|Home | About | Journals | Submit | Contact Us | Français|
In 1862 William James Little, a London orthopaedic surgeon wrote:
‘The object of this communication is to show that the act of birth does occasionally imprint upon the nervous and muscular systems of the nascent infantile organism very serious and peculiar evils. Nearly twenty years ago... I showed that premature birth, difficult labours, mechanical injuries during parturition to head and neck, where life had been saved, convulsions following the act of birth, were apt to be succeeded by a determinate affection of limbs of new-born children, spastic rigidity from asphyxia neonatorum, and assimilated it to the trismus nascentium and the universal spastic rigidity sometimes produced at later periods of existence1.’
This condition was known as Little's disease until William Osler coined the term cerebral palsy in 1888. He too noted the association with difficult deliveries and with asphyxia requiring prolonged resuscitation2. In the 1890s Sigmund Freud was the first to recognize that antepartum and postpartum factors could cause a similar condition. He postulated that most cases arose from difficult birth but speculated that the birth difficulty might have been caused by some underlying condition2. The asphyxia theory was subsequently given impetus when research in monkeys showed that perinatal asphyxia could cause brain damage3.
In the past 30 years, great increases in the use of fetal monitoring and caesarean section have been driven by the belief that early detection of asphyxia and speedy delivery will prevent brain damage. In parallel with these changes there has been an unprecedented rise in malpractice litigation. Surprisingly, over the same period the incidence of cerebral palsy in term infants has not changed4,5. This lack of impact prompted epidemiological studies which showed that asphyxia accounts for less than 10% of cases6. In 1999 an international consensus statement was published to provide an agreed reference for use by the courts and expert witnesses in birth injury litigation7. The intention was to provide a template that could be modified as new knowledge became available8.
How can the cause of cerebral palsy be established many years after the event? This is achieved by starting with the clinical condition and working backwards. The only type of cerebral palsy associated with intrapartum hypoxia is spastic quadriplegia, especially if accompanied by dyskinesia9. Mental retardation, epilepsy and learning disorders are not caused by birth asphyxia unless also accompanied by spastic quadriplegia. A statement of severity should not be made before 3-4 years, because mild to moderate cerebral palsy improves in the early years and dyskinesia is not always evident before then. Abnormal tone will, however, have been noticed earlier if the dyskinesia is caused by hypoxia. Furthermore, speech and cognitive development cannot be accurately assessed before age 3-4. There are several neurodegenerative and metabolic conditions that are slowly progressive and in their early phases may mimic cerebral palsy. Where there is doubt the child may need to be seen again after an interval. Syndromes such as Lesch—Nyhan, Rett and glutaric aciduria type 1 are examples.
Cerebral palsy caused by intrapartum hypoxia is always associated with a neonatal encephalopathy and seizures10,11. Newborn encephalopathy is defined by Nelson and Leviton as ‘a clinically defined syndrome of disturbed neurological function in the earliest days of life in the term infant, manifested by difficulty with initiating and maintaining respiration, depression of tone and reflexes, subnormal level of consciousness, and often by seizures’12. This definition is applicable only to term infants because feeding difficulty and abnormality of tone and reflexes are common in preterm infants.
The incidence of neonatal encephalopathy in a large study in Western Australia was 3.8/1000 term births13. That study identified certain preconceptional and antepartum risk factors for neonatal encephalopathy14, shown in Box 1. In 29% there were both antepartum and intrapartum risk factors while only intrapartum factors occurred in 4.9%. Spastic quadriplegia develops in about 10% of cases of neonatal encephalopathy, all with seizures11.
The inclusion of fetal acidaemia as an essential criterion in the consensus statement has been criticized, as pH measurements are often not available. The onus is now on maternity units to obtain that information at delivery. Acidaemia is defined as a pH <7 or base deficit >1215,16. A normal pH excludes hypoxic encephalopathy. By contrast, a pH <7 is associated with encephalopathy in only 10-20%15. The majority of severely acidotic infants, born with a base deficit >16, are also normal16.
In response to asphyxia there is an increase in fetal bloodflow to the heart, brain and adrenals at the expense of the kidney, liver, intestines and lung. Severe ischaemia frequently causes major organ dysfunction. Evidence of organ dysfunction provides confirmation of intrapartum hypoxia. It is not, however, an essential criterion, because there are instances of intrapartum hypoxia without evidence of organ dysfunction11,17.
Although meconium staining of the amniotic fluid is associated with increased risk of quadriplegic cerebral palsy, most children born with meconium in the liquor are normal18,19. The Apgar score, developed in 1952 by Virginia Apgar, remains a useful means of predicting neonatal survival, particularly when used in conjunction with pH values20. It is not, however, a sensitive method for predicting neurological outcome and was never intended for that purpose. Over 90% of infants with a 5-minute score of 0-3 will be normal21,22. Even among infants who do not breathe spontaneously for 20 minutes three-quarters of survivors will be normal23. It seems that there is a fine threshold between normality and death from asphyxia.
Fetal heart rate monitoring became established in the 1970s without proper evaluation, and led to a dramatic increase in emergency caesarean sections for fetal distress. According to a recent Cochrane review, the only benefit of electronic heart rate monitoring was a reduction in neonatal seizures24. Fetal heart rate monitoring lacks specificity. For every case of encephalopathy with an abnormal trace there are 83 normal babies with an abnormal trace25. In malpractice litigation, heart trace changes consistent with asphyxia frequently give rise to the claim that an earlier caesarean section would have prevented brain damage. There is no evidence to support such a contention6,9. The time between the decision to perform a caesarean section and delivery has assumed importance in the medicolegal arena. A 30-minute interval is regarded as the ‘gold standard’. Some hospitals have difficulty meeting that arbitrary standard, which seems to be based on what is generally achievable rather than evidence of potential harm26.
Sentinel hypoxic events are episodes of ischaemia that cause hypoxic brain injury in a neurologically intact fetus. For accurate timing and a judgment on possible sequelae, clear clinical signs are required. In addition, the fetal response to the event should be demonstrable by heart trace and pH evidence consistent with asphyxia. Such events, which seldom result in cerebral palsy, are cord prolapse, placental abruption and uterine rupture27.
Certain factors such as prematurity, intrauterine growth restriction and microcephaly at birth suggest a cause other than intrapartum asphyxia7. If two siblings have cerebral palsy, particularly the same type, a genetic cause is likely. Multiple pregnancy is associated with an excess risk of cerebral palsy, the risk being highest if one of the fetuses had died in utero. Fetal coagulation disorders and maternal autoimmune disorders have been linked with cerebral palsy. The coagulopathy induced by these disorders would explain reports of placental thrombi and brain thrombi in stillbirths and neonatal deaths28.
Infection (chorioamnionitis) is now known to be an important factor for cerebral palsy29. It can mimic all the essential and non-specific criteria of intrapartum birth asphyxia. In the past, many cases of cerebral palsy caused by infection were wrongly attributed to birth asphyxia9. There may be no history of prolonged rupture of membranes or clinical evidence of infection in the infant. Cytokines which are neurotoxic are generated by the fetus in response to infection30. Ultrasonography early in the neonatal period shows evidence of brain injury caused by infection31. The reason why the adverse effects are confined to a small minority of infants is not clear.
Imaging shortly after birth is useful in that it may reveal evidence of cerebral oedema—which confirms that the cerebral insult is of recent onset. Oedema develops in 6-12 hours and clears in 4 days7. Radiologists do not always agree on the interpretation of CT scans: the signs of cerebral oedema may not be clearcut. Ultrasonography, which is widely used in neonatology, is likewise open to differences in interpretation. In a recent follow-up study of normal babies, 20% had neonatal ultrasound abnormalities32.
After the neonatal period the main value of neuroimaging is to determine whether the cerebral palsy is caused by a developmental brain abnormality, intrauterine infection or some other congenital abnormality33,34. Since many children with cerebral palsy have brain malformations, neuroimaging is an essential part of legal proceedings35. MR imaging in the infant is now a good predictor of future neurological status, but when used many years later is not reliable in determining the cause or timing of a brain insult8.
The most important factors determining life expectancy are the degree of mental retardation, mobility and the ability to feed. Population-based cerebral palsy registers have been used to gather information about life expectancy. Differences in populations, data collection methods and definitions have resulted in wide variation between studies. The median survival for children who are immobile and tube-fed is about 7 years36. 80% of children with severe cognitive and ambulatory impairment survive to 18, most to 35 and beyond37. Life expectancy can be most accurately assessed by an individualized statistical assessment rather than by simply matching degree of disability with outcome in the published studies (Strauss D, personal communication).
From Little's observations in the middle of the 19th century, a childbirth litigation industry has been born. We now know that birth asphyxia accounts for only a small percentage of this poorly understood condition. Meanwhile the cost of medical negligence payments has risen sharply38. Payments to children with cerebral palsy are some of the largest. These payments now regularly exceed £3 million, and in some cases lawyers' fees exceed the award. In Ireland a child was recently awarded £2.1 million, with legal fees almost double at £4 million39. With the National Health Service currently facing a medical negligence bill of £2.6 billion, would not a no-fault compensation scheme be kinder to families and better for the Treasury?40