Abstract

OBJECTIVE. Subpial hemorrhages, typically seen in neonates, are rare but can harm the adjacent brain parenchyma. The purpose of this review is to summarize the anatomy and pathophysiology of subpial hemorrhage and highlight its characteristic neuro-imaging pattern.
CONCLUSION. The distinctive neuroimaging pattern of subpial hemorrhage is best appreciated on brain MRI, which shows the morphology over the cortex and injury to adjacent cortex and subcortical white matter. These findings do not occur in subarachnoid and subdural hemorrhages. Recognizing the pattern of subpial hemorrhages should guide prognostic precision, prognostication, and counseling.
Subpial hemorrhage is an intracranial hemorrhage subtype that, until recently, has been neither widely appreciated nor well understood. It is most common in neonates and infants [111] and can be clinically important because of the associated injury to adjacent brain parenchyma. Advances in cross-sectional multisequence MRI have facilitated identification of subpial hemorrhages, which are confined within a tight anatomic space [1, 3, 710, 1215]. Postmortem and neuroimaging studies have contributed to our understanding of subpial hemorrhage and its association with nearby cortical damage [1, 3, 12, 16]. However, the long-term clinical consequences of subpial hemorrhage are unclear. Although long-term sequelae may depend on the extent of cortical injury and overall volume of subpial hemorrhage, future studies are needed to assess long-term clinical outcomes and clinical and radiographic determinants of prognosis.
The earliest evidence of the existence of subpial hemorrhage came from a landmark article by Friede [1] in 1972 on the autopsy findings in nine infants. Autopsy studies dating to 1928 and 1934 also described subpial hemorrhage [2, 11]. Because subpial hemorrhage is difficult to detect with imaging modalities other than MRI, most cases have been historically grouped with subarachnoid hemorrhage under the broader term leptomeningeal hemorrhage [4, 9] or referred to as variants or subtypes of subarachnoid hemorrhage [10]. The earliest report of subpial hemorrhage recognized as a separate entity on cross-sectional neuro-imaging came from a case series published by Huang and Robertson [4] in 2004.
The purpose of this review is to describe the anatomic, pathophysiologic, clinical, and neuroimaging features of subpial hemorrhage that differentiate it from other extraaxial hemorrhages. To our knowledge, this is the first comprehensive neuroimaging review on the topic of subpial hemorrhage. The information provided should be of clinical relevance to radiologists, neonatologists, and neurologists caring for patients with this condition.
The meninges are critical structures for normal CNS organization and function [1719]. During the past several decades, electron microscopy and experimental data have improved our understanding of histologic, ultrastructural, and functional aspects of these CNS sheaths and their surrounding microenvironment [2030]. These findings have clarified the role of the meninges and better defined the relations between the meninges and adjacent blood vessels and extracellular matrix. This complex meningeal anatomy can affect the distribution of pathologic processes, such as hemorrhage and infection. Therefore, it is important to differentiate adjacent compartments, although the tight anatomic nature can pose a major challenge.

Anatomy of the Subpial Space

The subpial space (SpS) (Fig. 1) can be defined as the potential space bordered externally by the pia mater and internally by the glia limitans, which is also known as the external glial limiting membrane (i.e., the outermost layer of astrocyte foot processes in the neocortex [layer 1]) [5, 6, 21, 24, 29, 30]. This narrow anatomic zone is occupied primarily by collagen fiber bundles that contribute to the basement membranes related to both the pia mater and the glia limitans, making it a potential space for pathologic processes. It is distinct in nature from the fluid-filled compartment that composes the subarachnoid space, through which CSF flows. The SpS also contains segments of vessels (arterioles and small veins) and, occasionally, inflammatory cells [21, 24, 29, 30] (Fig. 1).
Fig. 1 —Schematic shows different types of extraaxial hemorrhages, providing providing comparison between epidural, subaxial, subarachnoid, and subpial hemorrhage. Subpial veins, unlike subpial arteries, lack pial covering. When hemorrhage collects in subpial space, it lifts pia mater. Splaying of cortex and cortical infarction are common with subpial hemorrhage, unlike subarachnoid hemorrhage (SAH). In SAH, blood flows in CSF-filled subarachnoid space overlying pia mater (green).
Ultrastructural studies by Alcolado, Hutchings, Zhang, and Weller and their associates [21, 24, 29, 30] showed that the pia mater forms a single thin layer of cells around the cortex, reflecting itself on the outer surface of both the subarachnoid vessels and trabeculae. This configuration creates a complete mechanical separation between the CSF-containing subarachnoid space and the predominantly collagen-containing SpS (Fig. 1).
Within the SpS (Fig. 1), the arteries that transit down to the cortex are coated by a largely continuous single layer of pial leptomeningeal cells, which separates them from the subpial collagen and encases them as far as the capillary bed level, where the endothelial basement membrane fuses with the basement membrane of the glia limitans. No perivascular (Virchow-Robin) space is present around the intracortical arteries [29, 30].
In contrast, the subpial segments of cortical veins are not encased in such a leptomeningeal sheath and therefore are in direct contact with SpS collagen (i.e., there is contiguity between the venous adventitia and the SpS) [29, 30]. These anatomic differences may contribute to the intrinsic risk of rupture that each of these vascular structures has during certain pathologic states.
Under normal circumstances, the pia mater and the glia limitans are closely apposed [24] (Fig. 1) without much intervening substance. The integrity of the pial-glial interface has a major role not only as a barrier [31] but also as an essential part of normal cortical development during the prenatal, neonatal, and postnatal periods. This interface is formed early during neocortical development [5, 6, 23] and is key for the proper arrangement of the cortical neurons [5]. Focal lesions of these related structures, such as subpial hemorrhage, can have serious developmental consequences, including acquired cortical dysplasias and leptomeningeal heterotopias and focally disorganized pial vasculature [5, 6, 23]. These cytoarchitectural changes caused by subpial abnormality can lead to cortical dysfunction and clinical consequences.

Incidence of and Risk Factors Associated With Subpial Hemorrhage

Few epidemiologic studies have described the incidence of neonatal subpial hemorrhage. The lack of precise data is caused in part by the taxonomy some authors use, in which subpial hemorrhage is considered together with subarachnoid hemorrhage, often under the term “leptomeningeal hemorrhage” [4, 9]. This broader group of hemorrhages is responsible for more than 40% of intracranial hemorrhages in fetuses and neonates and occurs most frequently in preterm and low-birth-weight groups and in neonates who die within 24 hours after birth [9]. The amalgamation of subpial hemorrhage and subarachnoid hemorrhage in epidemiologic studies makes it difficult to assess the burden of subpial hemorrhage in the neonatal population. In Friede's series [1], which is one of the few studies to differentiate these anatomically distinct types of hemorrhage, subpial hemorrhage was estimated to represent as much as 15% of perinatal intracranial hemorrhages.
Proposed risk factors associated with leptomeningeal hemorrhages include neonatal asphyxia, birth injuries, fetal head molding, clotting disorders, venous sinus compression, variations in intracranial pressure, and incomplete regression of the primary vascular network [9]. Subpial hemorrhage has also been reported in patients with abusive head trauma [3, 5, 32, 33]. Interestingly, subpial hemorrhage has been described in otherwise healthy term neonates, in whom it was presumed to have occurred secondary to birth trauma [4].

Pathophysiology of Subpial Hemorrhage

Differentiation of locations of leptomeningeal hemorrhage is important because of the distinct pathophysiology and clinical sequelae [1, 9, 34]. Subpial hemorrhage sits underneath an impermeable pia mater, which is otherwise closely adherent to the circumjacent glia limitans, preventing prompt resorption of blood and invariably causing some degree of cortical damage [1, 5, 6, 34], analogous to what happens in compartment syndrome. Conversely, subarachnoid hemorrhage can diffuse through the CSF and eventually be absorbed, preventing direct mass effect on the cortex but presenting other potential risks, such as hydrocephalus (Table 1).
TABLE 1: Main Differences Between Types of Leptomeningeal Hemorrhage
CharacteristicHemorrhage Type
SubpialSubarachnoid
LocationUnder pia materAbove pia mater (epipial)
DistributionMultifocal or unifocal; usually asymmetricUsually spreads through CSF spaces
ShapeCrescent-shaped, oval, roundLaminar
Mass effectYesNo
Subjacent cortical infarctionYesNo
Extension into sulciYes, with splaying of sulcal marginsYes, but filling only, without splaying of sulcal margins
It has been proposed [1, 5, 6] that subpial hemorrhage is a primary insult to the glia limitans end-feet or to glial precursors (radial glia). The insult ultimately leads to focal disruption of the basement membrane and rupture of perforating cortical vessels with consequent pooling of blood in the adjacent SpS. This theory makes subpial hemorrhage, in the strict sense, a primary intracortical hemorrhage that is so superficial that it easily extends outside the cerebral gray matter and is contained under the pia mater. The ensuing mass effect can cause secondary compression of subpial vessels. Specifically, subpial veins are more susceptible to this local pressure increase because, compared with their arterial counterparts, they work under a lower blood pressure system and lack a leptomeningeal coating [29, 30]. Therefore, subpial hemorrhage causes local venous congestion and hypertension due to blockage of cortical venous outflow, which is normally directed toward the SpS. The result is focal cortical or subcortical infarction characteristic of subpial hemorrhage (Fig. 2).
Fig. 2 —Chart shows proposed pathophysiologic course of subpial hemorrhage.
Although primary insult to the glia limitans seems to be a plausible hypothesis for the pathogenesis of subpial hemorrhage, the open cranial sutures may also increase susceptibility for venous compression in neonates [4, 35]. In addition to primary glia limitans injury and subpial venous congestion, traumatic pial-glial disruption, erythrocyte diapedesis (i.e., leakage of RBCs from vessels), and primary rupture of small vessels in the SpS [9] may be factors in the pathogenesis of subpial hemorrhage.

Neuroimaging Characteristics of Subpial Hemorrhage

Neonates are vulnerable to brain injury, which makes neuro-imaging an important assessment tool. Natural sleep MRI (also called feed-and-bundle and wrap-and-go MRI) enables detailed neonatal brain MRI without the risks of radiation and anesthesia and sedation. Early diagnosis of subpial hemorrhage allows better prognostication, especially during the neonatal period, which is important considering the potential long-term consequences of these lesions, such as seizures and cognitive and functional deficits specific to the location of the lesion.
Subpial hemorrhages, which emerge in diverse clinical scenarios, typically feature heterogeneous blood collections along the margin of the cerebral parenchyma. The collections are usually characterized by ellipsoid, semielliptic, or spherical shape (often resembling an egg or scoop of ice cream) with the long axis tangential to the brain; a smooth, regular contour on the external surface; and distinct margins demarcating their breadth of spread. Subpial hemorrhages invariably abut and deeply displace the underlying cortical ribbon (Fig. 1). These morphologic features are highly suggestive of localized hemorrhages apposed to gyral and sulcal edges and sufficiently contained to produce a local mass effect on the cortex, but they have an extraaxial appearance (grossly, in a supracortical location outside the brain parenchyma). The hemorrhage is contained under the pia mater sheath (i.e., in the SpS) (Figs. 3 and 4).
Fig. 3A —35-week-old boy with congenital hydrocephalus secondary to aqueductal stenosis who underwent ventriculoperitoneal shunting on day 28 of life. Postoperative concern for meningitis was evaluated with head sonography and MRI. Example of multifocal subpial hemorrhage in meningitis after shunt placement for hydrocephalus.
A, T2-weighted MR image shows multifocal bilateral predominantly hypointense subpial hemorrhages (arrows) abutting brain cortex. Oval to round shape (resembling egg or scoop of ice cream) and well-demarcated appearance are evident. Hemorrhages impinge on cortical ribbon with mass effect. Fluid-fluid (hematocrit) level (arrowhead) within subpial hemorrhage should not be confused with clefts secondary to parenchymal lacerations.
Fig. 3B —35-week-old boy with congenital hydrocephalus secondary to aqueductal stenosis who underwent ventriculoperitoneal shunting on day 28 of life. Postoperative concern for meningitis was evaluated with head sonography and MRI. Example of multifocal subpial hemorrhage in meningitis after shunt placement for hydrocephalus.
B, Axial diffusion tracer image (B) and ADC map (C) show infarction of cortex, which is buckled underneath subpial hemorrhage.
Fig. 3C —35-week-old boy with congenital hydrocephalus secondary to aqueductal stenosis who underwent ventriculoperitoneal shunting on day 28 of life. Postoperative concern for meningitis was evaluated with head sonography and MRI. Example of multifocal subpial hemorrhage in meningitis after shunt placement for hydrocephalus.
C, Axial diffusion tracer image (B) and ADC map (C) show infarction of cortex, which is buckled underneath subpial hemorrhage.
Fig. 4A —8-day-old boy born at term with possible meconium aspiration, true umbilical knot, hypotonia, and perinatal depression referred for therapeutic hypothermia protocol for neonatal hypoxic-ischemic encephalopathy.
A, T2-weighted MR image shows hemorrhagic focus (arrow) on left temporal lobe overlying cortex but separated from adjacent CSF by smooth border, suggesting subpial location.
Fig. 4B —8-day-old boy born at term with possible meconium aspiration, true umbilical knot, hypotonia, and perinatal depression referred for therapeutic hypothermia protocol for neonatal hypoxic-ischemic encephalopathy.
B, Susceptibility-weighted image shows substantial blooming artifact, making it difficult to separate subpial hemorrhage from underlying cortical ribbon (arrowhead).
Fig. 4C —8-day-old boy born at term with possible meconium aspiration, true umbilical knot, hypotonia, and perinatal depression referred for therapeutic hypothermia protocol for neonatal hypoxic-ischemic encephalopathy.
C, DW image (C) and ADC (D) map confirm cortical infarction (arrow).
Fig. 4D —8-day-old boy born at term with possible meconium aspiration, true umbilical knot, hypotonia, and perinatal depression referred for therapeutic hypothermia protocol for neonatal hypoxic-ischemic encephalopathy.
D, DW image (C) and ADC (D) map confirm cortical infarction (arrow).
Head sonography is an excellent screening tool that has been used extensively to detect various types of neonatal brain injury, including (but not limited to) hypoxic-ischemic encephalopathy, ischemic stroke, and various forms of intracranial hemorrhage. One limitation of head sonography is its limited utility in depicting the periphery of the brain, where most subpial hemorrhages occur. Although some head sonography findings raise suspicion of subpial hemorrhage, they may not allow definitive diagnosis of subpial hemorrhage (Fig. 5). In some cases, CT depicts subpial hemorrhage (Fig. 5); however, MRI provides the most precise anatomic detail, and combinations of sequences enable evaluation of the full spectrum of injury in subpial hemorrhage.
Fig. 5A —Term male neonate with in utero tetrahydrocannabinol exposure and maternal hemorrhage during delivery.
A, Coronal sonogram of head obtained with linear probe shows heterogeneously hyperechogenic extraparenchymal collection (asterisks) in close apposition to cortex (arrow) of right frontal lobe, initially thought to be subdural, and additional parenchymal hemorrhage (arrowhead).
Fig. 5B —Term male neonate with in utero tetrahydrocannabinol exposure and maternal hemorrhage during delivery.
B, Axial unenhanced CT image shows large extraaxial hematoma (red arrowhead) with associated mass effect and lobar white matter hemorrhage (star) and splaying underlying sulcus (black arrowhead). Findings are possible hint to subpial location.
Fig. 5C —Term male neonate with in utero tetrahydrocannabinol exposure and maternal hemorrhage during delivery.
C, Unenhanced T2-weighted MR image confirms large hypointense lobar subpial hemorrhage separated from CSF (arrowheads) and closely following cortical outline of right frontal lobe. Related mass effect produces mild subfalcine herniation also appreciable with CT. Note subcortical white matter hemorrhage associated with subpial hemorrhage (star).
Fig. 5D —Term male neonate with in utero tetrahydrocannabinol exposure and maternal hemorrhage during delivery.
D, Diffusion-tensor tracer image (D) and ADC map (E) show extensive cytotoxic edema of frontal lobe cortical ribbon (arrow). T2 hypointensity is present in subcortical white matter (star, C), mirroring CT finding and characterizing venous infarction with hemorrhagic conversion in this context, likely caused by impaired deep medullary venous flow.
Fig. 5E —Term male neonate with in utero tetrahydrocannabinol exposure and maternal hemorrhage during delivery.
E, Diffusion-tensor tracer image (D) and ADC map (E) show extensive cytotoxic edema of frontal lobe cortical ribbon (arrow). T2 hypointensity is present in subcortical white matter (star, C), mirroring CT finding and characterizing venous infarction with hemorrhagic conversion in this context, likely caused by impaired deep medullary venous flow.
Fig. 5F —Term male neonate with in utero tetrahydrocannabinol exposure and maternal hemorrhage during delivery.
F, Axial time-of-flight MR angiogram obtained 2 days after initial presentation shows linear hyperintensity (arrowhead) surrounding external contour of subpial hemorrhage and following approximate contour of cortex but clearly separated from cortical ribbon and from bulk of subpial hemorrhage. Findings are consistent with heavily blood-stained lifted-off pia mater (which appears exquisitely hyperintense on time-of-flight images because of T1 shortening effects related both to presence of blood products and to pulse sequence characteristics).
Because it cannot easily diffuse through the collagen-rich SpS, blood collects in a contained localized hematoma between the lifted-off pial lining and the subjacent cortex on either side. The resulting mass effect explains the unique imaging feature of cortical inward depression (cortical buckling) and patchy cortical and subcortical injury (Fig. 1). At MRI, this complication is well visualized on the ADC map as truly restricted diffusion that delineates an area of cortical ribbon and subcortical white matter in close contact with the subpial hemorrhage focus (Figs. 3 and 4).
The dual venous drainage of cerebral white matter typically protects the subcortical white matter from venous ischemic injury [36]. The preferentially centripetal flow of the deep medullary veins toward the subependymal regions renders the deep white matter fairly resistant to subpial hemorrhage–related mass effect and may also redirect part of the subcortical white matter drainage through medullary anastomotic veins [36]. In contrast, intracortical and immediately subcortical venous drainage cannot be easily shifted toward the deep venous system. The difference in the venous drainage systems explains the more selective cortical and subcortical involvement seen on imaging. Depending on the extent of the hemorrhage, global mass effect, and associated lesions, it is possible to have accompanying deep white matter damage that is venous ischemic or hemorrhagic in nature [16, 33] (Fig. 5).
Susceptibility-weighted imaging has become part of many routine neonatal brain MRI protocols because it shows hemorrhage better than other sequences do (Fig. 6). However, evaluating susceptibility-weighted imaging alone can be misleading because the blooming artifact inherent to this technique blurs the interface between the cortex and the adjacent subpial hemorrhage, falsely suggesting the presence of a peripheral intraparenchymal hemorrhage (Fig. 6). Therefore, the importance of careful evaluation of anatomic T2-weighted images cannot be overemphasized.
Fig. 6A —Term male neonate delivered vaginally with forceps presenting with suspected birth trauma and hypoxic-ischemic encephalopathy evaluated after application of therapeutic hypothermia protocol. Example of importance of T2-weighted imaging for anatomic localization.
A, T2-weighted image shows large subpial hemorrhage compressing cortex (arrow) and causing cytotoxic edema (ADC map not shown).
Fig. 6B —Term male neonate delivered vaginally with forceps presenting with suspected birth trauma and hypoxic-ischemic encephalopathy evaluated after application of therapeutic hypothermia protocol. Example of importance of T2-weighted imaging for anatomic localization.
B, Axial susceptibility-weighted image shows blooming artifact (asterisks) that makes it difficult to differentiate subpial hemorrhage from adjacent white matter and ventricular hemorrhages, limiting anatomic identification of subpial hemorrhage with this sequence.
Another frequent characteristic of subpial hemorrhage is multiplicity and multifocality (Figs. 3 and 7). This characteristic favors a pathologic mechanism that could account for preconditioning of glia limitans and subpial vessels to bleed, explaining the existence of multiple simultaneously affected sites, as proposed by Friede [1]. Although the distribution of the hemorrhages seems random, we have observed a higher frequency of temporal and parietal lesions, consistent with previous observations [34]. A curious imaging aspect of acute subpial hemorrhage was found in one patient who underwent additional investigation with time-of-flight MR angiography: a linear hyperintensity approximately mirroring the cortical contour was visualized and presumed to be the blood-stained pia mater (Fig. 5). This case can serve as proof of the concept that MRI can be used to morphologically differentiate subpial hemorrhage from other hemorrhage types. The time-offlight sequence could add diagnostic information in some cases.
Fig. 7A —3-month-old girl born at 27 weeks' gestation presenting with posthemorrhagic hydrocephalus after shunt placement.
A, T2-weighted MR image obtained 8 days after shunt placement shows newly developed bilateral ventricular hemorrhage and multifocal subpial hemorrhages (arrows), many of which have fluid clefts along their deep margins (arrowheads). Despite artifact arising from shunt valve, subpial hemorrhages are visible.
Fig. 7B —3-month-old girl born at 27 weeks' gestation presenting with posthemorrhagic hydrocephalus after shunt placement.
B, Axial T2-weighted image from follow-up MRI 2 months after shunt placement shows large cystic areas (asterisk) where subpial hemorrhages had occurred. Right occipital hemorrhage (arrow) communicates with ventricles, suggesting full-thickness parenchymal damage.
With the temporal evolution of subpial hemorrhage foci, a thin layer of additional simple fluid is often seen in the deep aspect of the subpial hematoma, typically interposed between the subpial hemorrhage and the cortex, creating an imaging sign that resembles a cleft (Fig. 3). This cleft sign can become more evident on follow-up studies, possibly reflecting degradation of blood products that starts from the parenchymal interface (presumed leading edge of natural thrombolysis) with entrapment of serous fluid (Fig. 7). In late stages, subpial cystic cavities can be detected, possibly reflecting chronic organized residual hematoma or retained interstitial fluid in the SpS, which can also merge with adjacent cystic encephalomalacia and ventricles (Fig. 7).
Another remarkable feature of neonatal subpial hemorrhage is that it occurs in diverse clinical scenarios, including complicated shunted hydrocephalus, infection, asphyxia, and in utero drug exposure. Although some studies have shown subpial hemorrhage in association with abusive head trauma [3, 33], which we have also seen (Fig. 8), subpial hemorrhage is not considered a pathognomonic feature of abusive head trauma.
Fig. 8A —2-month-old boy born at 33 weeks' gestation presenting with rapidly progressive increase of head circumference.
A, Axial T2-weighted fast spin-echo MR image shows bilateral subdural hematohygromas (asterisk, arrowhead). Arrows indicate foci of subpial hemorrhage.
Fig. 8B —2-month-old boy born at 33 weeks' gestation presenting with rapidly progressive increase of head circumference.
B, Axial susceptibility-weighted minimum-intensity-projection MR image shows stretching of bridging veins (arrows) in bilateral frontal poles, which is associated with blooming artifact. Finding represents cortical venous injury with thromboses and lollipop sign.
Fig. 8C —2-month-old boy born at 33 weeks' gestation presenting with rapidly progressive increase of head circumference.
C, Diffusion tracer image shows diminutive foci of subpial hemorrhage following contour of cortex and adjacent to discrete foci of cortical ischemia (arrow), more evident on left. ADC map (not shown) showed corresponding decrease in signal intensity. Constellation of findings is consistent with abusive head trauma.

Prognosis

Because of its frequent multifocal nature and direct effect on cortical structure, subpial hemorrhage may be associated with a poor functional prognosis. Moreover, there is consolidated pathologic evidence that in some cases, subpial hemorrhage is a primary driver of acquired focal cortical dysplasia [5, 6, 23], which could also presumably affect the developmental prognosis of young patients who survive the injury. Additionally, chronic repair of perinatal subpial hemorrhage is associated not only with foci of cortical dysplasia but also with local vascular disorganization [5, 6]. Whether these disorganized, repaired vessels might predispose children to recurrence of subpial hemorrhage is unknown. To our knowledge, data on prognosis and recurrence after subpial hemorrhage are currently unavailable.

Conclusion

Subpial hemorrhage is a poorly recognized type of intracranial hemorrhage that can have a major detrimental effect on cortical integrity. It occurs frequently in neonates and infants in a variety of clinical scenarios. Hemorrhage in the SpS can be diagnosed or at least suspected on the basis of the typical imaging pattern we describe, especially on MRI.
Knowledge of the neuroimaging features of neonatal subpial hemorrhage should aid radiologists in recognizing this distinct type of intracranial hemorrhage. Knowing the historical and anatomic context should improve understanding of this unique pathologic process. Further study of subpial hemorrhage is needed, not only regarding pathophysiology and risk factors but also with respect to the clinical profile and long-term outcomes of patients who sustain such an injury. Future research should aim to address this gap in knowledge.

References

1.
Friede RL. Subpial hemorrhage in infants. J Neuropathol Exp Neurol 1972; 31:548–556
2.
Glaser J. The cerebrospinal fluid of premature infants: with special reference to intracranial hemorrhage and pigmentation. Am J Dis Child 1928; 36:195–247
3.
Gunda D, Cornwell BO, Dahmoush HM, Jazbeh S, Alleman AM. Pediatric central nervous system imaging of nonaccidental trauma: beyond subdural hematomas. RadioGraphics 2019; 39:213–228
4.
Huang AH, Robertson RL. Spontaneous superficial parenchymal and leptomeningeal hemorrhage in term neonates. AJNR 2004; 25:469–475
5.
Marín-Padilla M. Developmental neuropathology and impact of perinatal brain damage. Part 1. Hemorrhagic lesions of neocortex. J Neuropathol Exp Neurol 1996; 55:758–773
6.
Marín-Padilla M. The child's brain: normal (unaltered) development and development altered by perinatal injury. Rev Neurol 2013; 57(suppl 1):S3–S15
7.
Vlasiuk VV. Pathomorphology and differential diagnosis of intracerebral hemorrhages in newborn infants. Arkh Patol 1986; 48:44–51
8.
Vlasiuk VV, Keshelava SD. Pathomorphology and pathogenesis of leptomeningeal hemorrhages in fetuses and newborn infants. Arkh Patol 1986; 48:36–42
9.
Vlasyuk VV. Birth trauma and perinatal brain damage. Springer, 2019
10.
Volpe JJ. Volpe's neurology of the newborn, 6th ed. Elsevier, 2018
11.
von Haam E. Pathology of intracranial hemorrhage in the newborn child. Am J Obstet Gynecol 1934; 27:184–194
12.
Larsen KB, Barber Z, Squier W. The pathology and aetiology of subcortical clefts in infants. Forensic Sci Int 2019; 296:115–122
13.
Roth P, Happold C, Eisele G, Nägele T, Weller M, Luft AR. A series of patients with subpial hemorrhage: clinical manifestation, neuroradiological presentation and therapeutic implications. J Neurol 2008; 255:1018–1022
14.
Squier W, Austin T, Anslow P, Weller RO. Infant subcortical cystic leucomalacia: a distinct pathological entity resulting from impaired fluid handling. Early Hum Dev 2011; 87:421–426
15.
Suzuki K, Matsuoka G, Abe K, Okada Y, Sakai S. Subpial hematoma and extravasation in the interhemispheric fissure with subarachnoid hemorrhage. Neuroradiol J 2015; 28:337–340
16.
Cain DW, Dingman AL, Armstrong J, Stence NV, Jensen AM, Mirsky DM. Subpial hemorrhage of the neonate. Stroke 2020; 51:315–318
17.
Adeeb N, Deep A, Griessenauer CJ, et al. The intracranial arachnoid mater: a comprehensive review of its history, anatomy, imaging, and pathology. Childs Nerv Syst 2013; 29:17–33
18.
Adeeb N, Mortazavi MM, Deep A, et al. The pia mater: a comprehensive review of literature. Childs Nerv Syst 2013; 29:1803–1810
19.
Mortazavi MM, Quadri SA, Khan MA, et al. Subarachnoid trabeculae: a comprehensive review of their embryology, histology, morphology, and surgical significance. World Neurosurg 2018; 111:279–290
20.
Albargothy NJ, Johnston DA, MacGregor-Sharp M, et al. Convective influx/glymphatic system: tracers injected into the CSF enter and leave the brain along separate periarterial basement membrane pathways. Acta Neuropathol 2018; 136:139–152
21.
Alcolado R, Weller RO, Parrish EP, Garrod D. The cranial arachnoid and pia mater in man: anatomical and ultrastructural observations. Neuropathol Appl Neurobiol 1988; 14:1–17
22.
Bakker EN, Bacskai BJ, Arbel-Ornath M, et al. Lymphatic clearance of the brain: perivascular, paravascular and significance for neurodegenerative diseases. Cell Mol Neurobiol 2016; 36:181–194
23.
Choi BH. Role of the basement membrane in neurogenesis and repair of injury in the central nervous system. Microsc Res Tech 1994; 28:193–203
24.
Hutchings M, Weller RO. Anatomical relationships of the pia mater to cerebral blood vessels in man. J Neurosurg 1986; 65:316–325
25.
Kida S, Weller RO, Zhang ET, Phillips MJ, Iannotti F. Anatomical pathways for lymphatic drainage of the brain and their pathological significance. Neuropathol Appl Neurobiol 1995; 21:181–184
26.
Morris AW, Sharp MM, Albargothy NJ, et al. Vascular basement membranes as pathways for the passage of fluid into and out of the brain. Acta Neuropathol 2016; 131:725–736
27.
Nicholas DS, Weller RO. The fine anatomy of the human spinal meninges: a light and scanning electron microscopy study. J Neurosurg 1988; 69:276–282
28.
Weller RO. Microscopic morphology and histology of the human meninges. Morphologie 2005; 89:22–34
29.
Weller RO, Sharp MM, Christodoulides M, Carare RO, Møllgård K. The meninges as barriers and facilitators for the movement of fluid, cells and pathogens related to the rodent and human CNS. Acta Neuropathol 2018; 135:363–385
30.
Zhang ET, Inman CB, Weller RO. Interrelationships of the pia mater and the perivascular (Virchow-Robin) spaces in the human cerebrum. J Anat 1990; 170:111–123
31.
Saunders NR, Liddelow SA, Dziegielewska KM. Barrier mechanisms in the developing brain. Front Pharmacol 2012; 3:46
32.
Squier W. The “shaken baby” syndrome: pathology and mechanisms. Acta Neuropathol 2011; 122:519–542
33.
Orman G, Kralik SF, Meoded A, Desai N, Risen S, Huisman TAGM. MRI findings in pediatric abusive head trauma: a review. J Neuroimaging 2020; 30:15–27
34.
Friede RL. Hemorrhages in asphyxiated premature infants. In: Developmental neuropathology, 2nd ed. Springer-Verlag, 1989:44–58
35.
Padget DH. The cranial venous system in man in reference to development, adult configuration, and relation to the arteries. Am J Anat 1956; 98:307–355
36.
Okudera T, Huang YP, Fukusumi A, Nakamura Y, Hatazawa J, Uemura K. Micro-angiographical studies of the medullary venous system of the cerebral hemisphere. Neuropathology 1999; 19:93–111

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 1056 - 1065
PubMed: 33566637

History

Submitted: February 20, 2020
Revision requested: March 18, 2020
Revision received: May 15, 2020
Accepted: June 1, 2020
Version of record online: February 10, 2021

Keywords

  1. MRI
  2. neonate
  3. neuroimaging
  4. subpial hemorrhage

Authors

Affiliations

André R. F. Barreto, MD
Russell H. Morgan Department of Radiology and Radiological Science, Division of Pediatric Radiology and Pediatric Neuroradiology, The Johns Hopkins School of Medicine, Charlotte R. Bloomberg Children's Center, The Johns Hopkins Hospital, 1800 Orleans St, Baltimore, MD 21287
Melisa Carrasco, MD, PhD
Division of Pediatric Neurology, The Johns Hopkins School of Medicine, Baltimore, MD
Ania K. Dabrowski, MD, PhD
Division of Pediatric Neurology, The Johns Hopkins School of Medicine, Baltimore, MD
Lisa R. Sun, MD
Division of Pediatric Neurology, The Johns Hopkins School of Medicine, Baltimore, MD
Aylin Tekes, MD
Russell H. Morgan Department of Radiology and Radiological Science, Division of Pediatric Radiology and Pediatric Neuroradiology, The Johns Hopkins School of Medicine, Charlotte R. Bloomberg Children's Center, The Johns Hopkins Hospital, 1800 Orleans St, Baltimore, MD 21287

Notes

Address correspondence to A. Tekes ([email protected]).
The authors declare that they have no disclosures relevant to the subject matter of this article.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media