Residents Papers

www. AJOG.org

Is cerclage height associated with the incidence of preterm

birth in women with an ultrasound-indicated cerclage?
Stacey Scheib, MD; John F. Visintine, MD; Gennady Miroshnichenko, MD;
Christopher Harvey, MD; Keith Rychlak, RD, MS; Vincenzo Berghella, MD
OBJECTIVE: Our aim was to determine whether there was a cerclage

RESULTS: There were 20 women in group 1 ( 18 mm), 25 in group 2

height threshold associated with spontaneous preterm birth in patients

with an ultrasound-indicated cerclage.

(13-17 mm), and 25 in group 3 ( 18 mm). Women with cerclage height

18 mm or greater had a lower incidence of spontaneous preterm birth less
than 35 weeks (4%) when compared with those with a cerclage height less
than 18 mm (33%) (relative risk, 0.69; 95% confidence interval, 0.55-0.86).

STUDY DESIGN: We performed a retrospective cohort study of women

with an ultrasound-indicated cerclage. Functional cervical length and

the cerclage height (distance from cerclage to the external cervical os)
were obtained. Our cohort was grouped into thirds, based on cerclage
height percentile. Our primary outcome was spontaneous preterm birth
less than 35 weeks.

CONCLUSION: Cerclage height of 18 mm or greater is associated with

a reduction in spontaneous preterm birth for women with an ultrasound-indicated cerclage.

Key words: cerclage, cervical length, preterm birth

he shortened cervix as determined by

transvaginal ultrasound is a wellestablished predictor of preterm birth.1
Cervical cerclage is 1 intervention that is
used for women found to have a shortened
cervix in the midtrimester, referred to as
the ultrasound-indicated cerclage. The efficacy of ultrasound-indicated cerclage has
been evaluated against no cerclage in randomized controlled trials, but the results
have been mixed.2-5 From a recent metaanalysis of patient level data, it was reported that cervical cerclage reduced the
risk of preterm birth for a select population, those with a singleton pregnancy, history of preterm birth, and a shortened cervix in the midtrimester.6

In addition to the patient population,

other factors such as cerclage position
may contribute to the variability observed in cerclage efficacy. With transvaginal ultrasound the distance from
cerclage to the external os, termed cerclage height, can be reliably visualized
and measured.7-10 Prior studies have
demonstrated variability in cerclage
height between patients,9,10 which may
affect cerclage efficacy. We hypothesized
that the greater the cerclage height placement, the more likely the cerclage would
be efficacious. Our objective was to determine, in patients with an ultrasoundindicated cerclage, whether there was a
cerclage height threshold associated with
spontaneous preterm birth.

From the Division or Maternal-Fetal

Medicine, Department of Obstetrics and
Gynecology, Thomas Jefferson University,
Philadelphia, PA (Drs Scheib, Visintine,
Harvey, Rychlak, and Berghella), and
Department of Obstetrics and Gynecology,
Lankenau Hospital, Wynnewood, PA (Dr
Miroshnichenko).

M ATERIALS AND M ETHODS

Presented at the 28th Annual Meeting of the

Society for Maternal-Fetal Medicine, Dallas TX,
Feb. 2, 2009.
Received June 23, 2008; revised Sept. 10,
2008; accepted Sept. 17, 2008.
Reprints not available from the authors.
0002-9378/free
2009 Mosby, Inc. All rights reserved.
doi: 10.1016/j.ajog.2008.09.021

e12

We performed a retrospective cohort

study of women who received an ultrasound-indicated cerclage. We defined
ultrasound-indicated cerclage as a cerclage placed for a short cervical length
( 25 mm), detected by transvaginal ultrasound between 14 and 236/7 weeks in
asymptomatic women. Our data sources
were the Thomas Jefferson University
Prematurity Database, which included
women from 1995 to 2007, and the Main
Line Hospital cerclage database, which
included women from 2003 to 2007.
History-indicated cerclage, transabdominal cerclage, multiple gestations,

American Journal of Obstetrics & Gynecology MAY 2009

indicated preterm birth, and pregnancies affected by a major fetal anomaly

were not included in the analysis to keep
our study population uniform.
Ultrasound measurements prior to
and after cerclage placement (if 24
weeks) were used for data analysis. All
transvaginal ultrasound measurements
were performed by experienced sonographers using standard technique. The
patients bladder was emptied prior to
visualization of the cervix. Only the minimum pressure necessary was used to obtain a clear image of the cervical canal in
the midsagittal plane. The first ultrasound obtained after placement of the
cerclage was performed within 2 weeks.
The following cervical ultrasound
measurements were considered in the
analysis: functional cervical length
(closed portion of the endocervical canal) and the cerclage height (the distance
from cerclage to the external cervical os)
(Figure 1). At least 3 measurements each
were obtained for the functional cervical
length and cerclage height, and the
shortest measurement was used. The
managing obstetricians were not blinded
to the measurement results. All data
analysis was based on the shortest measurements for each of the obtained measurements between 14 and 236/7 weeks.
We analyzed our cohort by grouping
patients into thirds, based on cerclage
height percentile prior to cerclage. Our

Residents Papers

www.AJOG.org

FIGURE 1

Schematic representation
of cervical length
and cerclage height
ultrasound measurements

The opaque circles represent the cerclage.

Scheib. Cerclage height and the incidence of preterm birth in
ultrasound-indicated cerclage. Am J Obstet Gynecol 2009.

primary outcome was spontaneous preterm birth less than 35 weeks and weeks
gained after placement of cerclage. The
2 or Fishers exact tests were performed
for categorical variables. Analysis of variance was performed for continuous variables. Type I error was set at 0.05 (2
sided). Logistic regression was performed to assess for confounders. Variables that were associated with cerclage
height (P .2) were considered as potential confounders and included in the
multivariable logistic regression model.
All statistical analysis was performed us-

ing SPSS software (version 16; SPSS Inc,

Chicago IL).

R ESULTS
Seventy women with ultrasound-indicated cerclage, singleton gestation, and
complete ultrasound measurements
were identified. There were 20 women in
group 1 ( 13 mm), 25 in group 2 (13-17
mm), and 25 in group 3 ( 18 mm). The
baseline characteristics were similar between groups, with the exception of
prior preterm birth and cervical length at
placement (Table). Sixty-three of the 70
cerclages placed were McDonald cerclages. It was unclear the cerclage type of
the remaining 7 cerclages. Most of the
cerclages were placed by residents as the
primary surgeon under the supervision
of a maternal fetal medicine attending.
The majority of the cerclages used Mersilene tape.
The incidence of spontaneous preterm delivery was 30% (6/20) in group
1, 36% (9/25) in group 2, and 4% (1/
25) in group 3 (Figure 2). A significant
difference was found in the incidence
of spontaneous preterm birth according to cerclage height groups (P
.018). There was no difference in the
incidence of spontaneous preterm delivery between groups 1 and 2 (P
.671), although there was a difference
in the incidence of preterm delivery between both groups 1 and 3 and 2 and 3

(P .034 and P .011, respectively).

Because there was no significant difference between groups 1 and 2, these 2
groups were then combined and compared with group 3 for further analysis
(P .0035). Women with cerclage
height of 18 mm or greater had a lower
incidence of spontaneous preterm
birth less than 35 weeks, 1 of 25 (4%),
when compared with those with a cerclage height less than 18mm 15 of 45
(33%) (relative risk, 0.69; 95% confidence interval, 0.55-0.86). Post hoc
analysis found the observed power to
be 1.
An analysis of outcomes according to
weeks gained after cerclage placement
was looked at to control for the gestational age at cerclage placement. Women
with a cerclage height of 18 mm or
greater had significantly greater amount
of time gained (17.2 weeks) when compared with those with a cerclage height of
less than 18 mm (15.2 weeks; P .047).
The potential confounders, prior preterm birth, gestational age at cerclage,
and functional cervical length prior to
cerclage, were included in the logistic regression analysis along with cerclage
height of 18 mm or greater. In our model
the associated reduction in preterm birth
less than 35 weeks remained only for
those with a cerclage height of 18 mm or
greater (adjusted odds ratio, 0.10; 95%
confidence interval, 0.01-0.94).

TABLE

Baseline characteristics by cerclage height group

Total

Group 1
(< 13mm)

Group 2
(13-17 mm )

Group 3
(> 18 mm)

70

20

25

25

Age (SD)

27 (5)

27 (4)

28 (6)

28 (4)

P value

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

.869

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

Parity (SD)

1.1 (1.2)

1.0 (1.0)

0.9 (1.1)

1.4 (1.5)

.417

................................................................................................................................................................................................................................................................................................................................................................................
c

African American race (%)

39 (58)

13 (65)

13 (52)

15 (60)

.668

Prior preterm birth 35 weeks (%)

19 (27)

9 (45)

7 (28)

3 (12)

.047

Prior cervical conization (%)

12 (18)

3 (16)

6 (25)

3 (12)

.488

................................................................................................................................................................................................................................................................................................................................................................................
d
................................................................................................................................................................................................................................................................................................................................................................................
d
................................................................................................................................................................................................................................................................................................................................................................................
d

Mullerian anomaly (%)

1 (4)

2 (8)

.455

Gestational age at cerclage in weeks (SD)

19 (1.9)

3 (4)

18 (2)

19 (1)

19 (1)

.120

Functional cervical length prior to cerclage

(mm) (SD)a

20 (5)

17 (7)

19 (4)

23 (3)

.002

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

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

Mean; b Analysis of variance; c 2 test; d Fishers exact test.

Scheib. Cerclage height and the incidence of preterm birth in ultrasound-indicated cerclage. Am J Obstet Gynecol 2009.

MAY 2009 American Journal of Obstetrics & Gynecology

e13

Residents Papers
FIGURE 2

Incidence of preterm birth

< 35 weeks according
to cerclage height
40%

Incidence of Preterm Birth

35%
30%
25%
20%
15%
10%
5%
0%
<13mm

13-17mm

18mm

Cerclage Height Groups

Bar graph representation of the incidence of

preterm birth 35 weeks according to cerclage
height group.
Scheib. Cerclage height and the incidence of preterm birth in
ultrasound-indicated cerclage. Am J Obstet Gynecol 2009.

C OMMENT
Placing an ultrasound-indicated cerclage
at a cerclage height (distance from the
external os) of 18 mm or greater was associated with a lower incidence of spontaneous preterm birth compared with
placing the cerclage closer to the external
os. Importantly, this held true even when
we controlled for functional cervical
length, which, if short, could make it
more difficult to place the cerclage close
to the internal os.
Whereas several randomized studies
have assessed the efficacy of ultrasoundindicated cerclage,2-5 there is paucity of
data on technical aspects of this procedure, which is something over which the
surgeon has some control. Few studies
have evaluated whether a successful cerclage depends on suture placement as
close as possible to the internal cervical
os, which can be represented by cerclage
height. As originally described in 1957 by
McDonald,11 the cerclage was placed as
high as possible to approximate to the
level of the internal os. After more than
50 years, our findings support this practice in women with ultrasound-indicated cerclage.
Similar to our study, Rust et al10 investigated ultrasound-indicated cerclage
position and the incidence of preterm
birth in 74 women. Their linear regression analysis demonstrated wide variability but no association between cerclage height and preterm birth.10
e14

www.AJOG.org
Because there did not appear to be a linear association between cerclage height
and preterm birth based on data from
the Rust article, we postulated that there
may be a cutoff value for cerclage height
that may be associated with preterm
birth. This was supported by our finding
that women with a cerclage height of 18
mm or greater had a much lower incidence of preterm birth.
Cerclage height has been evaluated
also in women with other indications for
cerclage. In 29 women, physical examination indicated (also called emergency)
cerclage, cerclage height was found to
positively correlate with gestational age
at delivery, but the association did not
persist in the multivariable analysis.9
Our hypothesis of why cerclage height is
associated with lower rates of preterm
birth is that the cerclage is closer to the
internal os and thus provides reinforcement. When comparing outcomes between women with similar prior failed
transvaginal cerclage who had either a
transabdominal or a transvaginal cerclage, the incidence of preterm birth was
lower in women with a transabdominal
cerclage, which is placed directly at the
internal cervical os.12
Other ultrasound measures of cerclage position aside from cerclage
height have been studied as predictors
of preterm delivery. Guzman et al9
found that the distance from the cerclage to the internal os (upper cervical
length) correlated minimally (r
0.31) with gestation age at delivery, but
this association was not significant (P
.10). An upper cervical length of less
than 10 mm was, however, associated
with delivery at less than 36 weeks. This
cutoff value was predictive only if measured in the first 48 hours after cerclage
placement because the upper cervical
length was found to shorten to less
than 10 mm before 28 weeks in all patients.9 Funneling to the cerclage has
also been found to be predictive of preterm delivery, but like shortening of
upper cerclage height, funneling may
develop at any time on subsequent ultrasounds and would require serial
scans after cerclage placement.13
Conclusions from our study need to
be considered in light of the relatively

American Journal of Obstetrics & Gynecology MAY 2009

small sample size. Although equally as

large as any similar study reported in the
literature, the 95% confidence interval
for our primary outcome is wide (0.010.94) as a result of the sample size. We
elected to narrow our patient population
by excluding patients with a history-indicated cerclage or a physical examinationindicated cerclage to keep our
study population homogenous. Other
variables such as inflammation could
have affected our results.
Our study suggests that a cerclage
height of 18 mm or greater was
achieved in about a third of women
who have an ultrasound-indicated cerclage and that these patients appear
have a much lower risk of preterm delivery compared with patients with a
shorter cerclage height. We suggest attempting to place the cerclage as close
as possible to the internal os. In the
greater than 13 years since we began
recording outcomes with ultrasoundindicated cerclage, with most placed by
residents as primary surgeons with supervision of a maternal fetal-medicine
attending physician, no bladder injuries or fistulas occurred. Transvaginal
ultrasound measurement of cerclage
height may provide the clinician with a
useful tool to evaluate the effectiveness
of the cerclage placement and provide
more accurate counseling regarding
the risk of preterm delivery following
an ultrasound-indicated cerclage. f
REFERENCES
1. 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. N Engl J Med
1996;334:567-72.
2. Berghella V, Odibo AO, Tolosa JE. Cerclage
for prevention of preterm birth in women with a
short cervix found on transvaginal ultrasound
examination: a randomized trial. Am J Obstet
Gynecol 2004;191:1311-7.
3. Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van Geijn HP. Final results of the Cervical Incompetence Prevention Randomized
Cerclage Trial (CIPRACT): therapeutic cerclage
with bed rest versus bed rest alone. Am J Obstet Gynecol 2001;185:1106-12.
4. To MS, Alfirevic Z, Heath VC, et al. Cervical
cerclage for prevention of preterm delivery in
women with short cervix: randomised controlled trial. Lancet 2004;363:1849-53.

Residents Papers

www.AJOG.org
5. Rust OA, Atlas RO, Reed J, van Gaalen J,
Balducci J. Revisiting the short cervix detected
by transvaginal ultrasound in the second trimester: why cerclage therapy may not help. Am J
Obstet Gynecol 2001;185:1098-105.
6. 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 Gynecol
2005;106:181-9.
7. Quinn MJ. Vaginal ultrasound and cervical
cerclage: a prospective study. Ultrasound Obstet Gynecol 1992;2:410-6.

8. Andersen HF, Karimi A, Sakala EP, Kalugdan

R. Prediction of cervical cerclage outcome by
endovaginal ultrasonography. Am J Obstet Gynecol 1994;171:1102-6.
9. Guzman ER, Houlihan C, Vintzileos A, Ivan J,
Benito C, Kappy K. The significance of transvaginal ultrasonographic evaluation of the cervix
in women treated with emergency cerclage.
Am J Obstet Gynecol 1996;175:471-6.
10. Rust OA, Atlas RO, Meyn J, Wells M, Kimmel S. Does cerclage location influence perinatal outcome? Am J Obstet Gynecol 2003;
189:1688-91.

11. McDonald IA. Suture of the cervix for inevitable miscarriage. J Obstet Gynaecol Br Emp
1957;64:346-50.
12. Davis G, Berghella V, Talucci M, Wapner
RJ. Patients with a prior failed transvaginal
cerclage: a comparison of obstetric outcomes with either transabdominal or transvaginal cerclage. Am J Obstet Gynecol 2000;
183:836-9.
13. OBrien JM, Hill AL, Barton JR. Funneling to
the stitch: an informative ultrasonographic finding after cervical cerclage. Ultrasound Obstet
Gynecol 2002;20:252-5.

MAY 2009 American Journal of Obstetrics & Gynecology

e15