- Email: [email protected]
Risks of antepartum cervical examination in multifetal gestations
Risks of antepartum cervical examination in multifetal gestations Harold A. Bivins, Jr., MD, Roger B. Newman, MD, Janna M. Ellings, BSN, CNM, Thomas C. Hulsey, SeD, and Andrea Keenan, BSPH Charleston, South Carolina OBJECTIVE: This study was performed to determine the risks associated with antepartum cervical examination in women with multifetal gestations attending a specialized antepartum Twin Clinic. STUDY DESIGN: Patients followed up in a specialized Twin Clinic from 1988 through 1991 (N = 89) received extensive preterm birth prevention education and routine cervical examination at each clinic visit. Obstetric outcome was compared between these patients and 288 other twin gestations followed up in the high-risk obstetric clinics between 1981 and 1991. In these control twin gestations cervical examination was done for obstetric indications only. RESULTS: Comparisons between the 89 Twin Clinic attendees and the 288 controls followed up in the high-risk obstetric clinic revealed no significant differences in patient demographics, medical complications, or infectious morbidity. Importantly, premature rupture of membranes was significantly less frequent in the Twin Clinic group, who had 7.6 ± 3.2 (mean ± SO) cervical examinations per patient. CONCLUSION: Routine cervical examination in multifetal gestations is not associated with increased rates of premature rupture of membranes or other obstetric complications. (AM J OBSTET GVNECOL 1993; 169:22-5.)
Key words: Twin gestation, cervical examination, premature rupture of membranes, twin clinic Routine antepartum cervical examination has been extensively investigated to assess its ability to predict premature labor. \-5 The predictive value of antepartum cervical examination has also been previously reported in multifetal gestations. 2 • 3 However, there is little information regarding the possible detrimental effects of routine cervical examination on the initiation of preterm labor, premature rupture of membranes, bleeding, or other infectious complications in the multifetal gestation. This study will report the frequency of these complications in a group of twin gestations receiving routine antepartum cervical examination and a second group receiving cervical examination for obstetric indications only.
Material and methods Between 1981 and 1991, 377 women with twin gestations who met inclusion criteria were followed up prenatally in the obstetric clinics of the Medical University of South Carolina. In 1988 a specialized antepartum twin clinic was established for women with multifetal
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Medical University of South Carolina. Received for publication October 6, 1992; revised December 23, 1992; accepted December 30, 1992. Reprint requests: Harold A. Bivins, Jr., MD, Department of Obstetrics and Gynecology, Medical University of South Carolina, 171 Ashley Ave., Charleston, SC 29425. Copyright © 1993 by Mosby-Year Book, 1nc. 0002-9378/93 $1.00 + .20 6/1/45203
22
gestations. This clinic was staffed by a single certified nurse-midwife providing primary patient care with supervision by members of the maternal-fetal medicine section. Since its inception, 89 women with twin gestations have been followed up through delivery. The other 288 twin gestations were followed up in the high-risk obstetric clinic, which is staffed primarily by obstetric residents and faculty. The majority of these twin gestations were followed up before the opening of the Twin Clinic in 1988 (n = 237). Fifty-one twin gestations were followed up in the high-risk obstetric clinics since the opening of the Twin Clinic. These gestations include some twins who were private obstetric patients of university-based faculty; however, because the Twin Clinic meets only on a single day each week, the majority of patients not referred to the Twin Clinic were unable to attend as a result of transportation or other logistic difficulties. General antepartum management guidelines for patients attending the Twin Clinic have been previously reported." The frequency of antepartum visits for twin gestations in both clinics was established at intervals of every I or 2 weeks beyond 20 weeks' gestation. Twins followed in the high-risk obstetric clinic underwent antepartum cervical examination for obstetric indications only. At each antepartum visit to the Twin Clinic a digital examination was performed to evaluate cervical status. These cervical examinations were performed with sterile gloves and KY lubricating jelly. The internal cervical os was palpated if possible, but the examining finger was not passed beyond the point at which cervical
Bivins et al.
Volume 169. Number I Am J Obstet Gynecol
23
Table I. Demographic characteristics of patients in the Twin Clinic (N = 89) and high-risk obstetric clinics (N = 288) Twin Clinic No.
Parity 0 1-3 4-7 Age (yr) <20 20-29 30+ Race Black White Inadequate prenatal care*
1
High-risk obstetric clinic %
No.
19 62 8
21.4 69.7 9.0
11 55 23 59 30 2
I
%
Significance
91 174 22
31.7 60.4 7.7
P = 0.084 P = 0.147 P = 0.852
12.4 61.8 25.8
61 167 60
21.2 58.0 20.8
P = 0.090 P = 0.606 P = 0.395
66.3 33.7 2.3
185 103 20
64.2 35.8 6.9
P = 0.820 P = 0.820 P = 0.163
*Based on the Prenatal Care Utilization Index. 7
resistance was identified. These cervical examinations were performed to obtain a cervical score. The cervical score represents the cervical dilatation in centimeters at the internal os subtracted from the cervical length in centimeters. The identification of a cervical score :5 0 before 34 weeks' gestation has been associated with an increased risk of pre term delivery in multifetal gestations. 2 • 3 Betamethasone for possible acceleration of fetal lung maturity was given prophylactically in the Twin Clinic to women whose cervical score was :5 0 before 34 weeks' gestation. Twin Clinic attendees underwent an average of 7.6 ± 3.2 (mean ± SD) cervical examinations each. By me.ans of the Medical University of South Carolina Perinatal Information Network the outcomes of the 89 twin gestations followed up in the Twin Clinic since 1988 were compared with those of the 288 twin gestations delivered at the Medical University Hospital since 1981 who did not attend the Twin Clinic. All patients were followed up in the antepartum obstetric clinics of the Medical University of South Carolina. Maternal transports were excluded, as were patients who failed to obtain prenatal care. Obstetric and neonatal data were extracted from antepartum and intrapartum records for inclusion in the computerized perinatal data base. Gestational age assessment was based on the date of the last menstrual period when confirmed by obstetric ultrasonography. When obstetric ultrasonography did not confirm menstrual dating, ultrasonography was used to establish gestational age and estimate the date of delivery. Preterm delivery was defined as delivery before 37 completed weeks of gestation. Premature rupture of membranes was defined as rupture of the amniotic sac > 1 hour before the onset of uterine contractions. Third-trimester bleeding was defined as the presence of unexplained vaginal bleeding before or during delivery, clinical evidence of placenta previa, or abruptio placentae. The adequacy of prenatal care was evaluated by means of a modification of the Prenatal Care Utilization Index. 7
Gestational age and birth weight were expressed as mean ± SD. Analyses of the data were performed with the Student t test, X2 test with Yates' correction, and the Fisher exact test. Significance was established at p < 0.05. Results
The demographic characteristics of the 377 twin gestations are presented in Table I. There were no significant demographic differences between the 89 twin gestations followed up in the Twin Clinic and the 288 followed up in the high-risk obstetric clinic. There were also no significant differences between the two groups in terms of inadequate prenatal care. The two groups also demonstrated no significant differences in the frequency of cesarean birth or the frequency of various medical complications of pregnancy (Table II). Gestational age at delivery was similar for both groups (35.6 ± 2.9 vs 35.0 ± 4.2 weeks, p = 0.204). The average birth weight of twin A in the Twin Clinic group was 2355 ± 535 gm and in the high-risk clinic group 2131.9 ± 730 gm (p = 0.002). The average birth weight for twin B in the Twin Clinic group was 2304.2 ± 527 gm and in the high-risk clinic group was 2094.0 ± 699 gm (p = 0.003). Obstetric complications for each group are presented in Table III. Routine antepartum cervical examination was not associated with any increase in obstetric complications potentially related to this procedure. There was significantly less premature rupture of membranes among the patients attending the Twin Clinic (12.4% vs 23.6%, P = 0.03). Women attending the Twin Clinic were less than half as likely to experience premature rupture of membranes compared with women attending the high-risk obstetric clinic (odds ratio 0.456, 95% confidence interval 0.229 to 0.907). Comment
As long ago as 1965 8 routine antepartum examination of the cervix was reported to help physicians detect
24
Bivins et al.
July 1993 Am J Obstet Gynecol
Table II. Comparison of medical and surgical complications of twin pregnancy Twin Clinic (N No.
Mode of delivery Vaginal Cesarean Preeclampsia Diabetes Anemia Substance abuse
= 89)
I
60 29 10 6 17 3
High-risk obstetric clinic (N
%
No.
67.4 32.6 11.2 6.7 19.1 3.4
195 93 41 12 44 9
I
= 288) %
Significance
P= P= P= P= P= P=
67.7 32.3 14.2 4.2 15.3 3.1
0.938 0.938 0.585 0.477 0.489 0.568
Table III. Pregnancy complications in twin gestations Twin Clinic (N = 89) No.
Preterm delivery Premature rupture of membranes Bleeding Intraamniotic infection Neonatal sepsis
69 11
2
3 1
1
High-risk obstetric clinic (N = 288)
%
No.
77.5 12.4 2.3 3.4 1.1
192 68 16 12 21
1
%
Significance
66.7 23.6 5.6 4.2 7.3
p = 0.071 P = 0.033*
P= P=
P=
0.160 0.511 0.056
*Statistically significant.
patients at increased risk of premature labor. Routine antepartum cervical examination has been integrated into many preterm birth prevention programs and has been shown to be a clinically useful predictor of patients at increased risk for preterm delivery.9.16 Unfortunately, possible detrimental effects of routine cervical examination on initiation of preterm labor, premature rupture of membranes, bleeding, and infectious morbidity is only superficially addressed in most of these reports and is especially lacking for multifetal gestations. Lenihan l ? reported that weekly cervical examinations beginning at 37 weeks' gestation significantly increased the incidence of premature rupture of membranes. He also reported a trend toward a higher incidence of primary cesarean section in the group receiving routine cervical examination with premature rupture of membranes. In a larger prospective study McDuffie et a1. 18 found no association between weekly cervical examinations at term and premature rupture of membranes or any other infectious morbidity. McColgin et a1. 19 have recently reported that "membrane stripping" at term results in a lower incidence of postdate pregnancies and an earlier onset of spontaneous labor without any increase in the incidence of premature rupture of membranes or infection. There is less information regarding the risk of preterm cervical examination. Holbrook et a1. 20 reviewed the findings of the San Francisco preterm birth prevention program and reported no increase in the rates of premature rupture of membranes, intraamniotic infection, or endometritis compared with a large control group not receiving routine antepartum cervical exam-
ination. In fact, there was a trend toward a lower incidence of premature rupture of membranes in the group receiving weekly cervical examinations. Our data confirm this trend by demonstrating a significant reduction in the frequency of premature rupture of membranes with routine cervical examination. The reason for this reduction in premature rupture of membranes is unknown, but it is hypothesized that it may be a result of heightened cervical surveillance and the earlier detection and treatment of preterm labor. The early detection and suppression of occult uterine activity may help avert the complication of preterm premature rupture of membranes. When the obstetric outcome of those women attending the Twin Clinic was compared with a contemporary control group followed without routine antepartum cervical examination, there were significant reductions in the rates of very early ( < 30 weeks' gestation) and very-Iow-birth-weight « 1500 gm) deliveries because of either preterm labor or premature rupture of membranes." This finding is consistent with the experience of several successful local and regional preterm birth prevention programs that have all emphasized routine antepartum cervical examinations.9. II, 13 Cervical examination has been shown to increase uterine activity.21 This is thought to be caused by increased prostaglandin release. 22 Therefore a reasonable postulate would be that more frequent cervical examination would increase the rate of preterm delivery, especially in a group of women already at high risk. Our data do not support this postulate. The preterm delivery rate was comparable between the patients in
Volume 169, Number I Am J Obstet Gynecol
the Twin Clinic and those twin gestations followed up in the high-risk obstetric clinic who had cervical examinations for obstetric indications only. There is also the theoretic risk of introducing bacteria into the cervical canal by digital examination, resulting in complications such as intraamniotic infection, neonatal sepsis, or a change in the cervical microftora that could lead to a higher incidence of postpartum infection. Creatsas et al. 23 have demonstrated a significantly higher incidence of bacterial pathogens in the endocervical canal of patients with very recent premature rupture of membranes compared with unruptured controls. However, we found no increase in the risk of premature rupture of membranes, intraamniotic infection, or neonatal sepsis among the multifetal gestations receiving routine cervical examination. Because all cervical examinations were performed by a single examiner in the Twin Clinic it is conceivable that how the cervical examination was performed may be another important variable, potentially off-setting the presumed adverse effects of multiple examinations. Unfortunately, this is a variable that we could not evaluate in this observational study, but our impression is that the cervical examination technique does not vary significantly among examiners. We believe that the cervical examinations performed in the Twin Clinic were thorough in that we obtained information on both cervical length and dilation at the internal cervical os. Obviously, many, if not most, of the twins followed up in the high-risk obstetric clinic received one or more antepartum cervical examinations. However, because cervical examination was not part of the routine management of multifetal gestations in the high-risk clinic, it must be assumed that antepartum cervical examination was quantitatively more frequent in the Twin Clinic. Because almost all physicians would not hesitate to perform a cervical examination if the obstetric situation required, this study really addresses what incremental risks may be encountered if routine antepartum cervical examination is used. The greatest risk confronting multifetal gestations is preterm delivery. Antepartum cervical examination has been demonstrated to be of predictive value for preterm labor in twin gestations!' 3 However, the risks of repetitive preterm cervical examinations have not been defined in this population to date. Our data support the safety of routine preterm cervical examination in the management of twin gestations.
REFERENCES 1. Creasy RK, Gummer BA, Liggins GC. System for predicting spontaneous preterm birth. Obstet Gynecol 1980;55: 692-5. 2. Newman RB, Godsey RK, Ellings ]M, Campbell BA, Eller DP, Miller MC. Quantification of cervical change: Relationship to preterm delivery in the multifetal gestation. AM ] OBSTET GVNECOL 1991;165:264-71.
Bivins et al.
25
3. Neilson ]P, Verkuyl DAA, Crowther CA, Bannerman C. Preterm labor in twin pregnancies: prediction by cervical assessment. Obstet Gynecol 1988;72:719-23. 4. Bouyer], Papiernik E, Dreyfuss], Collin D, Winisdoerffer B, Guegan S. Maturation signs of the cervix and the prediction of preterm birth. Obstet Gynecol 1986;68:20914. 5. Stubbs TM, Van Dorsten ]P, Miller MC. The preterm cervix and preterm labor: relative risks, predictive values, and change over time. AM] OBSTET GVNECOL 1986;155: 829-34. 6. Ellings ]M, Newman RB, Hulsey TC, Bivins HA, Keenan A. Reduction in very low birth weight deliveries and perinatal mortality in a specialized, multidisciplinary twin clinic. Obstet Gynecol 1993;81:387-91. 7. Alexander GR, Comely DA. Prenatal care utilization: its measurement and relationship to pregnancy outcome. Am ] Prev Med 1987;3:243-53. 8. Wood C, Bannerman RHO, Booth RT, Pinkerton ]HM. The prediction of premature labor by observation of the cervix and external tocography. AM ] OBSTET GVNECOL 1965;91:396-402. 9. Herron MS, Katz M, Creasy PK. Evaluation of a preterm birth prevention program: preliminary report. Obstet Gynecol 1982;59:452-6. 10. O'Connor MC, Arias E, Royston ]P, Dalrymple 1]. The merits of special antenatal care for twin pregnancies. Br] Obstet Gynaecol 1981;88:222-30. 11. Meis P], Ernest ]M, Moore ML, Michielutte R, Sharp PC, Buester PA. Regional program for prevention of premature birth in northwestern North Carolina. AM] OBSTET GVNECOL 1987;157:550-6. 12. Main DM, Gabbe S, Richardson D, et al. Can preterm deliveries be prevented? AM] OBSTET GVNECOL 1985;151: 892-8. 13. Papiernik E, Bouyer ], Dreyfuss ], et al. Prevention of preterm births: a perinatal study in Haguenau, France. Pediatrics 1985;76: 154-8. 14. Houlton MCC, Marivate M, Philpott RH. Factors associated with preterm labor and changes in the cervix before labor in twin pregnancy. Br] Obstet Gynaecol 1982;89: 190-4. 15. Papiernik E, Bouyer ], Collin D, Winisdoerffer G, Dreyfuss ]. Precocious cervical ripening and pre term labor. Obstet Gynecol 1986;67:238-42. 16. Goldenberg RL, Davis RO, Copper RL, Corliss DK, Andrews ]B, Carpenter AH. The Alabama preterm birth prevention project. Obstet Gynecol 1990;75:933-9. 17. Lenihan JP. Relationship of antepartum pelvic examinations to premature rupture of the membranes. Obstet Gynecol 1984;63:33-7. 18. McDuffie R, Osborne L, Nelson G, et al. Effect of routine cervical examinations at term on premature rupture of the membranes: a randomized controlled trial. Obstet Gynecol 1992;79:219-22. 19. McColgin SW, Hampton HL, McCaul ]F, Howard PR, Andrew ME, Morrison ]C. Stripping membranes at term: can it safely reduce the incidence of post-term pregnancies? Obstet Gynecol 1990;76:678-80. 20. Holbrook RH, Falcon], Herron M, Lirette M, Laros RK, Creasy RK. Evaluation of the weekly cervical examination in a preterm birth prevention program. Am ] Perinat 1987;4:240-4. 21. Aubry RH, Pennington ]C. High-risk pregnancy identification. Clin Obstet Gynecol 1973;16:16-8. 22. Mitchell MD, Flint APF, Bibby], et al. Rapid increases in plasma prostaglandin concentrations after the vaginal examination and amniotomy. BM] 1977;2:1183-5. 23. Creatsas G, Pavlatos M, Lolis D, et al. Bacterial contamination of the cervix and premature rupture of the membranes. AM] OBSTET GVNECOL 1981; 139:522-5.
Key words: Twin gestation, cervical examination, premature rupture of membranes, twin clinic Routine antepartum cervical examination has been extensively investigated to assess its ability to predict premature labor. \-5 The predictive value of antepartum cervical examination has also been previously reported in multifetal gestations. 2 • 3 However, there is little information regarding the possible detrimental effects of routine cervical examination on the initiation of preterm labor, premature rupture of membranes, bleeding, or other infectious complications in the multifetal gestation. This study will report the frequency of these complications in a group of twin gestations receiving routine antepartum cervical examination and a second group receiving cervical examination for obstetric indications only.
Material and methods Between 1981 and 1991, 377 women with twin gestations who met inclusion criteria were followed up prenatally in the obstetric clinics of the Medical University of South Carolina. In 1988 a specialized antepartum twin clinic was established for women with multifetal
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Medical University of South Carolina. Received for publication October 6, 1992; revised December 23, 1992; accepted December 30, 1992. Reprint requests: Harold A. Bivins, Jr., MD, Department of Obstetrics and Gynecology, Medical University of South Carolina, 171 Ashley Ave., Charleston, SC 29425. Copyright © 1993 by Mosby-Year Book, 1nc. 0002-9378/93 $1.00 + .20 6/1/45203
22
gestations. This clinic was staffed by a single certified nurse-midwife providing primary patient care with supervision by members of the maternal-fetal medicine section. Since its inception, 89 women with twin gestations have been followed up through delivery. The other 288 twin gestations were followed up in the high-risk obstetric clinic, which is staffed primarily by obstetric residents and faculty. The majority of these twin gestations were followed up before the opening of the Twin Clinic in 1988 (n = 237). Fifty-one twin gestations were followed up in the high-risk obstetric clinics since the opening of the Twin Clinic. These gestations include some twins who were private obstetric patients of university-based faculty; however, because the Twin Clinic meets only on a single day each week, the majority of patients not referred to the Twin Clinic were unable to attend as a result of transportation or other logistic difficulties. General antepartum management guidelines for patients attending the Twin Clinic have been previously reported." The frequency of antepartum visits for twin gestations in both clinics was established at intervals of every I or 2 weeks beyond 20 weeks' gestation. Twins followed in the high-risk obstetric clinic underwent antepartum cervical examination for obstetric indications only. At each antepartum visit to the Twin Clinic a digital examination was performed to evaluate cervical status. These cervical examinations were performed with sterile gloves and KY lubricating jelly. The internal cervical os was palpated if possible, but the examining finger was not passed beyond the point at which cervical
Bivins et al.
Volume 169. Number I Am J Obstet Gynecol
23
Table I. Demographic characteristics of patients in the Twin Clinic (N = 89) and high-risk obstetric clinics (N = 288) Twin Clinic No.
Parity 0 1-3 4-7 Age (yr) <20 20-29 30+ Race Black White Inadequate prenatal care*
1
High-risk obstetric clinic %
No.
19 62 8
21.4 69.7 9.0
11 55 23 59 30 2
I
%
Significance
91 174 22
31.7 60.4 7.7
P = 0.084 P = 0.147 P = 0.852
12.4 61.8 25.8
61 167 60
21.2 58.0 20.8
P = 0.090 P = 0.606 P = 0.395
66.3 33.7 2.3
185 103 20
64.2 35.8 6.9
P = 0.820 P = 0.820 P = 0.163
*Based on the Prenatal Care Utilization Index. 7
resistance was identified. These cervical examinations were performed to obtain a cervical score. The cervical score represents the cervical dilatation in centimeters at the internal os subtracted from the cervical length in centimeters. The identification of a cervical score :5 0 before 34 weeks' gestation has been associated with an increased risk of pre term delivery in multifetal gestations. 2 • 3 Betamethasone for possible acceleration of fetal lung maturity was given prophylactically in the Twin Clinic to women whose cervical score was :5 0 before 34 weeks' gestation. Twin Clinic attendees underwent an average of 7.6 ± 3.2 (mean ± SD) cervical examinations each. By me.ans of the Medical University of South Carolina Perinatal Information Network the outcomes of the 89 twin gestations followed up in the Twin Clinic since 1988 were compared with those of the 288 twin gestations delivered at the Medical University Hospital since 1981 who did not attend the Twin Clinic. All patients were followed up in the antepartum obstetric clinics of the Medical University of South Carolina. Maternal transports were excluded, as were patients who failed to obtain prenatal care. Obstetric and neonatal data were extracted from antepartum and intrapartum records for inclusion in the computerized perinatal data base. Gestational age assessment was based on the date of the last menstrual period when confirmed by obstetric ultrasonography. When obstetric ultrasonography did not confirm menstrual dating, ultrasonography was used to establish gestational age and estimate the date of delivery. Preterm delivery was defined as delivery before 37 completed weeks of gestation. Premature rupture of membranes was defined as rupture of the amniotic sac > 1 hour before the onset of uterine contractions. Third-trimester bleeding was defined as the presence of unexplained vaginal bleeding before or during delivery, clinical evidence of placenta previa, or abruptio placentae. The adequacy of prenatal care was evaluated by means of a modification of the Prenatal Care Utilization Index. 7
Gestational age and birth weight were expressed as mean ± SD. Analyses of the data were performed with the Student t test, X2 test with Yates' correction, and the Fisher exact test. Significance was established at p < 0.05. Results
The demographic characteristics of the 377 twin gestations are presented in Table I. There were no significant demographic differences between the 89 twin gestations followed up in the Twin Clinic and the 288 followed up in the high-risk obstetric clinic. There were also no significant differences between the two groups in terms of inadequate prenatal care. The two groups also demonstrated no significant differences in the frequency of cesarean birth or the frequency of various medical complications of pregnancy (Table II). Gestational age at delivery was similar for both groups (35.6 ± 2.9 vs 35.0 ± 4.2 weeks, p = 0.204). The average birth weight of twin A in the Twin Clinic group was 2355 ± 535 gm and in the high-risk clinic group 2131.9 ± 730 gm (p = 0.002). The average birth weight for twin B in the Twin Clinic group was 2304.2 ± 527 gm and in the high-risk clinic group was 2094.0 ± 699 gm (p = 0.003). Obstetric complications for each group are presented in Table III. Routine antepartum cervical examination was not associated with any increase in obstetric complications potentially related to this procedure. There was significantly less premature rupture of membranes among the patients attending the Twin Clinic (12.4% vs 23.6%, P = 0.03). Women attending the Twin Clinic were less than half as likely to experience premature rupture of membranes compared with women attending the high-risk obstetric clinic (odds ratio 0.456, 95% confidence interval 0.229 to 0.907). Comment
As long ago as 1965 8 routine antepartum examination of the cervix was reported to help physicians detect
24
Bivins et al.
July 1993 Am J Obstet Gynecol
Table II. Comparison of medical and surgical complications of twin pregnancy Twin Clinic (N No.
Mode of delivery Vaginal Cesarean Preeclampsia Diabetes Anemia Substance abuse
= 89)
I
60 29 10 6 17 3
High-risk obstetric clinic (N
%
No.
67.4 32.6 11.2 6.7 19.1 3.4
195 93 41 12 44 9
I
= 288) %
Significance
P= P= P= P= P= P=
67.7 32.3 14.2 4.2 15.3 3.1
0.938 0.938 0.585 0.477 0.489 0.568
Table III. Pregnancy complications in twin gestations Twin Clinic (N = 89) No.
Preterm delivery Premature rupture of membranes Bleeding Intraamniotic infection Neonatal sepsis
69 11
2
3 1
1
High-risk obstetric clinic (N = 288)
%
No.
77.5 12.4 2.3 3.4 1.1
192 68 16 12 21
1
%
Significance
66.7 23.6 5.6 4.2 7.3
p = 0.071 P = 0.033*
P= P=
P=
0.160 0.511 0.056
*Statistically significant.
patients at increased risk of premature labor. Routine antepartum cervical examination has been integrated into many preterm birth prevention programs and has been shown to be a clinically useful predictor of patients at increased risk for preterm delivery.9.16 Unfortunately, possible detrimental effects of routine cervical examination on initiation of preterm labor, premature rupture of membranes, bleeding, and infectious morbidity is only superficially addressed in most of these reports and is especially lacking for multifetal gestations. Lenihan l ? reported that weekly cervical examinations beginning at 37 weeks' gestation significantly increased the incidence of premature rupture of membranes. He also reported a trend toward a higher incidence of primary cesarean section in the group receiving routine cervical examination with premature rupture of membranes. In a larger prospective study McDuffie et a1. 18 found no association between weekly cervical examinations at term and premature rupture of membranes or any other infectious morbidity. McColgin et a1. 19 have recently reported that "membrane stripping" at term results in a lower incidence of postdate pregnancies and an earlier onset of spontaneous labor without any increase in the incidence of premature rupture of membranes or infection. There is less information regarding the risk of preterm cervical examination. Holbrook et a1. 20 reviewed the findings of the San Francisco preterm birth prevention program and reported no increase in the rates of premature rupture of membranes, intraamniotic infection, or endometritis compared with a large control group not receiving routine antepartum cervical exam-
ination. In fact, there was a trend toward a lower incidence of premature rupture of membranes in the group receiving weekly cervical examinations. Our data confirm this trend by demonstrating a significant reduction in the frequency of premature rupture of membranes with routine cervical examination. The reason for this reduction in premature rupture of membranes is unknown, but it is hypothesized that it may be a result of heightened cervical surveillance and the earlier detection and treatment of preterm labor. The early detection and suppression of occult uterine activity may help avert the complication of preterm premature rupture of membranes. When the obstetric outcome of those women attending the Twin Clinic was compared with a contemporary control group followed without routine antepartum cervical examination, there were significant reductions in the rates of very early ( < 30 weeks' gestation) and very-Iow-birth-weight « 1500 gm) deliveries because of either preterm labor or premature rupture of membranes." This finding is consistent with the experience of several successful local and regional preterm birth prevention programs that have all emphasized routine antepartum cervical examinations.9. II, 13 Cervical examination has been shown to increase uterine activity.21 This is thought to be caused by increased prostaglandin release. 22 Therefore a reasonable postulate would be that more frequent cervical examination would increase the rate of preterm delivery, especially in a group of women already at high risk. Our data do not support this postulate. The preterm delivery rate was comparable between the patients in
Volume 169, Number I Am J Obstet Gynecol
the Twin Clinic and those twin gestations followed up in the high-risk obstetric clinic who had cervical examinations for obstetric indications only. There is also the theoretic risk of introducing bacteria into the cervical canal by digital examination, resulting in complications such as intraamniotic infection, neonatal sepsis, or a change in the cervical microftora that could lead to a higher incidence of postpartum infection. Creatsas et al. 23 have demonstrated a significantly higher incidence of bacterial pathogens in the endocervical canal of patients with very recent premature rupture of membranes compared with unruptured controls. However, we found no increase in the risk of premature rupture of membranes, intraamniotic infection, or neonatal sepsis among the multifetal gestations receiving routine cervical examination. Because all cervical examinations were performed by a single examiner in the Twin Clinic it is conceivable that how the cervical examination was performed may be another important variable, potentially off-setting the presumed adverse effects of multiple examinations. Unfortunately, this is a variable that we could not evaluate in this observational study, but our impression is that the cervical examination technique does not vary significantly among examiners. We believe that the cervical examinations performed in the Twin Clinic were thorough in that we obtained information on both cervical length and dilation at the internal cervical os. Obviously, many, if not most, of the twins followed up in the high-risk obstetric clinic received one or more antepartum cervical examinations. However, because cervical examination was not part of the routine management of multifetal gestations in the high-risk clinic, it must be assumed that antepartum cervical examination was quantitatively more frequent in the Twin Clinic. Because almost all physicians would not hesitate to perform a cervical examination if the obstetric situation required, this study really addresses what incremental risks may be encountered if routine antepartum cervical examination is used. The greatest risk confronting multifetal gestations is preterm delivery. Antepartum cervical examination has been demonstrated to be of predictive value for preterm labor in twin gestations!' 3 However, the risks of repetitive preterm cervical examinations have not been defined in this population to date. Our data support the safety of routine preterm cervical examination in the management of twin gestations.
REFERENCES 1. Creasy RK, Gummer BA, Liggins GC. System for predicting spontaneous preterm birth. Obstet Gynecol 1980;55: 692-5. 2. Newman RB, Godsey RK, Ellings ]M, Campbell BA, Eller DP, Miller MC. Quantification of cervical change: Relationship to preterm delivery in the multifetal gestation. AM ] OBSTET GVNECOL 1991;165:264-71.
Bivins et al.
25
3. Neilson ]P, Verkuyl DAA, Crowther CA, Bannerman C. Preterm labor in twin pregnancies: prediction by cervical assessment. Obstet Gynecol 1988;72:719-23. 4. Bouyer], Papiernik E, Dreyfuss], Collin D, Winisdoerffer B, Guegan S. Maturation signs of the cervix and the prediction of preterm birth. Obstet Gynecol 1986;68:20914. 5. Stubbs TM, Van Dorsten ]P, Miller MC. The preterm cervix and preterm labor: relative risks, predictive values, and change over time. AM] OBSTET GVNECOL 1986;155: 829-34. 6. Ellings ]M, Newman RB, Hulsey TC, Bivins HA, Keenan A. Reduction in very low birth weight deliveries and perinatal mortality in a specialized, multidisciplinary twin clinic. Obstet Gynecol 1993;81:387-91. 7. Alexander GR, Comely DA. Prenatal care utilization: its measurement and relationship to pregnancy outcome. Am ] Prev Med 1987;3:243-53. 8. Wood C, Bannerman RHO, Booth RT, Pinkerton ]HM. The prediction of premature labor by observation of the cervix and external tocography. AM ] OBSTET GVNECOL 1965;91:396-402. 9. Herron MS, Katz M, Creasy PK. Evaluation of a preterm birth prevention program: preliminary report. Obstet Gynecol 1982;59:452-6. 10. O'Connor MC, Arias E, Royston ]P, Dalrymple 1]. The merits of special antenatal care for twin pregnancies. Br] Obstet Gynaecol 1981;88:222-30. 11. Meis P], Ernest ]M, Moore ML, Michielutte R, Sharp PC, Buester PA. Regional program for prevention of premature birth in northwestern North Carolina. AM] OBSTET GVNECOL 1987;157:550-6. 12. Main DM, Gabbe S, Richardson D, et al. Can preterm deliveries be prevented? AM] OBSTET GVNECOL 1985;151: 892-8. 13. Papiernik E, Bouyer ], Dreyfuss ], et al. Prevention of preterm births: a perinatal study in Haguenau, France. Pediatrics 1985;76: 154-8. 14. Houlton MCC, Marivate M, Philpott RH. Factors associated with preterm labor and changes in the cervix before labor in twin pregnancy. Br] Obstet Gynaecol 1982;89: 190-4. 15. Papiernik E, Bouyer ], Collin D, Winisdoerffer G, Dreyfuss ]. Precocious cervical ripening and pre term labor. Obstet Gynecol 1986;67:238-42. 16. Goldenberg RL, Davis RO, Copper RL, Corliss DK, Andrews ]B, Carpenter AH. The Alabama preterm birth prevention project. Obstet Gynecol 1990;75:933-9. 17. Lenihan JP. Relationship of antepartum pelvic examinations to premature rupture of the membranes. Obstet Gynecol 1984;63:33-7. 18. McDuffie R, Osborne L, Nelson G, et al. Effect of routine cervical examinations at term on premature rupture of the membranes: a randomized controlled trial. Obstet Gynecol 1992;79:219-22. 19. McColgin SW, Hampton HL, McCaul ]F, Howard PR, Andrew ME, Morrison ]C. Stripping membranes at term: can it safely reduce the incidence of post-term pregnancies? Obstet Gynecol 1990;76:678-80. 20. Holbrook RH, Falcon], Herron M, Lirette M, Laros RK, Creasy RK. Evaluation of the weekly cervical examination in a preterm birth prevention program. Am ] Perinat 1987;4:240-4. 21. Aubry RH, Pennington ]C. High-risk pregnancy identification. Clin Obstet Gynecol 1973;16:16-8. 22. Mitchell MD, Flint APF, Bibby], et al. Rapid increases in plasma prostaglandin concentrations after the vaginal examination and amniotomy. BM] 1977;2:1183-5. 23. Creatsas G, Pavlatos M, Lolis D, et al. Bacterial contamination of the cervix and premature rupture of the membranes. AM] OBSTET GVNECOL 1981; 139:522-5.