Intracranial Hemorrhage
Janet S. Soul
OVERVIEW
The incidence of intracranial hemorrhage (ICH) varies from 2% to >30% in newborns, depending on gestational age (GA) at birth and the type of ICH. Bleeding within the skull can occur: (i) external to the brain into the epidural, subdural, or subarachnoid spaces; (ii) into the parenchyma of the cerebrum or cerebellum; or (iii) into the ventricles from the subependymal germinal matrix or choroid plexus (Table 54.1). The incidence, pathogenesis, clinical presentation, diagnosis, management, and prognosis of ICH varies according to the ICH location and size, and the infant’s GA (1, 2). There is often a combination of two or more types of ICH, as an ICH in one location often extends into an adjacent compartment; for example, extension of a parenchymal hemorrhage into the subarachnoid space or ventricles. Although intraventricular hemorrhage (IVH) may be associated with parenchymal hemorrhage, this parenchymal hemorrhage is not an extension of the IVH, but a hemorrhagic venous infarction caused by obstruction of the terminal vein.
Diagnosis usually depends on clinical suspicion when an infant presents with typical neurologic signs, such as seizures, irritability, depressed level of consciousness, and/or focal neurologic deficits referable either to the cerebrum or brain stem. Diagnosis is confirmed with an appropriate neuroimaging study. MRI is the optimal imaging modality for almost all types of ICH, but ultrasound (US) is typically preferred for premature infants and critically ill infants who are not stable for transport to MRI. To avoid exposure of newborns to the ionizing radiation associated with CT, CT scan should be used only for emergent imaging studies when neither MRI or US is available/possible. The American Academy of Neurology (AAN) practice parameter states that all infants with a birth GA of <30 weeks should undergo routine cranial ultrasound (CUS) between 7—14 days and optimally repeated between 36 to 40 weeks postmenstrual age, but MRI is not recommended for routine surveillance (3). Our local practice is to obtain a CUS on every newborn with a birth GA of <32 weeks and birth weight < 1,500 grams.
Management varies according to the size and location of the ICH and the presenting neurologic signs. In general, only very large hemorrhages with clinical signs require surgical intervention for removal of the ICH itself. With a large ICH, pressor support or volume replacement (with normal saline, albumin, or packed red blood cells) may be required because of significant blood loss. More commonly, management is focused on treating complications such as seizures or the development of posthemorrhagic hydrocephalus. In general, although a large ICH is more likely to result in greater morbidity or mortality than a small one, the presence and severity of parenchymal injury—whether due to hemorrhage, infarction, or other neuropathology—is usually the best predictor of outcome.
I. SUBDURAL HEMORRHAGE (SDH) AND EPIDURAL HEMORRHAGE (EH)
Etiology and pathogenesis. The pathogenesis of SDH relates to rupture of the draining veins and sinuses of the brain that occupy the subdural space.
Vertical molding, fronto-occipital elongation, and torsional forces acting on the head during delivery may provoke laceration of dural leaflets of either the tentorium cerebelli or the falx cerebri. These lacerations can result in rupture of the vein of Galen, inferior sagittal sinus, straight sinus and/or transverse sinus, and usually a posterior fossa SDH. Breech presentation also predisposes to occipital osteodiastasis, a depressed fracture of the occipital bone or bones, which may lead to direct laceration of the cerebellum or rupture of the occipital sinus. Clinically significant SDH in the posterior fossa often results from trauma in the full-term infant, although small, inconsequential SDH is fairly common in uncomplicated deliveries (the true incidence in apparently well newborns is unknown). SDH in the supratentorial space usually results from rupture of the bridging, superficial veins over the cerebral convexity. Other risk factors for SDH include factors that increase the likelihood of significant forces on the infant’s head, such as large head size; rigid pelvis (e.g., in a primiparous or older multiparous mother); nonvertex presentation (breech, face, etc.); very rapid or prolonged labor or delivery; difficult instrumented delivery; or, rarely, a bleeding diathesis. MRI is the best modality to determine age of ICH. Postnatally, SDH and EH are almost always due to direct head trauma or shaking; hence, nonaccidental injury needs to be suspected in cases of acute presentation of SDH or EH beyond the perinatal period. However, care should be taken not to confuse an old chronic effusion from a birthrelated ICH with an acute postnatally acquired ICH. Careful interpretation of neuroimaging studies, particularly MRI, should distinguish acute SDH or EH from chronic effusion.
Table 54.1 Illustrating Neonatal ICH by Location, and Whether Each ICH Type Is Predominantly Primary (1°) or Secondary (2°) Source of Bleeding, and the Relative Incidence in Preterm (PT) or Term (T) Newborns
Type (location) of hemorrhage
Principal source of ICH
Relative Incidence in PT vs. T
1. Subdural and epidural hemorrhage
1°>2°
T> PT
2. Subarachnoid hemorrhage (SAH)
2°> 1°*
Unknown*
3. Intraparenchymal hemorrhage
Cerebral
2°> 1°
PT>T
Cerebellar
2°> 1°
PT>T
4. Germinal matrix/intraventricular hemorrhage
1°>2°
PT>T
* True incidence unknown, small 1° SAH may be more common than is recognized in both PT and T newborns.
Clinical presentation. When the accumulation of blood is rapid and large, as occurs with rupture of large veins or sinuses, the presentation follows shortly after birth and evolves rapidly. This is particularly true in infratentorial SDH, where compression of the brain stem may result in nuchal rigidity or opisthotonus, obtundation or coma, apnea, other abnormal respiratory patterns, and unreactive pupils and/or abnormal extraocular movements. With increased intracranial pressure (ICP), there may be a bulging fontanelle and/or widely split sutures. With large hemorrhages, there may be systemic signs of hypovolemia and anemia. When the sources of hemorrhage are small veins, there may be few clinical signs for up to a week, at which time either the hematoma attains a critical size, imposes on the brain parenchyma and produces neurologic signs, or hydrocephalus develops. Seizures may occur in up to half of neonates with SDH, particularly with SDH over the cerebral convexity. With cerebral convexity SDH, there may also be subtle focal cerebral signs and mild disturbances of consciousness, such as irritability. Subarachnoid hemorrhage probably coexists in the majority of cases of neonatal SDH, as demonstrated by a cerebrospinal fluid (CSF) exam (4). Finally, a chronic subdural effusion may gradually develop over months, presenting as abnormally rapid head growth, with the occipitofrontal circumference (OFC) crossing percentiles in the first weeks to months after birth.
Diagnosis. The diagnosis should be suspected on the basis of history and clinical signs and confirmed with a neuroimaging study. CT scan is the study of choice for diagnosing SDH or EH for acute emergencies, if MRI cannot be obtained quickly (3). Although CUS may be valuable in evaluating the sick newborn at the bedside, US imaging of structures adjacent to bone (i.e., the subdural space) is often inadequate. MRI has proven to be quite sensitive to small hemorrhage and can help establish timing of ICH. MRI is also superior for detecting other lesions, such as contusion, thromboembolic infarction, or hypoxic-ischemic injury that occurs in some infants with severe hypovolemia/anemia or other risk factors for parenchymal lesions. However, a CT scan is much quicker to obtain and usually adequate in an unstable infant with elevated ICP who may require neurosurgical intervention. When there is clinical suspicion of a large SDH, a lumbar puncture (LP) should not be performed until after the CT is obtained. The LP may be contraindicated if there is a large hemorrhage within the posterior fossa or supratentorial compartment. If a small SDH is found, an LP should be performed to rule out infection in the newborn with seizures, depressed mental status, or other systemic signs of illness, since small SDH are often clinically silent.
Management and prognosis. Most infants with SDH do not require surgical intervention and can be managed with supportive care and treatment of any accompanying seizures. Infants with rapid evolution of a large infratentorial SDH require prompt stabilization with volume replacement (fluid and/or blood products), pressors, and respiratory support, as needed. An urgent head CT and neurosurgical consultation should be obtained in any infant with signs of progressive brain stem dysfunction (i.e., coma, apnea, cranial nerve dysfunction); opisthotonus; or tense, bulging fontanelle. Open surgical evacuation of the clot is the usual management for the minority of infants with large SDH in any location accompanied by such severe neurologic abnormalities or obstructive hydrocephalus. When the clinical picture is stable and no deterioration in neurologic function or unmanageable increase in ICP exists, supportive care and
serial CT examinations instead of surgical intervention should be utilized in the management of posterior fossa SDH (5). Laboratory testing to rule out sepsis or a bleeding diathesis should be considered with large SDH. The infant should be monitored for the development of hydrocephalus, which can occur in a delayed fashion following SDH. Finally, chronic subdural effusions may occur rarely and can present weeks to months later with abnormally increased head growth. The outcome for infants with nonsurgical SDH is usually good, provided there is no other significant neurologic injury or disease. The prognosis for normal development is also good for cases in which prompt surgical evacuation of the hematoma is successful and there is no other parenchymal injury.
Epidural hemorrhage (EH). There are approximately 20 case reports of neonatal EH in the literature. EH appears to be correlated with trauma (e.g., difficult instrumented delivery), and a large cephalohematoma or skull fracture was found in about half the reported cases of EH. Removal or aspiration of the hemorrhage was performed in the majority of cases, and the prognosis was quite good except when other ICH or parenchymal pathology was present.
II. SUBARACHNOID HEMORRHAGE
Etiology and pathogenesis. Subarachnoid hemorrhage (SAH) is a common form of ICH among newborns, although the true incidence of small SAH remains unknown. Primary SAH (i.e., SAH not due to extension from ICH in an adjacent compartment) is probably frequent but clinically insignificant. In these cases, SAH may go unrecognized because of a lack of clinical signs. For example, hemorrhagic or xanthochromic CSF may be the only indication of such a hemorrhage in infants who undergo a CSF exam to rule out sepsis. Small SAH probably results from the normal “trauma” associated with the birth process. The source of bleeding is usually ruptured bridging veins of the subarachnoid space or ruptured small leptomeningeal vessels. This is quite different from SAH in adults, where the source of bleeding is usually arterial and, therefore, produces a much more emergent clinical syndrome. SAH should be distinguished from subarachnoid extension of blood from a germinal matrix hemorrhage (GMH)/IVH, which occurs most commonly in the preterm infant. SAH may also result from extension of SDH (e.g., particularly in the posterior fossa) or a cerebral contusion (parenchymal hemorrhage). Finally, subpial hemorrhage may occur, mostly in the otherwise healthy term newborn, and is usually a focal hemorrhage likely caused by local trauma resulting in venous compression or occlusion in the setting of a vaginal delivery (often instrumented) (6).
Clinical presentation. As with other forms of ICH, clinical suspicion of SAH may result because of blood loss or neurologic dysfunction. Only rarely is the blood volume loss large enough to provoke catastrophic results. More often, neurologic signs manifest as seizures, irritability, or other mild alteration of mental status, particularly with SAH or subpial hemorrhage occurring over the cerebral convexities.
Small SAH may not result in any overt clinical signs except seizures in an otherwise well-appearing baby. In these circumstance, the seizures may be misdiagnosed as abnormal movements or other clinical events.
Diagnosis. Seizures, irritability, lethargy, or focal neurologic signs should prompt investigation to determine whether there is a SAH (or other ICH). The diagnosis
is best established with a MRI (or CT) scan, or by LP to confirm or diagnose small SAH. CT scans are usually adequate to diagnose SAH, but as in the case of SDH/EH, an MRI is preferred because of the lack of radiation and is superior for the determination of whether there is evidence of any other parenchymal pathology. For example, SAH may occur in the setting of hypoxic-ischemic brain injury or meningoencephalitis, pathologies that are better detected by MRI than CT or US. CUS is not sensitive for the detection of small SAH, so should be used only if the patient is too unstable for transport to MRI/CT.
Management and prognosis. Management of SAH usually requires only symptomatic therapy, such as anticonvulsant therapy for seizures (see Chap. 56, Neonatal Seizures) and nasogastric feeds or intravenous fluids if the infant is too lethargic to feed orally. The majority of infants with small SAH do well with no recognized sequelae. In rare cases, a very large SAH will cause a catastrophic presentation with profound depression of mental status, seizures, and/or brain stem signs. In such cases, blood transfusions and cardiovascular support should be provided as needed, and neurosurgical intervention may be required. It is important to establish by MRI whether there is coexisting hypoxia-ischemia or other significant neuropathology that will be the crucial determinant of a poor neurologic prognosis, since a surgical procedure may not improve outcome if there is extensive brain injury in addition to the SAH. Occasionally, hydrocephalus will develop after a moderate-large SAH, and thus follow-up CUS scans should be performed in such infants, particularly if there are signs of increased ICP or abnormally rapid head growth.
III. INTRAPARENCHYMAL HEMORRHAGE
Etiology and pathogenesis
Primary cerebral hemorrhage is uncommon in all newborns, while cerebellar hemorrhage is found in 5% to 10% of autopsy specimens in the premature infant. An intracerebral hemorrhage may occur rarely as a primary event related to rupture of an arteriovenous malformation or aneurysm, from a coagulation disturbance (e.g., hemophilia, thrombocytopenia), or from an unknown cause. More commonly, cerebral intraparenchymal hemorrhage (IPH) occurs as a secondary event, such as hemorrhage into a region of hypoxic-ischemic brain injury. For example, IPH may occur as a result of venous infarction (since venous infarctions are typically hemorrhagic) either in relation to a large GMH/IVH (PT > T, see IV.) or as a result of sinus venous thrombosis (T > PT). Bleeding may occur secondarily into an arterial distribution infarction (term [T] > preterm [PT]) or, rarely, into an area of necrotic periventricular leukomalacia (PT > T). IPH is found not infrequently in infants undergoing extracorporeal membrane oxygenation (ECMO) therapy. Finally, cerebral IPH may occur as an extension of a large ICH in another compartment, such as large SAH or SDH, as rarely occurs with significant trauma or coagulation disturbance, and it may sometimes be difficult to identify the original source of hemorrhage.
Intracerebellar hemorrhage occurs more commonly in preterm than term newborns and may be missed by routine CUS, since the reported incidence is significantly higher in neuropathologic than clinical studies. The use of mastoid and posterior fontanelle views during CUS examination increases the
likelihood of detection of cerebellar hemorrhage (and posterior fossa SAH). Intracerebellar IPH may be a primary hemorrhage or may result from venous hemorrhagic infarction or from extension of GMH/IVH or SAH (PT > T). It is often difficult to distinguish the primary source or etiology of such hemorrhages by CUS. Cerebellar IPH rarely occurs as an extension of large SAH/SDH in the posterior fossa related to a trauma (T > PT).
Clinical presentation. The presentation of IPH is similar to that of SDH, where the clinical syndrome differs depending on the size and location of the IPH. In the preterm infant, IPH is often clinically silent in either intracranial fossa, unless the hemorrhage is quite large. In the term infant, intracerebral hemorrhage typically presents with focal neurologic signs such as seizures, asymmetry of tone/movements, or gaze preference, along with irritability or depressed level of consciousness. A large cerebellar hemorrhage (±SDH/SAH) presents as described above (see I.) and should be managed as for a large posterior fossa SDH.
Diagnosis. MRI is the best imaging modality for IPH, but CUS may be used in the preterm infant or when a rapid bedside imaging study is necessary. CT can be used for urgent evaluation when MRI is not available quickly, but the radiation exposure of CT should be avoided when possible. MRI is superior for demonstrating the extent and age of the hemorrhage and the presence of any other parenchymal abnormality. In addition, MR angiography/venography can be useful to demonstrate a vascular anomaly, lack of flow distal to an arterial embolus, or sinus venous thrombosis. Thus, MRI is more likely than CT or CUS to establish the etiology of the IPH and to determine accurately the long-term prognosis for the term infant. For the preterm infant, CUS views through the mastoid and posterior fontanelle improve the detection of hemorrhage in the posterior fossa. In cases where the etiology of IPH is unknown, an LP should be considered to rule out infection, unless there is significant mass effect or herniation.
Management and prognosis
Acute management of IPH is similar to that for SDH and SAH, where most small hemorrhages require only symptomatic treatment and support, while a large IPH with severe neurologic compromise should prompt neurosurgical intervention. It is important to diagnose and treat any coexisting pathology, such as infection or sinus venous thrombosis, as these underlying conditions may cause further injury that can have a greater impact on long-term outcome than the IPH itself. A large IPH, especially in association with IVH or SAH/SDH, may cause hydrocephalus, and thus head growth and neurologic status should be monitored for days to weeks following IPH. Follow-up imaging by MRI and/or CUS should be obtained in the case of large IPH, both to establish the severity and extent of injury and to rule out hydrocephalus or remaining vascular malformation.
The long-term prognosis largely relates to location and size of the IPH and GA of the infant. A small IPH may have relatively few or no long-term neurologic consequences. A large cerebral IPH may result in a life-long seizure disorder, hemiparesis or other type of cerebral palsy (CP), feeding difficulties, and cognitive impairments ranging from learning disabilities to significant intellectual disability, depending on the location. Cerebellar hemorrhage in the term newborn often has a relatively good prognosis, although it may result in cerebellar signs of ataxia, hypotonia, tremor, nystagmus, and mild cognitive deficits. There may be only minor deficits from small, unilateral cerebellar IPH in either preterm or term newborns. In contrast, a large cerebellar IPH that destroys a significant portion of the cerebellum (i.e., significant bilateral cerebellar injury) in a preterm newborn may result in severe cognitive and motor impairments for those infants who survive the newborn period (such infants often die of systemic illness rather than IPH) (7).
IV. GERMINAL MATRIX HEMORRHAGE/INTRAVENTRICULAR HEMORRHAGE
Etiology and pathogenesis
GMH/IVH is found principally in the preterm infant, where the incidence is currently 15% to 20% in infants born at <32 weeks’ GA, but is uncommon in the term newborn. The etiology and pathogenesis are different for term and preterm infants.
In the term newborn, primary IVH typically originates in the choroid plexus or in association with venous (±sinus) thrombosis and thalamic infarction, although IVH may also occur in the small remnant of the subependymal germinal matrix. The pathogenesis of IVH in the term infant is more likely to be related to trauma (i.e., from a difficult delivery) or perinatal asphyxia. However, at least 25% of infants have no identifiable risk factors. One study of CT imaging suggested that IVH might occur secondary to venous hemorrhagic infarction in the thalamus in 63% of term infants with clinically significant IVH (8). In such cases, there may be thrombosis of the internal cerebral veins, but occasionally there may be more extensive sinovenous thrombosis.
In the preterm infant, GMH/IVH originates from the fragile involuting vessels of the subependymal germinal matrix, located in the caudothalamic groove. There are numerous risk factors that have been identified in the etiology of IVH, including maternal factors such as infection/inflammation and hemorrhage, lack of antenatal steroids, external factors such as mode of delivery or neonatal transport to another hospital, and increasingly recognized genetic factors that predispose some newborns to IVH. However, these risk factors all contribute to the basic pathogenesis of GMH/IVH, which relates to alterations in blood flow and coagulation. Thus, the pathogenesis of GMH/IVH in preterm newborns has been shown to be largely related to intravascular, vascular, and extravascular factors (Table 54.2). The intravascular risk factors are probably the most important and are also the factors most amenable to preventive efforts.
The intravascular risk factors predisposing to GMH/IVH include ischemia/reperfusion, increases in cerebral blood flow (CBF), fluctuating CBF, and increases in cerebral venous pressure. Ischemia/reperfusion occurs commonly when hypotension is corrected quickly, whether due to disease or to iatrogenic intervention. This scenario often occurs shortly after birth, when a premature infant may have hypovolemia or hypotension that is treated with infusion of colloid, normal saline, or hyperosmolar solutions such as sodium bicarbonate. Rapid infusions of such solutions are thought to be particularly likely to contribute to GMH/IVH. Indeed, studies of the beagle puppy model showed that ischemia/reperfusion (hypotension precipitated by blood removal followed by volume infusion) reliably produces GMH/IVH (9). Sustained increases in CBF may contribute to GMH/IVH and can be caused by seizures, hypercarbia, anemia, and hypoglycemia, which result in
a compensatory increase in CBF. Fluctuating CBF has also been demonstrated to be associated with GMH/IVH in preterm infants. In one study, infants with large fluctuations in CBF velocity by Doppler US were much more likely to develop GMH/IVH than infants with a stable pattern of CBF velocity (10). The large fluctuations typically occurred in infants breathing out of synchrony with the ventilator, but such fluctuations have also been observed in infants with large patent ductus arteriosus or hypotension, for example. Increases in cerebral venous pressure are also thought to contribute to GMH/IVH. Sources of such increases include ventilatory strategies where intrathoracic pressure is high (e.g., high continuous positive airway pressure), pneumothorax, tracheal suctioning, and both labor and delivery, where fetal head compression likely results in significantly increased venous pressure (11). Indeed, a higher incidence of GMH/IVH is found in preterm infants with a longer duration of labor and in those delivered vaginally compared with those delivered via caesarean section. With all of these intravascular factors related to changes in cerebral arterial and venous blood flow, the role of a pressure-passive cerebral circulation is likely to be important. Several studies have shown that preterm infants, particularly asphyxiated newborns, have an impaired ability to regulate CBF in response to blood pressure changes (hence, “pressure-passive”) (12, 13). Such impaired autoregulation puts the infant at increased risk for rupture of the fragile germinal matrix vessels in the face of significant increases in cerebral arterial or venous pressure, and particularly when ischemia precedes such increased pressure. Finally, impaired coagulation and platelet dysfunction are also intravascular factors that can contribute to the pathogenesis of GMH/IVH.
Table 54.2 Factors in the pathogenesis of GMH/IVH
Intravascular factors
Ischemia/reperfusion (e.g., volume infusion after hypotension)
Fluctuating CBF (e.g., with mechanical ventilation)
Increase in CBF (e.g., with hypertension, anemia, hypercarbia)
Increase in cerebral venous pressure (e.g., with high intrathoracic pressure, usually from ventilator)
Platelet dysfunction and coagulation disturbances
Vascular factors
Tenuous, involuting capillaries with large luminal diameter
Extravascular factors
Deficient vascular support
Excessive fibrinolytic activity
Vascular factors that contribute to GMH/IVH include the fragile nature of the involuting vessels of the germinal matrix. There is no muscularis mucosa
and little adventitia in this area of relatively large diameter, thin-walled vessels; all of these factors make the vessels particularly susceptible to rupture.
Extravascular risk factors for GMH/IVH include deficient extravascular support and likely excessive fibrinolytic activity in preterm infants.
Pathogenesis of complications of GMH/IVH. The two major complications of GMH/IVH are periventricular hemorrhagic infarction (PVHI) and posthemorrhagic ventricular dilation (PVD). The risk of both complications increases with increased size of IVH. The pathogeneses of these two complications are discussed here.
PVHI has previously been considered an extension of a large IVH, hence referred to as a grade IV IVH. Although this designation is still used in much of the literature, careful neuropathologic studies have shown that the finding of a large, often unilateral or asymmetric hemorrhagic lesion dorsolateral to the lateral ventricle is not an extension of the original IVH, but is a separate lesion consisting of a venous hemorrhagic infarction. Neuropathologic studies demonstrate the fan-shaped appearance of a typical hemorrhagic venous infarction in the distribution of the medullary veins that drain into the terminal vein, resulting from obstruction of flow in the terminal vein by the large ipsilateral IVH. Evidence supporting the notion of venous obstruction underlying the pathogenesis of PVHI includes the observation that PVHI occurs on the side of the larger IVH, and Doppler US studies show markedly decreased or absent flow in the terminal vein on the side of the large IVH (14). Further neuropathologic evidence that PVHI is a separate lesion from the original IVH is that the ependymal lining of the lateral ventricle separating IVH and PVHI has been observed to remain intact in some cases, demonstrating that the IVH did not “extend” into the adjacent cerebral parenchyma. Hence, PVHI is a complication of large IVH, which is why some authors refer to it as a separate lesion rather than denoting PVHI to be a “higher” grade of IVH (i.e., a grade IV IVH). Risk factors for the development of PVHI include low birth GA, low Apgar scores, early life acidosis, patent ductus arteriosus, pneumothorax, pulmonary hemorrhage, and need for significant respiratory or blood pressure support (15).
Progressive posthemorrhagic ventricular dilation (PVD) or posthemorrhagic hydrocephalus (PHH—terminology varies), may occur days to weeks following the onset of GMH/IVH. Not all ventricular dilation progresses to established hydrocephalus that requires treatment, hence the terms are used with slightly different meanings (see IV.C.3. for clinical course of PVD). The pathogenesis of progressive posthemorrhagic ventricular dilation likely relates at least in part to impaired CSF resorption and/or obstruction of the aqueduct or the foramina of Luschka or Magendie by particulate clot (16). Recent work suggests that other mechanisms may play a role in the pathogenesis of PVD. High levels of TGF-β1 are found in the CSF following IVH, particularly in infants with PVD; TGF-β1 upregulates genes for extracellular matrix proteins that elaborate a “scar,” which may obstruct CSF flow and/or CSF reabsorption (16—18). In addition, restricted arterial pulsations (e.g., due to decreased intracranial compliance) have been proposed to underlie chronic hydrocephalus in hydrodynamic models of hydrocephalus (19). The pathogenesis of the brain injury resulting from PVD is probably related in large part to regional hypoxia-ischemia and mechanical distension of the
periventricular white matter, based on animal and human studies (20—23). In addition, the presence of non—protein-bound iron in the CSF of infants with PVD may lead to the generation of reactive oxygen species that in turn contribute to the injury of immature oligodendrocytes in the white matter (24). The brain injury associated with PVD is principally a bilateral cerebral white matter injury (WMI) similar to periventricular leukomalacia (PVL) with regard to both its neuropathology and long-term outcome (23, 25, 26, 27).
Clinical presentation
GMH/IVH in the preterm newborn is usually a clinically silent syndrome and thus is recognized only when a routine CUS is performed. The vast majority of these hemorrhages occur within 72 hours after birth, hence the use of routine CUS within 3—4 days after birth in many nurseries for infants with a GA <32 weeks. Infants with large IVH may present with decreased levels of consciousness and spontaneous movements, hypotonia, abnormal eye movements, or skew deviation. Rarely, an infant will present with a rapid and severe neurologic deterioration with obtundation or coma, severe hypotonia and lack of spontaneous movements, and generalized tonic posturing that is often thought to be seizure, but does not have an electrographic correlate by electroencephalogram.
The term newborn with IVH typically presents with signs such as seizures, apnea, irritability or lethargy, vomiting with dehydration, or a full fontanelle. Ventriculomegaly is often present at the time of IVH diagnosis in a term newborn. It is rare to find a catastrophic presentation unless there is another ICH, such as a large SDH or parenchymal hemorrhage.
Posthemorrhagic ventricular dilation (PVD) may develop over days to weeks following IVH, particularly in premature infants, and may present with increasing head growth (crossing percentiles on the growth chart), bulging fontanelle, splitting of sutures, decreased level of consciousness, impaired upgaze or sunsetting sign, apnea, worsening respiratory status, or feeding difficulties. However, PVD may be relatively asymptomatic in preterm newborns, as ICP is often normal in this population, particularly with slowly progressive dilation. Thus, serial CUS scans are critical for diagnosis of PVD in preterm infants with known IVH. A retrospective study of infants with birth weight < 1,500 g who developed IVH and survived at least 14 days showed that 50% of such infants will not show ventricular dilation, 25% will develop nonprogressive ventricular dilation (or stable ventriculomegaly), and the remaining 25% will develop PVD (28). The incidence of PVD increases with increasing severity of GMH/IVH; it is uncommon with grades I and II IVH (see Table 54.3) (up to 5% to 12%), but occurs in up to 75% of infants with grade III IVH ± PVHI. The incidence of PVD is also higher with younger GA at birth. Ventricular dilation may proceed rapidly (over a few days) or slowly (over weeks). About 40% of infants with PVD will have spontaneous resolution of PVD without any treatment. The remaining 60% generally require medical and/or surgical therapy (˜15% of this latter group do not survive).
Diagnosis
The diagnosis of GMH/IVH is almost invariably made by real-time portable CUS in the premature infant. We obtain routine CUS studies in all infants born at <32 weeks’ GA. In addition, CUS may be considered in
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
