European Journal of Obstetrics & Gynecology and Reproductive Biology, 42 (1991) 177-180 0 1991 Elsevier Science Publishers B.V. All rights reserved 0028-2243/91/$03.50
177
EUROBS 01234
Influence of vehicle form on efficiency of prostaglandin for cervical ripening
E, iYe1
Kristjar Skajaa, Astrid Mamsen and Niels J@rgen Secher Department of Obstetrics and Gynecology, University of A’rhus, AOrhus,Denmark
Accepted for publication 11 June 1991
Summary A double-blind randomized controlled trial compared the efficiency of prostaglandin E, (PGE,) gel in two different vehicle forms, one solid and the other more liquid, with respect to inducing labour in women with highly unfavourable cervical states. 43 received 0.5 mg PGE, in a solid gel (Cerviprost) and 37 received 0.5 mg PGE, in a more liquid cellulose-gel. After one application 18 (42%) and 9 (24%), respectively, went into labour (OR 2.8; 95% confidence interval: 0.8-6.0). After two applications of the gel 91 and 76%, respectively, were in active labour or had amniotomy performed as the cervical state had improved (OR 2.9; 95% confidence interval: 0.8-10.3). It is concluded that the solid gel was more efficient and safe than the more liquid gel. Prostaglandin
E,; Cervix; Labor
Introduction Several researchers recommend local application of prostaglandin E, (PGE,) for priming of the unripe cervix in order to induce labour [l]. Results have diverged from trial to trial, even when the same application route was used. This indicates that the viscosity of the vehicle rather than the application route should be investigated. Thus, the aim of the present study was to compare the efficiency of the PGE, gel in two different vehicle forms, one solid, the other more liquid, for inducing labour.
Correspondence:
Kristjar Skajaa, Department of Obstetrics and Gynecology, University of ,&rGlu, irhus, Denmark.
Material and Methods During the study period of 48 months, 7999 deliveries took place at the department. 420 (5.3%) were induced by amniotomy or by intracervical application of PGE, gel. Of these 80 (19%) fulfilled the inclusion criteria and were randomly allocated by the sealed envelope method to receive PGE, (0.5 mg) intracervically either in a cross-link starch polymer gel (Cerviprost R, Organon) which resulted in a rather solid substance, or in oxypropylmethylcellulose 3% which was more liquid than the former. Inclusion criteria comprised: a cervical score < 4 (modified Bishop), i.e., cervix not shorter than 1 cm and orificium not more than 1 cm dilated, a fetus in cephalic position, intact membranes, and no previous caesarean section.
178
The cervical state was evaluated by a midwife before the randomization, and later in accordance to the protocol this midwife, who did not know which kind of gel the woman had received, reassessed the cervical score. If the first portion of the gel did not induce active labour, characterized by frequent, intense contractions and improvement of the cervical score, a new cervical score was assessed after 6 h by the same midwife. If the cervical state had improved to 6 or more, amniotomy was performed, and an oxytocin drip was established about 2 to 3 h later, if necessary. If the cervical score was less than 6, the patient received a second application of the same gel. Patients who were not in active labour after 24 h and two applications of the gel had their cervical state evaluated once again. If amniotomy could not be performed, the patient was considered a failure and discharged from the protocol. These women received yet an application of the gel or had a caesarean section. Statistics The power calculation in the study was performed in accordance with other studies using PGE, gel but in different vehicle forms. Ekman et al. [2] reported a success rate of about 90% with Cerviprost R and Legarth et al. [3] and Stampe Sorensen et al. [4] a rate of about 60% with the cellulose gel. The risk of type 1 error and type 2 error was stipulated to 2a = 5% and p = 80%, respectively. Based on this and a desired minimal difference of 30% not to be overlooked, 80 women were required. Statistical analysis was by X2-test and by t-test for unpaired observations. Differences in the relative effects of the two PGE, gels were expressed as Odds Ratios (OR) with 95% confidence intervals (95% Cl). In stratified analysis, the MantelHaenzel test was used.
TABLE I Clinical features Cerviprost (n = 43) No. of primips Age (years, mean k SE) Gestat.age (days, mean f SE)
26 (70%) 28 (65%) P = 0.8 (X2-test) 27.0+ 0.9 28.3* 0.9 P = 0.3 (t-test) 284 + 2
43 women received Cerviprost and 37 received the cellulose gel. Clinical data concerning parity, age, gestational age, initial cervical score, and
280 +
3
P = 0.3 (r-test)
Initial cervical score (modified Bishop, mean f SE) Birth weight (g, mean f SE)
2.8* 3673
0.1
+91
2.8+ 3194
0.1
f134
P = 0.004 (t-test)
birth weight are presented in Table I. Indications for induction are listed in Table II. While no significant differences in the distribution of primips and multips or gestational age were discovered, the distribution of indications for the induction were different. This is reflected in the difference in birth weight. Thus, the calculation of OR was performed after stratification in accordance with indications for induction. 6 (14%) in the Cerviprost-group and 3 (8%) in the cellulose-gel group went into active labour after the first gel application and additionally 12 (28%) and 6 (16%), respectively, after amniotomy 6 h later as the cervical score had improved. The mean change in the score was 3.4 (SE 0.4) in the Cerviprost group and 3.6 (0.4) in the cellulose-gel group. Thus, 42 and 24% were successfully induced after one application of Cerviprost or cellulose gel, respectively (Table III). TABLE II Indications for induction Cetviprost 43)
(n =
Results
(Cellulose-gel (n = 37)
(I) Prolonged pregnancy (more than 42 weeks) (II) Hypertension/IUGR (III) Others
Cellulose gel (n = 37)
28 (65%) 17 (46%) 6 (14%) 14 (38%) 9 (21%) 6 (16%) P = 0.05 (X*-test)
179 TABLE III Outcome of induction In labour (frequent contractions and/or amniotomy) after first application of the gel: Cetiiprost (n = 43) Indication
Cellulose gel tn = 37)
group
(I)
13 (46%) 2 (33%)
(III
3 (33%) OR = 2.2 (0.8-6.0) Mantel-Haenzel
(III)
Indication
6.5 (1.4-30.7) 1.3 (0.2-10.3)
2 (12%) 4 (29%)
0.5 (4.4-0.06)
3 (50%) test
In labour (frequent contractions and/or after second application of the gel
(1) (II) (III)
OR (95% CI)
amniotomy)
Cerviprost (n = 25)
Cellulose gel (n = 28)
13 (87%)
11 (73%)
group 3 (75%) 5 (83%) OR = 2.6 (0.7-9.9) Mantel-Haenzel
2.4 (0.3-15.5)
1.3 (0.08-19.8) 10.0 (0.4-243)
7 (70%) 1 (33%) test
After the second application of the actual gel, an additional 8 of the remaining 25 (32%) and 10 of the 28 (36%), respectively, went into active labour within 24 hours. Thirteen (52%) with a mean cervix score of 5.5 (SE 0.4) in the Cerviprost group and 9 (32%) with a mean score of 5.4 (SE 0.4) in the cellulose-gel group had amniotomy performed after 24 h. Thus, 4 (9%) in the Cerviprost group and 9 (24%) in the cellulose-gel group were considered failures, as the cervical state had not improved properly for amniotomy even after 24 h (P = 0.06). Uterine hypertonia did not occur in any case after gel application. Of the 67 women still in the study, 6 (15.4%) in the Cerviprost group and 8 (28.6%) in the cellulose-gel group (NS) required caesarean section. In 2 and 5 cases, respectively, impending fetal distress was the indication for the caesarean section. The overall caesarean section rate in the department was 8.9% over the same period. Among women induced by primary amniotomy, the rate was 17% (unpublished data). One child in each group had an Apgar score below 7 after 1 min, and after 5 min the value was 8 for both children. The rest had Apgar scores of 10 after 1 and 5 min. In both groups the
pH venous blood from the cord was 7.29 + 0.02. There were no serious perinatal complications in either of the two groups. Discussion Based on the high ORs, our results suggest that PGE, administered in a solid gel for cervical ripening and/or induction of labour is more efficient and safe than gel in a more liquid form. Thus, the liquid gel seems more liable to move extra-amniotically and/or intravaginally, thereby reducing its influence on the cervical tissue. The indications for induction in the two treatment groups were different in spite of the randomization procedure. More cases with hypertension and/or IUGR were represented in the cellulose-gel group. These complications were also reflected by the lower birth weight in that group. Statistical analysis by the stratification for indication-for-induction class, however, took these differences into account, thus enabling a valid comparison. Our success rate was not as high as those reported in other investigations using a solid gel [2,5]. This could be due to different techniques of application. It is more likely, however, that our
180
study population had a greater incidence of very unripe cervices. The cervical score is vitiated with lack of objective parameters. Thus, the present design which involved blind assessment of the cervical state by a midwife, not otherwise involved in the investigation, should eliminate such bias. The inequality among the study populations makes comparison of the results from other study difficult, especially because the induction rate and the percentage of those induced with PGE, gel because of a unripe cervix, are infrequently stated. The caesarean section rate among those treated with PGE, gel was not, as might have been expected [l], reduced compared with those induced with primary amniotomy. It may reflect that the gel-treated women had very unfavourable cervical states, which actually was the reason why they were not induced with primary amniotomy. It seems that the solid gel is preferable, since more caesarean sections were per-
formed in the cellulose-gel group, although the difference was not significant. References Rayburn WF. Prostaglandin E, gel for cervical ripening and induction of labour: a critical analysis. Am J Obstet Gynecol 1989;160:529-534. Ekman G, Forman A, Marsal K et al. Intravaginal versus intracervical application of prostaglandin E, in viscous gel for cervical priming and induction of labour at term in patients with an unfavorable cervical state. Am J Obstet Gynecol 1983;147:657-662. Legarth J, GuldbEk E, Secher NJ. The efficiency of prostaglandin E, vaginal suppositories versus intracervical prostaglandin gel for induction of labour in patients with unfavourable inducibility prospects. Eur J Obstet Gynecol Reprod Biol 1988;27:93-98. Stampe Sorensen S, Brocks V, Lenstrup C. Induction of labour and cervical ripening by intracervically prostaglandin E,. Obstet Gynecol 1985:65;110-114. Ulmsten AU, Wingerup L, Ekman G. Local application of prostaglandin E, for cervical ripening or induction of term labour. Clin Obstet Gynecol 1983;26:95-105.
177
EUROBS 01234
Influence of vehicle form on efficiency of prostaglandin for cervical ripening
E, iYe1
Kristjar Skajaa, Astrid Mamsen and Niels J@rgen Secher Department of Obstetrics and Gynecology, University of A’rhus, AOrhus,Denmark
Accepted for publication 11 June 1991
Summary A double-blind randomized controlled trial compared the efficiency of prostaglandin E, (PGE,) gel in two different vehicle forms, one solid and the other more liquid, with respect to inducing labour in women with highly unfavourable cervical states. 43 received 0.5 mg PGE, in a solid gel (Cerviprost) and 37 received 0.5 mg PGE, in a more liquid cellulose-gel. After one application 18 (42%) and 9 (24%), respectively, went into labour (OR 2.8; 95% confidence interval: 0.8-6.0). After two applications of the gel 91 and 76%, respectively, were in active labour or had amniotomy performed as the cervical state had improved (OR 2.9; 95% confidence interval: 0.8-10.3). It is concluded that the solid gel was more efficient and safe than the more liquid gel. Prostaglandin
E,; Cervix; Labor
Introduction Several researchers recommend local application of prostaglandin E, (PGE,) for priming of the unripe cervix in order to induce labour [l]. Results have diverged from trial to trial, even when the same application route was used. This indicates that the viscosity of the vehicle rather than the application route should be investigated. Thus, the aim of the present study was to compare the efficiency of the PGE, gel in two different vehicle forms, one solid, the other more liquid, for inducing labour.
Correspondence:
Kristjar Skajaa, Department of Obstetrics and Gynecology, University of ,&rGlu, irhus, Denmark.
Material and Methods During the study period of 48 months, 7999 deliveries took place at the department. 420 (5.3%) were induced by amniotomy or by intracervical application of PGE, gel. Of these 80 (19%) fulfilled the inclusion criteria and were randomly allocated by the sealed envelope method to receive PGE, (0.5 mg) intracervically either in a cross-link starch polymer gel (Cerviprost R, Organon) which resulted in a rather solid substance, or in oxypropylmethylcellulose 3% which was more liquid than the former. Inclusion criteria comprised: a cervical score < 4 (modified Bishop), i.e., cervix not shorter than 1 cm and orificium not more than 1 cm dilated, a fetus in cephalic position, intact membranes, and no previous caesarean section.
178
The cervical state was evaluated by a midwife before the randomization, and later in accordance to the protocol this midwife, who did not know which kind of gel the woman had received, reassessed the cervical score. If the first portion of the gel did not induce active labour, characterized by frequent, intense contractions and improvement of the cervical score, a new cervical score was assessed after 6 h by the same midwife. If the cervical state had improved to 6 or more, amniotomy was performed, and an oxytocin drip was established about 2 to 3 h later, if necessary. If the cervical score was less than 6, the patient received a second application of the same gel. Patients who were not in active labour after 24 h and two applications of the gel had their cervical state evaluated once again. If amniotomy could not be performed, the patient was considered a failure and discharged from the protocol. These women received yet an application of the gel or had a caesarean section. Statistics The power calculation in the study was performed in accordance with other studies using PGE, gel but in different vehicle forms. Ekman et al. [2] reported a success rate of about 90% with Cerviprost R and Legarth et al. [3] and Stampe Sorensen et al. [4] a rate of about 60% with the cellulose gel. The risk of type 1 error and type 2 error was stipulated to 2a = 5% and p = 80%, respectively. Based on this and a desired minimal difference of 30% not to be overlooked, 80 women were required. Statistical analysis was by X2-test and by t-test for unpaired observations. Differences in the relative effects of the two PGE, gels were expressed as Odds Ratios (OR) with 95% confidence intervals (95% Cl). In stratified analysis, the MantelHaenzel test was used.
TABLE I Clinical features Cerviprost (n = 43) No. of primips Age (years, mean k SE) Gestat.age (days, mean f SE)
26 (70%) 28 (65%) P = 0.8 (X2-test) 27.0+ 0.9 28.3* 0.9 P = 0.3 (t-test) 284 + 2
43 women received Cerviprost and 37 received the cellulose gel. Clinical data concerning parity, age, gestational age, initial cervical score, and
280 +
3
P = 0.3 (r-test)
Initial cervical score (modified Bishop, mean f SE) Birth weight (g, mean f SE)
2.8* 3673
0.1
+91
2.8+ 3194
0.1
f134
P = 0.004 (t-test)
birth weight are presented in Table I. Indications for induction are listed in Table II. While no significant differences in the distribution of primips and multips or gestational age were discovered, the distribution of indications for the induction were different. This is reflected in the difference in birth weight. Thus, the calculation of OR was performed after stratification in accordance with indications for induction. 6 (14%) in the Cerviprost-group and 3 (8%) in the cellulose-gel group went into active labour after the first gel application and additionally 12 (28%) and 6 (16%), respectively, after amniotomy 6 h later as the cervical score had improved. The mean change in the score was 3.4 (SE 0.4) in the Cerviprost group and 3.6 (0.4) in the cellulose-gel group. Thus, 42 and 24% were successfully induced after one application of Cerviprost or cellulose gel, respectively (Table III). TABLE II Indications for induction Cetviprost 43)
(n =
Results
(Cellulose-gel (n = 37)
(I) Prolonged pregnancy (more than 42 weeks) (II) Hypertension/IUGR (III) Others
Cellulose gel (n = 37)
28 (65%) 17 (46%) 6 (14%) 14 (38%) 9 (21%) 6 (16%) P = 0.05 (X*-test)
179 TABLE III Outcome of induction In labour (frequent contractions and/or amniotomy) after first application of the gel: Cetiiprost (n = 43) Indication
Cellulose gel tn = 37)
group
(I)
13 (46%) 2 (33%)
(III
3 (33%) OR = 2.2 (0.8-6.0) Mantel-Haenzel
(III)
Indication
6.5 (1.4-30.7) 1.3 (0.2-10.3)
2 (12%) 4 (29%)
0.5 (4.4-0.06)
3 (50%) test
In labour (frequent contractions and/or after second application of the gel
(1) (II) (III)
OR (95% CI)
amniotomy)
Cerviprost (n = 25)
Cellulose gel (n = 28)
13 (87%)
11 (73%)
group 3 (75%) 5 (83%) OR = 2.6 (0.7-9.9) Mantel-Haenzel
2.4 (0.3-15.5)
1.3 (0.08-19.8) 10.0 (0.4-243)
7 (70%) 1 (33%) test
After the second application of the actual gel, an additional 8 of the remaining 25 (32%) and 10 of the 28 (36%), respectively, went into active labour within 24 hours. Thirteen (52%) with a mean cervix score of 5.5 (SE 0.4) in the Cerviprost group and 9 (32%) with a mean score of 5.4 (SE 0.4) in the cellulose-gel group had amniotomy performed after 24 h. Thus, 4 (9%) in the Cerviprost group and 9 (24%) in the cellulose-gel group were considered failures, as the cervical state had not improved properly for amniotomy even after 24 h (P = 0.06). Uterine hypertonia did not occur in any case after gel application. Of the 67 women still in the study, 6 (15.4%) in the Cerviprost group and 8 (28.6%) in the cellulose-gel group (NS) required caesarean section. In 2 and 5 cases, respectively, impending fetal distress was the indication for the caesarean section. The overall caesarean section rate in the department was 8.9% over the same period. Among women induced by primary amniotomy, the rate was 17% (unpublished data). One child in each group had an Apgar score below 7 after 1 min, and after 5 min the value was 8 for both children. The rest had Apgar scores of 10 after 1 and 5 min. In both groups the
pH venous blood from the cord was 7.29 + 0.02. There were no serious perinatal complications in either of the two groups. Discussion Based on the high ORs, our results suggest that PGE, administered in a solid gel for cervical ripening and/or induction of labour is more efficient and safe than gel in a more liquid form. Thus, the liquid gel seems more liable to move extra-amniotically and/or intravaginally, thereby reducing its influence on the cervical tissue. The indications for induction in the two treatment groups were different in spite of the randomization procedure. More cases with hypertension and/or IUGR were represented in the cellulose-gel group. These complications were also reflected by the lower birth weight in that group. Statistical analysis by the stratification for indication-for-induction class, however, took these differences into account, thus enabling a valid comparison. Our success rate was not as high as those reported in other investigations using a solid gel [2,5]. This could be due to different techniques of application. It is more likely, however, that our
180
study population had a greater incidence of very unripe cervices. The cervical score is vitiated with lack of objective parameters. Thus, the present design which involved blind assessment of the cervical state by a midwife, not otherwise involved in the investigation, should eliminate such bias. The inequality among the study populations makes comparison of the results from other study difficult, especially because the induction rate and the percentage of those induced with PGE, gel because of a unripe cervix, are infrequently stated. The caesarean section rate among those treated with PGE, gel was not, as might have been expected [l], reduced compared with those induced with primary amniotomy. It may reflect that the gel-treated women had very unfavourable cervical states, which actually was the reason why they were not induced with primary amniotomy. It seems that the solid gel is preferable, since more caesarean sections were per-
formed in the cellulose-gel group, although the difference was not significant. References Rayburn WF. Prostaglandin E, gel for cervical ripening and induction of labour: a critical analysis. Am J Obstet Gynecol 1989;160:529-534. Ekman G, Forman A, Marsal K et al. Intravaginal versus intracervical application of prostaglandin E, in viscous gel for cervical priming and induction of labour at term in patients with an unfavorable cervical state. Am J Obstet Gynecol 1983;147:657-662. Legarth J, GuldbEk E, Secher NJ. The efficiency of prostaglandin E, vaginal suppositories versus intracervical prostaglandin gel for induction of labour in patients with unfavourable inducibility prospects. Eur J Obstet Gynecol Reprod Biol 1988;27:93-98. Stampe Sorensen S, Brocks V, Lenstrup C. Induction of labour and cervical ripening by intracervically prostaglandin E,. Obstet Gynecol 1985:65;110-114. Ulmsten AU, Wingerup L, Ekman G. Local application of prostaglandin E, for cervical ripening or induction of term labour. Clin Obstet Gynecol 1983;26:95-105.
