|
|
||||||||
ORIGINAL RESEARCH |
From the Departments of Obstetrics and Gynecology, 1Radboud University Nijmegen Medical Centre, Nijmegen; and 2Erasmus Medical Center, Rotterdam, the Netherlands.
| ABSTRACT |
|---|
|
|
|---|
METHODS: This was an observational cohort study with historical controls of 101 pregnancies after transabdominal cervicoisthmic cerclage in 101 women with a classic history of cervical insufficiency and severe cervical defects precluding transvaginal cerclage.
RESULTS: Median gestational age at elective transabdominal cerclage (n = 95) was 14 (range 1216) weeks and at emergency cerclage (n = 6) was 18 (range 1722) weeks. Perioperative complications were blood loss 500 mL or more (n = 3) and rupture of membranes (n = 2). Patients were delivered by cesarean. Before cerclage 76% (95% confidence interval [CI] 70.281.1%) of births occurred before 32 weeks of gestation; total neonatal survival was 27.5% (95% CI 22.533.8%). After transabdominal cervicoisthmic cerclage 7% (95% CI 2.913.9%) of births took place before 32 weeks of gestation, and total neonatal survival was 93.5% (95% CI 85.596.6%).
CONCLUSION: In women with a classic history of cervical insufficiency and a traumatized cervix that precludes transvaginal cerclage, transabdominal cervicoisthmic cerclage is associated with successful outcome in the absence of procedure-related major complications.
LEVEL OF EVIDENCE: II-2
Surgical treatment of cervical insufficiency is generally accepted to consist of cervical cerclage in a subsequent pregnancy, although the benefit of the procedure remains disputed.3,6 The cervical suture is usually inserted by the transvaginal route. However, the vaginal approach can be surgically unfeasible or hazardous in women with an absent, very short, or severely lacerated intravaginal portion of the cervix as a result of developmental abnormality, previous surgery, or failed previous transvaginal cerclage. In these women transabdominal cervicoisthmic cerclage, as first described by Benson and Durfee,7 may be the only possibility for prophylactic or therapeutic surgical closure of the insufficient cervix.
In 1987 we reported our successful experience with transabdominal cervicoisthmic cerclage in 13 pregnant patients with 2 or more successive second-trimester losses, a clinical history of cervical insufficiency, and an extremely short, scarred, or partially absent cervix.8 In following years, we continued performing the procedure in carefully selected patients.
The aim of the present observational cohort study was to analyze the indications and maternal and fetal benefits and risks of transabdominal cervicoisthmic cerclage performed in 1 center in 101 pregnancies of 101 women with a history of cervical insufficiency in which transvaginal cervical cerclage was considered surgically unfeasible or hazardous because of severe cervical defects.
| MATERIALS AND METHODS |
|---|
|
|
|---|
The obstetric history of 101 pregnant women selected for transabdominal cervicoisthmic cerclage is presented in Table 1. All patients except 19 met the criteria of at least 2 lost pregnancies in the second- or early third-trimester, with a classic history of cervical insufficiency. Of 188 second- and early third-trimester deliveries, 88 (47%) in 52 women were associated with failed cervical cerclage. Fifteen women had only 1 previous second trimester loss. Seven of these had a vaginal portion of the cervix with a length of less than 2 mm and the characteristic cervical abnormalities of DES exposure in utero, 4 had undergone amputation or extensive exconization before the unsuccessful pregnancy, and 4 had a short and lacerated cervix after dilatation and curettage for 1 or more spontaneous miscarriages. Of the 4 nulliparous women, 1 was selected for transabdominal cervicoisthmic cerclage because of an almost absent cervix and a quadruplet pregnancy that was reduced to a twin pregnancy, 1 because of a short cervix associated with DES exposure in utero, long-standing infertility treatment, and extreme anxiety. The 2 remaining nulliparous women underwent emergency transabdominal cervicoisthmic cerclage because of a failed transvaginal cerclage of a practically absent cervix in 1 and a failed transvaginal cerclage of a cervix after exconization in the other. All patients had severe cervical deformities, 60% of which were attributable to damage caused by previous delivery with a cervical suture in situ, in most cases a Shirodkar cerclage.
|
We aimed at performing primary transabdominal cervicoisthmic cerclage at the end of the first trimester of pregnancy. Calculated gestational age, fetal cardiac activity, and the absence of detectable fetal anomalies were ascertained by ultrasonography. Before surgery, none of the patients had experienced abnormal uterine contractility, ruptured membranes, or uterine bleeding. Ultrasound measurement of cervical length before the transabdominal cervicoisthmic cerclage was not routinely performed and, if done, was not used as a criterion for selection.
All transabdominal cervicoisthmic cerclage operations were performed by either 1 or 2 of the authors (H.W. and F.L.) using the technique reported by Benson and Durfee7 with modifications as previously described.8 Under general or epidural anesthesia, the peritoneal cavity was entered by a low Pfannenstiel incision. The peritoneal bladder fold was opened and the bladder advanced downward to expose the cervicoisthmic junction. On both sides, the avascular area between the pulsating ascending and descending branch of the uterine artery and the isthmus, just below the uterine "waist," was identified by palpation. One hand was placed behind the pregnant uterus and, guided by a posterior finger, the uterine vasculature on 1 side was gently displaced laterally from anterior with the tip of a long right-angled clamp with tapered jaws, and the posterior leaf of the broad ligament was punctured. A 40-cm-long, 5-mm-wide Mersilene (Ethicon, Norderstedt, Germany) band was grasped between the jaws of the clamp and gently pulled through the paracervical space from posterior to anterior. The procedure was repeated on the other side. The band was pulled tight around the cervix, and the ascending branches of the uterine arteries were palpated to confirm the presence of pulsations. The band was then tied snugly on the anterior side of the cervix and the cut ends of the band were fixed to the band with thin nonabsorbable sutures. The peritoneal bladder fold was closed over the band, and on the posterior side the band was left uncovered by peritoneum. During the procedure the uterus was kept moist and remained inside the abdominal cavity.
Our initial routine of prophylactic tocolysis during the operation was abandoned after the first 20 procedures. Prophylactic antibiotics were administered only to the 6 patients who underwent an emergency procedure.
All patients received routine postoperative care and were usually discharged within 5 days. Standard antenatal care was provided by the referring obstetricians or by ourselves. Vaginal or ultrasound examinations of the cervix were not routinely performed, and reduction of physical or sexual activity was not recommended. We intended to perform an elective cesarean delivery by a transverse uterine incision above the level of the band between 36 and 38 weeks gestation and to leave the band in situ.
All data concerning patients' history, operation, course of pregnancy, delivery, postpartum period, and later follow-up were recorded in a standardized fashion over the course of 20 years. To ascertain completeness of follow-up, we reviewed all files of women who had remained in our care and, in 20032004, mailed out questionnaires to all referring gynecologists. For this study we analyzed only the data concerning the index pregnancy, in which transabdominal cervicoisthmic cerclage was performed. Totals, median values and ranges, and percentages with 95% confidence intervals were calculated for selected variables as appropriate.
| RESULTS |
|---|
|
|
|---|
Complications and obstetric outcome of transabdominal cervicoisthmic cerclage are presented in Table 2. The indications for the emergency operations were failed transvaginal cerclage in the first trimester of short and scarred cervices, followed by slippage or tearing out of the ribbon and cervical dilatation. Median blood loss was 50 mL; operative hemorrhage of 800 mL, 850 mL, and 1,500 mL, respectively, occurred in 3 patients. Bleeding was always caused by rupture of 1 or more thin-walled parametrial veins and occurred at the time of tunnelling the broad ligament or pulling the ribbon through. It usually stopped once the band was tied, but a few cases required stitches. All other perioperative complications occurred in the first 20 operations that were performed. In 1 case the scar of a previous cesarean delivery was opened accidentally during dissection of the bladder. The membranes remained intact, the scar was closed, the operation was continued successfully, and the pregnancy resulted in a term live infant. In 2 patients the membranes ruptured spontaneously 4 and 5 days, respectively, after an uncomplicated procedure. In both patients the uterus could be evacuated by transvaginal route, and the cerclage remained in situ. After the operation 3 women complained of frequent painful micturition that could not be attributed to bacterial cystitis. The bladder irritability gradually subsided within a few weeks. No other complications, such as fever, hemorrhage, or uterine contractions, occurred in the 4 weeks after the operation, including the 6 emergency procedures.
|
Seven women (95% CI 2.913.9) delivered before 32 weeks. Six pregnancies ended in delivery of 7 infants before 28 weeks of gestation. As noted above, in 2 patients the membranes ruptured and the fetuses died 4 and 5 days, respectively, after the operation. In 1 case, unexplained fetal death occurred 9 weeks after transabdominal cervicoisthmic cerclage at 22 weeks of gestation; the fetus and placenta were removed by aspiration. In 2 patients, 1 with a prophylactic transabdominal cervicoisthmic cerclage and 1 who underwent an emergency procedure, intrauterine infection and fetal demise necessitated delivery by the abdominal route at 20 weeks and 24 weeks, respectively. In 1 patient with a twin pregnancy, cesarean delivery was performed at 25 weeks of gestation because of rupture of membranes and intrauterine infection; 1 neonate survived. In the course of pregnancy, 20 women were admitted before 32 completed weeks of gestation for observation on suspicion of uterine contractions, and some were briefly treated with intravenous tocolysis. Preterm labor was confirmed in 1 woman who received intravenous tocolysis but delivered at 31 weeks.
In 94 women, pregnancy after transabdominal cervicoisthmic cerclage continued past 32 weeks of gestation, and 76 women delivered after 35 completed weeks, including 5 patients who underwent an emergency operation. The cerclage band was left in situ in all patients. Cesarean delivery was complicated by blood loss of more than 1,000 mL in 3 patients due to uterine hypotonia, placenta previa, and placenta accreta, respectively. In the first 2 cases hemorrhage was controlled with oxytocic agents, and the third patient underwent hysterectomy. One patient suffered pulmonary embolism 2 months after cesarean delivery and survived after treatment. One term neonate died 2 days after birth from a congenital cardiac abnormality. The total neonatal survival after transabdominal cervicoisthmic cerclage was 93.5% (95% CI 85.596.6%).
| DISCUSSION |
|---|
|
|
|---|
We performed elective transabdominal cerclage between 12 and 16 weeks of gestation. At that duration of pregnancy major fetal anomalies can be excluded by ultrasonography, and the risk of spontaneous abortion in a fetus with cardiac activity is small. Later in the second trimester it becomes more difficult to obtain adequate exposure, and manipulation of the larger uterus could increase the risk of uterine contractions. As an emergency procedure transabdominal cervicoisthmic cerclage was performed successfully up to 22 weeks of gestation in our patients and up to 26 weeks in other studies.13 There is no apparent indication to perform transabdominal cerclage before 12 weeks of gestation, because it is unlikely that spontaneous abortions at that time are related to cervical insufficiency. Some authors favor performance of transabdominal cervicoisthmic cerclage before conception. Published reports provide no evidence that the procedure is surgically easier or has fewer complications than when performed after the first trimester of pregnancy. An obvious disadvantage of preconception transabdominal cervicoisthmic cerclage is that pregnancy may not occur, either deliberately or involuntarily, as reported in the literature.15,16
The maternal risks of transabdominal cervicoisthmic cerclage associated with banding the cervicoisthmic junction in a highly vascular area close to the ureters should not be underestimated.9 Bleeding from parametrial veins is the main complication in reported studies, in particular when the uterine vasculature is dissected to localize a vessel-free paracervical space.7,17,18 In our experience, dissection is not necessary; the vessel-free area can be determined by palpation, and during puncture of the paracervical tissue, the uterine vessels and the ureters are avoided by gently pushing them laterally. The risk of damaging parametrial vessels may be further reduced by the use of fiberoptic lighting to transilluminate the broad ligament.19 After pulling the ribbon through, the uterine arteries should be palpated to make sure that they are not caught inside the ligature, which will cause uterine ischemia and may lead to fetal death.7,20 Our perioperative complications other than hemorrhage were encountered in the first series of 20 operations, which may reflect a learning curve. All operations were performed by 2 attending gynecologists and increasing experience with the relatively rare procedure may have contributed to the low complication rate. Recent reports suggest that the potential complications inherent in the laparotomy may be avoided by laparoscopic placement of a cervicoisthmic cerclage, either as a preconception interval procedure21 or in early pregnancy.22 Placement of the band close to the internal cervical os, with inclusion of part of the uterus in the cerclage when the cervix is completely absent, apparently does not compromise the development of the lower uterine segment. Cesarean delivery must be performed in all pregnancies in which fetal viability is reached, and hysterotomy may be required in case of fetal demise. The cerclage is not removed at cesarean delivery and may serve in future pregnancies, which many patients desire after the encouragement of a successful pregnancy.
Compared with 76% of births before 32 weeks of gestation in the history of our patients, 7% very preterm births after transabdominal cervicoisthmic cerclage constitutes marked improvement. In addition, 93% neonatal survival compared with 27% before transabdominal cerclage indicates a beneficial effect of the procedure and is in agreement with the average success rate of 89% in published reports.20 For our study that covers 2 decades, the rate of severe preterm delivery may be considered a better measure of success than neonatal survival because of significant improvements in neonatal care and survival in those years. We realize the potential bias inherent in using patients as their own historical controls. However, a randomized trial in our patient cohort would have required a no-cerclage group as the control sample, which would have been unacceptable in this high-risk population. Our study does not answer the question of whether transabdominal cervicoisthmic cerclage is more effective than the usual transvaginal cerclage. One study, in which women with short cervices were excluded, found that transabdominal cerclage was associated with a lower incidence of preterm rupture of membranes and preterm delivery compared with transvaginal cerclage.23
In contrast to most other reports, we only present the outcome of the pregnancy in which transabdominal cerclage was performed. Inclusion of successive pregnancies with the same cervicoisthmic cerclage in situ excludes procedure-related complications and carries the risk of selection bias; patients with a failed transabdominal cervicoisthmic cerclage may choose to refrain from future pregnancies, whereas women with successful transabdominal cervicoisthmic cerclage may opt for more pregnancies with the same cerclage in situ.
In conclusion, our results indicate that the benefits of transabdominal cervicoisthmic cerclage outweigh its inherent disadvantages and risks in carefully selected patients with a diagnosis of cervical insufficiency in whom a transvaginal approach is judged surgically unfeasible or hazardous. We recommend that women considered for transabdominal cervicoisthmic cerclage be referred for counseling and for the operation to a center with experience in performing the procedure.
| Footnotes |
|---|
doi:10.1097/01.AOG.0000206817.97328.cd
| REFERENCES |
|---|
|
|
|---|
2. Shennan A, Jones B. The cervix and prematurity: aetiology, prediction and prevention. Semin Fetal Neonatal Med 2004;9:4719.[Medline]
3. Drakeley AJ, Roberts D, Alfirevic Z. Cervical stitch (cerclage) for preventing pregnancy loss in women. Cochrane Database Syst Rev 2003;(1):CD003253.
4. Milhan D. DES exposure: implications for childbearing. Int J Childbirth Educ 1992;7:218.[Medline]
5. Kristensen J, Langhoff-Roos J, Wittrup M, Bock JE. Cervical conization and preterm delivery/low birth weight: a systematic review of the literature. Acta Obstet Gynecol Scand 1993;72:6404.[Medline]
6. Althuisius SM, Dekker GA, van Geijn HP. Cervical incompetence: a reappraisal of an obstetric controversy. Obstet Gynecol Surv 2002;57:37787.[Medline]
7. Benson RC, Durfee RB. Transabdominal cervico uterine cerclage during pregnancy for the treatment of cervical incompetency. Obstet Gynecol 1965;25:14555.
8. Wallenburg HC, Lotgering FK. Transabdominal cerclage for closure of the incompetent cervix. Eur J Obstet Gynecol Reprod Biol 1987;25:1219.[Medline]
9. Harger JH. Cerclage and cervical insufficiency: an evidence-based analysis. Obstet Gynecol 2002;100:131327.
10. McDonald IA. Suture of the cervix for inevitable miscarriage. J Obstet Gynaecol Br Emp 1957;64:34650.[Medline]
11. Golfier F, Bessai K, Paparel P, Cassignol A, Vaudoyer F, Raudrant D. Transvaginal cervicoisthmic cerclage as an alternative to the transabdominal technique. Eur J Obstet Gynecol Reprod Biol 2001;100:1621.[Medline]
12. Katz M, Abrahams C. Transvaginal placement of cervicoisthmic cerclage: report on pregnancy outcome. Am J Obstet Gynecol 2005;192:198992.[Medline]
13. Hole J, Tressler T, Martinez F. Elective and emergency transabdominal cervicoisthmic cerclage for cervical incompetence. J Reprod Med 2003;48:596600.[Medline]
14. Olsen S, Tobiassen T. Transabdominal isthmic cerclage for the treatment of incompetent cervix. Acta Obstet Gynecol Scand 1982;61:4735.[Medline]
15. Borruto F, Ferraro F. Transabdominal cervico-isthmic cerclage: state of the art. Clin Exp Obstet Gynecol 1990;17:1514.[Medline]
16. Groom KM, Jones BA, Edmonds DK, Bennett PR. Preconception transabdominal cervicoisthmic cerclage. Am J Obstet Gynecol 2004;191:2304.[Medline]
17. Watkins RA. Transabdominal cervico-uterine suture. Aust N Z J Obstet Gynaecol 1972;12:624.[Medline]
18. Mahran M. Transabdominal cervical cerclage during pregnancy: a modified technique. Obstet Gynecol 1978;52:5026.
19. Olatunbosun O, Turnell R, Pierson R. Transvaginal sonography and fiberoptic illumination of uterine vessels for abdominal cervicoisthmic cerclage. Obstet Gynecol 2003;102:11303.
20. Novy MJ. Transabdominal cervicoisthmic cerclage: a reappraisal 25 years after its introduction. Am J Obstet Gynecol 1991;164:163541.[Medline]
21. Mingione MJ, Scibetta JJ, Sanko SR, Phipps WR. Clinical outcomes following interval laparoscopic transabdominal cervico-isthmic cerclage placement: case series. Hum Reprod 2003;18:17169.
22. Cho CH, Kim TH, Kwon SH, Kim JI, Yoon SD, Cha SD. Laparoscopic transabdominal cervicoisthmic cerclage during pregnancy. J Am Assoc Gynecol Laparosc 2003;10:3636.[Medline]
23. 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;183:8369.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |