Red blood cell magnesium and hypoxic-ischaemic encephalopathy

Early Human Development 47 (1997) 287–296

Red blood cell magnesium and hypoxic-ischaemicencephalopathy

*Vincent Harrison , Gillian Peat

University of Cape Town, Department of Paediatrics and Child Health, Cape Town, South Africa

Received 29 March 1996; revised 21 June 1996; accepted 1 July 1996

Abstract

In developing countries, birth asphyxia is frequently associated with hypoxic ischaemicencephalopathy. This has been attributed to inadequate obstetric care but poor nutrition mayalso be important. This study determines the association between magnesium stores andhypoxic ischaemic encephalopathy. The level of red blood cell magnesium was measured on572 women in labour and on selected offspring in a teaching hospital in South Africa. Fifty fiveof the 572 women delivered infants with hypoxic ischaemic encephalopathy and hadsignificantly lower red blood cell magnesium levels (1.40 mmol / l) than controls. In the latterthe levels varied somewhat with the mode of delivery, vertex births 1.76 mmol / l, Caesareansections 1.67 mmol / l and vacuum extractions 1.61 mmol / l. Infants with hypoxic ischaemicencephalopathy had a significantly lower red blood cell magnesium level (1.39 mmol / l) thannormal infants (1.61 mmol / l). Fifty four of 55 babies were black and from poor socialcircumstances and nutritional deficiency may be relevant. Maternal height, age and the durationof labour did not influence the chance of hypoxic ischaemic encephalopathy and affectedinfants were more likely than normal ones to be meconium stained (50%), to have a low Apgarscore (58%) and to need endotracheal intubation at birth (54%). An intervention study in earlypregnancy may determine magnesium’s role in hypoxic ischaemic encephalopathy associatedwith asphyxia. 1997 Elsevier Science Ireland Ltd. All rights reserved

Keywords: Magnesium; Red blood cells; Hypoxic ischaemic encephalopathy; Pregnancy

*Corresponding author. Department of Paediatrics and Child Health, Institute of Child Health, Red CrossWar Memorial Children’s Hospital, Rondebosch 7700, South Africa. Tel: 1 27 21 6853026. Fax: 1 27 216891287.

0378-3782/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reservedPII S0378-3782( 96 )01787-2

288 V. Harrison, G. Peat / Early Human Development 47 (1997) 287 –296

1. Introduction

In the industrialised world intrapartum asphyxia is an infrequent cause of cerebralpalsy in term babies [1]. In rural South Africa cerebral palsy has been attributed tobirth asphyxia in over 30% of cases [2]. Spastic quadriplegia predominates presumab-ly from hypoxic-ischaemic damage to the brain. Hypoxic-ischaemic encephalopathy(HIE) is considered to be the best predictor of neurological handicap associated withasphyxia [3].

The incidence of HIE declines as socio-economic status improves [4]. This can beascribed to skillful obstetrics but it may also result from better nutrition. Intrapartuminsults are more likely to injure the brain of a vulnerable infant and the cerebrum ofan inadequately nourished fetus may be particularly susceptible to ischaemia andhypoxia. Magnesium is considered to protect the brain from cerebral palsy in verypremature infants [5] and we have chosen to study this cation in term pregnantwomen and in their offspring with HIE.

Our case study determines the amount of magnesium in the red blood cells (RBC)of women in active labour and compares the levels of those who had normal babieswith those who had HIE babies. The red blood cell is an acceptable source of storedmagnesium which is unaltered by changes in plasma volume and remains constantthroughout pregnancy [6].

2. Materials and methods

The study was conducted at Mowbray Maternity Hospital, a Teaching Unit of theUniversity of Cape Town. Most women are referred from midwife obstetric unitsbecause of complications in pregnancy and labour. Many are recent arrivals fromrural areas, one third have received little or no antenatal care and most are of lowsocio-economic status.

2.1. Mothers

Investigations were done on 576 healthy women in labour and 4 of these wereexcluded later because of Group B streptococcal sepsis in their infants. Theseconsecutive deliveries also excluded prematurity, multiple pregnancy, maternal illness(diabetes, cardiovascular disease, intrapartum infection) and those on medicationswhich might alter the neurological state of an infant such as magnesium sulphate,tranquilisers and sedatives. Obvious acute causes of asphyxia (abruptio, prolapsedcord) were not included. Of the 572 women, 380 were black, 6 white and 186 ofmixed race.

At the first antenatal visit maternal height was measured in centimeters and inlabour the uterine contractions, fetal heart rate, maternal pulse rate, blood pressureand temperature were checked every half-hour. Progress was evaluated at two tofour-hour intervals and the details were charted on a partogram. Active labour was

V. Harrison, G. Peat / Early Human Development 47 (1997) 287 –296 289

signaled by cervical dilatation (more than 3 cms) and three successive uterinecontractions (each over 40 s) within 10 min. The duration of labour was measuredfrom this point to the birth of the baby. The fetal heart rate pattern was monitoredthroughout active labour with an abdominal ultrasound transducer (Hewlett Packard8040A). The presence or absence of meconium was recorded and infant scalp bloodwas obtained for acid base measurements if fetal distress occurred. A pH of 7.15 orless necessitated immediate delivery.

2.2. Infants

Apgar scores were calculated at 1 and 5 minutes of age, weight was measured tothe nearest 10 g and head circumference was recorded in centimeters. Gestational agewas derived from physical characteristics [7].

Muscle tone, reflex activity and level of consciousness were assessed within 24 hand infants with characteristic signs of HIE [8] were re-examined hourly and gradedmild, moderate or severe on their worst features before anticonvulsant treatment orassisted ventilation. Asphyxiated babies were resuscitated at birth and those whoneeded anticonvulsants, intravenous fluid or intermittent positive pressure ventilationwere treated in an intensive care unit. A brain scan (ultrasound) was done on thosewith HIE within 5 days and repeated at 2 weeks of age. Survivors were assessed at afollow-up clinic.

Fifty-nine of 576 infants had signs of encephalopathy within 24 h of birth. Four ofthese were rejected because of Group B streptococcal sepsis and the remaining 55babies were studied. Of these 18 were graded as mild, 20 as moderate and 17 assevere HIE.

2.3. Magnesium

A 3 ml sample of venous blood (Lithium heparin) was obtained from each womanin active labour before full dilatation and centrifuged within 4 h. The plasma andupper layers of red blood cells were removed and the remaining pellet of older cellswas analysed. Magnesium was measured after deproteinisation with sodium tungstateat a low pH using a calmagite method (Mg-Erythrocytaire Kit, bioMeriex, Marcy,France). The absorbance of standards, controls and samples was determined induplicate at a wavelength of 500 nm using a Hitachi U1000 spectrophotometer andthe inter- and intra-assay errors were less than 5%.

Magnesium was measured in the red blood cells of infants with HIE (1 ml venousblood) and the results were disclosed after the infants had been graded. A sample ofvenous blood was also obtained from 43 control infants who required investigationfor serological status, jaundice or presumed infection. They were subsequentlycertified normal and not treated.

Data was analysed using confidence interval analysis (Gardner and Altman) andKruskal-Wallis one way analysis of variance.


Informed consent was obtained from each woman and the study was approved bythe Medical Ethics Committee, University of Cape Town.

3. Results

3.1. RBC magnesium (Table 1)

Levels were measured on 572 women. Of these 517 delivered normal babies andserved as controls. They were divided into normal vertex deliveries (NVD),Caesarean sections (CS) and vacuum extractions (VAC) because of an associationbetween magnesium levels and the mode of delivery. The NVD had higher levelsthan the CS (mean difference 0.09 mmol / l; 95% CI 0.03–0.15) and VAC (meandifference 0.11 mmol / l, 95% CI 0.02–0.2).

The HIE study group of 55 was combined (Fig. 1) as magnesium levels werecomparable irrespective of the type of delivery (p 0.33). Nevertheless values for thevarious modes of HIE deliveries are shown in Table 1. This group had significantlylower levels than the NVD control (mean difference 0.35 mmol / l, 95% CI 0.30–0.39), CS control (mean difference 0.26 mmol / l. 95% CI 0.21–0.31) and VACcontrol (mean difference 0.23 mmol / l, 95% CI 0.18–0.29).

In HIE babies the mean RBC magnesium of 1.39 mmol / l (1.05–1.49) wassignificantly less than that in normal babies 1.61 mmol / l (range 1.43–1.78, meandifference 0.22 mmol / l, 95% CI 0.16–0.27).

Fifty-four of 55 babies with HIE were black. Their mothers had significantly lowerlevels than control black mothers consisting of 160 NVD (mean difference 0.34mmol / l, 95% CI 0.29–0.39) and 130 CS (mean difference 0.24 mmol / l, 95%0.20–0.29) and 36 VAC (mean difference 0.23 mmol / l, 95% CI 0.17–0.29).

The black control levels were also lower than those of other ethnic groups (meandifference 0.07 mmol / l, 95% CI 0.04–0.10).

Plasma magnesium (Table 1): Levels were comparable in control and HIE groups(p 0.14).

Table 1Magnesium levels (mmol / l) in labour. The mean (standard deviation), range and confidence intervals (CI)are presented

Controls Hypoxic-ischemic encephalopathy

NVD CS VAC NVD CS VAC

Number 267 194 56 31 17 7RBC magnesium 1.76 (0.16) 1.67 (0.15) 1.64 (0.18) 1.42 (0.09) 1.43 (0.09) 1.36 (0.07)Range 1.41–2.36 1.32–2.29 1.31–2.25 1.05–1.53 1.25–1.56 1.22–1.4195% CI 1.74–1.77 1.65–1.69 1.59–1.69 1.38–1.46 1.38–1.46 1.29–1.42Plasma magnesium 0.80 (0.10) 0.80 (0.10) 0.80 (0.09) 0.79 (0.08) 0.80 (0.10) 0.76 (0.09)Range 0.58–1.29 0.62–1.11 0.62–1.09 0.63–0.97 0.66–0.97 0.65–0.9495% CI 0.78–0.81 0.78–0.83 0.75–0.82 0.75–0.82 0.74–0.86 0.65–0.86

NVD, Normal vertex delivery; CS, caesarean section; VAC, vacuum extraction; RBC, red blood cell.


Fig. 1. Maternal red blood cell levels of magnesium in labour. Ctrl, Control; NVD, normal vertex delivery;CS, Caesarean section; VAC, vacuum extraction; HIE, Hypoxic-ischaemic encephalopathy. The box coversthe middle 50% of the data values, between the 25th and 75th quartiles. The ‘whiskers’ extend out to theminimum and maxium values. The points are unusual values far from the bulk of the data. The horizontalbar is at the median. The notch is the width of the 95% confidence interval for the median. When notchesdo not overlap the means differ statistically at the 95% level (P , 0.05). the width of the box isproportional to the square root of the number of observations in the group.

3.2. Maternal factors (Table 2):

3.2.1. Duration of labourThe control NVD had shorter labours than the control CS (mean difference 155

min, 95% CI 10.3–320) and control VAC (mean difference 253 min, 95% CI 2-506)and HIE CS group (mean difference 261 min, 95% CI 2 198-717). The control NVD,HIE NVD and HIE VAC groups did not differ significantly (p 0.11). The effect of theduration of labour on outcome is uncertain because of wide mean changes.

3.2.2. Height and age (Table 2)These did not seem to influence the results (height p 0.4 and age p 0.36).

3.2.3. LiquorThis was stained with meconium in more of the HIE pregnancies (50%) than the

controls (NVD 27%, CS 34%, VAC 37%).

3.3. Infant factors (Table 3)

3.3.1. 1 min ApgarScores of 3 or less occurred more frequently in HIE babies (58%) than in controls

(NVD 2.2%, CS 18%, VAC 7%). The HIE infants (54%) were also more likely toneed endotracheal intubation at birth than controls (NVD 1.4%, CS 18%, VAC 3.5%).


Tab

le2

Mat

erna

lfac

tors

inhy

poxi

c-is

chae

mic

ence

phal

opat

hy.

Mea

n(s

tand

ard

devi

atio

n)ra

nge

and

95%

CI

pres

ente

d

CO

NT

RO

LS

HIE

NV

DC

SVA

CN

VD

CS

VAC

Hei

ght(

cm)

157

(6.3

)15

5(5

.2)

157

(6.8

)15

3(4

.5)

153

(6.8

)15

6(5

.6)

Ran

ge13

7–18

814

0–16

914

3–17

314

2–16

214

4–16

514

6–16

295

%C

I15

6–15

815

5–15

615

5–15

915

1–15

514

9–15

715

1–16

1A

ge(y

rs)

26.2

(6.4

)26

.4(6

.3)

23.5

(5.2

)22

.1(6

.5)

25.8

(7.7

)19

.1(3

.3)

Ran

ge14

–43

15–4

215

–36

13–4

217

–40

14–2

395

%C

I25

.5–2

725

.5–2

7.2

22.1

–24.

919

.7–2

4.5

21.8

–29.

816

.1–2

2.2

Lab

our(

min

)56

9(3

38)

725

(581

)82

3(3

68)

688

(258

)83

1(4

81)

833

(264

)R

ange

60–2

235

60–4

800

212–

2090

222–

1082

360–

1900

530–

1320

95%

CI

529–

611

640–

810

725–

922

592–

785

565–

1098

588–

1077

Tab

le3

Infa

nts

fact

ors

inhy

poxi

c-is

chae

mic

ence

phal

opat

hy.

Mea

n(s

tand

ard

devi

atio

n)ra

nge

and

95%

CI

pres

ente

d

Con

trol

sH

IE

NV

DC

SVA

CN

VD

CS

VAC

Wei

ght

(g)

3252

(521

)34

45(4

83)

3326

(505

)32

52(4

24)

3355

(486

)30

94(2

52)

Ran

ge19

40–4

920

2040

–524

021

00–4

460

2200

–414

023

40–4

320

2700

–344

095

%C

I31

90–3

316

3377

–351

531

91–3

462

3096

–340

831

06–3

606

2861

–332

7A

pgar

(1m

in)

8(1

.6)

6.6

(2.6

)7.

1(2

.1)

3.1

(2.0

)4.

4(3

.4)

4.8

(2.4

)R

ange

1–10

1–10

1–10

1–9

1–10

1–8

95%

CI

7.9–

8.3

6.3–

7.1

4.4–

102.

4–3.

92.

7–6.

32.

3–6.

8A

pgar

(5m

in)

9.5

(0.8

)8.

9(1

.5)

9.2

(1.2

)6.

0(2

.4)

7.1

(2.4

)7.

5(1

.1)

Ran

ge4–

103–

102–

102–

103–

106–

995

%C

I9.

4–9.

68.

8–9.

28.

9–9.

65.

2–7.

05.

9–8.

56.

5–8.

6


Within the controls the NVD scored better than the CS (mean difference 1.3, 95%CI 0.5–2.1) and VAC (mean difference 0.8, 95% CI 2 0.4–2.1).

3.3.2. 5 min ApgarScores of 3 or less occurred in 12% HIE babies in contrast to 2% control CS, 1.7%

control VAC and none of control NVD. The control groups had comparable scores (p0.24) but significantly higher ones than the equivalent HIE groups, NVD (meandifference 3.4, 95% CI 2.4–4.4), CS (mean difference 1.8, 95% CI 0.43–3.2) andVAC (mean difference 1.7, 95% CI 2 0.4–3.8).

3.3.3. Birthweight and gestationAll groups had comparable weights (p 0.37) and gestational ages (p 0.6), controls

39.960.7) weeks (range 39–41.2), HIE 40.060.7 weeks (range 38.7–41.1).Scalp pH values were not analysed because too few samples were obtained.

3.4. Encephalopathy

A tense fontanelle and changes in muscle tone occurred in all HIE babies and werepresent from birth in the moderate and severe ones. These groups also developedseizures, depressed reflexes, oliguria and blood and protein in the urine. All severecases required assisted ventilation. Most of the moderate (14 of 20) and severe (12 of17) cases were born NVD while the mild ones clustered in the CS group (10 of 18).

Maternal RBC magnesium averaged 1.42 mmol / l (1.22–1.56) in mild HIE, 1.43mmol / l (1.24–1.54) in moderate HIE and 1.38 mmol / l (1.05–1.49) in severe HIE.There was no correlation between the level of RBC magnesium and the severity ofHIE (p 0.26).

3.4.1. OutcomeInfants classified as mild recovered completely and had normal brain scans at two

weeks of age. Seven of the 20 moderately affected infants showed multicysticencephalomalacia and the remaining 13 appeared to be normal. In the severe group 4died and the surviving 13 have spastic quadriplegia with cystic degeneration in thecortical and subcortical regions.

4. Discussion

Birth asphyxia, and its sequelae is the bane of African obstretrics even insophisticated hospitals [9] where the incidence approaches 8%. In our study thepreponderance of HIE in black babies was striking despite comparable care for allwomen by the same nursing and medical staff. Labour was not significantly longer inthe HIE group nor was the outcome influenced by age or height. Many black womenwere from impoverished rural areas and had low stores of RBC magnesium. Nothingis known of their diet but it is unlikely that their staple food of processed maizecontains much magnesium. Other nutrients may also have been deficient. In a study


of well-nourished women on a home diet Ashe et al [10] established that the meanintake of magnesium during pregnancy was 60% of the recommended 450 mg/day.Of 47 seven-day balance periods only 3 were positive. It is unlikely that the blackwomen in our study had ingested magnesium-rich nuts, cheese, whole grain cereals orgreen vegetables.

The level of plasma magnesium could not predict the outcome of a pregnancy nordid it correlate with the intracellular content of magnesium. The red blood cells whichwere analysed reflected stored magnesium and probably contained comparableamounts of the cation while the discarded young ones have comparatively moremagnesium and are likely to be contaminated by reticulocytes [11].

Depleted stores of maternal magnesium increased the risk of HIE significantly butthe mechanism is unknown. The aetiology is difficult to explain in the absence of anobvious cause such as abruptio or prolapse of the cord. Intrapartum hypoxia is apossibility as suggested by meconium stained liquor and low Apgar scores. On theirown these are notoriously unhelpful features of hypoxia but the combination occurredfrequently in the HIE group.

If hypoxia or ischaemia are important factors their mechanism is uncertain. Innormal labour adequate perfusion of the placenta is ensured by the rise in maternalblood pressure and cardiac output. This compensates for the decreased blood flow inthe uterine artery when frequent contractions increase the amniotic pressure [12]. Theblood flow in the umbilical vessels remains uninterrupted and Brar et al [13] suggestthat intra-amniotic pressure is distributed evenly over the umbilical cord buffered byits Warton jelly. These delicate balances might be disturbed by hypomagnesaemia.Vasoactive substances such as Angiotensin 11 and Prostaglandin F2 constrict humanumbilical vessels in vitro if the medium lacks magnesium [14]. The placenta andumbilical vessels of a magnesium depleted woman may go into spasm during labourwhen vasoactive metabolites are released or alternatively hypomagnesaemia mayenhance the contractility of the uterus. When intra-amniotic pressure exceeds 60mmHg the diastolic flow velocity ceases in utero-placental vessels [13]. Adequatemagnesium decreases the frequency and amplitude of uterine contractions in vitro byblocking the entry of extra-cellular calcium into the myometrial cell [15]. Thisproposal is speculative and Doppler ultrasound together with fetal oxygen saturationmonitoring in labour would be informative.

A significant role for magnesium in uterine function is also suggested by thenumber of abnormal deliveries (Caesarean section and vacuum extraction) whichwere associated with low RBC levels of magnesium. This possibility requires furtheranalysis.

Acute hypoxia rarely damages the fetal brain but in combination with ischaemia itcan destroy cortical neurons within hours of the insult [16]. Hall [17] favours aconcept of vulnerability in HIE whereby a susceptible fetus, unlike a normal onecannot withstand a particular insult. A magnesium-depleted fetus might be en-dangered by hypoxia and progress to HIE despite adequate care and prompt delivery.Some of the accompanying biochemical derangements might be accentuated by adeficiency of magnesium. The ischaemic cells release glutamate [18] which binds tothe N. methyl D Aspartate (NMDA) receptors of neurons and permits sodium, water


and calcium to enter those channels. Magnesium inhibits depolarisation of the NMDAchannel thereby blocking the entry of calcium and by preventing an influx of sodiumit may reduce the risk of cytotoxic oedema. It also inhibits the release of glutamateand counteracts hypoxic damage to cells in culture [19]. Much of the destructionmight be related to the calcium overload which activates intracellular enzymes anddegrades lipids to oxygen-free radicals which can damage the cell membrane. Adeficiency of magnesium enhances susceptibility of hamster hearts to free radicaldamage [20] and this might also be applicable to neurons. The shortfall could depleteATP and inhibit the production of anaerobic energy sooner than expected.

Many HIE babies are born with encephalopathy and the efficacy of magnesiumtherapy after birth is unknown. Can idiopathic HIE be avoided by magnesiumsupplementation in pregnancy? Spatling [21] claims that it reduces the incidence ofasphyxia but bases this on too few cases. The evidence is unconvincing in well-nourished women but we have identified a population at risk which probably warrantsan intervention study in early pregnancy. This could determine whether magnesiumplays a role in HIE or merely acts as a marker.

Acknowledgments

We thank Dr. M. Wright, Department of Obstetrics and Gynaecology, University ofCape Town for advice and permission to study these patients and the nursing staff ofthe Labour Ward and Theatre, Mowbray Maternity Hospital for their invaluable help.The study was supported by a grant from the McCaul Bell Bequest.

References

[1] Nelson, K.B. (1988) What proportion of cerebral palsy is related to birth asphyxia? J. Pediatr., 112,572–574.

[2] Power, D.J. (1988) Asphyxia neonatorum in a developing world situation. MD Thesis, University ofCape Town.

[3] Levine, M., Komburg, J. and Williams, T.H. (1985) The incidence and severity of post-asphyxialencephalopathy in full-term infants. Early Hum. Dev., 11, 21–26.

[4] McDonald, H.M., Mulligan, J.C., Allen, A.C. and Taylor, P.M. (1980) Neonatal asphyxia 1.Relationship of obstretric and neonatal complications to neonatal mortality in 38405 consecutivedeliveries. J. Pediatr., 96, 898–907.

[5] Nelson, K.B. and Grether, J.G. (1995) Can magnesium sulfate reduce the risk of cerebral palsy invery low birthweight infants? Pediatrics, 95, 263–269.

[6] Boston, J.L., Beachene, R.E. and Cruikshank, D.P. (1989) Erythrocyte and plasma magnesium duringteenage pregnancy: relationship with blood pressure and pregnancy-induced hypertension. Obstet.Gynecol., 73, 169–173.

[7] Farr, V., Kerridge, D. and Mitchell, R. (1966) The value of some external characteristics in theassessment of gestational age at birth. Dev. Med. Child Neurol., 8, 657–660.

[8] Fenichel, G.M. (1983) Hypoxic-ischaemic encephalopathy in the newborn. Arch. Neurol., 40,261–266.

[9] Ballot, D.E., Rothberg, A.D., Davies, V.A., Smith, J. and Kirsten, G. (1993) Does hypoxemia preventbrain damage in birth asphyxia? Med. Hypotheses, 41, 334–347.


[10] Ashe, J.R., Schofield, F.A. and Gram, M.R. (1979) The retention of calcium iron phosphorus andmagnesium during pregnancy. The adequacy of perinatal diets with and without supplementation.Am. J. Clin. Nutr., 32, 286–291.

[11] Elin, R.J., Utter, A., Tan, H.K. and Corash, L. (1980) Effect of magnesium deficiency on erythrocyteageing in rats. Am. J. Pathol., 100, 765–777.

[12] Harbert, G.M. and Spisso, K.R. (1980) Biorythmns of the primate uterus (Macaca Mulatta) duringlabor and delivery. Am. J. Obstet. Gynecol., 138, 686–696.

[13] Brar, H.S., Platt, C.D., DeVore, G.R., Horenstein, J. and Medearis, A.L. (1988) Qualitative assessmentof maternal uterine and fetal umbilical artery blood flow and resistance in laboratory patients byDoppler Velocimetry. Am. J. Obstet. Gynecol., 158, 952–959.

[14] Altura, B.M., Altura, B.T. and Carella, A. (1983) Magnesium deficiency-induced spasms of umbilicalvessels: relation to preeclampsia hypertension growth retardation. Science, 221, 376–378.

[15] Popper, L.D., Batra, S.C. and Akerlund, M. (1989) The effect of magnesium on calcium uptake andcontractility in the human myometrium. Gynecol. Obstet. Invest., 28, 78–81.

[16] Vannucci, R.C. and Duffy, T.E. (1976) Cerebral oxidative and energy metabolism of fetal andneonatal rats during anoxia and recovery. Am. J. Physiol., 230, 1269–1275.

[17] Hall, D.M. (1994). Intrapartum events and cerebral palsy. Br. J. Obstet. Gynaecol., 1010, 745–747.[18] Benveniste, H., Drejer, J., Schousboe, A. and Diemer, N.H. (1984) Evaluation of extracellular

concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemiamonitored by intracellular microdialysis. J. Neurochem., 43, 1369–1374.

[19] Rothman, S.M. (1988) Synaptic activity mediates death of hypoxic neurones. Science, 220, 536–537.[20] Freedman, A.M., Cassidy, M.M. and Weglicki, W.B. (1991) Magnesium-deficient myocardium

demonstrates an increased susceptibility to an in vitro oxidative stress. Magnes. Res., 4, 185–189.[21] Spatling, L. and Spatling, G. (1988) Magnesium supplementation in pregnancy. A double-blind study.

Br. J. Obstet. Gynaecol., 74, 120–125.

Red blood cell magnesium and hypoxic-ischaemic encephalopathy

Documents

Transcript of Red blood cell magnesium and hypoxic-ischaemic encephalopathy