Role of Second-Trimester Uterine Artery Doppler in Assessing Stillbirth Risk

Singh, Tulika MD, MRCOG; Leslie, Karin MD, MRCOG; Bhide, Amar MD, FRCOG; D'Antonio, Francesco MD; Thilaganathan, Basky MD, FRCOG

Obstetrics & Gynecology:
doi: 10.1097/AOG.0b013e318242ad81
Original Research

OBJECTIVE: To estimate the association between uterine artery Doppler indices and stillbirth in routinely screened populations.

METHODS: Second-trimester uterine artery Doppler indices at 19 to 23 weeks of gestation were obtained from a large cohort of women. Pregnancy losses recorded on a mandatory national register were cross-linked to the Doppler database. Kaplan-Meier curves were constructed for the risk of stillbirth based on the uterine artery Doppler resistance indices. Logistic regression analysis was used to assess the influence of uterine artery Doppler indices and other more conventional risk factors on the likelihood of stillbirth.

RESULTS: Data were available from 15,835 women with 144 stillbirths (9.1 stillbirths per 1,000 births). Kaplan-Meier analysis showed that the risk of stillbirth (39.41 per 1000) in women with uterine artery Doppler indices greater than the 90th percentile was sevenfold higher (95% confidence interval 4.81–9.57) than the reference population (5.36 per 1000) with Doppler indices less than or equal to the 90th percentile. The positive predictive and negative predictive values for the 90th percentile uterine artery Doppler cut-off were 0.46% and 95.73%, respectively. The sensitivities of the 90th, 95th, and 99th percentile uterine artery Doppler resistance index cut-offs for the sensitivity of stillbirth were 46.2%, 35.4%, and 15.4%, respectively. Conventional risk factors for term stillbirth such as ethnicity, body mass index (BMI), and smoking no longer contributed to stillbirth risk when uterine artery Doppler indices were included in multivariable logistic regression analysis.

CONCLUSION: Elevated second-trimester Doppler indices, a proxy for impaired placentation, are more strongly associated with stillbirth than conventional risk factors. Risk factors such as ethnicity, maternal age, BMI, and smoking contribute to risk of term stillbirth through uteroplacental dysfunction.


In Brief

Elevated second-trimester Doppler indices are strongly associated with an increased risk of both preterm and term stillbirth.

Author Information

From the Fetal Medicine Unit, Academic Department of Obstetrics and Gynecology, St. George's Hospital Medical School, and the Newham University Hospital Trust, London, United Kingdom.

Corresponding author: Professor Basky Thilaganathan, Fetal Medicine Unit, St. George's University of London, Cranmer Terrace, London SW17 0RE, UK; e-mail:

Financial Disclosure The authors did not report any potential conflicts of interest.

Article Outline

The stillbirth rate has remained static over the past decade at approximately five per 1,000 births despite significant improvements in the delivery of antenatal care.1 One in every 200 pregnant women who gets beyond 22 weeks of gestation will experience an intrauterine death, a rate that is 10-fold greater than sudden infant death syndrome.2 Many stillbirths occur for no discernable reason, and are therefore classified as unexplained, which is frequently misinterpreted to imply that these stillbirths are also unavoidable. Common risk factors for stillbirth include parity, racial origin, prolonged gestation, smoking, and previous adverse pregnancy outcome.36 These risk factors have modest adjusted odds ratios between 1.6 and 2.9 for stillbirth, and those with higher odds ratios such as severe fetal growth restriction (adjusted odds ratio of 3.9) have relatively low population prevalence.7,8 Even though the policy of induction of labor from 41 weeks of gestation is specifically focused on preventing such stillbirths,9 it has not changed the prevalence of stillbirth at term.10 This is presumably because the majority of stillbirths at term occur before 41 weeks of gestation.

Poor uterine artery blood flow is associated with biochemical and cellular evidence of impaired placental development and function.11 Uterine artery Doppler assessment appears effective in screening populations at high-risk for preeclampsia.12 Poor uterine artery blood flow is also associated with adverse pregnancy outcome from other placental syndromes such as fetal growth restriction, abruption, and stillbirth.13 We used a large cohort of pregnancies from a single unit screened by second-trimester Doppler linked to a national stillbirth registry to explore the association between uterine Doppler indices and stillbirth.

Back to Top | Article Outline


This is a retrospective study performed between 2000 and 2008 in a single obstetric unit serving an inner city multi-ethnic population. The local Institutional Review Board, the Wandsworth Research and Ethics Committee, judged that ethical approval was not required because of the retrospective study design. Women with singleton pregnancies who were nulliparous or had a previous pregnancy with a history of placental syndromes (fetal growth restriction, preeclampsia, or stillbirth) underwent second-trimester uterine artery Doppler assessment. Doppler assessment was performed at the time of their routine anomaly scan between 19 and 23 weeks of gestation to assess the woman's baseline risk for preeclampsia, as recommended by national guidelines.9 The flow velocity waveform of the uterine arteries was identified as previously described14 and the resistance index recorded on a dedicated ultrasound database (Viewpoint). Women with uterine artery Doppler resistance greater than the 95th percentile for gestation were advised about the increased risk of preeclampsia and provided with verbal and written information regarding preeclampsia symptoms. For these women, an additional growth scan was scheduled for 32–34 weeks of gestation and induction of labor was advised from 41 weeks of gestation as per national protocol.9

Pregnancy outcomes were obtained from the hospital computerized maternity database, the general practitioner, or the patient. Data for all stillbirths were obtained from the hospital national reporting register, even for women who delivered in other health districts. Ethnicity was self-reported in all cases and the gestation of birth was used as a proxy for the gestation of stillbirth. The latter is based on an average 2-day time interval between fetal death and delivery in the third trimester.15 When the degree of tissue autolysis indicated a protracted interval between fetal demise and birth, the gestational age at death was classified as unknown. Cases of congenital abnormalities and unknown pregnancy outcome because of out-of-date contact details were excluded from the analysis.

Variables were compared using Mann–Whitney U or χ2 tests, as appropriate. Stepwise logistic regression analysis was carried out with stillbirth as the dependant variable. The mean of the left and right uterine artery resistance index was introduced as the explanatory continuous variable in the first step.14 Mean resistance index was dichotomized at the 90th, 95th, and 99th percentiles for gestation in this analysis. Maternal ethnicity, body mass index (BMI, calculated as weight (kg)/[height (m)]2), and smoking were introduced to explore the independent contribution of these factors to the risk of stillbirth. Mean uterine artery resistance index 90th percentile or less was treated as the reference group. The risk of stillbirth was compared using time-to-event analysis using the Kaplan-Meier and Cox proportional hazards model for the whole cohort and a subgroup from 36 weeks of gestation. This time series equivalent of multivariable logistic regression approach allows assessment of the relative risk accounting for variation in the duration of pregnancy.16

Back to Top | Article Outline


Uterine artery Doppler data were available from 15,835 women during the study period, with birth outcomes available in all but 39 pregnancies. The final cohort of 15,796 women included 144 antepartum stillbirths, with unknown gestation of death in 10 cases (excluded from analysis). The study population characteristics are shown in Table 1. The mean uterine artery resistance indices were significantly higher in the cases resulting in stillbirth (resistance index=0.63) compared with the live births (resistance index=0.53, P<.001). The sensitivity of the 90th, 95th, and 99th percentiles of uterine artery Doppler resistance index cut-off for the sensitivity of stillbirth were 46.2%, 35.4%, and 15.4%, respectively. The positive predictive and negative predictive values for the 90th percentile uterine artery Doppler cut-off were 0.46% and 95.73%, respectively. Approximately 50% of the stillbirths occurred in the cohort with uterine artery Doppler indices greater than the 90th percentile (Table 2). Logistic regression showed that both uterine artery resistance index and ethnicity had an independent association with stillbirth. The adjusted odds ratios (Table 2) for Afro-Caribbean race was 2.28 (95% confidence interval [CI] 1.56–3.33) and high uterine artery Doppler indices (greater than the 90th percentile) was 6.78 (95% CI 4.81–9.57). Spontaneous and iatrogenic preterm birth before 37 weeks of gestation occurred in 270 of 638 (4.8%) of the cohort and in 270 (16.6%) of the high uterine artery cohort (P<.01).

The gestational age distribution of stillbirths was bimodal, with the peak at 25 and 39 weeks of gestation (Fig. 1). The Kaplan-Meier curves demonstrated the pattern of higher fetal losses for uterine artery Doppler indices greater than the 90th percentile (Fig. 2) and Afro-Caribbean ethnic origin (Fig. 3). The analysis of the cohort of pregnancies progressing beyond 36 weeks of gestation showed that 35% of the stillbirths occurred in the cohort with uterine artery Doppler indices greater than the 90th percentile and that high uterine artery Doppler indices were the only factor associated with stillbirth (Fig. 4; adjusted odds ratio 5.82, 95% CI 3.02–11.20). A number of events and life table analyses based on second-trimester uterine artery Doppler indices are shown in Table 3. The receiver-operator characteristic curve for uterine artery Doppler resistance index in the prediction of stillbirth is shown in Figure 5 (area under curve=0.727).

Back to Top | Article Outline


The risk of stillbirth was higher among women with second-trimester uterine artery Doppler resistance indices in the highest decile. Conventional risk factors such as ethnicity, BMI, and smoking no longer contributed to stillbirth risk when uterine artery Doppler indices were included in multivariable logistic regression analysis. A protocol of induction of labor at 37 weeks of gestation in women with uterine artery Doppler indices greater than the 90th percentile has the potential to be more effective in preventing stillbirth than a policy of induction of labor for conventional risk factors such as maternal age, smoking, obesity, ethnicity, and prolonged pregnancy. Importantly, the most recent randomized control trials demonstrate that induced birth from 37 weeks of gestation is not associated with higher operative intervention or increased neonatal morbidity from late preterm delivery.17,18

This was a single-center study cohort with very good ascertainment of adverse pregnancy outcomes obtained from a national register. The study cohort consisted of only nulliparous and high-risk parous women with an obstetric history of placental syndromes. Both of these factors increase the risk of stillbirth by approximately twofold,7 explaining why the study cohort stillbirth rate of 9.1 per 1,000 births is higher than the 4.5 to 6.5 per 1,000 births expected for a hospital serving a multi-ethnic socially deprived population.1 Nulliparity and previous occurrence of a placental syndrome are recognized as conferring a higher risk for the subsequent development of preeclampsia and mandate more individualized risk assessment and plan of care.9 Uterine artery Doppler screening was routinely undertaken in nulliparous and high-risk parous women to identify and closely monitor women at increased risk.

There are two distinct gestational age distributions or epochs of increased stillbirths. The first, from 22 to 28 weeks of gestation, was correlated to both uterine artery Doppler indices and ethnic origin. This is in keeping with the strong association of abnormal uterine artery Doppler indices with early (less than 34 weeks) preeclampsia12 and severe fetal growth restriction.19 Previous population analyses in the same health region also showed an increased stillbirth rate in Afro-Caribbean and Asian women.20 In the study cohort, Asian ethnicity was not associated with an increased stillbirth risk, reflecting either improved access to health care in our unit or improved maternal health because of the effect of successive migrant generations.21 The second stillbirth epoch occurred from 36 weeks of gestation and was associated with high uterine artery Doppler indices, but not ethnic origin, BMI, maternal age, or smoking. This is a novel finding that suggests that the association between stillbirth and the latter factors are mediated through impaired placental function and confirms the assertion that fetal growth restriction attributable to poor placental function is the main risk factor for stillbirth at term.7 Ultrasound-estimated weight-based criteria are unlikely to recognize the majority of growth-restricted stillbirths at term because they fail to recognize fetuses that are smaller than they are meant to be, but are still normal in size for gestation by population standards.8 Although the use of customized birth weight percentiles may improve the detection of fetal growth restriction at term,22 the inaccuracy of ultrasound prediction of birth weight at term23 precludes its use as an effective screening tool.

Placental histology in term fetal growth restriction is only subtly different from that of normal pregnancy,24 unlike preterm fetal growth restriction.25 Similarly, preterm preeclampsia is strongly associated with fetal growth restriction and characteristic histological changes in the placenta, whereas preeclampsia at term is usually associated with normal birth weight and minimal placental changes.26 Given that placental histology and fetal size estimation are poor proxies for fetal growth restriction of placental origin, poor uterine artery blood flow may be the best currently available method of identification. High-resistance uterine artery waveforms have been associated with laboratory27,28 and clinical1113 evidence of poor placental development and function. Our study findings suggest that stillbirth is one such feature related to poor uteroplacental blood flow in mid pregnancy.

There has been only one previous study that systematically examined the association between uterine artery Doppler indices and stillbirth.13 Although, the latter study involved twice as many women, there were far fewer stillbirths (n=109, 3.6 stillbirths per 1,000 births) because of a healthier low-risk population or ascertainment bias. The authors also undertook serial scans, which may have altered pregnancy outcome. Importantly, they chose to define cases as only those with a histological diagnosis of placental syndromes (n=54) before undertaking multivariable analyses. They found a strong association between high-resistance uterine artery Doppler indices and stillbirth. However, their multivariable model had only 20% sensitivity for stillbirth after 32 weeks of gestation, for a 5% screen positive rate.

Our study was limited to nulliparous and high-risk parous women, raising the possibility that the findings are not applicable to the general population. However, there is substantial evidence that women with a previous normal pregnancy outcome with a short interpregnancy interval are at lowest risk for stillbirth.5 We would therefore argue that uterine artery Doppler screening was conducted in an appropriate high-risk group, but we acknowledge that these data need to be verified in lower-risk populations. Women with uterine artery Doppler indices greater than the 90th percentile were also routinely scanned at 32–34 weeks of gestation. This possibly explains the nadir in stillbirths at this gestation, because induced preterm birth may have prevented some stillbirths. However, a similar pattern of intrauterine deaths was observed by others,13 implying a bimodal distribution of stillbirths is a true pregnancy effect.

The main study finding is that second-trimester uterine Doppler is strongly associated with stillbirth at term. The strength of this association is such that it negates the significance of the majority of conventional risk factors such as ethnicity, maternal age, BMI, and smoking to predict stillbirth at term. The importance of the study association is that it has the potential to be used as a marker for women at high risk for stillbirth at term. This association deserves evaluation in a large prospective trial comparing early induction of labor in women with high-resistance Doppler indices compared with conventional management.

Back to Top | Article Outline


1. Centre for Maternal and Child Enquiries (CMACE). Perinatal mortality 2009. London (UK): CMACE; 2011.
2. Information and Statistics Division NHS Scotland. Scottish perinatal and infant mortality report 2000. Edinburgh (Scotland): ISD Scotland Publications; 2001.
3. Gray R, Bonellie SR, Chalmers J, Greer I, Jarvis S, Kurinczuk JJ, et al.. Contribution of smoking during pregnancy to inequalities in stillbirth and infant death in Scotland 1994–2003: retrospective population based study using hospital maternity records. BMJ 2009;339:b3754.
4. Goldenberg RL, McClure EM, Saleem S, Reddy UM. Infection-related stillbirths. Lancet 2010;375:1482–90.
5. Villamor E, Cnattingius S. Interpregnancy weight change and risk of adverse pregnancy outcomes: a population-based study. Lancet 2006;368:1164–70.
6. Smith GC, Pell JP, Dobbie R. Caesarean section and risk of unexplained stillbirth in subsequent pregnancy. Lancet 2003;362:1779–84.
7. Flenady V, Koopmans L, Middleton P, Frøen JF, Smith GC, Gibbons K, et al.. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet 2011;377:1331–40.
8. Gardosi J, Kady SM, McGeown P, Francis A, Tonks A. Classification of stillbirth by relevant condition at death (ReCoDe): population based cohort study. BMJ 2005;331:1113–7.
9. National Collaborating Centre for Women's and Children's Health (UK). Antenatal care: Routine care for the healthy pregnant woman. NICE Clinical Guidelines, No. 62. London (UK): RCOG Press; 2008.
10. Hussain AA, Yakoob MY, Imdad A, Bhutta ZA. Elective induction for pregnancies at or beyond 41 weeks of gestation and its impact on stillbirths: a systematic review with meta-analysis. BMC Public Health 2011;11(Suppl 3):S5.
11. Gebb J, Dar P. Colour Doppler ultrasound of spiral artery blood flow in the prediction of pre-eclampsia and intrauterine growth restriction. Best Pract Res Clin Obstet Gynaecol 2011;25:355–66.
12. Poon LC, Akolekar R, Lachmann R, Beta J, Nicolaides KH. Hypertensive disorders in pregnancy: screening by biophysical and biochemical markers at 11–13 weeks. Ultrasound Obstet Gynecol 2010;35:662–70.
13. Smith GC, Yu CK, Papageorghiou AT, Cacho AM, Nicolaides KH, Fetal Medicine Foundation Second Trimester Screening Group. Maternal uterine artery Doppler flow velocimetry and the risk of stillbirth. Obstet Gynecol 2007;109:144–51.
14. Napolitano R, Santo S, D'Souza R, Bhide A, Thilaganathan B. Sensitivity of higher, lower and mean second-trimester uterine artery Doppler resistance indices in screening for pre-eclampsia. Ultrasound Obstet Gynecol 2010;36:573–6.
15. Genest DR, Williams MA, Greene MF. Estimating the time of death in stillborn fetuses: I. Histologic evaluation of fetal organs; an autopsy study of 150 stillborns. Obstet Gynecol 1992;80:575–84.
16. Smith GC. Life-table analysis of the risk of perinatal death at term and post term in singleton pregnancies. Am J Obstet Gynecol 2001;184:489–96.
17. Boers KE, Vijgen SM, Bijlenga D, Van Der Post JA, Bekedam DJ, Kwee A, et al.. DIGITAT study group. Induction versus expectant monitoring for intrauterine growth restriction at term: randomised equivalence trial (DIGITAT). BMJ 2010;341:c7087.
18. Koopmans CM, Bijlenga D, Groen H, Vijgen SM, Aarnoudse JG, Bekedam DJ, et al.. HYPITAT study group. Induction of labour versus expectant monitoring for gestational hypertension or mild pre-eclampsia after 36 weeks' gestation (HYPITAT): a multicentre, open-label randomised controlled trial. Lancet 2009;374:979–88.
19. Melchiorre K, Leslie K, Prefumo F, Bhide A, Thilaganathan B. First-trimester uterine artery Doppler indices in the prediction of small-for-gestational age pregnancy and intrauterine growth restriction. Ultrasound Obstet Gynecol 2009;33:524–9.
20. Balchin I, Whittaker JC, Patel RR, Lamont RF, Steer PJ. Racial variation in the association between gestational age and perinatal mortality: prospective study. BMJ 2007;334:833.
21. Walsh J, Mahony R, Armstrong F, Ryan G, O'Herlihy C, Foley M. Ethnic variation between white European women in labour outcomes in a setting in which the management of labour is standardised-a healthy migrant effect? BJOG 2011;118:713–8.
22. Iraola A, González I, Eixarch E, Meler E, Illa M, Gardosi J, et al.. Prediction of adverse perinatal outcome at term in small-for-gestational age fetuses: comparison of growth velocity vs. customized assessment. J Perinat Med 2008;36:531–5.
23. Dudley NJ. A systematic review of the ultrasound estimation of fetal weight. Ultrasound Obstet Gynecol 2005;25:80–9.
24. Tomas SZ, Roje D, Prusac IK, Tadin I, Capkun V. Morphological characteristics of placentas associated with idiopathic intrauterine growth retardation: a clinicopathologic study. Eur J Obstet Gynecol Reprod Biol 2010;152:39–43.
25. Apel-Sarid L, Levy A, Holcberg G, Sheiner E. Term and preterm (<34 and <37 weeks gestation) placental pathologies associated with fetal growth restriction. Arch Gynecol Obstet 2010;282:487–92.
26. Sebire NJ, Goldin RD, Regan L. Term preeclampsia is associated with minimal histopathological placental features regardless of clinical severity. J Obstet Gynaecol 2005;25:117–8.
27. Whitley GS, Dash PR, Ayling LJ, Prefumo F, Thilaganathan B, Cartwright JE. Increased apoptosis in first trimester extravillous trophoblasts from pregnancies at higher risk of developing preeclampsia. Am J Pathol 2007;170:1903–9.
28. Prefumo F, Sebire NJ, Thilaganathan B. Decreased endovascular trophoblast invasion in first trimester pregnancies with high-resistance uterine artery Doppler indices. Hum Reprod 2004;19:206–9.
© 2012 by The American College of Obstetricians and Gynecologists.