See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/10695639

Etiology and risk factors for placenta previa: An overview and meta-
analysis of observational studies

Article  in  Journal of Maternal-Fetal and Neonatal Medicine · April 2003

DOI: 10.1080/jmf.13.3.175.190 · Source: PubMed

CITATIONS READS

183 1,342

2 authors, including:

Cande Ananth
Columbia University
638 PUBLICATIONS   14,133 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Placental abruption View project

Methods and applications View project

All content following this page was uploaded by Cande Ananth on 10 November 2014.

The user has requested enhancement of the downloaded file.

 The Journal of Maternal–Fetal and Neonatal Medicine 2003;13:175– 190

Etiology and risk factors for placenta previa: an

overview and meta-analysis of observational studies
A. S. Faiz 1 and C. V. Ananth 2
1
Division of Hematology, Department of Medicin e, Robert Wood J ohnson Medical School, University of Medicin e
and Dentistry of New J ersey, New Brunswick, New J ersey, USA
2
Section of Epidemiology and Biostatistics, Department of Obstetrics, Gynecology and Reproductive Sciences,
Robert Wood J ohnson Medical School, University of Medicin e and Dentistry of New J ersey, New Brunswick,
New J ersey, USA
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14

Objective: Several clinical and epidemiologic studies have reported disparate data on the preva-
lence rate as well as risk factors associated with placenta previa – a major cause of third-trimester
bleeding. We performed a systematic literature review and identified 58 studies on placenta previa
published between 1966 and 2000.
Study design: Each study was reviewed independently by the two authors and was scored (on the
basis of established criteria) on method of diagnosis of placenta previa and on quality of study
design. We extracted data on the prevalence rate of placenta previa, as well as associations with
various risk factors from each study. A meta-analysis was then performed to determine the extent
to which different risk factors predispose women to placenta previa.
Results: Our results showed that the overall prevalence rate of placenta previa was 4.0 per 1000
For personal use only.

births, with the rate being higher among cohort studies (4.6 per 1000 births), USA-based studies
(4.5 per 1000 births) and hospital-based studies (4.4 per 1000 births) than among case–control
studies (3.5 per 1000 births), foreign-based studies (3.7 per 1000 births) and population-based
studies (3.7 per 1000 births), respectively. Advancing maternal age, multiparity, previous Cesarean
delivery and abortion, smoking and cocaine use during pregnancy, and male fetuses all conferred
increased risk for placenta previa. Strong heterogeneity in the associations between risk factors
and placenta previa were noted by study design, accuracy in the diagnosis of placenta previa and
population-based versus hospital-based studies.
Conclusion: Future etiological studies on placenta previa must, at the very least, adjust for
potentially confounding effects of maternal age, parity, prior Cesarean delivery and abortions.

Key words: PLACENTA PREVIA ; ETIOLOGY ; EPIDEMIOLOGY ; META-ANALYSIS ; OVERVIEW

INTRODUCTION
Placenta previa is an obstetric complication in which the
maturing placenta partially or completely obstructs the
internal cervical os. It is a major cause of third-trimester
bleeding and has been associated with serious maternal
complications and adverse perinatal outcomes1. Maternal
complications associated with placenta previa include
hemorrhage requiring blood transfusion, disseminated
intravascular coagulation and partial or total hysterectomy.
Pregnancies complicated by placenta previa also result in
greater frequencies of prematurity and fetal and neonatal
death2–5. Despite early diagnosis and careful surveillance of
women with placenta previa and great advances in neo-
natal care, it is still challenging to avoid maternal compli-
cations and to improve perinatal outcomes among women
with this condition.
The etiology of placenta previa remains largely obscure,
but several clinical and epidemiological studies have
observed increased occurrence of placenta previa among
women with advanced maternal age, multiparity, male
fetuses, multiple pregnancy, prior Cesarean delivery and
prior spontaneous or induced abortion. Furthermore,
behavioral factors that have been implicated with
increased occurrence of placenta previa include maternal
smoking and drug use during pregnancy. Finally, women

Corr espondence: Dr A. S. Faiz, Division of Hem atology, Room 378, One Robert Wood J ohnson Place, PO Box 19, New Brunswick,
NJ 08903, USA
ã 2003 The Par thenon Publish ing Group 175 Received 24–06–02 Revised 28–10–02 Accepted 29–11–02
 Etiology and risk factors for placenta previa Faiz and Ananth

with a history of placenta previa in a prior pregnancy are at
higher risk of developing this condition in a subsequent
pregnancy.
The prevalence rate of placenta previa and the extent to
which various risk factors predispose women to develop
placenta previa vary greatly in the previous studies. We
therefore performed a systematic review of the literature
to estimate the prevalence of placenta previa and its asso-
ciation with various risk factors. A meta-analysis was
then performed to identify sources of heterogeneity across
studies, as well as to evaluate the strength and magnitude
of the association of placenta previa with these risk factors.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
previa was diagnosed in early pregnancy (first or second
trimesters), as these studies were predominantly focused
on evaluating the accuracy of an early diagnosis of placenta
previa by sonographic methods. If more than one study
had the same data source covering overlapping time
periods, then only one study was chosen for inclusion in the
meta-analysis. The choice of study in such instances was
based on the availability of data on risk factors for placenta
previa. If two or more studies using the same patient
population provided information on placenta previa
cross-classified by the same risk factors, then the study
that was published first was arbitrarily chosen for inclusion
in the meta-analysis.

MATERIALS AND METHODS

Literature review
Observational studies published in the English language
between January 1966 and March 2000 were potentially
eligible for inclusion in this overview. Identification of
such studies was based on a comprehensive MEDLINE
search, as well as by identifying studies cited in the
references of published papers. The MEDLINE search was
based on the following medical subject headings (MeSH):
For personal use only.
Scoring of studies
Both authors independently reviewed all the studies chosen
for the meta-analysis. A numeric score on a five-level
ordinal scale (1–5, with 5 being the best quality) on quality
of study was assigned for each study. The study quality was
based primarily on study design and adjustment for
four important confounding factors: maternal age, parity,
previous Cesarean delivery and previous spontaneous or
induced abortion. Similarly, each study was scored on a

placenta pr(a)evia, placental disorders, antepartum
h(a)emorrhage, and antepartum and uteroplacental
bleeding. The other key words used in conjunction with
previa were maternal age, gravidity, parity, C(a)esarean
delivery/section, uterine surgery, uterine instrumentation,
abortion, spontaneous abortion, induced abortion, elective
abortion, chronic hypertension, pregnancy-induced hyper-
tension, pre-eclampsia, eclampsia, cigarette smoking and
drug use. Both the primary author and the librarian in our
institution conducted the searches independently using
these key words.
Published case reports on placenta previa and studies on
placental abruption were excluded. No attempts were made
to locate any unpublished studies or those published in
abstract form. We excluded studies in which placenta
four-level ordinal scale (1–4, with 4 being the most accu-
rate) for the method of diagnosis of placenta previa used
in each study. The criteria used for scoring the study
quality and method of diagnosis of placenta previa were
determined a priori, and are shown in Tables 1 and 2,
respectively.
We were left with 58 studies4–61 for the meta-analysis
after excluding 42 studies2,3,62–101 that did not meet our
inclusion/exclusion criteria. Both authors gave identical
scores to 36 studies on study quality and to 48 studies for
method of diagnosis of placenta previa. The chance-
corrected kappa statistic with 95% confidence interval
(CI) for agreement in the scores between the two authors
on study design and diagnosis of placenta previa, respec-
tively, were 0.66 (95% CI 0.53, 0.79) and 0.85 (95% CI

Table 1 Numeric scoring of quality of studies according to study design

Cohort Case–control
studies studies Scoring criteria according to quality of study

5 5
4 4
3 3
2 2
1 1
studies in which no obvious biases could be identified or potential confounders were adequately
controlled for
studies in which adjustment was made for potential biases but residual confounding appeared likely
studies in which adjustment for potential confounders was done poorly
studies in which no adjustment was made for potential confounders
studies without appropriate control subjects

Studies that controlled for maternal age, parity, previous Cesarean delivery, previous spontaneous or induced abortion were considered
as those without obvious biases (score of 5)

176
J ournal of Maternal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa
Table 2 Numeric scoring of study’s quality based on the method
of diagnosis of placenta previa
Study
score Criteria for scoring
4 placenta previa confirmed at Cesarean delivery
3 placenta previa confirmed by third-trimester ultrasound
examination
2 placenta previa as reported by attendant at delivery or by
digital examination at the time of delivery
1 criteria for the diagnosis of placenta previa not well
defined
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
0.76, 0.94). The scoring was different for quality of study
in 22 studies and for diagnosis of placenta previa in ten
studies, and in such cases the two authors discussed
the reasons and reassigned the scores after resolving the
discrepancy.
Based on the scores for the quality of study, 12 studies
(20%) with the score of 4 and 5 were classified as
well-designed studies and 46 studies with scores of less than
3 were classified as poorly designed. Similarly, based on the
method of diagnosis of placenta previa, 25 studies (43%)
For personal use only.
with the score of 3 and 4 were classified as studies with
definite placenta previa and the remaining 33 studies
with scores of 1 and 2 were classified as studies with
probable placenta previa.
Data extraction
For each of those 58 studies that met our criteria, we
abstracted the year of publication, study design (cohort and
case–control study), geographical location where data
were collected (USA-based and foreign-based), source of
data (hospital-based, population-based, vital records) and
time and duration of data collection.
All studies were critically reviewed and information was
abstracted on total number of pregnancies and the number
of pregnancies complicated by placenta previa. Data on all
risk factors related to placenta previa examined in each
study were abstracted. Risk factors that were examined in
this study included maternal age, parity, prior Cesarean
delivery, prior spontaneous or induced abortion, smoking,
alcohol and cocaine use during pregnancy, hypertensive
disorders (chronic and pregnancy-induced hypertension
and pre-eclampsia/eclampsia), sex of the fetus, multiple
pregnancy and placental abruption. Studies that did
not provide sufficient information to carry out the meta-
analysis or those that did not provide data for the compari-
son group were excluded from the meta-analysis. However,
these studies were still included in the review as well as
for the calculation of rates (of placenta previa) when data
were available.
Faiz and Ananth
Statistical methods for meta-analysis
The prevalence rate of placenta previa (per 1000 preg-
nancies) was either abstracted from the study directly or
derived by dividing the number of cases of placenta previa
by the total number of pregnancies. Data on placenta
previa cross-classified by the presence or absence of risk
factors were abstracted from every study, and were
organized in 2 ´ 2 tables. Odds ratios (OR) with associated
95% CI were computed as the measure of effect.
A meta-analysis was then performed by pooling the data
for each risk factor across all studies. We did not pool the
data for multiple pregnancies, recurrent placenta previa
and placental abruption as risk factors for placenta previa,
owing to lack of data. Prior to pooling data across studies,
we tested the homogeneity of studies being pooled. The test
statistic for homogeneity, Qh, was then compared with
the c2 distribution, with degrees of freedom equal to the
number of studies being pooled minus one. Random effect
ORs were computed after pooling the studies. Random-
effect analysis accounts for heterogeneity among studies
being pooled and assumes that these studies are a sample
from a larger population of similar but unidentified studies.
We also performed subgroup analysis for each risk factor
by grouping studies based on study design (cohort vs. case–
control studies), geographic location (USA vs. foreign-
based studies) and source of data (hospital vs. population-
based studies). The dose–response effect of maternal
age and parity on the risk of placenta previa was also
evaluated from each study based on the covariance-
adjusted method102.
RESULTS
Fifty-eight studies met our criteria for inclusion in the
meta-analysis, and were scored and classified according
to geographic location, source of data and study design
(Tables 3 and 4). There were 30 United States-based, and
28 were foreign-based studies; 37 cohort and 21 case–
control studies; and 11 population-based and 47 hospital-
based studies.
The prevalence rate of placenta previa ranged between
2.8 and 19.7 per 1000 pregnancies in 58 studies (Table 5).
Of the 58 studies, there were 44 studies that reported data
on prevalence rates of placenta previa. There were a total
of 4.5 million pregnancies, of which 18 000 were identified
with placenta previa (4.0 per 1000 pregnancies). Amongst
the 44 studies that provided prevalence data, the frequency
of placenta previa was higher among 31 cohort studies (4.6
per 1000 pregnancies), 19 USA-based studies (4.5 per 1000
pregnancies) and 39 hospital-based studies (4.4 per 1000
pregnancies) than among 13 case–control studies (3.5 per
1000 pregnancies), 25 foreign-based studies (3.7 per 1000
pregnancies) and five population-based studies (3.7 per
177
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Table 3 Description of studies conducted within the USA and scores assigned to each study based on study design and diagnosis of
placenta previa
Placenta Scores assigned
Number of previa (per Criteria Criteria
Author Study years pregnancies 1000 births) for inclusion for exclusion Study design Diagnosis

Population-based case–control studies

Kramer10 1984–87 — — singleton live births — 5 1
Williams8 1987–88 singleton births 5 2
Zhang 16 1988–90 309 320 4.4 live births multiple births 4 2
Taylor7 1984–87 — — singleton live births 4 2
13
McMahon 1990 — — singleton live births nulliparous 4 2
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14

Hospital-based case–control studies

Grimes35 1975–79 28 665 3.4 > 500 g fetus — 3 3
Newton34 1981 — — — — 2 3
Parson45 1982–85 10 473 7.7 singletons — 2 1
Wolf4 1980–90 54 969 3.9 — multiple gestations 2 1
Williams22 1977–80 12 718 5.5 singletons — 5 2
Handler12 1988–90 55 314 5.9 — — 5 2
Chelmow23 1992–94 — — births ³ 20 weeks — 3 4
Macones24 1992–96 — — births ³ 20 weeks — 3 1
30
Sauer 1981–84 10 822 4.0 — — 2 4

Population-based cohort studies

Hemminki49 1976–83 — 9.9 live births — 1 2
For personal use only.

Iyasu41 1979–87 — 4.8 live and stillbirths — 2 1

36
Demissie 1989–92 447 963 6.0 singleton live births — 4 2

Hospital-based cohort studies

Niswander9 1959–65 17 265 6.0 singletons > 500 g — 2 1
Hibbard57 1948–53 102 670 4.6 live and stillbirths > 500 g — 2 2
1962–66
Crenshaw43 1950–69 208 837 3.7 — — 1 2
Brenner33 1962–69 31 070 6.0 singletons — 2 2
Cotton54 1975–78 38 398 4.5 — — 1 3
Barrett50 1972–74 — — — — 2 4
1979–80
Chervenak52 1978–82 16 029 4.3 — — 1 2
Silver53 1974–80 21 217 4.5 > 500 g fetus — 1 4
D’Angelo55 1979–81 — — > 28 weeks — 1 4
Clark14 1977–83 97 799 3.0 — — 2 4
McShane46 1975–82 42 058 3.5 > 20 weeks — 1 4
Jelsema40 — 6 576 16.8 — — 1 4
Frederiksen44 1976–97 93 384 5.5 — — 1 4

1000 pregnancies), respectively. Furthermore, the
prevalence of placenta previa showed no temporal trends
(Figure 1).
The odds ratios based on the covariance-adjusted
method showed that both advanced maternal age and high
parity were associated with an increased risk of placenta
previa. There were nine studies6–14 that reported an asso-
ciation between advancing maternal age and placenta
previa (Figure 2). Results showed that odds ratios were
between 1.2 and 2.7 for three cohort studies9,11,14, seven
USA-based studies7–10,12–14, six studies with probable
diagnosis of placenta previa7–10,12,13, and three poorly
designed studies6,9,14. Odds ratios ranged between 1.2 and
2.5 for six case–control studies6–8,10,12,13 and six well-
designed studies7,8,10–13. For two foreign based
studies6,11 and for three studies with definite diagnosis
of placenta previa6,11,14 pooled odds ratios ranged between
1.5 and 2.5.

178
J ournal of Maternal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Table 4 Description of studies conducted outside the USA and scores assigned to each study based on study design and diagnosis
of placenta previa
Placenta Scores assigned
Number of previa (per Criteria Criteria
Author Study years pregnancies 1000 births) for inclusion for exclusion Study design Diagnosis

Population-based case–control studies

61
Monica 1973–90 1 825 998 2.8 live and stillbirths — 3 1
> 28 weeks

Hospital-based case–control studies

Neri5 1968–77 29 486 5.4 — multiple gestations 2 3
Rose20 1978–82 20 377 3.9 — — 3 4
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14

Gorodeski17 1972–83 — 5.0 — — 2 2

Parazzini6 1979–91 49 765 2.8 — — 3 4
Takayama30 1974–94 15 255 5.8 — — 2 4
18
Abu Heija 1994–97 27 483 3.1 — — 3 1

Population-based cohort studies

Ananth11 1986–93 87 184 3.3 singletons multiple gestations 4 3
37
Ananth 1980–93 121 082 3.4 singletons multiple gestations 4 3

Hospital-based cohort studies

Gabert60 1956–66 22 437 4.0 — twins 1 2
Dutta58 1965–73 19 888 10.6 — — 1 2
Hill59 1971–77 26 155 9.1 — — 1 2
For personal use only.

Gorodeski21 1968–78 — 5.0 — — 1 2

Singh27 1973–78 12 040 19.7 — — 1 1
Nicolaides38 1973–81 24 549 8.2 — — 4 4
MacGillivary15 1976–83 114 470 3.4 singletons — 2 1
Nielsen28 1978–87 24 644 3.3 — — 1 4
Jackobovitz42 1972–86 26 858 5.4 — — 2 1
Leiberman48 1972–86 106 866 4.6 — — 1 2
Chattopadhay25 1988–92 41 206 5.4 — — 2 4
Makhseed19 1981–92 158 006 5.0 — — 1 1
Rosen56 1991–93 — — — — 1 3
Hershkowitz26 1985–92 58 633 4.8 — nulliparous 3 2
Khashoggi47 1988–91 13 032 7.4 — — 1 4
Love51 1991–93 15 930 3.6 — — 1 3
Zaideh29 1995–96 18 651 3.5 — — 1 3
Wang39 1989–98 25 844 4.9 singletons, primigravida congenital anomaly 4 2
Hendricks32 1993–97 16 169 1.0 singletons — 3 2

The relationship between parity and placenta previa was
provided in 13 studies6–18 (Figure 3). Among four cohort
studies9,11,14,15 and among ten studies with a probable
diagnosis of placenta previa7–10,12,13,15–18 the odds ratios
ranged between 1.1 and 2.9. Odds ratios for nine case–
control studies6–8,10,12,13,16–18 ranged from 0.9 to 2.3, and for
three studies with definite diagnosis of placenta previa6,11,14
they were between 0.9 and 2.0. Odds ratios were between
0.9 and 2.9 for five foreign-based studies6,11,15,17,18 and
six poorly designed studies6,9,14,15,17,18. Among the eight
USA-based studies7–10,12–14,16 and seven well-designed
studies7,8,10–13,16, odds ratios ranged between 1.1 and 2.3.
Prior Cesarean delivery as a risk factor for placenta
previa was evaluated in 21 studies6,7,10,12,14,16,18–32
(Table 6). The association was stronger among eight cohort
studies14,19,25–29,32 (pooled OR 6.2) and for 12 foreign-based
studies6,18–21,25–29,31,32 (pooled OR 4.8) compared with 13
case–control studies6,7,10,12,16,18,20–24,30,31 (pooled OR 2.0)
and nine studies7,10,12,14,16,22–24,30 conducted in the USA
(pooled OR 2.0). With Cesarean delivery, the risk of
placenta previa was higher in ten studies with definite
diagnosis of placenta previa6,12,14,20,23,25,28–31 (pooled
OR 3.0) and in 17 poorly designed studies6,12,14,18–21,23–32
(pooled OR 3.5).

179
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Table 5 Prevalence rate of placenta previa (per 1000 pregnancies) by type of study design and geographical location
Number of studies Rate of previa (derived) Number of studies Rate of previa (reported)

Overall 44 4.0 58 2.8 to 19.7

Cohort studies 31 4.6 37 3.0 to 19.7
Case–control studies 13 3.5 21 2.8 to 7.7
USA studies 19 4.5 30 3.0 to 16.8
Foreign-based studies 25 3.7 28 2.8 to 19.7
Population-based studies 5 3.7 11 3.1 to 9.9
Hospital-based studies 39 4.4 47 2.8 to 19.7
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
For personal use only.

Figure 1 Prevalence rate (per 1000 singleton births) of placenta Figure 3 Association between parity and placenta previa from
previa from each study based on year of data collection. The size of the 13 studies: odds ratios with 95% confidence intervals
the ‘bubble’ denotes the study size

previous abortions and risk of placenta previa. Previous

Figure 2 Association between maternal age and placenta previa
from the nine studies: odds ratios with 95% confidence intervals
Previous abortion conferred a 90% increase in the risk
of placenta previa in eight studies6,13,16,18,21,30,31,33 (Table
7). There were one cohort33 and seven case–control
studies6,13,16,18,21,30,31 that reported the association between
abortion was more significantly associated with placenta
previa in four foreign-based studies6,18,21,31 (pooled OR 2.7)
than in the four USA-based studies13,16,30,33 (pooled OR
1.8). The risk of placenta previa was similar in three
studies with definite diagnosis6,30,31 and in five studies with
probable diagnosis of placenta previa13,16,18,21,33 (pooled
OR 1.9). However, the risk was higher among six poorly
designed studies6,18,21,30,31,33 (pooled OR 2.4) than in two
well-designed studies13,16(pooled OR 1.8).
There was a higher risk of placenta previa with previous
spontaneous abortion (pooled OR 2.0) in seven
studies7,10,12,20,22,23,34 (Table 8). All the studies were
case–control studies and only one study20 was foreign
based. There was an increased risk of placenta previa with
spontaneous abortion in three studies with definite diagno-
sis of placenta previa20,23,34 and three poorly designed
studies20,23,34 (pooled OR 2.3) than in four studies with
probable diagnosis of placenta previa7,10,12,22 and in four
well-designed studies7,10,12,22 (pooled OR 2.0).
In the presence of previous induced abortion there was a
50% increase in placenta previa (Table 9). All eight studies
were USA-based case–control studies7,10,12,22–24,34,35. With

180
J ournal of Maternal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Table 6 Association between prior Cesarean delivery and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 21 60.3 20 < 0.001 2.7 (2.3, 3.2)

Cohort studies 8 0.2 7 0.999 6.2 (3.6, 10.5)
Case–control studies 13 11.7 12 0.470 2.0 (1.8, 2.3)
USA-based studies 9 7.1 8 0.525 2.0 (1.7, 2.3)
Foreign-based studies 12 12.7 11 0.313 4.8 (3.8, 6.2)
Definite previa 10 27.5 9 0.001 3.0 (2.3, 3.2)
Probable previa 11 25.0 10 0.009 2.3 (2.0, 2.7)
Well-designed studies 4 2.2 3 0.531 1.9 (1.7, 2.2)
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14

Poorly designed studies 17 35.0 16 0.006 3.5 (2.7, 4.6)

Qh, test statistic for homogeneity

Table 7 Association between previous abortions (spontaneous or induced) and placenta previa
Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 8 13.4 7 0.063 1.9 (1.7, 2.2)

For personal use only.

Cohort studies 1 — — — 2.1 (1.6, 2.9)

Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
7 12.8 6 0.046 1.9 (1.6, 2.1)
4 3.0 3 0.391 1.8 (1.6, 2.0)
4 4.5 3 0.212 2.7 (2.0, 3.7)
3 0.0 2 1.000 1.9 (0.8, 5.0)
5 13.4 4 0.009 1.9 (1.7, 2.2)
2 1.7 1 0.192 1.8 (1.6, 2.0)
6 5.7 5 0.336 2.4 (2.0, 3.0)

Qh, test statistic for homogeneity

Table 8 Association between prior spontaneous abortion and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 7 6.6 6 0.359 2.0 (1.7, 2.3)

Cohort studies
Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
— — — — —
7 6.6 6 0.359 2.0 (1.7, 2.3)
6 6.5 5 0.261 2.0 (1.7, 2.3)
1 — — — 2.1 (1.1, 4.4)
3 1.2 2 0.548 2.3 (1.4, 3.5)
4 5.0 3 0.172 2.0 (1.7, 2.3)
4 5.0 3 0.172 2.0 (1.7, 2.3)
3 1.2 2 0.753 2.3 (1.3, 3.5)

Qh, test statistic for homogeneity

181
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Table 9 Association between prior induced abortion and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 8 12.1 7 0.097 1.5 (1.3, 1.9)

Cohort studies
Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
— — — — —
8 12.1 7 0.097 1.5 (1.3, 1.9)
8 12.1 7 0.097 1.5 (1.3, 1.9)
— — — — —
3 5.1 2 0.078 1.3 (0.6, 2.7)
5 5.5 4 0.239 1.6 (1.4, 2.0)
4 5.5 3 0.139 1.6 (1.3, 2.0)

Poorly designed studies 4 5.1 3 0.164 1.2 (0.7, 2.3)

Qh, test statistic for homogeneity

Table 10 Association between maternal smoking and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 9 26.0 8 0.001 1.6 (1.4, 1.8)

For personal use only.

Cohort studies 1 — — — 1.2 (0.9, 1.5)

Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
8 18.8 7 0.009 1.7 (1.5, 1.9)
8 18.8 7 0.009 1.7 (1.5, 1.9)
1 — — — 1.2 (0.9, 1.5)
3 6.0 2 0.049 1.1 (0.8, 1.5)
6 5.6 5 0.347 1.8 (1.6, 2.0)
6 14.2 5 0.014 1.6 (1.5, 1.8)
3 8.1 2 0.017 1.4 (0.7, 2.9)

Qh, test statistic for homogeneity

previous induced abortion, the risk of placenta previa was
higher in five studies with probable diagnosis of placenta
previa7,10,12,22,24 and in four well-designed studies7,10,12,22
(pooled OR 1.6) than in studies with a definite diagnosis of
placenta previa23,34,35 (pooled OR 1.3) and in four poorly
designed studies23,24,34,35 (pooled OR 1.2).
Smoking during pregnancy was associated with a 60%
increase in the risk of placenta previa (Table 10). There
was one foreign-based cohort study11 and there were eight
USA-based case–control studies7,10,12,13,22–24,34. The risk of
placenta previa was higher among smokers in six studies
with probable diagnosis of placenta previa7,10,12,13,22,24
(pooled OR 1.8) and in six well-designed studies7,10–13,22
(pooled OR 1.6) than in three studies11,23,34 with definite
diagnosis of placenta previa (pooled OR 1.1) and in three
poorly designed studies23,24,34 (pooled OR 1.4). Three
case–control studies in the USA 12,23,24 (Table 11) also
demonstrated a higher risk of placenta previa with cocaine
use during pregnancy (pooled OR 2.9). There were two
studies with probable diagnosis of placenta previa23,24
(pooled OR 3.0) and two poorly designed studies23,24
(pooled OR 5.2).
There were seven studies4,6,11,15,22,33,36 describing the
association of male fetuses with risk of placenta previa
(Table 12). There were four cohort11,15,33,36 and three
case–control studies4,6,22 with similar risk of placenta previa
(pooled OR 1.2 and 1.1, respectively). Male fetuses were
associated with higher risk of placenta previa in three
foreign-based studies6,11,15 (pooled OR 1.3) than in four
USA-based studies4,22,33,36 (pooled OR 1.1). With male
fetuses there was a higher risk of placenta previa in two
studies with definite diagnosis of placenta previa6,11
(pooled OR 1.5) than in five studies with probable diag-
nosis of placenta previa4,15,22,33,36 (pooled OR 1.2).
However, the risk was similar for three well-designed11,22,36
and four poorly designed studies4,6,15,33 (pooled OR 1.2).
There was a 50% increased risk of placenta previa due to
chronic hypertension in three studies33,34,37 (Table 13).

182
J ournal of Maternal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Table 11 Association between maternal cocaine use and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 3 2.0 2 0.368 2.9 (1.9, 4.3)

Cohort studies
Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
— — — — —
3 2.0 2 0.368 2.9 (1.9, 4.3)
3 2.0 2 0.368 2.9 (1.9, 4.3)
— — — — —
1 — — — 4.3 (0.3, 72.0)
2 2.0 1 0.157 3.0 (1.8, 5.0)
1 — — — 2.5 (1.6, 3.9)

Poorly designed studies 2 0.0 1 1.000 5.2 (0.7, 39.0)

Qh, test statistic for homogeneity

Table 12 Association between sex of the fetus and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 7 7.7 6 0.261 1.2 (1.1, 1.3)

Cohort studies 4 6.1 3 0.107 1.2 (1.1, 1.3)
For personal use only.

Case–control studies 3 1.2 2 0.549 1.1 (0.9, 1.4)

USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
4 1.0 3 0.801 1.1 (1.0, 1.2)
3 2.1 2 0.349 1.3 (1.2, 1.5)
2 0.4 1 0.527 1.5 (1.2, 1.8)
5 2.4 4 0.662 1.2 (1.1, 1.3)
3 6.0 2 0.049 1.2 (1.1, 1.3)
4 1.3 3 0.729 1.2 (1.1, 1.5)

Qh, test statistic for homogeneity

Table 13 Association between chronic hypertension and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 3 2.4 2 0.301 1.5 (0.8, 2.7)

Cohort studies
Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
2 0.4 1 0.527 1.2 (0.7, 2.4)
1 — — — 4.7 (1.0, 22.0)
2 2.3 1 0.129 1.6 (0.8, 3.3)
1 — — — 1.2 (0.4, 3.7)
2 2.0 1 0.157 2.0 (0.7, 6.4)
1 — — — 1.3 (0.6, 2.7)
1 — — — 1.2 (0.4, 3.7)
2 2.2 1 0.138 1.9 (0.7, 5.0)

Qh, test statistic for homogeneity

Two studies were cohort studies33,37 (pooled OR 1.2) and tension in two studies with definite placenta previa34,37
two were USA-based studies33,34 (pooled OR 1.6). There (pooled OR 2.0) and in two poorly designed studies33,34
was significant risk of placenta previa due to chronic hyper- (pooled OR 1.9).

183
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

Pregnancy-induced hypertension showed a significant DISCUSSION

protective effect on the risk of placenta previa (Table 14).
Placenta previa is an obstetric condition that leads to
All three studies that examined pregnancy-induced
serious maternal and fetal complications. This condition
hypertension were USA-based cohort studies37–39 (pooled
has been reported to occur in 3–20 per 1000 pregnancies.
OR 0.4). All three were well-designed studies and
The wide variation in the prevalence rate can be attributed
pregnancy-induced hypertension had a similar protective
to inclusion of varying degrees of placenta previa, different
effect in two studies37,38 with a definite diagnosis and
methods and timing of diagnosis, and diversity of the
in one study39 with a probable diagnosis of placenta previa
patient population across studies. Many clinical and epi-
(pooled OR 0.3).
demiological studies have reported the rate of placenta
Pre-eclampsia also showed a protective effect on the
previa to be higher among older and multiparous women,
risk of placenta previa (Table 15). All three studies were
those with a previous Cesarean delivery, prior spontaneous
USA-based studies33,34,40 (pooled OR 0.9). There was one
or induced abortion as well as among women who smoked
case–control study34 and two cohort studies33,40. All
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14

or used cocaine during pregnancy.

studies were poorly designed and one study33 with a
This meta-analysis confirms the higher risk of placenta
probable diagnosis of placenta previa showed that the risk
previa associated with advancing maternal age. Among
of placenta previa increases with pre-eclampsia (pooled OR
older women, there may be atherosclerotic changes in the
1.3). However, studies with a definite diagnosis of placenta
uterine blood vessels causing compromised uteroplacental
previa34,40 confirmed the protective effect of pre-eclampsia
blood flow. This has been shown by microscopic studies of
on the risk of placenta previa (pooled OR 0.7).

Table 14 Association between pregnancy-induced hypertension and placenta previa

Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)
For personal use only.

Overall 3 4.8 2 0.091 0.4 (0.2, 0.5)

Cohort studies
Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
— — — — —
3 4.8 2 0.091 0.4 (0.2, 0.5)
— — — — —
3 4.8 2 0.091 0.4 (0.2, 0.5)
2 4.7 1 0.030 0.3 (0.2, 0.6)
1 — — — 0.3 (0.3, 1.0)
3 4.8 2 0.091 0.4 (0.2, 0.5)
— — — — —

Qh, test statistic for homogeneity

Table 15 Association between pre-eclampsia and placenta previa. Reprinted from reference 103, Ó 2003, with permission of
Elsevier Science
Test of homogeneity of pooled odds ratios Random-effects
pooled odds ratio
2
Number of studies c Qh Degrees of freedom p Value (95% CI)

Overall 3 4.0 2 0.135 0.9 (0.5, 1.4)

Cohort studies
Case–control studies
USA-based studies
Foreign-based studies
Definite previa
Probable previa
Well-designed studies
Poorly designed studies
2 4.0 1 0.045 0.8 (0.3, 2.1)
1 — — — 0.9 (0.5, 1.7)
3 4.0 2 0.135 0.9 (0.5, 1.4)
— — — — —
2 2.0 1 0.157 0.7 (0.4, 1.2)
1 — — — 1.3 (0.7, 2.7)
— — — — —
3 4.0 2 0.135 0.9 (0.5, 1.4)

Qh, test statistic for homogeneity

184
J ournal of Maternal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa
placentae from older women that have revealed utero-
placental underperfusion and large placental infarcts8. To
maintain optimal blood flow, an increased surface area may
be required for placental attachment, and this may result in
placental encroachment on the lower uterine segment8.
Our results also show a greater risk of placenta previa with
higher parity, confirming findings from earlier studies. This
may be due to endometrial scarring at the site of prior
placental attachments resulting in lower placental
implantation. The other possibility may be that blood
vessels at the sites of prior placental attachments undergo
changes that may lead to decreased uteroplacental blood
flow3. This, in turn, may result in a larger placenta
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
encroaching on the cervical os with repeated pregnancies.
In this meta-analysis we analyzed only maternal age and
parity as independent risk factors for placenta previa.
However, since women with higher parity are likely to
be older, it is possible that advanced maternal age and
increased parity are not independent risk factors for the
placenta previa risk. This combined effect of age and
gravidity on the risk of placenta previa was demonstrated in
a large (n = 37 956 020), population-based, cohort study
of singleton births from the USA (1989–98)103. This
study showed that the risk of placenta previa was not
For personal use only.
independent of maternal age and parity, but rather that
both factors exerted a joint influence on placenta previa
risk. In other words, increasing maternal age and increasing
parity conferred the greatest risk of placenta previa
compared with that of primigravid women aged < 20 years
(Table 16).
There is an increased risk of placenta previa among
women with a history of previous Cesarean delivery and
previous abortion. Our meta-analysis corroborates these
general findings reported in several previous studies as well
as a meta-analysis104. Damage and scarring to the endo-
metrial and myometrial lining during Cesarean delivery
and spontaneous and induced abortion are known to pre-
Table 16 Joint effects of maternal age and gravidity on the risk of placenta previa: adjusted relative risks with 95% confidence intervals
Gravidity (number of pregnancies)
Maternal age (years) 1 2
< 20 1.00 (referent) 1.72 (1.57, 1.89)
20–24 1.61 (1.50, 1.72) 2.21 (2.07, 2.37)
25–29 2.82 (2.63, 3.00) 3.59 (3.37, 3.83)
30–34 4.71 (4.41, 5.03) 5.24 (4.92, 5.59)
35–39 7.30 (6.78, 7.86) 7.87 (7.35, 8.44)
³ 40 10.41 (9.25, 11.72) 9.95 (8.93, 11.08)
Table reprinted from reference 103, Ó 2003, with permission of Elsevier Science
Relative risks were adjusted for the confounding effects of maternal anemia, intrapartum fever, preterm premature rupture of
membranes, hydramnios, maternal diabetes, placental abruption, unexplained uterine bleeding and male sex
Faiz and Ananth
dispose to the low implantation of the placenta in the
uterus.
There is a higher risk of placenta previa with cigarette
smoking and maternal cocaine use during pregnancy, as
previously reported. Placental enlargement has been noted
among women who smoke cigarettes and this has been
attributed to the vasoactive properties of nicotine and to
chronic hypoxia associated with carbon monoxide22,62,63. It
has also been observed that there are chronic hypoxic
changes in the uterine vasculature of smokers, resulting in a
larger placenta with increased likelihood of placental
encroachment on the cervical os22,74. Similarly, maternal
cocaine use is known to cause catecholamine-mediated
vasoconstriction and vasospasm in blood vessels innervated
by the sympathetic nervous system. This is likely to result
in underperfusion of the uteroplacental vessels and a larger
placenta encroaching on the cervical os12,22.
This meta-analysis confirms an increased male/female
ratio at birth among women with placenta previa. It has
been proposed that early and late insemination during the
menstrual cycle may cause an increase in the conception of
males as well as lower implantation of the placenta15. With
early insemination it may be possible that the embryo
reaches the lower uterine segment before the endometrial
lining is ready for implantation. Similarly, with late insemi-
nation, the ovum may be in the lower uterine segment
when it is fertilized, resulting in lower uterine implantation
in both cases.
There is an increased frequency of placenta previa
among women with pre-existing or chronic hypertension.
The exact mechanism that leads to lower implantation of
the placenta among women with chronic hypertension is
not clear. However, placenta previa has a protective effect
on the risk of pregnancy-induced hypertension and
pre-eclampsia. Although the precise mechanism is unclear,
it has been suggested that, owing to the wider diameter and
less restricted course of blood vessels, there is better
3 4 ³5
2.13 (1.86, 2.44) 2.87 (2.28, 3.60) 3.00 (2.05, 4.37)
2.99 (2.79, 3.22) 3.66 (3.37, 3.98) 4.66 (4.26, 5.08)
4.31 (4.04, 4.61) 4.99 (4.65, 5.36) 6.22 (5.80, 6.67)
6.02 (5.65, 6.43) 6.52 (6.08, 6.98) 7.82 (7.31, 8.35)
8.38 (7.82, 8.98) 9.24 (8.59, 9.94) 10.01 (9.35, 10.72)
11.96 (10.80, 13.24) 12.16 (10.90, 13.57) 12.62 (11.59, 13.74)
185
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa Faiz and Ananth

oxygenation of the placenta implanted in the lower uterine
segment48. With higher implantation of the placenta in
the uterine cavity there may be restricted blood flow,
causing hypoxia and release of vasoactive substances into
the blood stream, resulting in a greater risk of pregnancy-
induced hypertension and pre-eclampsia. A better blood
supply and oxygenation of the placenta in the lower uterine
segment prevents the release of vasoactive substances
into the blood stream and thus reduces the risk of
pregnancy-induced hypertension and pre-eclampsia in
cases of placenta previa47,48. An alternative hypothesis
suggests that venous drainage from the placenta implanted
in the fundal part of the uterus is through ovarian veins,
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
strength of association of all the risk factors with placenta
previa and directionality of association with population-
attributable risk are presented in Table 17.
The population-attributable risk for placenta previa
indicates that smoking has the greatest effect (26%)
followed by previous abortion (16%) and Cesarean delivery
(10%). This implies that, if all women were to quit smoking
during pregnancy, 26% of placenta previa cases would
potentially be preventable. As cigarette smoking is pre-
ventable, it would be beneficial to encourage pregnant
women to quit smoking, especially those women suspected
of being at increased risk for placenta previa. Similarly,
Cesarean deliveries and abortions should be performed

causing visceral congestion and vasoconstriction, resulting
in pre-eclampsia. However, when the placenta is implanted
in the lower uterine segment, drainage is through uterine
veins and there is no visceral congestion; therefore,
placenta previa has a protective effect on pre-eclampsia
and pregnancy-induced hypertension105.
There are a number of risk factors that increase the like-
lihood of developing placenta previa. These include
advanced maternal age, multiparity, smoking during
pregnancy, alcohol and drug use during pregnancy, prior
Cesarean delivery and abortion, multiple pregnancy,
For personal use only.
with more caution and only in cases where there is
evidence of fetal or maternal distress, in order to reduce
the risk of placenta previa in future pregnancies.
In this study, we attempted to identify all published
studies between 1966 and March 2000. However, in spite of
our best efforts we may have missed some that may have
reported data on placenta previa. To account for this bias,
the random-effects analysis formed the basis of pooling of
data across studies. Thus, we assume that studies included
in our meta-analysis are a (random) sample from a larger
population of similar studies. In this meta-analysis, each

prior placenta previa, maternal anemia and diabetes,
hydramnios, placental abruption and chronic hyper-
tension. Although we calculated pooled odds ratios for all
the risk factors for which data were available, there were
not enough data for the analysis of certain risk factors. The
study was individually reviewed and scored by the two
authors on the basis of established criteria for study quality
and for method of diagnosis of placenta previa.
The careful review of studies indicated that there is a
wide variation in the definition of placenta previa and

Table 17 Strength and direction of association between various risk factors and conditions associated with placenta previa
Associations
Population-attributable risk
Risk factors for placenta previa Strength Directionality (%)

Advanced maternal age (³ 35 years)* +++ ­ —

Multiparity (parity ³ 3)* +++ ­ —
Smoking during pregnancy + ­ 25.7
Alcohol use during pregnancy* + ­ —
Drug use during pregnancy + ­ 6.7
Prior Cesarean delivery ++ ­ 10.2
Prior abortion + ­ 16.0
Multiple pregnancy* + ­ —
Prior placenta previa† + ­ —
Maternal anemia* ++ ­ 2.7
Maternal diabetes* + ­ —
Hydramnios* + ­ 0.3
Placental abruption* ++ ­ 1.4
Chronic hypertension + ­ —
Pre-eclampsia/PIH - ¯ —

PIH, pregnancy-induced hypertension

*Data from reference 103
†
Data from references 21 and 61

186
J ournal of Maternal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa
method used for the diagnosis of placenta previa across
studies. This ranged from studies that did not give any
definition of placenta previa or method used for the
diagnosis of placenta previa to studies that explicitly
defined placenta previa and included only those women
with the placenta completely covering the internal cervical
os and who were diagnosed at the time of Cesarean
delivery. Similarly, risk factors including advanced mater-
nal age, higher parity, prior Cesarean delivery and the
potential confounders for placenta previa were controlled
in varying degrees across studies. This ranged from studies
that controlled for all potential confounders to studies that
had no control group. These discrepancies make it difficult
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
to compare the results of these individual studies.
In conclusion, meta-analysis is a useful technique not
only for quantitatively summarizing the results but also for
calculating the strength of association between placenta
previa and various risk factors. However, we recommend
that future studies should confirm a diagnosis of placenta
previa through sonographic visualization of the placental
position in the lower uterine segment, either partially or
completely covering the internal cervical os. In the absence
of such a diagnosis, placenta previa must be confirmed at
the time of Cesarean section. Potential confounders should
For personal use only.
also be controlled adequately in order to minimize possible
biases in the study findings. This will help to clarify the
epidemiology of placenta previa and its impact on adverse
maternal and perinatal outcomes.
ACKNOWLEDGEMENTS
At the time of the study, Ambarina Faiz MD, MPH was
a Primary Care/Health Services research fellow in the
Department of Family Medicine, UMDNJ, and was
supported through an NRSA Fellowship award (T32PE-
10011) from the Health Resources and Services Adminis-
tration. Cande Ananth, PhD, MPH is partly supported
through a grant (HD-38902) awarded to him from the
National Institutes of Health.
This study formed part of Dr Faiz’s post-doctoral
research under the direction of Dr Ananth.
REFERENCES
1. Cunningham FG, Gant NF, Leveno KJ, et al. Obstetrical
hemorrhage. In Williams Obstetrics. New York: McGraw
Hill, 1997:619– 69
2. Crane JM, Van Den Hof MC, Dodds L, et al. Neonatal
outcomes with placenta previa. Obstet Gynecol 1999;
93:541–4
3. Naeye RL. Placenta previa: predisposing factors and
effects on the fetus and the surviving infants. Obstet
Gynecol 1978;52:521– 5
Faiz and Ananth
4. Wolf EJ, Mallozzi A, Rodis JF, et al. Placenta previa is not
an independent risk factor for a small for gestational age
infant. Obstet Gynecol 1991;77:707– 9
5. Neri A, Gorodesky I, Bihari C, et al. Impact of placenta
previa on intrauterine fetal growth. Isr J Med Sci 1980;
16:429–32
6. Parazzini F, Dindelli M, Luchini L, et al. Risk factors
for placenta previa. Placenta 1994;15:321– 6
7. Taylor VM, Kramer MD, Vaughan TL, et al. Placenta
previa in relation to induced and spontaneous abortion:
a population-based study. Obstet Gynecol 1993;82:88– 91
8. Williams MA, Mittendorf R. Increasing maternal age as a
determinant of placenta previa: more important than
increasing parity? J Reprod Med 1993;38:425– 8
9. Niswander KR, Friedman EA, Hoover DB, et al. Fetal
morbidity following potentially anoxigenic obstetric
conditions. Am J Obstet Gynecol 1966;95:846– 52
10. Kramer MD, Taylor V, Hickok DE, et al. Maternal
smoking and placenta previa. Epidemiology 1991;2:221– 3
11. Ananth CV, Savitz DA, Luther ER. Maternal cigarette
smoking as a risk factor for placental abruption, placenta
previa, and uterine bleeding in pregnancy. Am J Epidemiol
1996;144:881– 9
12. Handler AS, Mason ED, Rosenberg DL, et al. The
relationship between exposure during pregnancy to
cigarette smoking and cocaine use and placenta previa.
Am J Obstet Gynecol 1994;170:884– 9
13. McMahon MJ, Li R, Schenck AP, et al. Previous cesarean
birth – a risk factor for placenta previa? J Reprod Med
1997;42:409– 12
14. Clark SL, Koonings PP, Phelan JP. Placenta previa/
accreta and previous cesarean section. Obstet Gynecol
1985;66:89– 92
15. MacGillivary I, Davey D, Isaacs S. Placenta previa and
sex ratio at birth. Br Med J 1986;292:371– 2
16. Zhang J, Savitz DA. Maternal age and placenta previa:
a population-based case–control study. Am J Obstet
Gynecol 1993;168:641– 5
17. Gorodesky IG, Bahari CM. The effect of placenta previa
localization upon maternal and fetal–neonatal outcome.
J Perinat Med 1987;15:169– 77
18. Abu-Heija AT, El-Jallad F, Ziadeh S. Placenta previa:
effect of age, gravidity, parity and previous cesarean
section. Gynecol Obstet Invest 1999;47:6– 8
19. Makhseed M, El-Tomi N, Moussa M. A retrospective
analysis of pathological placental implantation-site and
penetration. Int J Gynecol Obstet 1994;47:127– 34
20. Rose GL, Chapman MG. Aetiological factors in placenta
previa – a case controlled study. Br J Obstet Gynaecol
1986;93:586– 8
21. Gorodesky I, Bahari CM, Schacter A, et al. Recurrent
placenta previa. Eur J Obstet Gynaecol Reprod Biol 1981;
12:7–11
22. Williams MA, Mittendorf R, Leiberman E, et al. Cigarette
smoking during pregnancy in relation to placenta previa.
Am J Obstet Gynecol 1991;165:28– 32
187
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa
23. Chelmow D, Andrew DE, Baker ER. Maternal cigarette
smoking and placenta previa. Obstet Gynecol 1996;
87:703–6
24. Macones GA, Sehdev HM, Parry S, et al. The association
between maternal cocaine use and placenta previa. Am J
Obstet Gynecol 1997;177:1097– 100
25. Chattopadhyay SK, Kharif H, Sherbeeni MM. Placenta
previa and accreta after previous cesarean section. Eur J
Obstet Gynecol Reprod Biol 1993;52:151– 6
26. Hershkowitz R, Fraser D, Mazor M, et al. One or multiple
previous cesarean sections are associated with similar
increased frequency of placenta previa. Eur J Obstet
Gynecol Reprod Biol 1995;62:185– 8
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
27. Singh PM, Rodrigues C, Gupta AN. Placenta previa and
previous cesarean section. Acta Obstet Gynecol Scand
1981;60:367– 8
28. Nielsen TF, Hagberg H, Ljungblad U. Placenta previa and
antepartum hemorrhage after previous cesarean section.
Gynecol Obstet Invest 1989;27:88– 90
29. Zaideh SM, Abe-Heija AT, El Jallad MF. Placenta previa
and accreta: analysis of a two year experience. Gynecol
Obstet Invest 1998;46:96– 8
30. Sauer M, Parsons M, Sampson M. Placenta previa: an
analysis of three years experience. Am J Perinat 1985;
141:239–42
For personal use only.
31. Takayama T, Minakami H, Koike T, et al. Risks associated
with cesarean section in women with placenta previa.
J Obstet Gynaecol Res 1997;23:375– 9
32. Hendricks MS, Chow YH, Bhagavath B, et al. Previous
cesarean section and abortion as risk factors for developing
placenta previa. J Obstet Gynaecol Res 1999;25:137– 42
33. Brenner WE, Edelman DA, Hendricks CH. Characteris-
tics of patients with placenta previa and results of
expectant management. Am J Obstet Gynecol 1978;132:
180–9
34. Newton ER, Barss V, Cetrulo CL. The epidemiology and
clinical history of asymptomatic midtrimester placenta
previa. Am J Obstet Gynecol 1984;148:743– 8
35. Grimes DA, Techman T. Legal abortion and placenta
previa. Am J Obstet Gynecol 1984;149:501– 4
36. Demissie K, Breckenridge MB, Joseph L, et al. Placenta
previa: preponderence of male sex at birth. Am J Epidemiol
1999;149:824– 30
37. Ananth CV, Bowes WA Jr, Savitz DA, et al. Relationship
between pregnancy-induced hypertension and placenta
previa: a population based-study. Am J Obstet Gynecol
1997;177:997– 1002
38. Nicolaides KH, Faratian B, Symonds EM. Effect on low
implantation of the placenta on maternal blood pressure
and placental function. Br J Obstet Gynaecol 1982;89:
806–10
39. Wang JC, Hin LY, Ng KB. Pregnancy induced hyper-
tension and placenta previa: a racial and geographical
perspective. Int J Gynecol Obstet 1999;67:177– 8
188
J ournal of Maternal–Fetal and Neonatal Medicin e
Faiz and Ananth
40. Jelsema RD, Bhati RK, Zadore IE, et al. Is placenta previa
a determinant of preeclampsia? J Perinat Med 1991;
19:485–8
41. Iyasu S, Saftlas AK, Rowley DL, et al. The epidemiology
of placenta previa in the United States, 1979 through
1987. Am J Obstet Gynecol 1993;168:1424– 9
42. Jakobovits AA, Zubek L. Sex ratio and placenta previa.
Acta Obstet Gynecol Scand 1989;68:503– 5
43. Crenshaw C Jr, Darnell Jones DE, Parker RT. Placenta
previa: a survey of twenty years experience with improved
perinatal survival by expectant therapy and cesarean
delivery. Obstet Gynecol Surv 1973;28:461– 70
44. Frederickson MC, Glassenberg R, Stika CS. Placenta
previa: a 22 year analysis. Am J Obstet Gynecol 1999;
180:1432– 7
45. Parson MT, Winegar A, Siefert L, et al. Pregnancy out-
comes in short women. J Reprod Med 1989;34:357– 61
46. McShane PM, Heyl PS, Epstein MF. Maternal and
perinatal morbidity resulting from placenta previa. Obstet
Gynecol 1985;65:165– 82
47. Khashoggi T. Maternal and neonatal outcome in major
placenta previa. Ann Saudi Med 1995;15:313– 16
48. Leiberman JR, Fraser D, Kasis A, et al. Reduced frequency
of hypertensive disorders in placenta previa. Obstet
Gynecol 1991;77:83– 6
49. Hemminki E, Glebatis DM, Therriault GD, et al.
Incidence of placenta previa and abruptio placentae in
New York State. NY State J Med 1987;87:594– 8
50. Barrett JM, Boehm FH, Killam AP. Induced abortion: a
risk factor for placenta previa. Am J Obstet Gynecol 1981;
141:769–72
51. Love CDB, Wallace EM. Pregnancies complicated by
placenta previa: what is appropriate management? Br J
Obstet Gynaecol 1996;103:864– 7
52. Chervenak FA, Lee Y, Handler MA, et al. Role of
attempted vaginal delivery in the management of placenta
previa. Obstet Gynecol 1984;64:798– 801
53. Silver R, Depp R, Sabbagha RE, et al. Placenta previa:
aggressive expectant management. Am J Obstet Gynecol
1984;150:15– 22
54. Cotton DB, Read JA, Paul RH, et al. The conservative
aggressive management of placenta previa. Am J Obstet
Gynecol 1980;137:687– 95
55. D’Angelo LJ, Irwin LF. Conservative management of
placenta previa: a cost–benefit analysis. Am J Obstet
Gynecol 1984;149:320– 26
56. Rosen DMB, Peek MJ. Do women with placenta previa
without antepartum hemorrhage require hospitalization?
Aust NZ J Obstet Gynaecol 1994;34:130– 4
57. Hibbard LT. Placenta previa. Am J Obstet Gynecol 1969;
104:172–84
58. Dutta DC, Basu G, Konar H. Caesarean section in
placenta praevia in rural obstetric practice. J Ind Med
Assoc 1980;75:176– 9
 Etiology and risk factors for placenta previa
59. Hill DJ, Beisher NA. Placenta praevia without ante-
partum haemorrhage. Aust NZ J Obstet Gynaecol 1980;
20:21–3
60. Gabert HA. Placenta previa and fetal growth. Obstet
Gynecol 1971;38:403– 6
61. Monica G, Lilja C. Placenta previa, maternal smoking and
recurrence risk. Acta Obstet Gynecol Scand 1995;74:341– 5
62. Naeye RL. Maternal age, obstetric complications and
outcome of pregnancy. Obstet Gynecol 1983;61:210– 16
63. Christianson R. Gross differences observed in placenta of
smokers and non-smokers. Am J Epidemiol 1979;110:
178–87
64. Naeye R. Do placental weights have clinical significance?
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
Hum Pathol 1987;18:387– 91
65. Taylor VM, Peacock S, Kramer MD, et al. Increased risk of
placenta previa among women of Asian origin. Obstet
Gynecol 1995;86:805– 8
66. Strong TH, Brar HS. Placenta previa in twin gestations.
J Reprod Med 1989;34:415– 16
67. Zhang J, Fried DB. Relationship of maternal smoking
during pregnancy to placenta previa. Am J Prev Med
1992;8:278– 82
68. Gorodeski IG, Neri A, Bahary CM. Placenta previa – the
identification of low- and high-risk subgroups. Eur J Obstet
Gynecol Reprod Biol 1985;20:133– 43
For personal use only.
69. Ancona S, Chatterjee M, Rhee I, et al. The mid-trimester
placenta previa: a prospective follow-up. Eur J Radiol
1990;10:215– 16
70. Ruparelia BA, Chapman MG. Early low-lying placentae–
ultrasonic assessment, progress and outcome. Eur J Obstet
Gynecol Reprod Biol 1985;20:209– 13
71. Miller DA, Chollet JA, Goodwin TM. Clinical risk
factors for placenta previa–placenta accreta. Am J Obstet
Gynecol 1997;177:210– 14
72. Larvey JP. Placenta previa. Clin Obstet Gynecol 1990;
33:414–21
73. Oppenheimer LW, Farine D, Knox Ritchie JW, et al.
What is low lying placenta? Am J Obstet Gynecol 1991;
165:1036– 8
74. Naeye RL. Effects of maternal cigarette smoking on the
fetus and placenta. Br J Obstet Gynaecol 1978;85:732– 7
75. Naeye RL. Abruptio placenta and placenta previa:
frequency, perinatal mortality, and cigarette smoking.
Obstet Gynecol 1980;55:701– 4
76. Taylor VM, Kramer MD, Vaughan TL, et al. Placenta
previa and prior cesarean delivery: how strong is the asso-
ciation? Obstet Gynecol 1994;84:55– 7
77. Rizos N, Doran TA, Miskin M, et al. Natural history of
placenta previa ascertained by diagnostic ultrasound.
Am J Obstet Gynecol 1979;133:287– 91
78. Biswas R, Sawhney H, Dass R, et al. Histopathological
study of placental bed biopsy in placenta previa. Acta
Obstet Gynecol Scand 1999;78:173– 9
79. To WWK, Leung WC. Placenta previa and previous
cesarean section. Int J Gynecol Obstet 1995;51:25– 31
Faiz and Ananth
80. Yaegashi N, Cheba-Sekii A, Okamura K. Emergency
postpartum hysterectomy in women with placenta previa
and prior cesarean section. Int J Gynecol Obstet 2000;
68:49–80
81. Nelson HB, Huston JE. Placenta previa a possible solution
to the associated high fetal mortality. J Reprod Med 1971;
7:188–94
82. Robertson JG. Placenta praevia. Br Med J 1969;204:
383–92
83. Sand H. Ultrasound diagnosis of placenta praevia. Acta
Obstet Gynecol Scand 1977;56:109– 13
84. Comeau J, Shaw L, Marcell CC, et al. Early placenta previa
and delivery outcome. Obstet Gynecol 1983;61:577– 80
85. Dunstone SJ, Leibowitz CB. Conservative management of
placenta praevia with a high risk of placenta accreta. Aust
NZ J Obstet Gynaecol 1998;38:429– 33
86. Wing DA, Paul RH, Miller LK. Management of the
systematic placenta previa: a randomized, controlled trial
of inpatient versus outpatient expectant management.
Am J Obstet Gynecol 1996;75:806– 11
87. Besinger RE, Moniak CW, Paskiewicz LS, et al. The effect
of tocolytic use in the management of symptomatic
placenta previa. Am J Obstet Gynecol 1995;172:1777– 80
88. Mabie WC. Placenta previa. Clin Perinatol 1992;19:
425–35
89. Andres RL. The association of cigarette smoking with
placenta previa and abruptio placentae. Semin Perinatol
1996;20:154– 9
90. Magaan EF, Johnson CA, Gookin KS, et al. Placenta
previa: does uterine activity causes bleeding? Obstet
Gynecol Surv 1994;49:77– 8
91. Ballas S, Gitstein S, Jaffa AJ, et al. Midtrimester placenta
previa: normal or pathologic finding. Obstet Gynecol
1979;54:12– 14
92. Varma TR. The implication of a low implantation of
the placenta detected by ultrasonography in early
pregnancy. Acta Obstet Gynecol Scand 1981;60:265– 8
93. Zelop CC, Bromley B, Frigoletto FD Jr, et al. Second
trimester sonographically diagnosed placenta previa:
prediction of persistent previa at birth. Int J Gynecol
Obstet 1994;44:207– 10
94. Wexler P, Gottesfeld KR. Early diagnosis of placenta
previa. Obstet Gynecol 1979;54:231– 4
95. Wexler P, Gottesfeld KR. Second trimester placenta
previa: an apparently normal placentation. Obstet Gynecol
1977;50:706– 9
96. Baron F, Hill WC. Placenta previa, placenta abruptio.
Clin Obstet Gynecol 1998;41:527– 32
97. Bergsjo P. Acta sixty years ago: vaginal cesarean section in
cases of placenta previa. Acta Obstet Gynecol Scand
1990;70:723– 77
98. Crane JM, Van den Hof MC, Dodds L, et al. Maternal
complications with placenta previa. Am J Perinatol 2000;
17:101–5
189
J ournal of Mater nal–Fetal and Neonatal Medicin e
 Etiology and risk factors for placenta previa
99. Vengadasalam D, Khong CC. Maternal and perinatal
outcome resulting from placenta previa – a 5-year
retrospective survey in Alexandra Hospital. Singapore J
Obstet Gynecol 1989;20:20– 2
100. Abu-Heija AT, Jallad MF, Abukteish F. Maternal and
perinatal outcome of pregnancies after the age of 45.
J Obstet Gynaecol Res 2000;26:27– 30
101. Ananth CV, Wilcox AJ, Savitz DA, et al. Effect of
maternal age and parity on the risk of uteroplacental
bleeding disorders in pregnancy. Obstet Gynecol 1996;88:
511–16
102. Greenland S, Longnecker MP. Methods for trend estima-
tion from summarized dose response data, with application
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Columbia University on 10/14/14
to meta-analysis. Am J Epidemiol 1992;135:1301– 8
103. Ananth CV, Demissie K, Smulian JC, et al. Placenta
previa in singleton and twin births in the United States,
For personal use only.
190
J ournal of Maternal–Fetal and Neonatal Medicin e
View publication stats
Faiz and Ananth
1989–98: a comparison of risk factor profiles and asso-
ciated conditions. Am J Obstet Gynecol 2003;188:275– 81
104. Ananth CV, Smulian JC, Vintzileos AM. The association
of placenta previa with history of cesarean delivery and
abortion: a meta-analysis. Am J Obstet Gynecol 1997;
177:1071– 8
105. Bieniarz J. The patho-mechanism of late pregnancy
toxemia and obstetrical hemorrhages: contraindication in
the clinical pictures of eclampsia and placenta previa
depending upon the placental site. Am J Obstet Gynecol
1958;75:444– 53
106. Ananth CV, Smulian JC, Vintzileos AM. The effect of
placenta previa on neonatal mortality: a population-based
study in the United States, 1989–97. Am J Obstet Gynecol
2003;188:in press