White Matter Disease in the Premature Infant

1
White Matter Disease in the Premature Infant
Jeffrey M
Perlman MB Professor of
Pediatrics Weill Medical
College Cornell Medical
Center New York
2
BACKGROUND
1) Periventricular White Matter injury (WMI)
represents the most significant problem
contributing to both neonatal mortality as
well as long term neurodevelopmental
deficits. 2) Injury to white matter maybe
unilateral, bilateral and develop in the
presence or absence of hemorrhage. 3) WMI in
the presence of hemorrhage is often unilateral
and referred to as Grade IV IVH, and in the
absence of blood is referred to as
Periventricular Leucomalacia (PVL).
3
Outline

• Intraventricular Hemorrhage
• Pathogenesis
• Severe IVH
• Preventative
  strategies
• Periventricular Leucomalacia
• Definition
• Pathogenesis
• Preventative
  strategies
• Outcome

4
Periventricular-Intraventricular Hemorrhage
5
Background
1) Periventricular-Intraventricular hemorrhage
remains the most common serious neurologic
lesion of the premature infant. 2) The overall
incidence appears to be declining. 3) However
severe IVH remains a significant problem in
the very low birth weight infant lt1000gm.
6
Frequency of severe intraventricular hemorrhage
(IVH) in infants of birthweight 500-1500g
(1991-1992)

• 501-750g 751-1000g 1001-1250g 1251-1500g
• Grade of IVH (n1002) (n1084)
  (n1053) (n1299)
• Grade III IVH 13 6 5
  2
• Grade III IVH IPE 13 6 3
  1
• Total IVH 26 12 8 3
• IPE - Intraparenchymal echodensity - often
  referred to as Grade IV IVH.

(Adapted from Lemons et al Pediatrics 107 e12001)
7
Neuropathology
8
Blood in Lateral Ventricle
Germinal Matrix
Blood in Third Ventricle
Blood in Fourth Ventricle
9
Factors Contributing to PV-IVH
1) Vulnerable capillaries 2) Pressure Passive
Cerebral Circulation 3) Intravascular
pertubations
10
Pressure Passive Circulation
Refers to a state where cerebral blood flow
changes directly reflect changes in systemic
blood pressure This increases the potential for
injury during episodes of systemic hypertension
and/or systemic hypotension followed by increases
in systemic blood pressure
11
(No Transcript)
12
Beat to beat fluctuations in CBFV (top panel) and
simultaneous fluctuations in Arterial Blood
Pressure (bottom panel).
Stable pattern of Cerebral Blood Flow Velocity
(top) and Arterial Blood Pressure (bottom)
Relationship of Fluctuations in CBFV to
Development of IVH
CBFV Pattern IVH
No IVH
Fluctuations 21
2 Stable
7 20
3 developed IVH following a Pneumothorax
Perlman et al NEJM 1983
13
Schema of Factors Important In Pathogenesis Of
Periventricular-Intraventricular Hemorrhage

• Vascular Factors
• Involving capillary bed
• Apparent border zone region
• Extra-Vascular Factors ? Germinal Matrix
  Capillary ? Intravascular Factors
  
  Factors
• Poor vascular support
  Fluctuating CBF
• ? Fibrinolytic activity ? CBF
• ? Tissue pressure ? CBF
• Rupture ?
  Venous pressure
• 
  
• 
  
• 
  Germinal Matrix
• Hemorrhage
• IVH Periventricular
• Hemorrhagic Infarction

14
Cranial Ultrasound Scan - coronal view.
IPE
Note the Large Ipsilateral Echogenic Area in Left
Fronto-Parietal White Matter
15
Venous Drainage of the Deep White Matter
16
Pathogenesis of Periventricular White Matter
Injury (WMI)
1) Direct relationship to PV-IVH a) The
WMI is always noted concurrent with or following
a large PV-IVH. b) Rarely
is WMI observed prior to PV-IVH c) The WMI
is noted ipsilateral to the larger PV-IVH when
the lesion is
bilateral. 2) Evolution of WMI de novo a)
PV-IVH and WMI occur concurrently-both are
border zone regions which ? the
risk for hypoperfusion during hypotension-
hemorrhage then occurs as a reperfusion
injury. b) ? Hypoxanthine and Uric Acid
levels have been noted on day
one in infants who subsequently develop WMI


Guzetta et al Pediatrics 198678945, Perlman et
al Pediatrics 199391474 Russel et al Arch Dis
Childh 199267338, Perlman J Pediatr 1998132486
17
Prevention of PV-IVH and theAssociated White
Matter Injury

• Perinatal Interventions
• Labor and Delivery
• Postnatal Interventions

18
Perinatal and Postnatal interventions to prevent
PV-IVH
Intervention
Effect on IVH
Positive None Negative
Prenatal Glucocorticoids

Phenobarbital
Vitamin K
? Sympathomimetics

Postnatal Indomethacin
Phenobarbital
Vitamin E
? Ethamsylate
?
19
ANTENATAL STEROIDS PV-IVH
1) Data indicate a 50 ? in the occurrence of
severe IVH in infants exposed to antenatal
steroids. 2) This appears to be a dose dependent
effect. 3) Steroids may reduce IVH by
a) Enhancing
maturation of the germinal matrix b)
Altering energy substrates during episodes of
?cerebral perfusion, i.e. reducing basal
metabolism followed by an?in glucose
utilization during HI. c) Providing more
stable systemic BP d) A lesser requirement
for volume expanders.

Crowley et al Br J Obstet Gynaecol 19909711,
Wright L Am J Obstet Gynecol 1995173263 Garland
Jet al J Pediatr 1995126272
20
(No Transcript)
21
Labor and Delivery Room Management
1) There is conflicting data regarding the route
of delivery and subsequent PV-IVH. 2) Most
data are retrospective and were analysed
prior to the more frequent use of
glucocorticoids. 3) Several studies indicate ?
risk for IVH with ? duration of active phase
of labor, and ? risk for infants delivered
via CS prior to the active phase of labor. 4)
Recent data point to an important role of
placental inflammation, and in particular
fetal vasculitis which supercedes the
influence of route of delivery.
Hansen and Leviton Am J Obstet Gynec 1819971999
22
Postnatal Strategies
Postnatal strategies to prevent PV-IVH should
incorporate several important observations 1) The
majority of cases of severe IVH occur in the
smallest infants i.e. lt 1000 grams 2) The
condition of the infant at delivery in part
influenced by perinatal events significantly
impacts upon the development of severe
IVH. 3) Factors associated with increased risk
include ? BW, ? GA male sex, intubation, RDS.
23
Factors Modulating Severe PV-IVH

• Predictor P Value Odds Ratio
  Confidence
• Estimate Interval
• Steroids 0.017 0.167 0.03, 0.73
• PIH 0.003 0.108 0.01, 0.83
• Intubation 0.027 4.244 1.17, 15.31
• C/Section 0.625 0.183 0.40, 1.71
• GA (per week decrease) 0.0001 1.587 1.30,
  1.92

Chorioamnionitis/funisitis was not included in
this model
Perlman Clin Perinatol 29745,2002
24
Postnatal Interventions
1) Postnatal administration of medications have
included Phenobarbital, Vitamin E,
Ethamsylate and Indomethacin. 2) No consistent
beneficial effect has been shown with
Phenobarbital, Vitamin E or Ethamsylate. In one
study the incidence of IVH was ? in infants
who received Phenobarbital when compared to
controls. 3) Indomethacin has been shown to
reduce the incidence of severe IVH. However
at longer term follow up the incidence of
cerebral palsy and mental retardation is
comparable to controls.
Kuban et al Pediatrics 198677443, Allen et
al AJDC 1997151581 Schmidt et al N Engl J
Med 20013441966
25
Postnatal Indomethacin administration and
incidence of PV-IVH
Grade of PV-IVH Indomethacin
Placebo
(n209)
(n220)
No IVH
184(88) 182(82) 1
7(3.8)
13(5.5) 2
15(7.5)
16(7.5) 3
2(1.0)
1(0.5) 4
1(0.5) 10(4.5)
Cerebral Palsy at 36 months was noted in 8 who
received Indomethacin versus 11 of controls
Ment et al Pediatrics 199494543 Plt .01,
Allen Arch Pediatr. Adolesc Med 1997151580
26
Outcomes of Infants who received Indomethacin as
compared to Placebo
Outcome Event Rate
Odds Ratio
Indomethacin Placebo Adjusted P
Group
Group (95 CI) Value PV/IVH
236/569(41) 234/567(41)
1.0(.8-1.3) 0.86 Severe IVH
52/569(9) 75/567(13) 0.6(0.4-0.9
0.02 IPE/PVL/Cysts 125/563(22)
142/562(25) 0.8(0.6-1.1) 0.23 Hydrocephalus
15/470(3) 9/480(2)
1.7(0.7-3.9) 0.21 Seizure disorder
8/470(2) 7/483(1) 1.2(0.4-3.3)
0.76 Death/impairment 271/574(47)
261/569(46) 1.1(0.8-1.4) 0.61 Cerebral
Palsy 58/467(12) 55/477(12)
1.1(0.7-1.6) 0.64 MDIlt 70
118/444(27) 117/457(26) 1.0(0.4-2.5)
0.86
Adapted from Schmidt et al N Engl J Med
20013441966
27
PERIVENTRICULAR LEUCOMALACIA
28
Definitions

• Bilateral Cystic PVL refers to necrosis of
  white matter
• adjacent to the external angle of the lateral
  ventricles.
• Cysts maybe present upon ultrasound(US) at
  birth
• (representing an antenatal injury) or evolve
  postnatally
• usually within the first 2 to 3 weeks of
  life. Later
• appearing lesions also occur.
• Injury to white matter may evolve in the
  absence of
• cysts. The US will demonstrate progressive
  non
• obstructive lateral ventricle enlargement

29
Ultrasound changes in infants atrisk for White
matter injury

• Ultrasound Finding
  Incidence
• Hyperechogenicity
• Cystic change
  3 lt 1500g
• Ventriculomegaly
• Normal WM
• Non obstructive
  3 lt 1500g
• Ventriculomegaly
• Hyperechogenicity
• No Cysts
  5-10 lt 1500g
• Normal Ventricle size

30
TIMING OF WHITE MATTER INJURY BY ULTRASOUND

• 50 of the cases of cystic PVL are identified
  early, i.e., within the initial 72 hours of life
  by the appearance of periventricular
  hyperechogenicity
• 50 of the cases of cystic PVL are identified
  late, i.e., gt 28 days in the absence of prior
  ultrasonographic changes
• Nonobstructive ventriculomegaly is a slowly
  progressive change usually apparent beyond 28 days

31
Neuropathology

• PVL occurs in regions which represent
  arterial border
• zones which are termed watershed areas.
  These areas
• would be expected to be more susceptible to
  ischemic
• injury when there is a fall in systemic
  blood pressure
• and/or a decrease in cerebral blood flow.
• Both focal and diffuse involvement of white
• matter may occur. Diffuse injury is more
  commonly noted
• in premature infants requiring prolonged
  ventilator support.
• 
  

32
FOCAL Cystic PVL

• Occurs within the distribution of the end
  branches of
• the long penetrating arteries
• Common site - adjacent to the lateral
  ventricle -
• Anterior - at the level of foramen of
  Monroe
• Occipital- at the trigone
• Localized infarction
• Myelin loss and focal ventricular dilatation
  in the region
• of the trigone are long term sequelae.

33
DIFFUSE WHITE MATTER INJURY - MICROSCOPIC CHANGES

• 1) May occur without overt necrosis
• 2) Loss of oligodendrocyte
• 3) Reactive astrocytosis

34
PATHOGENESIS OF PVL

• Vascular Factors Intrinsic
  Vulnerability of the
  Oligodendrocyte
• Border zone region Free radicals
• Limited vasodilatory capacity Glutamate
• Pressure passive circulation Cytokines

35
(No Transcript)
36
CYSTIC PERIVENTRICULAR LEUKOMALACIA IN THE
PREMATURE INFANT - PERINATAL CHARACTERISTICS

• n 14
• n 10 (70) 4 (30)
• Minimal DR Intervention Hypotension
• Mild to Moderate Lung Disease
• No Overt Hypotension
• Incidence Diagnosis Noted on
• Cranial Sonogram
• Perlman et al, Pediatrics 1996

37
FACTORS ASSOCIATED WITH PVL

• Chorioamnionitis (OR 6.77 (CI 1.77-25) PPV 11.5
• PROM ( OR 6.59 (CI 1.96-22.10) PPV 11.5
• Hypocarbia
• Recurrent apnea and bradycardia
• Meningitis
• Ideopathic

38
INTRINSIC VULNERABILITY OF THE DIFFERENTIATING
OLIGODENDROCYTE

• Studies indicate that differentiating
• oligodendroglia in the process of myelination are
  the most vulnerable to injury. Potential factors
  important in the genesis of white matter injury
  include free radicals, excess extracellular
  glutamate and cytokine release.

39
Potential Mechanisms of Oligodendroglial Injury

• Free radical injury
• Excitotoxic injury
• Cytokines

40
FREE RADICAL METABOLISM

• Superoxide Dismutase
• O2 H2
  H2O2
• GSH-Peroxidase
• H2O2
  H2O O2
• Catalase
• Fe
• OH OH- Fe
• Fenton Reaction

41
DEVELOPMENT OF ANTIOXIDANTS

• 1) Superoxide dismutase immunoreactive glial
  cells are
• present in cerebellar white matter from
  25-26 wks
• and in temporal white matter from 31-32
  wks in
• humans (Takashima).
• 2) ? in catalase concurrent with onset of
  myelination in
• rat brain (Arnold).
• 3) Catalase immunoreactivity in glial cells is
  only
• present from 31-32 wks gestation in
  humans(Houdou)
• Takashima, et al. Brain Dev 12211-213, 1990.
• Arnold, et al. Brain Res 1551-17, 1978.
• Houdou, et al. Brain Res 556267-270, 1991.

42
FREE RADICAL INDUCED OLIGODENDROGLIAL INJURY

• 1) Vulnerability to free radical injury, e.g.,
  hydrogen
• peroxide, hydroxyl radical.
• 2) Free radical oligodendrocytic cell death
  appears to
• be mediated via apoptosis.
• 3) Free radical induced oligodendroglial injury
  can be
• prevented with scavengers i.e., superoxide
  
• dismutase, desferroxamine, Vitamin E.

43
Glutamate toxicity
Glutamate appears to target the immature
oligodendrocyte Mechanisms of glutamate
toxicity Non-receptor - 1) Glutamate enters
the cell in exchange for cystine resulting in
? intracellular cysteine. 2) Cysteine is the
precursor for glutathione. . Receptor - 1)
Cultured rat optic nerve oligodendrocytes express
kainate and AMPA subunits. 2)
Oligodendroglial death can be blocked by CNQX and
Ca removal.
IL-1 mediated injury
44
COMPARISON OF NMDA INDUCED STRIATAL AND
HIPPOCAMPAL INJURY IN ADENOVIRAL ENCODING
GALACTOSIDASE AND ADENOVIRAL ENCODING
INTERLEUKIN-1 ANTAGONIST-INFECTED ANIMALS

• n Left Right
• Striatal vol (mm3)
• Ad.Lac2 16 14.2 ? 0.7 5.9 ?
  0.7
• Ad.Lac2 Interleukin-1 19 13.2 ? 0.5 10.9 ? 0.2
• Hippocampal vol (mm3)
• Ad.Lac2 16 4.3 ? 0.3 2.4 ? 0.3
• Ad.Lac2 Interleukin-1 19 4.8 ? 0.3 4.5 ?
  0.3
• Plt0.0001 (two-way Anova)
• P. Hagan, et al. Neuroscience 751033, 1996.

45
Potential Mechanisms of Oligodendroglial Injury

• Free radical injury
• Excitotoxic injury
• Cytokines

46
Inflammation and Potential Brain Injury-
Experimental Observations

• Within CNS microglia release TNF ? , IL-1, IL-6
  and induce a reactive astrocytosis. These effects
  can be stimulated by endotoxin and blocked by
  specific antagonists.
• In adult rats, intracisternal administration of
  IL-1 and TNF? results in a dose dependent ? in
  blood brain permeability.
• IL-6 induces the bipotential oligodendrocyte
  precusor cell -2A towards the astrocytic rather
  than the oligodendrocytic pathway.
• PVL develops in a kitten model following
  endotoxin induced injury, in fetal rabbits
  following induction of maternal intrauterine
  infection and in fetal sheep following systemic
  endotoxemia.

47
Inflammation and Brain Injury-Clinical
Observations

• Chorioamnionitis and/or ? cord blood inflammatory
  cytokines i.e. IL-6, IL-1 ? and TNF-?
  concentrations have been associated with white
  matter injury (WMI) and/or cerebral palsy.
  Specifically ? umbilical cord IL-6 concentration
  was associated with a six-fold ? in WMI.
• The fetal inflammatory response (funisitis) is
  associated with the highest cytokine levels
  this response may be biologically more important
  than the maternal effects.
• Preterm infants were noted to have TNF ? , IL-1 ?
  , IL-6 expression within white matter in 15/17
  cases with PVL versus 3/17 cases without PVL

Yanowitz Pediatr Res 200251310, Leviton Pediatr
Res.199946566 Yoon, Am J Obstet Gynecol 96,
97
48
CHORIOAMNIONITIS (CA) AND PVL - WHAT IS THE
CLINICAL EVIDENCE?

• Study Odds Ratio 95 CI
• Perlman 5.14 1.27-20.79
• (Peds 96)
• Verma 2.23 1.23- 3.94
• (Am J Ob Gynecol 97)
• Alexander 3.40 1.60- 7.30
• (Obstet Gynecol 98)
• Leviton 5.8 1.00- 3.50
• (Seminars in Neurol 98)

49
PROPOSED PATHOGENESIS OF WHITE MATTER INJURY

• Ischemia
• Reperfusion Reperfusion
• Glutamate
• Cytokine Oligodendrogial Free
  Radical
• Release Cell Death Formation

50
Prevention of PVL
1) Likely to be difficult 2)
Hypotension is only noted in 30 of cases.
3) Although an association with
chorioamnionitis has been
demonstrated, the sensitivity
approximates 10. 4) Anti-oxidant therapy
has not been successful in the
treatment of other complications of
prematurity.


51
OUTCOME of Infants with White Matter Injury
52
Outcomes of Infants who received Indomethacin as
compared to Placebo
Outcome Event Rate
Odds Ratio
Indomethacin Placebo Adjusted P
Group
Group (95 CI) Value Severe IVH
52/569(9) 75/567(13)
0.6(0.4-0.9 0.02 IPE/PVL/Cysts
125/563(22) 142/562(25) 0.8(0.6-1.1)
0.23 Cerebral Palsy 58/467(12)
55/477(12) 1.1(0.7-1.6) 0.64 MDIlt 70
118/444(27) 117/457(26)
1.0(0.4-2.5) 0.86
Adapted from Schmidt et al N Engl J Med
20013441966
53
Ultrasound Status and Global Cognitive Functioning
Global Cognitive Total Normal IVH
PL/VD Outcomes (n597)
(n468) (N83) (N46)
Normal Intelligence 529 (89)
436 (93) 73 (88) 20 (43.5) Borderline
Intelligence 38 (6.0) 26 (5.6) 5
(6.0) 7 (15.2) Mental Retardation 30
(5.0) 6 (1.3) 5 (6.0) 19 (41.0)
Normal Intelligence Stanford Binet composite gt
84, Borderline lt 84, Mental Retardation lt 68
IVH germinal and intraventricular hemorrhage PL
Parenchymal lesion, VD Ventricular Dilation
From Whitaker et al. Pediatrics 199698719
54
Two Year Followup in Ten Infants with Cystic PVL

• Category Outcome
• Motor deficits 100
• Cognitive deficits 90
• Hearing deficits 40
• Visual deficits 80
• Perlman et al, Pediatrics 95

55
Outcome in Infants with Non Obstructive
Ventriculomegaly
1) Important predictor of adverse cognitive and
motor development at school age- e.g.
IQlt70(OR 19, 95CI4.5,80.6) 2) Increases the
risk for pshychiatric disorders e.g. ADHD
(OR 4.4, 95 CI 1.8,10.3)

1)
Ment et al Pediatrics 1042431999 Whitacker et
al Arch Gen Psychiatry 54847,1997 Stewart Lancet
3531651999
56
Cognitive Injury with White matter Injury
1) Adults with periventricular white matter
injury experience cognitive decline three
times more rapidly as compared to adults
without the lesion. 2) In a neuropathologic
study utilizing Paralbumin (PA) which is
expressed in a subpopulation of GABA- ergic
inhibitory neurons, revealed ?PA
immunoreactivity in somatosensory cortex in
infants with diffuse but not focal PVL
de Groot et al Ann Neurol 2002335-341, Takashima
Neuropediatrics 19993014
57
Quantitative 3D-MRI Volumes of Cerebral tissues
at Term in Preterms with PVL, No PVL and Healthy
Term infants
Preterms with Preterms
no Normal PVL at term
PVL at term Term
(n14) (n10)
(n14)

Cerebral Cortical 15741
21125 21821 Gray
Matter Myelinated White 144
237 2710 Matter CSF
6415
5224 3213


P0.0001 Adapted from Inder et al Ann Neurol
199946755
58
Conclusions

• White matter injury is the major cause of
  long-term
• motor and cognitive injury in the premature
  infant.
• The two major mechanisms of WMI are ischemia
  with
• resultant necrosis and/or free-radical
  induced cell
• death mediated via apoptosis.
• Funisitis and glucorticoids are two major
  perinatal
• factors modulating WMI.
• Preventative strategies have to begin in utero
  by
• identifying infection and/or inflammation.
  Postnatal
• strategies have to focus on those infants at
  greatest
• risk. for white matter injury.