Estimating risks of perinatal death

doi:10.1016/j.ajog.2004.08.014

American Journal of Obstetrics and Gynecology

Volume 192, Issue 1, January 2005, Pages 17-22

https://doi.org/10.1016/j.ajog.2004.08.014 Get rights and content

The relative and absolute risks of perinatal death that are estimated from observational studies are used frequently in counseling about obstetric intervention. The statistical basis for these estimates therefore is crucial, but many studies are seriously flawed. In this review, a number of aspects of the approach to the estimation of the risk of perinatal death are addressed. Key factors in the analysis include (1) the definition of the cause of the death, (2) differentiation between antepartum and intrapartum events, (3) the use of the appropriate denominator for the given cause of death, (4) the assessment of the cumulative risk where appropriate, (5) the use of appropriate statistical tests, (6) the stratification of analysis of delivery-related deaths by gestational age, and (7) the specific features of multiple pregnancy, which include the correct determination of the timing of antepartum stillbirth and the use of paired statistical tests when outcomes are compared in relation to the birth order of twin pairs.

Section snippets

Definition and classification of perinatal death

The World Health Organization definition of perinatal death is death of the offspring “occurring during late pregnancy (at 22 completed weeks gestation and over), during childbirth and up to seven completed days of life.” Stillbirths are subclassified as antepartum (ie, the fetus died before the onset of labor) or intrapartum (ie, the fetus died after the onset of labor but before birth). There can be uncertainty about the timing of death in cases in which the mother seeks care in established

Uses of perinatal mortality statistics

The perinatal mortality rate is the number of perinatal deaths divided by the total number of births and is used as an overall summary statistic of the risk of perinatal death. However, the perinatal mortality rate has certain drawbacks as a means of comparison of the outcome in relation to medical interventions. The most common causes of perinatal death are unexplained stillbirth and the pulmonary effects of prematurity. Collectively, these causes account for approximately 50% of perinatal

Numerators and denominators

Risk is defined as the probability of an adverse event. Probability is estimated by the number of individuals who experience an event divided by the number of individuals who were at risk of the event.⁷ It is clear that the numerator and the denominator for different types of perinatal death will differ according to the type of perinatal death. For example, renal agenesis is determined in embryonic life and leads to inevitable neonatal death at whatever gestational age the baby is delivered.

Antepartum stillbirth

Conventionally, the risk of stillbirth at a given week of gestation was estimated in relation to the total number of births during the given week. Yudkin et al⁸ argued that the population at risk of antepartum stillbirth at any week of gestation was all on-going pregnancies. They proposed that the risk of unexplained stillbirth at any given week of gestation should be estimated by the ratio of the number of unexplained stillbirths to the total number of on-going pregnancies at the start of the

Intrapartum stillbirth and neonatal death

Delivery-related perinatal death is defined as intrapartum stillbirth or neonatal death that is unrelated to congenital abnormality. Some authors have suggested that the risk of these events should also be estimated in relation to the number of on-going pregnancies at each week of gestation.¹⁷ The justification for the use of on-going pregnancies as the denominator is that both intrapartum stillbirth and neonatal death usually are due to obstetric events. However, this does not justify the

Stratification of delivery-related perinatal deaths by gestational age

The major cause of neonatal death is, as stated earlier, prematurity. Intrapartum stillbirth or neonatal death that is unrelated to congenital abnormality is rare at term and affects <1 per 1000 births.²⁰ It follows therefore that, in any analysis of delivery-related perinatal death that includes both term and preterm births, most of the events will be neonatal deaths because of prematurity. Surprisingly, when the risk of delivery-related perinatal death is compared between groups,

Multiple pregnancy

There are specific issues in the case of multiple pregnancy that should be addressed both in relation to the assessment of the risk of both antepartum and intrapartum events. The first is defining the timing of antepartum fetal death. In the analyses of antepartum stillbirth among singleton births, which was described earlier, it was assumed that all antepartum fetal deaths took place in the same week as the week of delivery. It is likely that some of the deaths will have preceded delivery by

Conclusions

Estimates of the risk of perinatal death from observational studies are used widely when women are counseled regarding obstetric intervention. However, these analyses frequently are flawed. The interpretation of the results of such studies requires the assessment of the definition of the event, the definition of those events that are deemed to be at risk, the use of appropriate statistical tests, and stratification by gestational age when appropriate.

References (24)

P.L. Yudkin et al.
Risk of unexplained stillbirth at different gestational ages
Lancet
(1987)
G.C.S. Smith
Life-table analysis of the risk of perinatal death at term and post term in singleton pregnancies
Am J Obstet Gynecol
(2001)
G.C.S. Smith et al.
Caesarean section and risk of unexplained stillbirth in subsequent pregnancy
Lancet
(2003)
J.C. Rageth et al.
Delivery after previous cesarean: a risk evaluation: Swiss Working Group of Obstetric and Gynecologic Institutions
Obstet Gynecol
(1999)
R.S. Hartley et al.
Perinatal mortality and neonatal morbidity rates among twin pairs at different gestational ages: optimal delivery timing at 37 to 38 weeks' gestation
Am J Obstet Gynecol
(2001)
Information and Statistics Division NHS Scotland
Scottish perinatal and infant mortality report 2000
(2001)
I. Ingemarsson et al.
Stillbirths and rate of neonatal deaths in 76,761 postterm pregnancies in Sweden, 1982-1991: a register study
Acta Obstet Gynecol Scand
(1997)
E.G. Raymond et al.
Effects of maternal age, parity, and smoking on the risk of stillbirth
BJOG
(1994)
C.R. Whitfield et al.
Perinatally related wastage- a proposed classification of primary obstetric factors
BJOG
(1986)
E.N. Hey et al.
Classifying perinatal death: fetal and neonatal factors
BJOG
(1986)

S.K. Cole et al.

Classifying perinatal death: an obstetric approach

BJOG

(1986)

S. Selvin

Statistical analysis of epidemiologic data

(1996)

Cited by (48)

Predicting and preventing stillbirth at term
2024, Seminars in Perinatology
Stillbirth at term affects ∼1 per 1000 pregnancies at term in high income countries. A range of maternal characteristics are associated with stillbirth risk. However, given the low a priori risk of stillbirth, the vast majority of women with clinical risk factors would not experience a stillbirth in the absence of intervention. Stillbirth is the end point of multiple pathways, including both fetal growth restriction and fetal overgrowth. In most term stillbirths there is no mechanistic understanding of the cause of death and a sizeable proportion are completely unexplained. Term stillbirth is potentially preventable by early delivery, providing a rationale for screening. “Omic” analyses of blood taken prior to the onset of some of the conditions associated with stillbirth may help identify women at high risk and allow the potentially harmful intervention of early term medically indicated delivery to be targeted to the pregnancies most likely to benefit.
Interventions to prevent stillbirth
2017, Seminars in Fetal and Neonatal Medicine
Stillbirth is one of the most distressing complications of pregnancy and still occurs far too frequently. The rate of stillbirth has been decreasing worldwide but room for improvement remains even in high-income countries. Risk factors for stillbirth have been identified in an effort to detect those women at increased risk. However, risk factors are non-specific and do not identify most stillbirths. Strategies employed to screen the general population such as assessment of fetal activity, fetal growth screening and biomarkers have also been used to identify increased risk for stillbirth. As with clinical risk factors, these methods are non-specific. Interventions to prevent stillbirth include antenatal testing of high-risk women, ultrasonographic assessments of fetal growth and Doppler velocimetry as well as iatrogenic preterm or term delivery. Additional research into the role of these interventions and better identification of those at high risk for stillbirth will help to achieve further stillbirth reduction.
Understanding perinatal mortality
2016, Obstetrics, Gynaecology and Reproductive Medicine
The term perinatal death is used to describe antepartum and intrapartum stillbirths, and early neonatal deaths. Although the overall rate of perinatal mortality is falling, a slower rate of reduction has been observed in stillbirth compared to neonatal death. Antenatal stillbirth contributes to a greater proportion of cases in high income countries and associated risk factors include maternal age, smoking, obesity and small for gestational age fetus. At term, intrapartum stillbirth and neonatal death are collectively referred to as delivery related perinatal death, and the incidence in nulliparous and multiparous women is approximately 1 in 1000 and 1 in 2000 births, respectively. Associated factors include advanced maternal age, small for gestational age, fetal macrosomia, breech labour and previous caesarean delivery. The impact of obstetric interventions in labour on delivery related perinatal death, including rising rates of caesarean delivery, is complex and unclear.
Prolonged and Postterm Pregnancy
2016, Obstetrics: Normal and Problem Pregnancies
Perinatal mortality: Extent and causes in Lubumbashi, Democratic Republic of Congo
2013, Revue d'Epidemiologie et de Sante Publique
La présente étude a été initiée en vue de déterminer le taux, les causes et les facteurs de risque de la mortalité périnatale (MP) dans la ville de Lubumbashi en République démocratique du Congo.
Nous avons réalisé une étude transversale au cours de laquelle les données ont été récoltées grâce à l’interview des femmes et la fouille des dossiers de maternité. Seules les femmes ayant accouché au cours de l’année 2010 et résidant à Lubumbashi au cours de cette même année, quelle qu’ait été l’issue de l’accouchement, et dont les nouveau-nés étaient âgés d’au moins sept jours, étaient incluses dans l’étude. Les femmes ont été recrutées dans les ménages sélectionnés par échantillonnage en grappes. Les grappes étaient constituées des aires de chaque zone de santé. La mortalité périnatale a été définie comme l’ensemble des mort-nés et des décès néonatals précoces pour 1000 naissances. Ses facteurs de risque ont été recherchés par les valeurs des OR ajustés par la régression logistique au seuil critique de 5 %.
De 11 536 femmes enquêtées, il y avait 11 633 naissances dont 177 mort-nés et 133 décès néonatals précoces. La mortalité périnatale était de 27 ‰ (IC95 % = 23,7–29,6 ‰). Les causes de cette mortalité étaient la détresse respiratoire (58,2 %), les infections néonatales (pneumonie et méningite néonatale, 13,5 %), les complications de la prématurité (9,0 %), le tétanos néonatal (1,6 %), les malformations congénitales (0,6 %) ; 17,1 % des décès périnatals étaient de causes inconnues. Le fait d’accoucher sans être mariée, l’accouchement à domicile, les complications de l’accouchement, l’accouchement par voie basse en cas de dystocie, la césarienne, les grossesses multiples, le faible poids de naissance et la prématurité étaient les facteurs de risque de la mortalité périnatale.
Des efforts sont nécessaires pour améliorer la disponibilité, l’utilisation et la qualité des soins obstétricaux et néonatals d’urgence. L’information des femmes concernant les signes de danger de la grossesse et de l’accouchement doit être également renforcée.
The present study was initiated in order to determine the rate, the causes and the risk factors for perinatal mortality in Lubumbashi, Democratic Republic of Congo.
Data for this cross-sectional study were collected by interviewing participating women and by analysis of medical files. Women who gave birth in 2010 and were residents of Lubumbashi during the same year were included. Women were included irrespective of the pregnancy outcome and perinatal survival was determined for newborns aged at least seven days. Women were recruited from households selected by cluster sampling for healthcare zones. Perinatal mortality was defined as stillbirths and early neonatal deaths per 1000 births. Risk factors were sought using the odds ratio method adjusted by logistic regression using a 5% threshold.
Among 11,536 surveyed women, there were 11,633 births including 177 stillbirths and 133 early neonatal deaths. Perinatal mortality was 27‰ (95%IC = 23.7–29.6‰). The causes of this mortality were respiratory distress (58.2%), neonatal infection (pneumonia and neonatal meningitis, 13.5%), complications of prematurity (9.0%), neonatal tetanus (1.6%), congenital malformations (0.6%). The cause of perinatal death was unknown for 17.1%. Risk factors for perinatal mortality were: unmarried mother; home delivery; complicated delivery; dystocia; caesarean-section; multiple pregnancy; low birth weight; prematurity.
Action should be taken to improve availability, use and quality of Emergency obstetrical and neonatal care. Women should be better informed concerning the danger signs of pregnancy and childbirth.
Understanding perinatal mortality
2013, Obstetrics, Gynaecology and Reproductive Medicine
The term perinatal death is used to describe antepartum and intrapartum stillbirths, and early neonatal deaths. At term, intrapartum stillbirth and neonatal death are collectively referred to as delivery related perinatal death, and the incidence in nulliparous and multiparous women is approximately one in 1000 and one in 2000 births, respectively. Associated factors include advanced maternal age, small for gestational age, fetal macrosomia, breech labour and previous caesarean delivery. The impact of obstetric interventions in labour on delivery related perinatal death, including rising rates of caesarean delivery, is complex and unclear. The incidence of overall perinatal death is falling mainly as a result of improvements in the management of premature neonates and from decreased deaths secondary to intrapartum anoxia at term. This review will provide an overview of perinatal mortality with a particular emphasis on delivery related perinatal death at term.

View all citing articles on Scopus

View full text

Editors' ChoiceEstimating risks of perinatal death

Section snippets

Definition and classification of perinatal death

Uses of perinatal mortality statistics

Numerators and denominators

Antepartum stillbirth

Intrapartum stillbirth and neonatal death

Stratification of delivery-related perinatal deaths by gestational age

Multiple pregnancy

Conclusions

Lancet

Am J Obstet Gynecol

Lancet

Obstet Gynecol

Am J Obstet Gynecol

Scottish perinatal and infant mortality report 2000

Stillbirths and rate of neonatal deaths in 76,761 postterm pregnancies in Sweden, 1982-1991: a register study

Acta Obstet Gynecol Scand

Effects of maternal age, parity, and smoking on the risk of stillbirth

BJOG

Perinatally related wastage- a proposed classification of primary obstetric factors

BJOG

Classifying perinatal death: fetal and neonatal factors

BJOG

Classifying perinatal death: an obstetric approach

BJOG

Statistical analysis of epidemiologic data

Editors' Choice
Estimating risks of perinatal death