Original Articles
Periventricular Hemorrhagic Infarction: Risk
Factors and Neonatal Outcome
Haim Bassan, MD*, Henry A. Feldman, PhD†, Catherine Limperopoulos, PhD*,
Carol B. Benson, MD‡, Steven A. Ringer, MD§, Elaine Veracruz, BS*, Janet S. Soul, MD*,
Joseph J. Volpe, MD*, and Adré J. du Plessis, MBChB, MPH*
The aim of this study was to define the incidence,
clinical associations, and short-term outcome of
periventricular hemorrhagic infarction in the modern
neonatal intensive care unit. From 5774 infants (birth
weight <2500 gm), periventricular hemorrhagic in-
farction diagnosed by cranial ultrasound was identified
and confirmed. gestational age-matched control infants
were identified with normal cranial ultrasounds and
detailed clinical data were obtained in both groups.
Periventricular hemorrhagic infarction was confirmed in
58 infants. Incidence was 0.1% (1500-2500 gm), 2.2%
(750-1500 gm), and 10% (<750 gm). Data across 6 study
years reveal increased incidence in infants <750 gm.
Compared with control infants, infants with periventricu-
lar hemorrhagic infarction had significantly greater as-
sociation with assisted conception, intrapartum factors
(emergency cesarean section, low Apgar scores), early
neonatal complications (patent ductus arteriosus, pneu-
mothorax, pulmonary hemorrhage), blood gas distur-
bances, and need for pressor, volume infusion, and respi-
ratory support. Neonatal mortality of this group was
40% (n ⴝ 23). Survivors had longer duration of mechan-
ical ventilation and critical care stay compared with
control subjects. Thirty-seven percent of survivors re-
quired cerebrospinal fluid shunt placement. Periventricu-
lar hemorrhagic infarction remains an important neuro-
logic complication of prematurity. A growing population
of survivors is apparent among infants with birth weight
<750 gm. Multiple hemodynamic factors associated with
periventricular hemorrhagic infarction cluster in the
intrapartum and early neonatal periods. © 2006 by
Elsevier Inc. All rights reserved.
From *Department of Neurology, Neonatal Neurology Research
Group, Children’s Hospital Boston and Harvard Medical School;
†
Clinical Research Program, Children’s Hospital Boston and Harvard
Medical School; ‡Department of Radiology, Brigham and Women’s
Hospital and Harvard Medical School; §Department of Neonatology,
Brigham and Women’s Hospital and Harvard Medical School, Boston
Massachusetts.
© 2006 by Elsevier Inc. All rights reserved.
doi:10.1016/j.pediatrneurol.2006.03.005 ● 0887-8994/06/$—see front matter
Bassan H, Feldman HA, Limperopoulos C, Benson CB,
Ringer SA, Veracruz E, Soul JS, Volpe JJ, du Plessis AJ.
Periventricular hemorrhagic infarction: Risk factors and neo-
natal outcome. Pediatr Neurol 2006;35:85-92.
Introduction
Periventricular hemorrhagic infarction is a serious com-
plication of germinal matrix-intraventricular hemorrhage,
with adverse neurodevelopmental sequelae approaching
90% in earlier reports [1]. The impact of this lesion is best
appreciated when one considers that approximately 20%
of the 55,000 premature infants born with birth weight
⬍1500 gm in the United States each year develop germi-
nal matrix-intraventricular hemorrhage [2]. Periventricular
hemorrhagic infarction is usually accompanied by a large
germinal matrix-intraventricular hemorrhage and was pre-
viously viewed as a simple “extension” of intraventricular
blood into the parenchyma, hence its earlier designation as
grade IV germinal matrix-intraventricular hemorrhage.
Our current understanding is that periventricular hemor-
rhagic infarction is a venous hemorrhagic infarct in the
drainage area of the periventricular terminal vein [2].
Prior reports addressing periventricular hemorrhagic
infarction as an extension of germinal matrix-intraventric-
ular hemorrhage have had two important limitations.
Firstly, data from infants with grade III germinal matrix-
intraventricular hemorrhage (without parenchymal in-
volvement) often were combined with data from infants
who had periventricular hemorrhagic infarction into de-
scriptions of a single entity, usually “severe germinal
Communications should be addressed to:
Dr. du Plessis; Department of Neurology, Fegan 11; Children’s
Hospital; 300 Longwood Avenue; Boston, MA 02115.
E-mail: adre.duplessis@childrens.harvard.edu
Received June 21, 2005; accepted March 7, 2006.
Bassan et al: PVHI in Preterm Infants 85
matrix-intraventricular hemorrhage” [3-5]. This clustering
occurred despite the two entities’ differing pathophysiol-
ogies, periventricular hemorrhagic infarction being a com-
plication of germinal matrix-intraventricular hemorrhage
and clearly differing in outcome; e.g., neurologic morbid-
ity approximates 35% in grade III germinal matrix-intra-
ventricular hemorrhage but can reach 90% in periventricu-
lar hemorrhagic infarction [2,6]. Secondly, many of these
previous studies did not adequately control for gestational
age, thus precluding an accurate understanding of the risk
factors associated with periventricular hemorrhagic infarc-
tion.
In addition, with advances in modern neonatal intensive
care such as the advent of surfactant, antenatal steroids,
and new ventilatory support techniques, there has been a
concurrent decrease in the overall incidence of germinal
matrix-intraventricular hemorrhage and an increase in the
survival rate of small premature newborns [2]. The effect
of these trends on the incidence, risk factors, and outcome
of periventricular hemorrhagic infarction in premature
infants in the modern neonatal intensive care unit has not
been defined clearly. For these reasons, the aim of the
present study was to define further the current character-
istics of periventricular hemorrhagic infarction. The goals
were (1) to determine the current incidence of periven-
tricular hemorrhagic infarction diagnosed by neonatal
cranial ultrasound in a modern neonatal intensive care
unit, (2) to identify and characterize perinatal factors of
periventricular hemorrhagic infarction in the current era,
and (3) to determine the short-term outcome of infants
with periventricular hemorrhagic infarction.
Study Design
In this retrospective, case-controlled study, an elec-
tronic database search of all neonatal cranial ultrasound
reports at Brigham and Women’s Hospital from January
1997 through December 2002 was performed, using as key
words periventricular hemorrhagic infarction, grade IV-
IVH, parenchymal hemorrhage/echodensity, venous in-
farction, and hemorrhagic infarction. For each identified
candidate, all neonatal cranial ultrasound studies were
reviewed simultaneously by two of the co-investigators
(C.B. and H.B.), both of whom were unaware of the
clinical course and outcome of these infants. In five cases
of disparity between these two reviewers, a third reviewer
(A.J.d.P.) was used as a tie-breaker. Inclusion criteria were
all preterm infants ⬍2500 gm admitted to the neonatal
intensive care unit over the 6-year study period with the
diagnosis of periventricular hemorrhagic infarction (see
diagnostic criteria below). Exclusion criteria were term
infants and preterm infants with known or suspected brain
malformation, dysmorphic features or congenital anoma-
lies suggestive of a genetic syndrome, metabolic disorders,
chromosomal abnormality, or confirmed central nervous
system infection within the first 6 days of life. The control
group was defined as infants with normal neonatal cranial
86 PEDIATRIC NEUROLOGY Vol. 35 No. 2
ultrasound throughout their neonatal intensive care unit
stay. We then matched to each study infant one control
infant according to gestational age at birth (⫾4 days) and
institution of birth. Gestational age was defined according
to the first day of the mother’s last normal menstrual
period. This study was approved by the Institutional
Review Board of Brigham and Women’s Hospital, Bos-
ton, Massachusetts.
Ultrasonographic Criteria
The cranial ultrasound protocol in our neonatal inten-
sive care unit remained consistent over the study period,
with a range of 1-4 studies performed during the first week
of life. Periventricular hemorrhagic infarction was defined
as an echodense lesion in the periventricular white matter
which is unilateral or, if bilateral, obviously asymmetric,
and associated with a germinal matrix-intraventricular
hemorrhage lesion which is usually ipsilateral or larger on
the ipsilateral side [2,7-9]. Cases that met criteria for
isolated periventricular leukomalacia (i.e., bilateral, usu-
ally symmetric echodensities in the white matter dorsolat-
eral to the lateral ventricles and sometimes but not
invariably associated with evolving germinal matrix-intra-
ventricular hemorrhage) were excluded [2]. Also excluded
were cases with transient echodensities (transient features
of periventricular hemorrhagic infarction) that resolved
without cystic residua and cases of congenital porence-
phalic cyst (periventricular cysts present within the first
days of life), owing to uncertainty regarding the precise
etiologic nature of these pathologies.
Clinical Data Collection
From the clinical records of all identified infants,
specific demographic, maternal and antenatal, labor and
intrapartum, and early postnatal data from the first 5 days
of life were abstracted. Data were collected from the
placental pathology reports for the presence of chorioam-
nionitis, placental abruption, or placental infarction. Intra-
uterine growth retardation was defined as birth weight
under the 10th percentile for gestational age. Fetal heart
rate abnormalities were defined as sustained fetal brady-
cardia ⬍100 beats/minute, decreased variability, and late
decelerations as noted in maternal records. Maternal fever
was defined as temperature ⬎38oC within 72 hours before
delivery. Patent ductus arteriosus was included when
confirmed by echocardiogram before day of life 5 or if a
clinical diagnosis before day of life 5 was confirmed by
subsequent echocardiogram. Infection was defined as
positive blood and/or urine culture, or if clinical suspicion
led to antibiotic treatment beyond the 48 hours routinely
used in our preterm infants. Next, survival rate and the
neurologic, gastrointestinal, ophthalmologic, and cardio-
respiratory complications of all survivors were docu-
mented.
Table 1. Demographics and biometric characteristics of infants with periventricular hemorrhagic infarction vs age-matched controls

Clinical Factor PVHI (n ⴝ 58) Control (n ⴝ 58) P Value

Median gestation, weeks (range) 25.6 (23–33) 25.5 (22.6–33) 0.318

Median birth weight, grams (range) 733 (450–2375) 788 (460–1690) 0.573
Median birth head circumference, cm (range) 23.5 (20–32) 23.5 (20–29.5) 0.263
Intrauterine growth retardation; number of cases (%) 5 (9%) 4 (7%) 0.690
Singleton; number of cases (%) 32 (55%) 38 (66%) 0.361
Females; number of cases (%) 21 (36%) 28 (48%) 0.281

Pairwise comparison of continuous variables by paired t test, dichotomies by McNemar test.

Abbreviation:
PVHI ⫽ Periventricular hemorrhagic infarction

Statistical Analysis
The trend in the rate of incidence of periventricular
hemorrhagic infarction was analyzed using logistic regres-
sion analysis. Simple comparisons between cases and
control infants were made for continuous variables by
Student paired t test and for dichotomous variables by the
McNemar chi-square test. To determine the independent
effects of the several variables associated with periven-
tricular hemorrhagic infarction, multivariable models were
constructed by conditional logistic regression analysis,
using case-control status as the dichotomous outcome.
Collinearity among groups of variables associated with
periventricular hemorrhagic infarction was investigated by
use of principal-components analysis. SAS (SAS Inc.,
Cary, NC) and SPSS (SPSS Inc., Chicago, IL) software
were used for all computations.
Results
Patient Demographics and Incidence
A total of 5774 premature infants ⬍2500 gm were born
at the study site in the period 1997-2002. The initial
electronic database search identified a total of 68 infants
with possible periventricular hemorrhagic infarction. By
the selection criteria described above, 10 infants were
excluded for the following reasons: criteria met for
periventricular leukomalacia (n ⫽ 2); meningitis (n ⫽ 1);
congenital periventricular cysts (n ⫽ 2); and transient
echodensities without evolution to cystic residua (n ⫽ 5).
Fifty-eight infants (1%) met the diagnostic criteria
for periventricular hemorrhagic infarction. The compar-
ison group included 58 gestational age-matched control
infants with normal neonatal cranial ultrasound. Demo-
graphic and biometric measures were comparable be-
tween the study and control groups (Table 1). Incidence
of periventricular hemorrhagic infarction as a function
of gestational age and birth weight is presented in Table
2 and Figure 1. A significant stepwise increase was
observed in frequency of periventricular hemorrhagic
infarction with decreasing birth weight: 0.1%, 2.2%,
10% for birth weights between 1500-2500 gm, 750-
1500 gm, and ⬍750 gm respectively (P ⬍ 0.001). Thus
incidence of periventricular hemorrhagic infarction
among infants ⬍2500 gm was 1% (0.4-1.6%; range
over 6 years) and for infants ⬍1500 gm was 4%
(1.3-7.7%; range over 6 years). Incidence revealed a
significant upward trend across the 6-year study period
only for infants ⬍750 gm (P ⫽ 0.018). For all other
birth weight categories, the incidence of periventricular
hemorrhagic infarction tended to fluctuate between the
years with no clear significant trend over the study

Table 2. Annual incidence of periventricular hemorrhagic infarction over study period by birth weight category

Birth Weight <750 gm Birth Weight 750–1500 gm Birth Weight 1500–2500 gm Birth Weight <2500 gm
Live Live Live Live
Year Births PVHI % Births PVHI % Births PVHI % Births PVHI %

1997 82 7 8.5 200 6 3 694 0 0.0 976 13 1.3

1998 44 2 4.5 180 1 0.6 788 1 0.1 1012 4 0.4
1999 59 2 3.4 158 1 0.6 679 2 0.3 896 5 0.6
2000 43 6 14.0 176 6 3.4 709 1 0.1 928 13 1.4
2001 36 4 11.1 154 3 1.9 756 0 0.0 946 7 0.7
2002 48 10 20.8 159 6 3.8 809 0 0.0 1016 16 1.6
Total 312 31 9.9 1027 23 2.2 4435 4 0.1 5774 58 1.0
Trend* 1.30 (1.05–1.61), P ⫽ 0.018 1.13 (0.89–1.45), P ⫽ 0.31 † 1.10 (0.94–1.28), P ⫽ 0.23

Abbreviation:
PVHI ⫽ Periventricular hemorrhagic infarction
* Odds ratio per calendar year (95% confidence interval). P value tests for significant trend.
†
Too few events to test for trend.

Bassan et al: PVHI in Preterm Infants 87

 Table 3. Labor and intrapartum factors of infants with
periventricular hemorrhagic infarction vs control infants

Clinical Factor PVHI Control

(n) Number (%) Number (%) P Value

Abnormal fetal 17 (33%) 8 (15%) 0.064

heart rate
tracings (52)
Maternal fever 8 (14%) 8 (14%) 1
(56)
Oxytocin 5 (9%) 11 (19%) 0.146
induction or
augmentation
(58)
Vaginal bleeding 18 (32%) 13 (23%) 0.359
(57)
Vaginal delivery 22 (39%) 30 (53%) 0.134
Figure 1. Incidence of periventricular hemorrhagic infarction (PVHI) (57)
as a function of gestational age. Emergent cesarean 17 (29%) 6 (10%) 0.007
section (58)
Meconium stained 4 (7%) 1 (2%) 0.375
amniotic fluid
period. Interestingly, among all 312 live births ⬍750 (57)
gm, the incidence of mortality between 1997-1999 was Apgar 1= ⱕ 5 (58) 42 (72%) 25 (43%) 0.005
50% (n ⫽ 92), and 79% of deaths occurred in the first Apgar 5= ⱕ 5 (58) 21 (36%) 6 (10%) 0.003
Respiratory 57 (98%) 43 (74%) 0.001
day of life. However, between 2000-2002, the mortality
resuscitation*
rate decreased to 36% (n ⫽ 45) and 55% of deaths (58)
occurred in the first day of life. Cardiac 7 (12%) 2 (4%) 0.18
compressions
(56)
Perinatal Factors Associated With Periventricular Medication used 3 (5%) 2 (4%) 1
Hemorrhagic Infarction (Univariate Analyses) for resuscitation
(57)
Maternal, Pregnancy, and Placental Factors: Pregnancy-
Pairwise comparison of dichotomous variables by McNemar test.
induced hypertension, antenatal infection, antenatal medica- Abbreviations:
tions (magnesium, beta agonists, indomethacin), chorioam- n ⫽ Number of pairs with available data for analysis
nionitis, placental abruption, placental infarction, and PVHI ⫽ Periventricular hemorrhagic infarction
placental weight were not significantly different between * Respiratory resuscitation ⫽ use of mask ventilation or intubation.
cases vs control infants. A complete course of antenatal
betamethasone was more common in the control group
(25/52, 48%) vs the periventricular hemorrhagic infarction in minimum carbon dioxide pressure values between the
group (19/52, 36%) but not statistically significant (P ⫽ two groups.
0.286). Assisted conception was significantly more common
in the periventricular hemorrhagic infarction group (P ⫽ Perinatal Factors Independently Associated With
0.012). Periventricular Hemorrhagic Infarction (Multivariate
Labor and Intrapartum Factors: Emergent cesarean Analysis)
section, low Apgar scores, and need for respiratory resus-
citation were significantly associated with periventricular The variables in each of the univariate domains were
hemorrhagic infarction (P ⬍ 0.05 for all variables) (Table strongly collinear, so that at most one from each domain
3). Fetal presentation (breech/vertex/transverse) and mode could be meaningfully included in a multivariate model.
of labor (induced/spontaneous/no labor) were not signifi- Variables that were too rare or too common to be included
cantly different between cases vs control subjects. in the model (e.g. pneumothorax, present in 0 controls;
Early postnatal factors (first 5 days of life) (Table volume expanders, used in 57 of 58 cases) were elimi-
4): Pressor support, volume infusion, high-frequency ven- nated, and we favored objective measurements (over
tilation, pneumothorax, pulmonary hemorrhage, and subjective ratings) and variables characterizing the patient
patent ductus arteriosus were all significantly more com- (rather than the treatment, thus avoiding confounding by
mon in infants who developed periventricular hemorrhagic indication). Sex and birth weight were included as oblig-
infarction (P ⬍ 0.05). Minimum pH, oxygen pressure, and atory covariates in the model. The final model included
HCO3⫺ values were significantly lower, whereas maximal assisted conception, patent ductus arteriosus, high-fre-
carbon dioxide pressure values were significantly higher quency ventilation, abnormal fetal heart rate, and early
in infants who developed periventricular hemorrhagic minimum pH. Strong independent contributions from
infarction (P ⬍ 0.02). No significant difference was found assisted conception (odds ratio ⫽ 14.5, 95% confidence

88 PEDIATRIC NEUROLOGY Vol. 35 No. 2

Table 4. Early clinical factors in first 5 days for infants with periventricular hemorrhagic infarction vs age-matched control infants

PVHI (n ⴝ 58) Control (n ⴝ 58)

Neonatal Clinical Factor in First 5 Days Number (%) or Mean ⴞ S.D. Number (%) or Mean ⴞ S.D. P Value

Surfactant treatment 57 (98%) 52 (90%) 0.063

Infants not intubated 1 (2%) 6 (10%) 0.063
High-frequency ventilation 37 (64%) 9 (16%) ⬍0.001
Reintubation 16 (28%) 12 (21%) 0.523
Pressor support 55 (95%) 33 (57%) ⬍0.001
Volume expanders 57 (97%) 46 (77%) 0.001
pH min* 7.12 ⫾ 0.13 7.23 ⫾ 0.1 ⬍0.001
pO2 min* 42 ⫾ 10 51 ⫾ 25 0.017
HCO3⫺ min* 17 ⫾ 3 19 ⫾ 3 0.001
pCO2 max* 69.4 ⫾ 20 59 ⫾ 15 0.003
pCO2 min* 32 ⫾ 8 33 ⫾ 8 0.5
Patent ductus arteriosus 23 (40%) 11 (19%) 0.012
Pneumothorax 6 (10%) 0 0.031
Pulmonary hemorrhage 11 (19%) 0 0.001
Infection 34 (59%) 27 (47%) 0.296

Pairwise comparison of continuous variables by paired t test, dichotomies by McNemar test.

Abbreviations:
pCO2 ⫽ Carbon dioxide pressure
pO2 ⫽ Oxygen pressure
PVHI ⫽ Periventricular hemorrhagic infarction
* Minimum and maximum values are derived from data of the first and second days of life.

interval: 1.3-163.1, P ⫽ 0.03) and patent ductus arteriosus
(odds ratio ⫽ 34.6, 95% confidence interval: 1.6-730.7, P
⫽ 0.02) were observed. The remaining predictor was
high-frequency ventilation (odds ratio ⫽ 10.2, 95% con-
fidence interval: 1.1-97, P ⫽ 0.04). Then, two reduced
models were generated (A: assisted conception, patent
ductus arteriosus, abnormal fetal heart rate, and early
minimum pH; B: assisted conception, patent ductus arte-
riosus, high-frequency ventilation) in which both assisted
conception and patent ductus arteriosus again revealed
strong and independent contributions. In these reduced
models, high-frequency ventilation could be confounded
by abnormal fetal heart rate and minimum pH over days
1-2. Controlling for multiple pregnancy or substituting
multiple pregnancy for assisted conception in the model
did not affect the statistical significance or odds ratio for
assisted conception.
Outcome in the Neonatal Intensive Care Unit
Neonatal mortality was 40% (n ⫽ 23) and 5% (n ⫽ 3)
for infants in the periventricular hemorrhagic infarction
and control groups, respectively. Twenty-two of 23 deaths
in the periventricular hemorrhagic infarction group, and
two of three deaths in the control group followed a
decision to withdraw life support. Death was not associ-
ated with sex (14 males, 9 females; P ⫽ 0.78). Median age
at death was 2.4 days (0.5-16 days, periventricular hem-
orrhagic infarction group) vs 19.5 days (13-26 days,
control group) (P ⫽ 0.054). Survivors with periventricular
hemorrhagic infarction required longer periods of mechan-
ical ventilation and had significantly longer length of
neonatal intensive care unit stay compared with their
control infants. Clinical neonatal seizures occurred in 19%
(n ⫽ 11) of all infants with periventricular hemorrhagic
infarction. Bronchopulmonary dysplasia, grade III retinop-
athy of prematurity, necrotizing enterocolitis, and patent
ductus arteriosus ligation were more common in infants
with periventricular hemorrhagic infarction but were be-
low the threshold for statistical significance for all vari-
ables (Table 5).
Discussion
This study sought to determine the impact of recent
advances in neonatal intensive care on the incidence and
early outcome of periventricular hemorrhagic infarction.
We hypothesized that modern perinatal care would be
associated with decreasing incidence and improved sur-
vival of infants with periventricular hemorrhagic infarc-
tion since the first major studies of this problem almost 20
years ago [1,10].
The data presented in this report suggest a decrease in
the incidence of periventricular hemorrhagic infarction
when compared with studies from the previous two de-
cades [1,7,11-15]. However, over each of the study years
1997-2002, the incidence of periventricular hemorrhagic
infarction for infants ⬍2500 gm fluctuated but did not
significantly change (P ⫽ 0.23), similar to the findings of
other recent studies [16,17]. Whether the incidence of
periventricular hemorrhagic infarction has reached a pla-
teau or there continues to be a decreasing incidence is yet
unclear; multicenter epidemiologic studies are required to
address this issue. Strikingly, across the 6 years of the
present study there was a significant increase in the
incidence of periventricular hemorrhagic infarction in

Bassan et al: PVHI in Preterm Infants 89

Table 5. Short-term outcome in neonatal intensive care unit of survivors with periventricular hemorrhagic infarction vs age-matched
control infants

PVHI (n ⴝ 35) Control (n ⴝ 35)

Number (%) or Number (%) or
Outcome Factor Mean ⴞ S.D. Mean ⴞ S.D. P Value

Neonatal seizures 6 (17%) 0 0.031

Ventriculoperitoneal shunt 13 (37%) 0 —
Repeated spinal taps and/or reservoir taps 16 (46%) 0 —
Patent ductus arteriosus ligation 5 (14%) 2 (6%) 0.375
Necrotizing enterocolitis* 8 (23%) 4 (11%) 0.4
Bowel surgery 4 (11%) 1 (3%) 0.38
Other surgery† 12 (34%) 5 (14%) 0.065
Grade III retinopathy of prematurity 7 (20%) 3 (9%) 0.29
Number days ventilated 32 ⫾ 22 22 ⫾ 19 0.044
Patients requiring oxygen at 36 weeks corrected age (n ⫽ 27) 11 (41%) 7 (26%) 0.4
Bronchopulmonary dysplasia (radiographic evidence) 28 (80%) 23 (66%) 0.125
Discharge days 96 ⫾ 45 69 ⫾ 28 0.001

Pairwise comparison of continuous variables by paired t test, dichotomies by McNemar test.

* Necrotizing enterocolitis: definite (radiologically diagnosed) or high clinical suspicion.
†
Includes inguinal hernia repair, gastrostomy tube, ophthalmologic laser therapy, central line placement.

infants with birth weight below ⬍750 gm. This trend
likely reflects the decrease in mortality among the sickest
and most immature infants, i.e., those at greatest risk for
periventricular hemorrhagic infarction [2].
The second major objective of the present study was to
determine whether associated factors for periventricular
hemorrhagic infarction had changed given advances in
neonatal care in recent decades. A number of previous
studies [4,18,19] have described certain prenatal factors
as protective against severe germinal matrix-intraventric-
ular hemorrhage (including periventricular hemorrhagic
infarction), such as pregnancy-induced hypertension and
administration of antenatal steroids, magnesium sulfate,
and beta-agonists. In this study, none of these factors was
significantly different between case and control infants,
nor was administration of antenatal indomethacin, a med-
ication previously identified as increasing the risk for
severe germinal matrix-intraventricular hemorrhage [20].
In the present study, assisted conception was found to be
an independent factor for periventricular hemorrhagic
infarction in the multivariate analysis, as previously sug-
gested [21,22]. The reason for this association remains
unclear. Finally, contrary to other studies [22,23], this
study found no association between maternal fever, doc-
umented maternal infection, pathologically confirmed
chorioamnionitis or neonatal infection and the develop-
ment of periventricular hemorrhagic infarction.
The intrapartum factors for periventricular hemorrhagic
infarction identified in the present study support fetal
distress as a common mechanism for this lesion. Specifi-
cally, emergent cesarean section, low Apgar scores, and
the need for respiratory resuscitation were significantly
more common in infants with periventricular hemorrhagic
infarction. These findings corroborate those from several
previous studies [13,14,24]. However, unlike others
[24,25] we did not find infants delivered vaginally or those
requiring cardiac resuscitation to be at greater risk for
periventricular hemorrhagic infarction, although our sta-
tistical power was relatively low to draw significant
conclusions.
Among the postnatal factors, pneumothorax, pulmo-
nary hemorrhage, and patent ductus arteriosus, all of
which were associated with periventricular hemorrhagic
infarction in this study, have previously been implicated as
causing pronounced disturbances in cerebral perfusion and
oxygenation [4,13,26]. Patent ductus arteriosus, an inde-
pendent factor in the multivariate model, is associated with
decreased diastolic blood flow [27] and low superior vena
cava flow [28]. Premature infants receiving high-fre-
quency ventilation were at increased risk for periventricu-
lar hemorrhagic infarction in the multivariate analysis,
confirming a previous report [29]. Although the require-
ment for high-frequency ventilation may reflect the sever-
ity of illness of these infants, it might also be associated
with an elevation of intrathoracic and hence central venous
pressure, thereby increasing cerebral venous pressure.
Cerebral venous pressure could also be increased by
pneumothorax and pulmonary hemorrhage.
In addition, acidosis and hypercarbia, identified as
factors associated with periventricular hemorrhagic infarc-
tion in our and other studies [11], are known to impair
cerebral pressure autoregulation [30], rendering the brain
more vulnerable to hemodynamic fluctuations. This obser-
vation is particularly relevant because the need for both
inotropic pressor support and intravascular volume ex-
panders was significantly associated with periventricular
hemorrhagic infarction in the present study as in others
[14]. In fact, nearly all our infants with periventricular
hemorrhagic infarction required these two interventions
The third objective of this study was to determine the
short-term outcome of periventricular hemorrhagic infarc-
tion. Although lower than the 60% mortality reported

90 PEDIATRIC NEUROLOGY Vol. 35 No. 2

10-20 years ago [1,13], our mortality data suggest that the
risk of neonatal death remains substantial, approaching
40% in the infants with periventricular hemorrhagic in-
farction in this study. These data corroborate the finding of
deVries et al. [15] of a mortality rate of 44% between
1990-1999. Nevertheless, a mortality rate as high as 69%
was recently reported from another center [16]. In the
present study, time of death was earlier than previously
reported [31]. Among survivors of periventricular hemor-
rhagic infarction in the present report, the neonatal mor-
bidity was significantly higher than in control infants, i.e.,
longer duration of mechanical ventilation and length of
neonatal intensive care unit stay. Overall, 37% of survi-
vors of periventricular hemorrhagic infarction required
ventriculoperitoneal shunts, similar to findings in a recent
report [32].
Prospective studies of periventricular hemorrhagic in-
farction are difficult to undertake given the relatively low
prevalence of this lesion. As a retrospective analysis, the
present study is subject to potential limitations. Timing of
cranial ultrasound was determined by neonatal intensive
care unit protocol and clinical judgment. Also, as infants
dying in the early hours of life may never undergo cranial
ultrasound, periventricular hemorrhagic infarction may be
underestimated because nearly 25% of such infants may
have periventricular hemorrhagic infarction at postmortem
examination [33]. Additionally, we cannot assign causality
to factors associated with periventricular hemorrhagic
infarction. Finally, the present work is based on data from
a single institution, and trends could vary across different
centers. However, the case-controlled design allowed us to
identify associations regardless of differences in gesta-
tional age.
In summary, this study suggests a decrease in both
overall incidence and mortality of periventricular hemor-
rhagic infarction from earlier reports two decades ago
[1,7]. Whether incidence of periventricular hemorrhagic
infarction has currently reached a nadir [16] or whether the
decline in incidence will persist is unclear. However, the
data presented here demonstrate that incidence of periven-
tricular hemorrhagic infarction significantly increases in
infants ⬍750 gm. Associated factors for periventricular
hemorrhagic infarction identified in this study are consis-
tent with the notion that disturbances in systemic and
cerebral hemodynamics clustering around the intrapartum
and early neonatal period are particularly important in the
genesis of periventricular hemorrhagic infarction. Con-
versely, factors associated with infection/inflammation
were not increased in infants with periventricular hemor-
rhagic infarction; however, these findings should be con-
firmed by measurements of inflammatory markers and a
larger study sample. Despite benefits associated with
advances in perinatal care, periventricular hemorrhagic
infarction remains an important complication of prematu-
rity, particularly extreme prematurity (⬍750 gm), and is
likely to increase in prevalence in this group as survival
rates of extremely low birth weight infants continue to
increase.
Haim Bassan is supported by the LifeBridge Fund. Statistical analysis for
this study was funded by grant MO1-RR02172 from the National Center
for Research Resources, National Institutes of Health, to the Children’s
Hospital Boston General Clinical Research Center. The authors thank
Shaye Moore for assistance with manuscript preparation and Amy
Kroeplin for data management.
References
[1] Guzzetta F, Shackelford G, Volpe S, Perlman JM, Volpe JJ.
Periventricular intraparenchymal echodensities in the premature new-
born: Critical determinant of neurologic outcome. Pediatrics 1986;78:
995-1006.
[2] Volpe JJ. Neurology of the newborn, 4th ed. Philadelphia: W.B.
Saunders, 2001:428 –93.
[3] Sheth RD. Trends in incidence and severity of intraventricular
hemorrhage. Neurology 1998;13:261-4.
[4] Weintraub Z, Solovechick M, Reichman B, et al. Effect of
maternal tocolysis on the incidence of severe periventricular/intraven-
tricular haemorrhage in very low birthweight infants. Arch Dis Child
Fetal Neonatal Ed 2001;85:F13-7.
[5] Fanaroff AA. The NICHD neonatal research network: Changes
in practice and outcomes during the first 15 years. Semin Perinatol
2003;27:281-7.
[6] Volpe JJ. Brain injury in the premature infant: Overview of
clinical aspects, neuropathology, and pathogenesis. Semin Pediatr Neurol
1998;5:135-51.
[7] de Vries LS, Roelants-van Rijn AM, Rademaker KJ, Van
Haastert IC, Beek FJ, Groenendaal F. Unilateral parenchymal haemor-
rhagic infarction in the preterm infant. Eur J Paediatr Neurol 2001;5:
139-49.
[8] Gleissner M, Jorch G, Avenarius S. Risk factors for intraven-
tricular hemorrhage in a birth cohort of 3721 premature infants. J Perinat
Med 2000;28:104-10.
[9] de Vries LS, Groenendaal F. Neuroimaging in the preterm infant.
Ment Retard Dev Disabil Res Rev 2002;8:273-80.
[10] McMenamin JB, Shackelford GD, Volpe JJ. Outcome of
neonatal intraventricular hemorrhage with periventricular echodense
lesions. Ann Neurol 1984;15:285-90.
[11] Levene MI, de Vries L. Extension of neonatal intraventricular
haemorrhage. Arch Dis Child 1984;59:631-6.
[12] Perlman JM, Volpe JJ. Intraventricular hemorrhage in ex-
tremely small premature infants. Am J Dis Child 1986;140:1122-4.
[13] Perlman JM, Rollins N, Burns D, Risser R. Relationship
between periventricular intraparenchymal echodensities and germinal
matrix-intraventricular hemorrhage in the very low birth weight neonate.
Pediatrics 1993;91:474-80.
[14] Bass WT, Schultz SJ, Burke BL, White LE, Khan JH, Karlow-
icz MG. Indices of hemodynamic and respiratory functions in premature
infants at risk for the development of cerebral white matter injury. J
Perinatol 2002;22:64-71.
[15] de Vries LS, Van Haastert IL, Rademaker KJ, Koopman C,
Groenendaal F. Ultrasound abnormalities preceding cerebral palsy in
high-risk preterm infants. J Pediatr 2004;144:815-20.
[16] Hamrick SE, Miller SP, Leonard C, et al. Trends in severe
brain injury and neurodevelopmental outcome in premature newborn
infants: The role of cystic periventricular leukomalacia. J Pediatr 2004;
145:593-9.
[17] Horbar JD, Badger GJ, Carpenter JH, et al. Trends in mortality
and morbidity for very low birth weight infants, 1991-1999. Pediatrics
2002;110:143-51.
[18] Perlman JM, Riser RC, Gee JB. Pregnancy-induced hyperten-
sion and reduced intraventricular hemorrhage in preterm infants. Pediatr
Neurol 1997;17:29-33.
Bassan et al: PVHI in Preterm Infants 91
 [19] Perlman JM. Antenatal glucocorticoid, magnesium exposure,
and the prevention of brain injury of prematurity. Semin Pediatr Neurol
1998;5:202-10.
[20] Norton ME, Merrill J, Cooper BA, Kuller JA, Clyman RI.
Neonatal complications after the administration of indomethacin for
preterm labor. N Engl J Med 1993;329:1602-7.
[21] Stromberg B, Dahlquist G, Ericson A, Finnstrom O, Koster M,
Stjernqvist K. Neurological sequelae in children born after in-vitro
fertilisation: A population-based study. Lancet 2002;359:461-5.
[22] Linder N, Haskin O, Levit O, et al. Risk factors for intraven-
tricular hemorrhage in very low birth weight premature infants: A
retrospective case-control study. Pediatrics 2003;111:e590-5.
[23] Dammann O, Kuban KC, Leviton A. Perinatal infection, fetal
inflammatory response, white matter damage, and cognitive limitations
in children born preterm. Ment Retard Dev Disabil Res Rev 2002;8:
46-50.
[24] Osborn DA, Evans N, Kluckow M. Hemodynamic and ante-
cedent risk factors of early and late periventricular/intraventricular
hemorrhage in premature infants. Pediatrics 2003;112:33-9.
[25] Finer NN, Horbar JD, Carpenter JH. Cardiopulmonary resus-
citation in the very low birth weight infant: The Vermont Oxford
Network experience. Pediatrics 1999;104:428-34.
[26] Pandit PB, O’Brien K, Asztalos E, Colucci E, Dunn MS.
Outcome following pulmonary haemorrhage in very low birthweight
92 PEDIATRIC NEUROLOGY Vol. 35 No. 2
neonates treated with surfactant. Arch Dis Child Fetal Neonatal Ed
1999;81:F40-4.
[27] Lipman B, Serwer GA, Brazy JE. Abnormal cerebral hemody-
namics in preterm infants with patent ductus arteriosus. Pediatrics
1982;69:778-81.
[28] Kluckow M, Evans N. Low superior vena cava flow and
intraventricular haemorrhage in preterm infants. Arch Dis Child Fetal
Neonatal Ed 2000;82:F188-94.
[29] Cools F, Offringa M. Meta-analysis of elective high frequency
ventilation in preterm infants with respiratory distress syndrome. Arch
Dis Child Fetal Neonatal Ed 1999;80:F15-20.
[30] Bassan H, Gauvreau K, Newburger JW, et al. Identification of
pressure passive cerebral perfusion and its mediators after infant cardiac
surgery. Pediatr Res 2004;54:1-8.
[31] Larroque B, Marret S, Ancel PY, et al. White matter damage
and intraventricular hemorrhage in very preterm infants: The EPIPAGE
study. J Pediatr 2003;143:477-83.
[32] Murphy BP, Inder TE, Rooks V, et al. Posthaemorrhagic
ventricular dilatation in the premature infant: Natural history and
predictors of outcome. Arch Dis Child Fetal Neonatal Ed 2002;87:
F37-41.
[33] Golden JA, Gilles FH, Rudewlli R, Leviton A. Frequency of
neuropathological abnormalities in very low birth weight infants. J Neu-
ropathol Exp Neurol 1997;56:472-8.