H E:

_::___

 

3;;__:__‘=:_E_S_:___:_______:_= mmm

 

 

 

’ “FF"?
- ‘ Pv‘.‘s¢
lbbn
I

 

V

Heller

LIBRARY
Michidan State

 

’31!

This is to certify that the
thesis entitled

THE INTERACTION OF CHORIOAMNIONITIS WITH
MATERNAL AND FETAL CHARACTERISTICS IN VERY-
LOW-BIRTHWEIGHT INFANTS ON THE RISK OF
INTRACRANIAL BRAIN LESIONS

presented by

BRIDGET MARIE PROTAS

has been accepted towards fulfillment
of the requirements for the

 

 

MS. degree in Epidemiologl

V Major Professor's Signature
3/2 (/0 8’
' I

Date

MSU is an affirmative-action, equal-opportunity employer

_.-.-o-.-n-u—-------u---.---—n--.-—~-.-

.c—v-----.--.-.-—---.-.--.-n-a--._.-.-.-n-u-o---A—.--.-.--I-A-—A

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K IProilAcc8Pres/CIRC/DateDue indd

THE INTERACTION OF CHORIOAMNIONITIS WITH MATERNAL AND FETAL
CHARACTERISTICS IN VERY-LOW-BIRTHWEIGHT INFANTS ON THE RISK OF
INTRACRANIAL BRAIN LESIONS
By

Bridget Marie Protas

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Epidemiology

2008

ABSTRACT

THE INTERACTION OF CHORIOAMNIONITIS WITH MATERNAL AND FETAL
CHARACTERISTICS IN VERY-LOW-BIRTHWEIGHT INFANTS ON THE RISK OF
INTRACRANIAL BRAIN LESIONS
By
Bridget Marie Protas

White matter damage (WMD) is the major neurological disorder in premature
infants. This study evaluated how the relationship between chorioamnionitis and WMD
was modified by maternal and fetal factors in very low birthweight infants. The 1,607
infants in the analysis were part of a previous study by the Developmental Epidemiology
Network. Infants weighed between 500 and 1500 grams and underwent at least one
cranial ultrasound scan. Chorioamnionitis was defined by histopathologic examination of
the placenta. Logistic regression modeling was limited to the 879 infants with gestational
age less than 29 weeks. In the unadjusted model, small-for-gestational age (SGA) infants
unexposed to chorioamnionitis were Significantly inversely related to WMD (OR 0.1;
95% CI 0.01 -0.6) while SGA infants exposed to Chorioamnionitis had a null association
with WMD (OR 0.9; 95% CI: 0.3-2.2). The appropriate-for-gestational age (AGA) and
large infants Showed no difference between chorioamnionitis status. After adjusting for
potential confounders, the odds ratio and confidence interval for the SGA infants
unexposed to Chorioamnionitis remained unchanged. Vaginal delivery persisted as a
Significant risk factor for WMD and receipt of a partial or full course of antenatal steroids
was inversely related to WMD. Distinctive features of this subgroup of SGA infants

should be investigated and should help inform the scientific community at large and

clinicians in their decision-making with obstetric care.

ACKNOWLEDGEMENTS

I would like to thank Dr. Nigel Paneth, Dr. Claudia Holzman and Dr. Hossein
Rahbar for their guidance and support that helped me complete this thesis. I am grateful
for their patience and encouragement. I have learned a great deal from their classes
which gave me a strong foundation in research that I can build upon.

I would also like to thank Dr. Alan Leviton for allowing me to use his data for my
thesis and Elizabeth Allred for her technical support.

Last but not least, I would like to thank my fiance', Matt, for his love and support.
He provides me with everything I need to be happy and successful and I could not have

done this without him.

iii

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. v
CHAPTER 1
INTRODUCTION ............................................................................................................... l
The problem of preterm infants ........................................................................................... 1
White matter damage and prematurity ................................................................................ 2
Preterm infants, morbidity, and mortality ...................................................................... 3
Pregnancy complications associated with WMD ................................................................ 4
Intrauterine growth restriction ........................................................................... 5
Pregnancy-induced hypertension/preeclampsia ................................................... 6
Pregnancy and labor complications associated with preterm birth .............................. 7
Mechanism of WMD .......................................................................................... 8
Research question ............................................................................................ 10
CHAPTER 2
MATERIALS AND METHODS
Study population ................................................................................................................ 13
Dependent variable ............................................................................................................ 13
Independent variables ........................................................................................................ 14
Approach to the analysis .................................................................................................... 15
RESULTS
Sample characteristics ........................................................................................... 16
Risk of WMD by infant characteristics in the entire sample (Table 2) ................. 18
Analysis stratified by gestational age (Table 3) .................................................... 20
Influences of chorioamnionitis on risk of WMD ................................................... 23
CHAPTER 4
DISCUSSION .................................................................................................................... 28
REFERENCES .................................................................................................................. 32

iv

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

LIST OF TABLES

Sample characteristics of 1607 VLBW infants, DEN data, 1991-93 ............. 17-18
Risk of WMD in 1607 VLBW infants, DEN data, 1991-93 .......................... 19-20
Risk of WMD in 1607 VLBW infants stratified by gestational age < 29

weeks and 2 29 weeks .................................................................................... 21-22
Relative risk of WMD in 1146 VLBW infants with and without any
chorioamnionitis stratified by gestational age, DEN data, 1991-93 .................... 23
Relative risk of WMD for 1607 VLBW infants with and without vaginal
delivery stratified by gestational age, DEN data, 1991-93 .................................. 25
Unadjusted ORS and 95% CI for the effect of birthweight z-scores in the

presence and absence of chorioamnionitis on WMD in 638 VLBW infants
< 29 weeks gestation, DEN data, 1991-93 .......................................................... 26

Adjusted ORS and 95% CI for the effect of birthweight z-scores in the
presence and absence of chorioamnionitis on WMD in 638 VLBW infants
< 29 weeks gestation, DEN data, 1991-93 .......................................................... 27

CHAPTER 1

INTRODUCTION

The Problem of Preterm Infants

Preterm labor, defined as birth before 37 weeks gestation," 2 accounts for
approximately 8-12% of live births in developed countries.3’ 4 Very preterm birth,
defined as birth before 32 weeks gestation, accounts for 2% of all births.5 Prematurity is
a perplexing phenomenon that is known to be the leading cause of neonatal morbidity and
mortality.4 It seems to be multifactorial in nature1 and is the major contributor to reduced
birthweight. Attempts to reduce the incidence of preterm birth have been unsuccessful as
the rates are steadily climbing in the United States.6 In the United States, the rate of
preterm birth rose from 9.7% to 11% in the years from 1990 to 2005.7 In addition,
preterm delivery disproportionately affects African Americans. The rate of preterm birth
and very preterm birth in African Americans in 2005 was 18.4% and 4.16%, respectively,
compared to 11.7% and 1.63%, respectively, in Caucasians.8

Advances in perinatal care in the 1990’s, which include regionalization of care
and the use of antenatal steroids, surfactant and assisted ventilation 9"0 have resulted in
increased survival in premature newborns, but the increase in survival has also increased
neurodevelopmental disability in the population. ” Stoelhorst et al compared two Dutch
cohorts of very preterm infants less than 29 weeks gestation, the Project on Prematures

and Small for Gestational Age Infants (POPS) of 1983 and the Leiden Follow-up Project

on Prematurity (LFUPP) of 1996. In the post-surfactant era, not only were there more

survivors for all gestational age categories, but also more preterm infants with
intraventricular hemorrhage (IVH) and necrotizing enterocolitis.l2

It has been estimated that 10-15% of the most premature babies, usually under 29
weeks gestation, develop cerebral palsy (CP).l3 White matter damage (WMD) underlies
at least 50% of CP cases and many studies have reported that the severity of preterm birth

is inversely related to the incidence of white matter damage.” '5’ '6’ '7

Therefore, very
preterm infants represent a very high risk group for brain abnormalities. A large Dutch
cohort study that followed preterm infants weighing less than 1500 grams to the age of 14

predicted that as many as 40% of the teenagers will be unable to become fully

independent adults.l8

White Matter Damage and Prematurity

White matter damage (WMD) is the major form of brain damage found in
premature infants. '9 Depending on how extensive it is, it may be associated with long-
term neuromotor, cognitive, and behavioral limitations including cerebral palsy.” 21’22’ 23’
24 The true incidence of WMD in very preterm infants is not completely known.25 With
the utilization of ultrasound scans and MRI, it has been estimated that approximately 8%
of VLBW infants will develop focal necrotic WMD, often described as periventricular
leukomalacia (PVL), and 20-50% will develop more diffuse WMD, which is a more
global loss of cells from the oligodendroglial lineage.” 27‘ 28 Very preterm infants who
die within the first few days of life are not included in these estimates and their high

frequency of WMD may contribute to an underestimation of the true incidence. In

addition, the white matter abnormalities seen on ultrasound and MRI scans may merely

be the “tip-of—the-iceberg” of the disruption that has occurred to the glial cells in the
developing fetal brain.29
White matter damage in preterm infants has a broad definition. Some authors
restrict attention to PVL only, while others have extended the definition to include
associated gray matter abnormalities.30 Furthermore, other researchers have also
included ventriculomegaly, usually defined as global enlargement of the lateral ventricles

31,15

in the definition. Ventricular enlargement has been hypothesized to be a

consequence of white matter damage30

Preterm Infants, Morbidity, and Mortality

Risk factors for white matter abnormalities share some of the same risk factors for
infant mortality. Morbidity and mortality has been associated with, but not limited to,
male gender, low gestational age and very low birthweight. Newborn preterm males
have been shown to suffer from higher mortality rates32 and higher incidences of other
perinatal complications such as cerebral palsy (CP)33 and intracranial hemorrhage34 than
females. In a study of 1,730 VLBW infants, mortality was higher in males than females
of similar birthweight and gestation by as much as 20%.}:2 Cerebral palsy has also been
shown to be higher among male infants compared to female infants in a large cohort
ascertaining cases from 9 registers in 5 European countries.35 Therefore, it is plausible
that male preterm infants could exhibit a higher incidence of WMD due to differences
involving sex hormones during early stages of development.36

Low birthweight is essentially caused either by shorter gestation, intrauterine

growth restriction or a combination of the two. Although the three different types of low

birthweight infants may arise from different causal pathways, they all have a higher risk
for mortality than infants of an appropriate weight. Babies weighing less than 1,500

grams have 100 times the risk of death of normal weight babies.37

Pregnancy Complications Associated with WMD

Among pregnancy complications known to have adverse effects on very preterm
infants and that may contribute to WMD are chorioamnionitis, fetal vasculitis, and
intrauterine growth restriction.

A simple definition of chorioamnionitis is inflammation of the placenta and extra
placental membranes.38 Clinical detection usually includes a combination of signs and
symptoms, most commonly maternal fever of at least 38°C on one or more occasions, an
elevated white blood cell count (usually 15,000 to 18,000 cells/mm3), maternal
tachycardia, fetal tachycardia, tender uterus, and/0r purulent/foul-smelling discharge.4’ 38
Chorioamnionitis is most commonly an ascending infection from normal vaginal flora.
Risk factors for chorioamnionitis include nulliparity, extended duration of labor and
rupture of membranes, pre-existing genital infections and low economic status.4 A large
fraction of chorioamnionitis is subclinical where the only apparent symptom is preterm
labor.39 Chorioamnionitis complicates approximately 1-5% of term pregnancies, but
accompanies anywhere from 3-30% of pregnancies, especially where labor is initiated by
premature rupture of membranes.4 In the very preterm neonate, 22-32 weeks gestation,
chorioamnionitis is present in approximately 50% of pregnancies.40

Some studies have shown that very low birthweight infants exposed to

. . . . . . 41. 42
chorioamnionitis had Significantly decreased mean performance IQS. Another recent

study found a significantly higher incidence of intraventricular hemorrhage and
retinopathy of prematurity in low birthweight infants exposed to histopathologically
defined chorioamnionitis.43 A meta-analysis by Wu et al found clinical and histological
chorioamnionitis to be significantly associated with cystic periventricular leukomalacia
(cPVL) (relative risk (RR) 2.6 and 1.6, respectively) and CP. Analyses of both clinical
and histological chorioamnionitis had borderline statistical heterogeneity between
studies, and Wu noted that no two studies had the same criteria for diagnosing clinical
chorioamnionitis. Five studies examined with the most stringent criteria of clinical
chorioamnionitis produced remarkably homogeneous results showing chorioamnionitis
was a risk factor for cPVL.44

Fetal vasculitis, inflammation of the fetal vessels in the placenta, has been shown
in one large study to be a better predictor of white matter damage in the preterm neonate
than a purely maternal inflammatory response.23 The fetal inflammatory response is
believed to be in the causal pathway indicating the most severe form of intrauterine
infection that has sequestered for some time and evolved to a fetal inflammatory

response.

Intrauterine Growth Restriction

Intrauterine growth restriction (IUGR) sometimes causes infants to be
iatrogenically delivered preterm. The major risk factors for IUGR are smoking/drug use
during pregnancy, pregnancy-induced hypertension, low pre-pregnancy weight and low
weight gain during pregnancy.45 The placenta is an active endocrine organ that is the

lifeline for the fetus.46 Disruption to the proper function and Size of the placenta will

inevitably have adverse impacts on the developing fetus that is dependent on it. In a
cohort of 685 preterm infants born less than 33 weeks gestation, WMD was inversely
associated with IUGR (measured as birthweight-for-gestational age below the 10th
percentile) and preeclampsia when compared to infants born after intrauterine infection
(odds ratio (OR)=0.08, 95% CI 0.02-0.41) after adjusting for gestational age, mode of
delivery, and antenatal antibiotic administration.47

Controversy exists as to whether IUGR is protective from neonatal morbidity by
forcing the fetus to mature. Several studies have refuted these inverse associations
showing that small-for-gestational age infants have higher mortality, higher morbidities
such as IVH and necrotizing enterocolitis, and increased length of hospital stay.“ 49
Jarvis and colleagues showed the risk of severe CP increased ll-fold for preterm infants
who are IUGR (birthweight below the 3lrd percentile) compared to preterm infants with
optimum growth of z-scores between 0.67 and 1.28 in their large cohort.35 This analysis
did not adjust for potential confounders and the case-only study design may limit the
generalizability of the results. Garite and colleagues analyzed infants using different
definitions of growth restriction: IUGR diagnosed antenatally, birthweight below the 10th
percentile as SGA, and a combination of both IUGR and SGA against a control group.

After adjusting for potential confounders, the growth restricted infants were not at an

increased risk of severe IVH or retinopathy of prematurity.

Pregnancy-induced Hypertension/Preeclampsia

It has been suggested that hypertensive disorders, such as pregnancy-induced

hypertension (PIH), impair normal fetal growth and that may result in a small, but more

mature, fetus.50 Studies have shown that infants exposed to PIH/preeclampsia have
increased morbidity and mortality,5 I’ 52 but others have shown no or inverse associations
with morbidity and mortality. Rates of IVH were similar among mothers with PIH and
normotensive mothers (2.2% vs. 2.2%) as was neonatal death (4.5% vs. 4.5%), while PV-
IVH was less common (16% vs. 30%).53‘ 54 Also, two other studies on pregnancies
complicated by preeclampsia showed a lower rate of periventricular/intraventricular
hemorrhage55 and cystic periventricular leukomalacia.56
Collins and Paneth57 argue that preeclampsia, being highly associated with IUGR,
may only appear protective of brain damage because the comparison group of preterm
infants was delivered for other reasons such as infection and the only true comparison
would be elective preterm deliveries. Furthermore, these infants are more likely to be of
greater gestational age which may be what is really providing protection. Another

important point is that iatrogenic delivery of IUGR/preeclamptic infants is a planned

event which may provide better and more immediate neonatal care.

Pregnancy and Labor Complications Associated with Preterm Birth

Many factors that may lead to prematurity operate through initiation of delivery
by spontaneous preterm labor, preterm premature rupture of membranes (PPROM),
hypertensive disorders, some other maternal complication or unknown pathway.
Spontaneous preterm labor and PPROM have been associated with vaginal delivery while
pregnancy-induced hypertension/preeclampsia has been associated with indicated

. . 8
delivery, most commonly, Cesarean section.5

The biggest predictor of spontaneous preterm labor is history of a prior
spontaneous preterm birth.59 Studies have also shown that Black race, bacterial
vaginosis, presence of fibronectin in cervicovaginal secretions, short cervix and
chorioamnionitis are also associated, but are not strong predictorsz’ 4’ 60’ 61’ 62

Risk factors associated with PPROM have been inconsistent among
epidemiologic studies2 and may due to the difficulty in distinguishing between onset of
labor and membrane rupture as the initiator of delivery.l Several characteristics have
been found in common between PPROM and spontaneous preterm labor and include
Black race, low socioeconomic status, smoking, and infections including
chorioamnionits.4 Chorioamnionitis is more common when PPROM occurs before 32
weeks gestation and the adverse effects of PPROM are also infection-related.4 These
dangers include WMD70, sepsis, and meningitis.

Hypertensive disorders in pregnancy affected 44 per 1,000 live births in a seven-
State reporting area in 200463 and include pregnancy-induced hypertension (PIH),
preeclampsia and eclampsia. These conditions are defined by a blood pressure of 140/90
mmHg, presence of proteinuria, and eventually seizures.64 Difficulties arise in
epidemiologic studies because the relationship between hypertensive disorders of
pregnancy, fetal growth restriction and WMD are not completely understood. The
hypertensive conditions are sometimes viewed as separate conditions that may affect

normal fetal growth or different severities or types of the same condition.65

Mechanism of WMD

The pathophysiology of the echolucent/echodense images and ventriculomegaly
in the preterm neonate are not well understood. There are two mechanisms involved that
may co-occur and account for much of the brain damage. First, the hypoxia-ischemia
theory hypothesizes that loss of oxygen leads to neuronal death, and is common in
preterm newborns.3 The prevalence of fetal asphyxia, defined as inadequate exchange of
blood gases measured by umbilical artery base greater than or equal to 12 mmol/L, in
infants from a study of 23,000 preterm pregnancies was 73 per 1,000, of which 50% were
moderate to severe asphyxia.66 Animal models have demonstrated that lack of oxygen
triggers an elevated release of excitatory neurotransmitters and contributes to brain
damage.67 Studies have found that infants who do not experience labor and vaginal
delivery are at a decreased risk of white matter damage.“ 69 Labor and delivery are
traumatic events for an infant and most likely, even more so for an infant who lacks
maturity. Low et a1 states that the developing fetus can compensate for asphyxia to a
certain degree, but intermittent exposure may be cumulative and surpass a threshold of
tolerance.66 Although several studies have found a null association with mode of delivery
and neonatal morbidity and mortality, they are limited by small sample sizes.70’ 7'

lschemia refers to reduced cerebral blood flow. Animal models have provided
evidence demonstrating the pathophysiologic relationship between reduced cerebral
blood flow and periventricular white matter damage. Using carotid artery occlusion,
neonatal rats had a 90% incidence rate of WMD.72 Ovine fetuses demonstrated WMD
proportional to the duration of ischemia also using carotid artery occlusion and
determined by the number of fluorescently labeled degenerating nuclei from the white

matter lesions (30 min:110 nuclei/mmz; 37 min: 1 51 nuclei/mmz; 45 minz294

nuclei/mmz).73 Back et al demonstrated a window of vulnerability to WMD by
autopsying human brains between 23 and 41 weeks gestational age. Immature
oligodendrocyte were approximately 3-times more prevalent in infants 28-41 weeks
gestation than infants 18-27 weeks gestation (30.9% vs. 9.9%; p<0.0001).74 The
immature oligodendrocytes may be a marker of protection from ischemia because their
development coincides with restricted localization of myelin sheaths to the
periventricular white matter. Concepts involved in ischemia such as free radical
generation as mediators of WMD are difficult to validate because timing and duration of
ischemia are only two of many factors affecting the extent of neurological damage.75
The inflammatory hypothesis postulates that polymorphonuclear leukocytes
invade the chorionic plate and cross over to invade the amnion and amniotic fluid as
well.26 These maternal leukocytes cause the release of large amounts of inflammatory
cytokines and are presumed to initiate preterm labor after an influx of prostaglandins.3
Increased local immunosuppression during pregnancy may facilitate the invasion of
ascending microorganisms from normal vaginal flora, leading to weakening of the
membranes that predispose them to rupture.76 In the Alabama Preterm Study, Andrews et
al found that women whose placental cord blood had a cytokine IL-6 level above 34.5
pg/mL had a 15-fold increase for membrane polymorphonuclear infiltration.77 Maternal
serum IL-6 has also been Shown to be a biomarker in women with premature rupture of
membranes whose infant is likely to develop funisitis.78 IL-6 has been thought to serve
as a messenger between mother and fetus, but a study by Aaltonen et a1 using placental
perfusion has shown that lL-6 remained on the same side of the placenta and did not

transfer to the fetal side.79

10

Whether the mechanism of WMD involves transplacental passage of cytokines,
compromised oxygen and/or blood flow to the fetus, more diffuse intermediary molecules

such as prostaglandins, or some combination of the above, further research is needed.80

Research guestion

Evidence is limited on the modifying effect of chorioamnionitis on WMD in very
preterm infants. A strong relationship between chorioamnionitis and preterm birth has
been demonstrated in many studies,4‘ 38 but the role of intrauterine infection on WMD has
conflicting results. This is mainly due to the use of different criteria for the diagnosis of
chorioamnionitis, the substantially different sample sizes, the inclusion and exclusion of
what constitutes white matter damage and the factors that each study chooses to control
for.

Assessment of the multiplicative effect of chorioamnionitis with other risk factors
in the literature is lacking. Qiu et al studied the association of labor and delivery with
parenchymal echodensities/lucencies and ventricular enlargement (PEL/V E) in the
Neonatal Brain Hemorrhage database and has been the only study found testing a
multiplicative effect of maternal infection with a fetal factor on WMD. Their results
showed when infants were stratified by small-for-gestational-age (SGA) versus
appropriate-for-gestational-age (AGA), the adjusted OR for the association of amnionitis
and PELN E was Significantly above 6.0 in SGA babies and non-significantly below 1.0
for AGA babies.69 The logistic regression models also Showed that mode of delivery
was not independently predictive of PEL/V E and no significant interaction between

amnionitis and mode of delivery was observed. Amnionitis was solely a clinical

11

diagnosis abstracted from medical records and infants with a birthweight up to 2,000
grams were included.

Many studies thus far focus on an additive relationship of maternal and fetal
independent variables on adverse outcomes in preterm infants. There is very little
literature addressing how the relationship between maternal infection and brain
abnormalities in the very premature infant are modified by factors such as mode of
delivery and fetal growth. This study has two aims:

1. To assess if a multiplicative interaction of chorioamnionitis with maternal and
fetal factors on the risk of echolucencies, echodensities and ventriculomegaly
exists.

2. If an interaction is present, to determine how much the relationship between the
independent variable and the dependent variable is modified by the

presence/absence of chorioamnionitis.

12

CHAPTER 2

MATERIALS AND METHODS

Study Population

The 1607 low birthweight infants were part of previous studies by the
Developmental Epidemiologic Network. The infants in the study weighed between 500
and 1500 grams and were live-born between January 1, 1991 and December 31, 1993.
The infants were recruited from the following five participating institutions in three
cities: Children’s Hospital, Brigham & Women’s Hospital, and Beth Israel Deaconess
Medical Center in Boston, MA; Babies Hospital and St. Luke’s Hospital in New York,
NY; and St. Peter’s Hospital in New Brunswick, NJ. All infants in the study had
undergone at least one of three cranial ultrasound scans. The first scan was performed on
postnatal days 1 to 3, the second on postnatal days 7 to 10 and the last scan was
performed between 3 and 8 weeks. Scans were read independently by two
ultrasonographers who were blind to any details of the study. In the event of either
ultrasonographer identifying brain abnormalities including intraventricular hemorrhage,
vetriculomegaly, or echolucent images, the scans were sent to a consensus committee that
agreed on an interpretation of the results. For full details, see reference 81. Placentas
from 1,131 mothers were collected and sent for morphologic examination.
Dependent Variable

The dependent variable of white matter damage consisted of infants having at
least one of the following three brain abnormalities: echolucent images, echodense

images, or vetriculomegaly. Echodense and echolucent images were also mutually

13

exclusive. If an echolucent image followed an echodense image or vice versa, then the

abnormality was classified by the most recent scan.

Hansaneviton

Independent Variables

Chorioamnionitis was defined by pathologic examination of the placenta as
presence of polymorphonuclear leukocytes in the chorion and amnion and was
also a binary variable.

Birthweight was measured in grams and 98% of gestational age was measured by
fetal ultrasound or date of mother’s last menstrual cycle.

Gestational age was divided into less than 29 weeks and greater than or equal to
29 weeks because the older gestational ages over-represent growth-restricted
infant, making studies of the effect of IUGR difficult.

Birthweight z-score is the number of standard deviations by which the birthweight
deviates from the median of the gestational age as proposed by Yudkin.82

Mode of delivery was either vaginal delivery or Cesarean section.

Initiators of preterm delivery were categorized as preterm labor, premature
rupture of membranes, pregnancy-induced hypertension/preeclampsia, or other.
Rupture of membranes was coded as less than 1 hour before delivery or greater
than or equal to 1 hour before delivery.

Race/ethnicity was categorized as Caucasian, African American, Hispanic, Mixed
and Other.

Maternal smoking consisted of two variables, smoking versus not smoking during
pregnancy and whether smoking was listed on the mother’s chart, which is an

assumption that she had been a smoker prior to conception.

l4

0 Previous premature infants, having a sexually transmitted disease ever or during
pregnancy, any versus no labor during delivery, any versus no exposure to
antenatal steroids, consumption of alcohol during last month of pregnancy, and
whether the mother was prescribed bedrest were all independent variables that
were analyzed.
Approachfito the Analysis

Data was analyzed using Chi-Square test for proportions to describe the
significance of associations among the independent variables and WMD and associations
among infants with a gestational age less than 29 weeks versus greater than or equal to 29
weeks. Data were stratified by infant gestational age less than 29 weeks versus greater
than or equal to 29 weeks and these strata were then each stratified by presence/absence
of chorioamnionitis. This same stratification technique was also used for mode of
delivery. Fisher’s exact probabilities were used when more than 25% of the cells
contained a value less than 5. P-values less than 0.05 were considered statistically
significant. Because the infants with a gestational age greater than or equal to 29 weeks
overrepresent growth-restricted infants, the logistic regression model used to identify risk
factors for WMD was limited to the 879 infants whose gestational age was less than 29

weeks. All analyses were run using SAS 9.1 (SAS Institute, Cary, NC).

15

CHAPTER 3
RESULTS

Sample characteristics (Table 1)

Of the 1,607 low birthweight infants, 114 (7%) had white matter damage. There
were 798 (49%) females and 809 (51%) males. The mean gestational age was 28.6
weeks (range 20.9 to 37.6) and 55% of the infants were less than 29 weeks gestation.
The mean birthweight z-score of the 1,597 infants with a calculated z-score using
Yudkin’s standards was -0.64 (range -9.84 to 6.26). The mean z-score for the infants
with a gestational age less than 29 weeks was -0.04 (range -3.9 to 6.3) and for infants
with a gestational age greater than or equal to 29 weeks was -1.36 (range -9.8 to 1.1).
The older infants having a mean z-score below 1 standard deviation from the median for
their gestational age demonstrates that they are mostly comprised of growth restricted
infants and splitting the sample at this age is necessary. Overall, 16% of the infants had a
birthweight z-score less than -2, meaning they were in the lowest 2.5% of the birthweight
distribution. For mode of delivery, 623 (39%) of the infants experienced vaginal delivery
and 984 (61%) were delivered by Cesarean section. Seventy five percent of the infants
experienced some amount of labor and 25% experienced none. Of the 1,146 infants with
complete information on presence of chorioamnionitis, 537 (47%) had
histopathologically defined chorioamnionitis and 609 (53%) had none. An almost equal
number of infants were exposed to partial or full antenatal steroids as infants whose

mothers did not receive any antenatal steroids. Delivery was initiated by preterm labor in

16

665 (42%) mothers, by preterm premature rupture of membranes in 508 (32%) mothers,
by pregnancy-induced hypertension in 321 (20%) mothers, and by other reasons in 108
(6%) mothers. Race of the mother was available for 1,581 subjects. There were 742
(47%) White mothers, 414 (26%) African American mothers, 309 (20%) Hispanic
mothers, and 116 (7%) mothers who were Mixed/Other race. Table 1 shows the sample

characteristics.

Table 1. Sample characteristics of 1607 VLBW infants, DEN Data, 1991-93

 

 

Characteristic % n
White matter damage
Yes 7 1 14
No 93 1493
Mean Gestational Age in weeks (Range) 28.6 (20.9-37.6)
Mean Birthweight in grams (Range) 1063.7 (500-1500)
Male 51 809
Female 49 798
Race
White 47 742
Black 26 414
Hispanic 20 309
Other/Mixed 7 1 I6
Missing=26

BIRTHWEIGHT AND GESTATION

Z Z 1 9 I42
1 > Z-score 2 0 24 379
0 > Z-score 2 -1 32 504
-l > Z-score Z -2 19 309
Z-score < -2 16 263
< 29 weeks 55 879
Z 29 weeks 45 728

DELIVERY CHARACTERISTICS

Vaginal delivery 39 623
Cesarean section 61 984

17

Table 1 (cont’d)

 

Initiator of preterm delivery

Preterm labor 42 665
PPROM* 32 508
PIHM 20 321
Other 6 108
Missing=5
Chorioamnionitis 47 537
No chorioamnionitis 53 609
Missing=46l
Membrane rupture Z 1 hour 51 807
Membrane rupture < 1 hour 49 789
Missing=1 l
Prescribed bedrest
Yes 60 890
No 40 583
Missing=l34
Presence of labor
Any 75 1206
None 25 401
Receipt of antenatal steroids
Full/partial 50 803
None 50 804

 

*Preterm premature rupture of membranes
"‘ *Pregnancy-induced hypertension
*“Percents were rounded to the nearest whole number

Risk of WMD bv infmr_t__characteristics in the entire sample (Table 2)

Table 2 shows the risk of WMD in relation to the exposure variables. There was
no significant association between WMD and gender, duration of membrane rupture
greater than or equal to 1 hour, consumption of alcohol last month of pregnancy (data not
shown), or presence or absence labor. Birthweight z-score less than -2, prescription of
bedrest, and receipt of partial or full antenatal steroids were significantly protective from

WMD. Significantly increasing the risk of WMD were African American race,

18

gestational age younger than 29 weeks, presence of chorioamnionitis, and vaginal
delivery. When compared to pregnancy-induced hypertension, preterm labor and preterm
premature rupture of membranes were associated with an elevated risk of white matter

damage.

Table 2. Risk of white matter damage in 1607 VLBW infants, DEN Data, 1991-93

 

 

Characteristic Risk of Brain Relative 95 % CI
Lesions Risk
N/Total
Gender
Female 52/798 0.065 ref
Male 62/809 0.077 1.1 (0.8, 1.7)
Race
White 43/742 0.058 ref
Black 36/414 0.087 1.9 (1.2, 2.9)
Hispanic 21/309 0.068 1.1 (0.7, 1.9)
Other/Mixed 12/1 16 0.103 1.7 (0.9, 3.2)
Missing=26
Gestational Age
2 29 weeks 20/728 0.027 ref
< 29 weeks 94/879 0.107 3.8 (2.4, 6.2)

GES TA T ION AND BIR T HWEIGHT

Z-scorez 1 13/142 0.09 0.7 (0.4, 1.3)
l >Z-scorez 0 43/336 0.12 ref

0 >Z-score Z -1 35/504 0.07 0.6 (0.4, 0.9)
-l >Z—score 2 -2 18/308 0.05 0.4 (0.2 ,0.7)
Z-score < -2 5/274 0.02 0.1 (0.06, 0.4)

DELIVERY CHARACTERISTICS

No Chorioamnionitis 32/609 0.053 ref

Chorioamnionitis 53/537 0.099 1.8 (1.2, 2.8)
Missing=46l

Initiator of preterm delivery

PIH“ 11/321 0.034 ref

Preterm Labor 53/665 0.080 2.3 (1.2, 4.4)
PPROM* 44/509 0.086 2.5 (1.3, 4.7)
Other 6/108 0.056 1.6 (0.6, 4.1
Missing=5

19

Table 2 (cont’d)

 

Mode of Delivery

Cesarean Section 48/984 0.049 ref
Vaginal Delivery 66/623 0.106 2.1 (1.5, 3.4)
Missing=3
Presence of labor
None 22/40] 0.055 ref
Any 92/1206 0.076 1.3 (0.8, 2.1)
Missing=17
Membrane Rupture < 1 hr. 48/789 0.061 ref
Membrane Rupture Z 1 hr. 65/807 0.081 1.3 (0.9, 1.9)
Missing=l 1
Any bedrest
No 53/530 0.10 ref
Yes 52/890 0.06 0.6 (0.4, 0.9)
Missing=l34
No antenatal steroids 71/804 0.088 ref
Any antenatal steroids 43/803 0.053 0.6 (0.4, 0.8)

 

*Preterm premature rupture of membranes
* *Pregnancy-induced hypertension

mlvsis stratified l_)y_gestational age (Table 3)
Infants < 29 weeks gestation:

After stratifying the data by gestational age less than 29 weeks versus greater
than or equal to 29 weeks, prescription of bed rest and any antenatal steroid
administration were all significantly inversely related to WMD for the younger infants.
The risk of WMD was 2-fold for those infants who experienced vaginal delivery. Risk of
WMD was also non-significantly increased among African American infants,

Mixed/Other race infants, and where delivery was initiated by PPROM and PTL. There

20

was also a non-significant dose-response of an inverse association with WMD for 2-
categories below zero.
Infants 2 29 weeks gestation:

For the more mature infants, birthweight z-scores were non-Significantly
inversely associated with WMD. The risk of WMD was increased among infants greater
than or equal to 29 weeks gestation whose mothers had histopathologically-defined
chorioamnionitis but was not statistically significant. Hispanic or Mixed/Other race and

where delivery was initiated by other mechanisms than PIH had elevated associations

with WMD.

Table 3. Risk of white matter damage for 1607 VLBW infants stratified by gestational age < 29
weeks and Z 29 weeks, DEN data, 1991-93

 

 

 

 

Gestational age < 29 wks Gestational age 2 29 wks

Characteristic Risk of WMD fl Risk of WMD R_R

Female 43/415 ref 9/3 83 ref

Male 51/464 1.0 11/345 1.3
Race

White 35/381 ref 8/361 ref

Black 32/249 1.4 4/166 1.0

Hispanic 16/173 1.0 5/136 1.6

Other/Mixed 9/66 1.3 2/33 2.7
BIRTHWEIGHT

Z-scoreZ 1 13/133 0.7 0/9 NC

1 >Z-score 2 0 39/298 ref 4/81 ref

0 >Z-score Z -1 29/278 0.8 6/226 0.5

-1 >Z-score Z -2 11/121 0.7 7/187 0.8

Z-score< -2 2/49 0.3 3/225 0.3
DELIVERY CHARACTERISTICS
No Chorioamnionitis 24/255 ref 8/354 ref
Chorioamnionitis 46/3 83 1.2 7/154 2.0
Initiators of preterm delivery

PIH 7/96 ref 4/221 ref

21

Table 3 (cont’d)

 

Preterm labor
PPROM
Other

Cesarean Section
Vaginal Delivery

Presence of labor
No labor
Any Labor

Membrane Rupture < 1 hr.
Membrane Rupture Z 1 hr.

Prescribed bed rest
No
Yes

No antenatal steroids
Any antenatal steroids

47/446
37/297
3/36

36/492
58/3 87

16/160
78/719
40/411
53/464
47/348

38/451

62/435
32/444

ref
1.0

ref
1 .1
ref

0.6

ref
0.5

6/219
7/21 1
3/72

12/492
8/236

6/241
14/487
8/3 78
12/343
6/235

14/439

9/360
1 1/348

1.5
1.8
2.3

ref
1 .4

ref
1.]
ref
1.6
ref

1.2

ref
1.2

 

* Bold indicates confidence interval does not include 1; PPROM denotes preterm premature

rupture of membranes; PIH denotes pregnancy-induced hypertension; NC refers to not enough

subjects to complete calculation.

22

Table 4. Relative risk of white matter damage for 1146 VLBW infants with and without any
chorioamnionitis stratified by gestational age, DEN data, 1991-I993

 

 

Characteristic Chorioamnionitis No Chorioamnionitis
Gestation<29 Gestation 229 Gestation<29 Gestation ,_>,29
wks wks wks wks
Risk RR Risk RR Risk RR Risk RR
Male 25/187 1.2 3/77 0.7 11/140 0.7 4/167 1.1
Female 21/196 4/77 13/115 4/187
White 16/148 ref 2/61 ref 10/129 ref 5/182 ref
Black 20/138 1.2 3/51 1.7 4/47 1.0 0/69 0.3
Hispanic 8/56 1.2 1/26 1.1 5/55 1.1 2/71 1.0
Mixed/Other 2/30 0.5 1/12 2.5 4/21 2.4 1/24 1.5
Z-score _>_l 5/67 0.5 0/3 NC 4/37 0.6 0/3 NC
l> Z-score 20 19/139 ref 3/23 ref 14/77 ref 1/24 ref
0> Z-score 2-1 16/130 0.9 2/74 0.2 6/67 0.5 2/92 0.5
-1> Z-score 2-2 4/34 0.9 1/36 0.2 0/50 NC 4/99 1.0
Z-score <-2 2/13 1.1 1/18 0.4 0/24 NC 1/136 0.2
Preterm labor 25/202 ref 5/59 ref 1 1/1 16 ref 1/96 ref
PPROM 18/168 0.9 2/87 0.3 8/50 1.7 3/64 4.5
PIH 2/4 4.0 0/4 NC 3/63 0.5 2/153 1.2
Other 1/6 1.3 0/4 NC 2/24 0.9 2/39 4.9
Delivery mode
Cesarean 12/144 ref 1 / 73 ref 15/195 ref 7/272 ref
Vaginal 34/227 1.9 6/81 5.4 9/60 1.9 1/82 0.4
No labor 5/42 ref 0/17 ref 7/83 ref 4/154 ref
Any labor 41/341 1.0 7/137 NC 17/172 1.1 4/200 0.7
Antenatal
steroids
None 3 l/ l 98 ref 4/63 ref 17/126 ref 2/200 ref
Any 15/185 0.5 3/91 0.5 7/129 0.4 6/154 3.8

 

*Preterm premature rupture of membranes; “Pregnancy-induced hypertension; Bold font
signifies RR is statistically significant; NC refers to not enough subjects to complete calculation.

I_nfluepces of chorioamnionitis op risk of WMD
In table 4, the data were further stratified by presence and absence of
chorioamnionitis among the gestational age strata. For the infants with gestational ages

less than 29 weeks in the presence of chorioamnionitis, vaginal delivery was associated

23

with WMD (RR 1.9, 95% CI 1.0-2.6). Maternal receipt of partial or full antenatal
steroids was significantly inversely associated with WMD for all infants with a
gestational age less than 29 weeks, with and without chorioamnionitis (RR 0.5, 95% CI
0.3-0.9; RR 0.4, 95% CI 0.2-0.9, respectively). The risk of WMD for infants with
gestational ages less than 29 weeks in the absence of chorioamnionitis was lower for all
birthweight z-score categories except the largest category compared to those in the
presence of chorioamnionitis. There were no cases of WMD in the two smallest z-score
categories among the young infants where chorioamnionitis was absent. For infants with
gestational ages greater than or equal to 29 weeks in the presence of chorioamnionitis,
vaginal delivery was non-significantly associated with WMD. There were also very few
cases of WMD among the older infants, regardless of chorioamnionitis status.

Table 5 shows the relative risks for WMD among the independent variables
stratified by mode of delivery then each stratum stratified by gestational age at the cutoff
of 29 weeks. Only maternal receipt of antenatal steroids was significantly inversely
associated with WMD for the infants with gestational ages less than 29 weeks who were
delivered abdominally (RR 0.3, 95% CI 0.1-0.7). For the infants delivered vaginally, the
gestationally older infants of all races compared to Whites had much higher non-
significant associations with WMD. There was no association between WMD and
chorioamnionitis for the infants under 29 weeks gestation regardless of mode of delivery.
Although, the older infants who were delivered vaginally and whose mothers had
chorioamnionitis were at an increased risk of WMD (RR 1.8; 95% CI 1.3-2.5), while the
older infants delivery by Cesarean were at a reduced risk (RR 0.6; 95% CI 0.1-3.7).

Because of the stratification in univariate analysis reduced the number of subjects in the

24

cells, it was difficult to obtain statistical significance for many of the independent

variables.

Table 5. Relative risk of white matter damage for 1607 VLBW infants with and without vaginal

delivery stratified by gestational age, DEN data, 1991-1993

 

Characteristic

Vaginal Deliver

Cesarean Delivery

 

Gestation < 29 Gestation Z 29 Gestation < 29 Gestation Z 29
wks wks wks wks
Risk RR Risk RR Risk RR Risk RR
Female 25/183 ref 4/1 17 ref 18/232 ref 5/266 ref
Male 33/204 1.2 4/1 19 1.0 18/260 0.9 7/226 1.7
White 20/140 ref 1/92 ref 15/241 ref 7/269 ref
Black 24/ 1 39 1.2 4/73 5.0 8/109 1.2 0/93 NC
Hispanic 7/71 0.7 2/51 3.6 9/102 1.4 3/85 1.4
Mixed/Other 6/30 1.4 1/16 5.8 3/36 1.3 2/34 2.3
Z-scorez 1 6/56 0.6 0/5 NC 7/77 1.0 0/4 NC
l>Z-score_>_ 0 25/139 ref 2/34 ref 14/159 ref 2/47 ref
0>Z~score2 -1 16/132 0.7 3/94 0.5 13/146 1.0 3/132 0.5
-1>Z-score2 -2 9/44 1.1 2/50 0.7 2/77 0.3 5/137 0.9
Z-score< -2 2/16 0.7 1/53 0.3 0/33 NC 2/172 0.3
No chorioamnion. 9/60 ref 1/82 ref 15/195 ref 7/272 ref
Chorioamnionitis 34/227 1.0 6/81 1.8 12/156 1.0 1/73 0.6
Initiator of
delivery
Preterm labor 33/231 ref 4/ 106 ref 14/215 ref 2/1 13 ref
PPROM* 24/141 0.7 4/97 1.1 13/156 1.3 3/114 1.5
P1H* 1/10 0.7 0/27 NC 6/85 1.1 4/198 1.1
Other 0/2 NC 0/5 NC 3/34 1.4 3/67 2 5
No steroids 36/205 ref 5/124 ref 26/230 ref 4/245 ref
Any steroids 22/182 0.7 3/112 0.7 10/262 0.3 8/247 2.0

 

*Preterm premature rupture of membranes; pregnancy-induced hypertension
Bold indicated confidence interval does not include 1; NC refers to not enough subjects to
complete calculation

In the unadjusted logistic regression model of infants less than 29 weeks

gestation, birthweight z-score categories were reduced to three categories where SGA

25

infants were those whose z-score was 5 -0.90 in order to obtain at least one case of WMD
in the stratum without chorioamnionitis. SGA infants without chorioamnionitis were
significantly protected from WMD (OR 0.1; 95% CI 0.01-0.6) while SGA infants with
chorioamnionitis had a null association with WMD (OR 0.9; 95% CI: 0.3-2.2). The AGA

and large infants showed no difference between chorioamnionitis status.

Table 6. Unadjusted ORS and 95% CI for the effect of birthweight z-scores in the
presence and absence of chorioamnionitis on white matter damage in 638 VLBW infants
< 29 weeks gestation, DEN Data, 1991-93

 

 

Variable OR 95% CI
No Chorioamnionitis
SGA 0.1 (0.01, 0.6)
AGA 1.0 ref
Large 0.7 (0.2, 2.3)
Chorioamnionitis
SGA 0.9 (0.3, 2.2)
AGA 0.9 (0.5, 1.7)
Large 0.7 (0.2, 2.0)

 

SGA is birthweight z-score $0.90; AGA is birthweight z-score >-0.90 and S 1; Large is birthweight z-
score 2 1

When adjustment was made for vaginal delivery and partial or full receipt of antenatal
steroids, SGA infants with chorioamnionitis became less associated with WMD (0.6;
95% CI 0.2-1.6). For the SGA infants without chorioamnionitis, the odds ratio was

remained unchanged (OR 0.1; 95% CI 0.01- 0.6).

26

Table 7. Adjusted ORS and 95% CI for the effect of birthweight z-scores in the presence
and absence of chorioamnionitis on white matter damage in 638 VLBW infants <29
weeks gestation, DEN Data, 1991-93

 

 

Variable OR 95% CI
No Chorioamnionitis
SGA 0.1 (0.01, 0.6)
AGA 1.0 ref
Large 0.4 (0.1, 1.1)
Chorioamnionitis
SGA 0.6 (0.2, 1.6)
AGA 0.7 (0.4, 1.4)
Large 0.4 (0.1, 1.1)
Vaginal delivery 1.9 (1.1, 3.4)

 

Partial/full receipt of AC8 0.4 (0.3,0.8)

 

SGA is birthweight z-score S-0.90; AGA is birthweight z-score >-0.90 and S 1; Large is birthweight z-
score 2 1; ACS is antenatal corticosteroids

27

CHAPTER 4

DISCUSSION

This study evaluated the interaction between maternal and obstetric factors,
focusing especially on fetal growth, with chorioamnionitis on the risk of WMD:
echolucencies, echodensities and ventriculomegaly in very low birthweight infants.
Several factors were found to be associated with WMD. In the overall sample, the risk of
WMD was increased among infants whose gestational age was less than 29 weeks (RR
3.8; 95% CI 2.4-6.2), had histopathologically-defined chorioamnionitis (RR 1.8; 95% CI
1.2-2.8) and experienced vaginal delivery (RR 2.1; 95% CI 1.5-3.4). Again, in the
overall sample, the lowest birthweight z-score category was significantly protective from
WMD (RR 0.1; 95% CI 0.06-0.4). When stratified by status of chorioamnionitis, the risk
of WMD was clearly modified by birthweight z-score for the infants with gestational
ages less than 29 weeks. Most notably, there were no cases of WMD among the infants
less than 29 weeks gestation who were not exposed to chorioamnionitis. The protective
effect from the absence of chorioamnionitis only existed in the SGA infants (OR 0.1;
95% Cl 0.01-0.8) and when controlling for potential confounders, that relationship held
significance.

This evidence of a protective effect of growth restriction on WMD supports the
work of Inder et al and Baud et a1 where SGA was shown to be inversely related to WMD
after adjustment for potential confounders (OR 0.58, 95% CI 0.25-1.15 and OR 0.08,
95% CI 0.02-0.41).“84 Oliver et al’s rat models showed the opposite effect, but did

Show that mild growth restriction was protective from more severe hypoxic-ischemic

28

insults and induced exitotoxic brain lesions.85 However, these results have severe
limitations. Hypoxia-ischemia was induced by a single event with an invasive technique
of uterine artery ligation to produce growth restricted rat pups, a circumstance that may
not be comparable to the situation in human infants where growth restriction may be
caused by more subtle, long-term, and possibly intermittent insults to the developing
human fetus. The statistical analysis was only univariate and potential confounders were
not controlled for. These results did provide evidence that hypoxic preconditioning may
play a role in cellular and molecular mechanisms in cell survival.

In human studies assessing WMD in preterm infants that also considered IUGR as
a risk factor, one study found no association of IUGR with WMD and the other found an
increased risk of WMD in SGA infants.15 ‘ 36 The study that found a null association was
a large, population-based cohort that adjusted only for gestational age. The latter study
did not adjust for any confounders in the analysis and did not have enough statistical
power to make meaningful comparisons. It also included infants up to 34 weeks
gestation, which are not typically considered very preterm infants.

Qui et al’s study that found SGA infants in the presence of infection increased the
risk of PEL/V E was most similar to the present study.69 Qui et al’s definition classified
cases of maternal infection only where clinical presentation of symptoms was obvious,
which would likely lead to misclassifying the more mild and sub-clinical cases of
infection as non-cases. This may explain why the cohort of preterm infants (less than 35
weeks gestation) had amnionitis in only 5% of the vaginal deliveries and 5% in the
Cesarean section deliveries. Stratification by SGA status in the multivariate model

subsequently led to an OR with an extremely wide confidence interval due to such a

29

small number of cases. Since the DEN data used histopathologic determination of
maternal infection, the null association seen among SGA infants in the presence of
chorioamnionitis may be a result of a bias created by including mild cases of
chorioamnionitis that may not increase the risk of WMD with the moderate and severe
cases of infection that may have more potential to increase the risk of developing WMD.
However, the infants not exposed to chorioamnionitis are truly unexposed and that
provides greater confidence in the inverse association with WMD seen in the SGA
infants.

The issue of defining chorioamnionitis is problematic when comparing studies
and, as mentioned above, strict definitions such as in this study may result in weak or null
associations which many studies have reported. Weak associations could also be
attributed in part to the causal chain of risk factors. Fetal vasculitis follows
chorioamnionitis and, therefore, should have a stronger association with WMD than
chorioamnionitis because it is closer to WMD in the causal pathway, as Paneth and
Leviton have shown.23

The strengths of the DEN data are the prospective collection of ultrasonic brain
scans, full agreement for diagnosis of white matter abnormalities in 92% of cases and
histopathologic determination of chorioamnionitis.

Future studies should make a closer examination of the determination of
gestational age, especially in the small fraction of infants where the determination was
made by the physician at birth. In very preterm infants, it is likely that the physician
could underestimate the gestational age by several weeks and misclassify some of the

older, more mature infants who have shown in other studies to exhibit some protection

30

from WMD. Also, the DEN data included 350 infants from a twin pregnancy and 100
infants from a triplet, quadruplet, or quintuplet pregnancy. Hierarchical modeling should
be considered to take into account multiple births nested within the same mother. Also, a
more appropriate growth standard should be used in calculation of z-scores. Yudkin’s
growth standards are sex-specific from 20 years ago out of Oxford, England where 96%
of the infants are white.82 The subjects in the DEN data are very diverse with only 47%
white and a more recent US. growth standard could arguably be used. Finally, in the
model of infants with a gestational age less than 29 weeks, 27% had missing placental
information that excluded them from analysis. This significant proportion of the sample
could affect the results, but inferences on how cannot be made.

All in all, the SGA infants in this analysis showed an inverse association with
WMD, while vaginal delivery persisted as a significant risk factor. Hansen et a1 had
similar findings with vaginal delivery as did Qiu et al.68‘ 69 Distinctive features of this
subgroup of SGA infants should be investigated and should help inform the scientific

community at large and clinicians in their decision-making with obstetric care.

31

REFERENCES

 

' Savitz DA, Dole N, Herring AH, et al; Should spontaneous and medically indicated
preterm births be separated for studying aetiology? Pediatric and Perinatal Epidemiology
2005; 19:97-105.

2 Creasy, Robert and Resnik. Matemal-Fetal Medicine. 4‘h Edition. Philadelphia: W.B.
Saunders, 1999.

3 Aprino C, D’Argenzio L, Ticconi C, et al; Brain damage in preterm infants: etiologic
pathways. Ann 1St Super Sanita 2005; 41 (2): 229-37.

 

" Abbrescia K and Sheridan B; Complications of second and third trimester pregnancies.
Emergency Medicine Clinics of North America 2003; 21: 695-710.

5 Martin JA, Hamilton, BE Sutton PD, et a1. Births: Final data for 2004. National Vital
Statistics Report 2006; 55: 1-104.

" Holzman C, Paneth N and Fisher R; Rethinking the concept of risk factors for preterm
delivery: antecedents, markers and mediators. Prenat Neonat Med 1998; 3: 47-52.

7 QiuckStat: Percentage of total births that were preterm, by gestational age—United
States, 1990 and 2005. MMWR 2007; 56(2): 33.

8 Hamilton BE, Martin JA, and Ventura SJ; Briths: Preliminary data for 2005. National
Vital Statistics Report 2006; 55 (11): 1-18.

 

9 Augustines LA, Lin YG, Rumney PJ, et al; Outcomes of extremely low-birthweight
infants between 500 and 750g. American Journ_al of Obstetrics and Gynecology 2000;
182: 1113-6.

’0 Finan A, Ledwidge M, Clarke T, et al; Perinatal factors influencing survival in
extremely low-birthweight infants. Journal of Obstetrics and Gynecology 1998; 18(3):
227-30.

H Wilson-Costello D, Friedman H, Minich N, et al; Improved survival rates with
increased neurodevelopmental disability for extremely low birth weight infants in the
19903. Pediatrics 2005; 115(4): 997-1003.

'2 Stoelhorst G, Rijken M, Martens SE, et al; Changes in neonatology: Comparison of two
cohorts of very preterm infants (gestational age <32 weeks): The project on preterm and
small for gestational age infants 1983 and the Leiden follow-up project on prematurity
1996-1997. Pediatrics 2005; 115(2): 396-404.

32

 

'3 Stoll B, Hansen N, Adams-Chapman I, et al; Neurodevelopmental and growth
impairment among extremely-low-birth-weight infants with neonatal infection. JAMA
2004; 292: 2357-2365.

"’ Lau J, Magee F, Qiu Z, et al; Chorioamnionitis with a fetal inflammatory response if
associated with higher neonatal mortality, morbidity, and resource use than

chorioamnionitis displaying a maternal response only. American Journal of Obstetrics
and Gynecology 2005; 193: 708-13.

'5 Larroque et al; White matter damage and intraventricular hemorrhage in very preterm
infants: the EPIPAGE study. The Journal of Pediatrics 2003; 143: 477-83.

‘6 Wilson-Costello D, Borawski E, Friedman H, et al; Perinatal correlates of cerebral
palsy and other neurologic impairment among very low birth weight children. Pediatrics
1998; 102: 315-22.

'7 Whitaker AH, Van Rossem R, Feldman J F, et al; Psychiatric outcomes in low-birth-
weight children at age 6 years: relation to neonatal cranial ultrasound abnormalities. Arch
Gen Psychiatry 1997; 54(9): 785-9.

‘8 Walther FJ, den Ouden AL, Verloove-Vanhorick SP; Looking back in time: outcome
of a national cohort of very preterm infants born in the Netherlands in 1983. Early Hum
w 2000;59:175 —191.

'9 Back SA, Riddle A and McClure MM; Maturation-dependent vulnerability of perinatal
white matter in premature birth. Stroke 2007; 38 (part 2): 724-30.

20 Darnmann O, Kuban K and Leviton A; Perinatal infection, fetal inflammatory
response, white matter damage, and cognitive limitations in children born preterm.
Mental Retardation and Developmental Disabilities Research Reviews 2002; 8: 46-50.

21 Wu YW, Croen LA, Shah SJ, et al; Cerebral palsy in a term population: risk factors
and neuroimaging findings. Pediatrics 2006; 118 (2): 690-696.

2"" Ramsey P et al; Chorioamnionitis increases neonatal morbidity in pregnancies
complicated by preterm premature rupture of membranes. American Journal of
Obstetrics and Gynecology 2005; 192: 1162-6.

23 Leviton A, Paneth N, Reuss L, et al; Maternal infection, fetal inflammatory response,
and brain damage in very low birth weight infants. Pediatric Research 1999; 46(5): 566-
74.

2‘ Dammann O and Leviton A; Inflammatory brain damage in preterm newboms-dry
numbers, wet lab, and causal inference. Early Human Development 2004; 79: 1-15

33

 

2’ Kinney HC;The near-term (late preterm) human brainand risk for periventricular
leukomalacia: a review. Seminars in Perin_atalogy 2006; 30 (2): 81-8.

2" Dammann O, Leviton A, Gappa M, et al; Lung and brain damage in preterm newborns,
and their association with gestational age, prematurity subgroup, infection/inflammation
and long term outcome. BJOG 2005; 112 (1): 1-4.

2’ Volpe JJ. Encephalopathy of prematurity includes neuronal abnormalities. Pediatrics
2005; 116: 221-225.

2" Silveira RC and Procianoy; Ischemic brain damage in very low birth weight preterm
newborn infants. Jornal de Pediatria 2005; 81 (1): 823-32.

2" Dambska M, Laure-Kamionowska M and Schmidt-Sidor B; Early and late
neuropathological changes in perinatal white matter damage. Journal of Child Neurology
1989; 4: 291-8.

3" Leviton A and Gilles F; Ventriculomegaly, delayed myelination, white matter
hypoplasia, and “periventricular” leukomalacia: how are they related? Pediatric
Neurology 1996; 15: 127-36

3' Bernstein IM, Horbar JD, Badger GJ, et al; Morbidity and mortality among very low
birth weight neonates with intrauterine growth restriction. The Vermont Oxford Network.
American Journal of Obstetrics and Gynecology 2000; 182 (1): 198-206.

’2 Fanaroff AA, Wright LL, Stevenson DK, et al; Very-low-birth-weight outcomes of the
National Institute of Child and Human Development Neonatal Research Network, May
1991 through December 1992. American Journal of Obstetrics and Gynecology 1995;
173: 1423-31.

3’ Costantine MM, How HY, Coppage K, et al; Does peripartum infection increase the
incidence of cerebral palsy in extremely low birthweight infants? American Joml of

Obstetrics and Gynecology 2007; 196 (5): e68.

3" Tioseco JA, Aly H, Essers J, et a1; Male sex and intraventricular hemorrhage. Pediatric
Critical Care Medicine 2006; 7 (1): 40-4.

3’ Jarvis S, Glinianaia SV, Amaud C, et al; Case gender and severity in cerebral palsy
varies with intrauterine growth. Archives of Disease in Childhood 2005; 90 (5): 474-9.

3" Johnston MV and Hagberg H; Sex and the pathogenesis of cerebral palsy.
Developmental Medicine and Child Neurology 2007; 49: 74-9.

’7 Basso O, Wilcox A and Weinberg C; Birthweight and mortality: causality or
confounding? American Journal of Epidemiology 2006; 164: 303-11

34

 

’3 Edwards RK; Chorioamnionitis and labor. Obstetrics and Gynecology Clinics 2005;
32(2): 287-96.

’9 Ustun C et al; Subclinical chorioamnionitis as an etiologic factor in preterm deliveries.
International Journal of Gynecology and Obstetrics 2001; 72 (2): 109-115.

4" De Felice C, Goldstein M, Parrini S, et a1; Early dynamic changes in pulse oximetry
signals in preterm newborns with histologic chorioamnionitis. Pediatric Critical Care
Medicine 2006; 7(2): 138-142

4' Versland LB, Sommerfelt K and Elgen I; Maternal signs of chorioamnionitis:
persistent cognitive impairment in low-birthweight children. Acta Paediatrica 2006; 95:
231-5.

’2 Sommerfelt K, Markestad T and Ellertsen B; Neuropsychological performance in low
birth weight preschoolers: a population-based, controlled study. European Joum_al of
Pediatrics 1998; 157(1): 53-8.

4’ Polam S, Koons A, Anwar M, et al; Effect of chorioamnionitis on neurodevelopmental
outcome in preterm infants. Arch Pediatr Adolesc Med 2005; 159: 1032-35.

 

4" Wu YW; Systematic review of chorioamnionitis and cerebral palsy. Mental
Retardation and Developmental Disabilites Research Reviews 2002; 8: 25-9.

4’ Kramer M, Platt R, Yang H, et al; Are all growth-restricted newborns created equally?
Pediatrics 1999; 103(3): 599-602.

‘6 Sizonenko S, Borradori-Tolsa C, Vauthay D, et al; Impact of intrauterine growth
restriction and glucocorticoids on brain development: insights using advanced magnetic
resonance imaging. Molecular and Cellular Endocrinology 2006; 254-255: 163-171.

’7 Baud O, Zupan V, Lacaze-Masmonteil T, et al; The relationships between antenatal
management, the cause of delivery and neonatal outcome in a large cohort of very
preterm singleton infants. BJOG 2000; 107: 877-884.

‘8 Gilbert WM and Danielson B; Pregnancy outcomes associated with intrauterine growth
restriction. American Journal of Obstetrics and Gynecology 2003; 188: 1596-1601.

’9 Garite JT, Clark R andThorp JA; Intrauterine growth restriction increases morbidity
and mortality among preterm neonates. American Journal of Obstetrics and Gynecology
2004; 191 (2): 481-7.

5° Simon NV, Hohman WA, Elser RC, et al; Fetal lung maturity in complicated

pregnancy,as predicted from microviscosity of amniotic fluid. Clinical Chemistry 1982;
28 (8): 1754-7.

35

 

 

5' Jain L; Effect of pregnancy-induced and chronic hypertension on pregnancy outcome.
Journal of Perinatology 1997; 17 (6): 425-7.

’2 Lydakis C, Beevers M, Beevers DG, et al; The prevalence of pre-eclampsia and
obstetric outcome in pregnancies of normotensive and hypertensive women attending a
hospital specialist clinic. Internatiognal Journal of Clinical Practice 2001; 55 (6): 361-7.

’3 Chen XK, Wen SW, Smith G, et a1; Pregnancy-induced hypertension is associated with
lower infant mortality in preterm singletons. BJOG 2006; 113: 544-51.

5" Friedman SA, Schiff E, Kao Lu, et a1; Neonatal outcome after preterm delivery for
preeclampsia. American Journal of Obstetrics and Gynecology 1995; 172: 1785-92.

’5 Spinillo A, Gardella B, Preti E, et a1; Preeclampsia and brain damage among preterm
infants: A changed panorama in a 20-year analysis. American Journ_al of Perin_atology
2007; 24 (2): 101-6.

’6 Murata Y, Itakura A, Matsuzawa K, et al; Possible antenatal and perinatal related

factors in development of cystic periventricular leukomalacia. Brain and Development
2005; 17-21.

’7 Collins M and Paneth N; Preeclampsia and cerebral palsy: are they related?
Developmental Medicine and Childhood Neurology 1998; 40: 207-11.

’8 Ananath CV, Demissie K, Smulian JC, et al; Relationship among placenta previa, fetal
growth restriction, and preterm delivery: A population-based study. Obstetrics and
Gynecology 2001; 98: 299-306.

 

’9 Goldenberg RL, Iams JD, Mercer BM, et al; The Preterm Pridiction Study: The value
of new vs. standard risk factors in predicting early and all spontaneous preterm births.
American Journal of Public Health 1998; 88 (2): 233-8.

 

6° Hillier SL, Nugent RP, Eschenbach DA, et al; Association between bacterial vaginosis
and preterm delivery of a low-birth-weight infant. New England Journal of Medicine
1995; 333: 1737-42.

6' Svare JA, Schmidt H, Hansen BB and Lose G; Bacterial vagoinosis in a cohort of
Danish pregnant women: Prevalence and relationship with preterm delivery, low
birthweight and perinatal infections. BJOG 2006; 133: 1419-25.

’2 Iams JD, Goldenberg RL, Mercer BM, et al; The Preterm Pridiction Study: Can low-

risk women destined for preterm birth be identified? American Journal of Obstetrics and
Gynecology 2001; 184: 652-5.

36

 

6’ Martin J A and Menacker F; Expanded health data from the new birth certificate, 2004.
National Vital Statistics Report 2007; 55 (12).

6“ Dolea C and Abouzahr C; Global burden of hypertensive disorders of pregnancy in the
year 2000. World Health Organization 2003: Geneva.

(’5 Villar J, Carroli G, Wojdyla D, et al; Preeclampsia, gestational hypertension, and
intrauterine growth restriction, related or independent conditions? American Journal of
Obstetrics and Gynecology 2006; 194: 921-31.

 

6" Low J A; Determining the contribution of asphyxia to brain damage in the neonate.
Journal of Obstetric and Gynaecologic Research 2004; 30 (4): 276-86.

’7 Johnston MV, Trescher WH, Ishida A, et al; Neurobiology of hypoxic-ischemic injury
in the developing brain. Pediatric Research 2001; 49 (6): 735-41.

6" Hansen A and Leviton A; Labor and delivery characteristics and risks of cranial
ultrasonographic abnormalities among very-low-birth-weight infants. American Journal
of Obstetrics and Gynecology 1999; 181 (4): 997-1006.

6’ Qiu H, Paneth N, Lorenz J and Collins M; Labor and delivery factors in brain damage,
disabling cerebral palsy, and neonatal death in low-birth-weight infants. American
Journal of Obstetrics and Gynecology 2003; 189 (4): 1143-49.

7° Locatelli A, Vergani P, Ghidini A, et al; Duration of labor and risk of cerebral white-
matter damage in very preterm infants who are delivered with intrauterine infection.
American Journal of Obstetrics and Gynecology 2005; 193: 928-32.

7' Wilson-Costello D, Borawski E, Friedman H, et al; Perinatal correlates of cerebral
palsy and other neurologic impairment among very low birth weight children. Pediatrics
1998; 102: 315-22.

’2 Uehara H, Yoshioka H, Kawasa S, et al; A new model of white matter injury in
neonatal rats with bilateral carotid artery occlusion. Brain Research 1999; 837 (1-2):
213-20.

’3 Riddle A, Luo NL, Manese M, et al; Spatial heterogeneity in oligodendrocyte lineage
maturation and not cerebral blood flow predicts fetal ovine periventricular white matter
injury. The Journal of Neuroscience 2006; 26 (11): 3045-55.

7" Back SA, Luo NL, Borenstein NS, et al; Late oligodendrocyte progenitors coincide
with the developmental window of vulnerability for human perinatal white matter injury.
The Journal of Neuroscience 2001; 21 (4): 1302-12.

37

 

7’ Traystman RJ, Kirsch JR and Koelher RC; Oxygen radical mechanisms of brain injury
following ischemia and reprofusion. Journal of Applied Physiology 1991; 71 (4): 1185-
95.

7" Stallmach T and Hebisch G; Placental pathology: its impact on explaining prenatal and
perinatal death. Virchows Arch 2004; 445: 9-16.

’7 Andrews W, Goldenberg R, Faye-Petersen O, et al; The Alabama preterm birth study:
polymorphonuclear and mononuclear cell placental infiltrations, other markers of

inflammation, and outcomes in 23-to 32-week preterm newborn infants. American
Journal of Obstetrics and Gynecology 2006; 195: 803-8.

’8 Murtha A, Sinclair T, Hauser E, et a1; Maternal serum cytokines in preterm premature
rupture of membranes. Obstetrics and Gynecology 2007; 109 (1): 121-27.

7’ Aaltonen R, Heikkinen T, Hakala K, et a1; Transfer of proinflammatory cytokines
across term placenta. Obstetrics and Gynecology 2005; 106 (4): 802-7.

8° Willoughby RE, Jr. and Nelson KB; Chorioamnionitis and brain injury. Clinics in
Perinatology 2002; 29: 603-21.

8' Kuban KC, Allred EN, Dammann O, et al; Topography of cerebral white-matter disease
of prematurity studied prospectively in 1607 very-low-birthweight infants. Journal of
Child Neurology 2001; 16: 401-8.

’2 Yudkin PL, Aboualfa M, Eyre J A, et al; Influence of elective and preterm delivery on
birthweight and head circumference standards. Archives of Disease and Childhood 1987;
62: 24-9.

8’ Inder TE, Wells SJ, Mogridge NB, et al; Defining the nature of the cerebral
abnormalities in the premature infant: a quantitative magnetic resonance imaging study.
Joumal of Pediatrics 2003; 143: 171-9.

8" Baud O, Zupan V, Lacaze-Masmonteil T, et al; The relationships between antenatal
management, the cause of delivery and neonatal outcome in a large cohort of very
preterm singleton infants. BJOG 2000; 107: 877-84.

8’ Oliver P, Baud O, Bouslama M, et al; Moderate growth restriction: deleterious and
protective effects on white matter damage. Neurobiology of Disease 2007; 26 (1): 253-
63.

8" Padilla-Gomez NF, Enriquez G, Acosta-Rojas R, et al; Prevalence of neonatal
ultrasound brain lesions in premature infants with and without intrauterine growth
restriction. Acta Paediatrica 2007; 96 (11): 1582-7.

38

IBHARIF .5;

l I

III

R

l

l

l

j .
l

l

l

7
2
2
5
6
l5
fl. II9
2
O
3
9
2
1
3

 

 

 

 

. i l
c . “Ml
. U l .1 IN;
.. Ell
D rlll Ill III-v
. Am will
I. 1'
V . \Jl Ill
ll
~ N IIIl
.v . Mill
.. HI
. . Ci
.
l
. . Ill 1
.
a
I
o
u
.
v
.
II
. .
. .
.
O
..
o.
to
I.
M. 0 O a I
. o
.
.-
o
.
.
a
u
r a
a .
O
.
O
.
..
. .-
.
a . .
z. . .1.
. ..
.. a
a .
c ..
I .
,
I . . . .
. . . .
v ..- . .
I .
o .
.
a
'
i . u.
I .
.
u. .
c . . . .
I
'
an
o
I n o
C
n. . .
I
‘ 0
.0
9 . .
.. .
. .
n .-
. .. .

 

 

;
.u.‘o
\o-.-‘It