Key words
misoprostol - labour induction - Cytotec
Introduction
The past 10 years have seen publication of nearly a dozen meta-analyses on use of
misoprostol for labour induction and its efficacy and safety compared to oxytocin,
dinoprostone and balloon catheters [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. Misoprostol can be administered both vaginally and orally and is considered the
most effective labour induction agent in cases of immature cervix [2], [3]. Like all medicinal products (prostaglandin E2, oxytocin), misoprostol may also
cause overstimulations resulting in changes in the CTG pattern. The risk of an overstimulation
is increased in particular with
vaginal administration and at higher dosage levels [3], [5], [11], [12]. Use of misoprostol for labour induction in women with prior caesarean sections
is not recommended [13], [14], [15], [16]. The main reason for this is that the sole randomized, controlled study (comparison
of vaginal misoprostol versus oxytocin) was prematurely discontinued following the
occurrence of two uterus ruptures and recruitment of 17 patients [17]. There are also mainly retrospective studies in which misoprostol was administered
only vaginally using various dosages and intervals [18], [19]. According to a Cochrane analysis, not a single uterus rupture
occurred following oral administration of misoprostol in 158 pregnant women [20]. Notwithstanding the fact that misoprostol has now been authorized in various countries
for labour induction, discussions of this theme arise repeatedly in German-speaking
countries, where it is/was only authorized for prevention and treatment of gastroduodenal
ulcers, but not for labour induction. Recent articles in the German press have discussed
the legality of use of misoprostol for labour induction as a hot button issue with
a focus on the lack of marketing authorization, lack of recommendations on dosage
and use and potential associations with complications (e.g. overstimulations, pathological
CTG, poor child outcomes) [21]. It is indeed not known how often misoprostol is used for labour induction in German
clinics, with data also lacking on how the drug is prepared, administered and dosed.
The last survey from 2013 revealed
that many different regimens were in use [22].
The objective of this survey was thus to establish a record of labour induction with
misoprostol in German clinics and determine the impact of the negative reporting on
everyday obstetric practice.
Material and Methods
Participants and setting
In this cross-sectional study, invitations were extended to 635 obstetrics and gynaecology
departments in Germany. The respective department heads were provided with a link
to the questionnaire in an email together with a cover letter explaining the objective
and design of the study. The questionnaire was developed based on national and international
recommendations and guidelines. To ensure clarity and feasibility, the questionnaire
was pre-tested by three experienced obstetricians who had not contributed to development
of the survey. Modifications were made based on the resulting feedback. The results
of these pre-tests were not taken into account in the final data evaluation. The final
questionnaire comprised a total of 19 multiple choice and open questions covering
the following topics:
-
Demographic aspects of the respective obstetric units (3 questions)
-
Use of misoprostol before and after current discussion (3 questions)
-
Misoprostol sourcing (1 question)
-
Misoprostol administration (6 questions)
-
Misoprostol dosage schemes (2 questions)
-
Clinic-specific labour induction management (1 question)
-
Clinic-specific labour induction alternatives (3 questions)
The survey was conducted pseudonymously. A maximum of two reminders were sent out
14 and 21 days after the first invitation to participate. No personal data were recorded.
Data collection
Data were collected from 24 February to 20 March 2020 on a voluntary basis with no
remuneration of the participating clinics. An online survey format was chosen to facilitate
Germany-wide participation. The participation link was available at www.soscisurvey.de (source: Stelzl P, Survey [Version 3.2.14i], https://www.soscisurvey.de, accessed 20 December 2020). This online platform ensured a high level of data protection
because the IP addresses of the participating clinics were not recorded. Each participant
was allowed to fill out the questionnaire just once during the 26-day survey period.
To ensure complete responses, a warning message reminded participants to furnish missing
responses before they could access the next page of the survey. A total of 262 of
the 635 solicited clinics completed the questionnaire for a response rate of 41.3%.
At the end of the survey period, the
collected data were exported to an Excel table and forwarded for statistical
analysis.
Statistical analysis
All statistical calculations and analyses were done with the statistics program package
SAS, Release 9.4 (SAS Institute Inc., Cary, North Carolina, USA). Both absolute and
relative frequencies are indicated for all responses to the multiple choice questions.
To compare two or more groups, an χ2 test was used or (if the conditions for that test type were not met) a Fisherʼs exact
test. The Mann-Whitney U test, or for more than two groups the Kruskal-Wallis test,
was used for ordinal scale characteristics. Missing values or responses such as “donʼt
know” were not considered for the analyses. A multiple logistic regression analysis
was performed for each outcome to determine which combination of impact parameters
(medical care level, births per year, labour induction rate) provided the best explanation
for the respective outcome. Generally speaking, a test result was considered significant
if the p value was below 0.05. A significance level of 0.10 was assumed
for the multiple regression analyses to render the combined impact of multiple
factors more recognizable.
Results
[Table 1] presents the demographic parameters of the clinics that either used or did not use
misoprostol in Cytotec prior to the critical reporting. There was no statistically
significant difference in use of misoprostol between the different care levels (Perinatal
Centre Level I, Perinatal Centre Level II, Clinic with Perinatal Focus or Obstetric/Private
Clinic; p = 0.2104). Use of the substance was also independent of the number of births
(p = 0.1845). A trend towards significance was seen in the impact of the labour induction
rate (p = 0.0518). It was observed that misoprostol was used particularly often in
clinics with moderate labour induction rates of between 20 and 30%.
Table 1 Comparison of the misoprostol subgroups in terms of demographic parameters. Percentages
refer to the subgroup named in the respective table header.
|
|
Misoprostol (n = 221, 84%)
|
No misoprostol (n = 41, 16%)
|
p values
|
|
Perinatal Centre Level I
|
76 (89%)
|
9 (11%)
|
0.2104
|
|
Perinatal Centre Level II
|
20 (77%)
|
6 (23%)
|
|
Clinic with Perinatal Focus
|
38 (78%)
|
11 (22%)
|
|
Obstetric/Private Clinic
|
87 (85%)
|
15 (15%)
|
|
Births per year
|
|
|
0.1845
|
|
|
16 (94%)
|
1 (6%)
|
|
|
72 (78%)
|
20 (22%)
|
|
|
49 (84%)
|
9 (16%)
|
|
|
42 (86%)
|
7 (14%)
|
|
|
27 (90%)
|
3 (10%)
|
|
|
15 (94%)
|
1 (6%)
|
|
Labour induction rate
|
|
|
0.0518
|
|
|
69 (78%)
|
20 (22%)
|
|
|
143 (88%)
|
19 (12%)
|
|
|
9 (82%)
|
2 (18%)
|
Sourcing and method of administration of misoprostol
Sourcing and administration of misoprostol varied in the different clinics as shown
in [Table 2]. Misoprostol was prepared in the clinic pharmacy in about half of the cases (54%)
or imported from other countries (46%). These figures did not differ amongst the different
care levels (p = 0.8185). Sourcing of the agent did not correlate with either number
of births per year or labour induction rate (p = 0.8398/p = 0.8795, [Tables 8] and [9]).
Table 2 Comparison of medical care levels in terms of use of misoprostol prior to reporting.
Percentages refer to the care level indicated in the respective table header.
|
|
Total (n = 221)
|
PNC Level I (n = 76)
|
PNC Level II (n = 20)
|
Clinic with Perinatal Focus (n = 38)
|
Obstetric/Private Clinic (n = 87)
|
p values
|
|
* Multiple responses possible, ** Comparison of 50 µg initial dose versus other dose
|
|
What was your source for misoprostol?
|
|
Preparation in clinic pharmacy
|
111 (54%)
|
41 (56%)
|
11 (58%)
|
17 (47%)
|
42 (53%)
|
0.8185
|
|
Import from other country
|
96 (46%)
|
32 (44%)
|
8 (42%)
|
19 (53%)
|
37 (47%)
|
|
What method(s) of administration did you use?*
|
|
Oral (tablet)
|
209 (95%)
|
73 (96%)
|
20 (100%)
|
35 (92%)
|
81 (93%)
|
0.5806
|
|
Oral (liquid)
|
11 (5%)
|
3 (4%)
|
1 (5%)
|
2 (5%)
|
5 (6%)
|
0.9664
|
|
Vaginal (tablet)
|
56 (25%)
|
22 (29%)
|
3 (15%)
|
9 (24%)
|
22 (25%)
|
0.6351
|
|
Vaginal (insert)
|
26 (12%)
|
13 (17%)
|
2 (10%)
|
4 (11%)
|
7 (8%)
|
0.3660
|
|
What dosage levels were used?*
|
|
25 µg
|
106 (48%)
|
45 (59%)
|
12 (60%)
|
12 (32%)
|
37 (43%)
|
0.0175
|
|
50 µg
|
183 (83%)
|
67 (88%)
|
13 (65%)
|
33 (87%)
|
70 (80%)
|
0.0802
|
|
75 µg
|
14 (6%)
|
7 (9%)
|
0
|
1 (3%)
|
6 (7%)
|
0.4763
|
|
100 µg
|
104 (47%)
|
38 (50%)
|
4 (20%)
|
14 (37%)
|
48 (55%)
|
0.0183
|
|
> 100 µg
|
11 (5%)
|
5 (7%)
|
2 (10%)
|
2 (5%)
|
2 (2%)
|
0.2966
|
|
Were dosage levels of 50 µg and more in general use?
|
|
Yes
|
196 (89%)
|
70 (92%)
|
15 (75%)
|
34 (89%)
|
77 (89%)
|
0.2192
|
|
No
|
25 (11%)
|
6 (8%)
|
5 (25%)
|
4 (11%)
|
10 (11%)
|
|
How did you obtain the desired dosage form?
|
|
Capsule/tablet dosed accordingly
|
129 (59%)
|
46 (61%)
|
15 (75%)
|
20 (53%)
|
48 (57%)
|
0.7987
|
|
Division of tablet
|
77 (35%)
|
25 (33%)
|
5 (25%)
|
16 (42%)
|
31 (37%)
|
|
Dissolution of tablet plus liquid in appropriate amount
|
11 (5%)
|
4 (5%)
|
0
|
2 (5%)
|
5 (6%)
|
|
Other
|
0
|
0
|
0
|
0
|
0
|
|
What dosing intervals were used for misoprostol?*
|
|
Every 2 hours
|
16 (7%)
|
3 (4%)
|
5 (25%)
|
2 (5%)
|
6 (7%)
|
0.0316
|
|
Every 4 hours
|
139 (64%)
|
50 (66%)
|
6 (30%)
|
26 (68%)
|
57 (66%)
|
0.0113
|
|
Every 6 hours
|
66 (30%)
|
26 (34%)
|
7 (35%)
|
10 (26%)
|
23 (26%)
|
0.7124
|
|
Other
|
8 (4%)
|
1 (1%)
|
2 (10%)
|
0
|
5 (6%)
|
0.1064
|
|
How many doses of misoprostol were administered per patient and day?
|
|
One
|
2 (1%)
|
0
|
1 (5%)
|
0
|
1 (1%)
|
0.4921
|
|
Two
|
27 (12%)
|
7 (9%)
|
3 (15%)
|
8 (22%)
|
9 (11%)
|
|
Three
|
107 (49%)
|
37 (49%)
|
9 (45%)
|
17 (46%)
|
44 (52%)
|
|
Four
|
65 (30%)
|
26 (34%)
|
4 (20%)
|
10 (27%)
|
25 (29%)
|
|
Five
|
6 (3%)
|
3 (4%)
|
0
|
1 (3%)
|
2 (2%)
|
|
Six
|
8 (4%)
|
2 (3%)
|
2 (10%)
|
1 (3%)
|
3 (4%)
|
|
More than six
|
3 (1%)
|
1 (1%)
|
1 (5%)
|
0
|
1 (1%)
|
|
On how many successive days was (only) misoprostol used for labour induction?
|
|
1 day
|
11 (5%)
|
3 (4%)
|
1 (5%)
|
2 (5%)
|
5 (6%)
|
0.0078
|
|
2 days
|
103 (47%)
|
26 (34%)
|
11 (55%)
|
23 (62%)
|
43 (51%)
|
|
3 days
|
84 (39%)
|
34 (45%)
|
8 (40%)
|
9 (24%)
|
33 (39%)
|
|
> 3 days
|
20 (9%)
|
13 (17%)
|
0
|
3 (8%)
|
4 (5%)
|
|
What is the substance amount for the initial dose?
|
|
25 µg
|
90 (45%)
|
36 (51%)
|
11 (61%)
|
12 (35%)
|
31 (39%)
|
0.1129**
|
|
50 µg
|
110 (55%)
|
33 (47%)
|
7 (39%)
|
22 (63%)
|
48 (61%)
|
|
100 µg
|
1 (0.5%)
|
1 (1%)
|
0
|
0
|
0
|
Table 8 Dependence of various parameters on annual number of births. The percentages quantify
the proportion of clinics for which the header is true in each case.
|
|
Use of dosage levels of 50 µg and more
|
Pharmacy preparation
|
Initial dose 50 µg
|
No misoprostol after reporting
|
No CTG check before or after prostaglandin dose
|
Use of castor oil
|
|
Clinics, total
|
196 (89%)
|
111 (54%)
|
110 (55%)
|
35 (17%)
|
35 (14%)
|
114 (46%)
|
|
Births per year:
|
|
|
|
|
|
|
|
< 500
|
15 (94%)
|
8 (57%)
|
6 (46%)
|
8 (57%)
|
3 (20%)
|
8 (53%)
|
|
500 – 999
|
64 (89%)
|
37 (55%)
|
41 (61%)
|
9 (69%)
|
11 (13%)
|
55 (63%)
|
|
1000 – 1499
|
43 (88%)
|
20 (45%)
|
25 (53%)
|
7 (15%)
|
9 (14%)
|
23 (41%)
|
|
1500 – 1999
|
35 (83%)
|
25 (61%)
|
18 (49%)
|
6 (15%)
|
7 (15%)
|
12 (26%)
|
|
2000 – 3000
|
25 (93%)
|
15 (56%)
|
10 (43%)
|
4 (15%)
|
3 (11%)
|
10 (36%)
|
|
> 3000
|
14 (93%)
|
6 (43%)
|
10 (71%)
|
1 (7%)
|
3 (19%)
|
6 (38%)
|
|
p value
|
0.8900
|
0.8398
|
0.6025
|
0.0537
|
0.9677
|
0.0002
|
Table 9 Dependence of various parameters on labour induction rate. The percentages quantify
the proportion of clinics for which the header is true in each case.
|
|
Use of dosage levels of 50 µg and more
|
Pharmacy preparation
|
Initial dose 50 µg
|
No misoprostol after reporting
|
No CTG check before or after prostaglandin dose
|
Use of castor oil
|
|
Total
|
196 (89%)
|
111 (54%)
|
110 (55%)
|
35 (17%)
|
35 (14%)
|
114 (46%)
|
|
Rate:
|
|
|
|
|
|
|
|
< 20%
|
62 (90%)
|
35 (55%)
|
39 (62%)
|
14 (21%)
|
10 (12%)
|
48 (57%)
|
|
20 – 30%
|
125 (87%)
|
71 (53%)
|
64 (50%)
|
20 (15%)
|
24 (16%)
|
63 (41%)
|
|
> 30%
|
9 (100%)
|
5 (56%)
|
8 (78%)
|
1 (11%)
|
1 (10%)
|
3 (30%)
|
|
p value
|
0.8900
|
0.8795
|
0.3992
|
0.2388
|
0.5527
|
0.0117
|
Misoprostol was administered orally in tablet form in nearly all clinics (95%) ([Table 2]). It was additionally administered vaginally in tablet form in one clinic in four
(25%), and more rarely in the form of inserts (12%). Administration of misoprostol
dissolved in liquid for drinking was practised in only a few clinics (5%).
Dosage of misoprostol
Dosage of misoprostol varied in the different clinics. Generally speaking, dosage
levels of 50 µg and above were used in most clinics (89%) independent of medical care
level (p = 0.2192). Misoprostol dosage levels of 25 µg (48%), 50 µg (83%), 75 µg (6%),
100 µg (47%) and > 100 µg (5%) were used, whereby low doses of 25 µg were administered
mainly in perinatal centres and higher doses of 100 µg were administered with notable
frequency in obstetric clinics (p = 0.0175/p = 0.0183). In most clinics, the dosage
forms were delivered prefabricated as tablets/capsules (58%). In the other cases,
the Cytotec tablet was divided (35%) or dissolved in water (5%).
In most cases, misoprostol was administered every four hours (63%) or every six hours
(30%), more rarely every two hours (7%) or at other intervals (4%). In Level 2 Perinatal
Centres, misoprostol was administered more frequently, i.e. every two hours or, more
rarely, every four hours (p = 0.0316/p = 0.0113). Depending on the dosage interval
used, in most cases three (49%), four (30%) or two (12%) doses were administered per
day. Use of misoprostol only was reported for two (47%) or three (39%) successive
days, rarely for more than three days (9%) or for only one day (5%).
An initial dose of 50 µg (55%) or 25 µg (45%) was selected in nearly all clinics.
The initial dose was 100 µg in one clinic only. The decision by a given clinic to
begin with a dose of at least 50 µg was found to be independent of the respective
medical care level (p = 0.1129); number of births and rate of labour inductions also
did not influence this decision (p = 0.6025/p = 0.3922, [Tables 8] and [9]).
Impact of media reporting on use of misoprostol
[Table 3] shows the impact of the critical reporting on use of misoprostol. Use of misoprostol
was discontinued in 17% of the clinics. This was independent of medical care level
(p = 0.9436) and labour induction rate (p = 0.2388, [Table 9]). It was, however, observed that a high percentage (about 60%) of clinics with fewer
than 1000 births per year discontinued use of misoprostol following the reporting
(p = 0.0537, [Table 8]). The main reasons for this were worry about patient reactions, avoidance of having
to justify decisions and fear of legal consequences (40% in each of these categories).
In 31% of the clinics, further use was disallowed by the boss or clinic management.
Changes were also instituted in the clinics that continued using misoprostol: Increased
efforts to provide information characterized the main change (80% of cases), with
only a few cases of a
different induction scheme (6%) or a lower initial dose for induction (4%) being
introduced.
Table 3 Comparison of medical care levels in terms of use of misoprostol following critical
reporting. Percentages refer to the care level indicated in the respective table header.
|
|
Total (n = 211)
|
PNC Level I (n = 73)
|
PNC Level II (n = 20)
|
Clinic with Perinatal Focus (n = 35)
|
Obstetric/
Private Clinic (n = 83)
|
p values
|
|
* Multiple responses possible
|
|
Is use of misoprostol continuing subsequent to the critical reporting?
|
|
Yes
|
176 (83%)
|
61 (84%)
|
17 (85%)
|
28 (80%)
|
70 (84%)
|
0.9436
|
|
No
|
35 (17%)
|
12 (16%)
|
3 (15%)
|
7 (20%)
|
13 (16%)
|
|
What has changed?
|
|
We have discontinued its use (n = 35), …*
|
|
(n = 12)
|
(n = 3)
|
(n = 7)
|
(n = 13)
|
|
|
… because it was disallowed (boss/clinic management/…)
|
11 (31%)
|
4 (33%)
|
2 (67%)
|
1 (14%)
|
4 (31%)
|
0.4963
|
|
… because we fear legal consequences
|
14 (40%)
|
2 (17%)
|
1 (33%)
|
4 (57%)
|
7 (54%)
|
0.1936
|
|
… to avoid having to justify decisions to patients, due to worry about patient reactions,
|
14 (40%)
|
7 (58%)
|
1 (33%)
|
3 (43%)
|
3 (23%)
|
0.3404
|
|
We have continued its use (n = 176) …*
|
|
(n = 61)
|
(n = 17)
|
(n = 28)
|
(n = 70)
|
|
|
… in lower single doses
|
7 (4%)
|
1 (2%)
|
1 (6%)
|
1 (4%)
|
4 (6%)
|
0.5116
|
|
… with a different treatment scheme (lower total dose)
|
10 (6%)
|
4 (7%)
|
1 (6%)
|
2 (7%)
|
3 (4%)
|
0.8932
|
|
… as before, but we must increase efforts to provide information
|
141 (80%)
|
51 (84%)
|
12 (71%)
|
23 (82%)
|
55 (9%)
|
0.6566
|
|
Other
|
39 (22%)
|
12 (20%)
|
5 (29%)
|
5 (18%)
|
17 (24%)
|
0.7479
|
Management for labour induction
Management of misoprostol use for labour induction is presented in [Table 4]. Written information was provided regarding the off-label use of misoprostol in
81% of cases. In cases of labour induction post prior caesarean section – regardless
of the method used – written information was provided in only 24% of the clinics.
Labour induction is an inpatient procedure in almost all such cases (97%). Medical
labour induction is almost always carried out accompanied by CTG checks: This was
done in 78% and 88% of the participating clinics before administration of misoprostol
and prostaglandin E2 respectively. CTG checks were performed just as frequently after
administration of a dose of misoprostol (76%) and prostaglandin E2 (88%). Cases in
which no CTG check was performed before or after administration of a dose of prostaglandin
showed no dependence on medical care level, number of births or labour induction rate
(p = 0.8414, p = 0.9677,
p = 0.5527). No misoprostol was administered in the presence of contractions
in 146 clinics (59%). 159 clinics (64%) also administered no prostaglandin E2 if labour
contractions were present.
Table 4 Comparison of medical care levels in terms of management of misoprostol use for labour
induction.
|
|
Total (n = 249)
|
PNC Level I (n = 81)
|
PNC Level II (n = 26)
|
Clinic with Perinatal Focus (n = 44)
|
Obstetric/
Private Clinic (n = 98)
|
p values
|
|
Provision of written information on off-label use when misoprostol is used
|
202 (81%)
|
73
|
18
|
31
|
80
|
0.0188
|
|
Provision of written information on off-label use when labour is induced post prior
caesarean section (condition post sectio)
|
60 (24%)
|
26
|
8
|
9
|
17
|
0.1010
|
|
Outpatient labour induction with misoprostol possible
|
8 (3%)
|
2
|
1
|
2
|
3
|
0.8347
|
|
CTG check before misoprostol dose
|
193 (78%)
|
69
|
16
|
29
|
79
|
0.0149
|
|
CTG check before prostaglandin E2 dose
|
218 (88%)
|
70
|
23
|
40
|
85
|
0.8895
|
|
CTG check after misoprostol dose
|
189 (76%)
|
68
|
15
|
30
|
76
|
0.0272
|
|
CTG check after prostaglandin E2 dose
|
218 (88%)
|
70
|
23
|
41
|
84
|
0.6374
|
|
No CTG check before or after prostaglandin dose
|
35 (14%)
|
11
|
3
|
5
|
16
|
0.8414
|
|
No misoprostol dose in the presence of contractions
|
146 (59%)
|
51
|
14
|
27
|
54
|
0.6798
|
|
No prostaglandin E2 dose in the presence of labour contractions
|
159 (64%)
|
51
|
16
|
31
|
61
|
0.7943
|
|
Raising of oxytocin dose every 10 – 20 minutes
|
74 (30%)
|
19
|
7
|
12
|
36
|
0.2557
|
|
Raising of oxytocin dose every 30 – 60 minutes
|
102 (41%)
|
39
|
10
|
15
|
38
|
0.4154
|
|
Raising of oxytocin dose at intervals > 60 minutes
|
15 (6%)
|
5
|
5
|
2
|
3
|
0.8676
|
|
Raising of oxytocin dose until contractions occur at 2 – 3 minute intervals
|
22 (9%)
|
9
|
1
|
3
|
9
|
0.7609
|
|
Raising of oxytocin dose until contractions occur at 4 – 5 minute intervals
|
48 (19%)
|
16
|
7
|
13
|
12
|
0.0693
|
|
Discontinuation of oxytocin administration for labour induction after 5 hours
|
78 (31%)
|
23
|
10
|
10
|
35
|
0.3437
|
|
Discontinuation of oxytocin administration for labour induction after 5 – 10 hours
|
36 (14%)
|
11
|
3
|
10
|
12
|
0.3839
|
|
Discontinuation of oxytocin administration for labour induction after 10 – 15 hours
|
4 (2%)
|
0
|
1
|
2
|
1
|
0.1100
|
|
Discontinuation of oxytocin administration for labour induction based only on clinical
indication and CTG
|
82 (33%)
|
31
|
4
|
17
|
30
|
0.1353
|
|
Castor oil (outpatient or inpatient)
|
114 (46%)
|
25
|
11
|
18
|
60
|
0.0006
|
Differences between the clinics were seen in particular regarding the use of labour
cocktails (outpatient or inpatient) (p = 0.0006): Their administration is comparatively
frequent (61%) in obstetric clinics.
Alternative methods of labour induction
Alternative methods of labour induction in cases of immature cervix (Bishop Score
< 3) are presented in [Table 5]. Frequent inpatient approaches in this situation are prostaglandin E2 (vaginal 84%,
or cervical 56%) and balloon catheter (53%). Balloon catheters are used most frequently
in Level I Perinatal Centres (77%) and infrequently in Clinics with Perinatal Focus
(34%, p < 0.0001). Further options include dilapan (38%), labour cocktail (35%) and
even oxytocin (29%). Outpatient management covering a variety of methods is only rarely
offered (in fewer than 10% of the clinics).
Table 5 Comparison of medical care levels regarding alternatives to misoprostol in cases
of immature cervix (Bishop Score 0 – 3).
|
|
Total (n = 249)
|
PNC Level I (n = 81)
|
PNC Level II (n = 26)
|
Clinic with Perinatal Focus (n = 44)
|
Obstetric/
Private Clinic (n = 98)
|
p values
|
|
* Multiple responses possible
|
|
Dilapan (outpatient)
|
19 (8%)
|
5
|
4
|
1
|
9
|
0.1975
|
|
Dilapan (inpatient)
|
94 (38%)
|
28
|
11
|
20
|
35
|
0.6013
|
|
Balloon catheter (outpatient)
|
9 (4%)
|
3
|
2
|
1
|
3
|
0.6628
|
|
Balloon catheter (inpatient)
|
133 (53%)
|
62
|
14
|
15
|
42
|
< 0.0001
|
|
Prostaglandin E2/dinoprostone (gel, tablet, insert) (outpatient)
|
4 (2%)
|
2
|
0
|
1
|
1
|
0.7931
|
|
Prostaglandin E2/dinoprostone (gel, tablet, insert) (inpatient)
|
208 (84%)
|
66
|
20
|
39
|
83
|
0.5718
|
|
Prostaglandin E2/dinoprostone (cervical gel) (outpatient)
|
4 (2%)
|
2
|
0
|
1
|
1
|
0.7931
|
|
Prostaglandin E2/dinoprostone (cervical gel) (inpatient)
|
140 (56%)
|
46
|
16
|
23
|
55
|
0.8991
|
|
Oxytocin
|
72 (29%)
|
23
|
7
|
9
|
33
|
0.4449
|
|
Castor oil (outpatient)
|
12 (5%)
|
5
|
0
|
1
|
6
|
0.5788
|
|
Castor oil (inpatient)
|
87 (35%)
|
16
|
10
|
12
|
49
|
0.0002
|
|
Other
|
15 (6%)
|
3
|
4
|
1
|
7
|
0.1340
|
Oxytocin (85%), vaginal prostaglandin E2 (73%) and labour cocktail (39%) are most
frequently selected for induction when the cervix is mature ([Table 6]). Approaches using cervical prostaglandin E2 (31%), balloon catheter (24%) and dilapan
(8%) are also used. Labour cocktails in cases of mature cervix are administered more
frequently in obstetric clinics (p = 0.0431).
Table 6 Comparison of medical care levels regarding alternatives to misoprostol in cases
of mature cervix.
|
|
Total (n = 249)
|
PNC Level I (n = 81)
|
PNC Level II (n = 26)
|
Clinic with Perinatal Focus (n = 44)
|
Obstetric/
Private Clinic (n = 98)
|
p values
|
|
Dilapan (outpatient)
|
3 (1%)
|
1
|
1
|
0
|
1
|
0.5040
|
|
Dilapan (inpatient)
|
21 (8%)
|
5
|
3
|
6
|
7
|
0.4339
|
|
Balloon catheter (outpatient)
|
5 (2%)
|
1
|
2
|
0
|
2
|
0.2142
|
|
Balloon catheter (inpatient)
|
59 (24%)
|
25
|
4
|
7
|
23
|
0.1891
|
|
Prostaglandin E2/dinoprostone (gel, tablet, insert) (outpatient)
|
6 (2%)
|
1
|
0
|
2
|
3
|
0.5629
|
|
Prostaglandin E2/dinoprostone (gel, tablet, insert) (inpatient)
|
181 (73%)
|
60
|
22
|
28
|
71
|
0.2885
|
|
Prostaglandin E2/dinoprostone (cervical gel) (outpatient)
|
1 (0.4%)
|
0
|
0
|
0
|
1
|
1.0000
|
|
Prostaglandin E2/dinoprostone (cervical gel) (inpatient)
|
77 (31%)
|
24
|
6
|
9
|
38
|
0.1166
|
|
Oxytocin
|
211 (85%)
|
69
|
24
|
38
|
80
|
0.5755
|
|
Castor oil (outpatient)
|
11 (4%)
|
2
|
0
|
2
|
7
|
0.3920
|
|
Castor oil (inpatient)
|
97 (39%)
|
23
|
9
|
17
|
48
|
0.0431
|
|
Other
|
16 (6%)
|
3
|
2
|
2
|
9
|
0.4753
|
Labour induction in condition post sectio
If a patient history includes a caesarean section, the favoured approaches to induction
are oxytocin (63%) and vaginal prostaglandin E2 (61%) as well as the mechanical methods,
balloon catheter (49%, especially at Level I Perinatal Centres at 73%) and dilapan
(35%) ([Table 7]). Only 2% of the clinics generally eschew labour induction in condition post sectio.
Table 7 Comparison of the medical care levels regarding labour induction methods used in
condition post sectio caesarea.
|
|
Total (n = 243)
|
PNC Level I (n = 81)
|
PNC Level II (n = 26)
|
Clinic with Perinatal Focus (n = 42)
|
Obstetric/
Private Clinic (n = 94)
|
p values
|
|
Dilapan (outpatient)
|
6 (2%)
|
1
|
0
|
1
|
4
|
0.6422
|
|
Dilapan (inpatient)
|
86 (35%)
|
25
|
13
|
16
|
32
|
0.3391
|
|
Balloon catheter (outpatient)
|
7 (3%)
|
2
|
1
|
1
|
3
|
0.9472
|
|
Balloon catheter (inpatient)
|
118 (49%)
|
59
|
12
|
14
|
33
|
< 0.0001
|
|
Prostaglandin E2/dinoprostone (gel, tablet, insert) (outpatient)
|
4 (2%)
|
1
|
0
|
1
|
2
|
1.0000
|
|
Prostaglandin E2/dinoprostone (gel, tablet, insert) (inpatient)
|
149 (61%)
|
48
|
21
|
23
|
57
|
0.1666
|
|
Prostaglandin E2/dinoprostone (cervical gel) (outpatient)
|
2 (1%)
|
1
|
0
|
0
|
1
|
1.0000
|
|
Prostaglandin E2/dinoprostone (cervical gel) (inpatient)
|
65 (27%)
|
23
|
6
|
8
|
28
|
0.5713
|
|
Oxytocin
|
152 (63%)
|
49
|
18
|
27
|
58
|
0.8674
|
|
Castor oil (outpatient)
|
3 (1%)
|
0
|
1
|
1
|
1
|
0.2618
|
|
Castor oil (inpatient)
|
77 (32%)
|
19
|
8
|
12
|
38
|
0.1092
|
|
Other
|
24 (10%)
|
8
|
1
|
6
|
9
|
0.6060
|
Generation of a multiple statistical model using logistic regression analysis was
only feasible for the outcome “Use of labour cocktail”, whereby the significance level
was set at 0.10. The observation was made that annual birth count (p < 0.0001) and
medical care level (p = 0.0893) impact this parameter independently. For the other
parameters (initial dose 50 µg, use of dosages of 50 µg and more, no misoprostol after
reporting) a multiple analysis revealed that only a single parameter was significant
in each case (whereby the significance level was set at α = 0.10).
Discussion
41% of the 635 clinics solicited for this national survey, which represented different
obstetric care levels, completed the questionnaire. It can be assumed that this study
provides a representative overview of labour induction as practised in Germany. Misoprostol
was used for labour induction in most of the clinics when the survey was conducted.
This was true regardless of care level and clinic size. This confirms that misoprostol
represents a standard method of medical labour induction.
A survey in 2013 revealed that a majority of clinics (66%) were already using misoprostol
for labour induction at that time [22].
In most cases, misoprostol was prepared in the clinicʼs own pharmacy (54%) or imported
from another country (45%) and administered orally in tablet form (95%). It was also
administered vaginally in tablet form in one of four clinics (25%) and in some clinics
in the form of inserts (12%). However, the misoprostol insert which has marketing
authorization is now no longer available.
The desired dosage was ensured with prefabricated tablets/capsules (59%). The Cytotec
tablet (200 µg) was divided (35%) or dissolved in water (5%) in the remaining cases.
This manual division of the Cytotec tablets is an ill-advised procedure now condemned
as such by both the current guideline recommendations and a “Red Hand Letter” issued
by the Federal Institute for Drugs and Medical Devices (Bundesinstitut für Arzneimittel und Medizinprodukte – BfArM)
[13], [15], [16], [23]. Despite the WHO recommendation, dissolving the tablet in water should also not
be done due to the resulting imprecision as to stability and active pharmaceutical
ingredient concentration [24].
There are no uniform recommendations for misoprostol dosage. 25 µg doses are recommended
internationally and it is reported that lower dosage levels (up to 50 µg) are associated
with outcomes similar to those obtained with higher dosages (100 µg) [24]. A further decisive factor in addition to dosage is the route of administration:
A very large meta-analysis (611 studies, 31 different methods) confirmed that vaginal
misoprostol in a dosage of ≥ 50 µg resulted in more cases of overstimulation than
placebo (OR 4.40, 95% CI 2.22 – 7.94), but revealed no differences in the rate of
transfers to paediatric clinics (OR 0.85, 95% CI 0.57 – 1.23) [11]. Similar data resulted for oral administration of misoprostol in a dosage of ≥ 50 µg
per tablet (OR 2.85, 95% CI 1.41 – 5.20 and OR 0.83, 95% CI 0.55 – 1.20). The Swiss
Association of Gynaecology and Obstetrics (Schweizerische Gesellschaft für Gynäkologie und
Geburtshilfe), in their Expert Brief No. 63 from 2019, recommend dosage levels of 25 – 50 µg vaginally
and 20 – 50 µg orally [13]. The S2k Induction of Labour Guideline describes single doses of 25 – 100 µg as
possible [15], [16]. An oral misoprostol preparation slated to become available in Germany this year
has received marketing authorization for single doses up to 50 µg and a maximum daily
dose of 200 µg. This preparation is already available in Austria [25]. The survey from 2013 revealed that many different regimens were in use [22]. Studies on labour induction with misoprostol in German clinics also revealed a
variety of treatment schemes [26], [27], [28]. This diversity has remained unchanged: According to the
current survey, the misoprostol dosages most frequently used were 25 µg (48%),
50 µg (83%) and 100 µg (47%). Dosages > 100 µg (5%) were the exception and should
be avoided according to current recommendations. The initial dose in nearly all treatment
schemes was 50 µg or 25 µg. The interval between doses was in most cases 4 hours (64%)
or 6 hours (30%), resulting in most cases in three (49%), four (30%) or two daily
doses (12%). Use of the different dosages or intervals was determined to be independent
of medical care level.
Misoprostol use in this context was off-label when the survey was carried out, making
provision of appropriate information obligatory. This information was provided in
written form as well in 81% of cases. Provision of such information in written form
is generally recommended in cases of off-label use [13], [15], [16].
Induction of labour with misoprostol was almost always done in an inpatient setting
(97%). Despite this being practised in outpatient settings as well internationally
[29] this is discouraged in current recommendations. Medical labour induction should
be performed in an inpatient setting under CTG control [13], [15], [16].
Earlier guidelines, now out of date, recommended dosing of prostaglandins accompanied
by CTG checks and not using prostaglandins in the presence of contractions [30]. However, a pre-dosing CTG check was performed in only 78% of the clinics surveyed
when misoprostol was administered and in 88% when prostaglandin E2 was used. Percentages
of CTG checks after dosing of misoprostol and prostaglandin E2 were similar (76% and
88%). Also, misoprostol was administered despite the presence of contractions in 41%
of the clinics, which figure was 36% for prostaglandin E2. This practice must be viewed
critically, since prostaglandins can cause overstimulations. CTG checks before and
after dosing of prostaglandins, and doing without prostaglandins in the presence of
contractions, thus raise the safety level of this medical labour induction practice
and should be done [15], [16]. It turns out
that complications associated with misoprostol, and with prostaglandins in general,
are not a matter of dosage, but rather of medical labour induction management in the
broader sense. This certainly underscores the importance of the information provided
in the new S2k Guideline regarding these points.
Labour induction in condition post sectio is associated with a raised risk of uterus
rupture, even though the absolute risk level is low. Accordingly, both earlier and
current labour induction guidelines characterize labour induction post sectio caesarea
as a possible option [15], [16], [31], [32]. In the current survey, nearly all clinics reported performing labour induction
in this situation (98%). However, information in written form regarding off-label
use with the available methods is provided in only 24%. This aspect could become legally
relevant, for which reason provision of this information in written form is recommended
[15], [16].
The critical reporting on Cytotec led to discontinuation of use of misoprostol for
labour induction in 17% of the clinics. The main reasons for this were worry about
patient reactions, avoidance of having to justify decisions and fear of legal consequences
(40% in each of these categories). In many cases (31%) further use was disallowed
by the boss or clinic management. This is an impressive demonstration of the power
of the press to impact obstetric medical care. In smaller clinics in particular, which
depend on every single birth, discontinuation of the drug for this indication was
observed above all in clinics with fewer than 500 births, but the critical reporting
resulted only in increased efforts to provide information accordingly in other clinics
(80%). There were only a small number of cases of shifts to other induction treatment
schemes (6%) or reduction of individual doses (4%).
This situation should be viewed critically, since one of the alternative methods of
labour induction was listed as the labour cocktail. The labour cocktail, for its part,
is uniquely guilty of the aspects criticized in the press: It has no marketing authorization
and evidence of safety and efficacy are lacking – this despite its use for labour
induction over nearly a century [33], [34]. The benefit of this method is not evidence-based [35] and adverse effects/complications are known [36]. The labour cocktail is therefore not recommended for labour induction in international
guidelines [37]. This is a plausible consequence in view of the fact that the active pharmaceutical
ingredient, ricinoleic acid, achieves its effect on muscle cells in the uterus and
intestine via prostaglandin receptors, so that the same
potential adverse effects expected with use of misoprostol and prostaglandin E2
apply to it as well. Ricinus oil (castor oil) is therefore only suitable for labour
induction in an inpatient setting and within the context of studies [15], [16].
Since the labour cocktail is used above all in smaller clinics (Obstetric/Private
Clinic, p = 0.0006) with fewer than 1000 births per year and with lower labour induction
rates (< 20%, p = 0.0117), the worry is justified that precisely those clinics that
decide to discontinue use of misoprostol because of the critical reporting will increasingly
turn to use of the labour cocktail [38]. This represents a sacrifice of quality in medical labour induction and puts patients
and children at greater risk.
In summary, this survey provides a good current overview of labour induction as practised
in German clinics. Other surveys are in some cases dated (from 2013 [22]) or were intended primarily for midwives [38]. Since this study did not aim to determine complication rates, future studies should
analyse the spectrum of adverse effects/complications associated with a labour induction,
in particular when misoprostol is administered.
Conclusion
This study demonstrates impressively that misoprostol, and use of prostaglandins generally
speaking, represents an established method in Germany. It also points up the need
for further improvement of certain procedures (e.g. dosage of misoprostol, monitoring).
These imperatives have been known for some time, for which reason development of the
S2k Induction of Labour Guideline was initiated, leading to its publication in December
2020. The Guideline will contribute to improvements in the labour induction procedure.
Some of the criticism expressed in the media may be justified, but the way it was
presented is itself deserving of criticism. It was revealed that the reporting resulted
in discontinuation of use of misoprostol mainly in smaller clinics, giving rise to
the concern that poorly investigated methods such as administration of ricinus oil
(castor oil) will take its place. This would represent a sacrifice of quality in medical
labour induction and put patients and
children at greater risk.
