- Email: [email protected]
Ultrasonographic assessment of cervical length in triplet pregnancies
Ultrasonographic assessment of cervical length in triplet pregnancies Kirk D. Ramin, MD, Paul L. Ogburn, Jr, MD, Tammy A. Mulholland, RDMS, Robert J. Breckle, RDMS, and Patrick S. Ramsey, MD Rochester, Minnesota OBJECTIVE: Our goal was to evaluate the utility of ultrasonographic assessment of cervical length in the management of triplet pregnancies and to compare these measurements with previously reported data for singleton pregnancies. STUDY DESIGN: The maternal records for all triplet pregnancies managed at the Mayo Medical Center from January 1993–January 1998 were reviewed. Cervical length assessment was undertaken at regular intervals during each pregnancy according to an established real-time transperineal ultrasonographic technique. Presence or absence of cervical funneling was noted at the time of the examination. Obstetric management and outcome data were assessed. RESULTS: Thirty-two triplet pregnancies were managed at our institution between January 1993 and January 1998. Average duration of pregnancy (±SD) was 32.4 ± 2.3 weeks. Progressive cervical shortening was noted with advancing gestational age; average cervical lengths (±SD) were 42.0 ± 5.0 mm at 10 weeks, 37.0 ± 8.0 mm at 20 weeks, 26.0 ± 10.0 mm at 25 weeks, and 21.0 ± 7.0 mm at 30 weeks. Comparison of triplet cervical length measurements with reported data from singleton pregnancies revealed a significant difference between the singleton and triplet data, respectively, at both 24 weeks (35.2 ± 8.3 mm vs 25.0 ± 8.0 mm, P < .001) and 28 weeks (33.7 ± 8.5 mm vs 28.0 ± 11.0 mm, P < .005). Cervical funneling was noted in 3 women with an average of 27 days from onset to delivery. CONCLUSIONS: Ultrasonographic assessment of cervical length is a useful adjuvant in the management of the triplet gestation. Triplet cervical length measurements are significantly different from those reported for gestational age–matched singleton pregnancies. Premature cervical shortening and the presence of cervical funneling are harbingers of premature delivery and should necessitate obstetric intervention. (Am J Obstet Gynecol 1999;180:1442-5.)
Key words: Triplets, multiple gestation, cervical length, ultrasonography, transperineal
Triplet gestations presently constitute 0.1% to 0.3% of all deliveries.1 In spite of the increasing prevalence of triplet pregnancies, relatively few advances have been set forth in the detection and prevention of preterm labor and delivery. Today the average duration of a triplet gestation remains between 32 and 34 weeks’ gestation.2 The resulting prematurity of neonates born at this and earlier gestational ages accounts for a major component of morbidity and mortality in neonates from triplet pregnancies. Several recent advances have shed light on potential avenues for the detection and prevention of preterm labor in patients with multifetal gestations. Both salivary estriol and cervical fetal fibronectin are potential bio-
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Mayo Medical Center. Poster Presentation, presented at the Sixty-sixth Annual Meeting of The Central Association of Obstetricians and Gynecologists, Kansas City, Missouri, October 15-17, 1998. Reprint requests: Kirk D. Ramin, MD, Mayo Medical Center, Charlton 4B, 200 1st St SW, Rochester, MN 55905. Copyright © 1999 by Mosby, Inc. 0002-9378/99 $8.00 + 0 6/6/98244
1442
chemical markers for the identification of patients at risk for preterm delivery.3, 4 Limited data, however, are present regarding the utility of these markers in multiple pregnancies. Perhaps a more important approach with respect to multiple gestations is the utility of cervical length assessment. Digital cervical assessment has been shown to be an effective tool for the evaluation of cervical length and to assess risk for preterm delivery.5, 6 Limitations of this technique, however, include the subjectivity of the assessment and the inability to detect intrinsic cervical shortening or funneling from above. To facilitate a more objective and accurate measure, ultrasonographic assessment of cervical length has been promoted.7, 8 Historically, the transabdominal approach has been compromised by the need for a distended bladder to facilitate visualization of the cervix, which in turn alters the actual cervical length measurement obtained. In contrast, both transvaginal and transperineal ultrasonographic assessment of cervical length have been shown to be effective tools to evaluate cervical length and improve our ability to detect patients at risk for preterm deliv-
Ramin et al 1443
Volume 180, Number 6, Part 1 Am J Obstet Gynecol
Fig 1. Cervical length assessment on triplet pregnancies. Mean cervical length (±SD) progressively shortened as gestational age progressed. Linear regression analyses were performed for mean cervical length, for –1 SD from the mean, and for –2 SD from the mean.
ery.7-9 Whereas data from these investigations are compelling, there is a paucity of data addressing the utility of these approaches in higher-order multiple gestations. Preliminary data with twin pregnancies has suggested a further role for ultrasonographic cervical length assessment.10, 11 To date, few known data exist regarding the normal cervical length changes for triplet pregnancies. Similarly, the use of cervical length assessment for prediction of preterm delivery in triplets has not been evaluated. To address this issue, we reviewed our experience with ultrasonographic surveillance of cervical length in triplet gestations and sought to delineate parameters to assist in identification of those gravid women at risk for preterm delivery. Material and methods We prospectively compiled data from the prenatal records of all women with triplet gestations at the Mayo Medical Center from January 1993–April 1998 to evaluate the cervical length changes during pregnancy. All patients were cared for according to a management plan similar to that outlined by Peaceman et al.12 Briefly, this management plan included extensive antepartum patient education, increased rest in an outpatient setting, abstinence from coitus, weekly prenatal visits beginning at 20 weeks’ gestation with routine ultrasonographic assessment of cervical length, and inpatient bed rest for progressive cervical change. More aggressive therapy (including magnesium sulfate, tocolytic therapy, corticosteroids, ampicillin, or cerclage or a combination of these) was implemented when deemed medically necessary. Delivery was carried out in those pregnancies progressing to 36 weeks. Reliable pregnancy dating criteria were established by either last menstrual period with confirmatory firsttrimester ultrasonography or by the date of embryo transfer in cases of assisted reproduction. Cervical length
Fig 2. Comparison of cervical length measurements—Singleton versus triplet gestations. Mean cervical length (±SD) for triplet gestation differed significantly from the reported data for gestational age—matched singleton pregnancies at both 24 (asterisk, P < .001) and 28 (asterisk, P < .005) weeks’ gestation.6
assessment was undertaken at the initial obstetric visit, at the time of the standard survey scan at 18 weeks, and weekly from 20 weeks until delivery. All ultrasonographic cervical assessments were performed with the use of an established real-time transperineal ultrasonographic technique in use at our institution.8, 9 Transperineal ultrasonography was chosen preferentially given the high degree of correlation with transvaginal assessment and to prevent iatrogenic cervical manipulation. A Toshiba (Toshiba Corp, Tochigi Ken, Japan) model UIBU277A ultrasound machine with a 3.75-MHz curvilinear transducer was used, with the maternal bladder empty and the maternal buttocks slightly elevated. A sagittal section through the cervix was visualized, and cervical length measurements were made in duplicate at each visit by a single examiner (T.A.M.) along with a real-time single observer (K.D.R.). Cervical length was measured from the notch demarcating the internal cervical os along the hypoechoic mucosal apposition to the external os as previously described.7, 8 Funneling was defined as herniation of the amniotic sac through the internal os and into the upper portion of the cervical canal. Routine cesarean delivery was performed on all triplet gestations during this time period when indicated by either failed tocolysis (>4 cm dilatation), fetal jeopardy, severe preeclampsia, or documented fetal lung maturity. Statistical analyses were performed with the Sigma Stat/Sigma Plot (SPSS Inc, Chicago, Ill) statistical package. Statistical analyses included the Student t test, linear regression analysis, and Fisher exact test where appropriate. Results Thirty-two triplet pregnancies were managed at our institution between January 1993 and January 1998. The average duration of pregnancy (±SD) was 32.3 ± 2.3 weeks. The majority of these deliveries occurred as a result of ei-
1444 Ramin et al
June 1999 Am J Obstet Gynecol
Table I. Comparison of triplet cervical length—Early delivery (<33 weeks) versus late delivery (≥33 weeks) Cervical length (mm, ±SD) Gestational age (wk)
Delivery before 33 wk (n = 11)
Delivery beyond 33 wk (n = 6)
10 20 24 28 30 32
41 ± 3 30 ± 3 25 ± 8 21 ± 6 18 ± 6 17 ± 2
41 ± 2 33 ± 11 27 ± 10 27 ± 9 31 ± 2 25 ± 8
in cervical length measurements between the 2 groups before the 20th week of gestation. Comment
Statistical significance P = .50 P = .03* P = .16 P = .04* P = .02* P = .08
*Statistically significant, Student t test.
ther refractory premature labor (31.0%) or preterm premature rupture of membranes (31.0%). Indications for delivery in the remaining patients included severe preeclampsia, nonreassuring fetal surveillance, or mature fetal lung maturity profile. Delivery in all instances was by cesarean section. Nine percent of the patient population had natural triplet gestations whereas the remainder were achieved through assisted reproductive techniques. Of the 32 pregnancies identified, serial data were available for only 21 patients. Additionally, 3 women remain pregnant at the time of manuscript preparation and are currently under evaluation; hence data were not available for assessment of preterm delivery risk on the basis of cervical length in these patients. Progressive cervical shortening was noted with advancing gestational age with average cervical length (±SD) of 42.0 ± 5.0 mm at 10 weeks, 37.0 ± 8.0 mm at 20 weeks, 26.0 ± 10.0 mm at 25 weeks, and 21.0 ± 7.0 mm at 30 weeks (Fig 1). Cervical funneling was noted in 14.3% of women (3/21) with an average of 27 days from onset to delivery. Cerclage was performed in 2 women and was placed before 18 weeks when funneling of membranes was noted and cervical length fell below 2.5 cm. Comparison of triplet cervical length measurements with reported data from singleton pregnancies revealed a significant difference between the singleton and triplet data, respectively, at both 24 weeks (35.2 ± 8.3 mm vs 25.0 ± 8.0 mm, P < .001) and 28 weeks (33.7 ± 8.5 mm vs 28.0 ± 11.0 mm, P < .005)(Fig 2). To evaluate the relationship between cervical length and risk for premature delivery in triplet pregnancies, we arbitrarily selected a cutoff of 33 weeks as a marker for early delivery. This gestational age represented an average for the reported duration of triplet pregnancies as reported in the literature.12 Patient data were separated into 2 groups: (1) delivery before 33 weeks and (2) delivery at or beyond 33 weeks. Cervical length data were computed and reanalyzed. Cervical length was significantly shorter at 20, 28, and 31 weeks in those women who subsequently delivered before 33 weeks as compared with those who delivered at ≥33 weeks (Table I). No significant difference was noted
Premature delivery is a major cause of perinatal morbidity and mortality in women with higher-order multifetal gestations. The incidence of low birth weight or very low birth weight infants resulting from triplet gestations approaches 50% as compared with a mere 2% for singleton pregnancies.13 Similarly, infant mortality is significantly greater in those neonates delivered of triplet gestations (19.4%) than in those from singleton gestations (1%).13 Cerebral palsy among survivors of triplet pregnancies occurs 47 times more often than in singletons.14 Clearly, interventions to detect or prevent premature delivery in this high-risk patient population are of utmost clinical importance. Use of ultrasonographic assessment of cervical length is an important clinical advance that has great potential application to the management of higher-order multiples. Iams et al7 elegantly documented the utility of this tool in the identification of patients with singleton gestations at risk for subsequent preterm delivery. In this series of 2915 gravid women, transvaginal ultrasonographic determination of cervical length was made at 24 weeks’ and again at 28 weeks’ gestation. Relative risk of preterm delivery based on the 24-week cervical length measurements revealed a dramatically increased risk in women with a shortened cervix. Similar findings were observed with the repeated ultrasonographic measurements at 28 weeks’ gestation. These data demonstrate the significance of a shortened cervix with respect to increased risk for preterm delivery. These authors raised the question of a continuum of cervical competence. The intrinsic competence of the cervix is that which is needed to maintain an individual pregnancy to term. Intuitively, this competency would remain constant in an individual woman; however, the load placed on the cervix could be altered (singletons vs triplets, hydramnios, and other factors). Impairment of factors that maintain this competence likely play a role in the development of preterm delivery.15 It is clear then that the cervix of patients with multifetal gestations is subjected to greater stressors and hence may exceed this competence at an earlier gestational age, resulting in premature delivery. This issue highlights the potential importance of ultrasonographic cervical assessment in this particular patient population. Several recent investigations have sought to evaluate this issue in twin gestations. Imseis et al10 evaluated 85 women with twin gestations with ultrasonographic cervical length assessment. Patients who subsequently delivered beyond 34 weeks’ gestation without intervention had a significantly greater cervical length (36.4 ± 5.8 mm) at the screening 24- to 26-week ultrasonographic assessment as compared with both those patients who delivered before 34 weeks (27.4 ± 8.5 mm) and those who
Ramin et al 1445
Volume 180, Number 6, Part 1 Am J Obstet Gynecol
delivered after 34 weeks but required intervention (27.7 ± 0.5 mm) (P < .001) When a cervical length cutoff of >35 mm was used as a marker for a low-risk population, a positive predictive value of 97% and a specificity of 94% were attained. Kushnir et al11 furthered investigation with twins through a comparative ultrasonographic cervical length evaluation between singleton and twin gestations. Similar to our data, this study demonstrated a significant difference between the cervical length of twins as compared with that of gestational age–matched singleton pregnancies.11 The results of our study further confirm the unique nature of the cervix in women with multifetal gestations. Progressive cervical shortening was observed with advancing gestational age. The degree of shortening observed was significantly greater than that previously reported for singleton pregnancies, which is consistent with impairment of cervical competence as a result of the multiple gestation.7 Additionally, crude comparisons between our data and those data from Kushnir et al11 suggest that the cervix in triplet pregnancies has a more rapid decline in length with advanced gestational age than is found in the cervix in twin pregnancies. Unlike the previously mentioned investigations concerning singleton pregnancies where no intervention was performed and thus the natural history of the gestation could be observed, we already were altering care by managing all of our triplet gestations according to a previously developed management scheme.12 Obviously, new standards are needed to establish normal and aberrant cervical changes in the higher-order gestations. Our data provide a basis for these standards. Similar to what has been reported for singleton and twin gestations, premature cervical shortening in triplets, as well as the development of cervical funneling, is a harbinger of premature delivery. Premature cervical shortening was noted in those women at risk for delivery before 33 weeks’ gestation, suggesting impairment of cervical competence as a predominant risk factor for early delivery in patients with triplet gestations. It is interesting that the relative changes noted between the patient cohort that delivered early and those that delivered beyond 32 weeks were not readily apparent until around the 20th week of gestation. These data support the management plan as proposed by Peaceman et al12 for triplets—reduced activity at approximately 20 weeks’ gestation with increased cervical surveillance. Whereas preliminary data have suggested that such a management plan can significantly reduce preterm delivery and improve perinatal outcome in triplet gestations, larger confirmatory series are needed.12 Ultrasonographic cervical length assessment repre-
sents an important advance in our clinical practice. Not only does the utility of cervical length assessment provide evidence of cervical competence or the lack thereof, it also provides us with a unique tool to monitor patients at risk for preterm delivery. Triplet and other higher-order pregnancies are exactly the high-risk situation in which data from ultrasonographic cervical assessment can provide early warning for impending preterm delivery such that appropriate obstetric intervention can be implemented. Future investigations should test potential obstetric interventions for preventing very preterm births in those triplet pregnancies at highest risk (funneling or cervical length <2.5 cm at 23-24 weeks’ gestation).
REFERENCES
1. Mordel N, Ezra Y, Dorembus D, Benshushan A, Schenker JG, Sadovsky E. Triplets are not so rare any more. J Perinat Med 1992;20:117-22. 2. Luke B. The changing patterns of multiple births in the United States: maternal and infant characteristics, 1973 and 1990. Obstet Gynecol 1994;84:101-6. 3. Lockwood CJ, Senyei AE, Dische MR, Casal D, Shah KD, Thung SN, et al. Fetal fibronectin in cervical and vaginal secretions as a predictor of preterm delivery. N Engl J Med 1991;325:669-74. 4. McGregor JA, Jackson GM, Lachelin GCL, Goodwin TM, Artal R, Hastings C, et al. Salivary estriol as risk assessment for preterm labor: a prospective trial. Am J Obstet Gynecol 1995;173:1337-42. 5. Leveno KJ, Cox K, Roark ML. Cervical dilation and prematurity revisited. Obstet Gynecol 1986;68:434-5. 6. Stubbs TM, Van Dorsten, Miller MC. The preterm cervix and preterm labor: relative risks, predictive values, and change over time. Am J Obstet Gynecol 1986;155:829-34. 7. Iams JA, Goldenberg RL, Meis PJ, 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. 8. Richey SD, Ramin KD, Roberts SW, Ramin SM, Cox SM, Twickler DM. The correlation between transperineal sonography and digital examination in the evaluation of the thirdtrimester cervix. Obstet Gynecol 1995;85:745-8. 9. Kurtzman J, Goldsmith LJ, Gall S, Spinnato JA. Transvaginal versus transperineal ultrasound: a blinded comparison in the assessment of cervical length at mid-gestation [abstract]. Am J Obstet Gynecol 1998;178(Pt 2):S15. 10. Imseis HM, Albert TA, Iams JD. Identifying twin gestations at low risk for preterm birth with transvaginal ultrasonographic cervical measurement at 24 to 26 weeks’ gestation. Am J Obstet Gynecol 1997;177:1149-55. 11. Kushnir O, Izquierdo LA, Smith JF, Blankstein J, Curet LB. Transvaginal sonographic measurement of cervical length: evaluation of twin pregnancies. J Reprod Med 1995;40:380-2. 12. Peaceman AM, Dooley SL, Tamura RK, Socol ML. Antepartum management of triplet gestations. Am J Obstet Gynecol 1992;167:1117-20. 13. Luke B, Keith LG. The contribution of singletons, twins and triplets to low birth weight, infant mortality and handicap in the United States. J Reprod Med 1992;37:661-6. 14. Petterson B, Nelson KB, Watson L, Stanley F. Twins, triplets, and cerebral palsy in births in Western Australia in the 1980’s. BMJ 1993;307:1239-43. 15. Craigo SD. Cervical competence and preterm delivery. N Engl J Med 1996;334:595-6.
Key words: Triplets, multiple gestation, cervical length, ultrasonography, transperineal
Triplet gestations presently constitute 0.1% to 0.3% of all deliveries.1 In spite of the increasing prevalence of triplet pregnancies, relatively few advances have been set forth in the detection and prevention of preterm labor and delivery. Today the average duration of a triplet gestation remains between 32 and 34 weeks’ gestation.2 The resulting prematurity of neonates born at this and earlier gestational ages accounts for a major component of morbidity and mortality in neonates from triplet pregnancies. Several recent advances have shed light on potential avenues for the detection and prevention of preterm labor in patients with multifetal gestations. Both salivary estriol and cervical fetal fibronectin are potential bio-
From the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Mayo Medical Center. Poster Presentation, presented at the Sixty-sixth Annual Meeting of The Central Association of Obstetricians and Gynecologists, Kansas City, Missouri, October 15-17, 1998. Reprint requests: Kirk D. Ramin, MD, Mayo Medical Center, Charlton 4B, 200 1st St SW, Rochester, MN 55905. Copyright © 1999 by Mosby, Inc. 0002-9378/99 $8.00 + 0 6/6/98244
1442
chemical markers for the identification of patients at risk for preterm delivery.3, 4 Limited data, however, are present regarding the utility of these markers in multiple pregnancies. Perhaps a more important approach with respect to multiple gestations is the utility of cervical length assessment. Digital cervical assessment has been shown to be an effective tool for the evaluation of cervical length and to assess risk for preterm delivery.5, 6 Limitations of this technique, however, include the subjectivity of the assessment and the inability to detect intrinsic cervical shortening or funneling from above. To facilitate a more objective and accurate measure, ultrasonographic assessment of cervical length has been promoted.7, 8 Historically, the transabdominal approach has been compromised by the need for a distended bladder to facilitate visualization of the cervix, which in turn alters the actual cervical length measurement obtained. In contrast, both transvaginal and transperineal ultrasonographic assessment of cervical length have been shown to be effective tools to evaluate cervical length and improve our ability to detect patients at risk for preterm deliv-
Ramin et al 1443
Volume 180, Number 6, Part 1 Am J Obstet Gynecol
Fig 1. Cervical length assessment on triplet pregnancies. Mean cervical length (±SD) progressively shortened as gestational age progressed. Linear regression analyses were performed for mean cervical length, for –1 SD from the mean, and for –2 SD from the mean.
ery.7-9 Whereas data from these investigations are compelling, there is a paucity of data addressing the utility of these approaches in higher-order multiple gestations. Preliminary data with twin pregnancies has suggested a further role for ultrasonographic cervical length assessment.10, 11 To date, few known data exist regarding the normal cervical length changes for triplet pregnancies. Similarly, the use of cervical length assessment for prediction of preterm delivery in triplets has not been evaluated. To address this issue, we reviewed our experience with ultrasonographic surveillance of cervical length in triplet gestations and sought to delineate parameters to assist in identification of those gravid women at risk for preterm delivery. Material and methods We prospectively compiled data from the prenatal records of all women with triplet gestations at the Mayo Medical Center from January 1993–April 1998 to evaluate the cervical length changes during pregnancy. All patients were cared for according to a management plan similar to that outlined by Peaceman et al.12 Briefly, this management plan included extensive antepartum patient education, increased rest in an outpatient setting, abstinence from coitus, weekly prenatal visits beginning at 20 weeks’ gestation with routine ultrasonographic assessment of cervical length, and inpatient bed rest for progressive cervical change. More aggressive therapy (including magnesium sulfate, tocolytic therapy, corticosteroids, ampicillin, or cerclage or a combination of these) was implemented when deemed medically necessary. Delivery was carried out in those pregnancies progressing to 36 weeks. Reliable pregnancy dating criteria were established by either last menstrual period with confirmatory firsttrimester ultrasonography or by the date of embryo transfer in cases of assisted reproduction. Cervical length
Fig 2. Comparison of cervical length measurements—Singleton versus triplet gestations. Mean cervical length (±SD) for triplet gestation differed significantly from the reported data for gestational age—matched singleton pregnancies at both 24 (asterisk, P < .001) and 28 (asterisk, P < .005) weeks’ gestation.6
assessment was undertaken at the initial obstetric visit, at the time of the standard survey scan at 18 weeks, and weekly from 20 weeks until delivery. All ultrasonographic cervical assessments were performed with the use of an established real-time transperineal ultrasonographic technique in use at our institution.8, 9 Transperineal ultrasonography was chosen preferentially given the high degree of correlation with transvaginal assessment and to prevent iatrogenic cervical manipulation. A Toshiba (Toshiba Corp, Tochigi Ken, Japan) model UIBU277A ultrasound machine with a 3.75-MHz curvilinear transducer was used, with the maternal bladder empty and the maternal buttocks slightly elevated. A sagittal section through the cervix was visualized, and cervical length measurements were made in duplicate at each visit by a single examiner (T.A.M.) along with a real-time single observer (K.D.R.). Cervical length was measured from the notch demarcating the internal cervical os along the hypoechoic mucosal apposition to the external os as previously described.7, 8 Funneling was defined as herniation of the amniotic sac through the internal os and into the upper portion of the cervical canal. Routine cesarean delivery was performed on all triplet gestations during this time period when indicated by either failed tocolysis (>4 cm dilatation), fetal jeopardy, severe preeclampsia, or documented fetal lung maturity. Statistical analyses were performed with the Sigma Stat/Sigma Plot (SPSS Inc, Chicago, Ill) statistical package. Statistical analyses included the Student t test, linear regression analysis, and Fisher exact test where appropriate. Results Thirty-two triplet pregnancies were managed at our institution between January 1993 and January 1998. The average duration of pregnancy (±SD) was 32.3 ± 2.3 weeks. The majority of these deliveries occurred as a result of ei-
1444 Ramin et al
June 1999 Am J Obstet Gynecol
Table I. Comparison of triplet cervical length—Early delivery (<33 weeks) versus late delivery (≥33 weeks) Cervical length (mm, ±SD) Gestational age (wk)
Delivery before 33 wk (n = 11)
Delivery beyond 33 wk (n = 6)
10 20 24 28 30 32
41 ± 3 30 ± 3 25 ± 8 21 ± 6 18 ± 6 17 ± 2
41 ± 2 33 ± 11 27 ± 10 27 ± 9 31 ± 2 25 ± 8
in cervical length measurements between the 2 groups before the 20th week of gestation. Comment
Statistical significance P = .50 P = .03* P = .16 P = .04* P = .02* P = .08
*Statistically significant, Student t test.
ther refractory premature labor (31.0%) or preterm premature rupture of membranes (31.0%). Indications for delivery in the remaining patients included severe preeclampsia, nonreassuring fetal surveillance, or mature fetal lung maturity profile. Delivery in all instances was by cesarean section. Nine percent of the patient population had natural triplet gestations whereas the remainder were achieved through assisted reproductive techniques. Of the 32 pregnancies identified, serial data were available for only 21 patients. Additionally, 3 women remain pregnant at the time of manuscript preparation and are currently under evaluation; hence data were not available for assessment of preterm delivery risk on the basis of cervical length in these patients. Progressive cervical shortening was noted with advancing gestational age with average cervical length (±SD) of 42.0 ± 5.0 mm at 10 weeks, 37.0 ± 8.0 mm at 20 weeks, 26.0 ± 10.0 mm at 25 weeks, and 21.0 ± 7.0 mm at 30 weeks (Fig 1). Cervical funneling was noted in 14.3% of women (3/21) with an average of 27 days from onset to delivery. Cerclage was performed in 2 women and was placed before 18 weeks when funneling of membranes was noted and cervical length fell below 2.5 cm. Comparison of triplet cervical length measurements with reported data from singleton pregnancies revealed a significant difference between the singleton and triplet data, respectively, at both 24 weeks (35.2 ± 8.3 mm vs 25.0 ± 8.0 mm, P < .001) and 28 weeks (33.7 ± 8.5 mm vs 28.0 ± 11.0 mm, P < .005)(Fig 2). To evaluate the relationship between cervical length and risk for premature delivery in triplet pregnancies, we arbitrarily selected a cutoff of 33 weeks as a marker for early delivery. This gestational age represented an average for the reported duration of triplet pregnancies as reported in the literature.12 Patient data were separated into 2 groups: (1) delivery before 33 weeks and (2) delivery at or beyond 33 weeks. Cervical length data were computed and reanalyzed. Cervical length was significantly shorter at 20, 28, and 31 weeks in those women who subsequently delivered before 33 weeks as compared with those who delivered at ≥33 weeks (Table I). No significant difference was noted
Premature delivery is a major cause of perinatal morbidity and mortality in women with higher-order multifetal gestations. The incidence of low birth weight or very low birth weight infants resulting from triplet gestations approaches 50% as compared with a mere 2% for singleton pregnancies.13 Similarly, infant mortality is significantly greater in those neonates delivered of triplet gestations (19.4%) than in those from singleton gestations (1%).13 Cerebral palsy among survivors of triplet pregnancies occurs 47 times more often than in singletons.14 Clearly, interventions to detect or prevent premature delivery in this high-risk patient population are of utmost clinical importance. Use of ultrasonographic assessment of cervical length is an important clinical advance that has great potential application to the management of higher-order multiples. Iams et al7 elegantly documented the utility of this tool in the identification of patients with singleton gestations at risk for subsequent preterm delivery. In this series of 2915 gravid women, transvaginal ultrasonographic determination of cervical length was made at 24 weeks’ and again at 28 weeks’ gestation. Relative risk of preterm delivery based on the 24-week cervical length measurements revealed a dramatically increased risk in women with a shortened cervix. Similar findings were observed with the repeated ultrasonographic measurements at 28 weeks’ gestation. These data demonstrate the significance of a shortened cervix with respect to increased risk for preterm delivery. These authors raised the question of a continuum of cervical competence. The intrinsic competence of the cervix is that which is needed to maintain an individual pregnancy to term. Intuitively, this competency would remain constant in an individual woman; however, the load placed on the cervix could be altered (singletons vs triplets, hydramnios, and other factors). Impairment of factors that maintain this competence likely play a role in the development of preterm delivery.15 It is clear then that the cervix of patients with multifetal gestations is subjected to greater stressors and hence may exceed this competence at an earlier gestational age, resulting in premature delivery. This issue highlights the potential importance of ultrasonographic cervical assessment in this particular patient population. Several recent investigations have sought to evaluate this issue in twin gestations. Imseis et al10 evaluated 85 women with twin gestations with ultrasonographic cervical length assessment. Patients who subsequently delivered beyond 34 weeks’ gestation without intervention had a significantly greater cervical length (36.4 ± 5.8 mm) at the screening 24- to 26-week ultrasonographic assessment as compared with both those patients who delivered before 34 weeks (27.4 ± 8.5 mm) and those who
Ramin et al 1445
Volume 180, Number 6, Part 1 Am J Obstet Gynecol
delivered after 34 weeks but required intervention (27.7 ± 0.5 mm) (P < .001) When a cervical length cutoff of >35 mm was used as a marker for a low-risk population, a positive predictive value of 97% and a specificity of 94% were attained. Kushnir et al11 furthered investigation with twins through a comparative ultrasonographic cervical length evaluation between singleton and twin gestations. Similar to our data, this study demonstrated a significant difference between the cervical length of twins as compared with that of gestational age–matched singleton pregnancies.11 The results of our study further confirm the unique nature of the cervix in women with multifetal gestations. Progressive cervical shortening was observed with advancing gestational age. The degree of shortening observed was significantly greater than that previously reported for singleton pregnancies, which is consistent with impairment of cervical competence as a result of the multiple gestation.7 Additionally, crude comparisons between our data and those data from Kushnir et al11 suggest that the cervix in triplet pregnancies has a more rapid decline in length with advanced gestational age than is found in the cervix in twin pregnancies. Unlike the previously mentioned investigations concerning singleton pregnancies where no intervention was performed and thus the natural history of the gestation could be observed, we already were altering care by managing all of our triplet gestations according to a previously developed management scheme.12 Obviously, new standards are needed to establish normal and aberrant cervical changes in the higher-order gestations. Our data provide a basis for these standards. Similar to what has been reported for singleton and twin gestations, premature cervical shortening in triplets, as well as the development of cervical funneling, is a harbinger of premature delivery. Premature cervical shortening was noted in those women at risk for delivery before 33 weeks’ gestation, suggesting impairment of cervical competence as a predominant risk factor for early delivery in patients with triplet gestations. It is interesting that the relative changes noted between the patient cohort that delivered early and those that delivered beyond 32 weeks were not readily apparent until around the 20th week of gestation. These data support the management plan as proposed by Peaceman et al12 for triplets—reduced activity at approximately 20 weeks’ gestation with increased cervical surveillance. Whereas preliminary data have suggested that such a management plan can significantly reduce preterm delivery and improve perinatal outcome in triplet gestations, larger confirmatory series are needed.12 Ultrasonographic cervical length assessment repre-
sents an important advance in our clinical practice. Not only does the utility of cervical length assessment provide evidence of cervical competence or the lack thereof, it also provides us with a unique tool to monitor patients at risk for preterm delivery. Triplet and other higher-order pregnancies are exactly the high-risk situation in which data from ultrasonographic cervical assessment can provide early warning for impending preterm delivery such that appropriate obstetric intervention can be implemented. Future investigations should test potential obstetric interventions for preventing very preterm births in those triplet pregnancies at highest risk (funneling or cervical length <2.5 cm at 23-24 weeks’ gestation).
REFERENCES
1. Mordel N, Ezra Y, Dorembus D, Benshushan A, Schenker JG, Sadovsky E. Triplets are not so rare any more. J Perinat Med 1992;20:117-22. 2. Luke B. The changing patterns of multiple births in the United States: maternal and infant characteristics, 1973 and 1990. Obstet Gynecol 1994;84:101-6. 3. Lockwood CJ, Senyei AE, Dische MR, Casal D, Shah KD, Thung SN, et al. Fetal fibronectin in cervical and vaginal secretions as a predictor of preterm delivery. N Engl J Med 1991;325:669-74. 4. McGregor JA, Jackson GM, Lachelin GCL, Goodwin TM, Artal R, Hastings C, et al. Salivary estriol as risk assessment for preterm labor: a prospective trial. Am J Obstet Gynecol 1995;173:1337-42. 5. Leveno KJ, Cox K, Roark ML. Cervical dilation and prematurity revisited. Obstet Gynecol 1986;68:434-5. 6. Stubbs TM, Van Dorsten, Miller MC. The preterm cervix and preterm labor: relative risks, predictive values, and change over time. Am J Obstet Gynecol 1986;155:829-34. 7. Iams JA, Goldenberg RL, Meis PJ, 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. 8. Richey SD, Ramin KD, Roberts SW, Ramin SM, Cox SM, Twickler DM. The correlation between transperineal sonography and digital examination in the evaluation of the thirdtrimester cervix. Obstet Gynecol 1995;85:745-8. 9. Kurtzman J, Goldsmith LJ, Gall S, Spinnato JA. Transvaginal versus transperineal ultrasound: a blinded comparison in the assessment of cervical length at mid-gestation [abstract]. Am J Obstet Gynecol 1998;178(Pt 2):S15. 10. Imseis HM, Albert TA, Iams JD. Identifying twin gestations at low risk for preterm birth with transvaginal ultrasonographic cervical measurement at 24 to 26 weeks’ gestation. Am J Obstet Gynecol 1997;177:1149-55. 11. Kushnir O, Izquierdo LA, Smith JF, Blankstein J, Curet LB. Transvaginal sonographic measurement of cervical length: evaluation of twin pregnancies. J Reprod Med 1995;40:380-2. 12. Peaceman AM, Dooley SL, Tamura RK, Socol ML. Antepartum management of triplet gestations. Am J Obstet Gynecol 1992;167:1117-20. 13. Luke B, Keith LG. The contribution of singletons, twins and triplets to low birth weight, infant mortality and handicap in the United States. J Reprod Med 1992;37:661-6. 14. Petterson B, Nelson KB, Watson L, Stanley F. Twins, triplets, and cerebral palsy in births in Western Australia in the 1980’s. BMJ 1993;307:1239-43. 15. Craigo SD. Cervical competence and preterm delivery. N Engl J Med 1996;334:595-6.