Main

Investigations using neuronal cell culture and animal models suggest a potential neuroprotective role for magnesium administration under conditions of hypoxia-ischemia. Magnesium, a noncompetitive antagonist of the N-methyl-D-aspartate receptor, has been studied as a potential neuroprotective agent because it prevents the toxicity of intraparenchymal administration of excitatory neurotransmitter receptor agonists in adult and neonatal animals(1,2). Investigations in adult animals support a role for systemically administered magnesium in attenuating brain injury from trauma(3) and stroke(4). In neonatal animals, i.p. administration of magnesium, in conjunction with scavengers of oxygen free radicals, appeared to ameliorate perinatal hypoxic-ischemic brain damage in rat pups(5). Retrospective studies of very low birth weight infants suggest a reduced incidence of cerebral palsy associated with exposure to magnesium sulfate (MgSO4) administered to the mother before delivery(6,7). In all these studies it was assumed that systemically administered magnesium enters the brain ECF, where it provides neuroprotection by interacting with N-methyl-D-aspartate receptors.

Knowledge of the effect of raising systemic magnesium concentration on both the rate of change and absolute increase in magnesium concentration around brain cellular elements is important for the development of protocols using MgSO4 infusions to therapeutically reduce brain damage. Equally important to the development of protocols specific for neonates is whether the effects of raising systemic magnesium concentration on brain ECF magnesium concentration differ in the newborn compared with the mature subject. Although age-related increases in CSF magnesium concentration with systemic magnesium administration have been observed in swine during the first month of life(8), it is unclear how this relates to magnesium concentrations in brain interstitial [Mg]ECF. The purpose of this study was to characterize and compare the entry and clearance of systemically administered MgSO4 into the brain ECF of newborn and more mature miniswine using cerebral microdialysis. Specifically, relative changes in [Mg]ECF and rates of magnesium entry and clearance were compared in two age groups for similar increases in [Mg]plasma. Because magnesium administration may alter cardiovascular hemodynamics(9,10), MAP and HR were monitored to characterize and compare the systemic effects associated with producing plasma magnesium concentrations of 4-6 mM in the two age groups. A preliminary report of this work has been reported in abstract form(11).

METHODS

Experimental protocol. The study protocol was approved by the Institutional Animal Care and Research Advisory Committee of the University of Texas Southwestern Medical Center at Dallas. A total of 18 miniswine was studied: nine newborn (age 4 ± 2 d, weight 1.3 ± 0.4 kg, mean ± SD) and nine more mature (approximately 1 month old) (age 39 ± 7 d, weight 6.3 ± 1.2 kg). Seven animals in each age group were infused with MgSO4 and two were infused with Na2SO4, as described below. Animals were initially sedated with ketamine (20 mg/kg i.m.), followed by local skin infiltration of the neck with 1% Xylocaine, rapid tracheotomy, and mechanical ventilation using a 30:70 oxygen:nitrous oxide mixture. Animals were provided analgesia with nalbuphine (0.15 mg/kg i.m.) and muscle relaxation with tubocurarine Cl (0.1 mg/kg i.v.). Rectal temperature was maintained at 38-39°C using a heating pad. Catheters were placed in the external jugular vein and carotid artery. After local scalp infiltration with 1% Xylocaine and administration of sodium pentothal (20 mg/kg i.v.), a burr hole was made and a loop-type microdialysis probe was inserted to a depth of 1 cm into the cerebral cortex and situated 1 cm anterior and 1 cm lateral to the coronal and sagittal cranial sutures, respectively, for measurements of brain [Mg]ECF. Animals were then allowed to stabilize for 3.5 h before obtaining baseline values of HR, MAP, hematocrit, [Mg]plasma, and [Mg]ECF over a 30-min period.

A target [Mg]plasma range of 4-6 mM was selected based on preliminary studies showing this to be the highest concentration of magnesium that could be tolerated without detrimental physiologic side-effects. At the same time, this range maximized the magnitude of any potential increases in [Mg]ECF that could be measured using microdialysis. MgSO4 (30 mg of anhydrous powder per mL of H2O; Sigma Chemical Co., St. Louis, MO) was infused i.v. at a dose of 275 mg/kg body weight for newborn and 200 mg/kg for 1-month-old swine for 30 min, followed by 100 and 75 mg/kg, respectively, for an additional 30 min in an effort to maintain [Mg]plasma at constant levels. Preliminary studies established that slightly more MgSO4 was required to achieve the target [Mg]plasma in newborns compared with 1-month-olds. Two additional 1-month-old swine were infused with Na2SO4 (30 mg of anhydrous powder per mL; Sigma) at 200 mg/kg/30 min, followed by Na2SO4 at 75 mg/kg/30 min. Two newborn swine were also infused with Na2SO4 at 275 mg/kg/30 min, followed by Na2SO4 at 100 mg/kg/30 min. HR and MAP were monitored continuously, with recordings made every 30 min. Arterial pH and blood gases were sampled every 30 min. [Mg]plasma was measured every 15 min beginning at the start of the infusion protocol. The decision to collect plasma samples at 15-min intervals was a compromise between maximizing the resolution for the following changes in magnesium versus the need to minimize physiologic changes that could occur with an excessive amount of blood withdrawn. Packed red blood cells removed for these measurements were readministered hourly via the carotid catheter. Changes in [Mg]ECF were monitored by collecting consecutive samples every 10 min via microdialysis as described below. At the end of the 1-h infusion period the animals were studied an additional 4 h with the same sequence of data collections. At the end of the protocol the animal was killed (sodium pentobarbital, 200 mg/kg, i.v.).

Microdialysis measurements. Loop-type microdialysis probes were constructed of commercially available hollow fiber dialysis tubing (ID = 150 µm, molecular weight cut-off = 17 kD; Spectrum Medical Industries Inc., Los Angeles, CA) formed into a 2-cm long loop and given mechanical stability by running a 100-µm diameter nylon thread through it. The effluent end of the fiber was attached to low volume tubing (1.2 µL per 10 cm) using cyanoacrylate-based glue; the inflow and effluent tubing (PE-20) were held together with a collar made from PE-260 tubing. After determination of the dead space from the probe to the effluent collection opening, probes were infused with artificial CSF (154 mM NaCl, 1.3 mM CaCl2, pH 7.2) at a rate of 5 µL/min and calibrated by dialysis against a standard solution with a known concentration of MgSO4 (154 mM NaCl, 1.3 mM CaCl2, and 1.0 mM MgSO4). Recovery factors were determined from the ratio of total magnesium concentrations in the effluent and standard solutions and the factor was then used to calculate the apparent [Mg]ECF from microdialysis fluid. Microdialysis data are presented at the midpoint of 10-min collection periods and corrected back to real time by using the dead space determinations for each probe (i.e. 6-8 µL or 1.2-1.6 min). Throughout the experiment, brain [Mg]ECF microdialysis samples of 50 µL volume were collected continuously every 10 min. At the end of several experiments, postmortem examination of the brain verified that dialysis sampling occurred primarily from the gray cerebral matter of the brain at a depth of 1 cm.

[Mg] measurement. Typically 45 µL of microdialysis fluid and 10 µL of blood were used for total magnesium determination at 285.2 nm via atomic absorption spectroscopy (Spectra AA, Varian Associates Inc., Palo Alto, CA) using an air/acetylene flame. Lanthanum oxide (0.5% w/v; Sigma) was added to reduce the potential for interferences from the incomplete dissociation of compounds in blood or microdialysis fluid, as recommended in the Varian analytical methods manual. Magnesium standard curves were prepared from stock MgSO4 solutions.

Data analysis. All systemic data (HR, MAP, pH, blood gases, and hematocrit), [Mg]plasma, and brain [Mg]ECF for each animal age group were analyzed by two-way repeated measures analysis of variance. Group interactions were further evaluated and localized by pair-wise comparisons (Student-Newman-Keuls). To facilitate and simplify the analysis of [Mg]plasma and [Mg]ECF compared with baseline, data collected in the postinfusion interval were grouped into 1-h blocks. To quantify the kinetics of increase in [Mg]plasma and [Mg]ECF during MgSO4 administration, and subsequent decrease after administration, half-lives for magnesium increase (t1/2 uptake) and decrease (t1/2 clearance) were calculated. The half-lives were determined directly from plots of the time course of changes in [Mg]plasma or [Mg]ECF versus time, as illustrated in Figure 1. [Mg]plasma and [Mg]ECF t1/2 values for uptake and clearance were compared between groups using unpaired t tests. p < 0.05 was considered significant. All data are expressed as mean ± SD.

Figure 1
figure 1

Calculation of t1/2 uptake and t1/2 clearance values from a plot of [Mg]ECF versus time for one animal. Times of the two different MgSO4 infusion rates are indicated by boxes located on the horizontal axis. The rate of increase and decrease in [Mg] were determined in a similar manner for [Mg]plasma.

Full size image

RESULTS

Baseline values for [Mg]plasma in groups infused with MgSO4 were the same for newborn (0.59 ± 0.22 mM) and 1-month-old (0.57 ± 0.11 mM) swine. During the infusion, [Mg]plasma rose to 5.72 ± 1.08 mM at 30 min and 5.39 ± 1.08 mM at 60 min of infusion for newborns and to 5.27 ± 0.76 and 4.43 ± 0.60 mM at 30 and 60 min of infusion, respectively for 1-month-old miniswine (p < 0.05 versus control for both groups at both times) (Fig. 2). Values for [Mg]plasma during MgSO4 infusion were not different between groups. From 30 to 60 min of infusion, [Mg]plasma did not change for newborns, but decreased for 1-month-old swine by the end of the infusion (p < 0.05). All values for [Mg]plasma after the MgSO4 infusion remained higher than control throughout the remainder of the experiment (4 h postinfusion). T1/2 uptake was approximately 2 min faster in 1-month-olds compared with newborns (Table 1). After the infusion, magnesium was cleared twice as fast for 1-month-old miniswine. As indicated in the upper panel of Figure 2, a group-time interaction (p < 0.0001) was present, reflecting the more rapid increase and decrease in [Mg]plasma for 1-month-old swine.

Figure 2
figure 2

Time course of changes in [Mg]plasma (upper) and [Mg]ECF (lower) for 1-month-old (squares) and newborn (circles) miniswine. Data are shown as mean ± SD. Due to differences between groups in baseline values for [Mg]ECF (see text), results are presented as percent change from baseline (latter set at 100%). The period of MgSO4 infusion is denoted by the thick horizontal bar. Brackets delineate grouping of postinfusion data for analysis. Asterisks indicate points that were different from control (p < .05) and crosses indicate that a group-time interaction was present (p < 0.0001 for [Mg]plasma and p = 0.0001 for [Mg]ECF).

Full size image
Table 1 Half lives for uptake and clearance of magnesium in plasma and brain ECF
Full size table

Baseline values for [Mg]ECF were calculated from the average of three measurements made during the 30-min baseline period. Values at control differed between age groups (0.23 ± 0.08 mM for newborn versus 0.15 ± 0.05 mM for 1-month-old, p < 0.05). Absolute peak [Mg]ECF values were not different between groups (0.40 ± 0.17 mM for newborn versus 0.35 ± 0.10 mM for 1-month-olds). To compare relative changes in [Mg]ECF for the two age groups, [Mg]ECF measurements obtained during the infusion and postinfusion period were expressed as a percentage relative to the baseline value obtained immediately before the infusion (Fig. 2). In both age groups, the infusion of MgSO4 resulted in significant increases in [Mg]ECF compared with control. Maximal values for [Mg]ECF rose to a similar extent; 253 ± 106% of control at 5 min after completion of the infusion for 1-month-old swine compared with 193 ± 76% of control at 25 min after completion of the infusion for newborns (p < 0.05 versus control for both groups). The t1/2 uptake for [Mg]ECF was more rapid for 1 month olds compared with newborns (Table 1), but t1/2 clearance values did not differ between age groups. As indicated in the bottom panel of Figure 2, a group-time interaction (p = 0.0001) was present and reflected the more rapid increase in [Mg]ECF in 1-month-old compared with newborn swine.

For the baseline period the MAP (mean of two measurements at t = -30 and t = 0 min) for 1 month olds was significantly higher than for newborns (p < 0.0001, unpaired t test) although HR did not differ (Fig. 3). During the infusion of MgSO4, MAP and HR for both age groups decreased. Expressed as a percentage relative to baseline, the decreases in MAP (approximately 20%) and HR (approximately 15%) during the infusion were the same in both age groups. When the MgSO4 infusion was terminated, MAP and HR gradually increased, although in both age groups, MAP remained lower than baseline at the end of the 4-h postinfusion period and HR was lower than baseline for 1 month olds.

Figure 3
figure 3

Time course of changes in MAP (upper) and heart rate (lower) for 1-month-old (squares) and newborn (circles) miniswine. Data are shown as mean ± SD. The period of MgSO4 infusion is denoted by the thick horizontal bar. Brackets delineate grouping of postinfusion data for analysis. Asterisks indicate points that were different from control (p < 0.05) and the cross indicates that a group-time interaction was present in the heart rate data (p < 0.0001).

Full size image

Arterial pH, blood gases, and hematocrit measured at selected times are listed in Table 2. All variables were the same between groups, except for expected age-related differences in hematocrit. At the completion of the MgSO4 infusion and at the completion of the study (post-Mg + 210 min), all variables were unchanged except for a small increase in pH in 1-month-old miniswine at post-Mg + 210 min.

Table 2 Arterial pH, blood gases, and hematocrit before, during, and after MgSO4 infusions
Full size table

The possible effects of Na2SO4 administration were explored for two animals of either age. No perturbations in hemodynamic parameters, blood gases, and pH were observed during the infusion. For all four animals, the [Mg]plasma at control was 0.25 ± 0.05 mM and was no different at 30 and 60 min of infusion. [Mg]ECF in the Na2SO4 infused animals of either age was similar to MgSO4-infused animals at baseline and also did not increase during the infusion.

DISCUSSION

This investigation demonstrates that increases in [Mg]plasma to 4-6 mM result in an elevation of brain [Mg]ECF. Baseline values of brain [Mg]ECF were higher in newborns, but the extent of increase in brain [Mg]ECF was similar between newborn and 1-month-old miniswine. The experimental protocol was designed to produce a large rapid increase in [Mg]plasma in the target range of 4-6 mM for both age groups and then maintain these levels for a 60-min period. For the most part this goal was achieved, although a slight decrease in [Mg]plasma from 30 to 60 min was observed for 1 month olds. Our underlying assumption was that a 60-min period would be sufficient to reach a constant elevated [Mg]plasma, and therefore the relative increases in [Mg]ECF for newborn and 1 month olds could be compared under identical steady-state conditions. Although constant levels of [Mg]ECF appear to be reached by approximately 30 min in 1 month olds, for newborns there was a continued increase in [Mg]ECF throughout the MgSO4 infusion. These age-related differences are reflected quantitatively in the t1/2 values (Table 1), with newborns having an 18-min longer t1/2 uptake compared with 1-month-old swine. The rate of clearance of magnesium from the brain was not different between age groups. Maturational differences were also observed in the rate of increase and decrease of [Mg]plasma. The plasma t1/2 clearance is 40 min longer in newborn than 1-month-old swine. In contrast, the difference in t1/2 uptake was less than 2 min, and given the 15- min time resolution for acquiring plasma samples, represents a minor group difference that does not explain the slower uptake of magnesium into ECF for newborns compared with 1 month olds. Factors that could effect maturational differences in [Mg]plasma include a greater volume of distribution for magnesium in the newborn group, and increased renal clearance of magnesium in 1-month-old miniswine. These factors presumably account at least in part for newborns requiring larger doses of MgSO4 to achieve similar peak [Mg]plasma compared with 1-month-old miniswine.

It is noteworthy that baseline values of [Mg]ECF for both newborn and 1-month-old miniswine were approximately 0.2 mM whereas literature values for total magnesium concentration in neonatal swine CSF range from 0.45 to 1.48 mM(8) and in humans from 1.0 to 1.2 mM(12) and does not appear to vary with age(13). This discrepancy in baseline [Mg]ECF may be intrinsic to perfusion with magnesium-free dialysis fluid, leading to drainage of endogenous magnesium and creation of a gradient of low extracellular magnesium in the immediate vicinity of the probe, similar to that hypothesized for calcium(14). There are two additional considerations. First, roughly 25% of the magnesium present in cerebrospinal fluid is bound to high molecular weight compounds, which would be unable to diffuse across a microdialysis membrane(15). A portion of the remaining magnesium is likely complexed to lower molecular weight ligands that may diffuse more slowly compared with ionized magnesium. A second issue is that the in vitro calibration as used does not accurately predict in vivo recovery due to inherent differences in the diffusibility of ionized magnesium in ECF versus calibration fluid(16). Although our measurement of relative changes in [Mg]ECF and the rate of entry and clearance does not require knowledge of absolute [Mg]ECF, it does assume that probe recovery remains constant throughout the protocol. Potential differences in recovery factors between probes would not invalidate comparisons of relative increases in [Mg]ECF in newborn versus 1-month-old miniswine or of rates of magnesium entry and clearance as reported.

A critical assumption in using microdialysis is that probe insertion does not disrupt the brain microenvironment sufficiently to affect the concentration or transport properties of the substance under investigation(17). Although the reliability of chronic microdialysis has been questioned(17), short-term studies examining BBB permeability have demonstrated that the BBB is intact 30 min to 1 h after probe insertion(18–20). Probe implantation causes the release of intracellular ions and metabolites, necessitating a 30-min period before the collection of baseline data(21). In our study, the protocol was commenced 3.5 h after probe implantation, and the baseline [Mg]ECF was constant for 30 min before infusion of MgSO4. Even if the BBB does not regain its original properties shortly after probe implantation, as contended by some(22,23), the comparison of relative magnesium entry and clearance between the two groups may still be valid. Administration of a hyperosmolar solution can increase the permeability of the BBB, but the threshold osmolarity of a bolus intracarotid infusion to cause BBB breakdown is approximately 1.6 M(24,25). The molar concentrations of the infused MgSO4 and Na2SO4 solutions were far less (0.25 M) and BBB integrity should not be altered.

A recent study reported that the i.p. injection of MgSO4 did not change brain [Mg]ECF as measured by intrastriatal microdialysis in the rat(26). However, the increase in [Mg]plasma was small (peak [Mg]plasma = 1.56 mM)(26). Our observation that an elevation of brain [Mg]ECF occurs when [Mg]plasma is increased is qualitatively consistent with studies of CSF magnesium concentration, although the magnitude of change in the CSF is smaller than our finding for ECF. For example, Oppelt et al.(27) observed that systemic administration of MgSO4 and MgCl2 was associated with maximal increases in the CSF magnesium concentration of only 21% despite elevations of [Mg]plasma that were 275-360% above control (maximal increases of approximately 2.6 mM). Small but significant increases in CSF magnesium concentration to 1.25 ± 0.05 mM have been demonstrated with systemic administration of MgSO4 to preeclamptic women when compared with control group values of 1.05 ± 0.08 mM(28). Rivera et al.(8) reported that increases in [Mg]plasma in neonatal swine increased CSF magnesium with the largest changes in younger animals, and less prominent changes with increasing age. A similar maturational difference in the extent of peak brain [Mg]ECF was not evident in the present study, and differences across age were evident only in the more rapid rate of increase in brain [Mg]ECF in 1 month olds. It may be difficult to directly compare changes in brain [Mg]ECF and CSF magnesium concentration because the composition of CSF is a complex combination of diffusion or secretion across the choroid plexus epithelium and of secondary exchanges between the ECF and the brain parenchyma(29).

The changes in MAP and HR observed with increases in [Mg]plasma to 4-6 mM are consistent with changes observed previously in swine(9,10). There were no differences between age groups in the extent of changes in HR and MAP during infusion of MgSO4. Both age groups demonstrated only partial recovery of hemodynamic parameters by 4 h postinfusion, with the heart rate of 1 month olds decreasing at study completion and giving rise to a group-time interaction. Arterial pH displayed a small increase in the 1-month-old animals of 0.06 units, which was within the physiologic range and indicated that the alterations in hemodynamic parameters caused by the MgSO4 infusions were well tolerated by the animals.

Investigations into mechanisms by which magnesium may serve as a neuroprotective agent have focused primarily on its ability to block excitatory neurotransmitter receptors and thus prevent the toxicity triggered by the excessive neurotransmitter release associated with ischemia or hypoxia-ischemia(1–5). Our study has demonstrated that increases in systemic magnesium concentration result in increases in brain [Mg]ECF, and places magnesium in an environment whereby interactions can readily occur at the neuronal cell surface. Magnesium, however, has other important effects that may ultimately influence the utility of magnesium as a neuroprotective agent. The significant decreases in MAP and HR indicate that further study is necessary to determine whether the magnitude of changes in systemic hemodynamics can alter the brain energy state. Magnesium is a vasoactive compound and may directly influence cerebral blood flow independent of changes in HR and blood pressure(30). The time course of brain [Mg]ECF for both age groups indicates that magnesium concentration declined over time; the pharmacologic kinetics of magnesium distribution may have implications regarding the need for continuous infusion or intermittent dosing of magnesium to maintain therapeutic levels in the brain ECF. Finally, further experiments are required to determine whether the magnitude of change in brain [Mg]ECF would be the same or less if [Mg]plasma were increased only to 2-3 mM, as is commonly used in clinical practice.