose, whereas one group of untreated animals with trauma was treated with 0.9% NaCl as continuous infusion beginning 10 min before until 24 h after lesion. 24 h after trauma the brain was removed from the skull, and the cerebral hemispheres were separated in the median plane for gravimetric assessment of hemispheric swelling. No significant reduction of hemispheric brain swelling (+7.4 +/- 2.9%) was found in animals receiving high-dose LF16-0687 as compared to the untreated controls. Brain swelling, however was significantly attenuated by the low-dose treatment, i.e. to +6.4 +/- 1.3%; vs. +9.3 +/- 1.1% found in the controls, (p < 0.05). The current data confirm that blocking of bradykinin B2-receptors by LF16-0687 is significantly attenuating vasogenic brain edema from a focal cold lesion. The therapeutical properties of the antagonist on brain edema formation cannot be attributed to a lowering of the blood pressure. Rather, specific blocking effects of B2-receptors in the brain appear to be involved. In conclusion, the understanding of secondary brain damage including brain edema in head injury has been markedly enhanced by the discovery of pathophysiologically active mediator compounds playing a role in its various manifestations. The current data confirm a pathophysiological function of bradykinin in vasogenic brain edema mediated by activation of B2-receptors. Currently it is studied whether LF16-0687 also reduces brain swelling when given after an insult.

Cerebral blood flow and the secondary growth of brain tissue necrosis after trauma.

A local brain tissue necrosis from trauma progresses during the following 24 hours or longer. A decrease in cerebral blood flow has been observed both in the necrotic as well as adjacent cortical region, which may influence expansion of the lesion into the perifocal brain tissue. Currently the regional cortical blood flow (rCBF) was assessed by using scanning laser Doppler fluxmetry. Brain tissue necrosis was induced by a highly standardised cold lesion. We attempted to inhibit the development of posttraumatic ischemia in and around the focal lesion by infusion of a hypertonic/hyperoncotic saline/starch solution. The infusion therapy resulted in a temporary improvement of posttraumatic blood flow in both necrotic and distant cortical regions. However, the expansion of the focal necrosis was not reduced. Additional investigations are in progress to determine whether further amelioration with a longer duration of rCBF increase is effective in combination with methods of neuroprotection to inhibit the secondary lesion growth after a traumatic insult.

Special aspects of severe head injury: recent developments.

Trauma in general, and head injury in particular, is the most frequent cause of mortality and morbidity in those aged up to 45 years. Outcome from severe head injury depends on the nature and severity of the primary lesion, and the manifestations of secondary brain damage of extra- and intracranial origin. The most important sequela is cerebral ischaemia resulting from intracranial hypertension caused by, for example, traumatic brain swelling or intracranial haemorrhage and/or systemic complications, of which arterial hypotension is the most significant. Because treatment so far is limited in principle to general symptomatic measures, continuing improvements in patient management is required on a comprehensive basis. In this context, major efforts are being made all over the world, not only to assess the current efficacy of, for example, logistics, organization and patient management in severe head injury, but also towards development of a consensus aimed at standardizing management and treatment procedures. With regard to the predominant influence of secondary ischaemia of the brain, recent experimental and clinical pathophysiological studies focus on the quality of cerebral blood flow, including the intriguing phenomenon of post-traumatic vasospasm. Other research objectives are concerned with the role of cytokines, leucocyte-endothelial interactions and molecular genetics in severe head injury (e.g. illuminated by the emerging role of the apolipoprotein E gene). Finally, the formation of international organizations, the American and European Brain Injury Consortium, is noteworthy. Although their primary objective is the development of guidelines for clinical trials, future objectives are conceivably more far spread and influential. It can be hoped, therefore, that the unacceptably poor outcome from severe head injury until now can be improved. Moreover, alleged management discrepancies between up-to-date trauma centres and rural hospitals may be eliminated.

The effect of dietary alpha-tocopherol on the experimental vasogenic brain edema.

It has become increasingly obvious that free radicals and lipid peroxidation contribute to brain damage from trauma by mediating edema formation and ischemia. It should, therefore, be expected that the actual level of endogenous antioxidants, as for example, vitamin C and E in plasma, has an influence on the extent of free radical-induced injury. In this communication we investigate the effect of dietary changes in the free radical scavenger alpha-tocopherol on posttraumatic cerebral swelling in Sprague-Dawley rats. Low, normal, and high plasma levels of alpha-tocopherol were established by respective diets supplied over 2 weeks. Animals of all groups received the same food without alpha-tocopherol. One group was fed a vitamin E-free diet. The pellet-food for the other animals was supplemented either with 5-mg alpha-tocopherol/100 g or 250-mg alpha-tocopherol/100 g dry mass, respectively. The vitamin E-free diet lowered the alpha-tocopherol level in plasma to 30% of control, whereas supplementation with 250 mg/100 g led to a plasma concentration of 200% of control. The animals were then subjected to a focal cold injury of the left cerebral hemisphere. Twenty-four hours after trauma the brain was removed and the water content of each hemisphere was determined by the wet-dry weight method. Swelling of the traumatized hemisphere was calculated as the difference in weight between the traumatized and contralateral control hemisphere. The 2-week alpha-tocopherol supplementation or -deletion diet, respectively, did not either afford significant reduction or lead to an enhancement of traumatic brain swelling. Likewise, the increase in brain water content of the traumatized hemisphere was not affected. It is concluded that supplementation or depletion of alpha-tocopherol for 2 weeks, resulting in a marked increase or decrease of the vitamin E plasma level, does not influence formation of posttraumatic vasogenic brain edema.

The penumbra zone of a traumatic cortical lesion: a microdialysis study of excitatory amino acid release.

A cortical tissue necrosis from focal trauma expands to 150% of its initial volume within 24 hrs. It is currently unknown, whether this phenomenon is part of the primary traumatic lesion or if it involves secondary mechanisms such as the release of excitatory amino acids into the traumatic penumbra zone. A microdialysis probe was inserted for that purpose in an oblique angle into the cortex of Sprague-Dawley rats, approximately 2 mm below the brain surface. One day later a highly standardized cortical freezing lesion was induced at the brain cortex above the microdialysis probe. Dialysate was collected prior to, during, and after trauma in 10 min intervals. In each animal, it was confirmed histologically, that the tip of the microdialysis probe was localized in the grey matter in close vicinity to the primary lesion. Following induction of the trauma a statistically significant increase of the dialysate level of aspartate, glutamate, glycine, and serine was observed, whereas that of alanine was not altered throughout the experiment. The posttraumatic increase of the excitatory neurotransmitters aspartate and glutamate indicates that these amino acids are involved in the secondary lesion growth after trauma. Confirmation of this hypothesis would require that specific antagonization of these excitotoxic amino acids is inhibiting growth of the lesion.

Assessment of regional cortical blood flow following traumatic lesion of the brain.

A brain tissue necrosis from trauma gradually expands during the subsequent 24 h. Among others, deterioration of perifocal blood flow could be involved in the secondary extinction of initially viable brain tissue. A highly standardized freezing lesion was made in cerebral cortex of rats for frequent measurements of regional cortical blood flow with high spatial resolution by laser Doppler scanning flowmetry. Following trauma a profound decrease in cerebral blood flow was found, not only in the lesion proper but also in the perifocal and distant brain areas, which would support a role of ischemia in the secondary lesion growth. Further studies, however, are required, particularly on whether therapeutic improvement of perifocal flow is affecting expansion of the traumatic tissue necrosis.

Treatment of vasogenic brain edema with the novel Cl- transport inhibitor torasemide.

The efficacy of the diuretic agent torasemide, which antagonizes the Na+/K+/Cl- cotransport and Cl- channels, was investigated to determine its inhibition of brain edema from a focal cerebral lesion. For this purpose, cold injury of the brain was induced in 50 Sprague-Dawley rats while monitoring arterial blood pressure. The brain was removed for gravimetric assessment of swelling of the traumatized hemisphere 24 h after trauma. The water content was also determined after drying the cerebral hemispheres for 24 h. Animals were divided into five groups. A control group with trauma received vehicle only; two other groups received 1.0 or 10.0 mg torasemide/kg body weight 30 minutes before and 6 h after trauma (n = 10-12). Administration of the drug after the insult was also investigated in animals with application of vehicle or 10.0 mg/kg of torasemide at 30 minutes and 6 h following the brain lesion (n = 8). Torasemide did not affect important physiologic variables, such as the arterial pO2, pCO2, pH, hemoglobin, hematocrit, or plasma osmolality, while increasing blood pressure (p < 0.01). The blood pressure response notwithstanding, treatment significantly attenuated hemispheric brain swelling from trauma. In control animals without treatment, cold injury led to hemispheric swelling of 8.89%. In animals with 1 mg torasemide/kg BW, brain swelling amounted to 8.51% and to 7.04% in animals receiving 10 mg/kg before and after the insult (p < 0.005). Treatment was also found to attenuate the increase in tissue water content from trauma, but without reaching statistical significance. Postinsult treatment with torasemide (10 mg/kg BW) at 30 minutes and 6 h after trauma was again associated with a significant reduction in hemispheric brain swelling, which in this group amounted to 7.46% compared with 9.76% in the untreated controls (p < 0.005). The increase in the cerebral water content from trauma was also significantly blunted in the latter experiments (p < 0.01). The present data indicate a remarkable therapeutic potential of the novel diuretic agent torasemide to reduce vasogenic brain edema from an acute cerebral lesion. It is surmised that the compound specifically interferes with Cl- transport mechanisms, which apparently are activated in edematous brain involving neuronal and glial cells, for example. This conclusion is supported by in vitro observations that torasemide inhibits the swelling of glial cells from acidosis. On the other hand, it is unlikely that gross dehydration of the brain secondary to the induction of diuresis by the agent played a role, because hematocrit and plasma osmolality were not found to be affected.

Growth kinetics of a primary brain tissue necrosis from a focal lesion.

Secondary brain damage, such as brain edema or impairment of the cerebral microcirculation may evolve from tissue necrosis of the brain induced by trauma or ischemia. This laboratory has provided novel information on the secondary increase of a primary brain tissue necrosis resulting from a focal lesion. We have presently investigated more closely the growth kinetics of this process during 24 h after trauma. Rats were subjected to a standardized focal freezing injury of the brain. Area and volume of the resulting necrosis were quantitatively assessed by morphometry after different periods of survival (i.e., 5 min, 3, 6, 12, 18 and 24 h after trauma). The maximal area of necrosis increased by 45% (p < 0.001) during the posttraumatic observation period. Growth of necrosis after trauma was not limited to the early period, but continued between 12 and 24 h, amounting then to 29% (p < 0.05). The volume of necrosis calculated on the basis of histological serial sections was also observed to increase by 45%. The current findings confirm that a primary brain tissue lesion induced by a standard cryogenic injury, studied as model of a contusion focus in severe head injury, is subjected to secondary growth within a period of 24 h after trauma, longer periods of survival were not investigated yet. Quantification of lesion growth makes possible not only to study underlying mechanisms, but also of whether this process can be therapeutically inhibited.

Therapeutical efficacy of a novel chloride transport blocker and an IP3-analogue in vasogenic brain edema.

The efficacy of torasemide, a novel chloride-channel blocker, and of PP56, an IP3 analogue, was currently examined in experimental brain edema. Following trephination in anesthesia rats were subjected to a focal cold injury of the left cerebral hemisphere. Animals of 4 experimental groups receiving either torasemide (i.v. at 30 min before and 6 h after lesion) or PP56 (continuous infusion beginning at 30 min before until 24 h after lesion) at two dose levels were compared with controls administered with i.v. saline. 24 h after trauma the brain was removed from the skull, and the hemispheres were separated in the median plane for gravimetric assessment of hemispheric swelling. Hct, blood gases and body temperature remained constant in all groups. Blood pressure was found to increase in a dose-dependent manner in animals with torasemide. No significant reduction of brain swelling was found in animals with low-dose torasemide (8.51 +/- 0.63%) or low- (7.91 +/- 0.60) and high-dose PP56 (6.85 +/- 1.05%) as compared to the untreated controls. Brain swelling, however, was significantly attenuated by high-dose torasemide to 7.04 +/- 0.36%, as compared to 8.89 +/- 0.29% of the untreated group (p < 0.005). It is currently studied whether torasemide reduces brain swelling when given after the insult.