Back to the list of reviews

This condition is now divided into ante-natal, intrapartum and post-natal, though these are not necessarily mutually exclusive. Indeed Stanley (1) suggests that a cluster of intrapartum events contributing to causal pathways can be found in 16% of children, but only in 10% is there no evidence of additional antenatal damage. Intrapartum and postnatal causes will not be considered further in this review.

ANTENATAL CAUSES

Genetic causes have been recognised in a few families, but no specific inheritance pattern has been identified, and inheritance appears to be multifactorial. Rates of fetal and neonatal loss are higher in families with all types of CP, as is a family history of being small for dates. It is unclear what the relative contribution of genetic, environmental and nutritional factors is in most families.

Congenital malformations have a strong association with CP. This suggests that damage to the fetus' neurological system is early in the pregnancy, rather than later. Data collected in Western Australia on CP showed 29% of those children with moderate or severe CP had a congenital malformation compared to 4.9% for the control group and normal population (2, 3). CP is also known to be more common in boys than girls.

Multiple pregnancies carry an increased risk of CP as an outcome. In a recent meta-analysis, Stanley calculated a relative risk for CP (compared to singletons) of 4.5 (95% CI 3.9, 5.2) for twins and 18.2 (10.4, 31.9) for triplets (4). The risk remains after correction for increased rates of prematurity. The risk is particularly high when one twin or triplet dies. The relative risk for a twin whose co-twin died is 11.4 (7.0, 18.8) compared to twins where both survive, and this is a 40 fold increase compared to the singleton population (3). This has led Pharoah (5) to postulate that 'the disappearing twin' - i.e. very early loss of a twin fetus - may be the cause of CP in a number of (apparent) singletons (5).

Maternal infections are well-recognised causal pathways of CP. Historically the TORCH infections were first described. However, congenital rubella is relatively rarely seen in the UK due to screening and childhood vaccinations. CMV (cytomegalovirus) infection is estimated to affect around 0.3% of fetuses in utero (6), damage being more severe if in the first or second trimester of pregnancy.

TOxoplasmosis Rubella Cytomegalovirus Herpes simplex

More recently, interest has been directed at other bacterial and mycoplasmal infections (chorioamnionitis, bacterial vaginosis, urinary tract infections) and whether they result in premature labour - the latter a risk factor for CP.

A study from California showed an association between CP and infection in term babies weighing over 2500gms. A fever of over 38 degrees in labour was associated with a 9-fold risk of CP overall and a 19 fold increase in quadriplegia. Chorioamnionitis specifically had a risk of 4.2 to 12 fold of CP, and up to 31-fold for quadriplegia (7). The group investigated this by looking for inflammatory markers in the neonatal blood spots of a group of CP children more than 3 years old, and comparing their concentrations with those found in the neonatal blood spots of a group of control children. A complex picture emerged, but there were significantly raised concentrations of two groups of chemicals in the blood spots of the CP children (8). 65% of the CP children had raised autoimmune antibodies and/or abnormal coagulation factors, compared to 3% of the controls. As a group they also had raised levels of a number of cytokines, including IL-1, IL-6, IL-8 and Tumour Necrosis Factor alpha (TNF). The inflammatory markers were thought to be secondary to the acute infections. The coagulation factors pointed, perhaps separately, to the significance of coagulopathy and interruption of cerebral blood supply - even at a microvascular level - as pathways to brain damage and consequent CP. Studies showing a relationship between peri- and post-natal infections have now been reported from the UK (9, 10).

One theoretical model to link this findings as cause and effect was proposed by Leviton in 1990 (11). He proposed that chorioamnionitis and/or maternal or uterine infections caused a release of TNF, either directly or as a result of endotoxin release. TNF might then stimulate prostaglandin release and cause preterm labour, and the preterm baby would then face a series of hazards (particularly hypotension and hypoxia) that could lead to brain damage. Alternatively the TNF might target specific areas of the brain and cause a localised inflammatory response resulting in permanent damage.

Infection and preterm labour

12

13

14

Treatment of asymptomatic bacteriuria probably prolongs pregnancy (15), and treatment of women with bacterial vaginosis and a previous preterm birth reduces the number of preterm deliveries (16). The current Cochrane review of the treatment of PROM with antibiotics does not advocate 'routine' treatment with antibiotics as there is no clear benefit for overall neonatal morbidity or mortality, although fewer babies need prolonged intensive care (17). However neither this nor the other reviews include good quality data on the long-term neurological outcome for the neonates. The results of the UK ORACLE study, coordinated by the author of the review on PROM, are still awaited. However, no long term benefit is recognised in giving antibiotics to women in preterm labour with intact membranes, although there is some reduction in NEC (18,19).

Infection and Periventricular leucomalacia

Support for the concept that chorioamnionitis might cause brain damage in the fetus came from studies in America and France in 1996. Perlman's team in Dallas identified bilateral cystic periventricular leucomalacia (CVPL) in 14 (2.3%) of 632 infants weighing less than 1750g at birth. Only 4 of these had ever been hypotensive, which was previously thought to be a major pathway to PVL, though other studies had not reported this consistently. They examined 20 perinatal risk factors for morbidity and univariate analysis showed that only chorioamnionitis and prolonged rupture of the membranes (PROM) were significantly associated with PVL (20). For PROM the OR was 6.6 (95% CI 2, 22.1) and for chorioamnionitis 6.8(1.8, 25.9). Similar results came from France (21).

Although not within the scope of this review, it is important to remember that post-natal hypotension can lead to PVL, and Grade 4 IVH or periventricular venous haemorrhagic infarction cause white matter damage with a high risk of neurological impairment in survivors.

Periventricular leukomalacia is 'softening of the white matter around the ventricles'. Cystic periventricular leukomalacia results from extensive necrosis of white matter adjacent to lateral ventricles in the brain. This area includes the cortico-spinal (pyramidal) tracts, and the optic radiations, so neurological consequences include motor deficits such as diplegia, and cortical blindness. Initial studies suggested hypoxia and underperfusion as risk factors associated with PVL, and it was therefore ischaemic in origin. PVL is seen after neonatal hypotension (mean BP < 30mmHg in the early neonatal period), but not consistently so, and not all studies have found hypotension to be a significant association. Recently neuro-chemically mediated injury has been recognised as a potential mechanism for white matter damage. Normally it is detected by ultrasound scans of the brains of infants undergoing intensive care. It shows as an echodensity that may or may not progress to cystic changes. Autopsy findings in infants who die with PVL show cell swelling and death of brain cells, with acute inflammatory changes. After death of the neural cells areas of gliosis (mild cases) or cyst formation (severe cases) develop. These may occur at the microscopic level with no obvious scan findings in the mild cases. If cysts develop, they become apparent from about 12 days after the causative insult. Thus a baby who develops cystic PVL in the first week of life has suffered an antenatal insult. The ultrasound findings are graded into: Flare Persistent flare (> 14 days) Cystic PVL These changes may be bilateral or unilateral These have some prognostic significance. In one study all 14 babies with cysts > 5mm developed CP, as did 2/11 babies with smaller cysts and 6/12 babies with prolonged flare (22). Other studies have reported similar but not identical outcomes with the size and site of the cysts being important.         Figure 1a: A preterm baby with a left sided flare. It shows the typical triangular shape flaring out from the lateral tip of the lateral ventricle.   Figure 1b: The same baby 3 weeks later with extensive cystic changes (arrowed) within the 'flare': this is cystic PVL. The ventricles are now dilated.         Figure 2: A baby with bilateral cystic changes (arrowed) (see reference to Perlman et al in show main text)   Figure 3 A simplified diagram of the cortico-spinal tracts, to et al in et how PVL and hydrocephalus may affect them.

Deficiencies of thyroid and iodine in the mother can have a profound effect on the fetus. The fetus can not produce its own thyroid hormone until late in the first trimester and relies on maternal T4 crossing the placenta; T4 being important for neuronal development. Iodine deficiency is linked with a spectrum of motor disorders.

Fetal growth

Growth restriction is one of the most important risk factors associated with CP, although the association appears more significant for term infants than for premature infants (23,24,25). This may be because the term infants have been under stress for a longer period of time or because the premature symmetrically growth restricted babies have poor survival rates and are therefore not reflected in the statistics. This link with CP may be that growth restricted babies are more vulnerable to insult, particularly intrapartum stress (24). What is not clear is whether CP causes the baby to grow poorly or whether IUGR makes the baby more susceptible to cerebral damage and CP.

Pre-eclampsia and Preventative Treatment

PET itself is associated with a decreased risk of CP, independent of magnesium sulphate usage (25). It is associated with decreased IVH (intraventricular haemorrhage) in the preterm infant and a decrease in infection (25) that might help explain this. Equally, there is a high incidence of elective delivery in this group (23) Mode of delivery, or more specifically, delivery without labour appears to have a protective effect against neurological damage. Antenatal steroids reduce IVH, RDS ( Respiratory distress syndrome) and necrotising enterocolitis (26). Steroids are given when there is a risk of premature delivery, and may be another confounding variable in assessing the effect of PET.

Retrospective and observational studies suggest that magnesium sulphate, when used for eclampsia or as a tocolytic, may have a protective effect against CP (9), with 3 possible mechanisms:

• Reversing cerebral arterial vasoconstriction of middle cerebral artery
• Releasing endothelial prostacyclins
• Inhibiting platelet aggregation

A MAGnet trial (Magnesium and neurological endpoints trial) was therefore set up as a double-blind RCT, comparing magnesium with a placebo. PET was excluded as a confounding variable. Unfortunately there were significantly higher deaths reported in the magnesium arm and the trial had to be stopped prematurely (27).

References

1. Stanley F, Blair E, Alberman E. Pathways to cerebral palsy involving signs of birth asphyxia. In: Cerebral Palsies: Epidemiology and Causal Pathways. Clin Devlop Med 2000;151:22-39.

2. Stanley FJ. Survival and cerebral palsy in low birthweight infants: implications for perinatal care. Paediatric and Perinatal Epidemiology 1992;6:298-310, Abstract

3. Palmer L. Antenatal antecedents of moderate and severe cerebral palsy. Paediatric and Perinatal Epidemiology 1993;16:95-102, Abstract

4. Stanley F, Blair E, Alberman E. The Special case of Multiple Pregnancy? In: Cerebral Palsies: Epidemiology and Causal Pathways. Clin Devlop Med 2000;151:109-123

5. Pharoah PO, Adi Y. Consequences of in-utero death in a twin pregnancy. Lancet 2000;355:1597-1622, Abstract

6. Logan S: personal communication 1998.

7. Grether JK, Nelson K Maternal infection and cerebral palsy in infants of normal birth weight. JAMA;1997;278:207-11, Abstract

8. Nelson K et al. Neonatal Cytokines and Coagulation in Children with CP. Ann Neurol 1998;44:665-675, Abstract

9. Murphy D. Case-control study of antenatal and intrapartum risk factors for cerebral palsy in very preterm singleton babies. Lancet 1995;346:1449-54, Abstract

10. Wheater M. Rennie JM. Perinatal infection is an important risk factor for cerebral palsy in very-low-birthweight infants. Developmental Medicine & Child Neurology. 2000; 42:364-7, Abstract

11. Leviton A, Paneth N White matter damage in preterm newborns: an epidemiologic perspective. Early Human Development 1990;24:1-22, Abstract

12. Hay PE, Lamont RF, Taylor-Robinson D, Morgan DJ, Ison C, Pearson J. Abnormal bacterial colonisation of the genital tract and subsequent preterm delivery and late miscarriage. BmedJ 1994;308:295-298, Abstract

13. Moller M, Thomsen AC, Borch K, Dinesen K, Zdravkovic M. Rupture of fetal membranes and premature delivery associated with Group B Streptococci in urine of pregnant women. Lancet 1984;Iii;69-70, Abstract

14. Stanley F, Blair E, Alberman E. Pathways to cerebral palsy involving preterm birth. In: Cerebral Palsies: Epidemiology and Causal Pathways. Clin Dev Med 2000;151:109-123, Abstract

15. Smaill F. Antibiotics for asymptomatic bacteriuria in pregnancy. Cochrane Database of Systematic Reviews, 2000.Abstract

16. Brocklehurst P, Hannah M, McDonald H. Interventions for treating bacterial vaginosis in pregnancy. Cochrane Database of Systematic Reviews, 2000, Abstract

17. Kenyon S. Antibiotics for preterm premature rupture of the membranes. Cochrane Database of Systematic Reviews, 2000, Abstract

18. King J. Antibiotics for preterm labour with intact membranes. Cochrane Database of Systematic Reviews, 2000, Abstract

19. Norman K, Pattinson RC, deSouza J, de Jong P, Moller G, Kirsten G. Ampicillin and Metronidazole treatment in preterm labour: a multicentre randomised controlled trial. Br J Obst Gynae 1994;101:404-8, Abstract

20. Perlmann JM, Risser R, Broyles RS. Bilateral cystic Periventricular Leukomalacia in the Premature Infant: Associated Risk Factors. Pediatr 1996;97:822-27,Abstract

21. Zupan V, Gonzalez P, Lacaze-Masmonteil et al. Periventricular leucomalacia: risk factors revisited. Dev Med Child Neurol 1996;38:1061-67, Abstract

22. Fazzi E, Orcesi S, Caffi L, Ometto A, Rondini G, Telesca C, Lanzi G. Neurodevelopmental outcome at 5-7 years in preterm infants with periventricular leukomalacia. Neuropediatrics. 1994;25:134-9.

23. Blair E. Intrauterine growth and spastic cerebral palsy 1: association with birth weight for gestational age. Am J Obstet Gynecol 1990;162:229-37,Abstract

24. Berg AT. Childhood neurological morbidity and its association with gestational age, intrauterine growth retardation and perinatal stress, Abstract

25. O'Shea TM. Antecedents of cerebral palsy in very low-birth weight infants. Clinics in Perinatology 2000;27(2):285-302, Abstract

26. Crowley P. Antenatal corticosteroid therapy: a meta-analysis of the randomised trials, 1972 to 1994. Am J Obstet Gynecol 1995;173:322-35, Abstract

27. Mittendorf R. Is tocolytic magnesium sulphate associated with increased total paediatric mortality? Lancet 1997;350(9090), Abstract

28. Spinillo A. Obstetric risk factors for periventricular leucomalacia among preterm infants. Br J Obstet Gynaecol 1998;105:865-71, Abstract