Prenatal hypoxic-ischemic brain injury is believed to cause permanent neurologic deficits in neonates. Cerebral palsy, severe mental retardation, and seizures have been associated with fetal hypoxic ischemic episodes.1,2 The occurrence of cerebral palsy and other forms of brain injury as a result of hypoxic-ischemic brain injury depends on several factors, including the severity and duration of the hypoxic process, the stage of pregnancy, previous neurologic damage, and other factors that can compromise the fetus (eg, infection, growth restriction).1,3–7
Hypoxic insult to the immature brain can result in seizure activity.1 It has been shown in the rat that hypoxia occurring early postnatally causes the highest seizure frequency and severity.8 In addition to decreased seizure threshold, acute hypoxia to the rat in this neonatal period is also associated with increased long-term susceptibility to seizures.8
Magnesium sulfate has been shown to protect the central nervous system (CNS) from oxygen toxicity and ischemic episodes.9–11 Magnesium sulfate is commonly used as a tocolytic agent for preterm labor and for prevention and treatment of eclamptic convulsions.9 We have shown in previous experiments that peripheral magnesium sulfate injections to the pregnant rat result in significant elevations of magnesium in fetal blood, amniotic fluid (AF), and the fetal brain.12 Fetal blood magnesium concentrations increased by 36% after 4 hours of continuous maternal peripheral magnesium administration. Magnesium concentrations in the fetal brain were significantly increased by 25% after 4 hours of sustained maternal levels.
The objectives of this study were to determine whether severe maternal hypoxia late in pregnancy affects fetal rat body and brain weight and size and causes neuronal damage in the fetal rat brain, and whether magnesium sulfate prevents or decreases those effects.
Materials and Methods
Female, time-pregnant, Long-Evans rats (from different litters) were obtained from Harlan Sprague Dawley (Indianapolis, IN). Rats were housed individually in polycarbonate cages in an environmentally controlled vivarium under 12-hour light and dark cycles. Animals were fed ad libitum throughout the experiments. The entire group of rats was carefully age matched, and the average weight was 350 g.
All pregnant rats (n = 21) were studied at 17 days gestation (total pregnancy length is 21 days). Rats were randomly assigned (based on a computer generated list) to one of four groups to receive saline injections and room air (n = 6), magnesium sulfate injections and room air (n = 5), saline injections and hypoxia chamber (n = 5), or magnesium sulfate injections and hypoxia chamber (n = 5).
Maternal magnesium sulfate injection protocol included a subcutaneous loading dose of 270 mg/kg followed by 27 mg/kg every 20 minutes for 4 hours. Control rats were injected with saline following the same volume and schedule. The 4-hour injection schedule was selected in order to reach significantly elevated concentrations of magnesium in the fetal brain, based on our previous experiments.12
Animals in the hypoxia groups were placed in a chamber containing a gas mixture of 9% oxygen, 3% carbon dioxide (CO2), and balanced nitrogen for 2 hours after the 4-hour injection protocol. The 2-hour time limit was established by our previous experience.13 Rats became severely hypoxic with this protocol without losing consciousness. Tail blood was collected to determine gas and magnesium levels at the beginning (time = 0) and conclusion (time = 2 hours) of the exposure period.
After 72 hours of recovery, at 20 days' gestation, animals were euthanized with CO2 until breathing stopped. The very short exposure to CO2 for euthanizing the rats was not associated with histologic brain changes because several days after the insult are needed for morphologic damage to appear.14 The advantage of this method as opposed to decapitation was the option to perfuse the cardiovascular system. Animals were then perfused transcardially with 0.1 mL heparin (1000 units/mL) followed by 100 mL of normal saline until the perfusate returned clear from a small right atrial incision. Cesarean deliveries were done, the uterine horns were opened, and fetuses were removed in sacs. Each fetal chest was opened and the animal's cardiovascular system was perfused transcardially with 2 mL of saline and 2 mL of 10% buffered formalin, to clean the brain vasculature of blood. After perfusion, the scalp skin and bone tissues were incised posteriorly and brains were removed intact.
All fetal body and brain weights and sizes were measured. A Mettler AE 240 scale (Mettler-Toledo, Inc., Hightstown, NJ) was used to determine fetal weight, and size was measured using a Vernier caliper (Bel-Art Products, Pequannock, NJ). Body size was measured before dissection, and brain size was measured after its isolation. Fetal rat brains and bodies were weighed separately.
Fetal brains were kept in formalin until sectioned. They were embedded in paraffin, sectioned in the coronal plane (8 μm thick), and stained with hematoxylin and eosin (Sigma Chemical Co., St. Louis, MO). Five sections were taken from each brain. Although all sections were evaluated, only one section (the best technically prepared) from each fetal brain was included in the final analysis. Histopathologic examination of the entire section was done with special focus on specific brain regions, including cortex-frontal lobe, parietal lobe, and entorhinal; hippocampuspyramidalis and dentate gyrus; basal ganglia; thalamus; hypothalamus/septum; and white matter. Each brain section was evaluated bilaterally. A neuropathologist masked to the experimental protocol performed histologic assessment of sections. This assessment included the extent of cellular damage within regions (isolated, laminar, diffuse), the nature of cell injury (cell shrinkage, nuclear pyknosis, cytoplasmic hypereosinophilia, and karyorrhexis), as well as reactive changes (inflammation and gliosis).
Blood gas parameters at baseline and after 2 hours of exposure to the gas mixture in the hypoxic chamber were compared, as well as magnesium levels before and after injections, using two-way analysis of variance. Fetal body and brain sizes and weights were compared among all four groups. Analysis of variance was used, and when significant, a multiple comparison procedure (Tukey/Kramer) was applied. The occurrences of histopathologic changes in fetal brains in the four groups were compared using a nonparametric test (χ2; changes were tabulated as 0 and 1 for analysis). Results are reported as mean ± one standard deviation (SD).
The sample size in each experiment was determined on the basis of our previous experience, the reported literature, and the following power analysis. For evaluation of statistical power we selected two important questions. The first was the main effect of the insult and the second was the main effect of drug type. For any given dependent variable we wanted sufficient power to be able to detect any large effect, which Cohen15 defines as d = 0.80 where d is the standardized difference between the population means. We assumed a type 1 error rate of α = 0.05, and an alternative hypothesis H1, d = 0.8. To test five animals per cell the power was calculated as > 0.80.
Exposure to the gas mixture resulted in a 41% decrease in mean maternal oxygen tension from 82.8 ± 20.0 mmHg before the start of the experiment to 49.2 ± 14.4 mmHg after 2 hours of hypoxia (P < .005). Mean maternal pH decreased from 7.37 ± 0.05 to 7.20 ± 0.04 (P < .001). There was no significant change in maternal carbon dioxide tension. Exposure to the magnesium sulfate injection protocol resulted in increased blood magnesium levels from 1.52 ± 0.2 to 3.77 ± 0.7 mg/dL (P < .001).
No significant effect of severe maternal hypoxia on fetal body and brain weight was noted (Table 1). The hypoxia protocol resulted in a significant decrease in fetal body and brain size compared with controls (Table 1). Maternal peripheral administration of magnesium sulfate before the hypoxic episode significantly reduced the effect of hypoxia on fetal brain without changing the body size.
Hypoxia also significantly increased the proportion of fetal rats that had histologic brain injury in the hippocampus (pyramidalis) and thalamus (Table 2). The brain damage in these areas included mainly shrinkage of cells and cells showing karyorrhexis. The data in Table 2 give the number of affected fetuses per number of fetuses examined. The number of maternal rats in each group was 6, 5, 5, and 5, respectively. The number of fetal rats in each group described in Table 1 was 46, 50, 41, and 51. However, only a fraction of those fetuses was included for the histopathologic evaluation (22, 22, 21, and 25, respectively). The denominators in Table 2 are not equal for each group because not all regions could be evaluated clearly in each fetal brain. Maternal magnesium sulfate administered peripherally reduced these deleterious effects on the fetal brain (Figure 1). No significant hypoxic injury effect was observed in the cortex and the white matter of the frontal and parietal lobes. There were no significant inflammatory or glial reactions.
We evaluated the effect of acute severe maternal hypoxia on the fetal brain. The results of this study suggest that severe maternal hypoxia was associated with decreased fetal body and brain size. Administration of peripheral maternal magnesium sulfate prevented the effect of hypoxia on fetal brain size, but no effect of magnesium sulfate on body size was noted. Severe maternal hypoxia also resulted in significantly more isolated fetal neuronal damage in the hippocampus (pyramidalis) and thalamus compared with controls. The neuronal damage included shrinkage of cells and karyorrhexis (fragmentation and breakage of the nucleus). This effect was reduced significantly by administration of peripheral maternal magnesium sulfate.
Hypoxic ischemic fetal brain injury associated with cerebral palsy results in several neuropathologic patterns.1,16 Five histopathologic subtypes have been described, including selective neuronal necrosis, parasagittal brain injury, periventricular leucomalacia, status marmoratus, and focal-multifocal ischemic brain necrosis.1,16,17 Selective neuronal necrosis is the most common neuropathologic fetal brain injury. The most vulnerable neurons are located in the hippocampus, thalamus, basal ganglia (caudate nucleus, putamen, and globus pallidus), and pons.1,18 The clinical effects of selective neuronal necrosis are mental retardation, seizure, and spasticity.1 In our hypoxic fetal rat model, acute hypoxia resulted in isolated neuronal damage with karyorrhexis in the hippocampus and thalamus.
Several mechanisms have been hypothesized to explain the pathophysiology of hypoxic brain damage. The design of this study made it impossible to evaluate fetal blood gases at the time of maternal hypoxia. However, each of the following proposed mechanisms could have directly affected the fetal brain after prolonged, severe, maternal hypoxia. Hypoxia is responsible for a series of biochemical reactions in the CNS that could lead to irreversible neuronal cell injury. Excitatory amino acid receptors, decreased amounts and/or activities of detoxification enzymes, oxygen-derived free radicals, increased intracellular concentrations of free calcium ions, and increased arachidonic acid metabolism have been implicated in the initiation of perinatal hypoxic-ischemic brain damage.19–23 Accordingly, many agents have been studied for the prevention of hypoxic brain injury, including excitatory amino acid receptor antagonists, calcium channel antagonists, anti-oxidants, free radical scavengers, and blockers of arachidonic acid metabolism.21,22
Magnesium sulfate injected subcutaneously into rats crosses the placenta within 2 hours if magnesium levels are sustained, crosses the fetal blood-brain barrier, and concentrates in the fetal rat brain.12 Parenteral magnesium sulfate administration had a definite protective effect on CNS oxygen toxicity.10 It reduced seizure duration and electroencephalogram amplitude in convulsions secondary to oxygen toxicity. A neuroprotective effect of magnesium was found even when it was administered 24 hours after an ischemic episode.11 Recent studies have shown that prenatal exposure to magnesium sulfate in very low birth weight infants was associated with reduction in cerebral palsy.24,25 Nelson and Grether,24 in an observational study, found that 7.1% of 42 infants later diagnosed with cerebral palsy were exposed in utero to magnesium sulfate, compared with 36% of the control survivors. They concluded that magnesium sulfate might have a protective effect against cerebral palsy in that population of infants with birth weights less than 1500 g. Schendel et al25 studied very low birth weight children aged 3–5 years. They found that among these survivors, those exposed to magnesium sulfate had a lower prevalence of cerebral palsy or mental retardation (0.9% and 1.8%, respectively) than those not exposed (7.7% and 5.8%, respectively). They also concluded that prenatal magnesium sulfate exposure was associated with reduced risk of cerebral palsy and possibly mental retardation among very low birth weight children. We have shown in this rat model that peripheral maternal magnesium sulfate administration reduced the level of injury to hypoxic fetal rat brain.
1. Diseases and injuries of the fetus and newborn. In: Cunningham FG, MacDonald PC, Gant NF, Leveno KJ, Gilstrap LC III, Hankins GDV, et al. eds. Williams obstetrics, 20th ed, Norwalk, Connecticut: Appleton & Lange, 1997;975–81.
2. Low JA, Froese AB, Galbraith RS, Smith JT, Sauerbrei EE, Derrick EJ. The association between preterm newborn hypotension and hypoxemia and outcome during the first year. Acta Paediatr 1993;82:433–7.
3. Murphy DJ, Sellers S, MacKenzie IZ, Yudkin PL, Johnson AM. Case-control study of antenatal and intrapartum risk factors for cerebral palsy in very preterm singleton babies. Lancet 1995;346:1449–54.
4. Morimoto Y, Yamamura T, Kemmotsu O. Influence of hypoxic and hypercapnic acidosis on brain water content after forebrain ischemia in the rat. Crit Care Med 1993;21:907–13.
5. Gunn AJ, Parer JT, Mallard EC, Williams CE, Gluckman PD. Cerebral histologic and electrocorticographic changes after asphyxia in fetal sheep. Pediatr Res 1992;31:486–91.
6. Towfighi J, Yager JY, Housman C, Vannucci RC. Neuropathology of remote hypoxic-ischemic damage in the immature rat. Acta Neuropathol 1991;81:578–87.
7. Van Geijn HP, Kaylor WM Jr, Nicola KR, Zuspan FP. Induction of severe intrauterine growth retardation in the Sprague-Dawley rat. Am J Obstet Gynecol 1980;137:43–7.
8. Jensen FE, Holmes GL, Lombroso CT, Blume HK, Firkusny IR. Age-dependent changes in long-term seizure susceptibility and behavior after hypoxia in rats. Epilepsia 1992;33:971–80.
9. American College of Obstetricians and Gynecologists. Hypertension in pregnancy. ACOG technical bulletin no. 219. Washington DC: American College of Obstetricians and Gynecologists, 1996.
10. Katz A, Kerem D, Sherman D. Magnesium sulfate suppresses electroencephalographic manifestations of CNS oxygen toxicity. Undersea Biomed Res 1990;17:45–9.
11. Tsuda T, Kogure K, Nishioka K, Watanabe T. Mg2+
administered up to twenty-four hours following reperfusion prevents ischemic damage of the CA1 neurons in the rat hippocampus. Neuroscience 1991;44:335–41.
12. Hallak M, Cotton DB. Transfer of maternally administered MgSO4
into the fetal compartment of the rat: Assessment of amniotic fluid, blood, and brain concentrations. Am J Obstet Gynecol 1993;169:427–31.
13. Hallak M, Kupsky WJ, Hotra JW, Irtenkauf SM. Fetal rat brain injury: Effect of transient maternal hypoxemia. Fetal Diagn Ther 1997;12:68–71.
14. Schwartz PH, Massarweh WF, Vinters HV, Wasterlain CG. A rat model of severe neonatal hypoxic-ischemic brain injury. Stroke 1992;23:539–46.
15. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, New Jersey: Lawrence Erlbaum Associates, 1988.
16. Volpe JJ. Perinatal hypoxic-ischemic brain injury. Pediatr Clin North Am 1976;23:383–97.
17. Edmonson SR, Hallak M, Carpenter RJ Jr, Cotton DB. Evolution of hydranencephaly following intracerebral hemorrhage. Obstet Gynecol 1992;79:870–1.
18. Hypoxic-ischemic encephalopathy: Neuropathology and pathogenesis. In: Volpe JJ, ed. Neurology of the newborn. 3rd ed. Philadelphia: WB Saunders Co. 1995;279–313.
19. MacDonald JW, Silverstein FS, Johnson MV. Neurotoxicity of N-methyl-D-aspartate is markedly enhanced in developing rat central nervous system. Brain Res 1988;459:200–3.
20. Silverstein FS, Buchanan K, Hudson C, Johnston MV. Flunarizine limits hypoxia-ischemia induced morphologic injury in immature rat brain. Stroke 1986;17:477–82.
21. Giacoia GP. Asphyxial brain damage in the newborn: New insights into pathophysiology and possible pharmacologic interventions. South Med J 1993;86:676–81.
22. Kjellmer I. Mechanisms of perinatal brain damage. Ann Med 1991;23:675–9.
23. Shan X, Aw TY, Smith ER, Ingelman-Sundberg M, Mannervik B, Iyanagi T, et al. Effect of chronic hypoxia on detoxication enzymes in rat liver. Biochem Pharmacol 1992;43:2421–6.
24. Nelson KB, Grether JK. Can magnesium sulfate reduce the risk of cerebral palsy in very low birthweight infants? Pediatrics 1995;95:263–9.
25. Schendel DE, Berg CJ, Yeargin-Allsopp M, Boyle CA, Decoufle P. Prenatal magnesium sulfate exposure and the risk for cerebral palsy or mental retardation among very low-birth-weight children aged 3 to 5 years. JAMA 1996;276:1805–10.