In term infants, about 75% of body weight at birth is made up by water (Dweck 1975). Weight loss in the first week of life is the result of a decrease of extracellular fluid via diuresis and a negative fluid balance. The calculation of fluid requirements must take into consideration the maintenance needs, losses and possible deficits. Therefore the difference between fluid losses, including urine production, insensible losses from the skin and lungs, and stool losses, and fluid intake from feeds or parenteral fluids needs to be carefully considered in order to avoid dehydration or overhydration. In the well term infant, fluid requirements in the first few days vary depending on the level of maturity, environmental temperature, body temperature and ambient humidity. Maintenance fluid requirements in these babies range from 60 to 120 ml/kg/d in the first three days (Beischer 1997).
The rationale for fluid restriction following perinatal asphyxia is the avoidance of fluid overload and the exacerbation of cerebral oedema. Cerebral oedema may occur as a result of a hypoxic-ischaemic insult. Klatzo defined two forms of cerebral oedema: 1. vasogenic oedema as a result of increased permeability of the blood brain barrier; 2. cytotoxic oedema manifested as intracellular swelling without increased permeability of the blood brain barrier (Klatzo 1967). It now seems evident that cytotoxic rather than vasogenic oedema may occur as a result of ischaemia (Kimelberg 1995). Increased intracranial pressure (ICP) does not introduce any acute functional neurological disturbances (Clancy 1988). Animal studies of hypoxia-ischaemia also do not support the view that cerebral oedema is a primary injury causing further cerebral insult, but is rather a consequence of the ischaemic damage in the immature animal (De Haan 1997, Mujsce 1990, Stonestreet 1992) .
In current clinical practice, restriction of maintenance fluid intake to anywhere between 40 to 70 ml/kg/day is recommended for neonates with HIE (Levene 2000, Gomella 1999). These recommendations are based on experience from the treatment of adults and children (Shenkin 1976, Yu 2000), or from animal models of cerebral hypoxia and/or ischaemia (Morse 1985). In adults and children, fluid restriction is commonly used in the brain injured patient (Shenkin 1976) and post-resuscitation from cardiac arrest (Hao-Hui 1980). In a series of cases using three different levels of fluid intake (Shenkin 1976) in 30 adult patients, it was found that fluid restriction (1055 ml/day) maintained the patients in a homeostatic balance. A larger fluid intake caused a decrease in serum osmolarity with only minor changes in urea nitrogen and hematocrit, indicating expansion of extracellular space with the potential to worsen cerebral oedema. However, in an adult rat model of experimentally induced cerebral oedema, hydration status did not influence the extent of cerebral oedema (Morse 1985). Neither study provided assessment of either short term or long-term neurological function. A significant decrease in mortality (from 63.5% to 17.2%) was found in a study comparing fluid intake of 60 ml/kg/day with an individualised approach to fluid therapy (40-208 ml/kg/d) in 3773 children aged 1 day to 13 years (Yu 2000). However, no comment was made in regards to aetiology or long-term outcome other than mortality in these children. There was also no mention of the effects on children of different age groups. Although the recommendations for fluid restriction in a neonate are based on the experience of restricting fluid intake in adults or older children, one has to keep in mind that the maintenance fluid requirement in neonates per kg bodyweight is much higher than in older children and adults. Quantitative recommendations for fluid restriction in neonates cannot be derived from the above-mentioned studies. The extrapolation from studies in adults, older children and animals to the human neonate is fraught with hazard due to the different physiology and mechanisms of injury.
The predictive value of clinical signs at or after birth, such as Apgar scores, cord pH, or clinical symptoms of encephalopathy, is low. Ultrasound has a low sensitivity in term babies with HIE. Only if cystic lesions are detected does it show a good correlation with spastic diplegia (Siegel 1984, de Vries 1985). CT can demonstrate diffuse cerebral oedema (Volpe 2000), injury to the cortex and white matter, and damage to central grey matter structures such as thalamus and basal ganglia (Roland 1998). MRI is superior to ultrasound and CT in assessing maturational changes because of its better visualisation of myelination and structural changes (Martin 1995). MRI is a useful tool in evaluating the extent of brain damage within the first 10 days of birth (Barkovich 1995) and is often used in conjunction with neurological assessment for prognostication of outcome.
i. severity of asphyxia (HIE grade 1, 2, or 3) (Sarnat 1976)
ii. degree of fluid restriction (50-59 ml/kg/day; < 50 ml/kg/day, no fluid restriction)
iii. length of fluid restriction (1 day, 1-3 days, > 3 days)
1. metabolic or mixed acidaemia (pH <7.0) on umbilical cord gas sampling if obtained
2. an Apgar score of 0 to 3 for more than five minutes
3. neonatal neurologic complications, e.g. seizures, coma or hypotonia; and
4. multisystem organ dysfunction, e.g. cardiovascular, hematologic, pulmonary, or renal system.
Primary outcomes:
Neonatal mortality (death during the first 28 days of life)
Infant mortality (death during the first year of life)
Severe neurodevelopmental disability at or equal to 12 months of age
or more. Severe neurodevelopmental disability was defined as: cerebral palsy,
developmental delay (DQ <70) or blindness (visual acuity < 6/60 in
both eyes), or any combination of these disabilities.
Secondary outcomes:
Hypoxic ischaemic encephalopathy, highest grade (Sarnat 1976)
Electrolyte disturbances:
a) Hyponatraemia (serum sodium concentration <130 mmol/L) or hypernatraemia (serum sodium concentration > 150 mmol/L)
b) Hypokalaemia (serum potassium concentration < 3 mEq/L) or hyperkalaemia (serum potassium concentration > 7 mEq/L)
The measurements were recorded within the first 24 hours of life as
a reflection of the asphyxial insult and compared with measurements made
at least 24 hours after the initiation of the therapy (result of the therapy).
Urine output (oliguria defined as urine output < 1 ml/kg/h) during
the first three days of life. The measurements were recorded within the first
24 hours of life as a reflection of the asphyxial insult and compared with
measurements made at least 24 hours after the initiation of the therapy (result
of the therapy).
Renal function (renal failure defined as serum creatinine > or = 1.5 mg/dL/133 micromol/L). The measurements were recorded within the first 24 hours of life as a reflection of the asphyxial insult and compared with measurements made at least 24 hours after the initiation of the therapy (result of the therapy).
Seizure activity (detection of seizures based on either clinical
grounds, detection by electroencephalogram, or by treatment of seizures with
antiepileptic treatment).
CT or MRI changes consistent with asphyxia.
infant, newborn (explode) [MeSH heading],
asphyxia (explode), [MeSH heading],
fluid therapy [MeSH heading],
ischaemia/ischemia (textword)
hypoxia (textword)
encephalopathy (explode) [MeSH heading]
Searches were made of previous reviews including cross-references and abstracts. The search was not limited to the English language; reports in foreign languages were translated. The title and abstract of each review were assessed for eligibility. If there was uncertainty, the full report was reviewed.
Weighted mean differences (and 95% confidence intervals) were to be reported for continuous variables such as duration of seizures. For categorical outcomes such as mortality, the relative risk (and 95% confidence intervals) and risk difference (and 95% confidence intervals) were to be reported. We were to use a fixed effect-model for meta-analysis and the heterogeneity statistic to help decide whether pooling was justified.
A strength of the review is the fact that the search strategy was broad and comprehensive. Using a broad definition of asphyxia and then combining this with fluid therapy or encephalopathy allowed us to search as widely as possible. All abstracts were reviewed by the authors and the full report read and examined if there was uncertainty about the abstract. Only review articles or recommendations for the treatment of perinatal asphyxia were found by using this search strategy. Missing publications before 1966 is a weakness of the review, especially if one considers that randomisation of acutely ill (i.e. babies in renal failure) babies may have been done then but may not considered to be ethical today.
Although most standard texts recommend fluid restriction as a treatment of neonates with HIE, it is obvious that these recommendations are not based on evidence from randomised controlled trials in neonates. They are much more likely to be based on evidence from adult studies. Avoiding fluid overload as a result of the renal dysfunction frequently seen after perinatal asphyxia will also play a role in current recommendations. However, the question remains why there are no randomised trials in neonates comparing fluid restriction with no fluid restriction. Reluctance to randomise severely asphyxiated infants to non-restricted fluids with the risk of worsening cerebral outcome may be the reason that no trials have been conducted. Renal dysfunction caused by perinatal asphyxia necessitates the restriction of fluids and, therefore, enrolling babies with renal failure into a randomised trial may not be ethical. If cerebral oedema is the consequence, rather than the cause of perinatal brain injury, then reduction of cerebral oedema will have little effect on neurological outcome. Supporting this observation, pharmacological manipulation of oedema has not been shown in animal studies to affect neuropathological outcome, despite decreasing the level of cerebral oedema by up to 70% (Vannucci 1993). However, one has to keep in mind that even if reduction of cerebral oedema may not affect neurological outcome, an increase in cerebral oedema to a degree that the compliant mechanisms of a neonatal skull are exhausted will result in an increase of the intracranial pressure and potentially contribute to cerebral damage.
Given the potential risk of worsening outcome, randomised studies in neonates may therefore not be ethical in all asphyxiated babies. However, current recommendations include all grades of HIE. Although fluid restriction may be beneficial for babies with multiorgan damage and severe HIE, this is not necessarily the case for babies with a milder degree of HIE without renal failure. 25% of all babies with HIE II will have an adverse neurodevelopmental outcome. These babies do not necessarily have signs of multi-organ damage and may not be as severely compromised at birth as babies with HIE III. It is unclear if in those babies fluid restriction is beneficial. These babies with HIE II are the ones that should be included in a randomised trial. These studies should investigate the effects of fluid management on outcomes such as mortality, seizure activity, evidence of cerebral damage on histology, and effects on renal function and electrolytes.
Barkovich AJ, Westmark K, Partridge C, Sola A, Ferriero DM. Perinatal asphyxia: MR findings in the first 10 days. AJNR. American Journal of Neuroradiology 1995;16:427-38.
Beischer NA, Mackay EV, Colditz PB. Obstetrics and the newborn. 3rd edition edition. Philadelphia: WB Saunders, 1997.
Clancy R, Legido A, Newell R, Bruce D, Baumgart S, Fox WW. Continuous intracranial pressure monitoring and serial electroencephalographic recordings in severely asphyxiated term neonates. American Journal of Diseases of Children 1988;142:740-7.
De Haan HH, Gunn AJ, Williams CE, Gluckman PD. Brief repeated umbilical cord occlusions cause sustained cytotoxic cerebral edema and focal infarcts in near-term fetal lambs. Pediatric Research 1997;41:96-104.
de Vries LS, Dubowitz LM, Dubowitz V, Kaiser A, Lary S, Silverman M, Whitelaw A, Wigglesworth JS. Predictive value of cranial ultrasound in the newborn baby: a reappraisal. Lancet 1985;2:137-40.
Donn SM, Naglie RA. Prevention of post-asphyxial hypoxic-ischemic encephalopathy. Indian Journal of Pediatrics 1986;53:573-86.
Dweck HS. Feeding the prematurely born infant. Fluids, calories, and methods of feeding during the period of extrauterine growth retardation. Clinics in Perinatology 1975;2:183-202.
Fedorova MV, Bykova GF. Infusion therapy of hypoxic and posthypoxic states in term newborn infants (Russian). Akusherstvo i Ginekologiia (Mosk) 1982;3:56-8.
Gomella TL. Neonatology: Management, procedures, on-call problems, diseases, and drugs. 4th edition. Stamford, Connecticut: Appleton and Lange, 1999.
Hao-Hui C. Dehydration therapy and hypotension in post-resuscitation cerebral oedema, and application of intra-ocular pressure measurement--a review of resuscitation work, part I. Resuscitation 1980;8:195-209.
Kimelberg HK. Current concepts of brain edema. Review of laboratory investigations. Journal of Neurosurgery 1995;83:1051-9.
Klatzo I. Presidental address. Neuropathological aspects of brain edema. Journal of Neuropathology and Experimental Neurology 1967;26:1-14.
Levene MI, Tudehope DI, Thearle MJ. Essentials of neonatal medicine. 3rd edition. Blackwell Science, 2000.
Martin E, Barkovich AJ. Magnetic resonance imaging in perinatal asphyxia. Archives of Disease in Childhood Fetal and Neonatal Edition 1995;72:F62-70.
Morse ML, Milstein JM, Haas JE, Taylor E. Effect of hydration on experimentally induced cerebral edema. Critical Care Medicine 1985;13:563-5.
Mujsce DJ, Christensen MA, Vannucci RC. Cerebral blood flow and edema in perinatal hypoxic-ischemic brain damage. Pediatric Research 1990;27:450-3.
Roland EH, Poskitt K, Rodriguez E, Lupton BA, Hill A. Perinatal hypoxic-ischemic thalamic injury: clinical features and neuroimaging. Annals of Neurology 1998;44:161-6.
Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Archives of Neurology 1976;33:696-705.
Shenkin HA, Bezier HS, Bouzarth WF. Restricted fluid intake. Rational management of the neurosurgical patient. Journal of Neurosurgery 1976;45:432-6.
Siegel MJ, Shackelford GD, Perlman JM, Fulling KH. Hypoxic-ischemic encephalopathy in term infants: diagnosis and prognosis evaluated by ultrasound. Radiology 1984;152:395-9.
Sinha SK, Singh J. Newer concepts and approaches to neonatal brain asphyxia. Indian Journal of Pediatrics 1998;65:55-62.
Stonestreet BS, Burgess GH, Cserr HF. Blood-brain barrier integrity and brain water and electrolytes during hypoxia/hypercapnia and hypotension in newborn piglets. Brain Research 1992;590:263-70.
Vannucci RC, Christensen MA, Yager JY. Nature, time-course, and extent of cerebral edema in perinatal hypoxic-ischemic brain damage. Pediatric Neurology 1993;9:29-34.
Volpe JJ. Neurology of the Newborn. 4th edition. Philadelphia: WB Saunders Company, 2000.
Yu PL, Jin LM, Seaman H, Yang YJ, Tong HX. Fluid therapy of acute brain edema in children. Pediatric Neurology 2000;22:298-301.
Dr Zsuzsoka Kecskes
Consultant Neonatologist
Department of Neonatology
The Canberra Hospital
Yamba Drive
Garran
Australian Capital Territory AUSTRALIA
2506
Telephone 1: +61 2 6244 4056
Facsimile: +61 2 6244 3112
E-mail:
zsuzsoka.kecskes@act.gov.au
| The review is published as a Cochrane review in The
Cochrane Library, Issue 3, 2005 (see http://www.thecochranelibrary.com for
information). Cochrane reviews are regularly updated as new evidence emerges
and in response to comments and criticisms, and The Cochrane Library should
be consulted for the most recent version of the Review. |