INTRODUCTION — Neonatal encephalopathy is a heterogeneous syndrome characterized by signs of central nervous system dysfunction in newborn infants. Clinical suspicion of neonatal encephalopathy should be considered in any infant exhibiting an abnormal level of consciousness, seizures, tone and reflex abnormalities, apnea, aspiration, feeding difficulties [1-3], and an abnormal hearing screen.
This topic will review the etiology and pathogenesis of neonatal encephalopathy. Other clinical aspects of this syndrome are discussed separately. (See "Clinical features, diagnosis, and treatment of neonatal encephalopathy".)
TERMINOLOGY — "Neonatal encephalopathy" has emerged as the preferred term to describe central nervous system dysfunction in the newborn period [2,4]. The American College of Obstetricians and Gynecologists (ACOG) describes neonatal encephalopathy as a clinically defined syndrome of disturbed neurologic function in the earliest days of life in an infant born at or beyond 35 weeks of gestation, manifested by a subnormal level of consciousness or seizures, and often accompanied by difficulty with initiating and maintaining respiration and depression of tone and reflexes [5].
Appropriately, the ACOG definition does not imply a specific underlying pathophysiology. The nature of brain injury causing neurologic impairment in a newborn is often unknown or poorly understood. While neonatal encephalopathy was once automatically ascribed to hypoxia-ischemia [6], it is now known that hypoxia-ischemia is only one of many possible contributors to neonatal encephalopathy.
Some investigators require stringent criteria for using the term neonatal encephalopathy, such as two or more symptoms of encephalopathy lasting over 24 hours [7], while others require no more than a low five-minute Apgar score [8]. However, the use of Apgar scores alone is problematic, as Apgar scores may be low due to maternal analgesia or prematurity, or can be normal in the presence of acute hypoxia-ischemic injury, or can be inflated in actual clinical practice [9].
Neonatal encephalopathy usually refers to central nervous system dysfunction in term and near-term infants, but for the purposes of this review, encephalopathy of the preterm infant has also been included.
When neonatal encephalopathy is indisputably due to hypoxic-ischemic (anoxic) brain injury (see 'Hypoxic-ischemic injury' below), it is appropriate to use the term hypoxic-ischemic encephalopathy (HIE) [10]. In addition to watershed area lesions, the presence of a thalamus L-sign is a proposed biomarker of partial, prolonged hypoxia-ischemia [11]. Since the precise cause and temporal onset of neonatal encephalopathy is unknown in most cases, some experts advocate calling the condition "presumed HIE" or "apparent HIE" when the clinical features and neonatal brain injury patterns on MRI suggest that HIE is the most likely mechanism [12]. It remains to be established whether neuroimaging or other testing can one day be used to determine when prenatal hypoxia, birth asphyxia, or hypoxic-ischemic brain injury is responsible for neonatal encephalopathy. (See "Clinical features, diagnosis, and treatment of neonatal encephalopathy".)
Timing of insult — A common but crucial problem is the inability to time the onset, duration, magnitude, and the single or repetitive nature of the exact insult that causes brain injury resulting in neonatal encephalopathy [13]. Early insults could explain why even after hypothermia treatment, outcomes remain suboptimal. (See "Clinical features, diagnosis, and treatment of neonatal encephalopathy", section on 'Range of outcomes'.)
The uncertain timing and etiology of brain injury in most cases of neonatal encephalopathy also fuels birth injury malpractice litigation. Malpractice cases, and too often clinicians, typically focus on events around the time of delivery, which happens to be the time (hours) when the majority of data from pregnant women are obtained, whereas the rest of pregnancy is relatively unmonitored [6]. However, it is usually unknown whether the ultimate brain injury is caused by the events only around delivery or by cumulative insults throughout pregnancy.
The definition of asphyxia is "…a condition of impaired blood gas exchange leading, if it persists, to progressive hypoxemia and hypercapnia. Diagnosis requires a blood gas" [14]. However, even with state-of-art monitoring, there is presently no reliable measure of brain function, brain oxygenation, or cerebral blood flow during the prenatal period or even in the intrapartum period. Therefore, the terms "birth asphyxia" and "fetal distress" are not always used appropriately [15].
Data from studies of neonatal encephalopathy using brain MRI, near-infrared spectroscopy and electroencephalogram monitoring suggest that the immediate perinatal period is important for evolution of brain injury in many cases [16]. One report evaluated 351 term infants with either neonatal encephalopathy (defined as the presence of abnormal tone, feeding difficulties, altered alertness, and at least three of several criteria suggesting possible perinatal hypoxic-ischemia) or seizures alone during the first three days of life [17]. Brain MRI was performed in the first one to two weeks after birth.
●In the group with encephalopathy, lesions suggestive of acute brain injury were found in 80 percent; most of the lesions were bilateral abnormalities in basal ganglia, thalami, cortex, or white matter, although focal infarction was detected in eight infants.
●In the group with only neonatal seizures, acute ischemic or hemorrhagic strokes were found in 69 percent.
Clinical signs that point to an early antenatal onset of neonatal encephalopathy include intrauterine growth restriction, small head size (if both head and body size are small then the insult could be in the first two trimesters of pregnancy), and features suggestive of arthrogryposis. (See 'Risk factors' below.)
RISK FACTORS — Few studies have adequately evaluated risk factors for neonatal encephalopathy other than hypoxia-ischemia. The studies evaluating prenatal and obstetric factors often include symptoms but not pathogenic events that could provide information regarding the timing of the hypoxic-ischemic event. Epidemiologic population studies of neonatal encephalopathy typically lack brain MRI data to determine the presence and degree of brain injury, and also lack information regarding long-term outcomes. By contrast, studies of neonatal encephalopathy that do include neuroimaging data are rarely population-based and are underpowered to determine the effect of a broad range of maternal antenatal risk factors.
●In a large population-based cohort of cases of neonatal encephalopathy from Western Australia, 69 percent had only antepartum risk factors, 25 percent had both antepartum and intrapartum risk factors, 4 percent had evidence of only intrapartum hypoxia, and 2 percent had no identified risk factors [18]. Thus, approximately 70 percent of neonatal encephalopathy cases were associated with risk factors arising before the onset of labor [19].
●Similarly, in a registry of over 4100 infants with neonatal encephalopathy, 46 percent had fetal risk factors and 27 percent had maternal risk factors predating the onset of labor, while only 15 percent had a clinically recognized sentinel event capable of causing asphyxia (35 percent if fetal bradycardia was included as an indicator) [20].
●A case-control study from Italy compared 27 term infants with neonatal encephalopathy and 100 control infants, suggesting a combination of antepartum and intrapartum events explain moderate to severe neonatal encephalopathy [21]. Compared with controls, neonates with encephalopathy had more frequent antepartum (74 percent versus 18 percent) and intrapartum (67 percent versus 19 percent) risk factors, including acute intrapartum events (33 percent versus 2 percent). On the whole, 26 percent of cases of neonatal encephalopathy had only antepartum risk factors, 22 percent had only intrapartum risk factors, and 44 percent had a combination of the two.
●In a case-control study in Ireland that compared 237 term infants with neonatal encephalopathy with 489 control infants, variables independently associated with neonatal encephalopathy included meconium, oligohydramnios, and obstetric complications, suggesting involvement of a combination of antepartum and intrapartum risk factors [22].
The disparate results of these reports are likely due to several reasons, including different inclusion/exclusion criteria among the studies and the assessment of variables that do not necessarily lead to critical brain injury (eg, shoulder dystocia, meconium aspiration, and abnormal fetal heart rate rhythms are ominous only if associated with fetal hypoxia, which is rare).
Antepartum — Most cases of neonatal encephalopathy have their antecedents in the prenatal period. It is unknown whether neonatal encephalopathy occurs as a result of a single insult (such as hypoxia-ischemia), multiple insults (eg, infection plus hypoxia-ischemia), or combinations of acute or chronic conditions. In cases with multiple insults, it is possible that the one closest to birth might be only a minor event that tips the balance to irreversible injury.
A population-based study that evaluated risk factors for neonatal encephalopathy compared 164 infants with neonatal encephalopathy and 400 randomly selected controls from term infants born in Western Australia [7]. The study identified a number of antepartum risk factors that can be grouped under categories based on the maternal-placental-fetal unit (figure 1):
●Maternal
•Preconceptual factors including maternal unemployment, family history of seizures or neurologic disorder, and infertility treatment
•Maternal thyroid disease
●Placental
•Severe preeclampsia
•Post-dates
•Abnormal appearance of the placenta
●Fetal
•Intrauterine growth restriction
•Predisposition of the fetus to injury
IUGR — Among these antepartum risk factors, intrauterine growth restriction (IUGR) was the strongest (relative risk [RR] 38.2, 95% CI 9.4-154.8) [7]. Although most babies with neonatal encephalopathy do not meet the criteria of IUGR, a small hospital-based case-control study found that a greater proportion of infants with neonatal encephalopathy were below the 10th percentile of growth potential compared with controls, and the difference was statistically significant [23]. These studies suggest that there are antenatal factors contributing to the brain injury. Unfortunately, IUGR provides no clue to the etiology (figure 2) because both external maternal and placental factors can affect fetal growth in addition to intrinsic factors. Of note, coronavirus disease 2019 (COVID-19) infections in mothers have rarely presented with newborn brain injury, but there are case reports of later encephalopathy at 29 days of life [24].
Genetic — The predisposition of the fetus to injury may be dependent on genetic and epigenetic causes. In a study of 28 patients with neonatal encephalopathy who had no risk factors for moderate to severe hypoxic-ischemic encephalopathy, whole exome sequencing identified pathogenic variants (including CDKL5, pyruvate dehydrogenase, CFTR, CYP21A2, ISY1, KIF1A, KCNQ2, SCN9A, MTFMT, and NPHP1) in 36 percent of cases [25]. In another study of 14 babies with neonatal encephalopathy without a history of perinatal asphyxia, pathogenic variants (SCN2A, KCNQ2, and GNAO1) and possible pathogenic variants (LIAS and CUL4B) were identified in 5 (36 percent) [26]. A similar study of 366 sick patients from China with neonatal encephalopathy found pathogenic or likely-pathogenic genetic variants (most commonly KCNQ2 and SCN2A) in 12 percent [27].
Genetic testing is not recommended for all patients with neonatal encephalopathy, given the challenges of cost, time pressure to screen patients before hypothermia, screening parents for acceptance of possible results, turn-around time of testing, unknown causation with finding of possible pathogenic variants, and ethical considerations regarding secondary findings (eg, pathogenic variants associated with adult-onset conditions) [28,29]. Whole exome sequencing have a role in the workup of neonatal encephalopathy without any identifiable risk factors for asphyxia or other prenatal causes [25], especially if intractable seizures are also present.
Placental — Placental findings of fetal vascular malperfusion, thrombosis, and inflammation have been associated with neonatal encephalopathy [30-34].
●In a study of prospectively collected placental pathologic specimens among 73 infants with neonatal encephalopathy and 253 unaffected control infants, lesions consistent with global fetal vascular malperfusion were more common in cases than in controls (20 versus 7 percent ) [30].
●In a hospital-based case-control study comparing 93 cases of neonatal encephalopathy to 387 controls, placental findings of fetal thrombotic vasculopathy, funisitis, and accelerated villous maturation were independently associated with neonatal encephalopathy [31].
●Another study found that the frequency of severe placental lesions was fivefold higher among 83 cases of neonatal encephalopathy from a medicolegal registry than among 250 controls (52 to 10 percent). These lesions included fetal thrombotic vasculopathy, chronic villitis with obliterative fetal vasculopathy, chorioamnionitis with severe fetal vasculitis, and meconium-associated fetal vascular necrosis [32].
●In a retrospective study of 100 term newborns who received hypothermia therapy for neonatal encephalopathy, placental abnormalities were more common among newborns (n = 49) who did not have a sentinel event (ie, a clinical history of disruption of blood flow to the fetus during delivery) such as placental abruption, uterine rupture, tight nuchal cord or cord prolapse [33]. As an example, an inflammatory pathology was significantly more frequent in infants without sentinel events (43 percent, versus 14 percent for infants with sentinel events).
Most of the placental lesions result in some form of hypoxic-ischemic damage. Placental lesions may underlie the finding of some studies that >41 weeks gestation is an antepartum risk factor [35]. A prospective case-control study of associations between placental abnormalities among infants with hypoxic-ischemic encephalopathy (n = 30) and three control groups (infants born by repeat cesarean section [n = 50], infants small for gestational age [n = 80], and infants requiring positive-pressure ventilation for >30 seconds at birth [n = 70]) found that fetal vascular malperfusion but not chorioamnionitis or maternal vascular malperfusion was distinctively observed in cases of hypoxia-ischemic encephalopathy, and that absence of placental findings was lowest in the hypoxia-ischemia group [36]. A case series described seven term newborns with neonatal encephalopathy who had been exposed to chronic or acute placental inflammation and/or hypoxic-ischemia. MRI showed T2 corticosubcortical hyperintensities (a pattern unlike the neuroimaging appearance of intrapartum hypoxic-ischemic encephalopathy) at postnatal days 0 to 4 that progressed to cystic encephalomalacia on days 9 to 12, pointing to an antenatal cause for newborn brain injury [37].
Intrapartum — Intrapartum risk factors for neonatal encephalopathy can be grouped as follows [18,20,35]:
●Persistent occipitoposterior position
●Shoulder dystocia
●Emergency cesarean delivery, which may include failed vacuum
●Operative vaginal delivery
●Acute intrapartum events or sentinel events (eg, uterine rupture, placental abruption, cord prolapse, tight nuchal cord, maternal shock/death)
●Inflammatory events (eg, maternal fever, chorioamnionitis, prolonged rupture of membranes)
An acute intrapartum event, such as a placental abruption or uterine rupture, conferred a fourfold increased risk of neonatal encephalopathy, but was present in only 8 percent of infants with neonatal encephalopathy [18]. Uterine rupture alone is associated with only a 2 to 3 percent incidence of neonatal death and a 2 to 6 percent incidence of neonatal encephalopathy [38,39]. In the series of 158 medicolegal cerebral palsy cases, sentinel intrapartum events were present in 11 percent [40].
Outcomes in another study of birth sentinel events with a minimum of 12 months follow-up included death in 20 percent, cerebral palsy in 41 percent, developmental delay in 15 percent, and normal development in 24 percent [41]. The latter two numbers suggest that plasticity and repair responses often determine outcome to well-defined single insults.
Some of the so-called intrapartum risk factors include obstetric treatments to prevent further fetal hypoxia, such as emergency cesarean delivery and operative vaginal delivery. These may or may not be true risk factors depending upon the duration of the underlying insult. In addition, increased duration of second stage of labor related to shoulder dystocia or failed vacuum may not necessarily result in critical brain injury unless accompanied by fetal hypoxia. Even after including newborn clinical criteria of distress in the diagnosis of suspected birth asphyxia, one report found that suspected birth asphyxia accounted for only 41 percent of the children with moderate or severe neonatal encephalopathy associated with later cerebral palsy [42].
Some inflammatory factors, such as prolonged rupture of membranes, may exert a pathogenic influence even before terminal labor. The importance of inflammation as a risk factor for neonatal encephalopathy is illustrated by the following reports [43]:
●In a population-based report that compared 1060 newborn cases of neonatal encephalopathy with 5330 unaffected control newborns, independent risk factors for neonatal encephalopathy were isolated intrapartum maternal fever (RR 3.1, 95% CI 2.3-4.2) and chorioamnionitis (RR 5.4, 95% CI 3.6-7.8) [44].
●In a cohort study that identified 25 cases of moderate to severe neonatal encephalopathy from 8299 term births, maternal fever had a sixfold increased risk of neonatal encephalopathy compared with a 12-fold increase for acidosis [45]. Although there was a multiplicative increased risk with acidosis (76-fold), the effect of maternal fever seemed to have no statistical interaction with acidosis, implying that maternal fever and acidosis represent different causal pathways.
●In a prospective study of infants exposed to maternal chorioamnionitis, there was a threefold increase in neonatal depression and neonatal intensive care unit admission for newborns with elevated temperature [46].
●A retrospective cohort study of 73 infants, 23 with mild and 50 with moderate-to severe hypoxic-ischemic encephalopathy, found that fetal inflammatory reaction of the placenta was higher in moderate-to-severe encephalopathy compared with mild encephalopathy (46 versus 17 percent, odds ratio 6.29, 95% CI 1.5-25), but the maternal inflammatory reaction was not significantly different (56 versus 34 percent) [47].
Other studies have found that maternal fever, often accompanied by a diagnosis of chorioamnionitis, is associated with low Apgar scores, neonatal seizures, and a diagnosis of "birth asphyxia" among infants who develop cerebral palsy [48,49].
An interaction between brain injury due to inflammation and hypoxia-ischemia has been suggested by the finding in case-control studies of an association between maternal chorioamnionitis and cerebral palsy in children with evidence of hypoxic-ischemic brain injury [48,50,51], and by the observation of increased cytokines in the cerebrospinal fluid of patients with neonatal encephalopathy [52]. Blood cytokines are associated with MRI injury in neonatal encephalopathy but the increase in both pro-inflammatory and anti-inflammatory cytokines make it hard to interpret antecedents of brain injury [53,54].
The need for resuscitation in the delivery room is itself a poor prognostic sign as it is associated with an increased risk at eight years of age of having a lower (<80) IQ score, even if the infant does not exhibit encephalopathy in the newborn period (odds ratio 1.65, 95% CI 1.13-2.43) [55].
Hypoxia-ischemia can result in fetal heart rate abnormalities and neonatal acidemia, which are often cited as risk factors. Fetal heart rate variables are not considered as good as umbilical cord acidemia for estimation of timing of birth insults [21,56] although both have deficiencies. The correlation of fetal heart rate abnormalities with umbilical acidemia may have a stronger association with the presence of intrauterine vascular disease (ie, preeclampsia, placental abruption, birth weight <10th percentile, or histologic evidence of placental infarction or severe vascular pathology) than with acute intrapartum events [57]. Neurologic morbidities and death are significantly more common in newborns with a pH <7.0 than in those with a pH ≥7.0, but the majority of acidemic neonates do not have any major morbidity [56].
HYPOXIC-ISCHEMIC INJURY — As noted earlier, it is appropriate to use the term hypoxic-ischemic encephalopathy (HIE) when neonatal encephalopathy is due to hypoxic-ischemic brain injury. (See 'Terminology' above.)
The signs and symptoms of HIE, as well as outcome, depend upon several factors:
●The immediate nervous system injury sustained during the hypoxic-ischemic insult
●Physiologic properties that lead to selective vulnerabilities of certain cell populations
●The presence of endogenous protective mechanisms
●Consequences of hypoxia-ischemia that lead to secondary injuries (eg, reperfusion injury, edema, increased intracranial pressure, abnormalities of autoregulation, and hemorrhage)
Hypoxia can result from ischemia (ie, a lack of sufficient blood flow to all or part of an organ), insufficient inspired oxygen, or inadequate blood oxygen-carrying capacity (eg, inadequate oxygen in inspired air, severe anemia, carbon monoxide poisoning). Regardless of the cause, cardiac and vascular compromise ultimately occur when hypoxia is prolonged. The result is hypotension, ischemia, and anaerobic metabolism leading to lactic acidosis. Thus, ischemia is both a cause and a result of hypoxia and compounds the complications of hypoxia by impairing the removal of metabolic and respiratory by-products (eg, lactic acid, carbon dioxide).
Fetal hypoxic-ischemic brain injury and subsequent HIE can occur by one or more of the following mechanisms (table 1):
●Maternal, via impaired oxygenation (eg, asthma, pulmonary embolism, pneumonia) or inadequate perfusion of maternal placenta (eg, cardiorespiratory arrest, maternal hypotension, preeclampsia, chronic vascular disease)
●Placental, via abruptio placenta, tight nuchal cord, cord prolapsed, true knot, or uterine rupture (see "Placental pathology: Findings potentially associated with neurologic impairment in children")
●Fetal, via impaired fetal oxygenation/perfusion (eg, fetomaternal hemorrhage, fetal thrombosis)
Antecedent events and risk factors — Brain injury from HIE may occur before the onset of labor. Supporting evidence comes from a neuropathologic study of 70 infants who died within seven days of birth [58]. Using criteria of low five-minute Apgar score and umbilical cord or initial blood gas pH <7.1 for asphyxia, findings consistent with brain damage before the onset of labor were present in all asphyxiated and encephalopathic infants. In addition, the same findings were present in 38 percent of term infants, 52 percent of preterm infants, and 1 of 12 infants without any evidence of birth asphyxia.
This report reveals that some cases of HIE die before manifesting brain injury or clinical signs of neonatal encephalopathy [58]. In addition, brain injury associated with HIE may be present without clinical signs of neonatal encephalopathy. However, there are limitations to extrapolating findings from autopsy studies in that the nature of death, failed resuscitation, and terminal drugs are not controlled for in most instances. What factors result in death in HIE is also unknown because quite a few infants have elective withdrawal of life support [59].
A major risk factor for HIE is that of multiple gestations, particularly the presence of monochorionic twins. A study of preterm infants of multiple gestations found that the incidence of antenatal white matter necrosis on cranial ultrasound was significantly higher in monochorionic than in dichorionic infants (30 versus 3.3 percent) [60].
In preterm infants meeting criteria for HIE, placental abruption is more likely to be identified as the antecedent event [61] than uterine rupture and cord prolapse, which are more common sentinel events among term infants diagnosed with HIE [41]. In preterm infants, HIE is associated with injury on 36-week MRI scan involving the basal ganglia (mostly severe), white matter (mostly mild), brainstem, and cortex in 75, 89, 44, and 58 percent, respectively [61].
These studies emphasize the importance of early brain imaging in documenting, and possibly timing, brain lesions, as well as the importance of postmortem examinations in cases of stillbirth and neonatal deaths. (See "Clinical features, diagnosis, and treatment of neonatal encephalopathy".)
Acute events — Determining whether an acute hypoxic-ischemic event contributed to neonatal encephalopathy is challenging, since there is no gold standard. The various clinical signs of HIE, including low Apgar scores, low cord pH, neonatal seizures and encephalopathy, are nonspecific and may occur in the absence of global hypoxic-ischemic brain injury or long-term neurologic sequelae. A consensus statement from the American College of Obstetricians and Gynecologists (ACOG) notes that neonatal encephalopathy related to acute hypoxia-ischemia is presumed to be associated with abnormal neonatal signs and contributing events, in close temporal proximity to labor and delivery, that are consistent with an acute hypoxic-ischemic event [5]. Markers that are helpful for determining the likelihood that an acute peripartum or intrapartum hypoxic-ischemic event contributed to the development of neonatal encephalopathy are as follows (table 2):
●Neonatal signs consistent with an acute peripartum or intrapartum event:
•Apgar score of <5 at 5 minutes and 10 minutes
•Fetal umbilical artery pH <7.0, or base deficit ≥12 mmol/L, or both
•Acute brain injury seen on brain MRI or magnetic resonance spectroscopy consistent with hypoxia–ischemia
•Presence of multisystem organ failure consistent with HIE
●Contributing factors consistent with an acute peripartum or intrapartum event:
•A sentinel hypoxic or ischemic event occurring immediately before or during labor and delivery
•Fetal heart rate monitor patterns consistent with an acute peripartum or intrapartum event
•Brain injury patterns based on imaging studies consistent with an etiology of an acute peripartum or intrapartum event
•No evidence of other proximal or distal factors that could be contributing
●Developmental outcome is spastic quadriplegia or dyskinetic cerebral palsy
Level and duration of hypoxia-ischemia — The level of hypoxia-ischemia that causes neonatal encephalopathy is unknown, but animal studies provide some information. There are two experimental paradigms that present with different pathophysiological pathways: umbilical cord occlusion and acute placental insufficiency.
●Occlusion of the umbilical cord results in cardiovascular compromise because of the removal of a low resistive vascular bed
●In sheep, fetuses can survive up to 30 minutes of occlusion with increasing brain damage observed in term sheep compared with premature sheep fetuses [62], while 20-minute occlusion may not cause any brain injury [63]
●In non-human primates, it was long believed that permanent neurologic injury occurred with occlusion of 12 to 17 minutes [64,65]
The clinical corollary of umbilical cord prolapse has a more varied response, probably because of the presence of some blood flow in the prolapsed cord. In humans, umbilical cord prolapse is an obstetric emergency except for the extreme premature gestation mother. In a chart review of 87 cases of cord prolapse among 36,500 deliveries, the median time from discovery to delivery was 15 minutes, with the longest being 14 hours [66]. There was no relation between time of discovery to delivery and postnatal mortality or morbidity [66]. The longest tolerated time of umbilical cord prolapse without major consequences (three days) was observed in an extremely premature infant [67], suggesting that the duration becomes critical only near term.
●Acute placental insufficiency results in fetal compromise due to impaired ability to exchange gas and nutrients.
●A study of acute placental insufficiency (via uterine ischemia) in rabbits at 70 percent gestation found that animals subjected to 30 minutes of global hypoxia-ischemia were no different than controls [68]. However, 40 minutes of global hypoxia-ischemia increased fetal mortality from 0 to 25 percent, and increased marked motor deficits at birth in the survivors from 0 to 75 percent [68]. In a population of normal fetuses, the susceptibility to injury is not dependent solely on the duration of hypoxia-ischemia. Rabbit fetuses at 79 percent gestation undergoing additional reperfusion-reoxygenation injury just after the cessation of hypoxia-ischemia have a greater chance of motor deficits that in those without reperfusion-reoxygenation injury [69].
●When experimental global hypoxia-ischemia is mild and chronic, it results in intrauterine growth restriction but may or may not result in brain injury [70,71].
Unfortunately, the onset of acute placental insufficiency states such as placental abruption in humans is almost always unknown.
The spectrum of hypoxic-ischemic injury and outcome can be summarized as in the Figure (figure 3). The intensity of the insult can be modified by prior events that may serve as a preconditioning stimulus. Also, there is a complex interaction of infection with hypoxia-ischemia. As an example, preeclampsia may be a protective factor for infants born to mothers with chorioamnionitis and at risk for cerebral palsy [72].
Vulnerable regions of developing brain — Hypoxia-ischemia may have deleterious effects on vulnerable cell populations peculiar to the developmental stage (figure 4), causing discrete injuries that could also affect seizure threshold or cognition.
●Late oligodendroglial progenitors are vulnerable to injury in early prematurity with resulting predominant white matter injury in premature infants [73].
●In the term neonate with ischemic brain injury, however, certain neurons in the deep gray nuclei and perirolandic cortex are most likely to be affected. Acute cell injury can trigger continuing loss of cells. There are neural-glial cell interactions that can increase the brain injury. Selective damage to neurons in the subcortical gray matter can directly contribute to long-term apoptosis in distal neuronal structures.
Both gray and white matter injury occur in preterm and term neonates with HIE. Ideally, MRI of patients with suspected neonatal encephalopathy should be done early and repeated at a later date to get a better sense of timing of insult and the severity of brain injury [37]. Early detection of recent hypoxic-ischemic insults usually depends upon apparent diffusion coefficient (ADC) measurements on MRI. Beyond a week after the onset of the insult, evidence of gray matter injury by neuroimaging is scant unless there is a significant decrease in volume of gray matter regions or an increase in ventricular size or obvious infarcts and hemorrhage. White matter injury is somewhat easier to detect by diffusion tensor imaging (DTI), as the fractional anisotropy of white matter bundles normally increase with age, leading to an ability to detect minor decreases in fractional anisotropy in WM regions. However, while DTI is performed at some tertiary centers, it is not yet widely available in clinical practice.
Mechanisms of neuronal injury — Hypoxia-ischemia initially causes energy failure and loss of mitochondrial function. This is accompanied by membrane depolarization, brain edema, an increase of neurotransmitter release and inhibition of uptake, and an increase of intracellular calcium that sets off additional pathologic cascades [74]. These include oxidative stress, with the production of reactive oxygen species and interaction with nitric oxide pathway to produce reactive nitrogen species [75].
It was once believed that reactive species caused damage only if antioxidant defenses were overwhelmed, thus upsetting the balance between oxidants and antioxidants. However, it is now realized that the interaction itself between reactive species and antioxidant defenses ultimately causes cellular injury and death (ie, the yin-yang theory of both being necessary) [76]. Reperfusion exacerbates the oxidative stress with a burst of reactive oxygen species.
The response of the fetus to the hypoxic-ischemic insult determines the subsequent injurious cascades and the clinical manifestations that result. One study monitored the response of rabbit fetus brains in utero to global hypoxia using MRI diffusion-weighted sequences and ADC mapping as a marker of ischemic injury [77]. Fetuses that showed a precipitous drop in brain ADC at the end of 40 minutes of global hypoxia manifested hypertonia and postural changes after birth, while those without a drop in ADC were relatively normal at birth. Thus, after the hypoxic-ischemic insult, the initial energy failure and oxidative stress probably play a critical role in subsequent cascades (figure 5) [69].
Excitotoxic injury — Excitotoxic cellular injury occurs via excess activation of glutamate receptors, which leads to several forms of cell death. There are four receptor types for glutamate [78], which are the N-methyl-d-aspartate (NMDA), alpha-amino-3-hydroxy-5-methyl-4-isoxazoleproprionic acid (AMPA), kainate, and metabotropic glutamate receptors. The metabotropic receptors are not directly coupled to ion channels.
●The NMDA receptors are the most avid and physiologically active. The channels activated by NMDA receptors are voltage-dependent and calcium-permeable. Their activation causes neuron depolarization [79]. Repeated depolarization of a neuron by unregulated glutamate release results in accumulation of intracellular calcium. During hypoxia-ischemia, there is failure to rapidly pump synaptically released glutamate back across the cell membrane, resulting in exposure of NMDA receptors to accumulated glutamate, which leads to lethal elevation of intracellular calcium levels. The cascade of events initiated by this process also can induce apoptosis [80].
●AMPA and kainate receptors are both coupled to sodium and potassium ion channels. Whereas NMDA receptors are always permeable to Ca2+, cation permeability of AMPA receptors depends on subunit composition. Ca2+ influx is differentially regulated by AMPA receptors compared with kainate receptors.
Oligodendroglia are particularly vulnerable to glutamate [81]. Preoligodendrocyte subtypes O4 and O1+ express subunits for both the AMPA (GluR1, GluR2, GluR3, and GluR4) and kainate (KA1, GluR5/6, and GLuR7) receptors but not NMDA receptors, whereas mature MBP+ oligodendrocytes have little to no expression of either NMDA receptors or non-NMDA receptors.
Mature oligodendrocytes in mixed cocultures die after exposure to kainate, but AMPA receptors are the most important mediators of cellular demise, with kainate receptors playing a smaller role [82]. In this paradigm, cell death occurs predominantly by necrosis, not apoptosis [82]. However, there is evidence that mature oligodendrocytes expressing myelin basic protein are resistant to excitotoxic injury produced by kainate, whereas earlier stages in the oligodendrocyte lineage are vulnerable to this insult [83].
Nitric oxide and oxygen-free radicals — Nitric oxide (NO) and oxygen-free radicals appear to play important roles in brain injury induced by hypoxia-ischemia.
Nitric oxide can behave as an oxidant as well as an antioxidant. Under pathological conditions there is excess NO production that results in cell toxicity through direct biochemical effects or through reactive nitrogen species that is formed from the reaction of NO and reactive oxygen species [84].
Nitric oxide is synthesized by nitric oxide synthase (NOS) from L-arginine in the presence of essential cofactor, tetrahydrobiopterin. Nitric oxide synthase exists in three isoforms: neuronal NOS (nNOS), endothelial NOS (eNOS), and inducible NOS (iNOS).
Available evidence suggests that eNOS has a predominant protective role in hypoxia-ischemia, whereas nNOS and iNOS have a facilitative role.
●Histopathologic studies have shown that nNOS knockout neonatal animals are protected from focal hypoxic-ischemic-induced histopathologic brain damage [85].
●Similarly, iNOS knockout animals show a reduction of focal ischemic brain damage and locomotor deficits [86].
●Animals lacking the eNOS gene have enlarged cerebral infarcts after ischemic injury [87].
The use of specific NOS inhibitors as neuroprotectants is currently being studied.
PERINATAL STROKE — Perinatal stroke is an increasingly recognized entity in term newborns with encephalopathy and cerebral palsy. Perinatal stroke occurs approximately once in 4000 births. (See "Stroke in the newborn: Classification, manifestations, and diagnosis", section on 'Epidemiology'.)
The majority of infants with ischemic perinatal stroke develop neonatal seizures. Additional signs of neonatal encephalopathy may also be present, such as lethargy, hypotonia, feeding difficulties, or apnea [88].
A specific cause for perinatal stroke is not identified in most affected newborns. Factors contributing to the risk include maternal conditions such as prothrombotic disorder and cocaine abuse; placental complications such as preeclampsia, chorioamnionitis and placental vasculopathy; and newborn conditions such as prothrombotic disorders, congenital heart disease, meningitis, and systemic infection [89]. During the delivery process, an infant may develop a cervical arterial dissection that leads to stroke. Perinatal stroke may be more likely to be associated with antepartum chronic events in the placenta compared with hypoxic-ischemic neonatal encephalopathy [90].
Potential long-term sequelae of perinatal arterial stroke include cerebral palsy, cognitive deficits, hemiparesis, and epilepsy. However, development is normal in approximately 19 to 33 percent of infants with neonatal ischemic infarction. (See "Stroke in the newborn: Management and prognosis", section on 'Prognosis of arterial ischemic stroke'.)
PROGRESSIVE ENCEPHALOPATHY — One must always consider the possibility of progressive disorders in cases of neonatal encephalopathy. Rapidly-evolving injury on MRI often helps identify these cases. These include metabolic, neurodegenerative, infectious or toxic etiologies that are rare, with a combined incidence of approximately 6 per 10,000 live births [91], but a much higher mortality rate than the general population [92]. A history of parental consanguinity is associated with a marked increase in the risk of progressive encephalopathy, and thus is an important clue suggesting metabolic and neurodegenerative disease [93].
Metabolic abnormalities — A large number of metabolic and genetic abnormalities may cause neonatal encephalopathy. However, metabolic and genetic disorders account only for a very small proportion of cases of neonatal encephalopathy.
Inborn errors of metabolism that present in the newborn period typically share strikingly similar clinical features, including decreased level of consciousness, seizures, poor feeding, hypotonia, and vomiting. Examples include:
●Disorders of amino acid metabolism (eg, maple syrup urine disease, phenylketonuria, nonketotic hyperglycinemia) (see "Overview of maple syrup urine disease" and "Overview of phenylketonuria")
●Hyperammonemia (eg, urea cycle defects) (see "Urea cycle disorders: Clinical features and diagnosis")
●Neonatal hypoglycemia (see "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia")
●Organic acidemias (see "Organic acidemias: An overview and specific defects")
●Mitochondrial disorders (see "Mitochondrial myopathies: Clinical features and diagnosis")
●Severe peroxisomal disorders (eg, Zellweger syndrome) (see "Peroxisomal disorders")
Specific disorders such as isolated sulfite oxidase deficiency and molybdenum cofactor deficiency (sulfite oxidase deficiency with other enzyme deficiencies) may produce neuroimaging and clinical findings that very closely mimic hypoxic-ischemic brain injury [94-96]. (See "Etiology and prognosis of neonatal seizures", section on 'Inborn errors of metabolism'.)
Genetic disorders such as Prader-Willi and chromosomal abnormalities may also present with newborn encephalopathy. (See "Prader-Willi syndrome: Clinical features and diagnosis", section on 'Infancy'.)
OTHER CAUSES — Given that neonatal encephalopathy is an umbrella term that includes any type of brain injury or insult resulting in central nervous system dysfunction, the list of brain disorders that can cause neonatal encephalopathy is quite long [97]. As examples, brain anomalies, intracranial hemorrhage, and infection can all lead to seizures and encephalopathy in the newborn period.
Intraventricular hemorrhage in term infants may cause symptoms of neonatal encephalopathy and is often related to sinovenous thrombosis as opposed to the more typical intraventricular hemorrhage associated with germinal matrix hemorrhage seen in preterm infants [98]. Intracerebral hemorrhage in a term infant is often idiopathic, but may be related to birth trauma, congenital vascular malformation, or a clotting disorder.
Finally, a variety of maternal toxins can cause encephalopathy in the newborn period. For instance, passive addiction to narcotics, barbiturates, alcohol, tricyclic antidepressants, and serotonin reuptake inhibitors can produce seizures and encephalopathy in the neonate [99].
SUMMARY
●Terminology – Neonatal encephalopathy is the preferred terminology to describe central nervous system dysfunction in the newborn period. It can result from a wide variety of conditions but often remains unexplained. The nature of brain injury causing neurologic impairment in a newborn is poorly understood. Hypoxia-ischemia is only one of many possible contributors to neonatal encephalopathy. Whether a particular newborn's encephalopathy can be attributed to hypoxic-ischemic brain injury is often controversial. (See 'Terminology' above.)
●Antepartum risk factors – Most cases of neonatal encephalopathy are associated with risk factors arising before the onset of labor (figure 1 and figure 2). These include:
•Maternal factors, including unemployment, family history of seizures or neurologic disorder, infertility treatment, and thyroid disease
•Placental conditions, including severe preeclampsia, post-dates, and abnormal appearance of the placenta
•Fetal problems, such as intrauterine growth restriction
Of these, intrauterine growth restriction is the strongest risk factor. (See 'Risk factors' above and 'Antepartum' above.)
●Intrapartum risk factors – Intrapartum risk factors for neonatal encephalopathy include the following:
•Acute intrapartum events or sentinel events (eg, uterine rupture, placental abruption, cord prolapse, tight nuchal cord, maternal shock/death)
•Inflammatory events (eg, maternal fever, chorioamnionitis, prolonged rupture of membranes)
•Persistent occipitoposterior position
•Shoulder dystocia
•Emergency cesarean delivery, which may include failed vacuum
•Operative vaginal delivery
An acute (ie, sentinel) intrapartum event, such as a placental abruption or uterine rupture, confers an increased risk of neonatal encephalopathy, but is present in only a minority of infants with neonatal encephalopathy. (See 'Intrapartum' above.)
●Perinatal stroke – Perinatal stroke is a separate recognized entity in term newborns with encephalopathy. (See 'Perinatal stroke' above.)
●Progressive encephalopathy – Metabolic and neurodegenerative disorders can underlie neonatal encephalopathy. (See 'Progressive encephalopathy' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Yvonne Wu, MD, MPH, who contributed to an earlier version of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟