The aetiology of cerebral palsy

Early HumanDevelopment36 ( 1994)8l-88

Review article

The aetiology of cerebral PlSY Fiona J. Stanleya’b ‘Western Australian Research Institute for Child Health. GPO Box 0184. Perth WA 6009, Australia ‘Department of Pediatrics, The University of Western Australia

(Received

Key work

7 May 1993; revision

received 4 August

1993; accepted

I October

1993)

Cerebral palsy; Prevalence; Low birth weight infants; Antenatal causes; Postnatal

causes

1. Introduction

The cerebral palsies are the most common and severe motor disabilities in childhood [25,31]. Cerebral palsy is the most life-limiting of childhood disabilities, exceeding asthma, epilepsy, blindness [21]. There is currently increased interest in their occurrence and patterns of likely causes for a variety of reasons. One is that the trends in cerebral palsy in those countries with population based data all show increases in prevalence in spite of increased use of perinatal practices aimed at reducing brain damage; another is that litigation for obstetric malpractice in cerebral palsy cases has reached proportions which are now having quite worrying effects on obstetric care [8] with debate focussing on whether the ‘new obstetrics’ introduced in the 1970s has had any major effects at all on cerebral palsy rates and the vexed causal association with perinatal asphyxia. The third main reason is that there have been universal reports of particularly significant increases in cerebral palsy rates in the low and very low birth weight infants who are surviving in increasing numbers [ 11,27,36]. Additional reasons for the international interest in cerebral palsy, but not particularly in what its causes might be, is that conductive education is being promulgated by parents and others as a significant advance in therapy. As well, selective posterior rhizotomy has now also started to be used on selected cases. Both of these treatments, but particularly the latter demand randomised trial evidence of effectiveness, particularly to ascertain subgroups who may benefit more than others as well as any harmful side effects. 0378-3782/94/$07.00 0 1994 Elsevier Science Ireland SSDI 0378-3782(93)01484-S

Ltd. All rights reserved

82

F.J Stanley/Early

Hum. Dev. 36 (1994) 81-88

This recent interest in cerebral palsy in the last decade is reflected in a significant increase in publications and in the UK, the founding of an organisation, called the Little Foundation (Little originally described cerebral palsy in 1862). The Little Foundation was established to encourage research into the causes and prevention of the cerebral palsies and so far has held, in 1991 and 1992 [14], a series of international workshops on various aspects. These have been most valuable in getting researchers together to pool their ideas and to encourage them to collaborate to test important hypotheses. 2. Definition Cerebral palsy is defined as a group of non-progressive motor disorders of movement or posture due to a defect or lesion of the developing brain [ 11.They have also been defined as a group of chronic disabilities characterised by aberrant control of movement or posture apearing in early life and not the result of progressive disease (181. 3. Cerebral palsy heterogeneity There are many syndromes, many individually very rare, that have as part of their clinical picture a non-progressive motor handicap. As defined, interference with the developing brain could include influences from conception into childhood (by which time cerebral palsy has usually been diagnosed), and thus causes can potentially be genetic, early pregnancy, perinatal or postnatal problems or a combination of factors. The cerebral palsies are thus aetiologically, pathologically and clinically a heterogeneous group of syndromes. The usefulness of this broad grouping of children with similar handicaps, but with a variety of causes has been for the purposes of management and planning of services. There is also considerable inter-observer variation in the diagnosis of cerebral palsy subtypes [31], which needs attention when comparing or pooling data from several sources. Few aetiological cerebral palsy syndromes have homogeneous clinical patterns; possibly the best examples are choreoathetosis and deafness associated with kernicterus due to Rh incompatibility (now preventable by the administration of anti-D after a sensitizing pregnancy) [30] and spastic diplegia with deaf mutism in the children born to iodine deficient mothers [26]. Few other known causes.result in identical clinical syndromes - for example, congenital rubella syndrome - a study in Western Australia (before vaccination was widespread) showed that 10% had cerebral palsy with a variety of motor handicaps [35]; birth asphyxia does not seem always to result in the same pattern of disability [9]; and the preterm infants who survive with cerebral palsy now show a heterogeneity of motor problems although the majority have spastic diplegia or quadriplegia [36,5]. Thus it seems that a variety of clinical pictures can arise from a single cerebral pathology and that many different cerebral pathologies can result in similar clinical entities. This heterogeneity is but one of the challenges in researching the causes of cerebral palsy!

F. J Stanley/Early

Hum. Dev. 36 (1994) 81-88

83

4. Prevalence

Recent trends in the birth prevalence of the cerebral palsies from the three population based registries with longitudinal data and similar methods of case ascertainment are similar with rates around 2 per thousand and rising in more recent birth cohorts [ 11,27,36]. Few Americans have responded to the exhortations of Paneth and Kiely (1984) [23], and little population data are available in USA. They would be of significant interest as there is a policy of more aggressive perinatal care there than elsewhere and differences in CP prevalence may be illuminating. A population based study of moderate and severe CP has been reported from California for children born from 1983 to 1985 [lo]. The prevalence by 3 years of age was 1.24 per 1000 live births. Interestingly this is comparable to 1.23 per 1000 livebirths for similarly classified children in the same years for Western Australia [33]. 5. Trends in low birth weight infants Total cerebral palsy rates mask the significant increases in very low birth weight infants. This is because the proportion of all children with cerebral palsy who had been of very low birth weight is still small, and these increases have made only a small contribution to the overall rate, and because in some series there has been a coincident fall in the proportion of cerebral palsy among heavier babies. Trends in the proportions of cerebral palsy in very low birth weight (c 1500 g) infants for the same three diverse populations are very similar, with significant increases in the proportions of very low birth weight cerebral palsy coincident with falls in mortality in the same birthweights following the introduction of neonatal intensive care [11,27,36]. In Western Australia cerebral palsy in infants weighing 5 1500 g rose from 15 per 1000 live births in 1967-70 to over 60 in 1983-85. The postnatal complications of preterm birth are well documented and include intraventricular haemorrhage and of particular relevance for cerebral palsy, periventricular leucomalacia [38,5]. However there is considerable disagreement as to whether the increase in cerebral palsy in these infants is all due to these complications or whether some preterm infants are already damaged before delivery [27,5,13]. These issues have considerable implications for those providing neonatal intensive care as its ability to reduce cerebral palsy in low birth weight infants hinges on the infant not being damaged before care commences [ 121.The increasing use of cerebral ultrasound and other scans in populations of surviving preterm infants should soon allow the natural history of intraventricular haemorrhage to be described in these children. 6. Known aetiologies of cerebral palsies The possible routes to irreversible brain damage are shown in the Figure [33]. Adverse antenatal events may either cause brain damage themselves or make the infant more vulnerable to the normal asphyxiating events of labour and delivery. In some

84

F.J Stanley/Early

Hum. Dev. 36 (1994) 81-88

infants, severe perinatal asphyxia may be the sole cause, although the reasons for its occurrence may never be known; and there are the groups of infants who have severe postnatal cerebral complications or suffer postnatal brain damage from a variety of situations. 7. Antenatal causes of cerebral palsy There are few well documented antenatal causes of cerebral palsy, although many of us now believe that the majority of cases probably stem from unrecorded antenatal events (how do you record abnormal neuronal migration in the second trimester?). The pure genetic contributions to cerebral palsy are rare and have been reviewed recently in Western Australia by Petterson (PhD thesis) [24]. Monogenic disorders explain only a very small proportion of the excess of cerebral palsy observed in multiple births. Chromosomal abnormalities, whilst likely to be associated with migration disorders, have so far not been looked for with new molecular techniques in large enough series to be properly enumerated. The higher rates in multiple births for Western Australia are shown in Table 1. These babies share genes and intrauterine environments, may have had drug and other methods of infertility treatments to be conceived, and are more at risk of poor growth, preterm birth and obstetric complications [28]. No study large enough to sort out these various contributions to cause has been done; but following the recent Little Foundation Workshop in Perth WA, a multicentre study is planned. My feeling is that much of the excess in multiple births will be explained by the high rates of low birth weight/preterm birth in multiple births [39] and antenatal death of a cotwin [28,16], rather than any genetic or obstetric care factors. Vascular factors such as twin-twin transfusion and disseminated intravascular coagulation have been thought responsible for the brain damage in the surviving twin [16]. However, in WA, we have found that the risk of cerebral palsy seems to be as high in unlike sex surviving twins as in like sex [39], so other mechanisms may also be involved. As most developed nations are showing a rise in multiple birth rates and falls in neonatal mortality in low birth weight infants then the level of neurological complications in these infants is of significant concern. As antenatal death of a cotwin or cotriplet carries such an increased risk of cerebral palsy, the increasing practice of fetocide needs to be investigated. Antenatal infections are well documented if rare causes of cerebral palsy. The most common was probably congenital rubella syndrome; the numbers of cases and

Table 1 Birth prevalence

of CP by plurality

(Western

Australia

1966-1985)

Plurality

N

Birth prevalence/ 1000 livebirths

(9%

Singleton Twins

877 50

2.4 6.3

(2.22-2.54) (4.51-8.03)

5

32

(3.7-59.5)

Triplets

CI)

F. J Stanley/Early

Hum. Dev. 36 (1994) 81-88

85

defects observed in a study done in Western Australia were significant before the vaccination program commenced [35]. Other intrauterine infections may also interfere with central nervous system development and cause cerebral palsy, cytomegalovirus being the most guilty candidate at the moment [6,7]. Chemical teratogenesis in early pregnancy has been shown to result in cerebral palsy with two well known and startling examples, the Minemata methyl mercury disaster from Japan [ 171and severe iodine deficiency in mothers which was observed in the Highlands of New Guinea [26] and reversed by iodised oil in early pregnancy. How important chemical exposures are to the total burden of cerebral palsy in developed countries is not quantifiable at present. Several recent studies have suggested that maternal thyroxine therapy in pregnancy is associated with cerebral palsy [29,4,26]. Maternal trauma during pregnancy has received a great deal of publicity recently but apart from case reports little data are available to enable any epidemiological description. Intrauterine growth retardation is significantly associated with cerebral palsy in the Western Australian data [3]; this was only of importance in those infants born after 34 weeks’ gestation. Many.others have also stressed the importance of poor intrauterine growth as a risk factor for cerebral palsy [37,5] and further stimulates us to look antenatally for causes of these handicaps. The role of pregnancy complications has been difficult to study; my reading of the literature and our own studies is that pre-eclampsia does not seem to be a risk factor, that antepartum haemorrhage is a major antecedent of preterm birth and early neonatal death but does not figure much as an antecedent for cerebral palsy, and that multicentre studies which can obtain reasonable data on large numbers of cases and controls to analyse by subcategories of cerebral palsy are needed to investigate the role of these antenatal factors. It is highly likely that many individually rarely occurring pregnancy factors may interrupt neuronal migration at a crucial stage. It is felt that most cases of cerebral palsy probably have their antecedents during pregnancy rather than during delivery [4,20]. 8. Perinatal causes of cerebral palsy Perinatal refers to the period around birth; in Australia this is officially from around 20 weeks’ gestation to 4 weeks after birth. In the UK it is from 28 weeks to 7 days postnatal. Until recently many people including parents and lawyers believed that most cerebral palsy originated at this time. The relationship between intraparturn asphyxia and later cerebral palsy has created as much interest as the level of cerebral palsy in preterm infants. Overestimation of the risk of impairment from intrapartum problems and obstetric mismanagement may have led to the inappropriate use of some obstetrical interventions or at least justified their increasing use [34]. There have been considerable problems in attempting to bring some rigour into this analysis. Our ability to measure birth asphyxia is limited and most measures (abnormal fetal heart rate traces, presence of meconium, low Apgar scores, low cord blood pH, delay in spontaneous breathing) occur from a variety of causes, are not measuring cerebral oxygen deprivation, do not correlate well with each other and

86

F.J Stanley/ Early Hum. Dev. 36 (1994) 81-88

are not good predictors of cerebral palsy (for example 45% of births in some studies have abnormal CTG traces) [ 151. The interpretation of fetal heart rate traces is notoriously unreliable with intra- and inter-observer variation almost negating their usefulness in epidemiological study [29,22]. Most studies now suggest that less than 10% of all cerebral palsy is likely to be due to birth asphyxia and can still not answer the vexed question of whether any obstetric intervention could have changed the outcome [32,2]. Many of the markers of asphyxia may well be results rather than causes; we need to look to their antecedents to identify the true causal sequences. Antenatal factors may contribute directly or they may make the child more vulnerable to perinatal events which by themselves would not cause problems. Nelson has said that newborn encephalopathy is the road between bad birth and bad outcome; if a baby shows no neurological signs after birth then it is unlikely that any subsequent problems are related to the birth. She demonstrated that it is only in the presence of signs of newborn encephalopathy that any obstetric factors had any relationship to cerebral palsy, and in most of these cases, antenatal factors were present [19]. A study of newborn encephalopathy is currently underway in Western Australia investigating both antenatal and intrapartum factors as well as outcomes. 9. Postnatal causes of cerebral palsy Postnatal refers to documented causes occurring after the neonatal period in a child thought to be normal before the event. These are the most preventable of all the cerebral palsy cases and are due to accidents, near drownings, encephalitis and other infections and increasingly post-surgery cases. 10. Summary I have attempted to give an overview of the latest thoughts on the aetiology of the cerebral palsies. These motor disabilities are of continuing interest and their prevalence is rising, particularly in low birth weight preterm singletons and multiple births. The likely multiplicity of causes demands intelligent investigation probably in collaborative population data bases. Ultrasound and other scans may provide better data on both site and timing of neonatal brain damage. However these are more likely to have been done on cerebral palsied children who were preterm than on those born at term. There are currently no clear preventive messages except those relating to postnatal cerebral palsy or encouraging strategies to reverse the increases in multiple births. The challenges now are to obtain better data on the antenatal factors and on causal sequences which may be important. The Little Foundation is seeking to encourage such collaborative studies. Other sugestions from our workshop [ 141included attempting to follow up those infants who have been participants in large randomised controlled trials (such as those evaluating surfactant) to the age when they can be confidently diagnosed as having cerebral palsy. If antenatal data have been collected on them they may form the basis for a cohort analysis relating risk to cerebral palsy occurrence, as well as to answer questions about whether the intervention influenced cerebral palsy occurrence.

F.J Stanley/ Early Hum. Dev. 36 (1994) 81-88

81

11. References 1 2 3 4 5 6 7

8 9 10 11 12

13 14 15

16 17

18 19 20 21 22 23

Bax, M. (1964): Terminology and classification of cerebral palsy. Dev. Med. Child. Neurol., 6. 295-297. Blair, E.B. and Stanley, F.J. (1988): lntrapartum asphyxia: a rare cause of cerebral palsy. J. Pediatr.. 112,515-519. Blair, E.B. and Stanley, F.J. (1990): Intrauterine growth and spastic cerebral palsy I. An association with birthweight for gestational age. Am. J. Obstet. Gynecol., 162, 229-237. Blair, E.B. and Stanley, F.J. (1993): When can cerebral palsy be prevented? The generation of causal hypotheses by multivariate analysis of a case control study. Paediatr. Peri. Epidemiol., 7: 272-301. Cooke, R.W.I(l990): Cerebral palsy in very low birthweight infants. Arch. Dis. Child, 65,2Ol-206. Dobbins, J.G., Stewart, J.A. and Demmler, G.J. (1992): Surveillance of congenital cytomegalovirus disease, 1990-1991. Morbid. Mortal. Weekly Rep., 41(SS-2), 35-44. Fowler, K.B., Stagno S, Pass, R.F., Britt, W.J., Boll, T.J. and Alford, C.A. (1992): The outcome of congenital cytomegalovirus infection in relation to maternal antibody status. N. Engl. J. Med., 326, 663-667. Freeman, J.M. and Freeman, A.D. (1992): Cerebral palsy and the ‘bad baby’ malpractice crisis. Am. J. Dis. Child, 146, 725-727. Freeman, J.M. and Nelson, K.B. (1988): lntrapartum asphyxia and cerebral palsy. Pediatrics, 82, 240-250. Grether, J.K., Cummins, S.K. and Nelson, K.B. (1992): The California cerebral palsy project. Paediatr. Perinat. Epidemiol., 6, 339-351. Hagberg, B., Hagberg, G., Olow, I. and von Wendt, L. (1989): The changing panorama of cerebral palsy in Sweden. V. The birth year period 1979-82. Acta Paediatr. Stand., 78, 283-290. Kitchen, W.A., Doyle, L.W., Fox, G.W., Richards, A.L., Lissenden, J.V. and Ryan, M.M. (1987): Cerebral palsy in very low birth weight infants surviving to 2 years with modern perinatal intensive care. Am. J. Perinatol., 4, 29-35. Leviton, A.L. and Paneth. N. (1990): White matter damage in pretenn newborns an epidemiologic perspective. Early Hum. Dev., 24, l-22. Little Foundation (1992): Report on workshop: How Can Epidemiology and Registers Assist in the Prevention of Cerebral Palsy? (available from Martin Bax or Fiona Stanley). Low, J.A. (1990): The significance of fetal asphyxia in regard to motor and cognitive deficits in infancy and childhood. In: Obstetrical Events and Development Sequelae, pp. 43-58. Editor: N. Tejani. CRC Press, Boca Raton, FL. Melnick, M. (1977): Brain damage in survivor after in-utero death in monozygous cotwin. Lancet, ii, 1287. Murakami, U. (1972): Organic mercury problems affecting intrauteirne life. In: Drugs and Fetal Development. Editors: M.A. Klingberg, A. Abromovic and J. Clark. Proceedings of an International Symposium on the Effects of Prolonged Drug Usage on Fetal Development. Plenum Press, New York. Nelson, K.B. and Ellenberg, J.H. (1978): Epidemiology ofcerebral palsy. Adv. Neurol., 19.421-435. Nelson, K.B. and Ellenberg, J.H. (1984): Obstetric complications as risk factors for cerebral palsy or seizure disorders. J. Am. Med. Assoc., 251, 1843-1848. Nelson, K.B. and Ellenberg, J.H. (1986): Antecedents of cerebral palsy. Mulivariate analysis of risk. N. Engl. J. Med., 315, 81-86. Newachek, P.W. and Taylor, W.R. (1992): Childhood chronic illness: prevalence, severity and impact. Am. J. Publ. Health, 82, 364-371. Paneth, N., Bommarito, M. and Stricker, J. (1993): Electronic fetal monitoring and later outcome. Can. Med. Assoc. J., in press. Paneth, N. and Kiely, J. (1984): The frequency of cerebral palsy: a review of population studies in industrialised nations since 1950. In: The Epidemiology of the Cerebral Palsies. Editors: F.J. Stanley, E.D. Alberman. Spastics International Medical Publications, Clinics in Developmental Medicine, No. 87. Blackwell Scientific Publications, Oxford.

88 24 25 26 21 28 29 30

31

32 33 34 35

36 31 38 39

F.J Stanley/Early

Hum. Dev. 36 (1994) 81-88

Petterson, B. (1992): Genetic aspects of cerebral palsy [dissertation]. University of Western Australia, Perth, Western Australia. Pharoah, P.O.D. (1985): The epidemiology of chronic disability in childhood. Int. Rehab. Med., 7, 11-17. Pharoah, P.O.D, Buttfield, I.H. and Hetzel, B.S. (1971): Neurological damage to the fetus resulting from severe iodine deficiency during pregnancy. Lancet, i, 308-310. Pharoah, P.O.D, Cooke, T., Cooke, R.W.1 and Rosenbloom, I. (1990): Birthweight specific trends in cerebral palsy. Arch. Dis. Child., 65, 602-606. Scheller, J.M. and Nelson, K.B. (1992): Twinning and neurologic morbidity [review]. Am. J. Dis. Child., 146, 11IO-I 113. Spencer, J.A.D (Editor) (1991): Fetal Monitoring. Physiology and Techniques of Antenatal and Intrapartum Assessment. Oxford Medical Publications, Oxford. Stanley, F.J. (1984): Perinatal risk factors in the cerebral palsies. In: The Epidemiology of the Cerebral Palsies. Editors: F.J. Stanley, Ed. Alberman. Spastics International Medical Publications, Clinics in Developmental Medicine, No. 87. Blackwell Scientific Publications, Oxford. Editors: F.J. Stanley, and E.D. Alberman. (1984): The epidemiology of the Cerebral Palsies. Spastics International Medical Publications, Clinics in Developmental Medicine, No. 87. Blackwell Scientific Publications, Oxford. Stanley, F.J. and Blair, E.B. (1991): Why have we failed to reduce the incidence of cerebral palsy? Med. J. Aust., 154, 623-626. Stanley, F.J. and Blair, E.B. (in press):. The epidemiology of the cerebral palsies. In: The Epidemiology of Childhood Disorders. Editor: I.B. Pless. Oxford University Press, Oxford. Birth asphyxia, cerebral palsy and a shot in the foot [editorial] (1989). Lancet, ii, 1251-1252. Stanley, F.J., Sim, M., Wilson, G. and Worthington, S. (1986): The decline in congenital rubella syndrome in Western Australia: an impact of the schoolgirl vaccination program? Am. J. Public Health, I, 35-39. Stanley, F.J. and Watson, L. (1992): Trends in perinatal mortality and cerebral palsy in Western Australia, 1967-1985. Br. Med. J., 304, 1658-1663. Uvebrant, P. and Hagberg, Cl. (1992): Intrauterine growth in children with cerebral palsy. Acta Paediatr., 81, 407-412. Volpe, J. (1990): Brain injury in the premature infant: Is it preventable? Paediatr. Res., 27 Suppl., S28-S33. Petterson, B., Nelson, K.B., Watson, L. and Stanley, F.J. (1993): Twins, triplets, and cerebral palsy in births in Western Australia in the 1980s. Br. Med. J., 307, 1239-1243.