Advertisement

Neonatal white matter damage and the fetal inflammatory response

Open AccessPublished:April 09, 2020DOI:https://doi.org/10.1016/j.siny.2020.101111

      Abstract

      In 1962 a long-recognized pathologic abnormality in neonatal brains characterized by multiple telencephalic focal white matter necroses was renamed periventricular leukomalacia (PVL) and the authors inappropriately asserted that their entity was caused by anoxia. They also failed to include three other white matter histologic abnormalities.
      In this essay, we identify the breadth of white matter pathology, especially in very preterm newborns, and show that none of the four histologic expressions of white matter damage, including focal necrosis, are associated with hypoxemia or correlates as hypotension, but are instead associated with markers of fetal or perinatal inflammation, particularly in preterm babies. We begin with the background needed to evaluate the evidence.

      1. Periventricular leucomalacia

      Betty Banker and Jeanne Claudia Larroche gave their 1962 article the title of “Periventricular leukomalacia of infancy. A form of neonatal anoxic encephalopathy” [
      • Banker B.Q.
      • Larroche J.C.
      Periventricular leukomalacia of infancy. A form of neonatal anoxic encephalopathy.
      ]. The term “Periventricular Leucomalacia” is a misleading redescription of an old pathologic entity of multiple focal white matter necroses in the neonatal brain. Schmorl, in 1904, summarized five late 19th century publications in noting the distribution of small focal necroses from subcortical U-fibers to the ventricular wall [
      • Schmorl C.G.
      Zur kenntniss des ikterus neonatorum, inbesondere der dabei auftretenden gehirnveranderungen.
      ]; Parrot (1868) pointed out that no associated traditional inflammatory cells were seen in or about these lesions, and thought the necroses resulted from a general metabolic disorder [
      • Parrot M.J.
      Etude sur la stèatose interstitielle diffuse de l'encèphale chez le nouveau-nè.
      ]; others attributed these focal necroses to thrombosis, omphalitis, emboli, or icterus [
      • Schmorl C.G.
      Zur kenntniss des ikterus neonatorum, inbesondere der dabei auftretenden gehirnveranderungen.
      ]. Subsequently, anoxia (usually following parturitional circulatory disturbance) [
      • Schwartz P.
      Birth Injuries of the Newborn, Morphology, Pathogenesis, Clinical Pathology and Prevention.
      ,
      • Clark D.B.
      • Anderson G.W.
      Correlation of complications of labor with lesions in the brains of infants.
      ] and endotoxemia were considered likely explanations [
      • Gluszcz A.
      On the periventricular septic necroses of the brain in premature infants.
      ,
      • Perrin E.V.
      • Landing B.H.
      The Schmorl lesion" in jaundiced infected infants.
      ]. None of these inferences were data-driven or based on comparisons to controls.
      Banker and Larroche drew the anoxic inference because all of the infants whose brains had periventricular leukomalacia had experienced apnea, cardiac arrest, or cyanosis, were given supplemental oxygen, or had lung pathology. They committed two inferential errors. First, they did not compare the frequency of these events/characteristics among children with PVL to those of children without PVL. Had they done so, they might have recognized that focal white matter necroses are not associated with episodes of hypoxia, or systemic hypotension [
      • Leviton A.
      • Gilles F.H.
      The epidemiology of the perinatal telencephalic leucoencephalopathy characterized by focal necroses.
      ,
      • Leviton A.
      • Gilles F.H.
      Acquired perinatal leukoencephalopathy.
      ,
      • Dammann O.
      • Leviton A.
      Perinatal brain damage causation.
      ]. It is unfortunate that their faulty inference continues to be offered as truth: “PVL is caused by a lack of oxygen or blood flow to the periventricular area of the brain, which results in the death or loss of brain tissue” []. This quote from a 2019 NIH webpage exemplifies “Fixation of Belief” [
      • Peirce C.S.
      The fixation of Belief.
      ] and the etiologic misinformation disseminated by Banker and Larroche in 1962.
      Second, they allowed terminal events to be considered “causes” of a disorder that most likely required days to develop. These inferences led to missing the clinical associations accompanying inflammatory events.
      Among additional errors made by Banker and Larroche was their failure to identify three other histologic abnormalities found in maturing telencephalic white matter. Apparently, they either did not see amphophilic globules, hypertrophic astrocytes, or acutely damaged glia in the brains of children, or if they had seen them, they did not consider them abnormal or important. To us, however, these histologic features were unquestionably abnormal [
      • Gilles F.
      • Murphy S.
      Perinatal telencephalic leucoencephalopathy.
      ,
      • Leviton A.
      • Gilles F.H.
      An epidemiologic study of perinatal telencephalic leucoencephalopathy in an autopsy population.
      ,
      • Leviton A.
      • Gilles F.
      Etiologic relationships among the perinatal telencephalic leucoencephalopathies.
      ,
      • Leviton A.
      • Gilles F.H.
      Classification of the perinatal telencephalic leucoencephalopathies.
      ]. The name (perinatal telencephalic) leucoencephalopathy (PTL) was intended to indicate white matter damage marked by any of the four histologic abnormalities.
      Hypoxia, or more appropriately hypoxemia, if sufficiently prolonged induces energy failure, and is associated with loss of neuronal function and death, much less in neonatal than in adult animals, and least in prematurely born animals [
      • Bennet L.
      • Roelfsema V.
      • Dean J.M.
      • Wassink G.
      • Power G.G.
      • Jensen E.C.
      • et al.
      Regulation of cytochrome oxidase redox state during umbilical cord occlusion in preterm fetal sheep.
      ]. The very preterm brain normally has much lower aerobic requirements than the term brain [
      • Gunn A.J.
      • Quaedackers J.S.
      • Guan J.
      • Heineman E.
      • Bennet L.
      The premature fetus: not as defenseless as we thought, but still paradoxically vulnerable?.
      ], but, to the best of our knowledge, hypoxia alone, with sustained cerebral perfusion, has not been shown to cause brain lesions, either in preterm, term, or adult humans or in experimental animals. While certain selective neuronal losses in the adult brain have been attributed to hypoxia, such specificity has yet to be verified. Some have conflated hypoxia with ischemia in the term “hypoxic-ischemic encephalopathy.” Ischemia, a general or focal restriction in tissue blood supply, also induces energy failure, but is far more complex than hypoxia in that it also diminishes blood component availability, as well as limits brain metabolic waste removal. Often this results from systemic hypotension or cerebral vascular occlusion. We have yet to find convincing evidence that hypoxia-ischemia contributes to brain damage in the preterm brain [
      • Gilles F.
      • Gressens P.
      • Dammann O.
      • Leviton A.
      Hypoxia-ischemia is not an antecedent of most preterm brain damage: the illusion of validity.
      ].
      We begin this issue of Seminars in Fetal & Neonatal Medicine by recounting how ignored publications in the following decades tried to change the conversation from faulty reasoning to data-driven inferences.

      2. Perinatal telencephalic leucoencephalopathy (PTL)

      The white matter abnormalities encountered in fetal brain after midgestation include widely distributed hypertrophic astrocytes in about 38% of neonatal brains, acutely damaged glial cells in about 38% of neonatal brains, amphophilic globules in about one-third of brains from the third trimester, and multiple focal necroses in only about 8%. Cortical necroses were seen in only about 2% [
      • Gilles F.
      • Murphy S.
      Perinatal telencephalic leucoencephalopathy.
      ,
      • Gilles F.H.
      Neural damage: inconstancy during gestation.
      ]. The proportion of brains with hypertrophic astrocytes increases with gestational age from 9% at 20 weeks to 63% at term. Amphophilic globules occur in about one-third of brains in the third trimester, while acutely damaged glial cells occur in about one-quarter of brains after 32 weeks’ gestation and multiple focal necroses occur in less than 10% of brains. These changes occur during rapid brain growth [
      • McLennan J.E.
      • Gilles F.H.
      • Neff R.
      A model of growth of the human fetal brain.
      ], when new vasculature is forming [
      • Kuban K.C.K.
      • Gilles F.H.
      Human telencephalic angiogenesis.
      ], and myelination is beginning [
      • Gilles F.
      • Nelson M.J.
      The developing human brain; Growth and Adversities.
      ].

      3. The nature of the four histologic abnormalities

      3.1 Hypertrophic astrocytes

      Like activated or hypertrophic (“reactive”) astrocytes at older ages, hypertrophic astrocytes in fetal telencephalic white matter have glial fibrils. However, they tend to have less cytoplasm [
      • Gilles F.
      • Murphy S.
      Perinatal telencephalic leucoencephalopathy.
      ,
      • Dąmbska M.
      Encephalic necrosis and inflammatory reaction in fetuses and newborns.
      ,
      • Ginsberg Y.
      • Khatib N.
      • Weiner Z.
      • Beloosesky R.
      Maternal inflammation, fetal brain implications and suggested neuroprotection: a summary of 10 Years of research in animal models.
      ].
      The time interval between an apparent insult and the presence of astroglial cell hypertrophy in newborns tends to be short (only 2–3 days in kittens) and varies with differing insults [
      • Cajal R.
      Degeneration and regeneration of the nervous system.
      ,
      • Gilles F.H.
      • Averill Jr., D.R.
      • Kerr C.S.
      Neonatal endotoxin encephalopathy.
      ]. These reactive astrocytes are not myelinating glia as asserted by some [
      • Ellison D.
      • Love S.
      Neuropathology.
      ], because myelinating glia have large pale nuclei, no eosinophilic cytoplasm, no GFAP positive glial fibrils, and do not have secular trends, but do have age and location trends. These hypertrophic astrocytes are limited to the white matter, are not found in the caudate, putamen, thalamus or cortex, and can sometimes be seen in abundance unaccompanied by other histologic indicators of white matter damage [
      • Leviton A.
      • Gilles F.H.
      Morphologic abnormalities in human infant cerebral white matter related to gestational and postnatal age.
      ]. They are distributed either diffusely (in about one-third of newborns) in fetal hemispheric white matter or are concentrated focally around necroses. When diffusely distributed, they are usually symmetrical in both hemispheres without much increase in the total number of glial cells, suggesting a modification of a pre-existing immature glial cell, such as a preoligodendroglial cell. See Fig. 1.
      Fig. 1
      Fig. 1A large hypertrophic astrocyte is marked by the wide arrow. There are multiple hypertrophic astrocytes with in the field. An acutely damaged glial cell is marked by the narrow arrow. The small nucleus has an intact nuclear membrane. The large open nuclei are myelination glia. Microglia are not present in this field.
      Differing brain injuries induce two different types of “reactive” astrocytes [
      • Kanski R.
      • van Strien M.E.
      • van Tijn P.
      • Hol E.M.
      A star is born: new insights into the mechanism of astrogenesis.
      ,
      • Liddelow S.A.
      • Guttenplan K.A.
      • Clarke L.E.
      • Bennett F.C.
      • Bohlen C.J.
      • Schirmer L.
      • et al.
      Neurotoxic reactive astrocytes are induced by activated microglia.
      ], each with its own transcriptome [
      • Zamanian J.L.
      • Xu L.
      • Foo L.C.
      • Nouri N.
      • Zhou L.
      • Giffard R.G.
      • et al.
      Genomic analysis of reactive astrogliosis.
      ]. A1 (neuroinflammatory) reactive astrocytes are induced by lipopolysaccharide (LPS) activated microglia, which release IL1a, TNF-α, and the complement component subunit 1q (C1q)(C3) and “help to drive the death of neurons and oligodendrocytes” [
      • Liddelow S.A.
      • Barres B.A.
      Reactive astrocytes: production, function, and therapeutic potential.
      ]. The JAK-STAT3 pathway activates A2 (ischemia-induced total tissue necrosis) scar-forming astrocytes.

      3.2 Acutely damaged glia

      Acutely damaged glial nuclei are pyknotic, irregular, hyperchromic, and faintly amphophilic with intact nuclear membranes, and are not apoptotic or associated with a histiocytic or macrophage response [
      • Gilles F.
      • Murphy S.
      Perinatal telencephalic leucoencephalopathy.
      ]. They share the same topographic distributions as hypertrophic astrocytes. It remains unclear if these acutely damaged glia are damaged premyelin glial cells or whether they are on their way to becoming hypertrophic astrocytes. See Fig. 1.

      3.3 Amphophilic globules

      Amphophilic globules are small amphophilic or basophilic deposits, sometimes laminated, usually in proximity to white matter vascular channels and may consist of damaged endothelium, other components of the blood brain barrier (BBB) or precipitation of protein leak [
      • Stolp H.B.
      • Dziegielewska K.M.
      • Ek C.J.
      • Habgood M.D.
      • Lane M.A.
      • Potter A.M.
      • et al.
      Breakdown of the blood-brain barrier to proteins in white matter of the developing brain following systemic inflammation.
      ] and subsequent mineralization [
      • McCartney E.
      • Squier W.
      Patterns and pathways of calcification in the developing brain.
      ] (Fig. 2). They contain carbohydrate residues, calcium, and protein, but no iron [
      • Gilles F.
      • Murphy S.
      Perinatal telencephalic leucoencephalopathy.
      ], and do not elicit a local glial response.
      Fig. 2
      Fig. 2At the bifurcation of this capillary there is a large mineralized globule marked by the arrow.

      3.4 Multiple focal white matter necroses

      White matter focal necroses range from simple small coagulative necroses to organized large cystic lesions, narrowly surrounded by hypertrophic astrocytes, frequently with nearby axonal retraction balls and other debris. They are usually multiple and are distributed from subcortical white matter to the lateral ventricular walls, with the largest near the ventricle. Some recent coagulative necroses occur at the edge of older organized necroses. In mid-telencephalic coronal sections, necroses tend to be linear [
      • Schwartz P.
      Die traumatischen schadigungen des zentralnerversystems durch die geburt.
      ,
      • Leech R.W.
      • Alvord Jr., E.C.
      Morphologic variations in periventricular leukomalacia.
      ,
      • Shuman R.M.
      • Selednik L.J.
      Periventricular leukomalacia. A one-year autopsy study.
      ], roughly perpendicular to the ventricular wall, seemingly following transmural vascular channels [
      • Kuban K.C.K.
      • Gilles F.H.
      Human telencephalic angiogenesis.
      ,
      • Nelson M.D.
      • Gonzalez-Gomez I.
      • Gilles F.H.
      The search for human telencephalic ventriculofugal arteries.
      ,
      • Gilles F.H.
      • Nelson M.D.J.
      • Gonzalez-Gomez I.
      Human telencephalic angiogenesis: an update.
      ]. In polar coronal sections they tend to be round or oblate.(Fig. 3). Although most of the hemispheric white matter may contain these lesions, nearby isocortex, hippocampus, striatum, globus pallidus, amygdala, and thalamus are relatively spared.
      Fig. 3
      Fig. 3In this mid-telencephalic slab through the thalamus focal necroses are seen in the white matter of this image by Schwartz. Note the linear nature of the necroses.
      While focal necroses occur in both hemispheres, they are rarely symmetric in location as might be expected following systemic hypotension and occur in almost all telencephalic white matter regions except for the anterior temporal lobe. Multiple small focal necroses are far more common than very large lesions. Focal white matter necroses are sometimes hemorrhagic, as expected when capillary or sinusoidal endothelium is damaged and erythrocytes leak into necrotic regions.
      Regardless of where along the axis from apoptosis through necroptosis [
      • Vandenabeele P.
      • Galluzzi L.
      • Vanden Berghe T.
      • Kroemer G.
      Molecular mechanisms of necroptosis: an ordered cellular explosion.
      ] and pyroptosis [
      • Denes A.
      • Lopez-Castejon G.
      • Brough D.
      Caspase-1: is IL-1 just the tip of the ICEberg?.
      ] they occur, focal white matter necroses by themselves give no clues about etiology, in part because necrosis results from many different metabolic, vascular, or inflammatory derangements, each often the consequence of multiple different antecedents [
      • Leviton A.
      Single-cause attribution.
      ]. Similarly, each of the three other histologic abnormalities has multiple sets of risk factors.
      Banker and Larroche and others have propagated the concept of deep white matter vascular border zones [
      • DeReuck J.
      The human periventricular arterial blood supply and the anatomy of cerebral infarctions.
      ,
      • Chattha A.S.
      • Richardson Jr., E.P.
      Cerebral white-matter hypoplasia.
      ,
      • Takashima S.
      • Tanaka K.
      Development of cerebrovascular architecture and its relationship to periventricular leukomalacia.
      ,
      • Volpe J.J.
      Neurology of the newborn.
      ] leading to a second assertion that deep hemispheric white matter contains border zones between recurrent collateral arteries and deep penetrators. This notion was based upon radiographs of thick brain slabs with incompletely barium-filled vascular beds without histologic controls, and the described recurrent collaterals (some even traversing the ventricle from the choroid plexus to the deep white matter) forming border zones in deep white matter [
      • DeReuck J.
      The human periventricular arterial blood supply and the anatomy of cerebral infarctions.
      ]. However, in histologic studies with completely filled vascular beds, no endarterial border zones or recurrent collaterals are present in deep white matter [
      • Kuban K.C.K.
      • Gilles F.H.
      Human telencephalic angiogenesis.
      ,
      • Gilles F.H.
      • Nelson M.D.J.
      • Gonzalez-Gomez I.
      Human telencephalic angiogenesis: an update.
      ,
      • Bär T.
      The vascular system of the cerebral cortex (Review).
      ,
      • Nelson Jr., M.D.
      • Gonzalez-Gomez I.
      • Gilles F.H.
      • Dyke Award
      The search for human telencephalic ventriculofugal arteries.
      ].
      We speculate, following Friede [
      • Friede R.L.
      Developmental neuropathology.
      ], that focal necroses follow transient obstruction of small transcerebral vascular channels [
      • Kuban K.C.K.
      • Gilles F.H.
      Human telencephalic angiogenesis.
      ], not a systemic abnormality such as anoxia, hypotension, or a diffuse change in pre-myelinated oligodendroglia. However, against this suggestion is the fact that only a few observers have found vascular obstructions. The absence of thrombi suggests that thrombi are transient such as platelets [
      • Sjobring U.
      • Ringdahl U.
      • Ruggeri Z.M.
      Induction of platelet thrombi by bacteria and antibodies.
      ], fibrin, shed endothelial cells, or thrombi that cannot be seen in conventional section stains, such as fat emboli. Another possibility is that muscularized leptomenigeal vessels constrict secondary to endothelin or thromboxane [
      • Feng S.Y.
      • Phillips D.J.
      • Stockx E.M.
      • Yu V.Y.
      • Walker A.M.
      Endotoxin has acute and chronic effects on the cerebral circulation of fetal sheep.
      ] prior to the origin of transpial vascular penetrators, or that pericytes of the BBB constrict.
      In light of recent developments, inflammatory phenomena are likely involved in contributing to coagulation/thrombosis [
      • van der Poll T.
      • Buller H.R.
      • ten Cate H.
      • Wortel C.H.
      • Bauer K.A.
      • van Deventer S.J.
      • et al.
      Activation of coagulation after administration of tumor necrosis factor to normal subjects.
      ] and necrosis [
      • Leviton A.
      • Fichorova R.
      • Yamamoto Y.
      • Allred E.N.
      • Dammann O.
      • Hecht J.
      • et al.
      Inflammation-related proteins in the blood of extremely low gestational age newborns. The contribution of inflammation to the appearance of developmental regulation.
      ,
      • Yanni D.
      • Korzeniewski S.J.
      • Allred E.N.
      • Fichorova R.N.
      • O'Shea T.M.
      • Kuban K.
      • et al.
      Both antenatal and postnatal inflammation contribute information about the risk of brain damage in extremely preterm newborns.
      ,
      • Leviton A.
      • Joseph R.M.
      • Allred E.N.
      • Fichorova R.N.
      • O'Shea T.M.
      • Kuban K.K.C.
      • et al.
      The risk of neurodevelopmental disorders at age 10 years associated with blood concentrations of interleukins 4 and 10 during the first postnatal month of children born extremely preterm.
      ,
      • Gussenhoven R.
      • Westerlaken R.J.J.
      • Ophelders D.
      • Jobe A.H.
      • Kemp M.W.
      • Kallapur S.G.
      • et al.
      Chorioamnionitis, neuroinflammation, and injury: timing is key in the preterm ovine fetus.
      ]. This might include the recruitment of circulating macrophages by circulating inflammatory proteins, the activation of resident microglia by proteins that crossed the endothelial barrier, and/or the relatively direct toxic effects of these inflammatory proteins on vulnerable pre-oligodendroglia [
      • Varol C.
      • Mildner A.
      • Jung S.
      Macrophages: development and tissue specialization.
      ,
      • Dean J.M.
      • Shi Z.
      • Fleiss B.
      • Gunn K.C.
      • Groenendaal F.
      • van Bel F.
      • et al.
      A critical review of models of perinatal infection.
      ,
      • Mallard C.
      • Tremblay M.E.
      • Vexler Z.S.
      Microglia and neonatal brain injury.
      ].

      4. Clustering and classification of these four markers of white matter damage

      The tendency of biologic phenomena to occur together can provide valuable information about systems, heterogeneity, and etiology. Cluster analysis of the four histologic features in the Boston Children's Hospital sample identified three clusters [
      • Leviton A.
      • Gilles F.H.
      Clustering of the morphological components of perinatal telencephalic leucoencephalopathy.
      ]. The cluster of acutely damaged glia and amphophilic globules did not occur preferentially with any other cluster. Hypertrophic astrocytes and amphophilic globules, on the other hand, clustered in the sample of brains that did not have any necrotic foci, while in the total sample the cluster of hypertrophic astrocytes and amphophilic globules clustered with necrotic foci [
      • Leviton A.
      • Gilles F.H.
      Clustering of the morphological components of perinatal telencephalic leucoencephalopathy.
      ].
      Initially, the name perinatal telencephalic leucoencephalopathy (PTL) was intended to indicate any white matter damage [
      • Gilles F.
      • Murphy S.
      Perinatal telencephalic leucoencephalopathy.
      ]. The subsequent recognition that the cluster of hypertrophic astrocytes and amphophilic globules was strong in its own right, and also clustered with necrotic foci, prompted limiting the PTL name to the entities characterized by the cluster of hypertrophic astrocytes and amphophilic globules, at least for early epidemiologic purposes. Subsequently, epidemiologic analyses of other clusters were reported [
      • Leviton A.
      • Gilles F.H.
      Acquired perinatal leukoencephalopathy.
      ].

      5. The epidemiology of histologically defined perinatal telencephalic leukoencephalopathy

      During the three years, 1965–1967, the frequency of PTL (defined as the finding of both hypertrophic astrocytes and amphophilic globules) in infants who died at Boston Children's Hospital varied very similarly to the frequency bacteria were recovered from cardiac blood aspirated at the time of postmortem examination, both peaking during the first quarter of 1967 [
      • Leviton A.
      • Gilles F.H.
      An epidemiologic study of perinatal telencephalic leucoencephalopathy in an autopsy population.
      ]. If postmortem bacteremia can be accepted as a surrogate for antenatal bacteremia, then bacteremia considerably before death can be seen as a risk factor for PTL (Table 1).
      Table 1The proportion of cases containing hypertrophic astrocytes and acutely damage glia increased throughout gestation while the proportion of cases containing focal necroses was largely a phenomenon of prematurity. In the same database, the proportion of cases containing gray matter necrosis of cortex or deep gray structures was only 2%.
      Percentage of cases in each gestational period
      Gestational Weeks20–2324–2728–3132–3536–3939 +
      Amphophilic Globules81431303235
      Hypertrophic

      Asrocytes
      91434465963
      Focal Necroses186861
      Acutely Damaged Glia2510202825
      On multivariate analyses, the risk of PTL was prominently elevated among children who had a clinical diagnosis of peritonitis, received kanamycin, received streptomycin, and had “postmortem” bacteremia [
      • Leviton A.
      • Gilles F.
      • Neff R.
      • Yaney P.
      Multivariate analysis of risk of perinatal telencephalic leucoencephalopathy.
      ]. Taken together, this combination of risk factors speaks strongly to the likelihood that the postmortem bacteremia reflected antenatal systemic infection/inflammation.
      The choice of antibiotics was seen as an indication that the children ’s physicians recognized gram-negative infection. This was in keeping with post-mortem laboratory reports that a gram-negative organism was recovered from the post-mortem aspirates of cardiac blood in 85% of the PTL infants who had post-mortem bacteremia.
      Even what was known 50 years ago suggested then that PTL might represent the adverse consequence of endotoxin on myelinogenesis or some other maturational process unique to infant white matter. Thus was born the “endotoxin hypothesis of neonatal white matter damage.” It was supported by the observation that infants with PTL tended to have lower weights of thymus and spleen, and were more likely to have thymic atrophy histologically [
      • Leviton A.
      • Gilles F.H.
      • Vawter G.F.
      The thymus in infants with perinatal telencephalic leukoencephalopathy.
      ].

      6. Lipopolysaccharide

      Because of the association between white matter abnormalities and neonatal bacteremia [
      • Leviton A.
      • Gilles F.H.
      An epidemiologic study of perinatal telencephalic leucoencephalopathy in an autopsy population.
      ], newborn kittens were infused intraperitoneally with a purified E. coli endotoxin, lipopolysaccharide (LPS), daily for 3–20 days [
      • Gilles F.H.
      • Leviton A.
      • Kerr C.S.
      Endotoxin leukoencephalopathy in the telencephalon of the newborn kitten.
      ]. All the kitten brains had cellular and fibrillary astrogliosis and amphophilic globules. Some also had focal areas of cystic necrosis, and deposits of eosinophilic or mineralized debris.
      The brains of kittens sacrificed two weeks following the insult tended to have diffuse white matter astrogliosis, either in patches or in larger regions. Large totally necrotic regions were also seen occasionally. The brains of kittens sacrificed two years after endotoxin exposure often had fibrillary gliosis and “a marked paucity of hemispheric white matter.” [
      • Gilles F.H.
      • Averill Jr., D.R.
      • Kerr C.S.
      Changes in neonatally induced cerebral lesions with advancing age.
      ] These are viewed as the late consequences of earlier white matter damage. While the focal white matter necroses from LPS administered to neonatal kitten are not associated with inflammatory cells [
      • Gilles F.H.
      • Averill Jr., D.R.
      • Kerr C.S.
      Neonatal endotoxin encephalopathy.
      ,
      • Gilles F.H.
      • Leviton A.
      • Kerr C.S.
      Endotoxin leukoencephalopathy in the telencephalon of the newborn kitten.
      ], those in neonatal dogs are [
      • Young R.S.
      • Yagel S.K.
      • Towfighi J.
      Systemic and neuropathologic effects of E. coli endotoxin in neonatal dogs.
      ].
      In a separate set of studies, 24 h after a single intraperitoneal LPS injection, rabbits and monkeys (Macaca mullatta) often had coagulative necrosis in various regions of the white matter [
      • Gilles F.H.
      • Averill Jr., D.R.
      • Kerr C.S.
      Neonatal endotoxin encephalopathy.
      ]. In less severely damaged brains, karyorrhectic nuclei were widely scattered in telencephalic white matter outside regions of coagulative necrosis.
      Others have found that LPS is linked to white matter damage, diminished oligodendrocyte density, gliosis, and hypomyelination [
      • Cai Z.
      • Pan Z.L.
      • Pang Y.
      • Evans O.B.
      • Rhodes P.G.
      Cytokine induction in fetal rat brains and brain injury in neonatal rats after maternal lipopolysaccharide administration.
      ,
      • Pang Y.
      • Campbell L.
      • Zheng B.
      • Fan L.
      • Cai Z.
      • Rhodes P.
      Lipopolysaccharide-activated microglia induce death of oligodendrocyte progenitor cells and impede their development.
      ,
      • Rousset C.I.
      • Chalon S.
      • Cantagrel S.
      • Bodard S.
      • Andres C.
      • Gressens P.
      • et al.
      Maternal exposure to LPS induces hypomyelination in the internal capsule and programmed cell death in the deep gray matter in newborn rats.
      ,
      • Wang X.
      • Rousset C.I.
      • Hagberg H.
      • Mallard C.
      Lipopolysaccharide-induced inflammation and perinatal brain injury.
      ,
      • Cardoso F.L.
      • Herz J.
      • Fernandes A.
      • Rocha J.
      • Sepodes B.
      • Brito M.A.
      • et al.
      Systemic inflammation in early neonatal mice induces transient and lasting neurodegenerative effects.
      ]. Mallard et al. also found two varieties of astrocytes [
      • Mallard C.
      • Welin A.K.
      • Peebles D.
      • Hagberg H.
      • Kjellmer I.
      White matter injury following systemic endotoxemia or asphyxia in the fetal sheep.
      ]. LPS itself barely crosses the BBB but cytokines induced by LPS do [
      • Banks W.A.
      • Robinson S.M.
      Minimal penetration of lipopolysaccharide across the murine blood-brain barrier.
      ,
      • Threlkeld S.W.
      • Lynch J.L.
      • Lynch K.M.
      • Sadowska G.B.
      • Banks W.A.
      • Stonestreet B.S.
      Ovine proinflammatory cytokines cross the murine blood-brain barrier by a common saturable transport mechanism.
      ].

      7. The national collaborative perinatal project

      Among the samples of newborn brains assessed specifically for histologic indicators of damage, only one had reasonably high-quality data collected routinely about pregnancy exposures and characteristics. The National Collaborative Perinatal Project (NCPP) recruited women from 12 clinical sites at their initial antenatal visit during the years 1959–1966 [
      • Niswander K.G.M.
      Women and their pregnancies.
      ]. The NCPP is the largest data file with potential clinical risk factor variables collected before a neonate's death, and contains more than 1200 clinical, maternal, delivery, and perinatal variables [
      • Leviton A.
      Epidemiologic methods.
      ], and collected 1100 neonatal brains. A major limitation of the NCPP sample's relevance to today is that sixty years ago, when the NCPP deliveries occurred (1959–1966), infants born much before term were highly unlikely to have been intensively resuscitated.

      8. A fetal brain inflammatory response circa 1960

      The odds ratios in Table 2 below are taken from the first table of each of chapters 19–24 in the report of the NCPP sample and are limited to infants who died within the first 30 days after birth [
      • Gilles F.H.
      • Leviton A.
      • Dooling E.C.
      Developing human brain: growth and epidemiologic neuropathology.
      ].
      Table 2Statistically-significant increased odds ratios of histologic abnormalities in the newborn's brain associated with indicators or correlates of urinary tract in infection in the gravida during this pregnancy compared to the risk in the offspring of gravida's who did not have the with indicator or correlate of urinary tract in infection. Data in this table are from Table 1 of each of chapters 19–24 [
      • Gilles F.H.
      • Leviton A.
      • Dooling E.C.
      Developing human brain: growth and epidemiologic neuropathology.
      ], and is limited to children who died within the first 30 days after birth.
      Histologic characteristic(s) of offspring's brain
      Pregnancy characteristicHypertrophic astrocytes (HA)Amphophilic globules (GL)Necrotic fociAcutely damaged glia (ADG)HA + GLHA + ADG
      CVA tenderness5.13.55.77.1
      Positive urine culture3.02.28.6
      Pyuria2.52.7
      Urinary tract infection (UTI)2.13.1
      UTI and temperature ≥ 100.4 F5.413.66.1
      Temperature ≥ 100.4 F5.7
      Newborn characteristic
      Septicemia3.59.24.12.76.2
      Because details collected about inflammation at the time of delivery in the NCPP sample from 50 years ago do not coincide with what would be collected today, we focused on the data collected routinely about two infections: cystitis and pyelonephritis. Costovertebral angle tenderness (an indicator of upper urinary tract infection), and febrile urinary tract infection (an indicator of a maternal systemic inflammatory response [
      • Stalenhoef J.E.
      • van Nieuwkoop C.
      • Wilson D.C.
      • van der Starre W.E.
      • van der Reijden T.J.K.
      • Delfos N.M.
      • et al.
      Procalcitonin, mid-regional proadrenomedullin and C-reactive protein in predicting treatment outcome in community-acquired febrile urinary tract infection.
      ]) tended to have the highest odds ratios for each histologic characteristic except necrotic foci. The failure to find any pregnancy associations for necrotic foci might represent the relatively small number of children who died during the first month after birth and had necrotic foci in their white matter (N = 31).
      Information was not available about when during the pregnancy the urinary tract infection occurred. Nevertheless, we offer this table as the best evidence that can be provided about the fetal brain response to maternal systemic inflammation before the current state of obstetrics.
      One possibility is that the inflammation is sufficient to contribute to brain damage. Another possibility is that this inflammatory exposure did not result in the histologic changes seen, but rather “primed the pump” by sensitizing the fetus/newborn to a subsequent brain damaging stimulus [
      • Eklind S.
      • Mallard C.
      • Arvidsson P.
      • Hagberg H.
      Lipopolysaccharide induces both a primary and a secondary phase of sensitization in the developing rat brain.
      ]. Either way, children whose mother had a UTI during pregnancy are at higher risk than others of a low intelligence quotient [
      • Broman S.H.N.P.
      • Shaughnessy P.
      • Kennedy W.
      Retardation in young children. A developmental study of cognitive deficit.
      ], a reading limitation (defined as a score one or more standard deviations below the expected mean on the Word Reading component of the WIAT-III) among normal intelligence children [
      • Leviton A.
      • Joseph R.M.
      • Allred E.N.
      • Fichorova R.N.
      • O'Shea T.M.
      • Kuban K.K.C.
      • et al.
      The risk of neurodevelopmental disorders at age 10 years associated with blood concentrations of interleukins 4 and 10 during the first postnatal month of children born extremely preterm.
      ], and cerebral palsy [
      • Polivka B.J.
      • Nickel J.T.
      • Wilkins 3rd, J.R.
      Urinary tract infection during pregnancy: a risk factor for cerebral palsy?.
      ,
      • Miller J.E.
      • Pedersen L.H.
      • Streja E.
      • Bech B.H.
      • Yeargin-Allsopp M.
      • Van Naarden Braun K.
      • et al.
      Maternal infections during pregnancy and cerebral palsy: a population-based cohort study.
      ,
      • Bear J.J.
      • Wu Y.W.
      Maternal infections during pregnancy and cerebral palsy in the child.
      ].
      We offer these findings about gestational urinary tract infection as an example of a fetal inflammatory stimulus that could be identified years before the modern era of perinatology and neonatology. During those years, telencephalic white matter damage was recognized as inflammatory, largely because bacteremia in human newborns [
      • Leviton A.
      • Gilles F.
      • Neff R.
      • Yaney P.
      Multivariate analysis of risk of perinatal telencephalic leucoencephalopathy.
      ], and systemic endotoxin (lipopolysaccharide) in kittens, rabbits, and monkeys [
      • Gilles F.H.
      • Averill Jr., D.R.
      • Kerr C.S.
      Neonatal endotoxin encephalopathy.
      ,
      • Gilles F.H.
      • Leviton A.
      • Kerr C.S.
      Endotoxin leukoencephalopathy in the telencephalon of the newborn kitten.
      ] were frequently followed by white matter damage. Also, decades before the modern era of perinatology and neonatology, fever was recognized as an effect of circulating pyrogens on the brain (specifically, the preoptic area of the anterior hypothalamus) [
      • Cooper K.E.
      Temperature regulation and the hypothalamus.
      ]. Only in the modern era, have some of those pyrogens been identified as cytokines originating in the amniotic cavity and capable of stimulating the fetal inflammatory response syndrome [
      • Lee S.E.
      • Romero R.
      • Jung H.
      • Park C.W.
      • Park J.S.
      • Yoon B.H.
      The intensity of the fetal inflammatory response in intraamniotic inflammation with and without microbial invasion of the amniotic cavity.
      ,
      • Jung E.Y.
      • Park K.H.
      • Han B.R.
      • Cho S.H.
      • Yoo H.N.
      • Lee J.
      Amniotic fluid infection, cytokine levels, and mortality and adverse pulmonary, intestinal, and neurologic outcomes in infants at 32 Weeks' gestation or less.
      ].

      9. Postnatal correlates of histologic abnormalities of the OFFSPRING’S brain

      In the NCPP sample, neonatal bacteremia was associated with increased risk of all three histologic abnormalities, hypertrophic astrocytes, acutely damaged glia, necrotic foci, and the combination of hypertrophic astrocytes and acutely damaged glia. None of the many NCPP clinical variables or those that can be viewed as indicators or close correlates of hypoxemia or hypotension (antenatally or postnatally) were associated with an increased risk of any of the four histologic abnormalities, or with their clusters [
      • Leviton A.
      • Gilles F.
      Etiologic relationships among the perinatal telencephalic leucoencephalopathies.
      ]. Therefore, hypoxia and hypotension are unlikely contributors to neonatal white matter abnormalities in humans. This inference has since been supported by studies of extremely preterm newborns who survived [
      • Gilles F.
      • Gressens P.
      • Dammann O.
      • Leviton A.
      Hypoxia-ischemia is not an antecedent of most preterm brain damage: the illusion of validity.
      ].
      Assuming the four histologic markers of white matter damage are signs of the fetal inflammatory response syndrome, other risk factors of each of these markers were apparent in the NCPP material. For instance, pre-pregnancy maternal obesity and excessive pregnancy weight gain are associated with increased risk of the most frequent of neonatal white matter abnormalities, namely, hypertrophic astrocytes [
      • Leviton A.
      • Gilles F.
      The epidemiology of the perinatal telencephalic leucoencephalopathy characterized by hypertrophic astrocytes.
      ]. White matter hypertrophic astrocytes reflect white matter damage. Prepregnancy high body mass index is also associated with chronic villitis and lower placental weight [
      • Brouwers L.
      • Franx A.
      • Vogelvang T.E.
      • Houben M.L.
      • van Rijn B.B.
      • Nikkels P.G.
      Association of maternal prepregnancy body mass index with placental histopathological characteristics in uncomplicated term pregnancies.
      ] and with placental amplification of trophoblastic Toll-like receptor 4 signaling pathways [
      • Yang X.
      • Li M.
      • Haghiac M.
      • Catalano P.M.
      • O'Tierney-Ginn P.
      • Hauguel-de Mouzon S.
      Causal relationship between obesity-related traits and TLR4-driven responses at the maternal-fetal interface.
      ]. Increasing body mass index among pregnant women is associated with significant increases in circulating IL-6,TNF, IL-1, and CRP except for the end of gestation [
      • Friis C.M.
      • Paasche Roland M.C.
      • Godang K.
      • Ueland T.
      • Tanbo T.
      • Bollerslev J.
      • et al.
      Adiposity-related inflammation: effects of pregnancy.
      ]. Maternal obesity is also associated in offspring with increased risk of a large number of neurodevelopmental and psychiatric abnormalities including schizophrenia (summarized in Ref. [
      • Edlow A.G.
      Maternal obesity and neurodevelopmental and psychiatric disorders in offspring.
      ]). Schizophrenia has widely spread white matter microstructural abnormalities [
      • Kelly S.
      • Jahanshad N.
      • Zalesky A.
      • Kochunov P.
      • Agartz I.
      • Alloza C.
      • et al.
      Widespread white matter microstructural differences in schizophrenia across 4322 individuals: results from the ENIGMA Schizophrenia DTI Working Group.
      ,
      • Koshiyama D.
      • Fukunaga M.
      • Okada N.
      • Morita K.
      • Nemoto K.
      • Usui K.
      • et al.
      White matter microstructural alterations across four major psychiatric disorders: mega-analysis study in 2937 individuals.
      ], which might reflect neonatal white matter damage.

      10. Conclusions

      We have reviewed our early data-driven contributions to the literature about telencephalic white matter damage including classification, biology, and epidemiology. We have found that none of the four histologic markers of white matter damage, including focal necroses, are associated with clinical markers of hypoxemia or hypotension, but all four are instead associated with clinical markers of fetal or perinatal inflammation and that perinatal white matter abnormalities are likely part of the fetal and neonatal inflammatory response syndrome.
      • Prevention of urinary tract and perhaps other gestational bacterial infections may possibly reduce the occurrence of markers of cerebral white matter damage.
      • Very early treatment of urinary tract and other gestational bacterial infections has not yet been established to reduce the occurrence of markers of cerebral white matter damage. Clinical trials need to assess such management.
      • The evaluation reported here does not address the potential contribution of extremely or very preterm birth or the phenomena that contribute such early delivery.
      • Does antibiotic prophylaxis given to pregnant women at risk for recurrent urinary tract infection reduce their offspring's risk of developmental (cognitive and other) limitations?
      • Determine whether the observations in preterm infants are the same or different in term newborns

      Acknowledgments

      The authors acknowledge support from the National Institute of Neurological Disorders and Stroke ( 5U01NS040069-05 ; 2R01NS040069-06A2 ), The National Eye Institute ( 1-R01- EY021820-01 ), and the National Institute of Child Health and Human Development ( 5P30HD018655-28 ) (AL).

      References

        • Banker B.Q.
        • Larroche J.C.
        Periventricular leukomalacia of infancy. A form of neonatal anoxic encephalopathy.
        Arch Neurol. 1962; 7: 386-410
        • Schmorl C.G.
        Zur kenntniss des ikterus neonatorum, inbesondere der dabei auftretenden gehirnveranderungen.
        Verhandl deutsch Path Gesellsch. 1904; 6: 109-115
        • Parrot M.J.
        Etude sur la stèatose interstitielle diffuse de l'encèphale chez le nouveau-nè.
        Arch Physiol Norm Patholol (Paris). 1868; 1 (622-42, 706-15): 530-550
        • Schwartz P.
        Birth Injuries of the Newborn, Morphology, Pathogenesis, Clinical Pathology and Prevention.
        Hafner, New York1961
        • Clark D.B.
        • Anderson G.W.
        Correlation of complications of labor with lesions in the brains of infants.
        J Neuropathol Exp Neurol. 1961; 20: 275-278
        • Gluszcz A.
        On the periventricular septic necroses of the brain in premature infants.
        in: Jacob H. International congress of neuropathology. Theime Verlag, Munich1961
        • Perrin E.V.
        • Landing B.H.
        The Schmorl lesion" in jaundiced infected infants.
        Am J Dis Child. 1962; 104: 551
        • Leviton A.
        • Gilles F.H.
        The epidemiology of the perinatal telencephalic leucoencephalopathy characterized by focal necroses.
        in: Gilles F.H. Leviton A. Dooling E.C. The developing human brain: growth and epidemiologic neuropathology. Wright PSG, Boston1983: 270-277
        • Leviton A.
        • Gilles F.H.
        Acquired perinatal leukoencephalopathy.
        Ann Neurol. 1984; 16: 1-8
        • Dammann O.
        • Leviton A.
        Perinatal brain damage causation.
        Dev Neurosci. 2007; 29: 280-288
        • Peirce C.S.
        The fixation of Belief.
        Popular Sci Mon. 1877; 12: 1-15
        • Gilles F.
        • Murphy S.
        Perinatal telencephalic leucoencephalopathy.
        J Neurol Neurosurg Psychiatr. 1969; 32: 404-413
        • Leviton A.
        • Gilles F.H.
        An epidemiologic study of perinatal telencephalic leucoencephalopathy in an autopsy population.
        J Neurol Sci. 1973; 18: 53-66
        • Leviton A.
        • Gilles F.
        Etiologic relationships among the perinatal telencephalic leucoencephalopathies.
        in: FH Gilles A.L. Dooling E.C. The developing human brain: growth and epidemiologic neuropathology. John Wright/PSG, Boston1983: 304-315
        • Leviton A.
        • Gilles F.H.
        Classification of the perinatal telencephalic leucoencephalopathies.
        in: Gilles F.H. Leviton A. Dooling E.C. The developing human brain: growth and epidemiologic neuropathology. Wright PSG, Boston1983: 244-250
        • Bennet L.
        • Roelfsema V.
        • Dean J.M.
        • Wassink G.
        • Power G.G.
        • Jensen E.C.
        • et al.
        Regulation of cytochrome oxidase redox state during umbilical cord occlusion in preterm fetal sheep.
        Am J Physiol Regul Integr Comp Physiol. 2007; 292: R1569-R1576
        • Gunn A.J.
        • Quaedackers J.S.
        • Guan J.
        • Heineman E.
        • Bennet L.
        The premature fetus: not as defenseless as we thought, but still paradoxically vulnerable?.
        Dev Neurosci. 2001; 23: 175-179
        • Gilles F.
        • Gressens P.
        • Dammann O.
        • Leviton A.
        Hypoxia-ischemia is not an antecedent of most preterm brain damage: the illusion of validity.
        Dev Med Child Neurol. 2018; 60: 120-125
        • Gilles F.H.
        Neural damage: inconstancy during gestation.
        in: Gilles F. Leviton A. Dooling E. The developing human brain: growth and epidemiologic neuropathology. Wright-PSG, Boston1983: 227-243
        • McLennan J.E.
        • Gilles F.H.
        • Neff R.
        A model of growth of the human fetal brain.
        in: Gilles F.H. Leviton A. Dooling E.C. The developing human brain: growth and epidemiologic neuropathology. Wright PSG, Boston1983: 43-58
        • Kuban K.C.K.
        • Gilles F.H.
        Human telencephalic angiogenesis.
        Ann Neurol. 1985; 17: 539-548
        • Gilles F.
        • Nelson M.J.
        The developing human brain; Growth and Adversities.
        Mac Keith Press, London2012
        • Dąmbska M.
        Encephalic necrosis and inflammatory reaction in fetuses and newborns.
        Pol Med J. 1968; 7: 404-434
        • Ginsberg Y.
        • Khatib N.
        • Weiner Z.
        • Beloosesky R.
        Maternal inflammation, fetal brain implications and suggested neuroprotection: a summary of 10 Years of research in animal models.
        Rambam Maimonides Med J. 2017; 8
        • Cajal R.
        Degeneration and regeneration of the nervous system.
        Hafner, New York1959: 1902
        • Gilles F.H.
        • Averill Jr., D.R.
        • Kerr C.S.
        Neonatal endotoxin encephalopathy.
        Ann Neurol. 1977; 2: 49-56
        • Ellison D.
        • Love S.
        Neuropathology.
        third ed. Mosby/Elsevier, Edinburgh2014
        • Leviton A.
        • Gilles F.H.
        Morphologic abnormalities in human infant cerebral white matter related to gestational and postnatal age.
        Pediatr Res. 1974; 8: 718-720
        • Kanski R.
        • van Strien M.E.
        • van Tijn P.
        • Hol E.M.
        A star is born: new insights into the mechanism of astrogenesis.
        Cell Mol Life Sci. 2014; 71: 433-447
        • Liddelow S.A.
        • Guttenplan K.A.
        • Clarke L.E.
        • Bennett F.C.
        • Bohlen C.J.
        • Schirmer L.
        • et al.
        Neurotoxic reactive astrocytes are induced by activated microglia.
        Nature. 2017; 541: 481-487
        • Zamanian J.L.
        • Xu L.
        • Foo L.C.
        • Nouri N.
        • Zhou L.
        • Giffard R.G.
        • et al.
        Genomic analysis of reactive astrogliosis.
        J Neurosci. 2012; 32: 6391-6410
        • Liddelow S.A.
        • Barres B.A.
        Reactive astrocytes: production, function, and therapeutic potential.
        Immunity. 2017; 46: 957-967
        • Stolp H.B.
        • Dziegielewska K.M.
        • Ek C.J.
        • Habgood M.D.
        • Lane M.A.
        • Potter A.M.
        • et al.
        Breakdown of the blood-brain barrier to proteins in white matter of the developing brain following systemic inflammation.
        Cell Tissue Res. 2005; 320: 369-378
        • McCartney E.
        • Squier W.
        Patterns and pathways of calcification in the developing brain.
        Dev Med Child Neurol. 2014; 56: 1009-1015
        • Schwartz P.
        Die traumatischen schadigungen des zentralnerversystems durch die geburt.
        Anat Untersuchungen Ergebn Inn Med u Kinderh. 1927; 31: 165-372
        • Leech R.W.
        • Alvord Jr., E.C.
        Morphologic variations in periventricular leukomalacia.
        Am J Pathol. 1974; 74: 591-602
        • Shuman R.M.
        • Selednik L.J.
        Periventricular leukomalacia. A one-year autopsy study.
        Arch Neurol. 1980; 37: 231-235
        • Nelson M.D.
        • Gonzalez-Gomez I.
        • Gilles F.H.
        The search for human telencephalic ventriculofugal arteries.
        Am J Neuroradiol. 1991; 12: 215-222
        • Gilles F.H.
        • Nelson M.D.J.
        • Gonzalez-Gomez I.
        Human telencephalic angiogenesis: an update.
        in: Fujisawa K. Morimatsu Y. Development and involution of neurons. Japan Scientific Societies Press, Tokyo1992: 31-41
        • Vandenabeele P.
        • Galluzzi L.
        • Vanden Berghe T.
        • Kroemer G.
        Molecular mechanisms of necroptosis: an ordered cellular explosion.
        Nat Rev Mol Cell Biol. 2010; 11: 700-714
        • Denes A.
        • Lopez-Castejon G.
        • Brough D.
        Caspase-1: is IL-1 just the tip of the ICEberg?.
        Cell Death Dis. 2012; 3: 5-9
        • Leviton A.
        Single-cause attribution.
        Dev Med Child Neurol. 1987; 29: 805-807
        • DeReuck J.
        The human periventricular arterial blood supply and the anatomy of cerebral infarctions.
        Eur Neurol. 1971; 5: 321-334
        • Chattha A.S.
        • Richardson Jr., E.P.
        Cerebral white-matter hypoplasia.
        Arch Neurol. 1977; 34: 137-141
        • Takashima S.
        • Tanaka K.
        Development of cerebrovascular architecture and its relationship to periventricular leukomalacia.
        Arch Neurol. 1978; 35: 11
        • Volpe J.J.
        Neurology of the newborn.
        2 ed. W B Saunders Cop, Philadelphia1987
        • Bär T.
        The vascular system of the cerebral cortex (Review).
        Adv Anat Embryol Cell Biol. 1980; 59: 1-62
        • Nelson Jr., M.D.
        • Gonzalez-Gomez I.
        • Gilles F.H.
        • Dyke Award
        The search for human telencephalic ventriculofugal arteries.
        AJNR Am J Neuroradiol. 1991; 12: 215-222
        • Friede R.L.
        Developmental neuropathology.
        Springer, Vienna1975
        • Sjobring U.
        • Ringdahl U.
        • Ruggeri Z.M.
        Induction of platelet thrombi by bacteria and antibodies.
        Blood. 2002; 100: 4470-4477
        • Feng S.Y.
        • Phillips D.J.
        • Stockx E.M.
        • Yu V.Y.
        • Walker A.M.
        Endotoxin has acute and chronic effects on the cerebral circulation of fetal sheep.
        Am J Physiol Regul Integr Comp Physiol. 2009; 296: R640-R650
        • van der Poll T.
        • Buller H.R.
        • ten Cate H.
        • Wortel C.H.
        • Bauer K.A.
        • van Deventer S.J.
        • et al.
        Activation of coagulation after administration of tumor necrosis factor to normal subjects.
        N Engl J Med. 1990; 322: 1622-1627
        • Leviton A.
        • Fichorova R.
        • Yamamoto Y.
        • Allred E.N.
        • Dammann O.
        • Hecht J.
        • et al.
        Inflammation-related proteins in the blood of extremely low gestational age newborns. The contribution of inflammation to the appearance of developmental regulation.
        Cytokine. 2011; 53: 66-73
        • Yanni D.
        • Korzeniewski S.J.
        • Allred E.N.
        • Fichorova R.N.
        • O'Shea T.M.
        • Kuban K.
        • et al.
        Both antenatal and postnatal inflammation contribute information about the risk of brain damage in extremely preterm newborns.
        Pediatr Res. 2017; 82: 691-696
        • Leviton A.
        • Joseph R.M.
        • Allred E.N.
        • Fichorova R.N.
        • O'Shea T.M.
        • Kuban K.K.C.
        • et al.
        The risk of neurodevelopmental disorders at age 10 years associated with blood concentrations of interleukins 4 and 10 during the first postnatal month of children born extremely preterm.
        Cytokine. 2018; 110: 181-188
        • Gussenhoven R.
        • Westerlaken R.J.J.
        • Ophelders D.
        • Jobe A.H.
        • Kemp M.W.
        • Kallapur S.G.
        • et al.
        Chorioamnionitis, neuroinflammation, and injury: timing is key in the preterm ovine fetus.
        J Neuroinflammation. 2018; 15: 113
        • Varol C.
        • Mildner A.
        • Jung S.
        Macrophages: development and tissue specialization.
        Annu Rev Immunol. 2015; 33: 643-675
        • Dean J.M.
        • Shi Z.
        • Fleiss B.
        • Gunn K.C.
        • Groenendaal F.
        • van Bel F.
        • et al.
        A critical review of models of perinatal infection.
        Dev Neurosci. 2015; 37: 289-304
        • Mallard C.
        • Tremblay M.E.
        • Vexler Z.S.
        Microglia and neonatal brain injury.
        Neuroscience. 2019; 405: 68-76
        • Leviton A.
        • Gilles F.H.
        Clustering of the morphological components of perinatal telencephalic leucoencephalopathy.
        J Neurol Neurosurg Psychiatr. 1971; 34: 642-645
        • Leviton A.
        • Gilles F.
        • Neff R.
        • Yaney P.
        Multivariate analysis of risk of perinatal telencephalic leucoencephalopathy.
        Am J Epidemiol. 1976; 104: 621-626
        • Leviton A.
        • Gilles F.H.
        • Vawter G.F.
        The thymus in infants with perinatal telencephalic leukoencephalopathy.
        Arch Neurol. 1978; 35: 377-381
        • Gilles F.H.
        • Leviton A.
        • Kerr C.S.
        Endotoxin leukoencephalopathy in the telencephalon of the newborn kitten.
        J Neurol Sci. 1976; 27: 183-191
        • Gilles F.H.
        • Averill Jr., D.R.
        • Kerr C.S.
        Changes in neonatally induced cerebral lesions with advancing age.
        J Neuropathol Exp Neurol. 1977; 36: 666-679
        • Young R.S.
        • Yagel S.K.
        • Towfighi J.
        Systemic and neuropathologic effects of E. coli endotoxin in neonatal dogs.
        Pediatr Res. 1983; 17: 349-353
        • Cai Z.
        • Pan Z.L.
        • Pang Y.
        • Evans O.B.
        • Rhodes P.G.
        Cytokine induction in fetal rat brains and brain injury in neonatal rats after maternal lipopolysaccharide administration.
        Pediatr Res. 2000; 47: 64-72
        • Pang Y.
        • Campbell L.
        • Zheng B.
        • Fan L.
        • Cai Z.
        • Rhodes P.
        Lipopolysaccharide-activated microglia induce death of oligodendrocyte progenitor cells and impede their development.
        Neuroscience. 2010; 166: 464-475
        • Rousset C.I.
        • Chalon S.
        • Cantagrel S.
        • Bodard S.
        • Andres C.
        • Gressens P.
        • et al.
        Maternal exposure to LPS induces hypomyelination in the internal capsule and programmed cell death in the deep gray matter in newborn rats.
        Pediatr Res. 2006; 59: 428-433
        • Wang X.
        • Rousset C.I.
        • Hagberg H.
        • Mallard C.
        Lipopolysaccharide-induced inflammation and perinatal brain injury.
        Semin Fetal Neonatal Med. 2006; 11: 343-353
        • Cardoso F.L.
        • Herz J.
        • Fernandes A.
        • Rocha J.
        • Sepodes B.
        • Brito M.A.
        • et al.
        Systemic inflammation in early neonatal mice induces transient and lasting neurodegenerative effects.
        J Neuroinflammation. 2015; 12: 82
        • Mallard C.
        • Welin A.K.
        • Peebles D.
        • Hagberg H.
        • Kjellmer I.
        White matter injury following systemic endotoxemia or asphyxia in the fetal sheep.
        Neurochem Res. 2003; 28: 215-223
        • Banks W.A.
        • Robinson S.M.
        Minimal penetration of lipopolysaccharide across the murine blood-brain barrier.
        Brain Behav Immun. 2010; 24: 102-109
        • Threlkeld S.W.
        • Lynch J.L.
        • Lynch K.M.
        • Sadowska G.B.
        • Banks W.A.
        • Stonestreet B.S.
        Ovine proinflammatory cytokines cross the murine blood-brain barrier by a common saturable transport mechanism.
        Neuroimmunomodulation. 2010; 17: 405-410
        • Niswander K.G.M.
        Women and their pregnancies.
        W.B. Saunders Company, Philadelphia1972
        • Leviton A.
        Epidemiologic methods.
        in: Gilles F. Leviton A. Dooling E. The developing human brain: growth and epidemiologic neuropathology. John Wright-PSG, Littleton, Mass1983: 10-16
        • Gilles F.H.
        • Leviton A.
        • Dooling E.C.
        Developing human brain: growth and epidemiologic neuropathology.
        John Wright-PSG Publishing Co, Boston1983
        • Stalenhoef J.E.
        • van Nieuwkoop C.
        • Wilson D.C.
        • van der Starre W.E.
        • van der Reijden T.J.K.
        • Delfos N.M.
        • et al.
        Procalcitonin, mid-regional proadrenomedullin and C-reactive protein in predicting treatment outcome in community-acquired febrile urinary tract infection.
        BMC Infect Dis. 2019; 19: 161
        • Eklind S.
        • Mallard C.
        • Arvidsson P.
        • Hagberg H.
        Lipopolysaccharide induces both a primary and a secondary phase of sensitization in the developing rat brain.
        Pediatr Res. 2005; 58: 112-116
        • Broman S.H.N.P.
        • Shaughnessy P.
        • Kennedy W.
        Retardation in young children. A developmental study of cognitive deficit.
        Lawrence Erlbaum Associates, Hillsdale, NJ1987
        • Polivka B.J.
        • Nickel J.T.
        • Wilkins 3rd, J.R.
        Urinary tract infection during pregnancy: a risk factor for cerebral palsy?.
        J Obstet Gynecol Neonatal Nurs. 1997; 26: 405-413
        • Miller J.E.
        • Pedersen L.H.
        • Streja E.
        • Bech B.H.
        • Yeargin-Allsopp M.
        • Van Naarden Braun K.
        • et al.
        Maternal infections during pregnancy and cerebral palsy: a population-based cohort study.
        Paediatr Perinat Epidemiol. 2013; 27: 542-552
        • Bear J.J.
        • Wu Y.W.
        Maternal infections during pregnancy and cerebral palsy in the child.
        Pediatr Neurol. 2016; 57: 74-79
        • Cooper K.E.
        Temperature regulation and the hypothalamus.
        Br Med Bull. 1966; 22: 238-242
        • Lee S.E.
        • Romero R.
        • Jung H.
        • Park C.W.
        • Park J.S.
        • Yoon B.H.
        The intensity of the fetal inflammatory response in intraamniotic inflammation with and without microbial invasion of the amniotic cavity.
        Am J Obstet Gynecol. 2007; 197 (294 e1-6)
        • Jung E.Y.
        • Park K.H.
        • Han B.R.
        • Cho S.H.
        • Yoo H.N.
        • Lee J.
        Amniotic fluid infection, cytokine levels, and mortality and adverse pulmonary, intestinal, and neurologic outcomes in infants at 32 Weeks' gestation or less.
        J Kor Med Sci. 2017; 32: 480-487
        • Leviton A.
        • Gilles F.
        The epidemiology of the perinatal telencephalic leucoencephalopathy characterized by hypertrophic astrocytes.
        in: Gilles F. Leviton A. Dooling E. The developing human brain: growth and epidemiologic neuropathology. Wright PSG, Boston1983: 251-261
        • Brouwers L.
        • Franx A.
        • Vogelvang T.E.
        • Houben M.L.
        • van Rijn B.B.
        • Nikkels P.G.
        Association of maternal prepregnancy body mass index with placental histopathological characteristics in uncomplicated term pregnancies.
        Pediatr Dev Pathol. 2019; 22: 45-52
        • Yang X.
        • Li M.
        • Haghiac M.
        • Catalano P.M.
        • O'Tierney-Ginn P.
        • Hauguel-de Mouzon S.
        Causal relationship between obesity-related traits and TLR4-driven responses at the maternal-fetal interface.
        Diabetologia. 2016; 59: 2459-2466
        • Friis C.M.
        • Paasche Roland M.C.
        • Godang K.
        • Ueland T.
        • Tanbo T.
        • Bollerslev J.
        • et al.
        Adiposity-related inflammation: effects of pregnancy.
        Obesity. 2013; 21: E124-E130
        • Edlow A.G.
        Maternal obesity and neurodevelopmental and psychiatric disorders in offspring.
        Prenat Diagn. 2017; 37: 95-110
        • Kelly S.
        • Jahanshad N.
        • Zalesky A.
        • Kochunov P.
        • Agartz I.
        • Alloza C.
        • et al.
        Widespread white matter microstructural differences in schizophrenia across 4322 individuals: results from the ENIGMA Schizophrenia DTI Working Group.
        Mol Psychiatr. 2018; 23: 1261-1269
        • Koshiyama D.
        • Fukunaga M.
        • Okada N.
        • Morita K.
        • Nemoto K.
        • Usui K.
        • et al.
        White matter microstructural alterations across four major psychiatric disorders: mega-analysis study in 2937 individuals.
        Mol Psychiatr. 2019;