Despite the imprecise evidence regarding benefits of cervical cerclage, the concept of cervical insufficiency is accepted among obstetricians without an objective diagnostic test for this condition. The lack of objective findings and clear diagnostic criteria make the incidence of cervical insufficiency difficult to ascertain and the performance of properly designed randomized studies for the necessity of cervical cerclage difficult to execute. Crude estimates of the number of cervical cerclage procedures per deliveries have ranged from 0.05% to 1.84%, with a more recent estimate of 0.46% being noted.1–3 This raises the following questions: Does the type of practice setting influence the use of cervical cerclage, and is cervical cerclage being overused in some venues or underused in others? It has been perceived that a higher rate of interventions, such as cesarean delivery, occurs in private practice obstetric settings compared with academic institutions.4 Our objective was to compare the use of cervical cerclage procedure between a community hospital and a university teaching hospital to estimate 1) if the rate of cervical cerclage differs, and 2) whether indications for cervical cerclage placement vary by institution.
MATERIALS AND METHODS
A comparison was performed between two hospitals, geographically only 1½ miles apart. Memorial Hermann Hospital is the primary teaching hospital for the University of Texas Health Science Center at Houston (here defined as the “university hospital”), which has an obstetrics and gynecology residency program. The Woman's Hospital of Texas is a community hospital with private physicians and no obstetric residents (here defined as the “community hospital”). Both hospitals have level III nurseries and maternal–fetal medicine physicians on staff, and both accept maternal-fetal transfers. A computerized search of medical records dated January 1, 2000, through December 31, 2004, was performed at both institutions to identify the diagnostic code “cervical incompetence” and the procedure codes “cerclage of cervix, during pregnancy; vaginal/abdominal.”5,6 Approval was obtained from both hospital institutional review boards.
Through medical record review, sociodemographic variables were extracted. Gravidity, estimated gestational age at placement of cerclage, and whether an obstetrician-gynecologist or a maternal–fetal medicine specialist placed the cerclage were noted. Cerclage was defined as elective if it was a planned procedure. This included both prophylactic and therapeutic cerclage. A prophylactic cerclage is a stitch inserted in asymptomatic women who are at risk of preterm delivery based on previous obstetric risk factors.7 A therapeutic cerclage is a stitch inserted in asymptomatic women in whom a short cervix has been detected by ultrasound or digital vaginal examination.7 Emergency or rescue cerclage describe a stitch inserted after advanced cervical dilation or bulging membranes through the cervix. The indication for the cerclage was the primary indication noted on the operative report by the physician. If it was a repeat cerclage it was classified as such, regardless of the reason for the previous cerclage. Categories for cerclage indications were as follows: one previous second-trimester loss, two or more previous second-trimester losses, previous cone biopsy or loop electrocautery excision procedure (LEEP), short cervix (either by digital or ultrasound examination), previous preterm delivery, multiple gestations, müllerian anomaly, in utero exposure to diethylstilbestrol, multiple terminations of pregnancy, cervical laceration, nonpregnant diagnosis (indicated by previous ease of dilation of the cervix with a dilator), or unknown. A combined category of advanced cervical dilation, funneling, or membranes at internal os of the cervix was done because these patients typically had two or more of these indications. For first-time cerclage placement, if two indications were documented on the operative report, the indication for which the cerclage was inserted was assumed to be the primary indication. For example, if a patient had a diagnosis of short cervix and a second diagnosis such as history of a previous second-trimester pregnancy loss, cone biopsy/LEEP, or multiple gestations, then the cerclage was classified as being placed for short cervix. The estimated gestational age at removal of cerclage, estimated gestational age at delivery, and birth weight were recorded. The incidence of cerclage was estimated as the number of cerclages per total deliveries over the 5-year study period.
The largest population estimate of cervical insufficiency from the medical literature reports an incidence of 0.46%.3 To detect a 50% difference in the proportion of cervical cerclage at P<.05 and a power of 80%, a sample size of 80 cerclages from each institution is needed. Data were analyzed by Student t test and χ2, as appropriate. Binary logistic regression was used to examine the likelihood that patients with a specific cerclage indication would be more likely to be found in a community hospital than in a university hospital setting. P<.05 was considered statistically significant. Analysis was performed using True Epistat 5.3 (Epistat Services, Richardson, TX) and SPSS 14.0 (SPSS Inc, Chicago, IL).
For the 5-year study period, 41,458 and 17,153 deliveries occurred at the community hospital and the university teaching hospital, respectively. Of the total deliveries, those occurring before 37 weeks of gestation were 15.5% (n=6,444) at the community hospital and 22.7% (n=3,892) at the university teaching hospital, P<.01. There were six maternal–fetal medicine specialists at the community and five at the university hospital. The groups were distinct, and no physician placed a cerclage at both hospitals. Cerclages were more often placed at the community hospital (2.6%, n=1,076) than at the university hospital (0.57%, n=98), P<.01, (unadjusted odds ratio [OR] 4.6, 95% confidence interval [CI] 3.8–5.7). The majority of cerclages at both hospitals were performed by a vaginal technique, 98.3% compared with 100% in the community hospital compared with the university hospital, respectively. The preponderance of the vaginal cerclages were performed using the McDonald technique, with a greater number being placed at the community hospital (99.7%) than at the university hospital (85.7%), P<.01. No difference was noted in the percentage of repeat cerclage between the community hospital and the university hospital. However, more elective cerclages were placed (94.6% versus 83.7%) at the community hospital than at the university hospital, P<.01. There were a larger number of emergency cerclages (16.3% versus 5.4%) placed at the university hospital than at the community hospital, P<.01. More maternal–fetal medicine physicians placed cerclages at the university hospital than at the community hospital setting (72.4% versus 39.6%), P<.01.
Characteristics of women receiving a cerclage between both hospitals are shown in Table 1. Women at the community hospital were older, had private insurance, a race designation of “white,” and were nulliparous, P<.05. Cerclages were placed earlier, and removed at a later estimated gestational age in the community hospital, P<.01. The most common indications for cerclages are shown in Table 2. Significantly more cerclages were placed at the university hospital, with the diagnoses of advanced cervical dilation, funneling, or membranes at the internal cervical os and for a history of two or more previous second-trimester losses, P<.01. More cerclages were placed at the community hospital for a history of cervical cone biopsy/LEEP, a diagnosis of short cervix, or for multiple gestations, P<.05. No statistical differences were noted for any of the other indications for cervical cerclage.
Results of the multivariate logistic regression model that evaluated the odds that patients with specific cerclage indications would be found in a specific hospital setting (university or community), while adjusting for the confounding effects of age, ethnicity, and insurance type, are displayed in Table 3. The type of cerclage indication that was attributed to each patient was the fourth variable entered into the regression equation, after age, ethnicity, and type of insurance. All four variables were regressed on the type of hospital setting, the group membership dependent variable. Women with the cerclage indication of advanced cervical dilation, funneling, or membranes at the internal cervical os were equally likely to be found in both hospital settings. Analysis revealed that there was an increase in the odds of being found in a community hospital of 4% with the cerclage indication of cone biopsy/LEEP, 19% with the cerclage indication of short cervix, and 12% with the cerclage indication of multiple gestations, P<.01. Because logistic regression requires that categorical independent variables specify a reference group within such variables, the odds of the reference group are not estimated. A recode of the reference group in the logistic analysis tested for the likelihood that two or more previous second-trimester losses would be found more likely in one hospital setting. The results of the analysis (P>.05) revealed that this group was not significant in distinguishing between the two hospital settings (ie, it was equally likely that two or more previous second-trimester losses would be found in both hospital settings).
The term cervical insufficiency has been used to describe the inability of the uterine cervix to retain a pregnancy in the absence of contractions or labor.8 Consequently, cervical insufficiency is described as a risk factor for recurrent second-trimester loss and early preterm delivery. Historically, pregnancy losses in the second or early third trimester with cervical dilation without contractions were considered sufficient to make the diagnosis.9 Various etiologies have been postulated for cervical insufficiency, such as trauma to the cervix, deficiencies in cervical collagen and elastin, congenital müllerian anomalies, and in utero exposure to diethylstilbestrol.8 Other evidence has suggested a relationship between cervical length in pregnancy and previous obstetric history.10 Patients with a short cervix are at increased risk for a second-trimester pregnancy loss or spontaneous preterm delivery. Thus, a short cervix could be considered as the expression of a spectrum of cervical disease or function.11 However, data confirming all of these associations are limited, and the role of prophylactic or therapeutic cervical cerclage in women for the prevention of second-trimester abortion or preterm delivery is unclear.7
In an attempt to describe obstetrician's preference for cerclage placement, physicians were presented various clinical scenarios under which practitioners would or would not place a cervical cerclage.12 For singleton pregnancies, respondents were most likely to recommend a cervical cerclage, and were most sure of their decision to do so, at earlier gestational ages (15–19 weeks), with a very short cervix (less than 1.5 cm), and with risk factors for preterm birth. However, this study demonstrated uncertainty among practicing obstetricians regarding the decision to place a cervical cerclage. It has been suggested that obstetricians are hesitant to do nothing in a situation where the risk of very early preterm birth is high.12 Yet, until more data are available, it has been recommended that cervical cerclage should not be offered to women considered at low or medium risk of second-trimester miscarriage or extreme preterm labor.7
Results from the current study suggest that obstetricians in a private practice setting use cervical cerclage more frequently than their colleagues in academic practice. Cervical cerclage was used more frequently in the community hospital for a previous cone biopsy/LEEP, for the diagnosis of short cervix, or for multiple gestations. Women with a previous cone biopsy or LEEP have not been shown to have a higher rate of preterm birth at less than 34 weeks of gestation when compared with low-risk controls.13 Women with a cone biopsy with or without cervical cerclage show no difference in the rate of delivery at less than 35 weeks of gestation.14,15 Contradictory results have been obtained from trials of therapeutic cerclage used to prevent preterm delivery in women with ultrasonographic findings of short cervix.15–18 Women with a cervical length of less than 25 mm have been shown to have some benefit from cerclage placement in some, but not all, populations.16,17 It has been suggested that either patient selection or small sample sizes may have caused the conflicting results in these two trials.17,18 In contrast, a cervical length of 15 mm or less is a predictor of cerclage failure.15,18 It follows that, because some benefit occurs with a cerclage with a cervical length less than 25 mm, waiting until a shorter cervical length develops before offering cerclage many not be essential.15 Prophylactic cerclage placement in twin or triplet pregnancies has not resulted in improved pregnancy outcomes.15,19 In twins, cerclage has been associated with a relative risk of greater than 2 for preterm birth at less than 35 weeks gestational age.15 It has been suggested that there may be a role for cervical cerclage in women considered “at very high risk” of second-trimester loss due to a cervical factor.7 Cerclage may have some beneficial effect in certain populations of pregnant women.14,15 The largest benefit was noted for pregnancies characterized by three or more second-trimester miscarriages or preterm deliveries.14 Similar benefits for cerclage were noted in a meta-analysis for groups of women with a prior preterm birth or second-trimester loss.15
Our study is limited by several factors. The study is retrospective and is restricted by the possibility of bias in choosing candidates for cervical cerclage. As indicated in Table 1, the populations of women who received a cerclage differ between the two institutions. Information on women who delivered during the study period at both institutions (greater than 58,000 deliveries), but who did not have a cerclage, were not collected. This precludes us from comparing between the populations confounding variables that may have impacted the crude incidence of cerclage. Thus, this finding should be interpreted cautiously. Risk factors predisposing women to cervical insufficiency were not examined and may have differed between the two populations. However, indications for first-time cerclage were based on risk factors for cervical insufficiency, and the majority of these indications did not differ between the two institutions. In addition, confounding variables between subjects in the study groups were controlled for by multivariate analysis. A larger number of women at less than 37 weeks of gestation delivered at the university hospital setting, suggesting a higher-risk population for preterm delivery, which in theory should have favored more cerclages in the university location. Then again, interpretation of this observation could be that the higher preterm birth rate at the university hospital may partly be attributable to an underuse of cerclage. It does not appear that cerclage was underused at the university setting because the overall rate of cerclage placement (0.57%) is similar to the best estimate of cervical insufficiency (0.46%) reported from a large database of more than half a million deliveries.3 Another limitation of our findings is the number of statistical comparisons done. Our sample size analysis was for the crude rate of cerclage placement between the two institutions. Subgroup analysis of the various cerclage indications results in insufficient sample size to detect small statistical differences. Because a P value of less than .05 could occur by chance once in 20 comparisons, the multiple comparisons from Table 2 should be interpreted with caution. Finally, our findings may be unique to the institutions compared. With the differences between institutions in interpreting and applying criteria justifying the diagnosis of cervical insufficiency, variations in cerclage use with other populations or institutions may occur.
With evidence suggesting that cerclage placement for a previous cone biopsy or LEEP, cervical length of less than 15 mm, or multiple gestations does not prevent preterm birth and may be associated with a higher incidence of preterm delivery, efforts should be focused on updating obstetricians in the community setting about effective use of cervical cerclage.14,15,18 The large number of nulliparous women in the current study who receive a prophylactic cerclage in the community setting initiates the cycle of future repeat cerclages. Obstetricians face a dilemma when caring for a pregnant woman who has had a cerclage in a previous pregnancy due to “cervical insufficiency.” The decision to repeat the cerclage, especially when the original indication is uncertain, is often an emotional one.20 Instruction and quality assurance for indications for cervical cerclage placement may be of benefit to practicing obstetricians. A prospective randomized trial for optimal indications for cervical cerclage is desperately needed.
1. Barter RH, Dusbabek JA, Riva HL, Parks J. Surgical closure of the incompetent cervix during pregnancy. Am J Obstet Gynecol 1958;75:511–21.
2. Toaff R, Toaff ME, Ballas S, Ophir A. Cervical incompetence: diagnostic and therapeutic aspects. Isr J Med Sci 1977;13:39–49.
3. Lidegaard Ø. Cervical incompetence and cerclage in Denmark 1980–1990. Acta Obstet Gynecol Scand 1994;73:35–8.
4. Cyr RM. Myth of the ideal cesarean section rate: commentary and historic perspective. Am J Obstet Gynecol 2006;194:932–6.
5. International classification of diseases, 9th revision. 6th ed. Salt Lake City (UT): Ingenix; 2002.
6. Current procedural terminology. 4th ed. Chicago (IL): American Medical Association Press; 2002.
7. Drakeley AJ, Roberts D, Alfirevic Z. Cervical stitch (cerclage) for preventing pregnancy loss in women (Cochrane Review). In: The Cochrane Library, Issue 2, 2004. Oxford: Update Software.
8. Harger JH. Cervical insufficiency. ACOG Practice Bulletin No. 48. American College of Obstetricians and Gynecologists. Obstet Gynecol 2003;102:1091–9.
9. Harger JH. Comparison of success and morbidity in cervical cerclage procedures. Obstet Gynecol 1980;56:543–8.
10. Iams JD, Goldenberg RL, Meis P, Mercer BM, Moawad A, Das A, et al. The length of the cervix and the risk of spontaneous premature delivery N Engl J Med 1996;334:567–72.
11. Iams JD, Johnson FF, 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;172:1097–106.
12. Pramod R, Okun N, McKay D, Kiehn L, Hewson S, Ross S, et al. Cerclage for the short cervix demonstrated by transvaginal ultrasound: current practice and opinion. J Obstet Gynaecol Can 2004;26:564–70.
13. Crane JMG, Delaney T, Hutchens D. Transvaginal ultrasonography in the prediction of preterm birth after treatment for cervical intraepithelial neoplasia. Obstet Gynecol 2006;107:37–44.
14. Final report of the Medical Research Council/Royal College of Obstetricians and Gynaecologists multicentre randomised trial of cervical cerclage. MRC/RCOG Working Party on Cervical Cerclage. Br J Obstet Gynaecol 1993;100:516–23.
15. Berghella V, Odibo AO, To MS, Rust OA, Althuisius SM. Cerclage for short cervix on ultrasonography: meta-analysis of trials using individual patient-level data. Obstet Gynecol 2005;106:181–9.
16. 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 dilation of the internal os. Am J Obstet Gynecol 2000;183:830–5.
17. Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van Geijn HP. Final results of the Cervical Incompetence Prevention Randomized Cerclage Trial (CIPRACT): therapeutic cerclage with bed rest versus bed rest alone. Am J Obstet Gynecol 2001;185:1106–12.
18. To MS, Alfirevic Z, Heath VCF, Cicero S, Cacho AM, Williamson PR, et al. Cervical cerclage for prevention of preterm delivery in women with short cervix: randomized controlled trial. Lancet 2004;363:1849–53.
19. Rebarber A, Roman AS, Istwan N, Rhea D, Stanziano G. Prophylactic cerclage in the management of triplet pregnancies. Am J Obstet Gynecol 2005;193:1193–6.
20. Fejgin MD, Gabai B, Goldberger S, Ben-Nun I, Beyth Y. Once a cerclage, not always a cerclage. J Reprod Med 1994;39:880–2.
© 2007 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
Figure. No caption available.