Natural History of Fetal Position During
Pregnancy and Risk of Nonvertex Delivery
Catherine Takacs Witkop, MD, MPH, Jun Zhang, PhD, MD,
and James Troendle, PhD
OBJECTIVE: To examine the natural history of fetal po-
sition throughout pregnancy and the likelihood for non-
vertex delivery.
METHODS: We examined fetal positions at 15–22 weeks,
at 31–35 weeks, and at delivery using data from the
Routine Antenatal Diagnostic Imaging with Ultrasound
trial. Characteristics of women with nonvertex and vertex
presentation at delivery were compared. Multivariable
logistic regression analysis was performed to determine
risk factors for nonvertex presentation at delivery, and
odds ratios were calculated for those risk factors found to
be statistically significantly associated with nonvertex
presentation.
RESULTS: Data for 7,045 women who underwent routine
prenatal ultrasound examinations were analyzed. We
found no association between nonvertex presentation at
15–22 and at 31–35 weeks of gestation. A nonvertex fetus
at 35 weeks had a 45% chance of spontaneous version by
delivery. Multivariable logistic regression analysis found
From the Department of General Preventive Medicine, Johns Hopkins Bloomberg
School of Public Health, Baltimore, Maryland; Division of Epidemiology,
Statistics and Prevention Research, National Institute of Child Health and
Human Development, National Institutes of Health, Bethesda, Maryland.
Supported in part by the Intramural Research Program at the National Institute
of Child Health and Human Development.
The authors thank the investigators of the Routine Antenatal Diagnostic Imaging
With Ultrasound (RADIUS) trial for allowing us to use the data. Principal
investigators of the RADIUS trial included B. G. Ewigman, MD, M. L.
LeFevre, MD, University of Missouri, Columbia; J. P. Crane, MD, Washington
University, St. Louis; F. D. Frigoletto, MD, Harvard Medical School, Boston;
R. P. Bain, PhD, George Washington University, Rockville; D. McNellis, MD,
National Institute of Child Health and Human Development, Bethesda.
The opinions and assertions contained herein are the expressed views of the
authors and do not necessarily reflect the views of the above investigators.
Corresponding author: Catherine Takacs Witkop, MD, MPH, Johns Hopkins
Bloomberg School of Public Health, Department of General Preventive Medicine,
615 N. Wolfe Street, Room WB602, Baltimore, MD 21205; e-mail:
katika@aya.yale.edu.
Financial Disclosure
The authors have no potential conflicts of interest to disclose.
© 2008 by The American College of Obstetricians and Gynecologists. Published
by Lippincott Williams & Wilkins.
ISSN: 0029-7844/08
VOL. 111, NO. 4, APRIL 2008
Wenyu Sun, MD, MPH,
that multiparous women had half of the risk of nonvertex
presentation as nulliparous women. Smoking during
pregnancy (odds ratio [OR] 1.47, 95% confidence interval
[CI] 1.10 –1.96), low volume of amniotic fluid at 31–35
weeks (OR 3.74, 95% CI 1.85–7.53), and fundal position of
the placenta at late ultrasound examination (OR 1.85,
95% CI 1.23–2.78) were all associated with significant
increases in the risk of nonvertex position at delivery.
CONCLUSION: Spontaneous version of a nonvertex fe-
tus at 35 weeks of gestation is still likely.
(Obstet Gynecol 2008;111:875–80)
LEVEL OF EVIDENCE: II
B reech or nonvertex presentation complicates ap-
proximately 3– 4% of term deliveries.1,2 The ma-
jority of these will be delivered by cesarean as was
seen in 2002 when 86.9% of women with breech
presentation underwent cesarean delivery.3 A meta-
analysis of planned cesarean delivery compared with
planned vaginal delivery for breech presentation
found decreased rates of short-term maternal morbid-
ity and mortality in the vaginal delivery group,4 but
two large studies showed low risk of severe maternal
complications in women undergoing cesarean deliv-
ery.5,6 The risks of neonatal morbidity and mortality
by mode of delivery are also under debate.6,7 External
cephalic version, or manually turning the fetus in
utero, is one method of reducing the rate of breech
delivery at term, but success rates vary from 35% to
86%.7
The uncertainty surrounding the ideal manage-
ment of breech presentation calls for a better under-
standing of the etiology of nonvertex position of the
fetus. It is hypothesized that the normal kicking
movements of the fetus allow it to assume a vertex
position by the time of delivery. Previously reported
risk factors for breech presentation include nulliparity
and grand multiparity, advanced maternal age, con-
tracted pelvis, and uterine anomalies or lesions. Preg-
nancy complications that have been associated with
OBSTETRICS & GYNECOLOGY 875
nonvertex presentation include preterm delivery,
congenital malformations, abnormal amniotic fluid
levels, and placenta previa or cornual implantation of
the placenta. The association of birth weight with
breech presentation has been somewhat mixed. A
previous population-based, case-control study of 11,771
patients using Washington State birth certificates ad-
ditionally found that risk of breech presentation was
increased with hydrocephalus, pre-existing diabetes,
congenital malformations, smoking, and late or no
prenatal care.8
The natural history of fetal position has not been
systematically studied in a large population. This
current paper presents the findings from a large
prospective study of women undergoing ultrasono-
graphic surveillance. We examined the natural his-
tory of fetal position throughout pregnancy and the
likelihood for nonvertex delivery by maternal and
fetal characteristics.
MATERIALS AND METHODS
The population for this study consisted of participants
from the Routine Antenatal Diagnostic Imaging with
Ultrasound (RADIUS) trial, a practice-based multi-
center study of pregnant women who were at low risk
for adverse outcomes of pregnancy. The trial was
carried out between November 1987 and May 1991
and was designed to test the hypothesis that routine
screening with standardized ultrasonography at two
time periods would reduce perinatal morbidity and
mortality. A detailed description of the trial is avail-
able elsewhere.9
In brief, English-speaking pregnant women 18
years of age or older, whose last menstrual period was
known to within 1 week and who were recruited
before 18 weeks of gestation, were eligible for this
study. Exclusion criteria included women who had
previous stillbirth, prior small-for-gestational age in-
fant, irregular menstrual cycles, greater than 3-week
discrepancy between uterine size and dates, pre-
existing diabetes mellitus, chronic hypertension, and
chronic renal disease. Eligible women were randomly
assigned to an ultrasound-screening group, in which
two routine ultrasound examinations were performed
at 15–22 weeks and at 31–35 weeks of gestation, or to
a control group with no scheduled ultrasounds. In
both groups, clinically indicated ultrasounds could be
performed at any time.
With limited exceptions, all ultrasound examina-
tions in this study were performed in one of the 28
ultrasound laboratories participating in the trial. The
evaluation was standardized and included assessment
of placental location, amniotic fluid volume according
876 Witkop et al Natural History of Fetal Position
to protocol by Phelan et al,10 uterine and adnexal
pathology, fetal number, fetal biometry, and anatomic
survey of the fetus.
Baseline information for the participants was col-
lected at recruitment and included demographic char-
acteristics and reproductive history. Antepartum and
intrapartum information was abstracted from medical
records. Quality control procedures for ultrasonogra-
phy and data abstraction were implemented. Institu-
tional review boards at each study site approved this
study.9
The RADIUS trial individually randomized
15,151 women by a computer-generated scheme;
7,617 women were in the screening group, and 7,534
were in the control group. There were no differences
in the baseline characteristics of the two groups. The
trial found no differences in perinatal outcomes be-
tween the two groups.
In this current study, we used data from women
who were assigned to receive ultrasound screening.
We excluded 69 women with multifetal pregnancies,
229 women whose infants were found to have con-
genital defects (postnatal diagnosis), and 274 women
who delivered before 36 weeks. The data from 7,045
women who received two routine ultrasound exami-
nations at 15–22 weeks and 31–35 weeks of gestation
and who delivered at 36 weeks of gestation or later
were analyzed. Information on 15,027 ultrasound
examinations, including some that were performed
for clinical indications, was available. Delivery infor-
mation was also available for all of the participants.
Subjects who underwent successful external cephalic
version at or beyond 36 weeks of gestation were
considered to have a nonvertex position at delivery.
We compared the percentage of ultrasounds
demonstrating the fetus in vertex, breech, transverse,
or variable positions at 15–22 weeks with the percent-
age of fetuses in the vertex or nonvertex position at
31–35 weeks. We examined the characteristics of
women with infants who were nonvertex at delivery
compared with those of women (controls) who deliv-
ered vertex infants. For smoking in pregnancy, any
smoking was considered a positive response. Amni-
otic fluid volume level was considered normal if
amniotic fluid index was greater than 6 cm. To
examine the association of estimated fetal weight and
birth weight with fetal position at term, subjects were
divided into percentiles of estimated fetal weight at
late ultrasound examination and percentiles of birth
weight: less than 10th percentile, 10th to 90th percen-
tiles, and 90th percentile or more.
For bivariate comparisons, we used ␹2 or Fisher
exact test for categorical data and Student t test for
OBSTETRICS & GYNECOLOGY
continuous variables. Multivariable log binomial re-
gression analysis was then carried out evaluating
maternal and fetal characteristics (smoking in preg-
nancy, number of previous pregnancies, number of
previous abortions, parity, birth weight, gestational
age, amniotic fluid level and placental position at late
ultrasound examination, and estimated fetal weight at
31–35 weeks of gestation) as risk factors for nonvertex
position at delivery. For those characteristics that
were found to be associated with nonvertex presen-
tation at delivery, odds ratios were calculated, adjust-
ing for the effects of the other characteristics. To
assess how well the logistic regression model fits the
data, Pearson ␹2 goodness-of-fit was used (P value
close to 1 indicates a good fit).11 Adjusted risk differ-
ence was also calculated. Statistical analysis was per-
formed with SAS 9.0 (SAS Institute Inc., Cary, NC).
RESULTS
Table 1 shows the change in fetal position between
early and late ultrasound examinations. In early preg-
nancy, defined as 15–22 weeks, 57% of fetuses had
nonvertex position, including breech, transverse, and
variable presentations. By 31–35 weeks, 10% had
nonvertex position, but there was no apparent asso-
ciation between nonvertex presentation early in preg-
nancy and nonvertex position at 31–35 weeks. Ninety
women underwent external cephalic version, with a
success rate of 60% (54 of 90).
Table 2 demonstrates the distribution of selected
maternal and fetal characteristics for infants that were
vertex and nonvertex at delivery. There were no
significant demographic differences between the two
groups. Mothers who delivered nonvertex infants
were more likely to have smoked during pregnancy
than those who delivered vertex infants (17% com-
pared with 12%, P⫽.005). Mothers delivering nonver-
tex infants were more likely to be nulliparous (59%)
compared with those whose infants who presented in
cephalic presentation (45%) (P⬍.001). Amniotic fluid
level was also found to be significantly associated with
Table 1. Change in Fetal Position Between Early
and Late Ultrasound Examinations
Fetal Position at 15–22 Fetal Position at 31–35 wk
wk
Position n (%) Vertex (%) Nonvertex (%)
Vertex 2,881 (43) 90 10
Breech 2,161 (33) 90 10
Transverse 968 (14) 87 13
Variable 636 (10) 90 10
Total [n (%)] 6,646 (100) 5,954 (90) 692 (10)
VOL. 111, NO. 4, APRIL 2008
fetal position at term, as was placental position at late
ultrasound.
Infants who were nonvertex at delivery had a
lower mean weight compared with those who were
vertex, but they were also more likely to have deliv-
ered at an earlier gestational age. Both estimated fetal
weight by ultrasound at 31–35 weeks and birth weight
percentile adjusting for gestational age at birth indi-
cate that there is no significant difference in fetal size
between vertex- and nonvertex-presenting infants.
Of all the variables examined, only four charac-
teristics (parity, smoking status during pregnancy,
decreased volume of amniotic fluid, and placental
position at late ultrasound examination) were found
to be associated with nonvertex presentation at deliv-
ery in the fully adjusted multivariable model (Table
3). As parity increased from zero to one previous
birth, there was a 42% (95% confidence interval [CI]
26 –55%) decrease in the risk of nonvertex delivery,
and in women with more than one previous birth, the
risk of nonvertex delivery decreased by 49% (95% CI
28 – 63%), compared with primiparous women. When
compared with those infants born to women who did
not smoke during pregnancy, infants born to those who
smoked during pregnancy had a 47% increase in risk of
nonvertex position at delivery (95% CI 10 –96%). The
absolute risk difference was 2%. A decreased volume of
amniotic fluid at 31–35 weeks increased the risk of
nonvertex position at delivery by a factor of 3.74 (95%
CI 1.85–7.53). Fundal position of the placenta at late
ultrasound examination was associated with a significant
increase in the risk of nonvertex position at delivery by
a factor of 1.85 (95% CI 1.23–2.78).
Nonvertex position at 31–35 weeks was a risk
factor for nonvertex position at delivery, and this risk
increased with each increasing week of gestation as
shown in Figure 1. However, by 35 weeks, a nonver-
tex fetus still had a 45% chance to convert spontane-
ously and deliver in a vertex position.
DISCUSSION
By examining a large cohort of women who received
ultrasound evaluation throughout pregnancy, we were
able to determine what characteristics might predict
term nonvertex presentation. Although some of the risk
factors are unchangeable, identifying potentially modi-
fiable risk factors might allow for preventive strategies.
Furthermore, these may contribute to understanding the
etiology of nonvertex presentation.
Controversy exists concerning the relationship
between fetal weight and risk of breech delivery.8,12,13
When adjusted for gestational age, birth weight in the
current study was not associated with nonvertex
Witkop et al Natural History of Fetal Position 877
Table 2. Patient Characteristics
Vertex at Delivery Nonvertex at Delivery
Maternal Characteristics (nⴝ6,669) (nⴝ376) P
Age (y, mean⫾SD) 28.5⫾4.1 28.5⫾3.8 .82
Race .35
White 6,311 (95) 360 (96)
Other 358 (5) 16 (4)
Education .65
High school graduate 1,944 (29) 99 (26)
Some college 1,964 (30) 128 (33)
College graduate 2,133 (32) 115 (31)
Graduate school 626 (9) 37 (10)
Gestational diabetes 135 (2) 4 (1) .19
Prepregnancy weight (kg, mean⫾SD) 62⫾12 62⫾12 .82
Smoking in pregnancy 823 (12) 65 (17) .005
Uterine scar from previous surgery 282 (4.2) 21 (5.6) .21
No. of previous pregnancies .02
0 2,333 (35) 162 (43)
1 2,408 (36) 120 (32)
2 1,219 (18) 53 (14)
3 or more 709 (11) 41 (11)
No. of previous abortions ⬍.001
0 5,846 (88) 307 (82)
1 681 (10) 55 (14)
2 or more 142 (2) 14 (4)
No. of miscarriages .93
0 5,705 (86) 321 (85)
1 or more 964 (14) 55 (15)
Parity ⬍.001
0 2,995 (45) 223 (59)
1 2,480 (37) 106 (28)
2 or more 1,194 (18) 47 (13)
Amniotic fluid volume level at late ultrasound examination ⬍.001
Normal 6,325 (98.6) 342 (95)
Decreased 39 (0.6) 10 (3)
Increased 51 (0.8) 6 (2)
Placental position at late ultrasound examination .04
Anterior 3,112 (48) 176 (49)
Posterior 2,799 (44) 142 (40)
Lateral 165 (3) 9 (2)
Fundal 339 (5) 31 (9)
Infant sex, male 3,382 (51) 182 (48) .38
Birth weight (g, mean⫾SD) 3,514⫾464 3,440⫾473 .003
Gestational age (wk, mean⫾SD) 39.5⫾1.4 39.1⫾1.4 ⬍.001
Percentile of estimated fetal weight at a given gestational .93
week at late ultrasound examination
Less than 10% 232 (4) 16 (4)
10–89.9% 4,930 (77) 271 (76)
90% or more 1,228 (19) 71 (20)
Percentile of birth weight at a given gestational week .58
Less than 10% 429 (6) 21 (6)
10–89.9% 5,606 (84) 318 (84)
90% or more 634 (10) 37 (10)
SD, standard deviation.
Data are expressed as mean⫾standard deviation or n (%).

position. Furthermore, there was no difference in the tion or percentiles of birth weight at a given gesta-
percentage of the infants who delivered vertex com- tional week. This was in direct contrast to a large
pared with nonvertex when divided into percentiles cross-sectional study examining fetal size as a risk
of estimated fetal weight at late ultrasound examina- factor for breech birth, in which term breech presen-

878 Witkop et al Natural History of Fetal Position OBSTETRICS & GYNECOLOGY

Table 3. Multivariable Analysis on Risks Factors for Nonvertex Position at Delivery and Adjusted Risk
Difference
Characteristics
Parity
0 1.0
1 0.58
2 or more
Smoking during pregnancy
Decreased volume of amniotic fluid
Placental position at late ultrasound examination
Posterior
Anterior
Fundal
Lateral
OR, odds ratio; CI, confidence interval.
Pearson ␹2 goodness-of-fit: P⫽.64.
* Positive values indicate a risk increase above that in the reference group, while negative values indicate a risk reduction below that in the
reference group.
tation was associated with smaller fetal size for gesta-
tional age when divided into birth weight gestational
age percentiles.12 It appears from the current study
that, in low-risk pregnancies at gestational ages near
term, weight is not associated with risk of breech
delivery.
This study found that smoking may be a modifi-
able risk factor for nonvertex delivery, even after
controlling for potential confounders such as fetal
weight. It is, therefore, unlikely that intrauterine
growth restriction related to smoking is the etiologic
factor for the increased risk of breech delivery at term.
Small studies have demonstrated an association be-
tween decreased intrauterine movement of the fetus
and cigarette smoking.14 –17 Smoking was also associ-
ated with breech birth in a large, population-based,
case-control study.8 Our hypothesis is that cigarette
smoking may decrease fetal muscle tone and move-
ments in utero. One can imagine that, with decreased
Fig. 1. Nonvertex position at 31–35 weeks and risk of
nonvertex position at delivery (with 95% confidence inter-
vals).
Witkop. Natural History of Fetal Position. Obstet Gynecol 2008.
VOL. 111, NO. 4, APRIL 2008
OR (95% CI) Risk Difference (%) (95% CI)*
Reference
(0.45 to 0.74) –2.7 (–1.5 to –3.8)
0.51 (0.37 to 0.72) –3.0 (–1.7 to –4.4)
1.47 (1.10 to 1.96) 2.0 (0.2 to 3.7)
3.74 (1.85 to 7.53) 12.0 (1.7 to 22.2)
1.0 Reference
1.11 (0.89 to 1.40) 0.5 (–0.5 to 1.6)
1.85 (1.23 to 2.78) 3.3 (0.5 to 6.2)
1.03 (0.51 to 2.06) 0.8 (–2.3 to 4.1)
kicking near term, the fetus may not have the motor
capacity to move into a vertex position. Smoking is
not the only risk factor for decreased movements in
the fetus. For example, diminished fetal movements
have been noted in diabetic pregnancies,18 and dia-
betes has been found to be a risk factor for breech
delivery.8
Decreased intra-amniotic fluid also appears to
increase the risk for nonvertex delivery. We hypoth-
esize that lower fluid level is associated with a reduc-
tion in the space available in the intrauterine environ-
ment for the fetus to accomplish the necessary
motions to move into a vertex position. This theory is
somewhat supported by numerous reports of in-
creased external cephalic version success with normal
or elevated fluid levels as compared with reduced
fluid levels.19,20
A recent multicenter randomized controlled trial
investigating external cephalic version beginning at
34 weeks of gestation compared with 37 weeks of
gestation found that early external cephalic version
could reduce risk of nonvertex presentation at deliv-
ery.21 Breech presentation at term is associated with
nonvertex presentation on ultrasound earlier in preg-
nancy, and this risk increases with gestational age
from 31 to 35 weeks. However, we found that even
with nonvertex presentation at 35 weeks, the rate of
spontaneous version is still high, arguing against early
intervention for nonvertex position.
This study is unique in that it offers a longitudinal
view of the natural history of fetal position throughout
pregnancy. The findings can aid clinicians in counsel-
ing patients regarding management of the nonvertex
fetus near term. The study also demonstrates yet
another potentially adverse outcome that may be
Witkop et al Natural History of Fetal Position 879
related to smoking in pregnancy. The association
between smoking and nonvertex presentation at term
may be useful in smoking cessation counseling during
the preconception and prenatal periods and may
motivate the pregnant woman who wants to reduce
her chances of cesarean delivery.
REFERENCES
1. Albrechtsen S, Rasmussen S, Dalaker K, Irgens LM. The
occurrence of breech presentation in Norway 1967–1994. Acta
Obstet Gynecol Scand 1998;77:410–5.
2. Cunningham FG, Levenko KJ, Bloom SL, Hauth JC, Gilstrap
LC, Wenstrom KD. Breech presentation and delivery. In:
Williams obstetrics. 22nd ed. New York (NY): McGraw-Hill;
2005. p. 509 –36.
3. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F,
Munson ML. Births: final data for 2002. Natl Vital Stat Rep
2003;52:1–113.
4. Cheng M, Hannah M. Breech delivery at term: a critical
review of the literature. Obstet Gynecol 1993;82:605–18.
5. Krebs L. Langhoff-Roos J. Elective cesarean delivery for term
breech. Obstet Gynecol 2003;101:690 – 6.
6. Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S,
Willan AR. Planned caesarean section versus planned vaginal
birth for breech presentation at term: a randomised multicen-
tre trial. Lancet 2000;356:1375–83.
7. American College of Obstetricians and Gynecologists. Exter-
nal cephalic version. ACOG Practice Bulletin 13. Washington,
DC: ACOG; 2000.
8. Rayl J, Gibson PF, Hickok DE. A population-based case-
control study of risk factors for breech presentation. Am J
Obstet Gynecol 1996;174:28–32.
9. Ewigman BG, Crane JP, Frigoletto FD, LeFevre ML, Bain RP,
McNellis D. Effect of prenatal ultrasound screening on perina-
tal outcome. N Engl J Med 1993;329:821–7.
880 Witkop et al Natural History of Fetal Position
10. Phelan JP, Smith CV, Broussard P, Small M. Amniotic fluid
volume assessment with the four-quadrant technique at 36 – 42
weeks’ gestation. J Reprod Med 1987;32:540–2.
11. Hosmer DW, Lemeshow S. Applied logistic regression. 2nd
ed. New York (NY): John Wiley & Sons; 2000. p. 145–7.
12. Roberts CL, Algert CS, Peat B, Henderson-Smart D. Small
fetal size: a risk factor for breech birth at term. Int J Gynecol
Obstet 1999;67:1–8.
13. Kian LS. The role of placental site in the aetiology of breech
presentation. J Obstet Gynaecol Br Commonw 1963;70:795–7.
14. Manning FA, Feyerabend C. Cigarette smoking and fetal
breathing movements. Br J Obstet Gynecol 1976;83:262–70.
15. Wood C, Gilbert M, O’Connor A, Walters WA. Subjective
recording of fetal movement. Br J Obstet Gynecol 1979;86:
836–42.
16. Graca LM, Cardoso CG, Clode N, Calhaz-Jorge C. Acute
effects of maternal cigarette smoking on fetal heart rate and
fetal body movements felt by the mother. J Perinat Med
1991;19:385–90.
17. Bocking AD. Observations of biophysical activities in the
normal fetus. Clin Perinatol 1989;16:583–94.
18. Devoe LD, Youssef AA, Castillo RA, Croom CS. Fetal bio-
physical activities in third trimester pregnancies complicated
by diabetes mellitus. Am J Obstet Gynecol 1994;171:298–303.
19. Healey M, Porter R, Galimberti A. Introducing external
cephalic version at 36 weeks or more in a district general
hospital: a review and an audit. Br J Obstet Gynaecol 1997;
104:1073–9.
20. Shalev E, Battino S, Giladi Y, Edelstein S. External cephalic
version at term— using tocolysis. Acta Obstet Gynecol Scand
1993;72:455–7.
21. Hutton EK, Kaufman K, Hodnett E, Amankwah K, Hewson
SA, McKay D, et al. External cephalic version beginning at 34
weeks’ gestation versus 37 weeks’ gestation: a randomized
multicenter trial. Am J Obstet Gynecol 2003;189:245–54.
OBSTETRICS & GYNECOLOGY