This case illustrates the course of an extremely premature infant who is at greatest risk for severe periventricular-intraventricular hemorrhage (PV-IVH) even when managed in the current era of neonatology. Although the overall incidence of PV-IVH in the premature infant decreased throughout the 1980s–1990s, it has been relatively static over the last two decades.¹ High-grade intraventricular hemorrhage remains a particularly common morbidity in the extremely preterm infant, particularly in cases of rapid delivery when the potential for full dosing of glucocorticoids around the time of delivery is not possible.^2,3 This chapter focuses on a brief review of the pathogenesis of germinal matrix hemorrhage (GMH) and PV-IVH. Various approaches or strategies for the prevention, diagnosis, and treatment as well as outcomes are discussed, with gaps in knowledge highlighted.

The overall incidence of GMH and PV-IVH declined during the era of the 1970s–1990s with the evolution of modern obstetric and neonatal intensive care practices, although severe intraventricular hemorrhage remains a significant clinical problem in the extremely preterm population.^1,4,5 For those born at or before 28 weeks’ gestation, 15% continue to have the most severe forms of hemorrhage, occurring more frequently at younger gestational ages.^1,5 This observation continues to be highly relevant because the survival of the infants born at the limits of viability continues to increase, with mortality and long-term neuromotor deficits among survivors more likely with severe hemorrhage.^6–8 However, evidence also points to neurodevelopmental deficits even with lower grades of hemorrhage and even when the cranial sonogram is interpreted as being normal (see later discussion).^7,9–11 These observations are important because they point to the limitations of cranial sonography in identifying more subtle injury to the cortex, deep gray matter, or cerebellum. These additional findings may not be clearly evident in infancy or may be more readily identified by MRI studies performed closer to term.^12,13

The primary lesion in PV-IVH is bleeding from small vessels in the subependymal germinal matrix (GM), a transitional gelatinous region that provides limited support for the luxurious but very immature capillary bed that courses through it.¹⁴ With maturation, this matrix region becomes less prominent and is essentially absent by term gestation. The hemorrhage, when it evolves, may be confined to the GM region (grade I IVH), or it may extend and rupture into the adjacent ventricular system (grade II or III IVH, depending on the extent of hemorrhage), or extend into the white matter (termed a grade IV or intraparenchymal echogenicity [IPE], more appropriately termed periventricular hemorrhagic infarction [PVHI]) (Fig. 3.1A).^2,15,16 PVHI, which is invariably unilateral, represents an area of hemorrhagic necrosis of varying size within periventricular white matter, dorsal and lateral to the external angle of the lateral ventricle (Fig. 3.1B).^2,17,18

The genesis of bleeding from capillaries within the GM is complex and includes a combination of intravascular, vascular, and extravascular influences. Intravascular factors, especially those that involve perturbations in cerebral blood flow (CBF), have a critical role in capillary rupture and hemorrhage. It has been shown, through the use of different methods to assess CBF, including Doppler ultrasonography, near-infrared spectroscopy, and xenon-enhanced computed tomography (CT), that the cerebral circulation of the sick infant is pressure passive; that is, CBF varies directly with changes in systemic blood pressure.19–21 This state would be expected to increase the vulnerability of the GM capillaries to periods of both hypotension and hypertension, and this is supported in experimental studies and clinical observations. In a beagle puppy model GM hemorrhage can be produced by systemic hypertension with or without prior hypotension.^22,23 In addition, clinical temporal associations have been demonstrated between fluctuations in systemic blood pressure and simultaneous fluctuations in CBF velocity as may occur in the ventilated premature infant with RDS, increases in CBF as may occur with rapid volume expansion or a pneumothorax, and the subsequent development of PV-IVH.^24–27 Conversely, decreases in CBF secondary to systemic hypotension, which may occur in utero or postnatally, may also play a prominent role in the genesis of PV-IVH in certain infants.²⁸ Hypercarbia, producing potential modulation of autoregulation and of CBF, increases the risk for severe IVH.^29,30 A presumed mechanism in this context is that of rupture upon reperfusion.^3,20 Finally, elevations in venous pressure may be an important additional intravascular mechanism of hemorrhage and may reflect the peculiarity of the anatomy of the venous drainage of GM and the white matter.² Thus at the level of the head of the caudate nucleus and the foramen of Monro, the terminal, choroidal, and thalamostriate veins course anteriorly to a point of confluence to form the internal cerebral vein. The blood flow then makes a U-turn at the usual site of hemorrhage, raising the possibility that an elevation in venous pressure increases the potential for venous distention with obstruction of the terminal and medullary veins and hemorrhagic infarction. Indeed, simultaneous increases in venous pressure have been observed in infants who exhibit variability in arterial blood pressure, such as when it occurs with RDS and associated complications, such as pneumothorax and pulmonary interstitial emphysema, or with mechanical or high-frequency ventilation.³¹ In addition, certain anatomic patterns of subependymal veins are more predisposed to hemorrhage, as seen in susceptibility-weighted imaging venography.³² To summarize, it is likely that both arterial and venous perturbations contribute to the genesis of IVH. Later evidence suggests that these intravascular responses may be modulated by inflammation or the administration of medications to the mother, such as glucocorticoids (see later discussion).^33–35

In addition to the intravascular factors, vascular and extravascular influences—the poorly supported blood vessels, excessive fibrinolytic activity noted within the matrix region, and a prominent postnatal decrease in extravascular tissue pressure—may all contribute to hemorrhage.^2,36,37

The pathogenesis of white matter injury associated with hemorrhage remains unclear but appears to be closely linked to the adjacent bleed. Two potential pathways have been proposed to explain this intricate relationship. The first suggests a direct relationship to the PV-IVH, on the basis of several clinical observations, as follows: (1) the white matter lesion is always noted concurrent with or following a large GM and/or IVH and is rarely, if ever, observed before the hemorrhage; and (2) the white matter injury is always observed ipsilateral to the side of the larger hemorrhage when there is bilateral involvement of the ventricular system.^2,3,18 This consistent relationship between the GM and the white matter injury may in part be explained by the venous drainage of the deep white matter (see earlier discussion). A second explanation is a de novo evolution of white matter injury. Thus it is thought that the PV-IVH and white matter injury generally occur concurrently. Because both the GM and the periventricular white matter are border-zone regions, the risk for ischemic injury is increased during periods of systemic hypotension, particularly in the presence of a pressure-passive cerebral circulation.^2,3,20 Hemorrhage in these regions may then occur as a secondary phenomenon, or reperfusion injury. In support of this theory is the fairly consistent observation of the simultaneous detection of PV-IVH and white matter injury on cranial ultrasonography. Moreover, elevated hypoxanthine and uric acid levels (as markers of reperfusion injury) have been observed on the first postnatal day in infants in whom white matter injury subsequently developed.^38,39

Identification of the mechanisms contributing to periventricular white matter injury is crucial to the prevention of this lesion. If the white matter injury is directly related to PV-IVH, then prevention of the latter should reduce the occurrence of the white matter injury. However, if the PV-IVH and the white matter injury occur simultaneously as a result of a primary ischemic event, with the hemorrhage occurring as a secondary phenomenon, then prevention of the secondary hemorrhage may not affect the primary ischemic lesion. Indeed, the two follow-up studies on indomethacin treatment to prevent IVH in the neonatal period are supportive of this latter concern. Thus although the incidence of severe IVH was reduced in infants treated with indomethacin in both studies, neurodevelopmental outcomes at 18-month follow-up, including cerebral palsy, were comparable in the indomethacin-treated group controls.^40,41

In most cases (up to 70% of less severe IVH cases) the diagnosis is made with a screening sonogram. In the earlier descriptions of PV-IVH the majority of cases, about 90%, evolved within the first 72 hours of postnatal life.⁴² However, the time to initial diagnosis of hemorrhage has shifted to a later onset in recent years.^15,43 Thus for premature infants weighing less than 1000 g, the IVH diagnosis is made early, within the first 24 hours in approximately 80% of infants. However, some cases are now noted after the 10th postnatal day. This changing pattern may reflect the complexity of disease in the tiniest infants and the extent of supportive medical care. Infants with the more severe IVH frequently exhibit clinical signs such as a bulging fontanel, seizures, a drop in hematocrit, hyperglycemia, metabolic acidosis, and pulmonary hemorrhage.³