RWI Cooke

Introduction

The planet Earth is unique in our solar system, in that it carries an atmosphere which is rich in oxygen. Oxygen is an abundant element in the universe, but because of its high affinity, is usually found in combination with other elements. Soon after the creation of Earth, the oxygen on our planet was also largely in the form of water, carbon dioxide and silicon and metal oxides. It was the occurrence of the simplest life-forms such as the blue-green algae, that allowed photosynthesis and the release of oxygen contained in carbon dioxide into the atmosphere. Other organisms were then free to develop metabolisms which used oxidation of organic compounds such as glucose to provide an energy source. The ability to photosynthesise is possessed by most higher plant forms and many algae today. Animals and most bacteria remain dependent (as well as plants themselves) on the oxygen they release. Oxidation reactions often give rise to the production of oxygen derived free-radicals (ODFR) which themselves can lead to further reactions. Even very simple organisms such as bacteria, have of necessity over the aeons, developed antioxidant systems to protect themselves from the unwanted effects of ODFRs on their structures and metabolism.

ODFRs have been implicated in a wide range of adult human disease from aging to cancer, heart disease and pancreatitis. Antioxidant therapies both for prophylaxis and treatment re used in both allopathic and "alternative" medicine, and antioxidant supplements are available in every supermarket. This article endeavours to investigate the evidence for ODFRs as a cause of neonatal disorders, and the prospects for antioxidant protection in this group.

Free-Radicals

A free-radical is defined as "any species capable of independent existence that contains one or more unpaired electrons".¹ Normally atoms and molecules have their nuclei surrounded by electrons arranged in concentric shells each containing a number of electron pairs. In a free-radical the outer shell has one or more of these electrons in an unpaired state. Molecular oxygen itself is technically a free-radical, but as it has two unpaired electrons, is' stable at temperatures below 300°C. Free-radicals may comprise a wide range of elements, but those involving oxygen in their make-up are of main importance in human disease.

Free-radicals combine rapidly in most cases with adjacent molecules when possible. This may result in "quenching" of the radical if this reaction involves receiving or taking an electron, or the free-radical may combine with the molecule to form a new free-radical, which is then able to combine with a further molecule, a so called "chain-reaction".² There are many ODFRs, but the two most studied in terms of disease are the superoxide radical and the hydroxyl radical. The latter is the more reactive, and can be produced from superoxide by means of a Fenton reaction, which is catalysed by the presence of small amounts of ''free'' iron.¹

Superoxide

The superoxide radical is produced during oxidative metabolism. Some 3% of electrons escape the mitochondrial electron transport chain, and are available to combine with oxygen to form superoxide radicals. Addition of two electrons produces the peroxide anion, which is readily converted to hydrogen peroxide, which is not an ODFR, but a "reactive oxygen species". Hydrogen peroxide may give rise to hydroxyl in the presence of "free" iron. In addition to mitochondrial transport and Fenton reactions, ODFRs are produced by polymorphonucleocytes, during prostaglandin synthesis, during reperfusion of ischaemic tissues, and through the actions of toxins and ionising radiation. Tissue injury and exposure to high oxygen concentrations also result in excess ODFR production.

Reperfusion injury as a source of ODFR generation has received particular study, as it is likely to occur in conditions as diverse as coronary thrombosis, cerebral thrombosis, ischaemic gut injury, renal transplantation and in ischaemic skin flaps in reconstructive surgery.³ The mechanism of injury in the so called "reperfusion syndrome" is the accumulation of hypoxanthine due to failure during hypoxia to reuse reduced high-energy phosphates such as adenosine triphosphate. On reperfusion of the tissue, the accumulated hypoxanthine is oxidised in the presence of xanthine oxidase, to uric acid, with the incidental release of superoxide. Xanthine oxidase is present in the endothelial cells of many blood vessels, but particularly in the liver and gut wall. One of the principal targets for ODFRs is the cell membrane with its abundance of phospholipids which are very vulnerable to oxidative damage due to their high number of double bonds. Cell membrane damage leads to cell dysfunction and ultimately cell death and necrosis, or to the stimulation of subsequent apoptosis or "programmed cell death".

The results of superoxide production are not all adverse. It has a role in the regulation of vascular tone, in the stimulation of fibroblast proliferation in healing, in down-regulation of lymphocytes and in bacterial killing through the "oxidative burst" of activated neutrophils.

An "ODFR Disease of Prematurity"

Many of the situations in which ODFR production is increased occur in the infant and in particular the preterm infant. Asphyxia-reperfusion, tissue injury, exposure to high environmental oxygen levels, and sepsis with neutrophil proliferation are all commonly seen. The term "ODFR disease of prematurity" has been coined to reflect the possible similarity in aetiology of many neonatal diseases.⁴

Many infants both term and preterm, experience intrapartum hypoxia/ischaemia, and suffer consequently from mainly cerebral damage, although other organs are also involved. The clinical appearances of these children often suggest an initial rapid recovery followed by a secondary deterioration over the subsequent hours, with cerebral oedema, seizures and often death. Magnetic resonance spectroscopy has demonstrated that this clinical picture correlates with the intracellular energy profile of the brain, which after initial recovery, later deteriorates with a secondary phase of energy failure. It is likely that the mechanism for this is the release of ODFRs after reperfusion of the brain.⁵ In the preterm infant where periventricular haemorrhage (PVH) and infarction (periventricular leucomalacia, PVL) are the predominant forms of cerebral injury, similar mechanisms are probably operating. PVH is seen more commonly after a period of hypotension after birth, and PVL is strongly associated with intrapartum ischaemic problems such as placental abruption. In either case the presence of free haemoglobin in brain tissue may lead to localised hydroxyl generation and further tissue injury.

Respiratory distress syndrome in the preterm is primarily due to lung immaturity and a consequent deficiency of pulmonary surfactants. Surfactant deficiency leads to a rapid desquamation of the lining cells of the bronchial tree, and a plasma leak from the pulmonary capillaries into the airways.⁶ Treatment involves exposure to high levels of inspired oxygen and mechanical ventilation. Both of these modalities will tend to increase ODFR production. Pulmonary surfactant, being a phospholipid, is rapidly degraded on exposure to pure oxygen. Secondary invasion of the lung by neutrophils and macrophages is stimulated by ODFRs, but also results in further ODFR release from the oxidative burst of activated white cells. The ensuing lung injury contributes to both the acute and chronic signs of this disease, and especially to the condition of chronic lung disease (CLD) of prematurity (bronchopulmonary dysplasia).^7,8

Retinopathy of prematurity (ROP, retrolental fibroplasia) has long been associated with exposure to high levels of environmental oxygen. With better control and management of oxygen therapy it became a rare condition. However, with the increasing survival of extremely preterm infants, it has again become prevalent, due to their increased susceptibility to the condition. The disease is unique to the preterm infant. The development of the retinal vessels outward from the centre of the optic cup becomes interrupted. This can be due to vasoconstriction due to hyperoxia, although it is likely that other adverse events leading to a reperfusion syndrome could have the same effect. After recovery, growth continues with the proliferation of an excess of new vessels. The release of fibroblast proliferating factors from these new vessels causes the development of fibrous tissue behind the developing retina resulting in eventual detachment and subsequent blindness. It is of interest that a number of epidemiological studies have shown an association between the number of whole blood transfusions given during the period of intensive care in these infants, and the risk of developing ROP.⁹ Frequent transfusions may lead to iron overload, with a greater risk of production of hydroxyl radicals due to Fenton reactions.

Necrotising enterocolitis (NEC) in infants is caused by secondary bacterial invasion of necrotic gut resulting from underperfusion. Animal models of this disorder suggest that it is an example of the reperfusion syndrome. It is particularly prevalent in infants who were shown as a fetus to be growth-retarded and to have retrograde aortic flow in diastole on doppler ultrasound studies.^10,11 Similarly it is frequently seen in preterm infants with a large ductal shunt resulting in a rapid aortic "run-off' in diastole, in term infants undergoing umbilical catheterisation and exchange transfusion, and in those with polycythaemia. All of these situations and procedures may result in gut underperfusion, followed by reperfusion when the situation improves.

Septicaemia results in neutrophil activation with ODFR release in many sick preterm infants. It is recognised to increase the risk of developing CLD, and is also associated with an increased incidence of PVL and a poorer outcome in NEC.

Rhesus incompatibility results in the Rhesus isoimmunisation syndrome seen mainly in North Europeans and their descendants. Antibodies to the Rhesus antigens on the red cell are generated by the Rh negative mother after "immunisation" by Rh antigen positive blood, usually from a previous pregnancy. Major haemolysis results in severely affected fetuses, with anaemia and fetal hydrops as late events. The causation of the fetal hydrops has always been poorly explained. It has been suggested that this simply represents "heart-failure" secondary to the anaemia, but in practice the relationship between the degree of anaemia and the severity of the hydrops is weak. Recent work has shown that in severely affected fetuses, there is a low level of iron binding capacity and a raised level of detectable free iron, presumably secondary to the haemolysis taking place.¹² Levels of lipid peroxidation products were also raised in affected individuals, indicating membrane phospholipid damage, probably explaining the capillary leak present which leads to hydrops.

Antioxidant Systems

Almost all organisms that utilise oxygen have developed antioxidant systems to avoid unwanted effects of ODFR action. These systems develop during gestation, and in newborn animals they are essentially fully developed. Only poor nutrition later in life appears to be associated with deficiencies of antioxidant protection. In the immature animal, however, many of the protective systems are far from mature, and it is likely that the preterm individuals predisposition to diseases involving ODFR injury is at least as likely to be due to a deficiency of antioxidant defences as to increased generation of ODFRs. Antioxidant systems fall into two groups; the enzymatic systems which operate mainly intracellularly, and the non-enzymatic systems which operate both intra- and extracellularly.¹

The main enzymatic antioxidants are superoxide dismutase, catalase, and glutathione peroxidase. Superoxide dismutase converts superoxide to hydrogen peroxide and oxygen. Catalase converts hydrogen peroxide to water and oxygen, and glutathione peroxidase uses glutathione to reduce hydrogen peroxide to water. Non-enzymatic antioxidants are less specific in their actions, but are characterised in the main by their large molecular size and availability of double bonds in their structure. Vitamin A (retinol), vitamin C (ascorbic acid), vitamin E (a-tocopherol), glutathione, uric acid, bilirubin and many lipids, aminoacids and proteins such as albumin may absorb the effects of ODFRs. Storage proteins such as transferrin and caeruloplasmin reduce the likelihood of free transition metal ions being free to catalyse Fenton reactions.

Vitamin A is an important "chain-breaking" antioxidant at low oxygen tensions, and also has important roles in immunity and repair.¹³ Vitamin C is a powerful reducing agent, reacting with superoxide, peroxyl and hydroxyl radicals. It may in some situations promote peroxidation by reducing ferric iron and stimulating Fenton reactions. It can be regenerated by glutathione. Vitamin E is a lipid soluble vitamin which is concentrated in phospholipid membranes, and is the most important "chain-breaking" antioxidant.¹⁴ It can be regenerated by glutathione and vitamin C.

There is plentiful evidence for immaturity of antioxidant systems in preterm animal models, but less in the human infant. Superoxide dismutase activity in the fetal lung at mid-gestation is about half that in the adult.¹⁵ Levels of glutathione in sick ventilated extremely preterm infants are less than half adult levels, but the ratio of reduced to oxidised glutathione (an indicator of oxidative stress) is ten-fold different.¹⁶ Serum transferrin is five times higher in the adult compared to the very preterm infant. Plasma vitamin A levels have been shown to be deficient in infants developing CLD.¹⁷ Vitamin E is present in plasma only in very low levels in preterm infants.¹⁸ Plasma levels of these vitamins may, however, not be a good indicator of their actual availability.

Measuring ODFR Activity in Infants

Measurement of free-radicals directly is difficult in vivo, due to their very short half-life, and evidence that a particular disorder is related to ODFR activity is usually sought by measuring products of oxidation, or the levels of antioxidants present. Measuring individual antioxidants, although often attempted, is unlikely to give a true picture of the antioxidant systems within an individual, as all need to be present and functioning for full protection. Tests such as TRAP and TAS (total antioxidant status) attempt to get round this by measuring directly the free-radical trapping ability of plasma.^19,20 They cannot take into account the effects of intracellular enzyme systems. Recent work has shown that TAS relates poorly to outcome in preterm infants.²² It is very affected by the bilirubin and uric acid levels which fluctuate rapidly in sick infants. There was however a weakly negative relationship between TAS and MDA-TBA detected (a measure of lipid peroxidation). Glutathione measured in blood or in tracheal aspirates is low in preterm infants, and lowest in those that go on to develop CLD. Superoxide dismutase (SOD) is low at birth in preterm infants, but rapidly increases in those exposed to raised oxygen levels. SOD levels in the red cells of infants with CLD are higher than those in adults.

ODFRs may combine with lipids, proteins and even DNA to produce measurable oxidant products. One of the most commonly used measures is of malondialdehyde (MDA), detected in the form of an adduct with thiobarbituric acid (TBA). It is easy to perform, but not very specific. High urinary MDAs have been demonstrated in infants receiving high inspired oxygen,²³ and high plasma levels demonstrated in infants who later died or developed CLD.²⁴ More recent work has not found a strong relationship between plasma and urinary MDAs and outcome. Also there was a poor correlation between plasma and urinary MDA on the same day, although a significant relationship between plasma values on one day and urinary values the next day. Plasma MDAs were highest and urinary values the lowest on day 0.²⁵

Raised plasma hypoxanthine levels result from periods of hypoxia and ischaemia, and are thought to be the substrate for rapid oxidation to uric acid with ODFR generation on reperfusion.⁴ In very preterm infants, significantly higher levels of plasma hypoxanthine were seen in those who subsequently developed PVL, porencephaly, ROP and CLD.²⁶

A more specific measure of the degree of lipid peroxidation due to ODFRs is the measurement of the alkanes ethane and pentane in the expired breath of ventilated infants.²⁷ These gases are generated by the beta-scission by ODFRs of linolenic and linoleic acids respectively. Trapped expired breath is passed through a gas chromatograph to measure the hydrocarbons present. Exhaled pentane peaks in ventilated infants on day 4 and declines slowly thereafter. Peak pentane levels were higher in infants who died later, had major PVH/PVL or ROP.²⁸ No significant correlation with CLD was observed. The administration of lipid infusions as part of total parenteral nutrition produced a sustained rise in pentane excretion. These findings support the hypothesis that lipid peroxidation is a feature of these diseases, but does not prove causation.²⁹ Only therapeutic intervention studies can do this.

Antioxidant Therapy in the Newborn Infant

Probably the earliest antioxidant therapy to be trialed in the human infant, was vitamin E for the prophylaxis of ROP in 1949. Six major trials have been reported since then with varying results. They have been summarised in a meta-analysis which gave an odds ratio of 0.84 (CI 0.58-1.23) for grade 3 or worse ROP with vitamin E prophylaxis.³⁰ In 1978, vitamin E was also tried as a prophylaxis for CLD, and seven further trials have been reported. Again a meta-analysis of all the trials suggests an odds ratio of 0.92 (CI 0.73-1.16) for CLD at 28 days postnatally.³¹ Prophylaxis of PVH has also been tried with vitamin E, and six studies published. All but one of these showed some reduction in PVII in at least some weight groups.³² Differences in study design have prevented a meta-analysis. The treatment has not met with widespread acceptance in neonatal practice.

Vitamin A prophylaxis was proposed for CLD when low levels of vitamin A were observed in those infants with CLD. Six trials have been published,³³ of which only one (the first) showed a benefit, with a reduction in CLD and ROP.³⁴ Superoxide dismutase obtained from cows, was used to prevent CLD by sub-cutaneous injection in infants in 1984. It appeared to be very successful, with a significant reduction in CLD and need for mechanical ventilation. This was somewhat surprising, as SOD is rapidly excreted in the urine after injection, and transport into cells seemed improbable. Further trials of SOD have awaited the introduction of recombinant human SOD which is now available. Intra-tracheal administration of this has been attempted, with demonstrable short term benefits of increased antioxidant levels in plasma and urine for 2-3 days, and a reduction in inflammatory markers in tracheal aspirate after one dose.³⁵

Allopurinol has been investigated, mainly in animal models, as a potential prophylaxis against the reperfusion syndrome. It is a competitive inhibitor of xanthine oxidase. In the newborn rat, piglet and gerbil, allopurinol when given at high doses parenterally (135 mgm/Kg) both before and after a hypoxic/ischaemic insult, was highly protective to the brain, greatly reducing later cortical atrophy.³⁶ An early published trial of allopurinol at the lower dose of 20 mgm/Kg by mouth, claimed to show a significant reduction in respiratory mortality.³⁷ More recently a large blinded randomised controlled trial involving 400 preterm infants was unable to show any benefit from the same oral dose.³⁸ Preliminary studies in term asphyxiated infants with a high dose of intravenous allopurinol have shown promising signs of cerebral protection.³⁹

Chelation therapy to remove free iron from the circulation, and to reduce the possible generation of hydroxyl radicals has been used successfully to control the problems associated with frequent transfusions in older children with thalasaemia. In the preterm infant, chelation with d-penicillamine has been attempted, to prevent ROP. In a small trial involving 104 patients, ROP of grade 2 or more did not occur at all in those treated with d-penicillamine.⁴⁰ Concern over toxicity of the drug, especially when used as a prophylactic in otherwise well infants has prevented further studies.

A non-specific, but very successful intervention in perinatal care has been the widespread use of antenatal steroids given to the mothers delivering prematurely.⁴¹ A complete course of treatment significantly reduces neonatal death, respiratory distress, necrotising enterocolitis and intracranial haemorrhage. Although a major function of steroid prophylaxis is to stimulate endogenous surfactant release, this is only likely to be one of its effects. Prophylactic therapy with surfactant immediately after birth does not have the same protective effect against cerebral haemorrhage and NEC. Among its many effects, antenatal steroid administration has been shown to stimulate antioxidant development in the lung and presumably elsewhere

Conclusions

There is some evidence to suggest that ODFRs are produced excessively in sick newborn infants, and more to suggest that antioxidant systems are immature especially in the preterm. A variety of neonatal disorders are likely to be associated with consequent oxidant damage. Therapeutic interventions to date have been disappointing, but this probably relates to the limited understanding of the reactions involved, and of the interactive nature of the many antioxidant systems. Better delivery systems for antioxidants, combination of antioxidants with pulmonary surfactants and multiple antioxidant delivery may provide solutions in the future.

References

1. Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine, Ed. Oxford; Clarendon Press, 1989.

2. McCord JM. Oxygen derived free radicals in post-ischemic injury. N Engl J Med 1989;312:159-63.

3. Saugstadt OD. Role of xanthine oxidase and its inhibitor in hypoxia: reoxygenation injury. Pediatrics 1996;98: 103-7.

4. Saugstadt OD. Hypoxanthine as an indicator of hypoxia its role in health and disease through free radical production. Pediatr Res 1988;23:143-50.

5. Wyatt JS, Edwards AD, Azzopardi D, Reynolds EOR. Magnetic resonance and near infra red spectroscopy for investigation of perinatal hypoxic-ischaemic brain injury. Arch Dis Child 1989;64:953-63.

6. Murch SH, Costelloe K, Klein NJ, et al. Mucosal tumour necrosis factor-alpha production and extensive disruption of sulphated glycosaminoglycans begins within hours of birth in neonatal respiratory distress syndrome. Pediatr Res 1996;40:484-9.

7. Bagchi A, Viscardi RM, Taciak V, et al. Increased activity of interleukin-6 but not tumor necrosis factor-alpha in lung lavage of premature infants is associated with the development of bronchopulmonary dysplasia. Pediatr Res 1994;36:244-52.

8. Kotecha 5, Wangoo A, Silverman M, et al. Increase in the concentration of transforming growth factor beta-I in the bronchalveolar lavage fluid before development of chronic lung disease of prematurity. J Pediatr 1996;128:464-9.

9. Cooke RWI. The apparent role of blood transfusions in the development of retinopathy of prematurity. Eur J Pediatr 1993;152:833-6.

10. Hackett GA, Campbell 5, Gamsu H, et al. Doppler studies in the growth retarded fetus and prediction of neonatal necrotising enterocolitis, haemorrhage, and neonatal morbidity. Br Med J 1987;294:13-6.

11. Coombs RC, Morgan ME, Durbin GM, et al. Abnormal gut blood flow velocities in neonates at risk of necrotising enterocolitis. J Pediatr Gastroenterol Nutr 1992;I 5:13-9.

12. Berger HM, Lindeman JH, van Zoeren-Grobben D, et al. Iron overload, free radical damage, and rhesus haemolytic disease. Lancet 1990;335:933-6.

13. Underwood BA. Vitamin A in animal and human nutrition. In: Sporn MB, Roberts AB, Goodman DS, eds. The Retinoids, vol 1. Orlando, Fla.: Academic Press, 1984;281-92.

14. Burton GW, Joyce A, Ingold KU. Is vitamin E the only lipid soluble, chain-breaking antioxidant in human blood plasma and erythrocyte membranes? Arch Biochem Biophys 1983;221:281-90.

15. Frank L, Sosenko IR. Development of lung antioxidant enzyme system in late gestation: Possible implications for the prematurely born infant. J Pediatr 1987;110:9-14.

16. Smith CV, Hansen TN, Martin NE, et al. Oxidant stress responses m premature infants during exposure to hypoxia. Pediatr res 1993;34:360-5.

17. Shenai JP, Chytil F, Stahlman MT. Vitamin A status of neonates with bronchopulmonary dysplasia. Pediatr Res 1985;19:185-8.

18. Moyer WT. Vitamin E levels in term and premature newborn infants. Pediatrics 1950;6:893-6.

19. Lindeman JH, van Zoeren-Grobben D, Schrijver J, et al. The total free radical trapping ability of cord blood plasma in preterm and term babies. Pediatr Res 1989;26:20-4.

20. Miller NJ, Rice-Evans C, Davies MJ, et al. A novel method for measuring antioxidant capacity and its application to monitoring the antioxidant status in premature infants. Clin Sci 1993;84:407-12.

21. Drury JA, Nycyk JA, Baines M, Cooke RWI. Does total antioxidant status relate to outcome in very preterm infants? Clin Sci 1988;94 (in press).

22. Jain A, Mehta T, Auld PA, et aT. Glutathione metabolism in newborns: evidence for glutathione deficiency in plasma, bronchoalveolar lavage fluid, and lymphocytes in prematures. Pediatr Pulmonol 1995;20:160-6.

23. Schlenzig JS, Bervoets K, von Loewenich V, Bohles H. Urinary malondiaaldehyde concentration in preterm neonates; is there a relationship to disease entities of neonatal intensive care? Acta Paediatr 1993;82:202-5.

24. Inder TE, Graham F, Sanderson K, Taylor BJ. Lipid peroxidation as a measure of oxygen free radical damage in the very low birth weight infant. Arch Dis Child 1994;70:F107-11.

25. Drury JA, Nycyk JA, Cooke RWJ. Comparison of urinary and plasma malondialdehyde in preterm infants. Clinica Chimica Acta 1997;263:177-85.

26. Russell GAB, Jeffers G, Cooke RWI. Plasma hypoxanthine: a marker for hypoxic-ischaemic induced periventricular leucomalacia. Arch Dis Child 1992;67:388-92.

27. Pikanen OM, Hallman M, Andersson SM. Determination of ethane and pentane in free oxygen radical-induced lipid peroxidation. Lipids 1989;24:157-9.

28. Pitkanen OM, Hallman M, Andersson SM. Correlation of free oxygen radical-induced lipid peroxidation with outcome in very low birthweight infants. J Pediatr 1990;116:760-4.

29. Wispe JR, Bell EF, Roberts RJ. Assessment of lipid peroxidation in newborn infants and rabbits by measurements of expired ethane and pentane: influence of parenteral lipid infusion. Pediatr Res 1985;19:374-9.

30. Sinclair JC, Bracken MB eds. Effective Care of the Newborn Infant. Oxford University Press, 1992, Oxford. p628.

31. Ibid p.401.

32. Ibid p.576.

33. Ibid p.403.

34. Shenai JP, Kennedy KA, Chytil F, Stahlman MT. Clinical trial of vitamin A supplementation in infants susceptible to bronchopulmonary dysplasia. J Pediatr 1987;111:269-77.

35. Davis JM, Rosenfeld WN, Richter SE, et al. Safety and pharmaconetics of multiple doses of recombinant human CuZn superoxide dismutase administered intratracheally to premature neonates with respiratory distress syndrome. Pediatrics 1997;100:24-30.

36. Palmer C, Towfighi J, Roberts RL, Heitjan DF. Allopurinol administered after inducing hypoxia-ischemia reduces brain injury in 7-day old rats. Pediatr Res 1993;33:405-11.

37. Boda D, Nemeth I, Hencz P, Denes K. Effect of allopurinol treatment in premature infants with idiopathic respiratory distress syndrome. Dev Pharmacol Ther 1984;7:357-67.

38. Russell GAB, Cooke RWI. Randomised controlled trial of allopurinol prophylaxis in very preterm infants. Arch Dis Child 1995;73:F27-31.

39. Van Bel F, Shadid M, Moison RM, et al. Effect of allopurinol on post asphyxial free radical formation, cerebral hemodynamics, and electrical brain activity. Pediatrics 1998;101:185-93.

40. Lakatos L, Hatrami I, Orosylan G, et al. Controlled trial of D-penicillamine to prevent retinopathy of prematurity. Acta Paed Hungarica 1986;27:47-56.

41. Crowley P, Chalmers I, Keirse MJ. The effects of corticosteroid administration before preterm delivery: an overview of the evidence from controlled trials. Br I Obstet Gynaecol 1990;97:11-25.