Volume 201, Issue 2 , Pages 163.e1-163.e4, August 2009
Interval to spontaneous delivery after elective removal of cerclage
Article Outline
Objective
The purpose of this study was to estimate the time interval between elective cerclage removal and spontaneous delivery.
Methods
Singleton pregnancies with McDonald cerclage were evaluated for the interval between elective cerclage removal (36-37 weeks) and spontaneous delivery. We also compared spontaneous delivery within 48 hours after cerclage removal between women with ultrasound-indicated vs history-indicated cerclage.
Results
We identified 141 women with elective cerclage removal. The mean interval between removal and delivery was 14 days. Only 11% of women delivered within 48 hours. Women with ultrasound-indicated cerclage were more likely to deliver within 48 hours, compared with women with history-indicated cerclage (odds ratio, 5.14; 95% confidence interval, 1.10-24.05).
Conclusion
The mean interval between elective cerclage removal and spontaneous delivery is 14 days. Women with cerclage who achieved 36-37 weeks should be counseled that their chance of spontaneous delivery within 48 hours after elective cerclage removal is only 11%.
Key words: cerclage, interval to delivery, preterm birth
Approximately 1% of patients receive cerclage per year for the prevention of preterm birth. Most of these pregnancies are successful and go to term or near-term with cerclage still in place. When Dr Shirodkar1 originally described his technique, he removed the cerclage suture prophylactically before labor onset, which is still the most common practice in the United States. There is very limited information regarding the interval of time between the elective removal of a cerclage suture and spontaneous delivery. This information may be helpful in treatment and for counseling women who undergo elective cerclage removal, because many women have questions regarding this interval and some women are under the impression that they may deliver soon after the cerclage is removed.
Our objectives were to estimate the interval of time between elective cerclage removal and spontaneous delivery, to compare delivery within 48 hours after cerclage removal between women with an ultrasound-indicated cerclage and women with a history-indicated cerclage, and to estimate whether a longer interval between elective cerclage removal and delivery was associated with better outcome in the future pregnancy.
Materials and Methods
A chart review of women at Thomas Jefferson University was conducted between 1995 and 2007. The Institutional Review Board of Thomas Jefferson University approved this study. Pregnancies with a McDonald cerclage, either history-indicated or ultrasound-indicated, were evaluated for the interval between elective cerclage removal (usually 36-37 weeks) and spontaneous delivery. Women received a history-indicated cerclage usually at approximately 12-15 weeks on the basis of poor obstetric history (usually ≥ 2 previous second-trimester losses). Women received an ultrasound-indicated cerclage at approximately 16-23 weeks for ≥ 1 previous preterm births and transvaginal ultrasound cervical length of < 25 mm.2 These women were under ultrasound surveillance of cervical length because of their obstetric history of previous second-trimester losses, history of cold knife cone, Müllerian defect, diethylstilbestrol (DES) exposure and multiple dilation and evacuation. Exclusion criteria were preterm birth at < 35 weeks that necessitated cerclage removal, removal at time of labor, multiple gestation, removal at the time of cesarean delivery before onset of labor, fetal death, loss to follow-up evaluation, and voluntary termination. No woman who went to labor with cerclage in place was included in our analysis.
All cerclage procedures were performed with the McDonald technique. In general, Mersilene tape was used, with 4 or 5 “bites” around the cervix at least 1-2 cm from the external os, and tied anteriorly. All procedures were done with the use of spinal anesthesia. Antibiotics were not administered at the time of cerclage. The usual practice at Thomas Jefferson University is to remove the cerclage suture in asymptomatic women at approximately 36-37 weeks of gestation in an outpatient setting. The patient is then counseled for labor precautions and sent home with a follow-up visit scheduled in 1 week. Gestational age was confirmed by ultrasound examination < 24 weeks.
Primary outcome was the interval, in days, between elective cerclage removal and spontaneous delivery. Secondarily, we compared delivery within 48 hours after cerclage removal between women with ultrasound-indicated and a history-indicated cerclage. Additionally, we estimated whether a longer interval between elective cerclage removal and delivery was associated with better outcome in the future pregnancy. To accomplish this goal, we evaluated gestational age at delivery in the subsequent pregnancy comparing the interval between elective cerclage removal and spontaneous delivery of ≤ 7 days or > 7 days. Continuous data were compared with the use of the Student t test and Mann-Whitney U test. The Fischer exact test and Pearson χ2 test were used when to compare categoric data. A probability value of < .05 was considered statistically significant. We then took all univariable with univariate associations (P ≤ .20) and entered them into the multiple logistic regression model. The variables were removed in a stepwise fashion, with a probability value of .15 as criterion for removal. Statistical analysis was performed with SPSS 14.0 software (version 16; SPSS Inc, Chicago, IL).
Results
Between 1995 and 2007, 299 cerclages that were eligible for inclusion in this study were identified. They represented 1.1% of all births (299/27,779) at Thomas Jefferson University. One hundred fifty-eight women were excluded for preterm birth (n = 117), labor at time of removal (n = 15), multiple gestation (n = 13), removal at time of cesarean delivery (n = 7), fetal death (n = 3), loss to follow-up (n = 2), and voluntary termination (n = 1), which left 141 women for analysis. Seventy-four women had a history-indicated cerclage, and 67 women had an ultrasound-indicated cerclage. Demographic characteristics and outcome variables are presented in Table 1. The incidence of previous preterm birth (14-35 weeks) overall in this analysis was 73%: 60% of patients who received an ultrasound-indicated cerclage, and 85% of patients who received a history-indicated cerclage. The overall gestational age at elective removal was 36.7 ±1.2 weeks (SD), and gestational age at delivery was 38.4 ± 1.6 weeks. No women had rupture of membranes at elective removal of cerclage. The mean interval between elective cerclage removal and spontaneous delivery was approximately 14.2 ± 9.6 days . Only 11% of patients spontaneously delivered within 48 hours after elective removal of cerclage, and no women had precipitous deliveries outside the hospital after elective cerclage removal.
TABLE 1. Demographics and outcome variables
| Variable | Cerclage | P value | |
|---|---|---|---|
| Ultrasound-indicated (n = 67) | History-indicated (n = 74) | ||
| Age (y)a | 28.8 ±6.0 | 31.5±5.4 | .005b |
| African American race (%) | 70 | 50 | .030c |
| Previous preterm birth (%) | 60 | 85 | .002c |
| Preterm births (n)a | 0.7±0.9 | 1.0±0.8 | .009b |
| Spontaneous pregnancy loss at 14-24 wks (%) | 46 | 80 | .005c |
| Smoking (%) | 22 | 9 | .102c |
| History of cold knife cone (%) | 6 | 8 | .748c |
| History of Müllerian defect (%) | 3 | 1 | .604c |
| Diethylstilbestrol exposure (%) | 2 | 7 | .259c |
| Gestational age of cerclage removal (wk)a | 36.7±1.1 | 36.7±1.3 | .939b |
| Gestational age at delivery (wk)a | 38.3±1.7 | 38.5±1.6 | .946b |
| Delivery within 48 h (%) | 18 | 5 | .038c |
| Time from elective cerclage removal to delivery (d)d | 13.4±10.2 | 15.0±8.9 | .341e |
aData are given as mean |
bStudent t test; |
cχ2 or Fisher exact test; |
ddata are given as median ± SD; |
eMann-Whitney U test. |
We then compared women who had an ultrasound-indicated cerclage with those who had a history-indicated cerclage. No significant difference was detected between both groups with respect to smoking, history of cold knife cone procedure, Müllerian defect, DES exposure, smoking, gestational age of cerclage removal, and gestational age at delivery. The mean interval in days between elective cerclage removal and delivery was 13.4 ± 10.2 days in the ultrasound-indicated group and 15.0 ± 8.9 days in the history-indicated group, which was also not statistically significant (P = .34). Differences were detected between both groups with respect to age, African American race, history of preterm birth, number of preterm births, spontaneous abortions, and delivery within 48 hours after elective cerclage removal. After multivariable analysis, only ultrasound-indicated cerclage remained an independent predictor of delivery within 48 hours (Table 2). Women with an ultrasound-indicated cerclage were 5 times more likely to deliver within 48 hours after elective cerclage removal than were women with a history-indicated cerclage (odds ratio, 5.14; 95% confidence interval [CI], 1.10-24.05).
TABLE 2. Adjusted odd ratios for the variables in the final multivariate model for the prediction of delivery within 48 hours
| Variable | Adjusted odds ratio | 95% CI |
|---|---|---|
| Age | 1.01 | 0.95-1.23 |
| African American race | 3.49 | 0.67-18.18 |
| Previous preterm birth | 0.55 | 0.08-3.67 |
| Number of preterm births | 0.58 | 0.26-1.33 |
| Spontaneous abortion at 14-24 wk | 3.18 | 0.70-14.36 |
| Smoking | 1.87 | 0.33-10.68 |
| Ultrasound-indicated cerclage | 5.14 | 1.10-24.05 |
We identified 20 women with a pregnancy subsequent to the index pregnancy that is reported here. Two pregnancies were excluded because of twin gestations. The gestational age at subsequent delivery was significantly less for those women with a previous interval from elective cerclage removal at ≤ 7 days than for those with an interval of > 7 days (Table 3). This occurred despite the fact that 5 of the 12 women with a previous interval of > 7 days did not receive any cerclage at all in the subsequent pregnancy; all 6 women with an interval to delivery of ≤ 7 days received cerclage.
TABLE 3. Effect of interval to delivery after elective cerclage removal on subsequent pregnancy
| Interval to delivery after elective cerclage removal in index pregnancy | P value | ||
|---|---|---|---|
| ≤ 7 d (n = 6) | > 7 d (n = 12) | ||
| Gestational age at delivery in subsequent pregnancy (mean ± SD) | 36.7 ± 1.0 | 38.8 ± 1.6 | .008 |
Comment
This study aimed to estimate the interval of time between elective cerclage removal and spontaneous delivery. There is a clinical myth held by some women and some clinicians that, because the cerclage was performed for prevention of preterm birth, once removed, the woman may possibly deliver right away. However, in this study only 11% of patients went into spontaneous labor within 48 hours of removal. In fact, the average interval of time for elective cerclage suture removal and spontaneous delivery for both cerclage types was surprisingly 14 days. Although this average was similar for history- vs ultrasound-indicated cerclage, the incidence of delivery within 48 hours of elective cerclage removal was 18% for ultrasound-indicated vs 5% for history-indicated cerclage. These data can be useful in counseling.
There is a paucity of data that estimates interval to delivery after elective cerclage removal in women with either history-indicated or ultrasound-indicated cerclage. Shirodkar1 reported that “the majority of patients deliver in two to eight days after cutting the tape.” He did not report the data on what to base this statement. He,3 as well as McDonald,4 placed (physical examination-indicated) cerclage in women with both previous second-trimester losses and cervical changes that were detected by manual examination during the subsequent pregnancy. We know now that these pregnancies are at even higher risk of cervical insufficiency and preterm birth than women with similar histories, but who receive ultrasound-indicated cerclage, because of earlier changes on transvaginal ultrasound cervical length.5 It is plausible that interval to delivery after removal after physical examination-indicated cerclage would occur earlier than what occurs after history-indicated or ultrasound-indicated cerclage. Another reason for a shorter interval to delivery after elective cerclage removal in Shirodkar's1 experience is the fact that he removed the tape electively at 38 weeks, although we removed it earlier, at 36-37 weeks.
In a study of 13 women who received “emergency” (probably ultrasound-indicated) cerclage, the interval between elective removal and delivery was 2.3 weeks.6 This is similar to our data, which have a larger sample size. In addition, our data show that women with a short cervix in the second trimester (who get an ultrasound-indicated cerclage) may have a cervix that provides less resistance, with a slightly shorter (approximately 1.5 days; nonsignificant) interval to delivery after cerclage removal and a 5-fold higher chance of delivering within 48 hours after removal than women with history-indicated cerclage.
Most obstetricians suggest removing cerclage electively once a gestational age of approximately 36-38 weeks is achieved.1, 7, 8, 9 Some obstetricians have argued to allow women with cerclage in place to go into labor. The largest such series reported a 6% incidence of cervical laceration (5/82) with removal at time of labor.10 Other obstetricians have reported hemorrhage, infection, poor wound healing, and even uterine rupture after allowing term labor with cerclage in place.7 We did not observe any cervical lacerations or other such complications in our series; another series with elective removal at 38 weeks reported a laceration rate of 3%.9
There are no data to assess whether the type of cerclage, McDonald or Shirodkar, is associated with an effect on interval to delivery. Moreover, the type of suture may also have an effect, but it has not been reported.
Strengths of our study are the novelty of the report, because none of the previous reports have assessed primarily interval to delivery after elective history-indicated or ultrasound-indicated cerclage removal. We also have a fairly homogeneous management, with similar technique of cerclage. Moreover, we compared ultrasound-indicated and history-indicated cerclage. A weakness of our study is its retrospective nature. Data that would have been interesting to evaluate were not recorded; for example, data on Bishop score or cervical length after cerclage removal were not available. Data in our database is recorded prospectively and is checked multiple ways for accuracy with multiple other electronic databases.
The findings and data that are presented in the last 2 rows of Table 1 should be helpful for the treatment and counseling of patients who undergo elective cerclage suture removal. There is no need to perform cerclage removal in the hospital only because of fear of spontaneous labor immediately after this procedure.
References
- . Cervical incompetence and its treatment. In: Sturgis SH, Taylor ML editor. Progress in gynecology. New York: Grune and Stratton; 1970;p. 478–497
- Prediction of preterm delivery with transvaginal ultrasonography of the cervix in patients with high-risk pregnancies: does cerclage prevent prematurity?. Am J Obstet Gynecol. 1999;181:809–815
- . A new method of operative treatment for habitual abortions in the second trimester of pregnancy. Antiseptic. 1955;52:299–300
- . Suture of the cervix for inevitable miscarriage. J Obstet Gynecol. 1957;64:346–350
- . Cervical cerclage in the second trimester of pregnancy: a historical cohort study. Am J Obstet Gynecol. 2001;184:1447–1456
- . The significance of transvaginal ultrasonographic evaluation of the cervix in women treated with emergency cerclage. Am J Obstet Gynecol. 1996;175:471–476
- . The incompetent internal os of the cervix, complication after repair. Am J Obstet Gynecol. 1961;81:465–471
- . Shirodkar cerclage in a multifactorial approach to the patient with advanced cervical changes. Am J Obstet Gynecol. 1990;162:1412–1420
- . Cervical cerclage: a review of 74 cases. J Reprod Med. 1984;29:103–106
- . Management of cervical cerclage at term: remove the suture in labor?. J Perinat Med. 2000;28:453–457
Authorship and contribution to the article is limited to the 6 authors indicated. There was no outside funding or technical assistance with the production of this article.
Cite this article as: Bisulli M, Suhag A, Arvon R, et al. Interval to spontaneous delivery after elective removal of cerclage. Am J Obstet Gynecol 2009;201:163.e1-4.
Reprints not available from the authors.
PII: S0002-9378(09)00408-6
doi:10.1016/j.ajog.2009.04.022
© 2009 Mosby, Inc. All rights reserved.
Volume 201, Issue 2 , Pages 163.e1-163.e4, August 2009
