Germinal matrix hemorrhage and intraventricular hemorrhage (GMH-IVH) in the newborn: Pathogenesis, clinical presentation, and diagnosis

INTRODUCTION — Germinal matrix hemorrhage and intraventricular hemorrhage (GMH-IVH; also referred to as intraventricular hemorrhage [IVH]) is an important cause of brain injury in the newborn, and in particular for preterm infants. Although the incidence has declined since the 1980s, GMH-IVH remains a significant problem, as improved survival of extremely preterm infants has resulted in a greater number of survivors with this condition [1,2].

The epidemiology, pathogenesis, clinical presentation, and diagnosis of GMH-IVH are discussed in this topic review. The management, complications, and outcome of GMH-IVH in the newborn are discussed separately. (See "Germinal matrix hemorrhage and intraventricular hemorrhage (GMH-IVH) in the newborn: Prevention, management, and complications".)

PATHOLOGY

Preterm infants — In preterm infants, the site of origin of bleeding is generally in small blood vessels in the germinal matrix (GM, also termed the ganglionic eminence), located between the caudate nucleus and the thalamus at the level of the foramen of Monro. Neuropathologic studies suggest that the hemorrhage is primarily within the capillary network that freely communicates with the venous system, although bleeding can also occur within the arterial circulation. Vessels in the GM occupy border zones between cerebral arteries and the collecting zone of the deep cerebral veins, and have increased permeability when subjected to hypoxia and/or increased venous pressure [3]. Bleeding can disrupt the ependymal lining and extend into the lateral ventricle. (See 'Germinal matrix fragility' below.)

Severity and grading of GMH-IVH — Severity of hemorrhage in preterm infants is based on the extent of bleeding (confined only to the germinal matrix region or extending into the adjacent ventricular system), involvement of the white matter (intraparenchymal), and/or with the presence of ventricular distension (table 1) [3].

The following grading system is used:

●Grade I – Bleeding is confined to the GM plus up to 10 percent of the ventricular area (image 1)

●Grade II – GMH and IVH occupies between 10 to 50 percent of the lateral ventricle volume (image 2)

●Grade III – GMH and IVH occupies more than 50 percent of the lateral ventricle volume and is associated with acute ventricular distension related to the volume of blood (image 3)

●Periventricular hemorrhagic infarction (PVHI, previously referred to as grade IV IVH) – Hemorrhagic infarction in periventricular white matter ipsilateral to a mostly large IVH (also called periventricular hemorrhagic infarction [PVHI]) (image 4 and image 5)

Mild Grade I to II corresponds to mild and grades III and PVHI (grade IV) to severe GMH-IVH. Each grade of GMH-IVH may be unilateral or bilateral, with either symmetric or asymmetric grades of GMH-IVH. Additional ischemic lesions in the white matter and hemorrhages in the cerebellum, which are not always visible with ultrasound, should be mentioned separately as they are not covered in this grading system and can have consequences for neurodevelopmental outcomes. (See 'Diagnosis' below and "Neonatal cerebellar hemorrhage".)

Periventricular hemorrhagic infarction (PVHI) — PVHI (grade IV IVH) is due to infarction caused by impaired venous drainage of the medullary veins in the white matter after GMH-IVH [4-6]. The circulatory disturbance occurs in the subependymal region where the medullary veins drain into the terminal vein, as demonstrated by Doppler ultrasound [7].

PVHI most often involves the parietal and frontal cerebral areas, and in a quarter of patients, PVHIs are bilateral [8-10]. Depending on the site of the lesion, PVHI may result in the destruction of the motor and associative white matter axons and may evolve into a single or multiple cysts, which may become confluent with the lateral ventricle (image 4 and image 5) [5]. PVHI is manifested clinically by cerebral palsy with spastic hemiparesis in approximately one-half of the infants or an asymmetric spastic quadriparesis in most infants with bilateral involvement [11]. Higher severity scores based on the extent of echodensity, unilateral versus bilateral involvement, and presence or absence of midline shift were associated with significant morbidity (pulmonary hemorrhage, neonatal seizures, and motor disability) and death [8,10]. In addition, these patients often have intellectual deficits. (See "Cerebral palsy: Epidemiology, etiology, and prevention", section on 'Prematurity'.)

Term infants — The origin and location of IVH in term infants differs from that of preterm infants. The following are the relative frequency of the location and origin of IVH in term infants [12,13]:

●35 percent choroid plexus

●24 percent thalamus

●17 percent GM

●14 percent periventricular cerebral parenchyma

●10 percent no origin determined

PATHOGENESIS

Preterm infant — The pathogenesis of GMH-IVH in preterm infants encompasses [14]:

●Germinal matrix fragility from a lack of structural support due to immaturity.

●Disturbances of cerebral blood flow (CBF), particularly related to hypoxia-ischemia and reperfusion, elevated arterial blood flow, elevated venous pressure, and impaired cerebral autoregulation.

Germinal matrix fragility — In preterm infants, GMH-IVH generally originates within the germinal matrix (GM); the highly cellular and richly vascularized layer in the subependymal and the subventricular zone that gives rise to neurons and glia during fetal development. As the fetus matures, the GM begins to involute starting at 28 weeks as its cellularity and vascularity decrease, and by term, it is generally absent [15].

In preterm infants, a deficient structural support system makes the GM vulnerable to hemorrhage and injury especially when there is hemodynamic instability. Within the GM, the capillary network consists of numerous thin-walled, large blood vessels that lack structural support and are highly metabolically active [3]. The microvasculature of the GM is particularly delicate because of the abundance of angiogenic blood vessels with a paucity of pericytes, immature basal lamina, and deficiency of tight junctions and glial fibrillary acidic protein (GFAP) in the astrocyte end-feet. In more immature infants, there are also fewer glial fibers, a structural component for blood vessels that normally develops with increasing maturation [3].

The fragile capillary network drains into a well-developed deep venous system that forms the terminal vein, which changes direction in a U-turn fashion as it empties into the internal cerebral vein. It is postulated that the venous system is prone to congestion and stasis, resulting in increased cerebral venous pressure (CVP), which contributes to GMH-IVH [3]. Magnetic resonance susceptibility-weighted imaging (SWI) venography demonstrated features of the subependymal venous system that may predispose to GMH-IVH in the preterm infant, including a narrower curvature of the veins in the thalamus compared with term infants [16].

Cerebral blood flow (CBF) instability — Fluctuations of CBF in preterm infants are associated with GMH-IVH [3,17,18]. Preterm infants are particularly vulnerable to alterations in CBF because they have impaired autoregulation of CBF compared with term infants. This impairment results in a pressure-passive circulation, in which the infant cannot sustain constant CBF during changes in systemic blood pressure [19,20]. As a result, increases or decreases in systemic blood pressure are reflected by similar changes in CBF, leading to injury of the fragile blood vessels of the GM.

Other factors that have been implicated in CBF fluctuations that can lead to GMH-IVH include anemia, hypercarbia, acidemia, hypoglycemia, asphyxia, and abrupt elevations in systemic blood pressure (due to noxious stimuli, rapid volume expansion with fluid boluses, and seizures).

The association of impaired autoregulation with GMH-IVH was demonstrated in studies of preterm infants that monitored mean arterial pressure (MAP) and CBF using near-infrared spectroscopy (NIRS).

●In one study of 32 preterm infants (gestational age [GA] between 23 and 31 weeks) who were mechanically ventilated, 8 of 17 patients with impaired autoregulation (based on concordant changes in CBF and MAP) developed severe GMH-IVH, cystic periventricular leukomalacia (c-PVL), or both [17]. In contrast, only 2 of 15 infants with apparently intact autoregulation developed severe lesions.

●In a subanalysis of a prospective study of 650 preterm infants (GA <32 weeks), the 30 infants with GMH-IVH were more likely to receive inotropic drugs prior to the diagnosis of hemorrhage compared with infants without GMH-IVH [21]. Monitoring of MAP and CBF using NIRS suggested that infants with severe bleeds had cerebral hyperperfusion and more blood pressure-passive brain perfusion, reflecting a lack of autoregulation.

●In another study of 88 preterm infants (GA <32 weeks), the risk of GMH-IVH was associated with greater cerebral pressure passivity (based on MAP and NIRS measurements), but not with MAP variations alone [18].

Term infant: — The pathogenesis of GMH-IVH in term infants is dependent on the underlying cause, including [13]:

●Trauma during delivery. (See "Neonatal birth injuries", section on 'Intraventricular hemorrhage'.)

●Hypoxic-ischemic encephalopathy (HIE). In a study of 157 term infants with HIE, brain magnetic resonance imaging (MRI) performed in 138 infants demonstrated subdural hemorrhage in almost half of the cohort (47 percent) and intraparenchymal hemorrhage (IPH) in approximately one-quarter (22 percent) [22]. The risk of IPH was highest in infants who received inotropes.

●Coagulation or platelet abnormalities (alloimmune thrombocytopenia, hemophilia [23], genetic mutations of other hemostatic genes). (See 'Genetic factors' below and "Clinical manifestations and diagnosis of hemophilia", section on 'Initial presentation' and "Neonatal thrombocytopenia: Etiology".)

●Therapeutic hypothermia during extracorporeal membrane oxygenation (ECMO) [24].

●Sinovenous thrombosis (particularly in infants with thalamic involvement) [12].

●Rare causes include:

•Gene mutation of the collagen genes (COL4A1 and COL4A2) and tight junction protein (JAM3) [25,26]

•Rupture of a vascular malformation

EPIDEMIOLOGY — GMH-IVH generally occurs in preterm infants, and the incidence increases with decreasing gestational age (GA) and birth weight (BW). Although prematurity is the predominant risk factor, there are additional factors that affect the risk of GMH-IVH, such as antenatal glucocorticosteroid and neonatal transport.

Gestational age

Preterm infants — GMH-IVH occurs most frequently in very low birth weight (VLBW) infants (BW <1500 g) and/or very preterm (VPT) infants (GA <32 weeks). However, the incidence of GMH-IVH has fallen including in extremely preterm infants (GA <28 weeks) who are the most susceptible population.

This was illustrated by a study from the National Institute of Child Health and Human Development (NICHD) neonatal research network of 34,636 infants with GAs between 22 and 28 weeks and BWs between 401 and 1500 g that reported that the overall rate of severe GMH-IVH dropped from 19 percent to 15 percent between 1993 to 2012 [27]. A study from the California Perinatal Quality Care Collaborative also reported a decrease in the incidence of severe IVH for infants between 22 and <32 weeks gestation between 2005 and 2015 [28]. In this cohort, the administration of antenatal glucocorticoid therapy was associated with a decreased risk for IVH at all GAs.

In a study from the Australian and New Zealand Neonatal Network of 60,068 VPT infants born between 1995 and 2012 who underwent brain imaging screening, the overall risk of IVH was 20.2 percent and the risk of severe GMH-IVH (grade III or IV) was 5.2 percent [29]. The risk of severe IVH decreased over three epochs during the study period for the full cohort of VPT infants (6.6, 5.7, and 5 percent, respectively) and for those born ≤27 weeks gestation (14, 13, and 11.6 percent, respectively). Factors associated with development of severe IVH included lack of administration of antenatal steroids, five-minute Apgar score <7, need for intubation at birth, extremely low GA, being outborn, and vaginal delivery.

Also, in a study from the Canadian Neonatal Network of infants (GA 30 to 32 weeks) born between 2011 and 2016, the reported rate of severe GMH-IVH (grade III or greater) was 3.1 percent among screened infants (285/9221) [30]. In this study, risk factors associated with severe injury included singleton birth, five-minute Apgar score <7, and the use of mechanical ventilation and/or vasopressors on day 1 of life.

Decreasing gestation increases GMH-IVH incidence — In preterm infants, the risk of GMH-IVH and its severity increases with decreasing GA and BW as noted by the following studies (see 'Severity and grading of GMH-IVH' above):

●In a study from the NICHD Neonatal Research Network of 9575 preterm infants with GA between 22 and 28 weeks and BW between 401 and 1500 g born between 2003 and 2007, the overall incidence of GMH-IVH increased with decreasing GA [31]. The prevalence of severe GMH-IVH (defined as grades III and IV [now referred to as periventricular hemorrhagic infarction; PVHI]) increased with decreasing gestation with rates of 38, 36, 26, 21, 14, 11, and 7 percent of survivors for infants with gestational ages (GA) 22, 23, 24, 25, 26, 27, and 28 weeks, respectively.

●In a population-based Swiss study of 2896 preterm infants (GA <32 weeks) born between 2000 and 2004, the rate of GMH-IVH decreased 3.5 percent with each added week of gestation [32].

●In a population-based prospective study of all preterm infants with gestational age below 27 weeks born in Sweden from 2004 to 2007, the incidence of GMH-IVH increased from 5.2 percent of survivors born at 26 weeks gestation to 19 and 20 percent for those born at 22 and 23 weeks gestation, respectively [33].

Term infants — In term infants, the incidence of symptomatic intracranial hemorrhages (epidural, subdural, subarachnoid, intraventricular, and intraparenchymal) is 0.27 to 0.49 per 1000 live births [34,35]. However, severe IVH occurs infrequently in term infants [36].

Other factors that affect incidence — Although prematurity is the predominant risk factor, there is good evidence that the following factors impact on the incidence of GMH-IVH.

Antenatal glucocorticoid therapy — Antenatal steroid therapy for mother at high risk of preterm delivery has repeatedly been shown to decrease the risk of GMH-IVH [32,37-39], even in pregnancies complicated by chorioamnionitis [40].

Neonatal transport — Infants who require transport are more likely to develop GMH-IVH.

●In a study from the US National Inpatient Sample Database of preterm 67,596 VLBW infants (BW <1500 g), multiple regression analysis showed that the transported group was more likely to develop GMH-IVH than the inborn group (27.4 versus 13.4 percent, adjusted odds ratio [OR] 1.75, 95% CI 1.64 -1.86) [41].

●Data from the Canadian Neonatal Network showed that among 2951preterm infants <29 weeks gestation, severe brain injury defined as grade III or IV IVH (PVHI) or persistent periventricular echogenicity or echolucency was more common in infants who were outborn relative to those who were inborn (25.3 versus 14.7 percent) [42].

●Data from the United Kingdom showed postnatal transfer within 48 hours of life in preterm infants (GA <28 weeks) was associated with a greater risk of death and severe brain lesions for survivors [43].

●In a study of 15,842 VLBW infants, the 668 outborn infants were more likely to have severe IVH (8.2 versus 4.1 percent, OR 3.45, 95% CI 1.16-10.3) [44].

Additional risk factors — Additional risk factors associated with an increased risk of GMH-IVH include:

●Respiratory distress with episodes of hypocapnia, hypercapnia, hypoxia, and/or acidemia [45-49]. These factors are thought to be associated with fluctuations in cerebral blood flow (CBF), elevated central venous pressure (CVP) and severe hypoxemia [49,50].

●Rapid increases in arterial blood pressure, which may be caused by noxious stimuli (eg, suctioning) or rapid fluid boluses, are associated with increased CBF [3] and evidence of cardiorespiratory instability based on extreme mean arterial blood pressure [51].

●Mechanical ventilation, likely by contributing to fluctuations in CBF and increased CVP [30,52].

●Cardiopulmonary resuscitation (CPR) in the delivery room increases the overall risk in extremely preterm infants (GA <28 weeks) of GMH-IVH and of severe IVH [53]. In the delivery room, severe IVH has been associated with increased numbers of intubation attempts, inadvertent delivery of high tidal volume during positive pressure ventilation, and large fluctuations of end-tidal carbon dioxide [54-56].

Off-peak delivery (delivery between midnight and seven o'clock am) is an independent reported risk factor (after adjusting for maternal, infant, and hospital characteristics) for severe IVH and the composite outcomes of death or severe IVH [57]. This may reflect a poorer quality of resuscitation efforts during this time period resulting, in an increased risk of GMH-IVH.

●Hypotension – Several observational studies have reported that hypotension, especially those requiring inotropic support, increases the risk of GMH-IVH in extremely preterm infants (GA <28 weeks) [30,58-60]. Another study showed that a strong response to dopamine (defined as an increase of mean arterial pressure [MAP] >10 mmHg) was associated with a reduced risk of GMH-IVH, while a failure to respond to dopamine increased the risk of IVH [61]. However, it is unclear whether hypotension and the use of inotropic support is a marker for severely ill patients [62]. (See "Assessment and management of low blood pressure in extremely preterm infants", section on 'Outcomes in treated versus untreated infants' and "Assessment and management of low blood pressure in extremely preterm infants".)

Possible risk factors — The evidence is less clear whether the following risk factors have a meaningful impact on the risk of GMH-IVH:

●Genetic predisposition – Hemostatic gene mutations

●Prenatal factors – Maternal preeclampsia, maternal use of indomethacin and aspirin

●Labor and delivery issues – Mode of delivery, breech presentation, delayed cord clamping

●Neonatal and postnatal factors – Bicarbonate therapy, metabolic acidosis [63], neonatal transport, hypothermia, pneumothorax, coagulation and platelet abnormalities, and low hematocrit

Genetic factors — Limited data suggest that genetic factors may contribute to GMH-IVH in both preterm and term infants [64].

Proposed genetic factors include:

●Hemostatic genes – There are conflicting reports on whether mutations of hemostatic genes predispose preterm infants to GMH-IVH, but these are likely to be of lesser importance than other risk factors. Case reports have implicated factor V Leiden mutations as a risk factor for GMH-IVH and PVHI [65-67].

However, a prospective study of 1008 very low birth weight (VLBW) infants (in whom DNA testing for factor V Leiden, prothrombin G20210A, factors VII and XIII were performed), reported no difference in the rate of GMH-IVH in the 178 infants with a mutation of one of the hemostatic genes compared with 830 infants without a mutation [68]. As an example, the rates of GMH-IVH in the 74 infants with mutations of factor V Leiden compared with those without were 19 versus 17.5 percent.

●Gene mutations – In several case reports, mutation in the collagen genes (COL4A1 and COL4A2) or tight junction protein (JAM3) has been associated with severe prenatal intracranial hemorrhage, but there are no reported studies with large cohorts [25,69,70]. In a candidate gene analysis, 13 single nucleotide polymorphisms were associated with severe IVH including five related to central nervous system (CNS) neuronal and neurovascular development [71].

●Inflammatory genes – Although data are conflicting, polymorphisms in the proinflammatory cytokine interleukin (IL)-6 have been proposed as genetic modifiers for the risk of IVH [71-73].

●Ethnicity – Infants born to mothers of African ancestry appeared to be at increased risk for grade II to IV GMH-IVH compared with those born to mothers who were White based on a prospective study that controlled for confounding variables. However, the risk was reduced by a single prenatal visit, suggesting that ethnicity may still be a marker for healthcare disparities, which increases the risk of GMH-IVH, rather than a truly independent factor.

Prenatal factors — It remains uncertain whether or not the following prenatal factors impact on the risk of GMH-IVH:

●Chorioamnionitis – Data are conflicting regarding whether chorioamnionitis increases the risk of GMH-IVH.

A prospective study from three tertiary centers reported that histologic chorioamnionitis was not associated with GMH-IVH or white matter injury near birth based on magnetic resonance imaging (MRI) [74]. In addition, chorioamnionitis was not associated with poorer motor and cognitive outcome based on Bayley-III assessments conducted between 18 and 24 months corrected age. A retrospective study also found no association between placental pathology (most commonly associated with chorioamnionitis) and IVH [75].

However, several prior observational studies have reported an increased risk of GMH-IVH associated with maternal intrauterine infection and/or amniotic sac inflammation [76-80]. Indirect evidence linking maternal intrauterine infection to GMH-IVH was provided in a meta-analysis that showed prenatal antibiotic use for prolonged rupture of membranes reduced the incidence of all grades of GMH-IVH [81]. In addition, a low neonatal neutrophil count (less than 1000 neutrophils/microL) within 2.5 hours of birth was associated with GMH-IVH and its severity [82].

A contributory role for maternal inflammation is supported by several studies that have shown an association between GMH-IVH and increased cytokine production and release (used as a biomarker for inflammation), and/or histologic evidence of inflammation [72,78-80,83].

A large multicenter Canadian prospective study of 3094 preterm infants (GA less than 33 weeks) born between 2005 and 2006 reported that the severity of GMH-IVH increases with chorioamnionitis [76]. Infants exposed to chorioamnionitis (15 percent of the cohort) were more likely to have severe IVH (defined as grades III and IV) compared with those without chorioamnionitis (22 versus 11 percent). A multivariate logistic regression analysis, which included adjustments for GA, BW, treatment with antenatal corticosteroids, and presence of maternal hypertension, showed that infants with chorioamnionitis had a 1.6-fold increased risk of severe IVH compared with those without chorioamnionitis.

●Maternal hypertensive disorders of pregnancy – Data are conflicting regarding the risk of GMH-IVH in preterm infants exposed to preeclampsia or born to mothers with HELLP syndrome (hemolysis with a microangiopathic blood smear, elevated liver enzymes, and a low platelet count) [84-87]. (See "Preeclampsia: Clinical features and diagnosis" and "HELLP syndrome (hemolysis, elevated liver enzymes, and low platelets)".)

Several retrospective studies have reported that GMH-IVH is increased in newborns of mothers with severe eclampsia and HELLP syndrome compared with mothers with normal blood pressure [85-87].

In contrast, multicenter studies reported a significant decreased risk of GMH-IVH if preeclampsia was present prior to birth [84] and if there was maternal hypertension during pregnancy [88].

●Maternal drug therapy – The following drugs used during pregnancy have been studied for their effects on GMH-IVH risk.

•Antenatal indomethacin therapy was shown in a meta-analysis of observational studies to increase the risk of severe IVH but not the overall risk of GMH-IVH [89].

•Antenatal low-dose aspirin does not appear to affect the rate of GMH-IVH based on a systematic review of trials of antiplatelet agents (primarily low-dose aspirin) to prevent preeclampsia. In this meta-analysis, the rate of GMH-IVH was not different between the treatment and control group (relative risk [RR] 0.88, 95% CI 0.63-1.22) [90].

Delivery

●Mode of delivery – It remains uncertain whether the route of birth affects the risk of GMH-IVH. During labor and vaginal delivery, compression of the fetal head by the uterus increases central venous pressure (CVP), which theoretically could promote GMH-IVH [3]. Although there is insufficient information from clinical trials based on systematic review of the literature [91,92], observational data predominantly report no differences in the risk of GMH-IVH based on the mode of delivery.

•Large observational studies primarily report similar rates of GMH-IVH between infants born by vaginal and those by caesarean delivery [93-96]. In the largest cohort of 20,231 preterm infants (GA between 24 and 34 weeks) born between 1995 and 2003, there was no difference in the risk of GMH-IVH between vaginal and cesarean deliveries after adjusting for confounding factors (adjusted odds ratio [OR] 0.99, 95 % CI 0.84-1.17).

•However, two smaller studies of preterm infants (GA <30 weeks) report that GMH-IVH is more likely to occur in infants born by vaginal versus caesarean delivery [97,98].

Limited evidence also suggests that there is no association between active labor and severe GMH-IVH [99-101].

●Delayed cord clamping – Although several studies reported that delayed cord clamping compared with matched controls was associated with a reduction in GMH-IVH [102-104], a large clinical trial did not show an association between delayed cord clamping and reduced risk of GMH-IVH [102,103,105]. In this trial of 1566 preterm infants (GA <30 weeks) there was no difference between cord clamping at <10 seconds and >60 seconds in the combined primary outcome of death or major morbidity at 36 weeks of gestation (relative risk [RR], 1.00; 95% CI 0.88-1.13) or the risk of GMH-IVH [105]. Nevertheless, delayed cord clamping in stable preterm infants is recommended for its potential associated benefits. (See "Delivery of the low birth weight singleton fetus", section on 'Optimal cord clamping' and "Labor and delivery: Management of the normal third stage after vaginal birth", section on 'Early versus delayed cord clamping'.)

●Cord milking – Cord milking or stripping the umbilical cord is associated with an increased risk of severe GMH-IVH in preterm infants (GA 23 to 27 weeks) [106,107]. (See "Labor and delivery: Management of the normal third stage after vaginal birth", section on 'Cord milking'.)

Neonatal and postnatal factors — Data are inconclusive regarding the impact of bicarbonate therapy, hypothermia, pneumothorax, and coagulation and platelet defects on the risk of GMH-IVH:

●Bicarbonate therapy – Bicarbonate may theoretically increase the risk of GMH-IVH, due to hyperosmolarity that may alter CBF [108].

●Hypothermia – Hypothermia following birth has been linked to increased risk of GMH-IVH in some studies [109], but not others [110]. Different definitions of hypothermia, size of study groups, and study design may explain conflicting results.

●Pneumothorax – Several reports have shown an association between pneumothorax and GMH-IVH (especially severe IVH), postulated to be related to increased CVP [111,112]. In contrast, in a study of 675 preterm infants (≤28 weeks), there was no increase in GMH-IVH occurrence in the 62 neonates with pneumothorax compared with those without this complication [113].

●Coagulation and platelet abnormalities – Thrombocytopenia and coagulation defects are common in preterm infants, especially those with other risk factors for GMH-IVH. Although several observational studies have shown an association between these two conditions and GMH-IVH [114-116], other data cast uncertainty regarding a causal role in the pathogenesis of GMH-IVH in preterm infants [117]. In particular, failure of procoagulant therapy to reduce GMH-IVH raises questions about a causal relationship [6,118]. (See "Neonatal thrombocytopenia: Clinical manifestations, evaluation, and management" and "Disseminated intravascular coagulation in infants and children".)

●Low hematocrit – Initial low hematocrit (below 45 percent) has been reported to increase the risk periventricular and IVH in extremely low birth weight infants (birth weight <1000 g) [119]. The proposed mechanism is the low hematocrit is associated with a lower intravascular volume that promotes cerebral hypoperfusion.

CLINICAL PRESENTATION

Prenatal hemorrhage — Prenatal GMH-IVH appears to be rare [120,121]. However, in utero severe periventricular hemorrhagic infarction in term infants may result in cerebral palsy [122,123]. A systematic review of the literature identified 240 prenatal cases of GMH-IVH: 27 cases with mild (grade I/II) IVH, 100 with grade III, 105 with PVHI, 8 eight cases in which the grade was not classified [121]. Developmental outcome was assessed in 129 of the 162 survivors at a median age of 17 months. Developmental delay was not observed in any child with mild IVH, but in more than half of the children with PVHI (28 of 77 cases).

Postnatal GMH-IVH: Preterm infant

Manifestations — GMH-IVH has three different postnatal presentations in the preterm infant [3]:

●Silent presentation – A clinically silent GMH-IVH without symptoms occurs in 25 to 50 percent of cases and is detected by routine ultrasound screening. (See 'Routine ultrasound screening' below.)

●Saltatory or stuttering course is the most common presentation and evolves over hours to several days. It is characterized by nonspecific findings including an altered level of consciousness, hypotonia, decreased spontaneous and elicited movements, and subtle changes in eye position and movement. Respiratory function is sometimes disturbed.

●Catastrophic deterioration is the least common presentation and evolves over minutes to hours. Signs include:

•Stupor or coma

•Irregular respirations, hypoventilation, or apnea

•Decerebrate posturing

•Generalized seizures, especially tonic seizures

•Flaccid weakness

•Cranial nerve abnormalities, including pupils fixed to light

•Other features include a bulging anterior fontanelle, hypotension, bradycardia, a falling hematocrit, metabolic acidosis, and inappropriate antidiuretic hormone secretion

Timing — Virtually all GMH-IVHs in preterm infants take place within the first few days after birth and the majority within the first day.

●A meta-analysis of very low birth weight (VLBW) infants (BW ≤1500 g) estimated that half of the cases of GMH-IVH occurred by six hours of life and only one-third of cases occurred after 24 hours [124]. Late-onset of GMH-IVH is typically associated with low cerebral blood flow (CBF), which may be due to low superior vena cava flow seen in infants with large ductal shunts or hypotension [125,126].

●In one series in which serial ultrasound examinations were performed beginning shortly after birth in infants with BW <1750 g, GMH-IVH was detected before one hour of age in 20 percent of patients [127].

Further progression over three to five days occurs in approximately 20 to 40 percent of cases [3].

Coexisting lesions — In neuropathologic studies, GMH-IVH rarely is an isolated lesion. The majority of infants who die more than one week after GMH-IVH also have associated white matter injury (periventricular leukomalacia [PVL]) or necrosis in the pons and the subiculum of the hippocampus. GMH-IVH is an independent risk factor for white matter injury in very preterm neonates born <28 weeks gestation [128]. Hemorrhages in the cerebellum also often coexist. Ultrasound imaging through the mastoid window (posterolateral fontanelle) is mostly required to diagnose cerebellar lesions with a size of >4 mm [129,130]. (See "Neonatal cerebellar hemorrhage".)

DIAGNOSIS

Cranial ultrasound — Cranial ultrasonography is most commonly used to diagnose GMH-IVH. It is the preferred imaging modality because of its high sensitivity for detecting acute GMH-IVH, its ease of portability (bedside imaging), and lack of ionizing radiation [3,131]. Coronal and parasagittal ultrasound views are obtained routinely to identify blood in the germinal matrix (GM), ventricles, or cerebral parenchyma, and any other echogenic abnormalities.

Ultrasonography is able to grade the severity of GMH-IVH based upon the location and extent of the GMH-IVH and presence of ventricular dilatation (table 1) [3]. Grade I (image 1), grade II (image 2), and grades III (image 3) and IV/periventricular hemorrhagic infarction (PVHI) (image 4) correspond to mild, moderate, and severe IVH, respectively.

However, cranial ultrasound is less sensitive than magnetic resonance imaging (MRI) in identifying low grade GMH-IVH and associated subtle lesions in the white matter and cerebellum [132]. (See 'Other radiographic studies' below.)

Routine ultrasound screening — Because up to one-half of GMH-IVH cases are clinically silent, routine ultrasound screening should be performed in preterm infants. (See 'Manifestations' above.)

In our neonatal intensive care unit (NICU), ultrasound is performed for all preterm infants with a gestational age (GA) <32 weeks. The timing and frequency of repeated scans are dependent on the GA at birth and clinical course of the preterm infants:

●For infants born <28 weeks GA, scans are performed on day of admission, two to three times during the first week, weekly until 34 weeks PMA, and one scan at discharge or at term-equivalent age (TEA). Once infants reach 34 weeks PMA, scans are performed every two weeks.

●For infants born >28 weeks, scans are performed on day of admission, day 4 to 7, weekly until 34 weeks PMA, and one scan at discharge or at TEA.

●Scans should be repeated following any acute clinical deterioration (eg, apneas, sepsis, NEC, surgery).

This strategy allows early detection, and if needed, early intervention, of complications including posthemorrhagic ventricular dilation (PHVD) (image 6) and optimal diagnosis of white matter injury [11]. Ultrasound is also performed in any patient with abnormal clinical findings consistent with GMH-IVH. (See "Germinal matrix hemorrhage and intraventricular hemorrhage (GMH-IVH) in the newborn: Prevention, management, and complications", section on 'Complications' and 'Manifestations' above.)

Although less frequent, GMH-IVH also occurs in preterm infants with a GA ≥32 weeks [133]. However, there is currently no specific guideline for cranial ultrasound screening in these moderate to late preterm infants. As a result, in our NICU, a cranial ultrasound is performed when there is a clinical suspicion for GMH-IVH based on subtle changes in neurologic or respiratory status, or for patients who have risk factors or conditions associated with GMH-IVH.

Society guideline — Although guidelines from pediatric societies have been published, they do not allow for early detection of GMH-IVH and its potential complication of PHVD because of the potential delay in the initial ultrasound screening and limited number of subsequent imaging tests. Optimal management of PHVD is dependent on early detection prior to the onset of clinical symptoms of increased intracranial pressure (ICP). As a result, we disagree with the suggested schedule and frequency of imaging and we continue to perform initial testing on the first day of admission with repeated scans during the first week after delivery and weekly or every other week until discharge. (See "Germinal matrix hemorrhage and intraventricular hemorrhage (GMH-IVH) in the newborn: Prevention, management, and complications", section on 'Posthemorrhagic ventricular dilatation (PHVD)'.)

●American Academy of Pediatrics - In a clinical report from the American Academy of Pediatrics (AAP), the following recommendations regarding cranial ultrasonography were published [134]:

•Routine ultrasound screening should be performed on all infants with a GA ≤30 weeks by 7 to 10 days of age. Screening before seven days of age may be indicated for infants with signs and symptoms of brain injury. Repeat screening is recommended at four to six weeks of age, at term equivalent, or before hospital discharge.

•Ultrasound screening is recommended for selected infants >30 weeks who have risk factors for brain injury (eg, placental abruption, received vigorous resuscitation, hypotension requiring vasopressor support, severe acidosis, prolonged mechanical ventilation, or confirmed sepsis or pneumothorax).

•For infants with abnormal ultrasound findings, repeat serial ultrasonography should be performed as clinically indicated.

•Standard cranial ultrasonographic screening should include views from the anterior and mastoid fontanelles. Additional posterior fontanelle and vascular imaging can be performed for additional information.

●Canadian Paediatric Society - In a revised position statement (revision available only online), the Canadian Paediatric Society recommends ultrasound imaging [135]:

•For all preterm infants <32 weeks gestation performed in the first 4 to 7 days after delivery. Suggested repeat imaging is performed 4 to 6 weeks after delivery.

•For infants born between 32 and 37 weeks gestation, cranial ultrasound is recommended only in the presence of risk factors for intracranial hemorrhage and is performed 4 to 7 days after delivery. A repeat ultrasound is performed only if the initial test is abnormal, at 4 to 6 weeks after delivery.

•However, when the first cranial ultrasound examination shows an abnormality (Grade 2 or higher IVH or white matter injury), a repeat ultrasound is recommended 7 to 10 days later. Cranial ultrasound is also performed weekly if ventricular dilation or worsening IVH or white matter injury is detected.

Other radiographic studies

●Magnetic resonance imaging (MRI) scans will identify more small GMH-IVHs in the temporal and occipital GM than ultrasound. MRI will identify additional white matter lesions, cerebellar hemorrhages, subdural or posterior fossa hemorrhages, and peripheral areas of infarction [130,131]. However, approximately one-half of the patients are not stable enough to be taken to the MR unit in the early stage. In addition, nonmetallic monitoring and support equipment appropriate for newborn infants is required. As a result, MRI is not the initial preferred diagnostic imaging modality.

●Computed tomography (CT) is not recommended in view of the need to transport infants to the scanner and particularly exposure to ionizing radiation. CT scanning is generally avoided now in newborns, except for emergencies (eg, neurosurgical emergency) or when ultrasound or MRI is not available.

Lumbar puncture — It is recommended to perform cranial ultrasonography before performing a lumbar puncture (LP). In case of a large third and small fourth ventricle, an aqueductal stenosis is likely and an LP should not be performed. If cranial ultrasonography is not available, LP has in the past been used to assist in the diagnosis of GMH-IVH. However, LP must be performed cautiously because of the small risk of herniation during pressure and fluid shifts in infants with large unilateral or posterior fossa hemorrhages. In IVH, the cerebrospinal fluid (CSF) typically contains numerous red blood cells and a high protein concentration. The CSF becomes xanthochromic several hours after the hemorrhage, and the glucose concentration may be reduced. (See "Lumbar puncture: Indications, contraindications, technique, and complications in children".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a parent might have about a given condition. These articles are best for parents who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for parents who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to the parents of your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

Basics topics (see "Patient education: Intraventricular hemorrhage in newborns (The Basics)")

SUMMARY AND RECOMMENDATIONS — Germinal matrix-intraventricular hemorrhage (GMH-IVH) is an important cause of brain injury in preterm infants.

●GMH-IVH occurs most frequently in infants born <32 weeks gestation or with a birth weight (BW) <1500 g. The risk of GMH-IVH increases with decreasing gestational age (GA). Additional risk factors include chorioamnionitis, lack of antenatal glucocorticoid therapy, neonatal transport, prolonged neonatal resuscitation, and respiratory distress requiring mechanical ventilation. (See 'Epidemiology' above and 'Other factors that affect incidence' above.)

●In preterm infants, GMH-IVH generally originates from the germinal matrix (GM). The risk of bleeding results from the structural fragility of the GM and cerebral blood flow (CBF) instability of the preterm infant. Severity of hemorrhage in preterm infants is based on the extent of bleeding (confined only to the GM region or extending into the adjacent ventricular system), involvement of the white matter (intraparenchymal), and/or with the presence of ventricular distension (table 1). (See 'Preterm infant' above and 'Severity and grading of GMH-IVH' above.)

●GMH-IVH is uncommon in the term infant and is primarily associated with birth trauma and perinatal hypoxia. Genetic disorders (coagulation and platelet disorders and structural mutations of collagen and tight junction proteins) are increasingly identified as contributors to GMH-IVH.

●The presentation of GMH-IVH can be clinically silent, saltatory, or catastrophic. A clinically silent syndrome occurs in 25 to 50 percent of cases, but is detected by routine ultrasound screening. (See 'Clinical presentation' above.)

●The diagnosis of GMH-IVH is made by cranial ultrasonography. The grading of the severity of GMH-IVH is based upon the location and extent of the GMH-IVH and presence of ventricular dilatation (table 1). (See 'Diagnosis' above.)

●Because approximately one-half of GMH-IVH is clinically silent, we recommend routine ultrasound screening in preterm infants <32 weeks gestation. In addition, cranial ultrasound should be performed for any older preterm or term infant who exhibit changes in neurologic or respiratory status, or who has any conditions or risk factors associated with GMH-IVH. (See 'Diagnosis' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Lisa M Adcock, MD, who contributed to an earlier version of this topic review.