Main

Considerable progress has been made in understanding the mechanisms causing brain damage. Medications have been tried for neuroprotection in experimental models as well as in clinical settings, but still there is no effective drug capable of protecting brain tissue against hypoxia-ischemia (1). This may be explained by the multifactorial mechanism of hypoxic-ischemic brain injury, difficulties in defining its onset, duration and severity, and the side effects reported in the use of investigational drugs.

Preservation of cerebral vascular reactivity and minimization of cerebral hemodynamic alterations caused by hypoxia-ischemia could be very important in cerebral protection. Alteration of cerebral hemodynamics accompanies substantial brain damage in experimental and clinical settings (2–4) with associated disruption of autoregulation, the most vulnerable regulatory mechanism of the cerebral circulation (3, 4). The level of cAMP detected in CSF correlates with the degree and the duration of asphyxia (5, 6) and has been suggested as a possible indicator of altered cerebral vascular reactivity (6–8) and brain metabolic and energy state (9). Also CSF cAMP concentrations have been correlated with long-term neurodevelopmental outcomes in neonates who had perinatal asphyxia (10). Various subtypes of sodium channels that use cAMP as second messenger have been detected in brain tissue and in vascular cells (11). Sodium channels have a role in the regulation of cerebral function, hemodynamics, and metabolism (11, 12). However, pharmacologic modulation of voltage-gated sodium channels for the purpose of neurovascular protection has received very little attention. In the present study, we hypothesized that sodium channel inhibition in severely asphyxiated newborn piglets would maintain CSF cAMP levels and would attenuate alterations in cerebrovascular response and reactivity during asphyxia and recovery/reventilation.

METHODS

The surgical and experimental procedures were reviewed and approved by the Animal Care and Use Committee of The University of Tennessee, Memphis. Piglets 2–5 d of age (1–2 kg) of either sex were anesthetized and instrumented before the experiments. Piglets were initially anesthetized with intramuscular ketamine hydrochloride (33 mg/kg) and acepromazine (3.3 mg/kg) and maintained on intravenous α-chloralose (50 mg/kg initially plus 3 mg·kg−1·h−1). The animals were ventilated with room air through a tracheotomy tube by using a Bournes BP200 infant ventilator. Catheters were inserted into the femoral vein for maintenance of anesthesia and fluid administration (5 mL·kg−1·h−1 of 5% dextrose in water) and into the femoral arteries to record BP and heart rate for blood gas sampling. Body temperature was maintained between 37 and 38°C.

Closed cranial window placement for measurements of pial arteriolardiameters and collection of periarachnoid CSF.

Each piglet had a closed cranial window inserted over the parietal cortex for measurements of pial arteriolar diameter, qualitative assessments of pial arteriolar flow, and collection of cortical periarachnoid CSF for cAMP determination. The methods for insertion of a closed cranial window have been described previously (5, 6, 8). Pial arterioles were observed with a Wild dissecting microscope, a video monitor, and a video microscaler. The images of the arterioles were displayed on the television monitor, and parallel lines projected by the microscaler bracketed the sides of the vessels. As the arteriolar diameter changed, the lines were moved manually to correspond with the vessel walls. Precalibration of the distance between the lines allowed determination of arteriolar diameter.

Protocol.

Three groups of piglets were studied, group 1 or asphyxia-control group (n = 5) and treatment groups 2 and 3 (n = 12). Animals in group 2 (n = 6) were treated with 50 mg/kg of novel nonselective sodium channel blocker HOE642 (cariporide mesilate, a Na+/H+ exchange blocker) before asphyxia, whereas animals in group 3 (n = 6) were treated at the end of asphyxia, the beginning of reventilation. Na+ channel blocker HOE642 was provided by Hoeschst Marion Roussel, Chemical Research, G838, 65926 Frankfurt, Germany. Group 1 animals or nontreated animals received an equal volume of normal saline infusion at the same rate. Thirty minutes after administration of normal saline or medication, asphyxia was induced by ventilating the piglets with a gas mixture of 10% CO2, 10% O2, and 80% N2 and by decreasing the minute volume. Respiratory support and gas mixture were manipulated to maintain systemic BP around 30 mm Hg and heart rate between 50 and 80 beats/min. At the end of 60 min of asphyxia, minute volume was increased, and piglets were ventilated with room air. Piglets were monitored during 60 min of recovery. Arterial blood was collected every 10 min for blood gases and pH measurement using an Instrumentation Blood Gas Analyzer (Instrumentation Laboratory, Lexington, MA). BP and heart rate were monitored continuously during the experiments. Cortical periarachnoid CSF was collected every 10 min for later measurement of cAMP. Diameters of two to three randomly selected pial arterioles were recorded every 10 min during the experiments. Also, when diameter measurements were being made, it was possible to characterize qualitatively the forward movement of RBC because of the thin arteriolar wall. RBC flow was described by one of the following: normal forward flow, increased forward flow, sluggish or decreased forward flow, or no visible forward flow. Further, it was noted whether sludging or clumping of RBC was observed. Vascular reactivity to topically applied isoproterenol (10−4 M) was also evaluated after 60 min of recovery.

cAMP assay.

Cortical periarachnoid CSF (0.4 mL) collected from under the cranial window was mixed with EDTA (5 μL) and stored at −60°C until assayed. cAMP was measured in CSF samples by using RIA procedures (6–8).

Statistical analysis.

Values reported as mean ± SEM were compared within group and between groups over time by 3-way ANOVA with repeated measures. Tukey's method was used for multiple comparisons. A p value < 0.05 was considered significant. The association between pial arteriolar diameters and CSF cAMP levels over time within group and between groups was analyzed using mixed linear models with random effects. The relationship between these variables within group was expressed as y =a +b (log cAMP), wherein y is the predicted pial arteriolar diameter, a the intercept, and b the slope. Statistics software used was S-PLUS (Mathsoft, Seattle, WA).

RESULTS

The groups of piglets did not differ in weight, baseline normal BP, pH, and blood gas values (Fig. 1). With the onset of asphyxia, BP, pH, and PO2 quickly decreased and were maintained at critically low levels by manipulating the degree of asphyxia with subsequent return to low normal level during recovery or reventilation. At each period after induction of asphyxia, BP, pH, PO2, and PCO2 changed significantly from baseline within group, but these changes were not statistically different among groups. These are also illustrated in Figure 1. Piglets pretreated with sodium channel blocker (group 2) tended to maintain slightly higher (although not statistically significant) pH values than in the other two groups throughout asphyxia. The base excess (BE) values were significantly higher during asphyxia in group 2 compared with the other two groups (p <.05); these differences were also observed during recovery/reventilation.

Figure 1
figure 1

Mean BP, pH, PaCO2, PaO2, and base excess (BE) throughout the experiment in the three groups of piglets. Baseline values are plotted at −30 min before asphyxia, and asphyxia was induced immediately at 0 min. +p < 0.05 at each period compared with baseline within all three groups; ★p < 0.05 compared with baseline within groups 1 and 3 only; ○p < 0.05, group 2 compared with either group 1 or 3.

Full size image

During asphyxia, CSF cAMP levels were significantly higher in the pretreated piglets compared with either group 1 or 3 (Fig. 2), with the peak CSF cAMP concentration occurring at 40 min of asphyxia. Levels in group 2 decreased as asphyxia continued, but a higher level was maintained compared with either group 1 or 3. CSF cAMP levels increased during recovery/reventilation in all groups, with levels higher in group 2 compared with group 1. CSF cAMP values in group 3 exceeded either those of group 1 or 2.

Figure 2
figure 2

Percent change in CSF cAMP concentrations from baseline measured every 10 min in the three groups of piglets. Values for each period within all three groups that are significantly different from baseline are indicated by +(p < 0.05). During asphyxia, the magnitude of increase in CSF cAMP is significantly greater in group 2 compared with groups 1 and 3. A significantly greater increase in levels occurred during recovery in groups 2 and 3 compared with group 1.

Full size image

The increase in pial arteriolar diameters with the onset of asphyxia (Fig. 3) was greater in groups 1 and 3 compared with group 2. Pial diameters in groups 1 and 3 decreased during the last one-half hour of asphyxia but remained greater than measurements in group 2 or values at baseline. During recovery, pial arteriolar diameters in groups 2 and 3 decreased toward baseline measurements, whereas diameters in group 1 piglets decreased but remained above baseline.

Figure 3
figure 3

Percent change in pial arteriolar diameter from baseline measured every 10 min throughout the experiment. For each designated period, values that are significantly different from baseline within each of the three groups were designated by +(p < 0.05). Pial arteriolar dilation during asphyxia occurred to a lesser degree in group 2 compared with groups 1 and 3.

Full size image

Analysis of the relationship between CSF cAMP and pial arteriolar diameter while controlling for time, using a linear mixed model with random effects, showed significant correlations between these measurements within group 1 (r2 = 0.38, p < 0.001) and group 2 (r2 = 0.30, p < 0.001) as illustrated in Figure 4. No significant correlation was noted between pial arteriolar diameters and CSF cAMP levels in group 3. In this group, the pial arteriolar diameter increased with increasing CSF cAMP during asphyxia, but, with treatment upon initiation of reventilation, the significant increase in CSF cAMP was associated with a decrease in pial arteriolar diameter from a maximally dilated state toward baseline values.

Figure 4
figure 4

Relationship between CSF cAMP concentrations and pial arteriolar diameter within each group over time. Highest correlation between CSF cAMP and pial arteriolar diameter was seen in group 1 (r2 = 0.38) but with a decreased but still significant correlation in group 2 (r2 = 0.30). No significant correlation was seen between the two variables in group 3.

Full size image

In group 1 at onset of asphyxia, RBC movement within the pial arterioles qualitatively appeared to increase initially, but, as asphyxia continued, RBC movement decreased, and cells became clumped together followed by complete cessation of RBC movement by the end of asphyxia. During recovery in group 1, initially some forward flow was noted but was transient, and ultimately RBC forward movement ceased and sludging was observed. Similar findings were observed during asphyxia in group 3 but with subsequent acceleration of RBC movement during the recovery phase; there was only minimal vessel engorgement. In group 2 piglets, however, persistent steady movement of RBC was observed throughout the whole period of asphyxia and recovery, with less engorgement and no intravascular RBC clumping.

After 60 min of recovery, determination of pial arteriolar response to isoproterenol showed significant increase in diameter in groups 2 and 3; this increase in diameter was significantly greater compared with group 1 (Fig. 5). The associated increase in CSF cAMP concentrations with pial dilatory response to isoproterenol is also shown in Figure 5. CSF cAMP concentrations increased significantly with isoproterenol administration in groups 2 and 3 but not in group 1.

Figure 5
figure 5

Percent change in pial arteriolar diameter and CSF cAMP concentrations to topical isoproterenol after 60 min of recovery from asphyxia. No significant change in diameter or CSF cAMP occurred in group 1. There were significant increases in pial arteriolar diameter and CSF cAMP in groups 2 and 3 from baseline (+p < 0.05). These percent changes were also significantly greater than changes observed in group 1 (○p < 0.05).

Full size image

DISCUSSION

We observed that compared with control piglets, pretreatment with sodium channel blocker resulted in maintaining higher BE and higher CSF cAMP levels, attenuation of pial arteriolar dilation, and preservation of forward movement of RBC during prolonged asphyxia. Treatment initiated at start of ventilation after asphyxia resulted in higher CSF cAMP concentrations during recovery compared with nontreated piglets. Furthermore, we observed preservation of cerebrovascular reactivity during recovery from asphyxia when sodium channel blocker was administered either before asphyxia or at the start of reventilation.

Different subtypes of sodium channel have been detected in all types of brain tissue as well as in vascular, cardiac, and skeletal muscle cells and are important for normal cellular function and metabolism (11, 12). A common mechanism of modulation of sodium channel activity on brain and vascular tissues is presumed to be cAMP-dependent phosphorylation (11, 13, 14). An increased level of cAMP reduces sodium influx in brain neurons, decreasing sodium channel number and activity and possibly attenuating excessive harmful sodium currents (15). Down-modulation of voltage-gated sodium channels is probably an effective way of reducing energy demands by decreasing the amount of energy used for maintenance of ionic gradient across the cell membrane (16, 17), and, therefore, it is possible that the metabolic requirement could be met by anaerobic metabolism (18, 19–22). However, if sodium channel stimulation is sustained during prolonged asphyxia, excessive sodium influx would block anaerobic glycolysis, thus abolishing this only remaining source of ATP. It follows, then, that anaerobic glycolysis may be maintained if sodium channel inhibitors are administered (23, 24). In pretreated piglets, we observed significantly less base deficit and a tendency for maintaining pH within an acceptable range even with associated significant decrease in mean arterial BP and PaO2 and increase in PaCO2 during prolonged asphyxia. These findings possibly reflect a better preservation of metabolism in general and in brain and vascular cell with administration of a nonselective sodium channel inhibitor. Perhaps with preservation of metabolism, there is also associated preservation of response to the other vasoactive stimuli other than cAMP during asphyxia. This, in part, may explain the decreased correlation between cAMP and vascular diameter in our pretreated animals (Fig. 4).

The changes in CSF cAMP in our control animals are consistent with previous reports (5–8, 25). Levels of CSF cAMP, however, were modified by administration of sodium channel blocker. Pretreatment with sodium channel blocker before asphyxia resulted in higher levels of CSF cAMP during prolonged asphyxia and during reventilation compared with nontreated animals. Treatment with sodium channel blocker at the end of asphyxia or the beginning of reventilation also resulted in significant elevation of CSF cAMP during recovery. Possible sources of cAMP in periarachnoid CSF are neurons, glial cells, and vascular endothelial and smooth muscle cells, and changes in cyclic nucleotide concentrations are likely reflective of intracellular changes (6, 7, 25–27). Vigorous intracellular cAMP generation during interruption of O2 supply coupled with ATP depletion has been described (7, 26). Also, cAMP is involved in the regulation of vasodilatory response to hypoxia-ischemia in newborn piglets (1, 8, 27). Cerebral circulation of newborn mammals has proven to be especially sensitive to the vasorelaxant action of cAMP (6–8, 27, 28), and topical application of cAMP to the cerebral surface was shown to cause pial arterial vasodilation ranging from 10 to 25% (8).

The role and function of sodium channels during hypoxia-ischemia when there are associated alterations in vascular hemodynamics need elucidation (2–4). Cerebral vasodilation appears to be a consistently observed phenomenon after hypoxia, hypercarbia, hypotension, and hypoxia-ischemia, both in experimental models and in human neonates (29–32). However, the physiologic significance of cerebral vascular arteriolar dilation in response to hypoxia-ischemia is not known. In experimental prolonged partial asphyxia with hypotension, a “no-reflow” phenomenon was observed during reventilation (5, 29, 33) despite cerebral dilation. Thus, cerebral vascular dilation does not necessarily indicate conservation of cerebral flow or autoregulation.

Low cerebral blood flow reported in term infants with hypoxic-ischemic encephalopathy during the first few days of life has been associated with poor outcome (34). However, in some studies, low cerebral blood flow with minimal oxygen requirement and oxygen consumption, as estimated from positron emission scan and near-infrared studies, appeared to be compatible with completely normal long-term neurologic outcome, as opposed to those infants who demonstrated increased oxygen consumption during the posthypoxic-ischemic period (19–22, 34–37). Perhaps limited or minimal cerebral vascular dilation with some conservation of blood flow may provide protection against breakdown in the blood-brain barrier in the course of hypoxia-ischemia. Compared with control animals in our study, pretreatment resulted in preservation of forward movement of RBC during prolonged asphyxia in the presence of minimal arteriolar dilation. Thus, there is a potential role for sodium channel blockers in maintaining blood flow during prolonged hypoxia-ischemia. The marked pial arteriolar dilation during asphyxia and loss of vasoreactivity during recovery in our control group probably reflect vascular paralysis as a result of hypoxia-ischemia. In contrast, sodium channel blocker administration either before asphyxia or in early reventilation was associated with preservation of vascular reactivity.

We can only speculate on the possible mechanisms by which sodium channel inhibition is associated with attenuation of cerebral vascular alterations and elevation of cAMP during hypoxia-ischemia. At the beginning of asphyxia, with perfusion pressure and O2 supply relatively preserved, a quickly rising PCO2 or hypercapnic response (27) results in CSF cAMP elevation with moderate physiologic cerebral vascular dilation. Early in asphyxia, this increase in cAMP also may lead to sodium channel down-regulation as one of the mechanisms of local metabolic vascular motor control, limiting vessels from further dilation, vascular paralysis, and disruption of the blood-brain barrier. However, with longer duration of hypoxia-ischemia, these adaptive mechanisms eventually fail, coinciding with a precipitous decrease in cAMP level in intracellular as well as extracellular fluids and associated with cerebral vascular paralysis. These changes were evident in the nontreated piglets throughout the whole experiment and during asphyxia in piglets who were treated at initiation of reventilation. Despite larger arteriolar lumen in the nontreated piglets, no reflow phenomenon was observed in our experiments as well as in other similar studies (5, 33, 4). Sodium channel blocker administration resulting in a marked increase in CSF cAMP levels followed by a delay in its precipitous decrease maintained arterial vascular tone associated with conservation of blood flow during prolonged asphyxia and preservation of physiologic response to other vasoactive stimuli during recovery or reventilation. Thus, from our findings in the first experiment of sodium channel blocker administration in the newborn large mammal model of neonatal asphyxia, pharmacologic modulation of sodium channels may have a role in therapeutic or preventive management in newborn hypoxic-ischemic injury.