- Email: [email protected]
Intracisternal versus intracarotid infusion of L-arginine in experimental cerebral vasospasm
Journal of Clinical Neuroscience 14 (2007) 556–562 www.elsevier.com/locate/jocn
Laboratory study
Intracisternal versus intracarotid infusion of L-arginine in experimental cerebral vasospasm ¨ zu¨m ¨ nal O U a
¨ zen Karadag˘ a, Mustafa Gu¨relik a, , Adem Aslan a, O b Aysßenur Tasß , H. Zafer Kars a
a,*
Department of Neurosurgery, Cumhuriyet University Faculty of Medicine, Sivas, Turkey Department of Neurology, Cumhuriyet University Faculty of Medicine, Sivas, Turkey
b
Received 28 December 2005; accepted 7 March 2006
Abstract Aim: The effect of short term intracisternal and intracarotid L-arginine infusion on experimental cerebral acute phase vasospasm in a rabbit subarachnoid haemorrhage model is investigated, and the two groups compared. Materials and Method: Subarachnoid haemorrhage was produced by intracisternal injection of autologous blood in New Zealand rabbits. On the fourth day after subarachnoid haemorrhage, cerebral blood flow was monitored using transcranial Doppler ultrasonography during intracisternal and intracarotid saline and L-arginine infusions. Result: Cerebral blood flow measurements revealed resolution of vasospasm with short-term intracisternal and intracarotid L-arginine infusion. No significant difference was found between the effects of intracisternal and intracarotid L-arginine infusions, however intracarotid L-arginine infusion created a more potent vasodilatation towards the end of infusion. Conclusion: Both intracisternal and intracarotid short term L-arginine infusion significantly improve acute phase cerebral vasospasm after experimental subarachnoid haemorrhage. Intracarotid L-arginine infusion is more potent and safer as large amounts of intracisternal L-arginine may lead to overproduction of nitric oxide by inducible nitric oxide synthase with the production of free radicals. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Nitric oxide; L-arginine; Cerebral vasospasm; Subarachnoid haemorrhage
1. Introduction The regulation of cerebral blood flow (CBF) is significantly dependent on cerebral vascular tone, modulated by a balance between contraction and relaxation effects.1 Basal cerebral vascular tone requires continuous release of nitric oxide (NO) by endothelial nitric oxide synthase (eNOS). Neuronal nitric oxide synthase (nNOS) together with eNOS, control CBF by autoregulation and chemoregulation. A disturbance in the contraction-relaxation
*
Corresponding author. Fax: +90 346 2191333. ¨ zu¨m). ¨. O E-mail address: [email protected] (U
0967-5868/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2006.03.013
balance in favour of contraction is considered to be the cause of cerebral vasospasm after subarachnoid haemorrhage (SAH).2,3 Nitric oxide is the soluble endogenous neurotransmitter gas produced by oxidation of L-arginine by the enzyme NOS. NO is thought to be involved in many physiological events, particularly in regulation of vascular tone. NO activates soluble guanylyl cyclase (sCG). Soluble guanylyl cyclase converts guanlylate 3 0 , 5 0 triphosphate into cyclic guanlylate 3 0 , 5 0 monohosphate (cGMP). Elevated cGMP in vascular smooth muscle causes relaxation via a number of pathways.4 The role of NO in the pathophysiology of vasospasm after SAH is still uncertain. Decreased NO availability
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
through increased consumption and/or decreased production in the cerebral arterial wall has been proposed. NO, its donors and precursors delivered via intra-arterial, intravenous and intrathecal or intraventricular routes, have been investigated to prevent cerebral vasospasm after SAH.3,5–9 Intracisternal and intrathecal administration of L-arginine, a NOS substrate, has been used to increase NO production. Intracisternal and intracarotid L-arginine administration increased CBF in preclinical studies.8 A previous study by Go¨ksel et al. showed that continuous intracisternal L-arginine infusion improves experimental cerebral vasospasm.6 In this study the effect of short-term intracisternal and intracarotid L-arginine infusion on experimental acute phase vasospasm in a rabbit SAH model is investigated, and the two groups compared. 2. Materials and methods Forty New Zealand albino rabbits of both sexes, weighing between 2350–3050 grams were used in this study. The animals were kept at the Animal Care Centre of Cumhuriyet University Medical Faculty throughout the experiment. Environmental temperature and humidity levels were the same for all animals under normal daylight conditions. Through the experiment, the rabbits had free access to food and water. The experimental protocol was approved by the Regional Animal Care and Use Committee. 2.1. Animal groups Three groups were designated for this study: control (Cont) group (n = 8); SAH-intracisternal infusion (Cis) group (n = 16); and SAH-intracarotid infusion (Car) group (n = 16). Animals in Cis group were randomly assigned to SAHintracisternal saline infusion (CisS) group (n = 8), and SAH-intracisternal L-arginine infusion (CisLA) group (n = 8). Animals in Car group were randomly assigned to SAH-intracarotid saline infusion (CarS) group (n = 8) and SAH-intracarotid L-arginine infusion (CarLA) group (n = 8). 2.2. Anaesthesia Intramuscular injections of 50 mg/kg ketamine-HCl plus 5 mg/kg xylazine were given for all procedures. 2.3. Experimental SAH A modified model of Chan et al. was used to create experimental vasospasm.10 Fresh autologous, non-heparinized blood was drawn from each animal’s cannulated femoral artery, and was slowly given percutaneously into the cisterna magna with a 24-gauge steel needle at a dose of 0.7 mL/kg after removal of an equal amount of cerebrospi-
557
nal fluid. The animals were then tilted with tail up for 15 min in order to diffuse the blood into the subarachnoid space. 2.4. Surgical procedures In Cis groups, an intracisternal catheter was placed on the fourth day after SAH. This procedure was performed under the guidance of a surgical microscope. A midline dorsal neck incision was made. With dissection of muscles, the dura mater between the foramen magnum and the C1 lamina was exposed. The dura was perforated with a 22gauge steel needle. After observation of cerebrospinal fluid (CSF) flow from this hole, a 2 F silicone catheter (Bard Inc. Murray Hill, NJ, USA) was advanced for 5 millimetres into the cisterna magna. The catheter was sealed to the dura with tissue glue (Beriplast P, Aventis-Behring, King of Prussia, PA, USA) and incision was closed with layered sutures. The outer end of the catheter was sealed off and covered with sterile gauze until use. In Car groups, an intracarotid catheter was placed on the fourth day after SAH. Under guidance of a surgical microscope, a right paramedian incision, extending approximately 3 cm inferiorly from the angle of the mandible, was made. The right carotid bifurcation was exposed lateral to the trachea and oesophagus. Both the internal and external carotid arteries were identified. The internal carotid artery was temporarily ligated just above the bifurcation and was perforated with a 22-gauge steel needle approximately 1 cm superior to the bifurcation. A 2 F silicone catheter (Bard Inc.) was advanced 2 cm into the internal carotid artery. The catheter was sealed to the artery with tissue glue (Beriplast P, Aventis-Behring) and the incision was closed with layered sutures. Heparinized saline was infused into the catheter in order to prevent coagulation. The tip of the catheter was checked under fluoroscopy. The outer end of the catheter was sealed off and covered with sterile gauze until use. In Cis and Car groups an intraparenchymal transducer was placed on the fourth day after SAH to monitor intracranial pressure and brain temperature (ICP-BT). A burr hole was opened in the right frontal bone 2 mm anterior to the coronal suture and 3 mm lateral to the midline. After the dura was exposed and perforated, a transducer (Camino 110-4BT, Integra NeuroCare, San Diego, CA, USA) was advanced 5 mm, and fixed to the calvarium. A monitoring device (Model MPM-1, Camino with MPM-1 Serial Data Capture Software Version 2.0, Integra NeuroCare) was used for data recording. The right femoral artery was percutaneously cannulated in CisLA, and CarLA groups. 2.5. Cerebral blood flow measurement Flow in the right internal carotid artery (ICA) was evaluated using transcranial Doppler (TCD) ultrasonography via a right transorbital window. After flow signals from
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
558
the right ICA were detected, the clearest signals were then recorded. Measurement of mean flow velocity (MFV) in the right ICA was used, and all TCD studies were performed in a blinded fashion. A transcranial Doppler (TCD) ultrasonography device with 2 MHz frequency probe (Smart-Lite TCD System, RIMED, Raanana, Israel) was used.
accepted at p < 0.05. SPSS for Windows (Version 10.0) (SPSS Inc., Chicago, IL, USA) was used. 3. Results Mean MFV in the control group was 25.62 ± 0.88 cm/ sec. In CisS and CisLA groups mean MFV values were 53.12 ± 2.39 cm/sec and 51.25 ± 2.06 cm/sec at time zero (0 min), respectively. There were significant differences between the mean MFV values of the control group, and CisS and CisLA groups (p < 0.05) at the beginning of the infusion due to severe vasospasm in the right ICA. Saline infusions did not create any significant MFV change (p > 0.05) (Fig. 1). When L-arginine infusion commenced, the mean MFV began to decrease and this trend continued until the 30th min of infusion. Thereafter, mean MFV values was stable until the end of infusion. There was no significant difference between the mean MFV value in the control group and the mean MFV values at the 30th, 40th, 50th and 60th min in the CisLA group, respectively (p > 0.05). Thirty minutes after the infusion was completed (90th min of the experiment), the mean MFV value was higher than that at the 60th min, but lower than the value at 0 min in the CisLA group (Fig. 1). The mean MFV value recorded 30 min after the infusion was completed (90th min of the experiment), was significantly different from the mean MFV values of all minutes in the CisLA group (p < 0.05). Alterations in ICP-BT, body temperature, mean arterial pressure, and arterial PO2, PCO2, HCO 3 , SaO2, and pH in the CisLA group were statistically insignificant (p < 0.05) (Table 1). In CarS and CarLA groups mean MFV values were 53.25 ± 1.86 cm/sec and 53.50 ± 2.12 cm/sec at time zero (0 min), respectively. There were significant differences in mean MFV between the control group, and CarS and
2.6. Experimental design In the control group, CBF was evaluated using TCD without any surgical procedure and/or infusion. In Cis and Car groups, TCD was performed on the fourth day after SAH, and 2 h after catheter and ICP-BT transducer placement, the time of which was accepted as zero CBF evaluation. After this study, sterile saline was infused into the cisterna magna of animals in the CisS group and internal carotid artery of animals in CarS group. 300 lmol L-arginine was infused into the cisterna magna of animals in CisLA infusion group and into the internal carotid artery of animals in CarLA group for 60 min (infusion rate = 1 mL/h). To prevent an uncontrolled vasoactive effect, an L-arginine solution in water was used which had a pH of 11.4. During saline and L-arginine infusions TCD was performed at 10-min intervals. TCD was repeated 30 min after L-arginine infusion was completed. ICP-BT, body temperature, and mean arterial pressure were monitored at 10-min intervals; arterial PO2, PCO2, HCO 3 , SaO2, and pH were monitored at 20-min intervals in CisLA, and CarLA groups. 2.7. Statistical analysis The Kruskal-Wallis test, Mann-Whitney U-test and variance analysis of repeated measurements with the Bonferroni correction were used for statistical analysis. Significance was
60
MFV (cm/sec)
50
CisS CisLA Control
40
30
20 0
10
20
30
40
50
60
70
80
90
Time (min) Fig. 1. Mean flow velocity (MFV) curves with standard error in CisS (intracisternal saline), and CisLA (intracisternal L-arginine) groups. Interrupted line shows MFV in the control group.
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
559
Table 1 Physiological parameters (mean ± standard error) in the CisLA group Time (min)
0 10 20 30 40 50 60
ICP
ICT
7.1 ± 0.29 5.2 ± 0.16 7.2 ± 0.45 6.2 ± 0.55 5.3 ± 018 5.3 ± 026 6.0 ± 018
BT
36.6 ± 0.05 36.5 ± 0.02 36.6 ± 0.02 36.4 ± 0.03 36.6 ± 0.04 36.7 ± 0.03 36.6 ± 0.03
MAP
36.7 ± 0.04 36.5 ± 0.12 36.6 ± 0.04 36.5 ± 0.01 36.7 ± 0.05 36.8 ± 0.02 36.5 ± 0.07
Arterial
85.3 ± 0.53 88.2 ± 0.25 89.0 ± 0.18 86.3 ± 0.37 89.1 ± 0.39 88.3 ± 0.92 87.3 ± 0.46
PO2
PCO2
HCO 3
SaO2
pH
86.3 ± 0.55
28.0 ± 0.27
23.3 ± 0.26
95.5 ± 0.26
7.43 ± 0.04
84.1 ± 0.37
30.2 ± 0.19
21.5 ± 0.38
93.3 ± 0.70
7.40 ± 0.00
88.1 ± 0.74
29.7 ± 0.39
22.6 ± 0.28
96.1 ± 0.39
7.42 ± 0.03
85.8 ± 0.86
30.4 ± 0.29
24.5 ± 0.15
94.3 ± 0.32
7.44 ± 0.03
o
o
ICP, Intracranial pressure (mmHg); ICT, Intracranial temperature (C ); BT, Body temperature (C ); MAP, Mean arterial pressure (mmHg); PO2, Pressure of O2 (mmHg); PCO2, Pressure of CO2 (mmHg); HCO 3 , Bicarbonate concentration (mmol/lt); SaO2, Oxygen saturation (%); pH, Blood pH.
MFV at 30 min after the infusion was completed (90th min of the experiment) was significantly different from the mean MFV values of all minutes in the CarLA group (p < 0.05). Alterations in ICP-BT, body temperature, mean arterial pressure, and arterial PO2, PCO2, HCO 3 , SaO2, and pH in the CarLA group were statistically insignificant (p > 0.05) (Table 2).
CarLA groups (p < 0.05) at the beginning of the infusion due to severe vasospasm in the right ICA. Saline infusion did not produce any significant MFV change (p > 0.05) (Fig. 2). When L-arginine infusion commenced, the mean MFV began to decrease and this trend continued until the 30th min of infusion. Thereafter mean MFV was stable until the 60th min of infusion. There was no significant difference between the mean MFV values in the control group and the mean MFV values at the 30th, 40th and 50th min in the CarLA group, respectively (p > 0.05). At the end of the infusion, the mean MFV value was lower than that at the 30th, 40th, 50th min of infusion. The mean MFV value at the 60th min of infusion was also lower than the mean MFV of the control group which means that an additional dilatation was created in the right ICA at the 60th min of L-arginine infusion (Fig. 2). There were significant differences between the mean MFV at the 60th min of L-arginine infusion and that in the control group and the 30th, and 40th min of L-arginine infusion (p < 0.05). Thirty minutes after the infusion was completed (90th min of the experiment) the mean MFV value was higher than the values at the 60th min, but lower than the value at time zero in the CarLA group (Fig. 2). The mean
3.1. Intracisternal versus intracarotid L-arginine infusion As soon as the infusion was started, mean MFV values began to decrease and this trend continued until the 30th min of infusion in both the CisLA, and CarLA groups. After the 30th min of infusion, MFV remained relatively stable in the CisLA group, but decreased further in the CarLA group. The difference between the mean MFV values of CisLA and CarLA groups at the 60th min was significant (p < 0.05). Thirty minutes after the infusion was completed (90th min of the experiment), the mean MFV in both the CisLA, and CarLa groups was higher than that at the 60th min of infusion, but lower than at time zero (Fig. 3).
60
MFV (cm/sec)
50 CarS CarLA
40
Control
30
20 0
10
20
30
40
50
60
70
80
90
Time (min) Fig. 2. Mean flow velocity (MFV) curves with standard error in CarS (intracarotid saline), and CarLA (intracarotid L-arginine) groups. Interrupted line shows MFV in the control group.
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
560
Table 2 Physiological parameters (mean ± standard error) in the CarLA group Times (min)
0 10 20 30 40 50 60
ICP
ICT
6.0 ± 0.26 5.1 ± 0.22 4.5 ± 0.18 4.8 ± 0.22 5.1 ± 0.22 6.1 ± 0.22 6.2 ± 0.25
BT
36.5 ± 0.05 36.5 ± 0.02 36.6 ± 0.03 36.5 ± 0.03 36.7 ± 0.03 36.6 ± 0.03 36.7 ± 0.04
MAP
36.5 ± 0.01 36.7 ± 0.14 36.5 ± 0.02 36.5 ± 0.01 36.8 ± 0.03 36.6 ± 0.01 36.9 ± 0.02
Arterial
86.5 ± 0.26 88.0 ± 0.26 89.1 ± 0.22 87.0 ± 0.32 88.6 ± 0.26 85.0 ± 0.26 88.1 ± 0.51
PO2
PCO2
HCO 3
SaO2
pH
92.1 ± 0.05
31.3 ± 0.01
24.2 ± 0.03
96.1 ± 0.22
7.43 ± 0.05
90.5 ± 0.11
30.6 ± 0.02
23.5 ± 0.04
95.0 ± 0.26
7.40 ± 0.05
92.4 ± 0.07
28.7 ± 0.05
24.7 ± 0.04
95.1 ± 0.22
7.42 ± 0.04
88.3 ± 0.10
30.9 ± 0.05
24.1 ± 0.06
94.6 ± 0.32
7.43 ± 0.04
o
o
ICP, Intracranial pressure (mmHg); ICT, Intracranial temperature (C ); BT, Body temperature (C ); MAP, Mean arterial pressure (mmHg); PO2, Pressure of O2 (mmHg); PCO2, Pressure of CO2 (mmHg); HCO 3 , Bicarbonate concentration (mmol/lt); SaO2, Oxygen saturation (%); pH, Blood pH.
60
CisLa
50
MFV (cm/sec)
CarLa Control
40
30
20 0
10
20
30
40
50
60
70
80
90
Time (min) Fig. 3. Mean flow velocity (MFV) curves with standard error in CisLA (intracisternal L-arginine), and CarLA (intracarotid L-arginine) groups. Interrupted line shows MFV in the control group.
4. Discussion Nitric oxide decreases during the first 30 min after SAH. This decrease may result from scavenging by haemoglobin and anaerobic metabolism into nitrosylhemoglobin. NO is restored 60 min after SAH.11,12 This may result from activation of inducible nitric oxide synthase (iNOS) by haemoglobin and oxyhaemoglobin in macrophages and activated microglia. Thereafter, overproduction of NO by activated iNOS results in elevated NO in CSF and brain tissue.13–16 The half-life of NO is reduced under conditions of oxidative stress.17 Many researchers have found elevated nitric oxide metabolites that reflect high NO levels after SAH.18–21 Overproduction and overconsumption of NO may occur simultaneously after SAH. While the elevation of NO in the early stages of SAH prevents vasospasm, overproduction of NO may activate production of free radicals which may cause pathological changes in the vascular endothelium and smooth muscle cells.22 This contributes to pathological changes in the cell wall of endothelium and smooth muscle cells in the media layer of arteries. Alteration of the cell wall structure may in turn disturb the diffusion of NO from endothelial cell to smooth muscle cell.23,24
Intracisternal L-arginine may be utilized by upregulated nNOS and/or activated iNOS to produce NO. Furthermore L-arginine may diffuse to the endothelium and eNOS may also produce NO. However, haematoma around the vessels and inflammation in all layers of the vessel wall may be a barrier to L-arginine diffusion into the endothelium. Pluta et al. reported that in a primate model of SAH, immunoreactivity for nNOS was virtually absent in the nervi vasorum of the middle cerebral artery (MCA) in spasm.25 Therefore, iNOS probably utilises intracisternal L-arginine. But, intracisternal L-arginine may result in large amounts of peroxynitrite that can be toxic to cell structure. A significant correlation was found between elevated NO metabolites in CSF after SAH and poor gradeSAH.13,26 Also in different experimental models of cerebral trauma, neuronal tissue damage is followed by increase of NO production.27,28 Although there is evidence for dysfunction of eNOS in SAH in the literature, intracarotid infusion of L-arginine caused vasodilatation in the present study.29,30 Studies on a partially purified preparation of NOS in a cell-free system indicated that the Michaelis constant (Km) of NOS for L-arginine was in the micromolar range.31 It has also been shown that intracellular L-arginine is higher
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
than plasma L-arginine. L-arginine is present in the millimolar range within the endothelial cell. As available intracellular L-arginine concentration is always in considerable excess of NOS, extracellular L-arginine administration should drive NO production, a phenomenon known as the ‘‘arginine paradox’’.17 Several explanations have been proposed to account for this paradox. One of them is the effect of endogenous inhibitor of NO, asymmetric dimethylarginine (ADMA).32 It is possible that ADMA antagonizes the normal intracellular concentration of L-arginine, and additional L-arginine supplementation is required to overcome a functional defect of NOS substrate. In this study we aimed to compare the therapeutic effect of short-term intracisternal and intracarotid L-arginine infusion in experimental acute phase cerebral vasospasm. Both intracisternal and intracarotid short-term L-arginine infusion significantly improved cerebral vasospasm after SAH. Intracisternal L-arginine infusion improves cerebral vasospasm, but large amounts of intracisternal L-arginine may lead to overproduction of NO by iNOS which may trigger the production of free radicals which can destroy cell structure. Therefore, intracarotid L-arginine infusion is probably safer. In this study, intracarotid L-arginine infusion is proposed to have induced an NO level sufficient to dilate the right ICA to the control artery diameter at the 30th min. At this time, the plasma L-arginine level may overcome functional defects of NOS substrate, and contraction and relaxation factors might balance. With continuing intracarotid L-arginine infusion, this balance favours relaxation factors. Several explanations can be proposed to account for this additional vasodilatation. First, L-arginine may inhibit peripheral sympathetic tone leading to vasodilatation via its metabolites.30 Second, the plasma level of L-arginine at the 60th min of intracarotid L-arginine infusion might overwhelm the inhibition effects of ADMA and also cause additional dilatation. Intracisternal L-arginine infusion does not cause additional dilatation at the 60th min of infusion. In conclusion, we speculate that short-term continuous intracarotid infusion of L-arginine, a substrate of NOS, may be a treatment for acute phase cerebral vasospasm. Due to the short infusion time of a single dose of L-arginine, this study is preliminary, and cannot be extrapolated for human treatment. The use of intracarotid L-arginine infusion in the clinical setting may be possible after investigation of the efficacy and safety of various doses and durations of intracarotid L-arginine infusion. References 1. Brain JE, Faraci FM, Hesitad DD. Recent insights into the regulation of cerebral circulation. Clin Exp Pharmacol Physiol 1996;23:449–57. 2. Edward DH, Byrne J, Griffith TM. The effect of chronic subarachnoid hemorrhage on basal endothelium-derived relaxing factor activity in intratechal cerebral arteries. J Neurosurg 1992;76:830–7. 3. Pluta RM, Thompson BG, Afshar JK, et al. Nitric oxide and vasospasm. Acta Neurochir Suppl 2001;77:67–72.
561
4. Iadecola C. Bright and dark sides of nitric oxide in ischemic brain injury. Trends Neurosci 1997;20:132–9. 5. Afshar JKB, Pluta RM, Boock RJ, et al. Effect of intracarotid nitric oxide on primate cerebral vasospasm after subarachnoid hemorrhage. J Neurosurg 1995;83:118–22. _ The therapeutic effect of ¨ zu¨m U ¨ ztoprak I. ¨, O 6. Go¨ksel HM, O continuous intracisternal L-arginine infusion on experimental cerebral vasospasm. Acta Neurochir (Wien) 2001;143:277–85. 7. Kajita Y, Suzuki Y, Oyama H, et al. Combined effect of L-arginine and superoxide dismutase on the spastic basilar artery after subarachnoid hemorrhage in dogs. J Neurosurg 1994;80:476–83. 8. Pluta RM, Oldfield EH, Boock RJ. Reversal and prevention of cerebral vasospasm by intracarotid infusion of nitric oxide donors in a primate model of subarachnoid hemorrhage. J Neurosurg 1997;87:746–51. ¨ nal B, Inalo¨z S, et al. Therapeutic effect of intracarotid 9. So¨nmez OF, U infusion of spermine/nitric oxide complex on cerebral vasospasm. Acta Neurochir 2002;144:921–8. 10. Chan RC, Durty FA, Thompson GB, et al. The role of the prostacyclin-thromboxane system in cerebral vasospasm following induced subarachnoid hemorrhage. J Neurosurg 1984;61:1120–8. 11. Schwartz AY, Sehba FA, Bederson JB. Decreased nitric oxide availability contributes to acute cerebral ischemia after subarachnoid hemorrhage. Neurosurgery 2000;47:208–15. 12. Sehba FA, Schwartz AY, Chereshnev I, et al. Acute decrease in cerebral nitric oxide levels after subarachnoid hemorrhage. J Cereb Blood Flow Metab 2000;20:604–11. 13. Peterson JW, Kwun BD, Hackett JD, et al. The role of inflammation in experimental cerebral vasospasm. J Neurosurg 1990;72:767–74. 14. Seifert V, Stolke D, Kunz U, et al. Influence of blood volume on cerebrospinal fluid levels of arachidonic acid metabolites after subarachnoid hemorrhage: experimental study on the pathogenesis of cerebral vasospasm. Neurosurgery 1988;23:313–21. 15. Sills AK, Clatterbuck RE, Thompson RC, et al. Endothelial cell expression of intercellular adhesion molecule in experimental post hemorrhagic vasospasm. Neurosurgery 1997;41:453–61. 16. Yamashima T, Yamamato S. Cerebral arterial pathology in experimental subarachnoid hemorrhage. J Neurosurg 1983;30:843–50. 17. Cooke JP. Does ADMA cause endothelial dysfunction? Arterioscler Thromb Vasc Biol 2000;20:2032–7. 18. Khaldi A, Zauner A, Reinert M, et al. Measurement of nitric oxide and brain oxygen tension in patients after severe subarachnoid hemorrhage. Neurosurgery 2001;49:33–40. 19. Ng WH, Moochhala S, Yeo TT, et al. Nitric oxide and subarachnoid hemorrhage: Elevated levels in cerebrospinal fluid and their implications. Neurosurgery 2001;49:622–6. 20. Suzuki Y, Osuka K, Noda A, et al. Nitric oxide metabolites in the cisternal cerebral spinal fluid of patients with subarachnoid hemorrhage. Neurosurgery 1997;4:807–12. 21. Woszczyk A, Deinsberger W, Bo¨ker DK. Nitric oxide metabolites in cisternal CSF correlate with cerebral vasospasm in patients with a subarachnoid hemorrhage. Acta Neurochir 2003;145:257–64. 22. Beckman JS, Beckman TW, Chen J, et al. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide superoxide. Proc Natl Acad Sci USA 1990;87:1620–4. 23. Dawson VL, Dawson TD, London ED, et al. Nitric oxide mediates glutamate neurotoxicity in primary cortical cultures. Proc Natl Acad Sci USA 1991;88:6368–71. 24. Radi R, Beckman JS, Bush KM, et al. Peroxynitrite oxidation of sulfhydyls: the cytotoxic potential of superoxide and nitric oxide. J Biol Chem 1997;266:4244–50. 25. Pluta RM, Thompson BG, Dawson TM, et al. Loss of nitric oxide synthase immunoreactivity in cerebral vasospasm. J Neurosurg 1996;84:648–54. 26. Sadamitsu D, Kuroda Y, Nagamitsu T, et al. Cerebrospinal fluid and plasma concentrations of nitric oxide metabolites in postoperative patients with subarachnoid hemorrhage. Crit Care Med 2001;29:77–9. 27. Gahm C, Holmin S, Matheisen T. Nitric oxide synthase expression after human brain contusion. Neurosurgery 2002;50:1319–26.
562
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
28. Khaldi A, Chiueh CC, Bullock MR, et al. The significance of nitric oxide production in the brain after injury. Ann N Y Acad Sci 2002;962:53–9. 29. Hino A, Tokuyama Y, Weir B, et al. Change in endothelial nitric oxide synthase mRNA during vasospasm after subarachnoid hemorrhage in monkeys. Neurosurgery 1996;36:562–8. 30. Khurana VG, Sohni YR, Mangrum WI, et al. Endothelial nitric oxide synthase gene polymorphisms predict susceptibility to aneuris-
mal subarachnoid hemorrhage and cerebral vasospasm. J Cereb Blood Flow Metab 2004;24:291–7. 31. Pollock JS, Forstermann U, Mitchell JA, et al. Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelium cells. Proc Natl Acad Sci USA 1991;88:10480–4. 32. Vallance P, Chan N. Endothelial function and nitric oxide: clinical relevance. Heart 2001;85:342–50.
Laboratory study
Intracisternal versus intracarotid infusion of L-arginine in experimental cerebral vasospasm ¨ zu¨m ¨ nal O U a
¨ zen Karadag˘ a, Mustafa Gu¨relik a, , Adem Aslan a, O b Aysßenur Tasß , H. Zafer Kars a
a,*
Department of Neurosurgery, Cumhuriyet University Faculty of Medicine, Sivas, Turkey Department of Neurology, Cumhuriyet University Faculty of Medicine, Sivas, Turkey
b
Received 28 December 2005; accepted 7 March 2006
Abstract Aim: The effect of short term intracisternal and intracarotid L-arginine infusion on experimental cerebral acute phase vasospasm in a rabbit subarachnoid haemorrhage model is investigated, and the two groups compared. Materials and Method: Subarachnoid haemorrhage was produced by intracisternal injection of autologous blood in New Zealand rabbits. On the fourth day after subarachnoid haemorrhage, cerebral blood flow was monitored using transcranial Doppler ultrasonography during intracisternal and intracarotid saline and L-arginine infusions. Result: Cerebral blood flow measurements revealed resolution of vasospasm with short-term intracisternal and intracarotid L-arginine infusion. No significant difference was found between the effects of intracisternal and intracarotid L-arginine infusions, however intracarotid L-arginine infusion created a more potent vasodilatation towards the end of infusion. Conclusion: Both intracisternal and intracarotid short term L-arginine infusion significantly improve acute phase cerebral vasospasm after experimental subarachnoid haemorrhage. Intracarotid L-arginine infusion is more potent and safer as large amounts of intracisternal L-arginine may lead to overproduction of nitric oxide by inducible nitric oxide synthase with the production of free radicals. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Nitric oxide; L-arginine; Cerebral vasospasm; Subarachnoid haemorrhage
1. Introduction The regulation of cerebral blood flow (CBF) is significantly dependent on cerebral vascular tone, modulated by a balance between contraction and relaxation effects.1 Basal cerebral vascular tone requires continuous release of nitric oxide (NO) by endothelial nitric oxide synthase (eNOS). Neuronal nitric oxide synthase (nNOS) together with eNOS, control CBF by autoregulation and chemoregulation. A disturbance in the contraction-relaxation
*
Corresponding author. Fax: +90 346 2191333. ¨ zu¨m). ¨. O E-mail address: [email protected] (U
0967-5868/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2006.03.013
balance in favour of contraction is considered to be the cause of cerebral vasospasm after subarachnoid haemorrhage (SAH).2,3 Nitric oxide is the soluble endogenous neurotransmitter gas produced by oxidation of L-arginine by the enzyme NOS. NO is thought to be involved in many physiological events, particularly in regulation of vascular tone. NO activates soluble guanylyl cyclase (sCG). Soluble guanylyl cyclase converts guanlylate 3 0 , 5 0 triphosphate into cyclic guanlylate 3 0 , 5 0 monohosphate (cGMP). Elevated cGMP in vascular smooth muscle causes relaxation via a number of pathways.4 The role of NO in the pathophysiology of vasospasm after SAH is still uncertain. Decreased NO availability
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
through increased consumption and/or decreased production in the cerebral arterial wall has been proposed. NO, its donors and precursors delivered via intra-arterial, intravenous and intrathecal or intraventricular routes, have been investigated to prevent cerebral vasospasm after SAH.3,5–9 Intracisternal and intrathecal administration of L-arginine, a NOS substrate, has been used to increase NO production. Intracisternal and intracarotid L-arginine administration increased CBF in preclinical studies.8 A previous study by Go¨ksel et al. showed that continuous intracisternal L-arginine infusion improves experimental cerebral vasospasm.6 In this study the effect of short-term intracisternal and intracarotid L-arginine infusion on experimental acute phase vasospasm in a rabbit SAH model is investigated, and the two groups compared. 2. Materials and methods Forty New Zealand albino rabbits of both sexes, weighing between 2350–3050 grams were used in this study. The animals were kept at the Animal Care Centre of Cumhuriyet University Medical Faculty throughout the experiment. Environmental temperature and humidity levels were the same for all animals under normal daylight conditions. Through the experiment, the rabbits had free access to food and water. The experimental protocol was approved by the Regional Animal Care and Use Committee. 2.1. Animal groups Three groups were designated for this study: control (Cont) group (n = 8); SAH-intracisternal infusion (Cis) group (n = 16); and SAH-intracarotid infusion (Car) group (n = 16). Animals in Cis group were randomly assigned to SAHintracisternal saline infusion (CisS) group (n = 8), and SAH-intracisternal L-arginine infusion (CisLA) group (n = 8). Animals in Car group were randomly assigned to SAH-intracarotid saline infusion (CarS) group (n = 8) and SAH-intracarotid L-arginine infusion (CarLA) group (n = 8). 2.2. Anaesthesia Intramuscular injections of 50 mg/kg ketamine-HCl plus 5 mg/kg xylazine were given for all procedures. 2.3. Experimental SAH A modified model of Chan et al. was used to create experimental vasospasm.10 Fresh autologous, non-heparinized blood was drawn from each animal’s cannulated femoral artery, and was slowly given percutaneously into the cisterna magna with a 24-gauge steel needle at a dose of 0.7 mL/kg after removal of an equal amount of cerebrospi-
557
nal fluid. The animals were then tilted with tail up for 15 min in order to diffuse the blood into the subarachnoid space. 2.4. Surgical procedures In Cis groups, an intracisternal catheter was placed on the fourth day after SAH. This procedure was performed under the guidance of a surgical microscope. A midline dorsal neck incision was made. With dissection of muscles, the dura mater between the foramen magnum and the C1 lamina was exposed. The dura was perforated with a 22gauge steel needle. After observation of cerebrospinal fluid (CSF) flow from this hole, a 2 F silicone catheter (Bard Inc. Murray Hill, NJ, USA) was advanced for 5 millimetres into the cisterna magna. The catheter was sealed to the dura with tissue glue (Beriplast P, Aventis-Behring, King of Prussia, PA, USA) and incision was closed with layered sutures. The outer end of the catheter was sealed off and covered with sterile gauze until use. In Car groups, an intracarotid catheter was placed on the fourth day after SAH. Under guidance of a surgical microscope, a right paramedian incision, extending approximately 3 cm inferiorly from the angle of the mandible, was made. The right carotid bifurcation was exposed lateral to the trachea and oesophagus. Both the internal and external carotid arteries were identified. The internal carotid artery was temporarily ligated just above the bifurcation and was perforated with a 22-gauge steel needle approximately 1 cm superior to the bifurcation. A 2 F silicone catheter (Bard Inc.) was advanced 2 cm into the internal carotid artery. The catheter was sealed to the artery with tissue glue (Beriplast P, Aventis-Behring) and the incision was closed with layered sutures. Heparinized saline was infused into the catheter in order to prevent coagulation. The tip of the catheter was checked under fluoroscopy. The outer end of the catheter was sealed off and covered with sterile gauze until use. In Cis and Car groups an intraparenchymal transducer was placed on the fourth day after SAH to monitor intracranial pressure and brain temperature (ICP-BT). A burr hole was opened in the right frontal bone 2 mm anterior to the coronal suture and 3 mm lateral to the midline. After the dura was exposed and perforated, a transducer (Camino 110-4BT, Integra NeuroCare, San Diego, CA, USA) was advanced 5 mm, and fixed to the calvarium. A monitoring device (Model MPM-1, Camino with MPM-1 Serial Data Capture Software Version 2.0, Integra NeuroCare) was used for data recording. The right femoral artery was percutaneously cannulated in CisLA, and CarLA groups. 2.5. Cerebral blood flow measurement Flow in the right internal carotid artery (ICA) was evaluated using transcranial Doppler (TCD) ultrasonography via a right transorbital window. After flow signals from
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
558
the right ICA were detected, the clearest signals were then recorded. Measurement of mean flow velocity (MFV) in the right ICA was used, and all TCD studies were performed in a blinded fashion. A transcranial Doppler (TCD) ultrasonography device with 2 MHz frequency probe (Smart-Lite TCD System, RIMED, Raanana, Israel) was used.
accepted at p < 0.05. SPSS for Windows (Version 10.0) (SPSS Inc., Chicago, IL, USA) was used. 3. Results Mean MFV in the control group was 25.62 ± 0.88 cm/ sec. In CisS and CisLA groups mean MFV values were 53.12 ± 2.39 cm/sec and 51.25 ± 2.06 cm/sec at time zero (0 min), respectively. There were significant differences between the mean MFV values of the control group, and CisS and CisLA groups (p < 0.05) at the beginning of the infusion due to severe vasospasm in the right ICA. Saline infusions did not create any significant MFV change (p > 0.05) (Fig. 1). When L-arginine infusion commenced, the mean MFV began to decrease and this trend continued until the 30th min of infusion. Thereafter, mean MFV values was stable until the end of infusion. There was no significant difference between the mean MFV value in the control group and the mean MFV values at the 30th, 40th, 50th and 60th min in the CisLA group, respectively (p > 0.05). Thirty minutes after the infusion was completed (90th min of the experiment), the mean MFV value was higher than that at the 60th min, but lower than the value at 0 min in the CisLA group (Fig. 1). The mean MFV value recorded 30 min after the infusion was completed (90th min of the experiment), was significantly different from the mean MFV values of all minutes in the CisLA group (p < 0.05). Alterations in ICP-BT, body temperature, mean arterial pressure, and arterial PO2, PCO2, HCO 3 , SaO2, and pH in the CisLA group were statistically insignificant (p < 0.05) (Table 1). In CarS and CarLA groups mean MFV values were 53.25 ± 1.86 cm/sec and 53.50 ± 2.12 cm/sec at time zero (0 min), respectively. There were significant differences in mean MFV between the control group, and CarS and
2.6. Experimental design In the control group, CBF was evaluated using TCD without any surgical procedure and/or infusion. In Cis and Car groups, TCD was performed on the fourth day after SAH, and 2 h after catheter and ICP-BT transducer placement, the time of which was accepted as zero CBF evaluation. After this study, sterile saline was infused into the cisterna magna of animals in the CisS group and internal carotid artery of animals in CarS group. 300 lmol L-arginine was infused into the cisterna magna of animals in CisLA infusion group and into the internal carotid artery of animals in CarLA group for 60 min (infusion rate = 1 mL/h). To prevent an uncontrolled vasoactive effect, an L-arginine solution in water was used which had a pH of 11.4. During saline and L-arginine infusions TCD was performed at 10-min intervals. TCD was repeated 30 min after L-arginine infusion was completed. ICP-BT, body temperature, and mean arterial pressure were monitored at 10-min intervals; arterial PO2, PCO2, HCO 3 , SaO2, and pH were monitored at 20-min intervals in CisLA, and CarLA groups. 2.7. Statistical analysis The Kruskal-Wallis test, Mann-Whitney U-test and variance analysis of repeated measurements with the Bonferroni correction were used for statistical analysis. Significance was
60
MFV (cm/sec)
50
CisS CisLA Control
40
30
20 0
10
20
30
40
50
60
70
80
90
Time (min) Fig. 1. Mean flow velocity (MFV) curves with standard error in CisS (intracisternal saline), and CisLA (intracisternal L-arginine) groups. Interrupted line shows MFV in the control group.
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
559
Table 1 Physiological parameters (mean ± standard error) in the CisLA group Time (min)
0 10 20 30 40 50 60
ICP
ICT
7.1 ± 0.29 5.2 ± 0.16 7.2 ± 0.45 6.2 ± 0.55 5.3 ± 018 5.3 ± 026 6.0 ± 018
BT
36.6 ± 0.05 36.5 ± 0.02 36.6 ± 0.02 36.4 ± 0.03 36.6 ± 0.04 36.7 ± 0.03 36.6 ± 0.03
MAP
36.7 ± 0.04 36.5 ± 0.12 36.6 ± 0.04 36.5 ± 0.01 36.7 ± 0.05 36.8 ± 0.02 36.5 ± 0.07
Arterial
85.3 ± 0.53 88.2 ± 0.25 89.0 ± 0.18 86.3 ± 0.37 89.1 ± 0.39 88.3 ± 0.92 87.3 ± 0.46
PO2
PCO2
HCO 3
SaO2
pH
86.3 ± 0.55
28.0 ± 0.27
23.3 ± 0.26
95.5 ± 0.26
7.43 ± 0.04
84.1 ± 0.37
30.2 ± 0.19
21.5 ± 0.38
93.3 ± 0.70
7.40 ± 0.00
88.1 ± 0.74
29.7 ± 0.39
22.6 ± 0.28
96.1 ± 0.39
7.42 ± 0.03
85.8 ± 0.86
30.4 ± 0.29
24.5 ± 0.15
94.3 ± 0.32
7.44 ± 0.03
o
o
ICP, Intracranial pressure (mmHg); ICT, Intracranial temperature (C ); BT, Body temperature (C ); MAP, Mean arterial pressure (mmHg); PO2, Pressure of O2 (mmHg); PCO2, Pressure of CO2 (mmHg); HCO 3 , Bicarbonate concentration (mmol/lt); SaO2, Oxygen saturation (%); pH, Blood pH.
MFV at 30 min after the infusion was completed (90th min of the experiment) was significantly different from the mean MFV values of all minutes in the CarLA group (p < 0.05). Alterations in ICP-BT, body temperature, mean arterial pressure, and arterial PO2, PCO2, HCO 3 , SaO2, and pH in the CarLA group were statistically insignificant (p > 0.05) (Table 2).
CarLA groups (p < 0.05) at the beginning of the infusion due to severe vasospasm in the right ICA. Saline infusion did not produce any significant MFV change (p > 0.05) (Fig. 2). When L-arginine infusion commenced, the mean MFV began to decrease and this trend continued until the 30th min of infusion. Thereafter mean MFV was stable until the 60th min of infusion. There was no significant difference between the mean MFV values in the control group and the mean MFV values at the 30th, 40th and 50th min in the CarLA group, respectively (p > 0.05). At the end of the infusion, the mean MFV value was lower than that at the 30th, 40th, 50th min of infusion. The mean MFV value at the 60th min of infusion was also lower than the mean MFV of the control group which means that an additional dilatation was created in the right ICA at the 60th min of L-arginine infusion (Fig. 2). There were significant differences between the mean MFV at the 60th min of L-arginine infusion and that in the control group and the 30th, and 40th min of L-arginine infusion (p < 0.05). Thirty minutes after the infusion was completed (90th min of the experiment) the mean MFV value was higher than the values at the 60th min, but lower than the value at time zero in the CarLA group (Fig. 2). The mean
3.1. Intracisternal versus intracarotid L-arginine infusion As soon as the infusion was started, mean MFV values began to decrease and this trend continued until the 30th min of infusion in both the CisLA, and CarLA groups. After the 30th min of infusion, MFV remained relatively stable in the CisLA group, but decreased further in the CarLA group. The difference between the mean MFV values of CisLA and CarLA groups at the 60th min was significant (p < 0.05). Thirty minutes after the infusion was completed (90th min of the experiment), the mean MFV in both the CisLA, and CarLa groups was higher than that at the 60th min of infusion, but lower than at time zero (Fig. 3).
60
MFV (cm/sec)
50 CarS CarLA
40
Control
30
20 0
10
20
30
40
50
60
70
80
90
Time (min) Fig. 2. Mean flow velocity (MFV) curves with standard error in CarS (intracarotid saline), and CarLA (intracarotid L-arginine) groups. Interrupted line shows MFV in the control group.
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
560
Table 2 Physiological parameters (mean ± standard error) in the CarLA group Times (min)
0 10 20 30 40 50 60
ICP
ICT
6.0 ± 0.26 5.1 ± 0.22 4.5 ± 0.18 4.8 ± 0.22 5.1 ± 0.22 6.1 ± 0.22 6.2 ± 0.25
BT
36.5 ± 0.05 36.5 ± 0.02 36.6 ± 0.03 36.5 ± 0.03 36.7 ± 0.03 36.6 ± 0.03 36.7 ± 0.04
MAP
36.5 ± 0.01 36.7 ± 0.14 36.5 ± 0.02 36.5 ± 0.01 36.8 ± 0.03 36.6 ± 0.01 36.9 ± 0.02
Arterial
86.5 ± 0.26 88.0 ± 0.26 89.1 ± 0.22 87.0 ± 0.32 88.6 ± 0.26 85.0 ± 0.26 88.1 ± 0.51
PO2
PCO2
HCO 3
SaO2
pH
92.1 ± 0.05
31.3 ± 0.01
24.2 ± 0.03
96.1 ± 0.22
7.43 ± 0.05
90.5 ± 0.11
30.6 ± 0.02
23.5 ± 0.04
95.0 ± 0.26
7.40 ± 0.05
92.4 ± 0.07
28.7 ± 0.05
24.7 ± 0.04
95.1 ± 0.22
7.42 ± 0.04
88.3 ± 0.10
30.9 ± 0.05
24.1 ± 0.06
94.6 ± 0.32
7.43 ± 0.04
o
o
ICP, Intracranial pressure (mmHg); ICT, Intracranial temperature (C ); BT, Body temperature (C ); MAP, Mean arterial pressure (mmHg); PO2, Pressure of O2 (mmHg); PCO2, Pressure of CO2 (mmHg); HCO 3 , Bicarbonate concentration (mmol/lt); SaO2, Oxygen saturation (%); pH, Blood pH.
60
CisLa
50
MFV (cm/sec)
CarLa Control
40
30
20 0
10
20
30
40
50
60
70
80
90
Time (min) Fig. 3. Mean flow velocity (MFV) curves with standard error in CisLA (intracisternal L-arginine), and CarLA (intracarotid L-arginine) groups. Interrupted line shows MFV in the control group.
4. Discussion Nitric oxide decreases during the first 30 min after SAH. This decrease may result from scavenging by haemoglobin and anaerobic metabolism into nitrosylhemoglobin. NO is restored 60 min after SAH.11,12 This may result from activation of inducible nitric oxide synthase (iNOS) by haemoglobin and oxyhaemoglobin in macrophages and activated microglia. Thereafter, overproduction of NO by activated iNOS results in elevated NO in CSF and brain tissue.13–16 The half-life of NO is reduced under conditions of oxidative stress.17 Many researchers have found elevated nitric oxide metabolites that reflect high NO levels after SAH.18–21 Overproduction and overconsumption of NO may occur simultaneously after SAH. While the elevation of NO in the early stages of SAH prevents vasospasm, overproduction of NO may activate production of free radicals which may cause pathological changes in the vascular endothelium and smooth muscle cells.22 This contributes to pathological changes in the cell wall of endothelium and smooth muscle cells in the media layer of arteries. Alteration of the cell wall structure may in turn disturb the diffusion of NO from endothelial cell to smooth muscle cell.23,24
Intracisternal L-arginine may be utilized by upregulated nNOS and/or activated iNOS to produce NO. Furthermore L-arginine may diffuse to the endothelium and eNOS may also produce NO. However, haematoma around the vessels and inflammation in all layers of the vessel wall may be a barrier to L-arginine diffusion into the endothelium. Pluta et al. reported that in a primate model of SAH, immunoreactivity for nNOS was virtually absent in the nervi vasorum of the middle cerebral artery (MCA) in spasm.25 Therefore, iNOS probably utilises intracisternal L-arginine. But, intracisternal L-arginine may result in large amounts of peroxynitrite that can be toxic to cell structure. A significant correlation was found between elevated NO metabolites in CSF after SAH and poor gradeSAH.13,26 Also in different experimental models of cerebral trauma, neuronal tissue damage is followed by increase of NO production.27,28 Although there is evidence for dysfunction of eNOS in SAH in the literature, intracarotid infusion of L-arginine caused vasodilatation in the present study.29,30 Studies on a partially purified preparation of NOS in a cell-free system indicated that the Michaelis constant (Km) of NOS for L-arginine was in the micromolar range.31 It has also been shown that intracellular L-arginine is higher
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
than plasma L-arginine. L-arginine is present in the millimolar range within the endothelial cell. As available intracellular L-arginine concentration is always in considerable excess of NOS, extracellular L-arginine administration should drive NO production, a phenomenon known as the ‘‘arginine paradox’’.17 Several explanations have been proposed to account for this paradox. One of them is the effect of endogenous inhibitor of NO, asymmetric dimethylarginine (ADMA).32 It is possible that ADMA antagonizes the normal intracellular concentration of L-arginine, and additional L-arginine supplementation is required to overcome a functional defect of NOS substrate. In this study we aimed to compare the therapeutic effect of short-term intracisternal and intracarotid L-arginine infusion in experimental acute phase cerebral vasospasm. Both intracisternal and intracarotid short-term L-arginine infusion significantly improved cerebral vasospasm after SAH. Intracisternal L-arginine infusion improves cerebral vasospasm, but large amounts of intracisternal L-arginine may lead to overproduction of NO by iNOS which may trigger the production of free radicals which can destroy cell structure. Therefore, intracarotid L-arginine infusion is probably safer. In this study, intracarotid L-arginine infusion is proposed to have induced an NO level sufficient to dilate the right ICA to the control artery diameter at the 30th min. At this time, the plasma L-arginine level may overcome functional defects of NOS substrate, and contraction and relaxation factors might balance. With continuing intracarotid L-arginine infusion, this balance favours relaxation factors. Several explanations can be proposed to account for this additional vasodilatation. First, L-arginine may inhibit peripheral sympathetic tone leading to vasodilatation via its metabolites.30 Second, the plasma level of L-arginine at the 60th min of intracarotid L-arginine infusion might overwhelm the inhibition effects of ADMA and also cause additional dilatation. Intracisternal L-arginine infusion does not cause additional dilatation at the 60th min of infusion. In conclusion, we speculate that short-term continuous intracarotid infusion of L-arginine, a substrate of NOS, may be a treatment for acute phase cerebral vasospasm. Due to the short infusion time of a single dose of L-arginine, this study is preliminary, and cannot be extrapolated for human treatment. The use of intracarotid L-arginine infusion in the clinical setting may be possible after investigation of the efficacy and safety of various doses and durations of intracarotid L-arginine infusion. References 1. Brain JE, Faraci FM, Hesitad DD. Recent insights into the regulation of cerebral circulation. Clin Exp Pharmacol Physiol 1996;23:449–57. 2. Edward DH, Byrne J, Griffith TM. The effect of chronic subarachnoid hemorrhage on basal endothelium-derived relaxing factor activity in intratechal cerebral arteries. J Neurosurg 1992;76:830–7. 3. Pluta RM, Thompson BG, Afshar JK, et al. Nitric oxide and vasospasm. Acta Neurochir Suppl 2001;77:67–72.
561
4. Iadecola C. Bright and dark sides of nitric oxide in ischemic brain injury. Trends Neurosci 1997;20:132–9. 5. Afshar JKB, Pluta RM, Boock RJ, et al. Effect of intracarotid nitric oxide on primate cerebral vasospasm after subarachnoid hemorrhage. J Neurosurg 1995;83:118–22. _ The therapeutic effect of ¨ zu¨m U ¨ ztoprak I. ¨, O 6. Go¨ksel HM, O continuous intracisternal L-arginine infusion on experimental cerebral vasospasm. Acta Neurochir (Wien) 2001;143:277–85. 7. Kajita Y, Suzuki Y, Oyama H, et al. Combined effect of L-arginine and superoxide dismutase on the spastic basilar artery after subarachnoid hemorrhage in dogs. J Neurosurg 1994;80:476–83. 8. Pluta RM, Oldfield EH, Boock RJ. Reversal and prevention of cerebral vasospasm by intracarotid infusion of nitric oxide donors in a primate model of subarachnoid hemorrhage. J Neurosurg 1997;87:746–51. ¨ nal B, Inalo¨z S, et al. Therapeutic effect of intracarotid 9. So¨nmez OF, U infusion of spermine/nitric oxide complex on cerebral vasospasm. Acta Neurochir 2002;144:921–8. 10. Chan RC, Durty FA, Thompson GB, et al. The role of the prostacyclin-thromboxane system in cerebral vasospasm following induced subarachnoid hemorrhage. J Neurosurg 1984;61:1120–8. 11. Schwartz AY, Sehba FA, Bederson JB. Decreased nitric oxide availability contributes to acute cerebral ischemia after subarachnoid hemorrhage. Neurosurgery 2000;47:208–15. 12. Sehba FA, Schwartz AY, Chereshnev I, et al. Acute decrease in cerebral nitric oxide levels after subarachnoid hemorrhage. J Cereb Blood Flow Metab 2000;20:604–11. 13. Peterson JW, Kwun BD, Hackett JD, et al. The role of inflammation in experimental cerebral vasospasm. J Neurosurg 1990;72:767–74. 14. Seifert V, Stolke D, Kunz U, et al. Influence of blood volume on cerebrospinal fluid levels of arachidonic acid metabolites after subarachnoid hemorrhage: experimental study on the pathogenesis of cerebral vasospasm. Neurosurgery 1988;23:313–21. 15. Sills AK, Clatterbuck RE, Thompson RC, et al. Endothelial cell expression of intercellular adhesion molecule in experimental post hemorrhagic vasospasm. Neurosurgery 1997;41:453–61. 16. Yamashima T, Yamamato S. Cerebral arterial pathology in experimental subarachnoid hemorrhage. J Neurosurg 1983;30:843–50. 17. Cooke JP. Does ADMA cause endothelial dysfunction? Arterioscler Thromb Vasc Biol 2000;20:2032–7. 18. Khaldi A, Zauner A, Reinert M, et al. Measurement of nitric oxide and brain oxygen tension in patients after severe subarachnoid hemorrhage. Neurosurgery 2001;49:33–40. 19. Ng WH, Moochhala S, Yeo TT, et al. Nitric oxide and subarachnoid hemorrhage: Elevated levels in cerebrospinal fluid and their implications. Neurosurgery 2001;49:622–6. 20. Suzuki Y, Osuka K, Noda A, et al. Nitric oxide metabolites in the cisternal cerebral spinal fluid of patients with subarachnoid hemorrhage. Neurosurgery 1997;4:807–12. 21. Woszczyk A, Deinsberger W, Bo¨ker DK. Nitric oxide metabolites in cisternal CSF correlate with cerebral vasospasm in patients with a subarachnoid hemorrhage. Acta Neurochir 2003;145:257–64. 22. Beckman JS, Beckman TW, Chen J, et al. Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide superoxide. Proc Natl Acad Sci USA 1990;87:1620–4. 23. Dawson VL, Dawson TD, London ED, et al. Nitric oxide mediates glutamate neurotoxicity in primary cortical cultures. Proc Natl Acad Sci USA 1991;88:6368–71. 24. Radi R, Beckman JS, Bush KM, et al. Peroxynitrite oxidation of sulfhydyls: the cytotoxic potential of superoxide and nitric oxide. J Biol Chem 1997;266:4244–50. 25. Pluta RM, Thompson BG, Dawson TM, et al. Loss of nitric oxide synthase immunoreactivity in cerebral vasospasm. J Neurosurg 1996;84:648–54. 26. Sadamitsu D, Kuroda Y, Nagamitsu T, et al. Cerebrospinal fluid and plasma concentrations of nitric oxide metabolites in postoperative patients with subarachnoid hemorrhage. Crit Care Med 2001;29:77–9. 27. Gahm C, Holmin S, Matheisen T. Nitric oxide synthase expression after human brain contusion. Neurosurgery 2002;50:1319–26.
562
¨ zu¨m et al. / Journal of Clinical Neuroscience 14 (2007) 556–562 ¨.O U
28. Khaldi A, Chiueh CC, Bullock MR, et al. The significance of nitric oxide production in the brain after injury. Ann N Y Acad Sci 2002;962:53–9. 29. Hino A, Tokuyama Y, Weir B, et al. Change in endothelial nitric oxide synthase mRNA during vasospasm after subarachnoid hemorrhage in monkeys. Neurosurgery 1996;36:562–8. 30. Khurana VG, Sohni YR, Mangrum WI, et al. Endothelial nitric oxide synthase gene polymorphisms predict susceptibility to aneuris-
mal subarachnoid hemorrhage and cerebral vasospasm. J Cereb Blood Flow Metab 2004;24:291–7. 31. Pollock JS, Forstermann U, Mitchell JA, et al. Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelium cells. Proc Natl Acad Sci USA 1991;88:10480–4. 32. Vallance P, Chan N. Endothelial function and nitric oxide: clinical relevance. Heart 2001;85:342–50.