The Role of the Intrauterine and Perinatal Environment in Cerebral Palsy

Author DisclosureDrs Paneth, Korzeniewski, and Hong did not disclose any relationships relevant to this article.After completing this article, readers should be able to: The special importance of CP to clinicians whose practices encompass pregnancy and the perinatal or neonatal period is that no other neurodevelopmental disorder bears as close a causal relationship to the hazards of intrauterine and perinatal life. Although the strength of that relationship has been exaggerated at times, especially in relation to obstetric trauma and birth asphyxia, there can be no serious doubt that adverse events in pregnancy and the perinatal period do occasionally damage the fetal or neonatal brain. A distinctive feature of brain damage produced between early gestation and infancy is destruction of cerebral white matter or extrapyramidal tracts, thereby producing a clinical syndrome in which disturbance of motor control is a prominent and probably a necessary feature. Depending on the nature and extent of the insult, the motor disorder may be predominantly spastic, dyskinetic, or ataxic and may involve all four limbs (quadriplegia), one side (hemiplegia), or be most prominent in the legs (diplegia).CP severe enough to be recognized as a disability is found in about 1 in 500 school-age children in most developed countries, with little variation between countries or over the past 30 years. (1) Because CP often is accompanied by other neurodevelopmental disabilities, particularly seizure disorders, varying degrees of mental retardation, blindness, or deafness, affected children constitute the single largest group of children in the population who have major disabilities. (2) More often than not, however, children who have CP have normal intelligence, but the difficulty of assessing cognition in children who cannot communicate easily because of dysarthria sometimes has hampered recognition of this truth. Computer technologies have made it possible for some children who have even severe dysarthria to express their intelligence in words. (3)In recent years, population-based CP registers have reported on the prevalence of CP in several countries, primarily in Europe and Asia, (4) but only one recent study has reported on CP time trends in a United States population. (5) In general, most studies converge on a prevalence of about 2 per 1,000 school-age children. The increasing survival rate of preterm infants is probably contributing to a slight increase in prevalence, which may be counteracted by a decline in the risk of CP among preterm survivors. (6)During the past 2 decades, our understanding of CP has been advanced greatly by the use of computed tomography (CT) and magnetic resonance imaging (MRI). (1)(7) Many neuroimaging studies have described the neuroanatomic findings that appear to underlie clinical CP, supplementing an earlier fragmented literature that described the neuropathologic abnormalities in children who had CP in whom autopsies had been performed (Table 1). (8) The principal findings emerging from these case series are: About 50% of CP cases occur in infants born after 37 weeks of gestation; the other 50% occur in the approximately 10% of babies born preterm. (15) The relative risk of CP increases steadily with decreasing gestational age at birth; at the lowest gestational ages at which survival now occurs, the risk of CP in survivors is 5% to 15% or at least 50 times the risk of infants born at term. (16) The distinctively high risk of CP in preterm infants suggests that it is useful to consider risk factors separately in preterm and term children (Table 2).The National Collaborative Perinatal Project (NCPP) (more than 50,000 children born from 1959 to 1966 and followed to age 7 y) showed conclusively that clinical measures reflective of birth asphyxia (eg, fetal bradycardia, low Apgar score, delayed time to first breath) occurred in only a minority of children who had CP. (17) About 70% of children who had CP had Apgar scores of 7 or more at 5 minutes. (18) Recent estimates of the proportion of CP attributable to birth asphyxia have ranged from 6% (15) to 8% (19) to 25%. (20) In spite of an apparent improvement in Apgar scores in recent years, no decline in CP has been noted in developing countries. Neonatal encephalopathy, with abnormal neurologic findings as described by Sarnat and Sarnat, (21) is an essential mediator of the relationship of birth asphyxia and CP; absent neonatal neurologic abnormalities, it is unlikely that a case of CP can be attributed to any acute metabolic abnormality acquired at or around birth. (22) At the same time, it is important to keep in mind that not all neonatal encephalopathy is attributable to birth asphyxia. (23)Congenital malformations of many organ systems are more common among children who have CP than among controls, suggesting that factors operating early in gestation may be important in the genesis of CP. (17)(24)(25) The inclusion of malformations of the brain under the diagnostic rubric of CP has been controversial because well-known brain malformation syndromes such as neural tube defects usually are not classified as CP, even when motor findings compatible with CP are present. The recognition, however, that the presence of malformations cannot be excluded without brain imaging has changed perspectives on this diagnostic dilemma, and the most recent (2004) international conference on CP definition and classification decided to include brain malformations in the CP category if they produce the clinical motor findings characteristic of the disorder. (26) Among the more common brain malformations in children who have CP are lissencephaly, polymicrogyria, and schizencephaly. Environmental factors producing this type of CP must, of course, manifest in the first half of pregnancy.Viral and parasitic infections during pregnancy have the potential to damage the brain and occasionally to cause CP. About 1% to 2% of CP in some case series is attributable to prenatal cytomegalovirus infection; (27) prior to widespread rubella vaccination, congenital rubella was the cause of an occasional case of CP. Toxoplasma gondii characteristically produces an infection in the fetus that leads to hydrocephalus and CP via aqueductal obstruction in the brain. (28) It has been estimated that these viral and parasitic prenatal infections might account for up to 5% of CP. (20)Meta-analyses of the relationship of chorioamnionitis and CP recently were published by Wu and Colford (29) and by Wu. (30) Although most of the reviewed studies were of preterm infants, two studies in term infants yielded a pooled odds ratio (OR) of 4.6 for the relationship of chorioamnionitis to CP. Since that review, a case-control study of CP in children born after 36 weeks of gestation found a similar multivariate OR of 4.1 for the presence of chorioamnionitis. (31)A different approach to assessing the infection hypothesis is provided by studies that assess the fetal or neonatal inflammatory response in relation to CP, as measured by the presence of elevated levels of proinflammatory cytokines. Assaying stored blood that had been obtained for newborn genetic screening in 31 predominantly term children who had CP and 65 controls, Nelson and associates (32) found higher levels of the interleukins (ILs) 1, 8, and 9; tumor necrosis factor (TNF)-alpha, and RANTES in all CP cases than in any of the controls. In another analysis of the same sample, interferons alpha, beta, and gamma were found to be higher in 14 of the 31 CP cases than in any control. (33)It has been estimated that iodine deficiency results in a global loss of 10 to 15 intelligence quotient points at the population level and constitutes the world’s greatest single cause of preventable brain damage and mental retardation. (34) Although most cases of cretinism in the developed world are due to malformations affecting fetal and neonatal thyroid hormone production, in iodine-deficient regions, neurologic deficits are produced in utero by the maternal hypothyroxinemia caused by severe lack of dietary iodine. Postnatal infant hypothyroidism produces mental retardation and severely retarded development of many organ systems, but prenatal exposure to maternal iodine deficiency hypothyroidism often leaves the fetus with an intact thyroid gland (and, therefore, euthyroid), but with major neurologic deficits (referred to as neurologic cretinism) that include a form of CP in which spasticity of the legs is a particularly common finding. Affected children also are mentally retarded and have sensorineural hearing loss. Some imaging studies suggest damage to basal ganglia. (35) This form of CP can be prevented completely by iodine supplementation in the form of iodized salt or, where the indigenous salt market cannot be penetrated, by iodized oil injections to women of child-bearing age. (34)(36)(37) An unanswered question is whether milder levels of iodine deficiency in developed countries currently may contribute to neurodevelopmental disorders.Unconjugated bilirubin crosses the blood-brain barrier only during a brief window of postnatal development of uncertain duration (probably days). Bilirubin in the brain stains the developing basal ganglia, causing the pathologic entity kernicterus, which can be fatal. (38)In developed countries, the specific form of dyskinetic CP that can occur in survivors of exposure to very severe hyperbilirubinemia is primarily of historical interest due to the successful use of Rh immune globulin to control rhesus hemolytic disease and careful attention to, and management of, neonatal hyperbilirubinemia via phototherapy and exchange transfusion. (39)(40) However, the practice of very early discharge of babies from the hospital, often unaccompanied by an appropriate plan to monitor the baby for jaundice, may have placed some United States infants at risk of kernicteric brain damage. (41)(42)Consumption by pregnant mothers of foods contaminated by MeHg has produced spastic quadriplegia with mental retardation in Japan (where the food source was fish contaminated with industrial effluent containing MeHg) and Iraq (where grain was contaminated accidentally). (43)(44) There is as yet no evidence of an association of CP with the lesser levels of MeHg exposure occurring in the United States (principally via fish consumption), although this topic has not been much studied.PS is defined as a cerebrovascular event occurring between 28 weeks of gestation and 28 days of postnatal age, often involving the middle cerebral artery and commonly the result of thromboembolism from an intracranial or extracranial vessel, the heart, or the placenta. (32) The CP that results from PS usually is hemiplegic. Surveys from the United States and Canada converge on a prevalence estimate for PS of about 1 per 4,000 live births. (32)(45)(46) Acute neonatal illnesses, congenital heart disease, birth asphyxia, and sickle cell anemia play roles in PS, (45) and a growing body of literature suggests associations as well with inherited thrombophilias, such as factor V Leiden heterozygosity and antithrombin-III deficiency and, for hemorrhagic strokes, protein-C and protein-S deficiencies. Maternal antiphospholipid antibodies, including lupus anticoagulant and anticardiolipin, are not rare, and because they interfere with normal coagulation, they also can increase the risk for PS. Although some studies have found such polymorphisms in children who have CP, it is not yet established that the frequency is higher than in the background population. (32)(46)(47)(48)Genetic factors are believed to play a minor role in CP etiology; the risk of recurrence in siblings is estimated to be only 1%. (49)(50) A CP case-control study in Western Australia between 1980 and 1986 failed to find recurrence of CP, congenital malformations, or reproductive loss in families in which a child had CP. (51) Familial recurrence of CP is likely to be more common in CP resulting from a cerebral malformation.Multiple-birth infants are at an increased risk of CP. Yokoyama and colleagues (52) found a steadily increasing prevalence of CP from 0.9% among twins to 3.1% among triplets to 11.1% among quadruplets. Similarly, Pharoah and Cook (53) found a gradient in prevalence from 0.23% in singletons to 1.3% in twins and 4.5% in triplets. It appears that once gestational age and birthweight are accounted for, multiple birth status conveys virtually no additional independent risk of CP. (54) However, two circumstances unique to twins—a common circulation (twin-twin transfusion syndrome) and death of a co-twin in utero—may elevate the risk of CP. The risk of CP has been estimated to be 14 times higher when a co-twin is known to have died in utero, (55)(56) probably as a result of toxic products of decomposition. (57) It has been hypothesized that in some cases of CP in apparent singletons, a co-twin died in utero early in pregnancy before twinning could be diagnosed. Support for this concept comes from a careful study of early pregnancy that documented a twin dying in utero in 71% of twin pregnancies diagnosed before the 10th week of pregnancy. (58) This figure, if it can be generalized, implies that for every set of twins diagnosed at birth, 2.45 singletons are born who were former twins (71%/29%). With a singleton birth/twin births ratio of 88:1, as many as 2.8% of singletons might be former twins and at elevated risk of CP. However, the relative risk of CP in infants whose co-twin died very early in utero is unknown.The Surveillance of Cerebral Palsy in Europe (SCPE) group recently published the largest study of CP cases (n=4,503), pooling data from several European CP registries. (59) The study showed an excess of CP in children below the 10th percentile in fetal growth for all gestational ages beyond 32 weeks, confirming earlier observations. (60) The finding that some term infants who have CP have periventricular white matter loss similar to that of preterm infants (61) suggests the hypothesis that injury to developing oligodendrocytes in the late second/early third trimester may not always lead to preterm delivery, but may be marked by poor fetal growth subsequent to the injury. A surprising result of the SCPE study was an increased risk of CP in children above the 97th percentile of birthweight for gestational age.The relationship of fetal growth to CP in preterm infants is less secure and more complex than in term infants because some risk factors for preterm birth associated with impaired fetal growth, such as preeclampsia, actually might convey diminished risk for CP. (62) Two of the better-controlled studies failed to find an excess risk of CP in growth-retarded preterm infants; (63)(64) the SCPE study also did not find an excess of CP in growth-retarded babies younger than 32 weeks’ gestation unless fetal (rather than neonatal) growth standards were used to define intrauterine growth retardation. (60)Preterm infants long have been known to have both a high risk of CP and a high prevalence of chorioamnionitis, especially when the latter is defined by pathologic findings of inflammation of membranes, cord, and vessels of the placenta, rather than by clinical signs of infection in the mother. Cerebral palsy in the NCPP survivors was found to be 3.4 times more common among low-birthweight infants who had experienced chorioamnionitis. (65) However, the pooled OR for clinical chorioamnionitis and CP in preterm infants in a review of 21 studies, although statistically significant, was only 1.9. (29) A review of this topic by Willoughby and Nelson was more skeptical of the relationship between chorioamnionitis and CP in preterm infants, seeing it as “equivocal.” (66)As in term infants, attempts have been made to link perinatal inflammatory markers to CP in preterm infants. Nelson and associates (67) failed to find the association of CP with neonatal cytokines that they had found previously for term infants. In contrast, Minagawa and colleagues (68) found that cord blood IL-18 was strongly linked to CP in infants younger than 35 weeks’ gestation, especially among those younger than 30 weeks’ gestation. Yoon and coworkers (69)(70) documented a strong relationship (significant ORs ranging between 5.9 and 6.4) between several cytokines (IL-6, IL-1-beta, IL-8, TNF-alpha) in amniotic fluid and CP in a cohort of preterm (<35 wk) singleton infants in whom amniocentesis had been performed for medical problems in pregnancy and who were followed to age 3 years. A more recent examination of the same cohort revealed a gestational age-adjusted OR of about 6 for amniotic fluid metalloproteinase-8 and CP. (71)Severely preterm infants commonly experience very low levels of thyroid hormone in the early postnatal weeks. Three large studies have shown a relationship between the severity of the hypothyroxinemia and risk of neurodevelopmental disorder, (72)(73)(74) although only one investigation specifically implicated CP among the outcomes. (72) Because low levels of thyroid hormone are associated with other risk factors for brain damage in preterm infants, it is unlikely that this issue will be resolved without a randomized trial of thyroid hormone administration in at-risk infants. (72)(75)Several studies have suggested that hypocapnia (low levels of Pco2) may be associated with periventricular leukomalacia in preterm infants. (76)(77)(78) Increasing evidence suggests the prudence of attempting to maintain Pco2 levels above 35 mm Hg by judicious mechanical ventilation practices. Because hyperoxia also was found in one study to increase the risk of CP, maintaining Po2 levels between 50 and 60 mm Hg may be similarly prudent. (76)Because Mg is a blocker of the N-methyl d-aspartate (NMDA) receptor in the brain and the NMDA receptor is implicated in the excitotoxic amino acid cascade that contributes to asphyxial brain damage, clinical observations suggesting that preterm infants exposed to Mg (eg, because of maternal preeclampsia or for tocolysis) might be at lower risk of CP have been received with considerable interest. However, these findings have not been supported universally, and a recent large randomized trial of Mg administration in labor failed to produce a significant reduction in CP. (79)Investigations into the causation of CP are continuing works in progress. The primary advance of recent years has been the availability of neuroimaging, which has confirmed further the antenatal origin of many cases of CP. In term infants, a modest fraction of CP is related to congenital malformations of the brain. Fetal growth retardation, neonatal encephalopathy (sometimes caused by birth asphyxia), prenatal infection, and perinatal stroke may be contributors to CP in term infants. In preterm infants, ventilator management, perinatal infection/inflammation, and transient thyroid hormone deficiency may be important risk factors. There is no conclusive evidence that CP, which because of its severity and frequency is the single most important neurodevelopmental handicap of childhood, is yet on the decline.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.