Objective
The purpose of this study was to evaluate the benefit of cervical cerclage for cervical length ≤25 mm at the time of fetoscopic laser photocoagulation (FLP) for twin-twin transfusion syndrome.
Study Design
A multicenter, retrospective cohort study was conducted with 163 patients with a short cervix before FLP for twin-twin transfusion syndrome. Seventy-nine of the patients (48%) had cerclage placement at the surgeon’s discretion. The outcome measures that were compared were gestational age at delivery and perinatal mortality rates for patients with cerclage and those who were treated conservatively. Outcomes were evaluated with the use of comparative statistics.
Results
There were no differences in the preoperative variables, except cerclage was performed more frequently for a cervical length of ≤15 mm ( P < .001). There were no differences in the gestational age at delivery (28.8 ± 5.4 vs 29.1 ± 5.6 weeks with and without cerclage, respectively; P = .15); perinatal mortality rates were similar between the 2 groups.
Conclusion
The benefit of cerclage for patients with short cervix before FLP remains questionable.
Fetoscopic laser photocoagulation (FLP) has become the standard treatment for severe twin-twin transfusion syndrome (TTTS). Preterm birth that is due to polyhydramnios-induced cervical shortening in TTTS has been reported as an important cofactor in the determination of perinatal survival after FLP. Robyr et al described that a short cervical length before FLP in TTTS was associated with more premature deliveries, especially if the cervical length was <20 mm. However, Chavira et al, in a retrospective cohort study, noted that a cervical length of <25 mm did not affect the perinatal outcome, which included procedure-to-delivery interval, gestational age at delivery, and neonatal survival. Both studies, however, may have been biased, because cerclages were placed in 50-80% of patients with a cervix of <20 mm.
Salomon et al conducted the only study that compared the benefit of emergent cerclage vs conservative management for a cervical length of <15 mm in patients with TTTS who underwent FLP. This was a retrospective review of 14 patients: 5 patients without cerclage (before 2004) and 9 patients with cerclage (after 2004). The study showed clear improvement in both prolongation of the pregnancy and survival to the end of the neonatal period. However, the sample size in this study was small, and the conclusions have not been validated by any other study. Therefore, it remains unknown whether cervical cerclage for a short cervix in cases of TTTS with FLP prolongs the pregnancy and improves perinatal mortality rates.
The hypothesis of our study was that cervical cerclage for a short cervix (≤25 mm) significantly prolongs the latency period of procedure-to-delivery compared with no cerclage in the setting of TTTS with FLP.
Materials and Methods
This was a retrospective cohort study from 6 North American fetal centers ( Table 1 ). The study was approved by the institutional review boards at Baylor College of Medicine (H-26949), which was the primary data center, and at the participating institutions. Patients with TTTS who were referred to a fetal center underwent a comprehensive ultrasound examination that included Doppler studies for diagnosis and staging that were based on the Quintero system. The cervical length was assessed routinely with transvaginal ultrasound scanning for all patients with suspected TTTS before FLP. This was performed by a sonographer, and the images were approved by a perinatologist at the respective center. The technique for endoscopic laser ablation of placental vessels was a single-port fetoscopy with either local anesthetic and intravenous sedation or regional or general anesthesia. The placental vascular equator was identified, and neodynium-yttrium aluminum garnet or diode laser was used to ablate all vascular anastomoses. During the study period, all centers used a random (nonsequential) selective ablation method. The decision to perform a cerclage before or within 1 week of FLP, the type of suture that was used used, and the type of cerclage that was performed (McDonald or Shirodkar ) were at the surgeon’s discretion. One of the following suture materials was used: Mersilene tape (Ethicon Inc, Somerville, NJ), #5 Ticron (Covidian, North Haven, CT) or 0-Prolene (Ethicon Inc). All patients returned to their referring physicians for follow-up evaluation in the remainder of the pregnancy. Information about pregnancy outcomes was collected from the patients or the referring physicians as a part of clinical care. Study data from all centers were de-identified and transferred to the primary data center, where they were stored on a secure digital drive.
| Centers | Study period | Included patients, n | Cerclage, n (%) |
|---|---|---|---|
| Children’s Hospital of Philadelphia, Philadelphia, PA | 04/2004–7/2010 | 4 a | 0 |
| Cincinnati Children’s Hospital, Cincinnati, OH | 09/2005–11/2010 | 39 | 29 (75) |
| Evergreen Hospital, Kirkland, WA | 04/2004–09/2009 | 24 | 12 (50) |
| Mount Sinai Hospital, Toronto, Canada | 01/2004–11/2010 | 34 | 19 (56) |
| Texas Children’s Fetal Center, Houston, TX | 10/2006–12/2010 | 43 | 16 (37) |
| University of Maryland School of Medicine, Baltimore, MD | 05/2005–10/2010 | 19 | 3 (16) |
| Total | 163 | 79 (48.5) |
a The center did not offer laser surgery for cervical length of <20 mm.
The inclusion criterion for the study was a cervical length of ≤25 mm that was measured by transvaginal ultrasound scanning before FLP for TTTS. Exclusion criteria were patients with triplets, patients with a prophylactic cerclage that was placed before the diagnosis of TTTS, or patients who were treated only with amnioreduction. We collected the following preoperative variables for each patient: maternal demographics that included age, parity, previous preterm deliveries, race, smoking, major medical illness, Quintero TTTS stage, recipient maximum vertical pocket (MVP), and cervical length (measured transvaginally). Operative variables included the type of anesthesia, MVP after procedure, timing of cerclage (before or after FLP), suture type and number of cerclages placed, and procedure-related complications. Postoperative variables included preterm premature rupture of membranes (PPROM; defined as clinical confirmation of rupture at <34 weeks’ gestation), chorioamnionitis, abruption, fetal death, gestational age at delivery, and neonatal survival.
The data for preoperative, intraoperative, and postoperative variables were extracted to SPSS software (version 11.0; SPSS Inc, Chicago, IL). Parameters were compared with the use of the χ 2 test for categoric variables; the Fisher exact test was used when an expected frequency was <5. The Mann-Whitney U test was used to compare nonparametric continuous variables. An unpaired Student t test was performed for continuous variables, which satisfied the criteria for normal distribution by histogram, kurtosis, skewness, and probability-probability plots. A subgroup analysis was performed to compare the outcomes between different ranges of cervical length: ≤15 mm, 16-20 mm, and 21-25mm. The level of statistical significance was taken as a probability value of < .05. Kaplan-Meier survival plots were created and compared with the use of the Mantel-Cox log rank to test the equality of survival time for both groups. An adjustment for preoperative covariates that included cervical length, recipient’s MVP, maternal age, placentation, and maternal body mass index was performed with the Cox regression model for procedure-to-delivery interval in the group that had significant difference between cerclage and noncerclage groups on a univariate analysis.
Results
A total of 163 patients were included in the study. Seventy-nine patients (48.5%) received a cerclage for a short cervix; 6 patients had a Shirodkar cerclage placed, and 73 patients had a McDonald cerclage placed. The remaining 84 patients (51.5%) were treated conservatively. The timing of cerclage was 41 placements (52%) before and 38 placements (48%) after FLP procedure (6 patients received the cerclage a week after the procedure). The distribution of patients based on each center, study interval, and number of patients who had a cerclage are shown in Table 1 . The rate of cerclage was statistically significantly different between the centers ( P < .001). There were no differences in the patient demographics and the outcomes between the centers ( Appendix ). The distribution of patients with or without cerclage was uniform over the consecutive case series for each center ( Figure 1 ) , except for the Children’s Hospital of Philadephia, which did not offer cerclage.
Preoperative variables were compared between the 2 groups ( Table 2 ). There were no differences between the 2 groups, except that the cerclage group had a shorter cervical length compared with noncerclage group ( P < .001). Intraoperative and postoperative outcomes were also compared ( Table 3 ). There were significant differences in the type of anesthesia that was used for the laser procedure between the 2 groups, with the cerclage group more likely to receive regional or general anesthesia ( P < .001). On a post-hoc analysis, there were more cases of spinal anesthesia in the cerclage group ( P = .001) and more cases of intravenous sedation in the noncerclage group ( P < .001). There were no differences in the gestational age at delivery or in perinatal mortality rates between the 2 groups. Kaplan-Meier survival analysis showed no difference in the procedure-to-delivery interval between the 2 groups ( Figure 2 ; P = .8).
| Variable | Cerclage | No cerclage | P value |
|---|---|---|---|
| n (%) | 79 (48) | 84 (52) | |
| Maternal age, y a | 28.5 ± 5.2 | 27.5 ± 5.5 | .27 |
| Body mass index, kg/m 2 a | 29.5 ± 6.3 | 27.8 ± 6 | .1 |
| Gravidity, n b | 2 (1–9) | 2 (1–6) | .85 |
| Parity, n b | 1 (0–4) | 1 (0–3) | .74 |
| Previous preterm births, n (%) | 4 (5) | 5 (6) | .8 |
| Race, n (%) | .78 | ||
| White | 61 (77) | 61 (73) | |
| African American | 7 (9) | 12 (14) | |
| Hispanic | 4 (5%) | 5 (6) | |
| Asian | 7 (9) | 6 (7) | |
| Tobacco use, n (%) | 2 (3) | 6 (7) | .17 |
| Gestational age at procedure, wk a | 21.4 ± 2.3 | 22.1 ± 2.4 | .07 |
| Twin-twin transfusion syndrome Quintero stage, n (%) | .35 | ||
| I | 13 (16) | 14 (17) | |
| II | 10 (13) | 19 (23) | |
| III | 53 (67) | 47 (56) | |
| IV | 3 (4) | 4 (5) | |
| Anterior placenta, n (%) | 33 (42) | 41 (49) | .08 |
| Recipient maximum vertical pocket before fetoscopic laser photocoagulation, cm a | 12.2 ± 9.4 | 13.2 ± 3 | .05 |
| Cervical length, mm a | 13.9 ± 5.2 | 18.4 ± 5.1 | < .001 |
a Data are given as mean ± SD;
| Variable | Cerclage | No cerclage | P value |
|---|---|---|---|
| n | 79 | 84 | |
| Anesthesia, n (%) | .001 | ||
| General | 7 (9) | 3 (4) | |
| Spinal | 40 (51) | 22 (26) | |
| Intravenous sedation | 32 (40) | 59 (70) | |
| Recipient of maximum vertical pocket after fetoscopic laser photocoagulation, cm a | 5.6 ± 1.5 | 6 ± 1.4 | .08 |
| Net amnioreduction volume, mL a | 1802 ± 969 | 2085 ± 1149 | .1 |
| Preterm premature rupture of membranes, n (%) | 28 (35) | 23 (27) | .4 |
| Chorioamnionitis, n (%) | 1 (1) | 4 (5) | .4 |
| Abruption, n (%) | 0 | 5 (6) | .08 |
| Gestational age at preterm premature rupture of membranes, wk a | 23.4 ± 5.4 | 25.5 ± 4.5 | .15 |
| Gestational week at delivery a | 28.8 ± 5.4 | 29.1 ± 5.6 | .73 |
| Procedure to delivery, d a | 51.1 ± 36.4 | 49.3 ± 37.8 | .76 |
| Survival to birth, n (%) | .09 | ||
| 0 | 9 (14) | 21 (26) | |
| 1 | 15 (18) | 15 (19) | |
| 2 | 55 (68) | 48 (53) | |
| Survival to 28 days birth, n (%) | .62 | ||
| 0 | 18 (21) | 25 (28) | |
| 1 | 16 (19) | 15 (24) | |
| 2 | 45 (57) | 44 (48) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
