Objective
The purpose of this study was to evaluate the effectiveness of laparoscopic abdominal cerclage placement in the prevention of recurrent preterm birth.
Study Design
We conducted a multicenter cohort study with retrospective Dutch (32 patients) and Boston (34 patients) cohorts who had undergone preconceptional laparoscopic abdominal cerclage placement. Eligible patients had at least 1 second/third trimester fetal loss or delivered at <34 weeks of gestation because of cervical insufficiency and/or a short or absent cervix. Primary outcome was delivery of an infant at ≥34 weeks of gestation with neonatal survival. Secondary outcome measures included surgical and pregnancy outcomes and patients’ satisfaction (Dutch cohort).
Results
Surgical outcomes of 66 patients were excellent, with 3 minor complications. After preconceptional laparoscopic abdominal cerclage, 35 pregnancies were evaluated. Twenty-five patients (71.4%) delivered at ≥34 weeks of gestation; 3 patients (8.6%) experienced a second-trimester fetal loss. The total fetal survival rate was 90.0%.
Conclusion
Preconceptional laparoscopic abdominal cerclage shows encouraging and favorable perinatal outcomes in patients with a poor obstetric history.
Preterm birth (defined as the birth of an infant at <37 weeks of gestation) is an important determinant of perinatal morbidity and death. One predominant factor of preterm delivery is cervical insufficiency, which is estimated to complicate 0.1-1.0% of all pregnancies. Cervical insufficiency is characterized by acute, painless cervical dilation in the absence of uterine activity. Without specialized treatment, cervical insufficiency has a high likelihood of reoccurrence in subsequent pregnancies. The traditional treatment for cervical insufficiency is placement of a vaginal cerclage. In patients in whom a vaginal cerclage previously has failed or in whom a vaginal cerclage is technically not feasible because of an extremely short or absent cervix, an abdominally placed cerclage provides an alternative. The advances in minimally invasive surgery have led to the increasing use of the laparoscopic approach for abdominal cerclage placement, for which reports were first published in 1998.
The aim of this study was to assess the effectiveness of preconceptional laparoscopic abdominal cerclage (LAC), in terms of surgical and pregnancy outcome and patient satisfaction.
Materials and Methods
Study methods
Two separate cohort studies have been performed, both of which had a retrospective design. The Dutch cohort had consecutive inclusion of all patients who received an abdominal cerclage in the period of June 1997 to December 2011. After June 1997, all patients received LAC; before that date, laparotomy was performed. In the Boston cohort, all patients with LAC from May 2007 to December 2010 have been included. In this period, all patients in both the Dutch and the Boston cohort received a laparoscopic approach; LAC was not applied in the participating hospitals in this period. In both cohorts, patients were divided into 2 main indication groups for LAC: (1) previous failed vaginal cerclage (defined as a previous vaginal cerclage that resulted in a second- or third-trimester fetal loss, immature delivery, or premature delivery [delivery at <34 weeks of gestation]) and (2) previous cervical surgery (defined as recurrent loop electrosurgical excision procedure, (laser) conization of the cervix, or trachelectomy). A part of the latter group included patients with a previous immature or premature delivery.
Inclusion of patients and method of data collection
Eligible patients were women with a failed vaginal cerclage (defined as a previous vaginal cerclage that resulted in a second- or third-trimester fetal loss or immature or premature delivery [defined as delivery at <34 weeks of gestation]) and all patients in whom vaginal cerclage insertion was technically not possible because of an extremely short, scarred, or absent cervix because of recurrent loop electrosurgical excision procedure, (laser) conization of the cervix, or trachelectomy. Because some of these patients also had an immature or preterm delivery in their obstetric history, the judgment was made that the risk of an immature or premature delivery in a next pregnancy justified the placement of an abdominal cerclage ( Supplementary Table provides details of obstetric history in the Appendix ). All diagnoses of cervical insufficiency and the indication for placement of an LAC took place in the Dutch and the Boston clinics. All patients in whom LAC was placed during pregnancy were excluded. No patients had a previous successful vaginal cerclage (defined as delivery at ≥34 weeks of gestation). In the Dutch preconceptional cohort, all patients for whom LAC was planned were asked to participate consecutively; all of the patients agreed to participate ( Figure 1 ). Approval was obtained from the Medical Ethical Exam Committee (trial no. 39306). Data were collected with the use of Case Report Forms and the local perinatal information network starting in September 2009. Additionally, telephone interviews were executed with all patients in the Dutch preconceptional cohort from September 2009 to December 2011, starting from at least 6 months after surgery and repeated every 6 months. These interviews included questionnaires on patient satisfaction and physical complaints after the procedure, duration of fertility wish, and pregnancy outcomes ( Figure 2 ). In the Boston cohort, the records of all patients who had LAC inserted in a nonpregnant state in the Brigham and Women’s Hospital (Boston, MA) from May 2007 to December 2010 were reviewed retrospectively after the approval of the appropriate institutional review board had been obtained.
The follow-up period of patients varied from 2-168 months in the Dutch cohort and from 12-48 months in the Boston cohort.
Surgical technique
All procedures were performed with general anesthesia as an interval procedure (ie, placed in nonpregnant patients) in the Dutch cohort (H.A.M.B. or J.A.F.H) and in the Boston cohort (J.I.E) according to a standard surgical and perioperative protocol.
Dutch preconceptional cohort
The laparoscopic instruments (one 12-mm umbilical trocar and 2 additional 5-mm trocars in the left lower quadrant and to the left of the umbilical trocar, respectively) and a small simple uterine mobilizer were inserted. The peritoneal surface of the bladder was incised to identify the appropriate position of the suture placement. The avascular space between the ascending and descending branch of the uterine artery was dissected at both sides. Precisely above the level of the sacrouterine ligaments at the cervicocorporal junction, a polyester suture (Braun cervix set, USP 6, 0.5 cm; B. Braun Melsungen AG, Melsungen, Germany) was directed through the paracercival tissue with a laparoscopic Deschamps needle. After placement, the suture was tension-free tied at the posterior site with 8 knots. Prophylactic antibiotic treatment was not applied.
Boston preconceptional cohort
Minor details differ from the Dutch technique. Instead of the use of a laparoscopic Deschamps needle, Mersilene tape (Ethicon, Somerville, NJ) with curved needles was used. The needles had to be straightened to fit through the trocar and to enable correct placement through the paracervical tissue at the cervicocorporal junction. A 5-mm wide Mersilene tape suture was placed medial to the uterine vessels. The needles were cut off and removed; 7 knots were placed anteriorly. The ends of the Mersilene tape were secured to the lower uterine segment with a 2-0 silk suture. The vesicouterine peritoneum was then closed with Monocryl suture (Ethicon) tied intracorporeally.
Participating centers
The following hospitals were included in the Dutch cohort: VU University Medical Center, Amsterdam, The Netherlands (n = 25); Haga Ziekenhuis, The Hague, The Netherlands (n = 1); Maxima Medisch Centrum, Veldhoven, The Netherlands (n = 1); University Medical Center Groningen, Groningen, The Netherlands (n = 1); Ziekenhuis Oost-Limburg, Genk, Belgium (n = 4). In the Boston cohort all patients (n = 34) were included in the Brigham and Women’s Hospital, Boston, MA.
Outcome measures
The primary outcome measure was defined as delivery at ≥34 weeks of gestation with neonatal survival to hospital discharge.
Secondary outcome measures included surgical outcome parameters (eg, total operation time [defined as the total time in the operating room of LAC placement], excluding surgeries during which a laparoscopic myomectomy or cyst removal was also performed, total blood loss, hospitalization, and complications), pregnancy rate after surgery, and complications during pregnancy (eg, preterm contractions, use of tocolysis, preterm premature rupture of membranes). Deliveries at <34, <30, and <28 weeks of gestation were calculated for all pregnant patients, according to the most frequently used cutoff values in recent publications.
In the Dutch cohort, secondary outcome measures also included patient satisfaction and experiences regarding surgery and pregnancy ( Figure 2 ).
Definitions
Early pregnancy was defined as a positive (urine or blood) pregnancy test. Ongoing pregnancy was defined as a vital intrauterine pregnancy at 12 weeks of gestation. Cutoff values for first-, second-, and third-trimester loss were defined as fetal loss at ≤13 weeks of gestation, fetal loss at 14-22 weeks of gestation, and fetal loss at ≥23 weeks of gestation, respectively, because these cut-off values have been used most frequently in the recent literature. Fetal survival rate per pregnancy was defined as the total number of live born infants who survived the neonatal period (to hospital discharge) divided by the total number of all pregnancies. Fetal survival rate per ongoing pregnancy was defined as the total number of live born infants who survived the neonatal period divided by the total number of all ongoing pregnancies (excluding first-trimester losses).
Statistical analysis
Results of both cohorts are presented separately and combined (total preconceptional cohort). Outcomes of both preconceptional cohorts were compared with the use of the Student t test for continuous data in case the parameter was distributed normally; otherwise the Mann-Whitney U test was applied. Categoric variables were compared with the use of the Fisher exact test. We compared the outcome between the Dutch and Boston cohort using logistic regression analysis with respect to delivery at ≥34 weeks of gestation and second-trimester fetal losses. Linear regression analysis was performed to compare the outcome of both cohorts with respect to gestational age at cesarean delivery. In these 3 regression analyses, previous failed vaginal cerclage and maternal age were included as covariables ( Table 1 ). All tests were 2-sided; a probability value of < .05 was considered significant. For the analyses of pregnancy outcome, we used only the results of the first pregnancy (miscarriage or delivery) after preconceptional LAC.
| Variable | Regression coefficient (B) | SE | P value |
|---|---|---|---|
| Logistic regression analysis for delivery at ≥34 weeks of gestation | |||
| Dutch-Boston cohort | 1.29 | 0.81 | .11 |
| Dutch-Boston cohort a | 1.16 | 0.83 | .16 |
| Dutch-Boston cohort b | 0.81 | 0.93 | .38 |
| Logistic regression analysis for 2nd-trimester fetal loss | |||
| Dutch-Boston cohort | 0.63 | 1.28 | .62 |
| Dutch-Boston cohort a | 0.98 | 1.30 | .45 |
| Dutch-Boston cohort b | –0.11 | 1.57 | .94 |
| Linear regression analysis for gestational age at cesarean delivery | |||
| Dutch-Boston cohort | –2.34 | 0.51 | < .01 |
| Dutch-Boston cohort a | –2.28 | 0.50 | < .01 |
| Dutch-Boston cohort b | –2.05 | 0.56 | < .01 |
a Covariable: previous failed vaginal cerclage;
Results
Demographic results
Sixty-six patients had a preconceptional LAC placed ( Figure 1 ). Patients in the Dutch cohort were significantly younger than patients in the Boston cohort (33.3 vs 36.1 years; P < .01). At least 60% of all patients had 1 previous second- or third-trimester pregnancy loss. In the group with indication of a previous failed vaginal cerclage, all patients had experienced a previous delivery at 14-34 weeks of gestation, despite their vaginal cerclage. More than one-half of all patients had a previously failed vaginal cerclage. Apart from age, demographic characteristics were not significantly different between the Dutch and Boston cohorts. Details are presented in Table 2 .
| Variable | Preconceptional cohort | Total (n = 66) | P value d | |||||
|---|---|---|---|---|---|---|---|---|
| Dutch | Boston | |||||||
| Previous failed vaginal cerclage (n = 16) a | Previous cervical surgery (n = 10) a | Total Dutch cohort (n = 32) b | Previous failed vaginal cerclage (n = 23) a | Previous cervical surgery (n = 2) a | Total Boston cohort (n = 34) c | |||
| Demographic data | ||||||||
| Study group, n | 16 | 10 | 32 | 23 | 2 | 34 | 66 | |
| Mean age ± SD, y e | 32.7 ± 4.0 | 34.9 ± 2.8 | 33.3 ± 4.2 | 36.3 ± 3.3 | 40.5 ± NA | 36.1 ± 3.8 | 34.8 ± 4.2 | .006 |
| Ethnicity | .13 | |||||||
| White, % | 76.9 | 100 | 84.0 | 65.0 | 100 | 92.0 | 88.1 | |
| Non-white, % | 23.1 | 0 | 16.0 | 35.0 | 0 | 8.0 | 11.9 | |
| Body mass index, kg/m 2 f , g | 26.5 ± 8.3 | 26.6 ± 15.2 | 26.3 ± 7.0 | 23.7 ± 11.3 | NR | 23.8 ± 10.7 | 25.2 ± 9.3 | .98 |
| Gravidity, n g | 3.0 ± 3.0 | 0.5 ± 1.3 | 2.0 ± 2.5 | 3.0 ± 2.0 | NA | 3.0 ± 2.0 | 2.0 ± 2.0 | .58 |
| Parity, n g | 2.0 ± 2.0 | 0.5 ± 2.0 | 2.0 ± 2.0 | 2.0 ± 1.0 | NA | 1.0 ± 1.0 | 1.0 ± 1.0 | .91 |
| Obstetric history | ||||||||
| Patients with ≥1 pregnancy loss during 2nd/3rd trimester, % | 66.7 | 50.0 | 64.3 | 69.6 | 50.0 | 60.6 | 62.3 | .80 |
| Patients with delivery at 14-34 weeks of gestation, % h | 100 | 50.0 | 75.9 | 100 | 50.0 | 72.7 | 74.2 | .88 |
| 1st trimester losses, n g | 0.0 ± 1.0 | 0.0 ± 0.3 | 0.0 ± 0.8 | 0.0 ± 1.0 | NA | 0.0 ± 1.0 | 0.0 ± 1.0 | .39 |
| 2nd and 3rd trimester losses, n g | 2.0 ± 2.0 | 0.5 ± 1.3 | 1.0 ± 2.0 | 1.0 ± 2.0 | NA | 0.9 ± 0.9 | 1.0 ± 2.0 | .34 |
| Indication cerclage | ||||||||
| Patients with a previous failed vaginal cerclage, % i | 100 | 0 | 50.0 | 100 | 0 | 67.6 | 61.3 | .19 |
| Patients with surgical (cervical) history, % j | 31.3 | 100 | 46.4 | 30.4 | 100 | 27.3 | 36.1 | .18 |
a Indication placement of laparoscopic abdominal cerclage;
b Includes 6 patients with “other” indication: 4 patients with previous immature or premature delivery and/or instrumental termination of pregnancy that lead to clinical assessment of a technically unfeasible cervix to insert a vaginal cerclage and 2 patients with no reported data of indication of laparoscopic abdominal cerclage placement;
c Includes 9 patients with “other” indication: 8 patients with previous immature or premature delivery and/or instrumental termination of pregnancy that lead to clinical assessment of a technically unfeasible cervix to insert a vaginal cerclage and 1 patient with no reported data of indication of laparoscopic abdominal cerclage placement;
d Preconceptional total Dutch/total Boston cohort;
e Data are given as mean ± SD;
g Data are given as median ± interquartile range;
h Includes all 2nd- and 3rd-trimester fetal losses and all (immature and premature) deliveries at 14-34 weeks of gestation (excluding 1st trimester losses);
i Defined as a previous 2nd or 3rd trimester fetal loss, immature or premature delivery (defined as delivery at 14-34 weeks of gestation) with a vaginal cerclage in situ; the higher percentage of failed vaginal cerclage compared with previous 2nd/3rd trimester losses in the Boston preconceptional cohort can be clarified by the fact that not all failed vaginal cerclage resulted in a pregnancy loss, but other complications, such as prematurity (defined as delivery at 34 weeks of gestation or less), instead;
j Surgical cervical history includes recurrent loop electrosurgical excision procedure, (laser) conisation of the cervix, and trachelectomy.
Surgical outcomes
Surgical outcomes report on complications that occurred to 2 weeks after surgery ( Table 3 ). Mean total operation time (the total time in the operating room) in the Dutch and the Boston cohorts was 88.8 vs 96.6 min ( P = .29), respectively. Mean blood loss in all patients was <40 mL. Severe hemorrhage (defined as blood loss of >400 mL) did not occur in any case. In 3 of 66 patients (4.5%) with a preconceptional LAC, minor perioperative complications occurred (perforation of the uterus and a pelvic infection). According to the protocol, all patients went home 1 day after the surgical procedure in the Dutch cohort and on the day of the procedure in the Boston cohort. No additional antibiotics or medication were used in the perioperative period. LAC could be placed without major problems in all patients and without conversion to a laparotomy.
| Preconceptional cohort | ||||
|---|---|---|---|---|
| Outcome | Total Dutch (n = 32) | Total Boston (n = 34) | Total (n = 66) | P value a |
| Total operation time, min b , c | 88.8 ± 22.7 | 96.6 ± 23.3 | 93.1 ± 23.1 | .29 |
| Blood loss during surgery, mL | 26.1 ± 22.1 | 35.5 ± 21.8 | 31.6 ± 22.2 | .12 |
| Severe hemorrhage >400 mL, n c | 0 | 0 | 0 | NA |
| Blood transfusion required, d n | 0 | 0 | 0 | NA |
| Additional complications, n (%) d , e | 2 (7.7) | 1 (3.0) | 3 (4.5) | .58 |
| Median nights of admission to hospital after surgery | 0 | 0 | 0 | .23 |
a Preconceptional total Dutch/total Boston cohort;
b Total time in the operating room; all surgeries in which a laparoscopic myomectomy or cyst removal was performed were excluded from operation time analysis;
c Data are given as mean ± SD;
d During surgery or until 2 weeks after surgery;
e Other than severe hemorrhage (>400 mL), for example, uterine perforation, pelvic infection, fever.
Pregnancy outcomes
With the exception of mean gestational age at cesarean delivery, pregnancy outcomes were not significantly different between the Dutch and Boston cohorts ( Table 4 ); results have been combined in the total preconceptional cohort. A total of 56 patients (84.8%) had a complete follow-up examination after preconceptional surgery. Four patients were lost to follow up, and 36 of 52 patients (69.2%) became pregnant after LAC placement. One patient was in her second trimester of pregnancy at submission of this article and therefore could not be included in all analyses ( Figure 1 ). Three of 35 patients (8.6%) had a second-trimester miscarriage; all 3 patients had a LAC based on a previous failed vaginal cerclage. The first patient had bulging membranes (without uterine contractions) through the cerclage on a routine checkup in the hospital at 16 weeks + 4 days of gestation and signs of intrauterine infection, given elevated C-reactive protein plasma levels. After counseling, the pregnancy was evacuated by curettage. The second patient was admitted to the hospital at 15 weeks + 6 days of gestation for abdominal pain. On ultrasound scanning, bulging membranes were visible. Despite immobilization and intravenous adminstration of antibiotics, the uterine contractions persisted, and membranes were bulging into the introitus. After counseling, she chose for termination of the pregnancy, and the fetus (consistent with 12 weeks of gestation) was evacuated by curettage. Both procedures were complicated with a total blood loss of 1000 mL. The third patient had bulging membranes at 21 weeks of gestation without uterine contractions and was admitted to the hospital. Because of progressive cervical dilation in the following 2 days of hospitalization, the fetus was evacuated by uncomplicated dilation and extraction. No third-trimester losses occurred.
| Outcome | Preconceptional cohort | |||||||
|---|---|---|---|---|---|---|---|---|
| Dutch | Boston | Total (n = 56) | P value d | |||||
| Previous failed vaginal cerclage a (n = 12) | Previous cervical surgery a (n = 8) | Total b (n = 27) | Previous failed vaginal cerclage a (n = 21) | Previous cervical surgery a (n = 2) | Total c (n = 29) | |||
| Patients lost to follow up, n | 0 | 0 | 0 | 0 | 0 | 4 | 4 | |
| Pregnant patients after interval procedure, n/N (%) e | 12/16 (80.0) | 4/8 (50.0) | 19/27 (70.4) | 13/21 (61.9) | 2/2 (100.0) | 17/25 (68.0) | 36/52 (69.2) | .75 |
| Complications during pregnancy | ||||||||
| Total pregnant patients, n | 12 f | 4 | 19 f | 13 | 2 | 17 | 36 | |
| Total fetal loss per trimester, n/N (%) | 2/11 (18.2) | 0 | 3/18 (16.7) | 5/13 (38.5) | 0 | 5/17 (29.4) | 8/35 (22.9) | .44 |
| 1st | 0 | 0 | 1/19 (5.3) d | 4/13 (30.8) | 0 | 4/17 (23.5) | 5/36 (13.9) | .17 |
| 2nd | 2/11 (18.2) | 0 | 2/18 (11.1) | 1/13 (7.7) | 0 | 1/17 (5.9) | 3/35 (8.6) | .52 |
| 3rd | 0 | 0 | 0 | 0 | 0 | 0 | 0 | NA |
| Premature preterm rupture of membranes g | 1/11 (9.1) | 0 | 1/18 (5.6) | 0 | 0 | 0 | 1/35 (2.9) | .58 |
| Chorioamnionitis | 0 | 0 | 0 | 0 | 0 | 0 | 0 | NA |
| Premature contractions at <37 wk of gestation, n/N (%) h | 3/11 (27.3) | 0 | 3/18 (16.7) | 2/13 (15.4) | 0 | 2/17 (11.8) | 5/35 (14.3) | .64 |
| Medication i | 3/11 (27.3) | 0 | 3/18 (16.7) | 1/13 (7.7) | 0 | 2/17 (11.8) | 5/35 (14.3) | .63 |
| Immobilization j | 2/11 (18.2) | 1/4 (25.0) | 3/18 (16.7) | 3/13 (23.1) | 0 | 4/17 (23.5) | 7/35 (20.0) | .67 |
| Other complications k | 0 | 1/4 (25.0) | 2/18 (11.1) | 1/13 (7.7) | 0 | 1/17 (5.9) | 3/35 (8.6) | .61 |
| Pregnancy outcome | ||||||||
| Total pregnancies with complete follow up, n l | 11 | 4 | 18 | 13 | 1 | 17 | 35 | |
| Total deliveries m | ||||||||
| Gestational age ≥34 wk, n/N (%) | 9/11 (81.8) | 4/4 (100) | 15/18 (83.3) | 6/13 (46.2) | 1/1 (100.0) | 10/17 (58.8) | 25/35 (71.4) | .15 |
| Gestational age <34 wk, n/N (%) | 0 | 0 | 0 | 2/13 (15.4) | 0 | 2/17 (11.8) | 2/35 (5.7) | .23 |
| Gestational age <30, n/N | 0 | 0 | 0 | 0 | 0 | 0 | 0 | NA |
| Gestational age <28 wk, n/N | 0 | 0 | 0 | 0 | 0 | 0 | 0 | NA |
| Patients with cesarean delivery, n | 9 | 4 | 15 | 8 | 1 | 12 | 27 | |
| Mean gestation at cesarean delivery, wk n | 37.9 ± 0.9 | 38.5 ± 1.0 | 38.2 ± 0.9 | 35.5 ± 1.8 | 37.0 ± NA | 35.9 ± 1.6 | 37.2 ± 1.7 | < .001 |
| Complications during cesarean delivery, n/N (%) o | 2/9 (22.2) | 0 | 2/15 (13.3) | 2/8 (25.0) | 0 | 3/12 (25.0) | 5/27 (18.5) | .63 |
| Fetal outcome: survival rate, % | ||||||||
| Pregnancy p | 81.8 | 100 | 83.3 | 88. | 100 | 70.6 | 77.1 | .38 |
| Ongoing pregnancy q | 81.8 | 100 | 88.2 | 61.5 | 100 | 92.3 | 90.0 | .72 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
