Transvaginal ultrasonographic evaluation of the cervix before labor: Presence of cervical wedging is associated with shorter duration of induced labor

Fereshteh Boozarjomehri, MD, Han Timor-Tritsch, MD, Conrad R. Chao, MD, and Harold E. Fox, MD

New York, New York

OBJECTIVE: Our purpose was to test the hypothesis that transvaginal ultrasonographically determined characteristics of the cervix are associated with duration of induced labor. STUDY DESIGN: Fifty-three patients scheduled for induction of labor underwent transvaginal ultrasonography and digital cervical examinations before labor induction. Cox proportional-hazards multiple regression analysis was performed to determine the variables that made a significant contribution to the prediction of latent-phase and total labor duration. In the analysis the possible confounding effects of exogenous prostaglandin, previous vaginal delivery, and previous termination of pregnancy were controlled. RESULTS: Latent-phase and total labor duration were significantly associated with the presence of cervical wedging noted on transvaginal ultrasonography and administration of prostaglandin but not with the result of digital examination of cervical effacement or dilatation. Latent-phase duration was also associated with cervical length measured by transvaginal ultrasonography. The presence of wedging was significantly associated with shorter latent (15.9 ± 1.7 vs 34.1 ± 3.8 hours, p = 0.0001) and total (22.0 ± 1.8 vs 38.3 ± 3.6 hours, p = 0.0001) labor length. CONCLUSION: The presence of wedging and decreased cervical length observed by transvaginal ultrasonography is associated with a shorter duration of induced labor and may be useful in the evaluation of induction candidates. (AM J OesTET GVNECOL 1994;171 :1081-7.)

Key words: transvaginal ultrasonography, cervical wedging, shorter labor

Digital examination has been the standard method of assessment of the cervix before and during induction of labor. Studies have demonstrated differences between digital and transvaginal ultrasonographic assessment of the cervix.'-4 We hypothesized that characteristics of transvaginal ultrasonographic cervical assessment in candidates for induction of labor are associated with duration of labor.

Material and methods During the 4 months of the study all induction

candidates seen during research times were assessed for eligibility to be included in the study. Fifty-four patients with a viable fetus, intact fetal membranes, and no vaginal bleeding were asked to participate. One patient

From the Division of Maternal-Fetal Medicine, Department of Ob­stetrics and Gynecology, College of Physicians and Surgeons of Columbia University, and the Sloane Hospital for Women of the Presbyterian Hospital. Presented at the Fourteenth Annual Meeting of the Society of Perinatal Obstetricians, Las Vegas, Nevada, January 24-29, 1994. Reprint requests: Ilan E. Timor-Tritsch, MD, Department of Obstet­rics and Gynecology, Sloane Hospital for Women at Columbia Presbyterian Medical Center, 622 W. 168th St., New York, NY 10032. Copyright © 1994 by Mosby-Year Book, Inc. 0002-9378/94 $3.00 + 0 6/6/57839

declined the institutional review board-approved con­sent. Fifty-three pregnant patients in the third trimester (51 patients > 38 weeks of gestation, one at 34 weeks, one at 35 weeks) were studied. The indications for the induction were postdates (34), preeclampsia or preg­nancy-induced hypertension (7), oligohydramnios (5), and other medical or fetal problems necessitating de­livery (diabetes mellitus, Rh sensitization, seizure disor­der, fetal hydrocephalus, macrosomia, intrauterine growth retardation, fetal diaphragmatic hernia) (7).

Digital cervical evaluation was performed by the pri­vate physician or the admitting physician before induc­tion. Induction of labor was defined as initiation of labor by administration of intravaginal prostaglandin gel or intravenous oxytocin. The agent was selected on the basis of the initial digital evaluation of the cervix. Induction of labor was performed according to stan­dard practice in this hospital. The prostaglandin dose was 4 mg of gel administered vaginally every 4 hours for three doses. Volumetrically controlled oxytocin was ad­ministered intravenously at 1 mU/min and increased by 1 mU/min every 30 minutes until labor was established. Thirty-eight patients received both prostaglandin and oxytocin; 10 were given oxytocin only and five patients were delivered after prostaglandin administration only.

1081

1082 Boozarjomehri et al. October 1994 Am J Obstet Gynecol

Fig. 1. Transvaginal ultrasonography images of different shapes of cervix. A, Nonwedging. B, Wedging. C, U-shape. D, Mucus plug in internal os obscuring wedge pattern. E, Hypertrophic cervical glands giving the appearance of wedging cervix. F, Schematic representation of lower uterine segment, cervix, and vagina.

In our study the effacement and dilatation components of the Bishop5 score were studied in comparison to transvaginal ultrasonographic cervical length and inter­nal os descriptive anatomy. Dilatation in digital evalu­ation was defined as the ability of the examiner to open or be able to assess the internal os of the cervix by digital examination. Effacement was defined as the cervical length estimated by the examiner.

All transvaginal ultrasonography procedures were performed by one of the authors (F.B.), before induc­tion, by means of standard technique. 6 •s We used an Hitachi EVB 410 ultrasonography machine (Hitachi, Tokyo) with a 6.5 MHz transvaginal transducer. The best image quality was obtained with the area of interest between 2 and 7 cm from the tip of the probe, which is the focal zone of the 6.5 MHz probe.B• 9 The length of the cervix by transvaginal ultrasonography was mea­sured from the internal os to the external os, the furthest points at which the cervical walls were juxta­posed.B• 10 In cases of cervical wedging the thickness of the anterior or posterior lip ofthe cervix (depending on which one was better visualized) was used for consistent measurement of cervical length. The length was mea­sured during uterine diastole. Dilatation in transvaginal ultrasonography was defined 'as visualization of any opening in the entire cervical canal. The first recorded image of the cervix in the sagittal section (which was the only image used for statistical evaluation) was obtained before induction. This was followed by periodic sooo­grams every 4 to 6 hours in the early part of induction with milder contractions and every 0.5 to 1 hour once

any cervical change was noted on transvaginal ultra­sonography. The number of examinations in each pa­tient varied depending on clinical need for examina­tion, as determined by the physicians managing the labor, or on the rate of progress of the latent phase of labor. Each scan lasted 1 to 5 minutes. Onset of labor was defined as the time when induction started. The study of cervical effacement and dilatation was contin­ued through completion of the latent phase of labor, which was defined as 90% to 100% effacement and 3 to 4 cm dilatation of the cervix as measured by both I transvaginal ultrasonography and digital examination. This study took from a few hours to several days in each case.

The subjects were categorized in two groups accord­ing to the configuration of the internal os on the initial transvaginal ultrasonographic cervical image only. (1) Nonwedgingwas defined as a closed internal os (Fig. 1, A) . (2) Wedging or funneling was defined as any triangle "V pattern" at the area of the internal os with its apex anywhere along the cervical canal (Fig. 1, B). In characterizing different patterns of opening of the in­ternal os, we observed a "U pattern" of the internal os, which seemed to be exaggerated "wedging" or "bal­looning" of the upper portion of the cervix and short­ening of the cervical canal" (Fig. 1, C). We classified both V and U patterns as "wedging."

Statistical analysis Cox regression. To assess the effect of cervical wedging

on labor length while controlling for confounding vari­ables and censored observations, Cox proportional-

Volume 171, Number 4 Am J Obstet Gynecol

Table I. Characteristics of patients with and without cervical wedge

No. of patients Age (yr)

Mean ± SEM

Gestational age (wk) Parity Previous vaginal delivery Previous voluntary termination Dilatation (em) Effacement (%) Length by transvaginal ultrasonography (mm) Prostaglandins given Time to active phase (hr) Time to delivery (hr)

Wedge present

31 24.9 ± 1.0 40.3 ± 0.3 0.77 ± 0.19

14 (45%) 18 (58%)

1.00 ± 0.19 42 ± 4

28.0 ± 1.0 22 (71%)

15.9 ± 1.7 22.0 ± 1.8

Table II. Multivariable model of latent phase and total labor duration

Variable*

Duration of latent phase Variables in the equation

Wedge presence Previous voluntary termination Cervical length by transvaginal ultrasonography Use of prostaglandin

Variables not in the equation Previous vaginal delivery

Cervical dilatation (digital examination) Cervical effacement (digital examination)

Duration of total labor Variables in the equation

Wedge presence Use of prostaglandin

Variables not in the equation Previous voluntary termination Previous vaginal delivery Cervical dilatation (digital examination) Cervical effacement (digital examination) Cervical length by transvaginal ultrasonography

Significance

p = 0.0097 P = 0.0136 P = 0.0413 P = 0.0027

P = 0.9254 P = 0.3285 P = 0.4196

p = 0.0000 p = 0.0017

P = 0.0670 P = 0.3352 P = 0.4482 P = 0.4460 P = 0.3667

Exp (B)

1.7140 1.6024 0.9383 0.5505

2.0981 0.5396

Boozarjomehri et aI. 1083

Wedge absent

22 25.4 ± 1.2 39.7 ± 0.4 0.27 ± 0.16

3 (14%) 3 (14%)

0.48 ± 0.14 39 ± 6

34.8 ± 1.5 21 (91%)

34.1 ± 3.8 38.3 ± 3.6

Significance

p = 0.754 P = 0.214 P = 0.026 P = 0.034 P = 0.002 P = 0.022 P = 1.000 P < 0.001 P = 0.059 P < 0.001 P < 0.001

Exp (B) 95% confidence limit

Lower

1.1394 1.1018 0.8827 0.3728

1.4872 0.3670

I Upper

2.5783 2.3303 0.9975 0.8131

2.9600 0.7934

Exp (B) = e to the B power. For continuous variables (dilatation, effacement, length) Exp B indicates percent change in hazard rate for I-unit change in variable. Values < 1 indicate decrease in rate of achievement of active phase or delivery for each increase in that variable; values > 1 indicate increase in those rates for each increase in variable. In case of cervical length Exp B = 0.9383 indicates that likelihood of progression to active phase is less by approximately 6.2% per hour for each millimeter increase in cervical length. For dichotomous variables (wedge, prostaglandin, termination of pregnancy, delivery) Exp B indicates relative risk of progression to active phase or delivery for that variable. In case of wedge presence Exp (B) = 1.7140 indicates that relative likelihood of progression to active phase for patient with wedging is 1.71 times that of patient without wedging. Variables for which confidence intervals do not include 1 are significantly related to the outcome (latent phase or total labor length).

*For dichotomous variables presence of attribute was coded as 1 and absence of attribute was coded 'as O.

hazards multiple regression analysisl2 was performed, with length of latent phase and duration of total labor as dependent variables in separate analyses and pres­ence of wedging, history of voluntary termination of pregnancy, history of previous vaginal delivery, initial cervical length as measured by transvaginal ultrasonog­raphy, initial cervical dilatation as measured by digital examination, initial cervical effacement as measured by digital examination, and use of prostaglandin as inde­pendent variables.

Other analyses. Least-squares regression analysis was performed to examine the relationships between vari-

abIes suspected of having linear correlations (transvagi­nal ultrasonographically derived parameters and their clinical examination counterparts). Mean values of par­ity, effacement, and dilatation (the three variables for which the data were not normally distributed in both wedge and nonwedge groups) were compared by Mann-Whitney U test. Normally distributed continuous variables such as length and dilatation by transvaginal ultrasonography, age, and gestational age were com­pared with unpaired t test. Significance was inferred in cross tabulations by X2 or Fisher's exact test with Yates' correction as appropriate (fable I).

1084 Boozarjomehri et al

45

40

35 en L..

5 30 ~ c: 25 0

:oJ 03 L.. 20 :J Cl L.. 0 15 .0 03

...J

10

5

0 WEDGE NO WEDGE

Latent Phase

I.~

WEDGE NO WEDGE

Total Labor

45

40

35

30

25

20

15

10

5

o

October 1994 Am J Obstet Gynecol

Fig. 2. Latent and total labor duration in wedge and nonwedge groups (n = 53).

Results

Characteristics of patients with and without wedg­ing. Thirty-one of the 53 patients exhibited wedging (demographic data presented in Table I) . Patients dem­onstrating wedging were not different in maternal age or gestational age from nonwedge patients but were more likely to have had a previous delivery, and there was a strong trend toward increased parity in the wedging group; however, in both groups average parity was less than one, indicating the relatively high preva­lence of nulliparity in this study population. Patients with wedging were more likely to have had a previous voluntary termination of pregnancy. There was a trend toward greater use of prostaglandins in the nonwedge patients (Table I).

The presence of wedging was significantly associated with a greater mean cervical dilatation (Table I). Cer­vical effacement was not different between wedge and nonwedge patients. On the other hand, transvaginal ultrasonographic cervical length was significantly greater in the nonwedge patients. In addition, cervical length, as assessed by transvaginal ultrasonography, showed no relationship to cervical effacement measured by digital examination (r = 0.227, P = 0.102). Cervical dilatation assessed by transvaginal ultrasonography also showed no relationship to cervical dilatation measured by digital examination (r = 0.168, P = 0.230).

Univariate analysis. The presence of a wedge, as demonstrated by transvaginal ultrasonography, was as­sociated with a significantly shorter duration of the latent phase (15.9 ± 1.0 vs 34.1 ± 3.8 hours, P =

0.0001) and total labor (22.0 ± 1.8 vs 38.3 ± 3.6

hours, P = 0.0001) duration (Table I and Fig. 2). Cer­vical dilatation as assessed by digital examination was also associated with shorter latent (r = -0.4215, P = 0.003) and total (r = - 0.3987, P = 0.003) labor. Cervical effacement by digital examination was not significantly associated with duration of either latent or total labor duration (r = - 0.0820, P = 0.576 and r =

- 0.0720, P = 0.608, respectively) . The original question posed in this study was to

determine the value of cervical wedging in the predic­tion of induced labor duration. Numerous factors might

I confound this study (digital examination, cervical length by transvaginal ultrasonography, historic and demographic factors). We controlled for the effect of possible confounders by Cox proportional-hazards mul­tiple regression analysis.

The analysis for latent-phase duration (Table II) revealed that of the variables examined presence of wedge and previous voluntary termination of preg­nancy were significantly associated with shorter latent labor duration [Exp (B) > 1, see footnote to Table II]. Prostaglandin use and increasing transvaginal ultra­sonographically measured cervical length were signifi­cantly associated with a longer latent phase [Exp (B) < 1]. Parity (number of previous vaginal deliveries) and findings on digital cervical examination did not dem­onstrate a significant relationship to the duration of the latent phase. Similarly, for total labor duration (Table II) the presence of wedging was significantly associated with shorter labor length, and the use of prostaglandin was associated with longer labor length. History of voluntary termination of pregnancy, parity, digital cer-

Volume 171. Number 4 Am J Obstet Gynecol

(1) en I1l .r; c.. .... c

~ ...J

.!: 01 c 'c; 'Cij E (1)

a:: en 1: (1)

~ c..

Boozarjomehri et al. 1085

1,2

1.0

.8

.6

.4

.2

0,0 WEDGE

• PRESENT

-,2 o ABSENT

a

Time (hours)

Fig. 3. Survival function plot for duration of latent phase. '

1,2

1,0

"'0 (1) '-(1)

.8 .2: Q3 "'0 C

::J .6 01

c 'c; 'Cij E (1)

.4 a:: en 1: .2 (1)

~ c..

0,0 WEDGE

• PRESENT

-.2 o ABSENT

a 20 40 60 80 100

Time (hours)

Fig. 4. Survival function plot for total labor duration.

vical examination, and transvaginal ultrasonographic assessment of cervical length were not significantly associated with the total duration of labor. Survival function plots (Figs. 3 and 4), stratified for the presence of wedging, demonstrated that for essentially all time periods significantly more patients with wedging pro­gressed to the active phase or were delivered compared with patients without wedging.

Comment

The Bishop scoring system,S in the case of a poor Bishop score, is not helpful in predicting the response to induction of labor. Several studies addressed the relationship between preinduction cervical scoring sys­tems and the course of labor. According to some ob­servers 13, 15 the Bishop score has a poor predictive value for the outcome of induction. Paterson-Brown et aI.'

1086 Boozarjomehri et al.

have found that among the Bishop score components only dilatation and length (effacement) have shown a significant correlation with successful vaginal delivery. Lang et al. 16 found that of the five components in the Bishop score, dilatation was found to be the most important component, and it was also believed that it is the most simple, accurate, and reproducible com­ponent.

Anatomically, in closed cervices only the distal one half of the true cervical length (vaginal portion) is palpable (Fig. 1, F)" 6. 17; therefore neither the configu­ration of the internal os (e.g., wedging, V or U shape) nor the entire length of the cervix can be judged by palpation alone. The technique of transvaginal ultraso­nographic assessment of the cervix is easily applicable to all patients. During scanning the vaginal transducer is in close proximity to the cervix; it avoids the pitfalls of a transabdominal scan requiring a full bladderlO. 18. 19

and the shadowing of the perineal scan in which the content of the rectosigmoid colon may interfere with the image.2o

As to the discrepancy between cervical dilatation as­sessed by digital examination and the dilatation seen on transvaginal ultrasonography, it may be related to stretching of the cervical os during digital examination, which implies cervical softening and therefore ripening. This observation is also reported by Anderson and Ans­bacher. 7 It is apparent that these are different observa­tions that may be similar but not congruent. The correla­tion between cervical effacement by digital examination and transvaginal ultrasonography, as far as the initial measurement of cervical length is concerned, was poor both in the "wedged" and "nonwedged" groups. This is in agreement with previously published studies 14 and is possibly because only 50% ofthe cervix (vaginal portion) is palpable by digital examination," 6.17 and also because effacement is a subjective evaluation with high interob­server variability.2 The measurement of cervical length by transvaginal ultrasonography is a reproducible and objective evaluation. 7 The sonologist must carefully as­sess wedging, which at times may be obscured by mucus or confused with cervical glands (Figs. 1, D and E). Care­ful real-time imaging and liberal use ofthe different gain controls clarifies these conditions.

Our study suggests that the length of the latent phase correlates with transvaginal ultrasonographically mea­sured cervical length, whereas length of total labor does not. This is consistent with the observation of Paterson­Brown et al. 4 that a relationship could not be demon­strated between induction-to-delivery interval and cer­vicallength measured by transvaginal ultrasonography.

Although duration of spontaneous labor is known to be shorter in multiparous women, this is not necessarily true in those undergoing induction. In the study by Lang et al. 16 the following is stated:

October 1994 Am J Obstet Gynecol

The influence of parity on duration of labor was found to be very constant and highly significant, whereas it seemed to have only little influence on inducibility, and the only factor of significant importance of inducibility was the preiabor cervical condition.

Similarly, in our results parity did not demonstrate a significant relationship to induced latent-phase and total labor duration. By means of transvaginal ultra­sonographically defined characteristics of the cervix we studied the preiabor cervical condition and demon­strated the association of the presence of a wedge with significantly shorter induced latent and total labor. We therefore conclude that cervical wedging detected by transvaginal ultrasonography is an early anatomic sign of cervical maturation or ripening. Our finding suggests that transvaginal ultrasonographic cervical assessment may be helpful in planning induction for women with long-closed cervices.

REFERENCES 1. Lim BH, Mahmood TA, Smith NC, Beat I. A prospective

comparative study of transvaginal ultrasonography and digital examination for cervical assessment in the third trimester of pregnancy. ] Clin Ultrasound 1992:20:599-603.

2. Kushnir 0, Vigil DA, Izquierdo L, Schiff M, Curet LB. Vaginal sonographic assessment of cervical length changes during normal pregnancy. AM] OBSTET GYNECOL 1990:162: 991-3.

3. Jackson GM, Ludmir ], Bader TJ. The accuracy of digital examination and ultrasound in the evaluation of cervical length. Obstet Gynecol 1992:79:214-8.

4. Paterson-Brown S, Fisk NM, Rodeck CH, Rodeck E. Pre­induction cervical assessment by Bishop's score and trans­vaginal ultrasound. Eur ] Obstet Gynecol Reprod Bioi 1991:40:17-23.

5. Bishop E. Pelvic scoring for elective induction. Obstet Gynecol 1964:24:266-8.

6. Sonek ]D, lams ]D, Blumenfeld M, ] ohnson F, Landon M, Gabbe S. Measurement of cervical length in pregnancy: comparison between vaginal ultrasonography and digital examination. Obstet Gynecol 1990:76:172-5.

7. Anderson HF, Ansbacher R. Ultrasound: a new approach to the evaluation of cervical ripening. Semin Perinatol 1991:2:140-8.

8. Anderson HF, Nugent CE, Wanty SD, Hayashi RH. Pre­diction of risk for preterm delivery by ultrasonographic measurement of cervical length. AM ] OBSTET GYNECOL 1990:163:859-67.

9. Timor-Tritsch IE, Rottem S. Transvaginal sonography. New York: Elsevier, 1987:1-13.

10. Anderson HF. Endovaginal and transabdominal ultra­sonography of the uterine cervic during pregnancy.] Clin Ultrasound 1991:19:77-83.

11. Brown ]E, Thieme GA, Shah DM, Fleischer AC, Boehm FH. Transabdominal and transvaginal endosonography: evaluation of the cervix and lower uterine segment in pregnancy. AM] OBSTET GYNECOL 1986:155:721-6.

12. Pedhazur E. Multiple regression in behavioral research. Fort Worth, Texas: Harcourt-Brace, 1982.

13. Friedman EA, Niswander KR, Bayonet-Rivera NP, et al. Relationship of prelabour evaluation to inducibility and the course of labour. Obstet Gynecol 1966:28:495-501.

14. Hughey M], McElin TW, Bird CC. An evaluation of pre­induction scoring system. Obstet GynecoI1976:48:635-41.

Volume 171, Number 4 Am J Obstet Gynecol

15. Dhall K, Mittal SC, Kumar A. Evaluation of preinduction scoring system. Aust N Z j Obstet Gynecol1987 ;27 :309-11.

16. Lang AP, Secher Nj, WestergaardjG. Prelabor evaluation of inducibility. Obstet GynecoI1982;60:137-47.

17. Michaels WH, Montgomery C, Karo j, Temple j, Ager J. Ultrasound differentiation of the competent from the incompetent cervix: prevention of preterm delivery. AM J OBSTET GYNECOL 1986;154:537-46.

Hales et al.

18. Confino E, Maydon KL, Giglia RV, Vermesh M, Gleicher N. Pitfalls in sonographic imaging of the incompetent uterine cervix. Acta Obstet Gynecol Scan 1986;65:593-7.

19. Zemlyn S The effect of urinary bladder in obstetrical sonography. Radiology 1978;128:167-75.

20. Jeanty F, D'Alton M, Romero R, Hobbins J. Perineal scanning. Am J Perinatol 1986;3:289-95.

Double-blind comparison of intracervical and intravaginal prostaglandin E2 for cervical ripening and induction of labor

Kurt A. Hales, MD: William F. Rayburn, MD," b Gayla L. Turnbull, RN," H. Dix Christensen, PhD," b and Edna Patatanian, RPhC

Oklahoma City, Oklahoma

OBJECTIVE: Our purpose was to compare the safety and effectiveness of prostaglandin E2 delivered sequentially as an intracervical (0.5 mg) or intravaginal (2.5 mg) gel. STUDY DESIGN: Hospitalized patients with an unfavorable cervix (Bishop score 054) at <!:35 weeks and requiring induction of labor were assigned to receive two 2.5 ml doses of gel intracervically and intravaginally in a double-blind, placebo-controlled manner. Second and third doses were given at 6-hour intervals until there were either regular uterine contractions or a Bishop score change > 3 points. RESULTS: The 100 evaluable cases received prostaglandin E2 either intracervically (n = 52) or intravaginally (n = 48). Difficulty with exact gel instillation was present with intracervical gel only, where spillage occurred in 85% of cases. Compared with intracervical therapy prostaglandin E2 given intravaginally was more likely to significantly change the Bishop score (60.4% vs 40.4%, P = 0.04) and stimulate regular contractions (72.9% vs 48.1%, P = 0.01). Uterine hyperstimulation was present in one case in each group. CONCLUSION: Although each was safe, instillation of prostaglandin E2 gel was better at a higher intravaginal dose than a lower intracervical dose because of its greater ease of administration and higher likelihood of cervical change. (AM J OBSTET GYNECOL 1994;171 :1087-91.)

Key words: Prostaglandin E2, cervical ripening, induction of labor

Control of the onset of labor represents a major challenge to the obstetrician. A pregnancy requiring induction of labor with an unfavorable cervix presents a management dilemma similar to the inhibition of pre­term labor. Prostaglandin E2 (PGE2) has been shown to be efficacious in promoting preinduction cervical rip-

From the Departments of Obstetrics and Gynecology,· Pharmacology, b

and Hospital Pharmacy,' UniversityofOklahoma College of Medicine. Supported by the John W. Records Perinatal Research Fund. Presented at the Fourteenth Annual Meeting of the Society of Perinatal Obstetricians, Las Vegas, Nevada, January 24-29, 1994. Reprint requests: Kurt A. Hales, MD, Santa Barbara Cottage Hospital, Perinatal Center, P.O. Box 689, Santa Barbara, CA 93102.

ening and in initiating labor. I -3 Various routes of deliv­ery and doses of the drug have been tried, but topical application as either an intracervical or intravaginal gel in low doses has gained the most widespread accep­tance. Use of such preparations is safe, with uterine hyperstimulation being the primary yet uncommon and reversible side effect. I. 2 An intracervical preparation (Prepidil, Up john, Kalamazoo, Mich.) is now available commercially and, because of its lower dose (0.5 mg vs 2.5 mg), may be associated with less uterine hyper­stimulation than a widely used intravaginal preparation.

Copyright © 1994 by Mosby-Year Book, Inc. 0002-9378/94 $3.00 + 0 6/6/57840

Few studies with PGE2 have been performed compar­ing success with cervical ripening when it is given intracervically or intravaginally. These have been un­dertaken in an unblinded manner using a single dose.

1087