SMFM Papers

www. AJOG.org

Identifying the causes of stillbirth: a comparison

of four classification systems
Patrizia Vergani, MD; Sabrina Cozzolino, MD; Elisa Pozzi, MD; Maria Serena Cuttin, MD;
Massimiliano Greco, MD; Sara Ornaghi, MD; Valeria Lucchini, MD
OBJECTIVE: To identify the classification protocol for stillbirth that
minimizes the rate of unexplained causes.
STUDY DESIGN: All stillbirths at 22 weeks from 1995-2007 underwent a workup inclusive of fetal ultrasonography, amniocentesis for
karyotype and cultures, placental histology, fetal autopsy, skin biopsy,
total body X-ray, maternal testing for thrombophilias, TORCH, Parvovirus spp, thyroid function, indirect Coombs, Kleiheuer-Betke test, and
genital cultures. To such a cohort, we applied the 4 most commonly
used classification protocols.

RESULTS: The stillbirth rate during the study period was 0.4% (154/
37,958). The RoDeCo classification provided the lowest rate of unexplained stillbirth (14.3%) compared with Wigglesworth (47.4%), de Galan-Roosen (18.2%), and Tulip (16.2%) classifications. Mean gestational
age at stillbirth in unexplained vs explained stillbirth was similar in the 4
protocols.
CONCLUSION: Adoption of a consistent and appropriate workup protocol can reduce the rate of unexplained stillbirth to 14%.

Key words: Cause of death, classification systems, stillbirth

Cite this article as: Vergani P, Cozzolino S, Pozzi E, et al. Identifying the causes of stillbirth: a comparison of 4 classification systems. Am J Obstet Gynecol 2008;
199:319.e1-319.e4.

tillbirths are one of the most common adverse pregnancy outcomes,

with estimated rates of 5.3 per 1000 deliveries in developed countries and 25.5
per 1000 in developing countries.1
Despite the intensification of obstetric surveillance, the stillbirth rate at
22 weeks has remained constant for the
past 3 decades. Detection of causes of
stillbirths is important to identify deficiencies in the provision of care, to focus attention on areas in which improvements are possible, and to
indicate areas in which new developments or knowledge may be expected
to lead to preventive measures to lower
perinatal mortality.2 However, classifi-

From the Departments of Obstetrics and

Gynecology (Drs Vergani, Cozzolino, Pozzi,
Greco, and Ornaghi) and Pathology (Drs
Cuttin and Lucchini), University of MilanoBicocca, Monza, Italy.
Presented at the 28th Annual Meeting of the
Society for Maternal-Fetal Medicine, Dallas, TX,
Jan. 28-Feb. 2, 2008.
Received Feb. 28, 2008; revised May 1, 2008;
accepted June 30, 2008.
Reprints not available from the author.
0002-9378/$34.00
2008 Mosby, Inc. All rights reserved.
doi: 10.1016/j.ajog.2008.06.098

cation of causes of stillbirth is often

ambiguous because of the complicated
pathophysiologic processes encountered in the mother, fetus, and placenta, and the fact that stillbirths often
result from interaction of different
processes. The classification systems
proposed differ in approaches, definitions, and levels of complexity.3-8
However, no single system is universally accepted and the comparison
among systems is difficult.
The aim of this study is to compare
different classifications systems on a cohort of stillbirths undergoing a consistent and comprehensive workup to establish which classification minimizes
rates of unexplained stillbirth, that is
unknown cases of fetal death.

M ATERIALS AND M ETHODS

We conducted a retrospective study on a
cohort of stillbirths diagnosed at the Department of Obstetrics and Gynecology
of the San Gerardo Hospital. Monza, Italy, between January 1995 and December 2007. The diagnosis of stillbirth was
based on the World Health Organization (WHO) recommendations and
was defined as fetal death at 22 weeks of
gestation or greater, or birthweight
500 g if the gestational age was un-

known.9,10 Patient information recorded included date of delivery, gestational age, maternal demographic
characteristics, including body mass
index, neonatal sex and birthweight,
and pregnancy details.
During the study period, all stillbirths
underwent evaluation according to a
uniform and comprehensive protocol
that included fetal ultrasonography, amniocentesis for karyotype and cultures,
placental histology, autopsy, skin biopsy,
total body X ray, maternal testing for inherited and acquired thrombophilias,
TORCH, Parvovirus spp, thyroid function, indirect Coombs, Kleiheuer-Betke
test, and genital cultures. Growth restriction was defined as fetal weight below the
10th customized percentile, using the
gestation-related optimal weight software (GROW), version 7.4.2 (www.
gestation.net), which calculates the fetal
growth potential by adjusting for the fetal sex and parental constitutional characteristic known at the beginning of the
pregnancy.
To our cohort, we applied 4 classification systems published in the obstetric
and pathology literature during the past
20 years for the identification of causes of
stillbirth.3-8 A panel of 2 obstetricians reviewed all stillbirth cases and classified

SEPTEMBER 2008 American Journal of Obstetrics & Gynecology

319.e1

SMFM Papers

www.AJOG.org

TABLE 1

Contribution of individual tests in the identification

of causative or risk factors for stillbirths
Test

Placental histology

119

77.3

Severe lesions

75

48.7

Mild lesions

44

28.6

.....................................................................................................................................................................................................................................
a
.....................................................................................................................................................................................................................................
b

..............................................................................................................................................................................................................................................

Autopsy findings (with normal karyotype)

38

24.7

Abnormal karyotype

18

11.7

Total body X- ray

Skin biopsy

29

18.8

0.6

44

28.6

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

TORCH, Parvovirus spp, and genital cultures

..............................................................................................................................................................................................................................................

Kleiheuer-Betke test

..............................................................................................................................................................................................................................................

Maternal testing for inherited and acquired thrombophilias

..............................................................................................................................................................................................................................................

Total

154

..............................................................................................................................................................................................................................................
a

Severe lesions: severe vascular or inflammatory lesions, placental abruptio, true knots of the umbilical cord, thrombosis
of umbilical vessels.

Mild lesions: mild vascular or inflammatory lesions.

Vergani. Identifying the causes of stillbirth. Am J Obstet Gynecol 2008.

the causes according to the 4 selected

classification systems.

R ESULTS
During the 12 years of the study period,
there were 154 stillbirths of a total of
37,958 births, yielding a stillbirth rate of
4 per 1000. Clinical records were available for all deaths. Median gestational
age at delivery was 31 weeks and 2 days
(range, 22-42 weeks). Median age of the
mothers was 31 years (range, 18-43

years). Autopsy and placental examination were performed in 152 (98.7%)

cases.
The contribution of individual tests in
the detection of risk or causative factors
for stillbirth is delineated in Table 1.
Tables 2-5 display the distribution of
causes of death in our cohort according
to the classifications of Wigglesworth,3
de Galan-Roosen,6 ReCoDe,7 and Tulip.8 The extended Wigglesworth classification, which is the classification

TABLE 2

Extended Wigglesworth classification: application to current study

Code

Classification

154

..............................................................................................................................................................................................................................................

Congenital defect/malformation (lethal or severe)

55

35.7

Unexplained antepartum fetal death

70

45.5

Death from intrapartum asphyxia, anoxia, or trauma

0.6

Immaturity

1.3

Infection

10

6.5

Death caused by other specific causes

13

8.4

Death caused by accident or nonintrapartum causes

Sudden infant death, cause unknown

Unclassifiable

1.9

Problematic

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

10

Vergani. Identifying the causes of stillbirth. Am J Obstet Gynecol 2008.

319.e2

American Journal of Obstetrics & Gynecology SEPTEMBER 2008

system most commonly used internationally, found unknown causes (unexplained or unclassifiable) in 73 cases
(47.4%) in our cohort. The 3 more recent classification systems allocated a
proportion of these 73 cases to known
causes of stillbirth. The de Galan-Roosen
classification allocated 42 of 73 cases to
placental causes; the ReCoDe classification allocated 20 of 73 cases to fetal
growth restriction; and the Tulip classification identified placenta/umbilical
cord causes in 45 of 73 cases. The 3 more
recent classification systems had thus
fewer unknown cases of stillbirth than
the extended Wigglesworth classification, with percentages of 18.2% for de
Galan-Roosen classification, 14.3% for
ReCoDe classification, and 16.2% for
Tulip classification.
The median gestational ages were similar in the group with unknown vs
known causes of stillbirth (31.4 [range,
25.0-40.0] vs 31.2 [range, 22.3-42.2]);
however, all cases of unknown stillbirths in all 4 classifications occurred after 25 weeks.

C OMMENT
We have found that the commonly used
Wigglesworth classification3 of stillbirths results in a higher proportion of
unexplained stillbirths compared with
newer classification systems that include
fetal growth restriction and placental
disease as causative processes of fetal
death. Indeed, in our cohort, the
Wigglesworth classification3 failed to
identify a cause of stillbirth in nearly half
of the cases, whereas such rate decreased
to 18%, 14%, and 16% using the classifications proposed by de Galan Roosen,6
ReCoDe,7 and Tulip,8 respectively.
The reduction in rates of unexplained
stillbirth in the more recent classification
systems results from the inclusion of abnormalities in fetal growth and pathologic changes of placenta as causal categories. Indeed, both conditions are large
contributing factors to stillbirths, and
the newer classifications acknowledge
the vital role of the placenta in determining optimal fetal development. The Tulip classification8 focused on placental
causes of death, validating the impor-

SMFM Papers

www.AJOG.org

TABLE 3

TABLE 4

de Galan-Roosen classification: application to current study

ReCoDe classification: application

to current study

Code

Cause of death

Group

154

Classification

..............................................................................................................................................................................................................................................

Trauma

Infection

13

Placenta/cord disease

57

Maternal immune system disease

Congenital malformations

Prematurity/immaturity complications

Unclassifiable

..............................................................................................................................................................................................................................................

8.4

..............................................................................................................................................................................................................................................

37

154

...........................................................................................................

Fetus

93

60.4

Fetal growth
restriction

26

Umbilical cord

13

8.4

Placenta

19

12.3

...........................................................................................................
a

..............................................................................................................................................................................................................................................

0.6

...........................................................................................................

52

33.8

...........................................................................................................

1.9

28

18.2

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

Vergani. Identifying the causes of stillbirth. Am J Obstet Gynecol 2008.

...........................................................................................................

Amniotic fluid

3.9

Uterus

Mother

Intrapartum

0.7

Trauma

Unclassified

...........................................................................................................
...........................................................................................................
...........................................................................................................

tance of the placental examination in a

protocol of investigation of stillbirth.
The ReCoDe classification7 used customized percentiles for birthweight,
which assess the contribution of poor fetal growth better than standardized
growth curves. Of interest, the ReCoDe
classification system does not consider
factors such as abnormalities in fetal
growth and pathologic changes of placenta, as causes but as relevant conditions in stillbirths. Our findings suggest
that stillbirth classification systems that
do not include placental abnormalities
or fetal growth restriction are inadequate
as they may prevent identification of important factors related to fetal death,
which may have an impact on counseling
after a fetal death. In line with our conclusion, other groups have emphasized
the importance of an adequate classification system for causes of stillbirth to
know if and which preventive measures
can be implemented to lower the risk of
perinatal mortality in subsequent pregnancies.6,11 In fact the classification systems are simply a more organized way to
collect the evidence and place it in order of importance in the most likely
causative pathways, always in the context
of the patients medical and obstetric history. Identification of processes with a
potential for recurrence may alter dramatically the management of a subsequent pregnancy.
The benefit of classification systems is
precisely that they provide a method by
which the obstetrician can be guided in
the identification of possible causal pathways for a stillbirth in the presence of

multiple risk factors. Because the individual tests are not mutually exclusive,
the total of positive tests is greater than
the number of stillbirth cases. In the
presence of multiple factors, causation is
ascribed to the most relevant. It is important also to emphasize that detection of a
positive test does not necessarily explain
the stillbirth, such as, for example, mild
placental lesions. On the yield of the individual tests in our series, placental pathology and autopsy are of utmost importance in the evaluation of stillbirths,
whereas total body X-ray and skin biopsy
provide a negligible contribution.
The cause of stillbirth is often difficult
to identify because more than 1 factor
may contribute to stillbirth in any individual case, and because different conditions may be considered causes of or risk
factors for stillbirth, such as growth restriction and hypertension. For these
reasons, in the presence of multiple contributing factors to stillbirth, newer classification systems have attempted to establish a hierarchy of relevance in the
causes, so as to call the attention to the
primary causes of the adverse event. For
example, in the presence of a stillbirth
associated with preeclampsia and fetal
growth restriction, the classification system would attribute the outcome primarily to poor fetal growth; conversely,
in the presence of an appropriately
grown stillbirth associated with preeclampsia and placental abruption, the
primary cause of death would be the abruption. In the Tulip system, placental
disease determines the primary cause of
abnormal fetal development, so that

...........................................................................................................

22

14.3

...........................................................................................................
...........................................................................................................
a

17% of total.

Vergani. Identifying the causes of stillbirth. Am J

Obstet Gynecol 2008.

poor fetal growth is considered subordinate to any abnormal histopathologic

placental findings. Such a system allows
the identification of the cause of stillbirth
in a more specific way than systems
based on the clinical conditions (eg, ReCoDe). On the other hand, the system
requires a competent placental pathologist and adequate knowledge of placenta
disease by the obstetricians. Minimizing
the rates of unexplained stillbirths with
adequate classification systems may
eventually open the door to research
TABLE 5

Tulip classification: application

to current study
Code

Cause of
death

154

...........................................................................................................

Congenital
anomaly

52

33.8

Placenta

61

39.6

Prematurity/
immaturity

2.6

...........................................................................................................
...........................................................................................................

...........................................................................................................

Infection

4.5

Other

3.2

Unknown

25

16.2

...........................................................................................................
...........................................................................................................

Vergani. Identifying the causes of stillbirth. Am J

Obstet Gynecol 2008.

SEPTEMBER 2008 American Journal of Obstetrics & Gynecology

319.e3

SMFM Papers
protocols aimed at evaluating the pathogenic processes at play in fetal death, and
possibly at preventative strategies.
f
REFERENCES
1. Stanton C, Lawn JE, Rahman H, et al. Sillbirths rates: delivering estimates in 190 countries. Lancet 2006;367:1487-94.
2. Whitfield CR, Smith NC, Cockburn F, et al.
Perinatally related wastagea proposed classification of primary obstetric factors. Br J Obstet Gynaecol 1986;93:694-703.
3. Wigglesworth JS. Monitoring perinatal mortality: a pathophysiological approach. Lancet
1980;2:684-6.

319.e4

www.AJOG.org
4. Confidential Enquiry into Stillbirths and
Deaths in Infancy. The confidential enquiry into
stillbirths and deaths in infancy. London:
CESDI; 1993:1-163.
5. Hey EN, Lloyd DJ, Wigglesworth JS. Classifying perinatal death: fetal and neonatal factors.
Br J Obstet Gynaecol 1986;93:1213-23.
6. de Galan Roosen AEM, Kuijpers JC, van der
Straaten, et al. Fundamental classification of
perinatal death: validation of a new classification
system of perinatal death. Eur J Obstet Gynecol
2002;103:30-6.
7. Gardosi J, Kady SM, McGeown P, et al.
Classification of stillbirth by relevant conditions
at death (ReCoDe): population based cohort
study. BMJ 2005;331:1113-7.

American Journal of Obstetrics & Gynecology SEPTEMBER 2008

8. Korteweg FJ, Gordijn SJ, Timmer A, et al.

The Tulip classification of perinatal mortality: introduction and multi-disciplinary inter-rater
agreement. BJOG 2006;113:393-401.
9. World Health Organization. Definition and indicators in family planning maternal and child
health and reproductive health. Geneva: WHO
Press; 2006.
10. World Health Organization. Neonatal and
perinatal mortality: country, regional and global
estimates. Geneva: WHO Press; 2006.
11. Korteweg FJ, Gordijn SJ, Timmer A, et al.
A placental cause of intra-uterine fetal
death depends on the perinatal mortality classification system used. Placenta 2008;29:
71-80.