Accepted Manuscript

Uterocervical angle: a novel ultrasound screening tool to predict spontaneous preterm

birth

Margaret Dziadosz, MD, Terri-Ann Bennett, MD, Cara Dolin, MD, Anne West Honart,
MD, Ms Amelie Pham, Ms. Sarah S. Lee, Ms. Sarah Pivo, MA, Ashley S. Roman, MD

PII: S0002-9378(16)00525-1
DOI: 10.1016/j.ajog.2016.03.033
Reference: YMOB 11012

To appear in: American Journal of Obstetrics and Gynecology

Received Date: 1 March 2016

Accepted Date: 17 March 2016

Please cite this article as: Dziadosz M, Bennett T-A, Dolin C, West Honart A, Pham A, Lee SS, Pivo S,
Roman AS, Uterocervical angle: a novel ultrasound screening tool to predict spontaneous preterm birth,
American Journal of Obstetrics and Gynecology (2016), doi: 10.1016/j.ajog.2016.03.033.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1

Uterocervical angle: a novel ultrasound screening tool to predict spontaneous preterm birth

Margaret DZIADOSZ, MD

Terri-Ann BENNETT, MD

PT
Cara DOLIN, MD

RI
Anne WEST HONART, MD

Ms. Amelie PHAM

SC
Ms. Sarah S LEE

Ms. Sarah Pivo, MA

Ashley S ROMAN, MD
U
AN
Affiliation:

New York University Langone Medical Center

M

Division of Maternal Fetal Medicine

D

Department of Obstetrics & Gynecology

550 1st Avenue

TE

New York, NY 10016

EP

The authors report no conflicts of interest.

C

There was no financial support of this study.

AC

Presentation: “The 2016 36th Annual Pregnancy Meeting,” Society of Maternal Fetal Medicine,

Atlanta, Georgia USA February 5, 2016.

Control ID 1037, Program ID 46, Oral presentation Fellows Plenary II

 ACCEPTED MANUSCRIPT
2

Corresponding author:

Margaret DZIADOSZ, MD

New York University Langone Medical Center

Department of Obstetrics and Gynecology

PT
Division of Maternal Fetal Medicine

RI
550 1st Avenue New York, NY 10016

margaret.dziadosz@gmail.com

SC
Work Phone: 212-263-0223

U
Mobile Phone: 973-934-2393
AN
Fax: 212-562-2754
M

Word count:

Abstract: 301
D

Main Text: 2693

TE
C EP
AC
 ACCEPTED MANUSCRIPT
3

Condensation

Measurement of transvaginal uterocervical angle in second trimester is predictive of

spontaneous preterm birth

Short version of title

PT
Uterocervical angle screen

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
4

Abstract

Background: Mechanical alteration of the cervical angle has been proposed to reduce

spontaneous preterm birth (sPTB). Performance of the uterocervical angle (UCA) as measured

PT
by ultrasound for predicting sPTB is poorly understood.

RI
Objective: To determine whether a novel ultrasonographic marker, UCA, correlates with risk of

sPTB in a general population.

SC
Study Design: We conducted a retrospective cohort study from May 2014 to May 2015 of

singleton gestations between 16 0/7 – 23 6/7 weeks undergoing transvaginal ultrasound (TVU)

U
AN
for cervical length (CL) screening. Images were re-measured for UCA between the lower uterine

segment and the cervical canal. Primary outcome was prediction of sPTB <34 weeks and <37
M

weeks by UCA and secondary outcome evaluated CL and sPTB.

D

Results: A total of 972 women were studied. The rate of sPTB in this cohort was 9.6% for
TE

delivery <37 weeks and 4.5% for <34 weeks. A UCA of ≥95o was significantly associated with

sPTB <37 weeks with sensitivity of 80% (p<0.001, CI 0.70-0.81, NPV 95%). A UCA of ≥105 o
EP

predicted sPTB <34 weeks with sensitivity of 81% (p<0.001, CI 0.72-0.86, NPV 99%). CL ≤25mm

significantly predicted sPTB <37 weeks (p<0.001, sensitivity 62%, NPV 95%) and <34 weeks
C

(p<0.001, sensitivity 63%, NPV 97%). Regression analysis revealed a significant association of
AC

maternal age, nulliparity, race, and obesity at conception with sPTB and UCA. There was no

correlation identified between history of dilation and curettage, abnormal pap smear results,

excisional cervical procedures, smoking, or obesity at delivery on sPTB and UCA.

Conclusion: A wide UCA ≥95o and ≥105o detected during the 2nd trimester was associated with
 ACCEPTED MANUSCRIPT
5

an increased risk for sPTB <37 and <34 weeks, respectively. UCA performed better than CL in

this cohort. Our data indicate that UCA is a useful, novel transvaginal ultrasonographic marker

that may be used as a screening tool for sPTB.

PT
RI
SC
Key words:

Uterocervical angle, preterm birth, transvaginal ultrasound, cervical length

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
6

Introduction

Preterm birth remains the leading cause of neonatal morbidity and mortality worldwide1. The

March of Dimes reports a preterm birth rate of 9.6% in the United States2. Currently, the ability

PT
to identify women at risk of sPTB is limited and includes a detailed history, transvaginal

ultrasound (TVU), or symptoms such as preterm contractions3,4. We now have several

RI
interventions to decrease the risk of spontaneous preterm birth (sPTB) when at-risk women are

identified1,5, including progesterone supplementation, cerclage placement, and vaginal pessary

SC
insertion6-11.

U
Mechanisms underlying spontaneous preterm birth are complex. Cervical tissue, composed of a
AN
matrix of collagen fibers, is supported by the cardinal and uterosacral ligaments24-26. The cervix

experiences pressures from surrounding pelvic organs and withstands forces from the growing
M

uterus. A combination of physiologic pressures and individual anatomy affect the internal os
D

and cervical function25,26. Clinical cervical function can be described through cervical structure
TE

integrity as is displayed through ultrasonographic cervical length and uterocervical angle24. A

wide, or obtuse, uterocervical angle (UCA) lends a more direct, linear outlet of uterine contents
EP

onto the cervix. A narrower, or acute, UCA supports an anatomical geometry that would exert

less direct force on the internal os, which may be protective from cervical deformation.
C
AC

As early as the 1950’s, pessaries were thought to create an immunological barrier and

mechanically change the inclination of the cervical canal, thereby distributing pelvic force away

from the cervix12-17. Altered uterocervical angle to a more acute, or narrowed, angle after

vaginal pessary placement has been confirmed by MRI12,18-23.

 ACCEPTED MANUSCRIPT
7

Pelvic angles can be visualized and measured during TVU examination performed in pregnancy.

Sochacki-Wojcicka et al. have suggested that angles between the uterus and cervix may be

related to gestational age at delivery27. In many practices, second trimester universal TVU

PT
screening for cervical length (CL) is offered to identify women at increased risk of preterm

birth3,4,28. Our objective was to evaluate whether UCA can predict risk of sPTB in a general

RI
population of singletons and to evaluate its performance for predicting sPTB relative to CL.

SC
Materials and Methods

This retrospective cohort study was designed to evaluate the performance of uterocervical

U
angle measurement during routine TVU screening for CL in singleton gestations. Institutional
AN
Review Board approval was obtained with exemption prior to initiating the study. Consecutive

women who delivered at our institution between May 1, 2014 and May 30, 2015 by university-
M

practice associated practitioners were eligible. Women were identified from institutional
D

computerized lists of deliveries by provider and date of delivery.

TE

TVU CL screening is routinely performed in all women with singleton gestations who are

scheduled for an ultrasound between 16 0/7 and 23 6/7 weeks’ gestation4,28 in our practice.
EP

TVU CL was performed in a uniform fashion according to Cervical Length Education and Review

(CLEAR) criteria by RDMS accredited sonographers who are monitored by Maternal Fetal
C

Medicine attending staff29. Transvaginal images were obtained with the 4-9MHz IC5-9D wide
AC

view transducer (Voluson e8 GE Healthcare) or 4-8MHz C8-4v wide view transducer (IU22

Phillips Healthcare). Images used to report CL from visits for anatomical survey studies were

accessed and re-measured for UCA.

 ACCEPTED MANUSCRIPT
8

The UCA is the triangular segment measured between the lower uterine segment and the

cervical canal, yielding a measurable angle (Figure 1). The first ray was placed from the internal

os to the external os. The calipers were placed where the anterior and posterior walls of the

PT
cervix touch the internal and external os along the endocervical canal. If the cervix was curved,

the first ray was also drawn from the internal os to the external os as a straight line. A second

RI
ray was then drawn to delineate the lower uterine segment. This ray was traced up the anterior

SC
uterine segment to a distance allowed by the preloaded image. Ideally, the second ray would

reach 3cm up the lower uterine segment in order to establish an adequate measurement. The

U
anterior angle in between the two rays was measured with a protractor.
AN
In the presence of funneling, the first ray was placed to measure the length of remaining cervix.

The second caliper was placed from the innermost portion of measurable cervix and extended
M

to the lower uterine segment. In the event that the lower uterine segment was found to be
D

irregular, the second caliper was placed centrally along the segment. In the event of a
TE

retroverted uterus, the angle should be measured in a similar fashion with the first ray along

the measurable cervix and the second ray traced along the lower uterine segment. Unlike in an
EP

anteverted or axial positioned uterus, however, the posterior side of the angle closer to the

intrauterine contents should then be measured. Inter and intra-observer UCA measurements
C

were compared for variability.

AC

Women were included in the analysis if they were between the ages of 18 to 50 years, carrying

a singleton gestation, delivered at our institution between May 1, 2014 and May 30, 2015, and

had TVU CL screening images performed at our antenatal care center between 16 0/7 and 23

6/7 weeks. Women were excluded from the study if no TVU CL image was available for review
 ACCEPTED MANUSCRIPT
9

or if an available image was considered sub-optimal due to an inability to visualize landmarks

defined by Iams et al, such as the lower uterine segment anteriorly or if there was excess

compression of the cervix30. Women were also excluded from analysis if they experienced

PT
medically indicated preterm birth. Demographic and pregnancy outcome data was collected via

chart review.

RI
Our primary outcome was the prediction of sPTB <34 and <37 weeks’ gestation by UCA

SC
measured during routine TVU for CL during the second trimester. A secondary outcome

formulated prior to data collection included evaluation of sPTB predicted by TVU CL ≤25mm.

U
Receiver operating characteristic (ROC) curves were developed to determine an optimal UCA
AN
for prediction of sPTB at <34 and <37 weeks’ gestation. Fisher’s exact test, χ2 test, student t

test, Kappa coefficient, Spearman’s correlation, likelihood ratios and stepwise linear regression
M

were performed as appropriate with p<0.05 defined as significance using SPSS 21.0 (2012,
D

version 21; IBM Corp, Armonk, NY).

TE

Results

A total of 1109 women were eligible for inclusion during the study period. After 137 (12%)
EP

women were excluded, 115 for lack of adequate TVU image availability and 22 due to medically

indicated preterm birth, 972 (88%) women remained for inclusion. The Kappa coefficient was
C

0.94 for masked inter-observer variability and 0.90 for masked intra-observer variability, both
AC

evident of strong agreement between measured UCA.

The rate of sPTB <37 weeks was 9.6% (n=84) and 4.5% (n=43) at <34 weeks. Women who

delivered preterm were similar to those who delivered ≥37 weeks with respect to age, race,

nulliparity, mode of conception, smoking, cervical procedures, maternal diabetes, and maternal
 ACCEPTED MANUSCRIPT
10

hypertension. However, compared with women who delivered ≥37 weeks, women who

delivered preterm were more likely to have had a prior sPTB, a history of dilation and curettage

(D&C), a shorter CL in the second trimester, a higher BMI, and a cesarean delivery (Table 1).

PT
ROC curves were generated to evaluate UCA and gestational age at delivery. When women who

delivered <37 weeks were compared to those who delivered ≥37 weeks, the area under the

RI
curve was 0.78 (p <0.001, SE ±0.03, 95% CI 0.70 - 0.81). A point was chosen on the ROC curve

corresponding with a UCA of 95o for optimal sensitivity (81%) and minimal false positive rates

SC
(Figure 2A). For women who delivered <34 weeks, the ROC curve generated an area under the

U
curve of 0.787 (p <0.001, SE ±0.03, 95% CI 0.72 – 0.86). The optimal UCA cut off point was
AN
chosen at 105o with a sensitivity of 80% (Figure 2B).

A UCA of >95o was significantly associated with sPTB <37 weeks with a sensitivity of 80%
M

(p<0.001, RR 4.3, specificity 53%, PPV 14%, NPV 95%). In this group, 484 women had a UCA
D

>95o and 68 (14%) of these women delivered <37 weeks, while of 488 women with UCA <95o,
TE

only 16 (3.3%) delivered <37 weeks’ gestation. A UCA >105o predicted sPTB <34 weeks with a

sensitivity of 81% and was significant (p<0.001, RR 7.2, specificity 65%, PPV 10%, NPV 99%). Of
EP

366 women with a UCA >105o, 35 (9.6%) delivered <34 weeks, while of 606 women with UCA

<105o, only 8 (1.3%) women delivered <34 weeks’ gestation.

C

Stepwise linear regression was performed to evaluate for confounders to UCA and its
AC

prediction of sPTB. Analysis revealed a significant association between maternal age, obesity at

conception, nulliparity, and race with UCA and sPTB. However, condition indices for the

included variables were all <17, which does not imply a strong level of correlation between

variables. There was a significant association noted between a more narrow UCA and a history
 ACCEPTED MANUSCRIPT
11

of prior cesarean delivery, however this association was excluded with stepwise linear

regression. There was no correlation identified with short CL, prior preterm birth,

administration of supplemental progesterone, mode of conception, a history of abnormal pap

PT
smear results, LEEP procedure, cervical conization, history of D&C, smoking or mode of delivery

(Table 2).

RI
In a secondary analysis, TVU CL measured during routine second trimester anatomical survey

SC
was evaluated for association with sPTB. Of 84 women who delivered <37 weeks, 71 (85%) had

a CL ≥25mm and 13 (15%) had a CL ≤25mm in the second trimester. ROC curves for CL

U
measurement in the second trimester and sPTB <37 weeks and <34 weeks resulted in areas
AN
under the curve of 0.372 and 0.315 respectively, indicating a poor performance of this test in

our population. CL ≤25mm was found to be associated with sPTB <37 weeks with a sensitivity of
M

15% and a specificity of 98% (p<0.001, RR 6.7, PPV 46%, NPV 92%). A CL ≤25mm in the second
D

trimester was also associated with sPTB <34 weeks (p<0.001, RR 7.7, sensitivity 19%, specificity
TE

98%, PPV 29%, NPV 96%).

As neither CL nor UCA measurements followed a linear pattern and both had outlying data
EP

points, Spearman’s correlation was calculated and found to be not significant between the two

testing parameters (p 0.12). The correlation coefficient was -0.05 which implies that as CL
C

decreases, UCA increases. This is consistent with the findings of our study when predicting
AC

sPTB.

We examined the likelihood ratios of both UCA and CL in order to compare their performance.

The positive likelihood ratios of CL ≤25mm were 9.2 and 8.3 for sPTB <37 and <34 weeks’. The

negative likelihood ratios of UCA < 95o and <105o for sPTB <37 and <34 weeks’ were 0.36 and
 ACCEPTED MANUSCRIPT
12

0.29. We evaluated testing in two other ways: (a) either a positive measurement for UCA or a

short CL (b) both a positive measurement for UCA and a short CL. The best performance was

demonstrated by combined testing, yielding positive likelihood ratios that ranged 17 – 20 (Table

PT
3-4).

Comments

RI
We performed a retrospective cohort study to determine whether TVU UCA measurement

SC
performed between 16 and 23 6/7 weeks’ gestation could predict sPTB. We propose that

positioning and shape of pelvic organs affect the mechanical function of the cervix in

U
pregnancy. Physiologic loading forces of pregnancy lend compression and tension to the cervix,
AN
while personal anatomy drives specific patterns of stress onto the internal os. The more acute

the uterocervical angle, the less direct loading pressure rests on the internal os, while a wider
M

UCA that contributes a larger load may predict spontaneous preterm birth due to aberrant
D

physiologic pressures.
TE

Principle findings

As a single measurement, UCA ≥95o and ≥105o was a significant predictor of sPTB at <37 and at
EP

<34 weeks respectively. Our data indicates that UCA measurement is significantly associated

with risk of sPTB in women with singleton gestations.

C

Several statistical parameters need to be considered in order to evaluate a screening test. In

AC

prior studies evaluating CL as a screening tool, when TVU was performed in the second

trimester, a CL ≤25mm in low risk women, yielded a sensitivity of 37% for spontaneous preterm

birth31, whereas CL evaluation of women at high risk of sPTB yielded a sensitivity of only 19%32.
 ACCEPTED MANUSCRIPT
13

The PPV of CL screening has ranged between 18% in studies on low risk women to 75% in high-

risk women31,32.

The primary results of our study showed that UCA had greater sensitivity of 80-81% than did CL

PT
in our population for prediction of sPTB <37 and <34 weeks’. Assessment of secondary

outcomes in our cohort revealed that TVU CL ≤25mm has a sensitivity of 15 – 19%, which is

RI
similar to previously published data. PPV of UCA was low at 10% and the PPV of TVU CL was

SC
46%. As PPV is dependent on the prevalence of sPTB in the population studies, confounders

including maternal age, nulliparity, race and maternal weight may have biased our data.

U
Importantly, the NPV of UCA ranged between 95-99% in our study population. This indicates
AN
that with a negative UCA screen, there is a very low likelihood of developing sPTB. Therefore,

with an acute measure of the UCA, there may be minimal concern for sPTB <34 and <37 weeks.
M

Clinical Implications
D

In our study, UCA performed well compared to CL as a screening tool. In fact, UCA performed
TE

better than CL with a higher sensitivity and NPV. Also, the negative likelihood ratios of UCA

ranged from 0.29 – 0.36 for prediction of sPTB. These data imply that when a patient does not
EP

screen positive with an obtuse UCA measurement, it is highly unlikely that she will go on to

have a sPTB and may not require additional cervical monitoring. Regression analysis showed
C

that unlike CL, UCA measurement is not affected by a history of D&C, abnormal pap smears, or
AC

cervical instrumentation. As many women undergo such procedures or will have an abnormal

pap smear result during their childbearing years, it is beneficial that UCA may be used without

concern for these confounding effects. Though CL ≤25mm has a strong positive likelihood ratio,

it has a worse negative likelihood ratio than UCA for predicting sPTB. The use of CL as a
 ACCEPTED MANUSCRIPT
14

screening tool has been extensively studied and the combination of both measurements may

be the best predictor of a risk of sPTB in our patients.

Strengths & Limitations

PT
This study has several strengths. It is the first study to evaluate UCA as a predictor of sPTB, a

well-defined outcome, in a large cohort of singleton gestations. In our study population, the

RI
two groups of women were similar and allowed for the evaluation of a robust sample of TVU

SC
images. The sPTB rates exhibited by our population reflect those reported in the United States.

All of the sonographers used a previously defined protocol to measure CL. In addition, all of the

U
UCA measurements were performed by a single provider.
AN
There are weaknesses to our study. Due to its retrospective nature, there is potential for both

selection and information bias. Since the information was gathered from computerized charts,
M

data is dependent on the accuracy of information input by medical personnel. We were also
D

limited to previously obtained TVU images for a wide range of gestational ages in the second
TE

trimester. Since we only evaluated UCA at one point in time in each pregnancy, it is unknown

whether serial evaluation of the UCA would correlate more strongly with risk of sPTB.
EP

The syndrome of sPTB is a multifactorial phenomenon. It encompasses maternal risk factors

such as race and parity that are not modifiable. However, it also includes clinical conditions,
C

such as cervical insufficiency, for which we may screen, intervene and initiate treatment. We
AC

conclude that measurement of the UCA may contribute to risk assessment for sPTB in women

with singleton gestations. An acute UCA may reflect cervical competence and resistance to

passage of the fetus through the cervical outlet.

 ACCEPTED MANUSCRIPT
15

Conclusions

Screening of transvaginal UCA measurement in the second trimester is predictive of sPTB in our

cohort. When measured during routine second trimester TVU, UCA may serve as a novel

PT
screening tool. When TVU UCA measurement is combined with CL measurement, stronger

prediction of risk for sPTB would result. Future prospective trials are needed to confirm our

RI
findings and to define the role of UCA as a useful screening tool with more certainty. Until

SC
prospective trials are completed, the use of UCA as a screening test for the prediction of sPTB

should remain investigational.

U
AN
References

1. WHO Guidelines Approved by the Guidelines Review Committee. WHO

M

Recommendations on Interventions to Improve Preterm Birth Outcomes. Geneva: World

Health Organization 2015

2. March of Dimes. Premature Birth Report Card marchofdimes.org/reportcard. 2015

D

3. Practice bulletin no. 130: prediction and prevention of preterm birth. Obstetrics and
TE

gynecology. Oct 2012;120(4):964-973.

4. Berghella V. Universal cervical length screening for prediction and prevention of
preterm birth. Obstet & gynecol survey. Oct 2012;67(10):653-658.
EP

5. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm
birth. Lancet. Jan 5 2008;371(9606):75-84.
6. Meis PJ, Klebanoff M, Thom E, et al. Prevention of recurrent preterm delivery by 17
alpha-hydroxyprogesterone caproate. NEJM. Jun 12 2003;348(24):2379-2385.
C

7. Sibai BM, Istwan NB, Palmer B, Stanziano GJ. Pregnancy outcomes of women receiving
compounded 17 alpha-hydroxyprogesterone caproate for prophylactic prevention of
AC

preterm birth 2004 to 2011. Amer J of perinat. Sep 2012;29(8):635-642.

8. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH. Progesterone and the risk of
preterm birth among women with a short cervix. NEJM. Aug 2 2007;357(5):462-469.
9. Hassan SS, Romero R, Vidyadhari D, et al. Vaginal progesterone reduces the rate of
preterm birth in women with a sonographic short cervix: a multicenter, randomized,
double-blind, placebo-controlled trial. Ultrasound in Obstetrics and Gynecology. Jul
2011;38(1):18-31.
 ACCEPTED MANUSCRIPT
16

10. Berghella V, Odibo AO, To MS, Rust OA, Althuisius SM. Cerclage for short cervix on
ultrasonography: meta-analysis of trials using individual patient-level data. Obstet and
gynecol. Jul 2005;106(1):181-189.
11. Schaaf JM, Hof MH, Mol BW, Abu-Hanna A, Ravelli AC. Recurrence risk of preterm birth
in subsequent singleton pregnancy after preterm twin delivery. AJOG. Oct
2012;207(4):279.e271-277.

PT
12. Cannie MM, Dobrescu O, Gucciardo L, et al. Arabin cervical pessary in women at high
risk of preterm birth: a magnetic resonance imaging observational follow-up study.
Ultrasound in Obstetrics and Gynecology. Oct 2013;42(4):426-433.
13. Goyal A, Fishwick J, Hurrell R, Cervellione RM, Dickson AP. Antenatal diagnosis of

RI
bladder/cloacal exstrophy: challenges and possible solutions. Journal of pediatric
urology. Apr 2012;8(2):140-144.

SC
14. Arabin B, Alfirevic Z. Cervical pessaries for prevention of spontaneous preterm birth:
past, present and future. Ultrasound in Obstetrics and Gynecology. Oct 2013;42(4):390-
399.
15. Sieroszewski P, Jasinski A, Perenc M, Banach R, Oszukowski P. The Arabin pessary for the

U
treatment of threatened mid-trimester miscarriage or premature labour and
miscarriage: a case series. The journal of maternal-fetal & neonatal medicine Jun
AN
2009;22(6):469-472.
16. Abdel-Aleem H, Shaaban OM, Abdel-Aleem MA. Cervical pessary for preventing preterm
birth. The Cochrane database of systematic reviews. 2013;5:Cd007873.
M

17. Goya M, Pratcorona L, Merced C, et al. Cervical pessary in pregnant women with a short
cervix (PECEP): an open-label randomised controlled trial. Lancet. May 12
2012;379(9828):1800-1806.
D

18. Kyvernitakis I, Khatib R, Stricker N, Arabin B. Is Early Treatment with a Cervical Pessary
an Option in Patients with a History of Surgical Conisation and a Short Cervix?
TE

Geburtshilfe und Frauenheilkunde. Nov 2014;74(11):1003-1008.

19. Araujo Junior E, Santana EF, Nardozza LM, Moron AF. Association of progesterone,
pessary, and antibiotic for treating pregnant woman with short cervix syndrome:
importance of magnetic resonance imaging in the assessment of pessary position.
EP

Journal of clinical imaging science. 2013;3:27.

20. Liem SM, van Pampus MG, Mol BW, Bekedam DJ. Cervical pessaries for the prevention
of preterm birth: a systematic review. Obstetrics and gynecology international.
C

2013;2013:576723.
21. Komesu YM, Ketai LH, Rogers RG, Eberhardt SC, Pohl J. Restoration of continence by
AC

pessaries: magnetic resonance imaging assessment of mechanism of action. AJOG. May

2008;198(5):563.e561-566.
22. House M, Bhadelia RA, Myers K, Socrate S. Magnetic resonance imaging of three-
dimensional cervical anatomy in the second and third trimester. European journal of
obstetrics, gynecology, and reproductive biology. May 2009;144 Suppl 1:S65-69.
23. House M, O'Callaghan M, Bahrami S, et al. Magnetic resonance imaging of the cervix
during pregnancy: effect of gestational age and prior vaginal birth. AJOG. Oct
2005;193(4):1554-1560.
 ACCEPTED MANUSCRIPT
17

24. Myers KM, Feltovich H, Mazza E, et al. The mechanical role of the cervix in pregnancy.
Journal of biomechanics. Jun 25 2015;48(9):1511-1523.
25. Fernandez M, House M, Jambawalikar S, et al. Investigating the mechanical function of
the cervix during pregnancy using finite element models derived from high-resolution
3D MRI. Computer methods in biomechanics and biomedical engineering. Mar
2016;19(4):404-417.

PT
26. House M, McCabe R, Socrate S. Using imaging-based, three-dimensional models of the
cervix and uterus for studies of cervical changes during pregnancy. Clinical anatomy
(New York, N.Y.). Jan 2013;26(1):97-104.
27. Sochacki-Wojcicka N, Wojcicki J, Bomba-Opon D, Wielgos M. Anterior cervical angle as a

RI
new biophysical ultrasound marker for prediction of spontaneous preterm birth.
Ultrasound in Obstetrics and Gynecology. Sep 2015;46(3):377-378.

SC
28. Orzechowski KM, Boelig R, Nicholas SS, Baxter J, Berghella V. Is universal cervical length
screening indicated in women with prior term birth? AJOG. Feb 2015;212(2):234.e231-
235.
29. Timor-Tritsch IE, Bashiri A, Monteagudo A, Arslan AA. Qualified and trained

U
sonographers in the US can perform early fetal anatomy scans between 11 and 14
weeks. AJOG. Oct 2004;191(4):1247-1252.
AN
30. Iams JD, Grobman WA, Lozitska A, et al. Adherence to criteria for transvaginal
ultrasound imaging and measurement of cervical length. AJOG. Oct
2013;209(4):365.e361-365.
M

31. Iams JD, Goldenberg RL, Meis PJ, et al. The length of the cervix and the risk of
spontaneous premature delivery. National Institute of Child Health and Human
Development Maternal Fetal Medicine Unit Network. NEJM. Feb 29 1996;334(9):567-
D

572.
32. Owen J, Yost N, Berghella V, et al. Mid-trimester endovaginal sonography in women at
TE

high risk for spontaneous preterm birth. JAMA. Sep 19 2001;286(11):1340-1348.

C EP
AC
 ACCEPTED MANUSCRIPT
18

Table 1 Demographic data

Demographic Birth ≥37 weeks Birth <37 weeks P value

n = 888 (91%) n = 84 (8.6%)
Mean maternal age, y 33 (± 5) 33 (± 5) 1
Race/ethnicity 0.41

PT
White 615 (69) 54 (64)
Non-white 273 (31) 30 (36)
Nulliparous 548 (62) 50 (60) 0.77

RI
Prior spontaneous 58 (6.5) 13 (16) 0.005
preterm birth
Natural conception 785 (88) 73 (87) 0.82

SC
Tobacco use 6 (0.6) 1 (1) 0.46
Prior dilation and 165 (19) 25 (30) 0.01
curettage
Prior cervical 6 (0.7) 2 (2.4) 0.14

U
conization
AN
Prior cervical LEEP 47 (5.3) 4 (4.8) 0.54
Abnormal pap smear 207 (23) 17 (20) 0.61
Mean gestational age 20 (± 3) 21 (± 3) <0.001
at TVU, weeks
M

Mean CL at TVU, cm 40 (± 7) 36 (± 9) <0.001

Mean UCA at TVU 93 (± 26) 120 (± 27) <0.001
Mean body mass index 24 (± 5) 26 (± 5) <0.001
D

at conception, kg/m2
Mean body mass index 29 (± 13) 33 (± 30) 0.04
TE

at delivery, kg/m2
Diabetes 59 (6) 10 (12) 0.11
Hypertensive disorder 115 (13) 13 (15) 0.62
EP

Cesarean delivery, 295 (33) 45 (54) <0.001

primary
Cesarean delivery, 56 (6.5) 8 (9) 0.36
C

repeat
Mean gestational age 39 (± 2.5) 33 (± 4) <0.001
AC

at delivery
Female fetus 729 (48) 38 (45) 0.60

NICU admission 94 (11) 54 (64) <0.001

Data are n (± standard deviation) or n (%)

 ACCEPTED MANUSCRIPT
19

Table 2 Stepwise linear regression analysis of potential confounders on uterocervical angle

Variable Beta Coefficient 95% Confidence P value

Interval
Maternal age (years) 0.14 0.39 - 1.1 <0.001

PT
Nulliparity -0.11 -9.6 - -2.4 0.001
Race -0.10 -9.7 - -2.1 0.002
Obese BMI kg/m2, 0.12 4.7 - 16 0.006

RI
conception

U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
20

Table 3

Test characteristics of uterocervical angle and cervical length for prediction of spontaneous
preterm birth <37 weeks

PT
Parameters Sensitivity Specificity Positive Negative Positive Negative
Predictive Predictive likelihood likelihood
Value Value ratio ratio

RI
UCA >95o 80% 53% 14% 95% 1.7 0.36
(0.71 – (0.52 – (0.12 – (0.95 –
0.88) 0.54) 0.15) 0.98)

SC
CL ≤ 25mm 15% 98% 46% 92% 9.2 0.86
(0.095 – (0.98 – (0.27 – (0.92 –

U
0.22) 0.99) 0.65) 0.93)
P value <0.001 <0.001 <0.001 <0.001
AN
UCA >95o or 62% 66% 15% 95% 1.8 0.58
CL≤ 25mm (0.51 – (0.65 – (0.12 – (0.93 –
M

0.72) 0.67) 0.17) 0.96)

o
UCA >95 12% 99% 63% 92% 18 0.89
and CL≤ (0.069 – (0.98 – (0.36 – (0.92 –
D

25mm 0.16) 0.99) 0.84) 0.93)

(95% Confidence Interval)
TE
C EP
AC
 ACCEPTED MANUSCRIPT
21

Table 4

Test characteristics of uterocervical angle and cervical length for prediction of spontaneous
preterm birth <34 weeks

Parameters Sensitivity Specificity Positive Negative Positive Negative

Predictive Predictive likelihood likelihood

PT
Value Value ratio ratio
UCA >105o 81% 65% 10% 99% 2.3 0.29
(0.66 – (0.64 – (0.08 – (0.98 –

RI
0.91) 0.65) 0.11) 0.99)
CL ≤ 25mm 19% 98% 29% 96% 8.3 0.83
(0.093 – (0.97 – (0.14 – 0.96 – 0.97)

SC
0.31) 0.98) 0.47)
P value <0.001 <0.001 <0.001 <0.001

U
UCA >105o 63% 65% 7% 97% 1.8 0.57
or CL≤ (0.47 – (0.64 – (0.057 – (0.96 –
AN
25mm 0.76) 0.65) 0.093) 0.98)
UCA >105o 23% 98% 48% 97% 19 0.78
and CL≤ (0.13 – (0.98 – (0.27 – (0.96 –
M

25mm 0.33) 0.99) 0.68) 0.97)

(95% Confidence Interval)
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT
22

Legend

Figure 1 Transvaginal ultrasound measurement of the uterocervical angle (UCA) technique

Measure the triangular segment between the lower uterine segment and the cervical canal

A. 18 5/7 weeks UCA 75o B. 16 5/7 weeks UCA 118o (CL) cervical length

PT
RI
Figure 2 Receiver Operating Characteristic Curves

Evaluating uterocervical angle (UCA) and gestational age at delivery A. UCA for sPTB <37 weeks

SC
B. UCA for sPTB <34 weeks

U
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC