Transabdominal cerclage: Can we predict who fails?

ANDREA L. FICK1, AARON B. CAUGHEY2, & JULIAN T. PARER2

1Department of Obstetrics and Gynecology, University of Iowa, Iowa City, IA, USA and 2Division of Maternal-Fetal

Medicine, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, CA, USA

(Received 17 May 2006; revised 3 September 2006; accepted 14 September 2006)

Abstract

Objective. To examine the outcome of pregnancies in women with transabdominal cerclage (TAC) and to determinewhether aspects of the obstetric history predict failure.

Methods. This was a cohort study of pregnant women referred for a transabdominal cerclage between 1978 and 2004. Recordswere reviewed for obstetric history and maternal demographics. Predictor variables were prior pregnancy loss, prior vaginalcerclage, associated factors for TAC, and maternal age. The outcome variable was delivery of an infant beyond 24 weeks whosurvived the neonatal period. Outcomes were compared using Student’s t-test, standard z-test, and Chi-square test.

Results. Eighty-eight women delivered 96 pregnancies after TAC placement. The fetal salvage rate prior to TAC was 18%,93% after the procedure (p5 0.001). Delivery beyond 37 weeks occurred in 70% of pregnancies. Maternal age, priorcerclage history, associated factors for TAC, or previous delivery of a viable infant did not predict the eight failures out of the96 pregnancies.

Conclusion. Women with TAC had a higher rate of successful pregnancies than prior to TAC. Neither maternal age norprior pregnancy loss predicted failure. However with such a high success rate, we would have needed 948 women to do so.TAC is an option for women with a poor obstetric history including failed vaginal cerclage.

Keywords: Cerclage, cervical insufficiency

Introduction

Cervical cerclage for cervical insufficiency remains a

subject of controversy and uncertainty. The incidence

of cervical insufficiency described by various authors

is 0.5–1.0% [1,2]. Randomized clinical trials have

revealed variable findings [3–5], although the largest

trial demonstrates improved outcomes for women at

moderate risk for preterm delivery, and particularly

for women with three or more mid-trimester losses

or preterm deliveries. In this subgroup, the preterm

delivery rate was 15% in women with cerclage, and

32% in those randomized to no cerclage [3].

A recent American College of Obstetricians and

Gynecologists Practice Bulletin [6] states that one

may consider offering prophylactic cerclage to

women with three or more unexplained mid-trime-

ster losses or preterm deliveries. It also states that

another option is to obtain serial ultrasounds of

cervical length in women with risk factors, beginning

at 16–20 weeks, and that emergency cerclage may be

offered to such women after counseling them about

the paucity of data to support such treatment and

poor outcomes after urgent/emergent cerclage. This

highly conservative approach has clearly not been

obstetrical practice, and many women have had

cerclages placed for lesser indications than these,

based on the number of procedures performed and

reported in the literature [7–14].

Management of cervical insufficiency becomes

more uncertain when a patient has had one or more

failed pregnancies after vaginal cerclage, or when

vaginal cerclage is not feasible due to anatomic

defects of the cervix. There are no prospective

studies to guide management in these women. There

has been one retrospective cohort study of abdominal

versus vaginal cerclage [15]. The authors compared

women who had had more than one prior failed

vaginal cerclage who received either repeat vaginal

cerclage or transabdominal cerclage (TAC) in a

subsequent pregnancy. The abdominal cerclage

group had a statistically significant decreased rate

of premature rupture of membranes and was

delivered at a later gestational age (36.3 vs. 32.8

weeks) than the vaginal cerclage group. The authors

concluded that TAC is superior in women with failed

Correspondence: Andrea L. Fick, MD, University of Iowa Hospitals and Clinics, Department of Obstetrics and Gynecology, 200 Hawkins Drive, Iowa City, IA

52242-1080, USA. Tel: þ1 319 353 6620. Fax: þ1 319 353 6759. E-mail: andrea-fick@uiowa.edu

The Journal of Maternal-Fetal and Neonatal Medicine, January 2007; 20(1): 63–67

ISSN 1476-7058 print/ISSN 1476-4954 online � 2007 Informa UK Ltd.

DOI: 10.1080/14767050601059156

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

prior vaginal cerclage and women should be offered

the procedure. In a systematic review of pregnancy

outcomes with TAC after a previous failed vaginal

cerclage, Zaveri and colleagues [16] concluded that

TAC conferred a lower risk of perinatal death or

delivery 524 weeks, though it may be associated

with a higher risk of operative complications.

With this background, we examined a group of

women with fetal wastage due to cervical insuffi-

ciency based on history and physical exam managed

with a transabdominal cerclage. A number of these

women had pregnancy failure despite the TAC and

the aim of the present analysis was to determine if

there are risk factors to explain such failures.

Methods

The women in this series were evaluated at the

University of California San Francisco and informa-

tion on prior fetal wastage and preterm delivery,

medical history, and physical exam were collected.

These women were offered a TAC due to cervical

insufficiency if they had a history consistent with at

least one previous second or third trimester preg-

nancy loss due to painless dilation of the cervix.

Women who did not have this history yet had

additional risk factors associated with cervical in-

sufficiency and were judged to be poor candidates for

vaginal cerclage, were also offered a TAC procedure.

These factors included anatomic defects of the

cervix, scarred cervix, diethylstilbestrol (DES) ex-

posure, and prior cervical conization [17–20].

We designed a retrospective cohort study to

examine predictors of TAC failure. Medical records

were reviewed from women who had undergone

TAC placement between 1978 and 2004 at the

University of California San Francisco. The Com-

mittee for Human Research at the University of

California San Francisco approved this study. The

first 23 women in the series had been examined

previously for validity of the indications for TAC

[21,22]. Gravidity, parity, prior vaginal cerclage,

prior pregnancy losses, and number of living children

were recorded for each patient. The women in our

cohort did not routinely have an ultrasound for

cervical length in the pre-operative evaluation for

TAC. Operative and hospital records were examined

for gestational age, complications, and estimated

blood loss. Pregnancy outcome information included

gestational age at delivery, birth weight, delivery

indication, and evidence of preterm labor or pre-

mature rupture of membranes. TAC failure was

defined as delivery of a nonviable infant or neonatal

death. We considered any infant delivered prior to 24

weeks gestation to be nonviable. The index preg-

nancy was the pregnancy in which the TAC was

initially placed. The fetal salvage rate was defined as

the number of viable infants who delivered beyond

24 weeks and survived the neonatal period in the

index pregnancy divided by the number of total

pregnancies that progressed beyond the first trime-

ster. The first trimester pregnancy losses (�14

weeks) were not included since the pathophysiology

is presumed to be different from second and third

trimester losses. All data were entered into Excel

(Microsoft, Redmond, WA, USA) and outcomes

were compared using Student’s t-test, standard z-

test, and Chi-square test.

All procedures were performed by one surgeon

(JTP), who was assisted by maternal–fetal staff,

fellows, and residents. Each procedure was per-

formed under general, spinal or epidural anesthesia.

A vertical midline skin incision was utilized in the

majority of women. The uterus was gently elevated

from the abdomen. The bladder peritoneum was

incised followed by blunt and sharp dissection to

expose the lower uterine segment. The outline of the

cervix was palpated. Using blunt dissection, tunnels

were created bilaterally through the paracervical

tissue, medial to the uterine blood supply and

slightly below the level of the internal cervical os. A

5-mm polyester fiber suture (Mersilene, Ethicon,

Somerville, NJ, USA) was passed through the

tunnels from posterior to anterior. The band was

tied anteriorly and the ends secured with nonabsorb-

able suture. Hemostasis was assured and the uterus

was returned to the abdomen. The fascia and skin

layers were closed. Prophylactic antibiotics were not

routinely used. Prophylactic tocolytics were not

routinely used prior to, during, or after the proce-

dure. Indomethacin was used if the patient had

severe cramping after the procedure. All women

were delivered by cesarean section and the TAC was

left in place unless the patient requested removal, a

bilateral tubal ligation was planned, or when the

band had eroded into the uterus or vagina.

Results

Our cohort included 96 pregnancies in 88 women

who had undergone the TAC procedure. Mean

maternal age at the time of procedure was 33.2 years.

The women had a mean gravidity of 4.3 with a range

of 1–11 pregnancies. All TACs were placed between

11 and 17 weeks with a mean gestational age of 13.4

weeks. The fetal salvage rate prior to TAC placement

was 18% (Table I). After TAC placement the

fetal salvage rate in the same women was 93%

(p5 0.001). Eight women had a TAC placed during

the index pregnancy, delivered their infants and

subsequently became pregnant again and delivered

viable infants.

Forty-two women had a history consistent with

cervical insufficiency and no further associated

64 A. L. Fick et al.

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

factors. The remaining 46 women had additional

factors associated with cervical insufficiency. We

compared gestational age at delivery and delivery of a

viable infant in these women in relation to the

associated factors for TAC (Table II). Prior cervical

conization was the most common associated factor

(n¼ 27) and scarred cervix was the least common

(n¼ 2). Eleven women had an anatomic defect of the

cervix and 13 women had a history of in utero DES

exposure. Using the Chi-square test for multiple

proportions, there was no difference between the

rates of delivery 537 weeks or delivery of a viable

infant between the associated factors in the women

with TAC.

Maternal age was examined as a possible predictor

of outcome in all pregnancies after TAC (Table II).

There were 44 pregnancies delivered by women who

were equal to or greater than 35 years of age at the

time of TAC placement. Women who were less than

35 years of age delivered 52 pregnancies. When

comparing number of viable infants and delivery

after 37 weeks, there was no statistical difference

between groups.

History of a prior vaginal cerclage was studied as a

predictor of TAC success in the index pregnancy

(n¼ 88). Forty-three women had one prior vaginal

cerclage, 15 women had two prior cerclages, three

women had three prior cerclages, and 26 women had

no history of previous vaginal cerclage. Women

without a prior vaginal cerclage did not have a

statistically different proportion of viable births (25

(96%)) in the index pregnancy after TAC placement

as compared to women with previous vaginal

cerclages (one prior cerclage, 42 (98%); two prior

cerclages, 13 (87%); three prior cerclages, 3 (75%);

p4 0.05). If the women were divided into two

groups, no prior/one prior cerclage vs. two or more

vaginal cerclages, there was no statistical difference

in delivery of viable infants or delivery 537 weeks;

23% vs. 32%, p4 0.05. The Chi-square test for

multiple proportions was utilized to evaluate each

group and outcome separately.

When obstetric history was considered, 49 women

had at least one living child prior to TAC placement

and 56 women had no living children. Two women

in each group delivered nonviable infants after

(TAC) placement.

Nine pregnancies were delivered before 31 weeks

due to preterm premature rupture of membranes

(PPROM), preterm labor (PTL), fetal demise,

uterine rupture and/or chorioamnionitis. All were

delivered via hysterotomy except in the case of

uterine rupture. Three infants delivered due to

PTL or PPROM at 24, 26 and 27 weeks, survived

the neonatal period. One woman who was delivered

due to PPROM and PTL at 24 weeks with a neonatal

death due to extreme prematurity has subsequently

conceived two pregnancies and delivered these

infants beyond 37 weeks. In another patient not

included in this group, the fetal membranes pro-

lapsed during the TAC procedure. An attempt was

made to reduce the membranes but rupture occurred

and the fetus and placenta delivered. The TAC was

placed and after a period of recovery the patient

became pregnant again and delivered a viable infant

at term.

Hemorrhage was the most common complication

at the time of TAC placement. The median

estimated blood loss was 100 mL with a range of

50 to 1300 mL. Seven women had a blood loss

greater than 400 mL. Four women required blood

transfusion. No women since 1988 have been

transfused. Blood loss in these women was due to

uterine vessel puncture, bleeding in the suture

tunnel, or fetal membrane prolapse/rupture during

the procedure. An additional patient experienced

pancytopenia and required four units of packed red

blood cells.

Table I. Fetal salvage rates before and after transabdominal

cerclage (TAC).

Before TAC After TAC

Pregnancies 340 96

1st trimester losses 100 0

2nd and 3rd trimester losses 196 8

Living children 44 88

Fetal salvage rate* 44 (18%) 88 (93%)

*Chi-square, p5 0.001.

Table II. Factors associated with outcome in women with a transabdominal cerclage.

Associated factor Viable % (n) Nonviable % (n) Delivered 437 weeks % (n) Delivered 537 weeks % (n) Total

None 38 (90%) 4 (10%) 34 (81%) 4 (10%) 42

Anatomic defect 11 (92%) 1 (8%) 9 (75%) 3 (25%) 12

Scarred cervix 2 (100%) 0 1 (50%) 1 (50%) 2

DES 13 (100%) 0 8 (62%) 5 (38%) 13

Prior cone or multiple LEEPs 26 (96%) 1 (4%) 24 (89%) 3 (11%) 27

Maternal age �35 42 (95%) 2 (5%) 33 (75%) 11 (25%) 44

Maternal age 535 48 (92%) 4 (8%) 35 (67%) 17 (33%) 52

*Chi-square, p4 0.05. DES, diethylstilbestrol; LEEPs, loop electrosurgical excision procedures.

Transabdominal cerclage 65

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

In two women the polyester fiber suture was found

to have eroded into the lower uterine segment. The

suture of the first woman was visualized vaginally at

22 weeks during speculum exam after PPROM. The

fetus was delivered via hysterotomy and the cerclage

was removed. The second woman had PPROM at 21

weeks and was delivered at 24 weeks due to

chorioamnionitis. The suture was noted to have

eroded through the anterior wall of the uterus at the

time of hysterotomy and was removed. A new suture

was placed and the woman subsequently had two

near-term living children. In both women the fetuses

died shortly after birth of extreme prematurity.

All women were delivered by cesarean section. Of

the total pregnancies, 70% were delivered after 37

weeks. The mean gestational age at delivery was

36+ 4 weeks. Survival analysis revealed the median

gestational age at delivery to be 37 weeks (Figure 1).

When categorized by previous cerclage history, the

mean gestational age at delivery was 36 weeks for all

groups except for women with three previous

cerclages, who delivered at a mean gestational age

of 37 weeks. There were three twin gestations and

delivery occurred at 33, 35 and 38 weeks, respec-

tively.

Discussion

We examined pregnancy outcomes in 88 women and

96 pregnancies after placement of a transabdominal

cerclage and calculated a fetal salvage rate of 93%.

This is in contrast to the fetal salvage rate prior to

TAC of 18%. These results are highly suggestive that

TAC at least partially contributes to the improved

fetal salvage rate. Given the disastrous fetal salvage

rate at baseline among these women, one might

expect there to be some aspect of the obstetric history

that would influence the lower fetal salvage rate

among subgroups of these women. However, the

comparisons of associated risk factors were neither

clinically nor statistically significant. Moreover,

neither maternal age, previous cerclage history, nor

previous living children were associated with the

subsequent delivery of a viable infant, or a preterm

delivery, after TAC.

Our data are similar to those published by

Lotgering et al. [23]. The authors examined a cohort

of 101 women with a history of at least two second or

early-third trimester pregnancy losses who under-

went a TAC procedure, whereas we also included

women with abnormalities of the cervix associated

with cervical insufficiency. The neonatal survival

rates before and after TAC placement, 27.5% and

93.5%, respectively, were similar to our study. In

contrast to our cohort, fewer women, 52 (51%), had

a history of a previous failed vaginal cerclage and

eight women had a TAC placed emergently after a

failed vaginal cerclage in the same pregnancy. It is

difficult to determine the significance of these cohort

differences as Lotgering et al. [23] did not examine

the effect of prior vaginal cerclage on neonatal

survival and both cohorts are underpowered to

detect differences in pregnancy failure before and

after TAC.

While we have reported on a large case series of

TAC, our negative findings could be attributed to an

insufficient number of women for statistical power.

For example, to detect a two-fold difference in failed

pregnancy between a control group and TAC group,

given our fetal salvage rate of 93%, we would have

needed a sample size of 474 women in each group to

achieve 80% power. Considering that this study

represents over 25 years of experience at our

institution, we would need a 25-site multicenter trial

to run for 10 years, which is neither clinically nor

economically feasible. Even if this trial were possible,

we would argue that women with a poor pregnancy

history, failed vaginal cerclage, or who are not

candidates for vaginal cerclage may not agree to be

randomized. In addition, it may be argued that it is

unethical to expectantly manage this group of

women who appear to have enormous gains from

this intervention.

Cervical length as measured by ultrasound is now

used to evaluate patient risk factors for preterm labor

[24–26] but has not been examined in the context of

transabdominal cerclage. A recent meta-analysis of

prophylactic vaginal cerclage for a shortened cervix

on ultrasound revealed the cerclage does not

decrease the incidence of preterm birth and perinatal

mortality [26]. The early women in our cohort did

not routinely have an ultrasound for cervical length

in the pre-operative evaluation for TAC as the

diagnostic technique was not available. Although

we routinely measure cervical length now, the

predominant indication for TAC remains a history

of cervical insufficiency and either a failed previous

vaginal cerclage or a cervix so anatomically abnormal

that an effective vaginal cerclage is deemed impos-

sible.Figure 1. Survival analysis for pregnancies with a transabdominal

cerclage.

66 A. L. Fick et al.

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.

We feel that TAC is a reasonable option in women

with a history consistent with cervical insufficiency

and a history of failed vaginal cerclage or who are not

candidates for vaginal cerclage. In our group of 88

women the fetal salvage rate was higher after TAC.

Furthermore, since none of the variables we exam-

ined predicted TAC failure, we do not find that any

of the factors we examined should limit TAC use in

women with any of the findings we describe,

including indication for prior fetal loss, anatomic

defects, number of prior losses or maternal age. With

an experienced operator and strict patient selection

the risks of the procedure are minimized. The

transabdominal cerclage procedure can produce a

high rate of pregnancy success, results in few

pregnancy failures, and can be offered to women

who meet the indication criteria.

Acknowledgements

Dr Caughey is supported by the National Institute of

Child Health and Human Development, Grant #

HD01262 as a Women’s Reproductive Health

Research Scholar.

References

1. Ludmir J. Sonographic detection of cervical incompetence.

Clin Obstet Gynecol 1988;31:101–109.

2. Lidegaard O. Cervical incompetence and cerclage in Den-

mark 1980–1990: A register based epidemiological survey.

Acta Obstet Gynecol Scand 1994;73:35–38.

3. MRC/RCOG Working Party on Cervical Cerclage. Final

report of the Medical Research Council/Royal College of

Obstetricians and Gynaecologists multicentre randomised

trial of cervical cerclage. Br J Obstet Gynaecol 1993;100:

516–523.

4. Rust OA, Atlas RO, Jones KJ, Benham BN, Balducci J. A

randomized trial of cerclage versus no cerclage among patients

with ultrasonographically detected second-trimester preterm

dilatation of the internal os. Am J Obstet Gynecol 2000;

183:830–835.

5. Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van

Geijn HP. Final results of the Cervical Incompetence

Prevention Randomized Cerclage Trial (CIPRACT): Ther-

apeutic cerclage with bed rest versus bed rest alone. Am J

Obstet Gynecol 2001;185:1106–1112.

6. American College of Obstetricians and Gynecologists. Prac-

tice Bulletin No. 48. Cervical insufficiency. ACOG; Novem-

ber 2003.

7. Surico N, Ribaldone R, Arnulfo A, Baj G. Uterine malforma-

tions and pregnancy losses: Is cervical cerclage effective? Clin

Exp Obstet Gynecol 2000;27:147–149.

8. Kurup M, Goldkrand JW. Cervical incompetence: Elective,

emergent or urgent cerclage. Am J Obstet Gynecol 1999;181:

240–246.

9. Harger JH. Comparison of success and morbidity in cervical

cerclage procedures. Obstet Gynecol 1980;56:543–548.

10. Cardwell MS. Cervical cerclage: A 10 year review in a large

hospital. South Med J 1988;81:15–19.

11. Whitehead KD, Wise RB, Dunnihoo DR, Otterson WN.

Retrospective analysis of cervical cerclage procedures at the

Louisiana State University, Shreveport (1980 to 1987). South

Med J 1990;83:159–160.

12. Angel JL, Petrantoni M, O’Brien WF, Knuppel RA. Cervical

cerclage in early pregnancy. J Perinatol 1991;11:258–261.

13. Treadwell MC, Bronsteen RA, Bottoms SF. Prognostic

factors and complication rates for cervical cerclage: A review

of 482 cases. Am J Obstet Gynecol 1991;165:555–558.

14. Guzman ER, Forster JK, Vintzileos AM, Ananth CV, Walters

C, Gipson K. Pregnancy outcomes in women treated with

elective vs. ultrasound-indicated cervical cerclage. Ultrasound

Obstet Gynecol 1998;12:323–327.

15. 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;183:836–839.

16. Zaveri V, Aghajafari F, Amankwah K, Hannah M. Abdominal

versus vaginal cerclage after a failed transvaginal cerclage: A

systematic review. Am J Obstet Gynecol 2002;187:868–872.

17. Moinian M, Andersch B. Does cervix conization increase the

risk of complications in subsequent pregnancies? Acta Obstet

Gynecol Scand 1982;61:101–103.

18. Kaufman RH, Adam E, Hatch EE, Noller K, Herbst AL,

Palmer JR, Hoover RN. Continued follow-up of pregnancy

outcomes in diethylstilbestrol-exposed offspring. Obstet Gy-

necol 2000;96:483–489.

19. Grunberger W, Riss P. Cervical incompetence after previous

cervical dilatation and curettage (author’s transl). Wien Med

Wochenschr 1979;129:390–392.

20. Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A,

Das A, Thom E, McNellis D, Copper RL, Johnson F, et al.

The length of the cervix and the risk of spontaneous

premature delivery. N Engl J Med 1996;334:567–572.

21. Cammarano C, Herron M, Parer JT. Validity of indications

for transabdominal cervicoisthmic cerclage for cervical in-

competence. Am J Obstet Gynecol 1995;172:1871–1875.

22. Herron M, Parer JT. Transabdominal cerclage for fetal

wastage due to cervical incompetence. Obstet Gynecol

1988;71:865–868.

23. Lotgering FK, Gaugler-Senden IP, Lotgering SF, Wallenburg

HC. Outcome after transabdominal cervicoisthmic cerclage.

Obstet Gynecol 2006;107:779–784.

24. Iams JD, Johnson FJ, Sonek J, Sachs L, Gebauer C, Samuels

P. Cervical competence as a continuum: A study of

ultrasonographic cervical length and obstetric performance.

Am J Obstet Gynecol 1995;172:1097–1106.

25. Iams JD. Prediction and early detection of preterm labor.

Obstet Gynecol 2003;101:402–412.

26. Berghella V, Odibo A. Cerclage for short cervix on ultra-

sound: A meta-analysis of the randomized trials. Am J Obstet

Gynecol 2004;191:S106.

Transabdominal cerclage 67

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y T

ufts

Uni

vers

ity o

n 10

/28/

14Fo

r pe

rson

al u

se o

nly.