Keywords
castor oil - induction of labor - multiparity - postdate pregnancies - inpatients
Schlüsselwörter
Rizinusöl - Geburtseinleitung - Multiparität - Überschreitung des errechneten Geburtstermins
- stationäre Patientinnen
Introduction
Induction of labor (IOL) is indicated when the benefits of delivery outweigh the benefits
of prolonged pregnancy [1]
[2]
[3]
[4]
[5] or the risk of an adverse outcome increases with further prolongation of pregnancy,
like in cases of diabetic pregnancies, macrosomia, or placental insufficiency [6]. As shown by Grobman et al., elective IOL at 39 weeks of gestation in low-risk women
does not increase the risk of perinatal adverse events but significantly reduces the
number of caesarean sections (CS) compared with expectant management [7]. However, women often still perceive induction of labor as an interference with
the natural course of labor and evaluate it negatively on an emotional level.
The hormone inducing labor is prostaglandin. Thus, current induction methods are mainly
based on drug administration containing prostaglandins or mechanical devices triggering
the release of endogenous prostaglandins [2]. Labor induction using prostaglandins has shown to be effective and safe in numerous
studies [8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
and thus administration of prostaglandins is recommended to induce labor in most international
guidelines [16]. Depending on the route of administration or the type of synthetic prostaglandin
used, the safety profile of mechanical induction has been shown to be more favorable
but may not be as effective as using prostaglandins for IOL [13].
Castor oil, extracted from the Ricinus communis plant, was the first reported medical procedure to induce labor in the first half
of the last century and was already used by the ancient Egyptians [17]
[18]. Castor oil exerts its effects through ricinoleic acid, a hydroxylated fatty acid.
Research on mice has shown that ricinoleic acid specifically activates the EP3 prostanoid
receptor [19]
[20]. It has been shown to not only have a laxative effect but is also effective on the
uterus to induce labor [2]. In the 1960s, oxytocin replaced castor oil in obstetric care primarily to mitigate
gastrointestinal side effects [18]. Misoprostol, developed in 1973 for treating gastric ulcers, entered the market
in the mid-1980s and was described as mimicking natural labor [17]
[21]. Misoprostol is now the most commonly used drug in the setting of labor induction
at term and has been increasingly criticized recently by the public.
Both as a result of the public debate and independently of it, castor oil is increasingly
explicitly requested by women as an alternative to prostaglandins, as it is accepted
as an “old household remedy”. Consequently, castor oil cocktail is often applied in
outpatient settings at home, ahead of hospital admission for labor induction. It has
been shown that depending on the quantity and quality of castor oil used to induce
labor, effectiveness and severity of side effects do vary [22], leaving it to be a risky practice, especially without monitoring of fetal and maternal
well-being.
To diminish the occurrence of unmonitored outpatient inductions utilizing castor oil
and to accommodate women’s preferences, we began offering castor oil as one of the
options for labor induction in our routine clinical practice, using a standardized
recipe containing solely quality-tested castor oil. Based on findings from a prior
perspective, randomized, double-blind, placebo-controlled clinical trial conducted
by Gilada et al. [23], which demonstrated the ineffectiveness of castor oil in primiparous women, we restricted
its administration to multiparous women.
The objective of this study was to assess the safety and effectiveness of castor oil
for labor induction specifically in multiparous women analysing the real-world data
collected in our cohort.
Material and Methods
In a retrospective analysis, we compared a cohort of pregnant women receiving a castor
oil cocktail for IOL with a cohort of pregnant women induced by guideline recommended
methods. Subjects in the control group were identified by 1 : 2 matching by parity,
age of the mother, and year of treatment. We included data of singleton pregnancies
after completed 37 weeks of gestation indicated for IOL at our tertiary care obstetric
unit between 2014 and 2021. We excluded cases with a history of caesarean section
and primiparous women.
In our institution induction by castor oil cocktail is offered to multiparous women
asking for alternative methods of labor induction since 2010. It was implicated into
our institutional guidelines and consent forms in 2012. Women receiving castor oil
cocktail for labor induction all consented and signed for application of this method.
The castor oil cocktail was prepared according to our standardized recipe ([Fig. 1]) using quality-tested castor oil from the university hospital’s pharmacy. If onset
of labor has not occurred within 12 hours after consumption of the cocktail, IOL was
continued using guideline standard methods, like mechanical IOL with a double balloon
catheter, misoprostol or dinoprostone. Oral misoprostol was given at an initial dose
of 50 µg, followed by subsequent doses of 100 µg administered every 4 h for a maximum
of 48 h. 10 mg dinoprostone was inserted vaginally releasing 0.3 mg of dinoprostone
per hour, for a maximum application time of 48 hours. The vaginally placed double
balloon catheter remained for 12 hours, followed by prostaglandins if onset of labor
was not established. Monitoring during IOL was performed according to the German guidelines
irrespective of the method applied.
Fig. 1
Castor oil cocktail recipe.
As primary endpoints we analyzed the rate of vaginal deliveries and perinatal outcome
data. Secondary outcome parameters described are the frequency of cases in which labor
was not induced by the castor oil cocktail after 12 hours and the duration from initiation
of induction to delivery.
Data collection
Clinical data were collected from the electronic patient records. Clinical information
collected included maternal age, gravidity, parity, body mass index (BMI), gestational
age at delivery, reasons for and methods of IOL. BMI was calculated from maternal
height and pre-pregnancy weight. Gestational age was calculated using the last menstrual
period or crown-rump-length retrieved from earliest available ultrasound.
Perinatal outcome included mode of delivery, application of epidural analgesia, non-reassuring
fetal status, frequency of fetal scalp blood testing, time interval from induction
to delivery, umbilical cord arterial pH, Apgar score, and postnatal admission to neonatal
intensive care unit (NICU). The induction to delivery interval was calculated from
the time the woman was admitted to the labor ward for IOL to the time of delivery.
For further analysis, the cases were grouped as follows: delivery on the day of admission,
delivery within two days of admission and delivery later than two days after admission.
Data handling and analysis was in adherence with the Declaration of Helsinki. Ethical
approval for this study was given by the ethical committee of our institution (2022–2679-Daten).
Statistical evaluation
Statistical analysis was performed with SPSS 27.0 (IBM Corp., released 2020. IBM SPSS
Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp.). Chi2 test or Fisher exact test were used to compare categorical data. Most of the continuous
data were not normally distributed; therefore, our data are presented using median
and interquartile range. Nonparametric tests were used to compare continuous data
between groups. A p value < 0.05 was considered to indicate statistical significance
(2-tailed).
Results
During the study period from 2014 to 2021, 1140 multiparous women meeting the inclusion
criteria underwent IOL at our university hospital. 148 (13%) requested and received
a castor oil cocktail for IOL. Therefore, the matched control group, which was induced
using established methods, consisted of 296 women who met the relevant criteria.
Baseline characteristics
[Table 1] shows maternal and pregnancy characteristics. Median age, gravidity, and parity
did not differ between groups. Gestational age at delivery was higher in the castor
oil group (median 286 days vs. 281 days; p < 0.01), BMI was significantly lower in
the castor oil group (median 22.7, min. 16, max. 40 vs. median 25.1, min. 16, max.
55; p < 0.01). Predominantly, IOL was initiated due to late or post term pregnancy,
premature rupture of membranes (PROM), placental insufficiency, estimated fetal weight
exceeding 4 kg, diabetes mellitus (DM), request of the mother or hypertensive diseases
in pregnancy. The most common cause of IOL was post-date pregnancy, which was significantly
more common in the castor oil group (57% vs. 32%; p < 0.01), followed by PROM, which
was significantly more common in the group of women receiving established methods
for IOL (18% vs. 4%; p < 0.01).
Table 1
Maternal and pregnancy characteristics.
|
Variable
|
Castor oil cocktail
|
Established methods
|
p value
|
|
Data are n (%) or median and interquartile range (IQR) unless otherwise specified
BMI = Body Mass Index; PROM = Premature Rupture Of Membranes
p < 0.05 is significant and bold
|
|
Total, n
|
148
|
296
|
|
|
Median age in years, median (IQR)
|
34 (30–36)
|
33 (30–36)
|
0.86
|
|
BMI before pregnancy in kg/m², median (IQR)
|
22.7 (20.8–25.7)
|
25.1 (22.1–29.6)
|
< 0.01
|
|
Gestational age in days, median (IQR)
|
286 (281–289)
|
281 (274–287)
|
< 0.01
|
|
Gestational age in weeks, median (IQR)
|
40 (40–41)
|
40 (39–40)
|
< 0.01
|
|
Gravidity, median (IQR)
|
2 (2–3)
|
3 (2–3)
|
0.71
|
|
Parity, median (IQR)
|
1 (1–2)
|
1 (1–2)
|
0.99
|
|
Indications for induction of labor
|
|
|
|
|
Post-date pregnancy
|
84 (57)
|
94 (32)
|
< 0.01
|
|
PROM
|
6 (4)
|
54 (18)
|
< 0.01
|
|
Placental insufficiency
|
24 (16)
|
50 (17)
|
0.89
|
|
Estimated fetal weight exceeding 4 kg
|
5 (3)
|
25 (8)
|
0.46
|
|
Insulin dependent diabetes mellitus
|
7 (5)
|
14 (5)
|
0.99
|
|
Request of the mother
|
4 (3)
|
18 (6)
|
0.16
|
|
Hypertensive diseases in pregnancy
|
7 (5)
|
13 (4)
|
0.99
|
|
Others
|
8 (5)
|
21 (7)
|
0.55
|
|
Unknown
|
3 (2)
|
7 (2)
|
0.99
|
Results of labor induction
Onset of labor was established in 109 (74%) women receiving solely a castor oil cocktail.
In contrast, the castor oil cocktail did not induce labor in 39 women (26%) after
12 hours. Therefore, IOL was continued with oral misoprostol in 33 cases, three with
a double balloon catheter and one with vaginal dinoprostone. The remaining methods
of IOL are displayed in [Fig. 2].
In the control group 178 (60%) women received misoprostol for IOL, in 71 (24%) a double
balloon catheter was placed, 32 (11%) women received dinoprostone, seven (2%) women
oxytocin, and five (2%) women were induced by amniotomy. Of the group being induced
by double balloon catheter, this method failed to establish labor in 54 (76%) women.
IOL was continued with either misoprostol in 46 women or dinoprostone in eight women
([Fig. 2]).
Fig. 2
Applied methods of labor induction in the study cohort.
Perinatal outcome data
Perinatal outcome data are summarized in [Table 2]. The rate of vaginal deliveries in both groups (93% vs. 89%) is comparable. Overall,
a good perinatal outcome was observed. The induction to delivery interval was significantly
shorter in women receiving a castor oil cocktail for IOL in comparison to the control
group (p = 0.04). The number of cases assigned to the group delivering in “less than
one day” was significantly higher in the castor oil cocktail group than in the control
group (46% vs. 34%; p = 0.01). Control cases were more often assigned to the subgroup
“one to two days” (p = 0.04). There was no difference in the proportion of cases assigned
to the group “more than two days” (p = 0.74).
Table 2
Perinatal outcome.
|
Variable
|
Castor oil cocktail
|
Established methods
|
p value
|
|
Data are n (%) or median and interquartile range (IQR) unless otherwise specified.
p < 0.05 is significant and bold
|
|
Mode of delivery
|
|
|
0.35
|
|
Vaginal
|
137 (93)
|
262 (89)
|
|
|
Operative Vaginal Delivery
|
5 (3)
|
12 (4)
|
|
|
Caesarean section
|
6 (4)
|
22 (7)
|
|
|
Amniotomy
|
8 (5)
|
24 (8)
|
0.33
|
|
Epidural analgesia
|
11 (7)
|
29 (10)
|
0.39
|
|
Non-reassuring fetal status
|
32 (22)
|
64 (22)
|
0.99
|
|
Fetal scalp blood testing, n (%)
|
7 (5)
|
12 (4)
|
0.80
|
|
Induction to delivery interval, n (%)
|
|
|
0.04
|
|
< 1 day
|
68 (46)
|
99 (34)
|
0.01
|
|
1–2 days
|
64 (44)
|
160 (55)
|
0.04
|
|
> 2 days
|
14 (10)
|
32 (11)
|
0.74
|
|
Arterial pH, median (IQR)
|
7.23 (7.18–7.29)
|
7.22 (7.17–7.28)
|
0.35
|
|
1-min APGAR, median (IQR)
|
9 (8–9)
|
9 (8–9)
|
0.17
|
|
5-min APGAR, median (IQR)
|
10 (9–10)
|
9 (9–10)
|
0.24
|
|
10-min APGAR, median (IQR)
|
10 (10)
|
10 (9–10)
|
0.28
|
|
NICU admission
|
3 (2)
|
24 (8)
|
0.01
|
There were significantly more neonates admitted to the NICU in the control group (8%
vs. 2%; p = 0.01) ([Table 2]). Indications for NICU admission are depicted in [Table 3]. The most common reason was respiratory distress in both groups.
Table 3
Comparison of indications for NICU admission.
|
Variable
|
Castor oil cocktail (n = 3)
|
Established methods (n = 24)
|
p value
|
|
Data are n (%) or median and interquartile range (IQR) unless otherwise specified.
p < 0.05 is significant and bold
|
|
Indication for admission
|
|
|
0.05
|
|
Respiratory Distress
|
2 (67)
|
14 (58)
|
|
|
Hypoglycemia
|
0
|
3 (13)
|
|
|
Neonatal infection
|
1 (33)
|
0
|
|
|
Congenital malformation
|
0
|
4 (16)
|
|
|
Others
|
0
|
3 (13)
|
|
Discussion
The objective of this study was to evaluate the everyday practice of IOL by using
the consumption of castor oil cocktails as a frequently requested alternative to conventional
methods, aiming to generate real-world data on the safety and effectiveness of castor
oil.
We could convincingly demonstrate the safety and effectiveness of castor oil cocktail
for IOL in a cohort of 148 multiparous women who received a castor oil cocktail compared
to 296 multiparous women induced by conventional methods. In both groups there was
a high rate of vaginal deliveries with a corresponding low rate of caesarean section
(4% vs. 6%) which did not differ between groups ([Table 2]). The German perinatal statistics of 2022 report a caesarean section rate of 4%
for the Robson-category 3 including term deliveries in multiparous women, which corresponds
to our data. Neonates in the castor oil group were at significantly lower risk to
be admitted to NICU (2% vs. 8%; p = 0.01). However, this is most likely explained
by the higher, although not statistically significant, amount of uncomplicated pregnancies
in the castor oil cocktail group ([Table 2]). The overall percentage of neonates requiring NICU admission in our hospital is
7%, consistent with the Cochrane review on IOL methods, where oral misoprostol induction
resulted in an 8% admission rate [13]. Finally, the induction-to-delivery interval was significantly shorter following
castor oil cocktail initiation for IOL.
While several studies demonstrate the effectiveness and safety of castor oil cocktails
[22]
[24]
[25]
[26]
[27]
[28]
[29]
[30], a retrospective cohort study by Boel et al. found no difference in labor duration,
perinatal outcomes, or adverse events between women receiving castor oil and those
managed expectantly [31]. Gilada et al. conducted a
prospective, randomized, double-blind, placebo-controlled clinical trial demonstrating
the safety and efficacy of castor oil for IOL in multiparous post-date women, with
similar perinatal outcomes and adverse effects between groups. However, this effect
was not observed in primiparous women [23]. Furthermore, Knaus et al. demonstrated a significant interaction between castor
oil and parity, with a higher rate of cesarean sections in primiparous women receiving
castor oil cocktail [32]. Consequently, castor oil cocktail is exclusively offered to multiparous women in
our hospital.
Despite limited studies on the safety and efficacy of castor oil and although current
guidelines do explicitly not recommend its use, it is widely administered by midwives
to stimulate labor. A national survey of members of the American College of Nurse-Midwives
revealed that 93% reported using castor oil to stimulate labor [22]. The refusal to employ castor oil cocktails in clinical settings often leads to
unsupervised consumption at home, where non-standardized recipes prevail and the quality
of castor oil preparations remains untested by certified methods. Our data confirm
the safety and effectiveness of IOL applying castor oil cocktail in a cohort of multiparous
pregnant women induced at term using a standard recipe with quality tested castor
oil in a clinical setting of a tertiary care hospital.
The main limitations of this study include its retrospective, single-center design
and the general reliability of electronic patient records. However, a notable strength
is the relatively large cohort of multiparous women. Including high-risk cases resulted
in a heterogeneous study cohort, further strengthening our study. As a result of the
matching criteria applied, the study cohort differed in BMI and gestational age at
birth. However, since mean BMI was in both groups below the threshold for obesity
and the difference in gestational age of birth was not of clinical relevance we did
not consider this as a confounder. Additionally, all women received a castor oil cocktail
prepared according to a standardized recipe containing quality-tested castor oil,
ensuring consistent administration of the pharmacologically active substance. Compared
to other studies, we used a lower volume of castor oil (20 ml vs. 60 ml) [23]
[24]. Due to the retrospective nature of the study, the induction-to-delivery interval
was calculated as the difference between admission and delivery dates, rather than
in hours, which is another limitation of our study. However, our data show that the
use of castor oil for labor induction does not lead to a prolongation of IOL. While
the lack of documentation of adverse effects or patient satisfaction may be considered
a further limitation, the authors can report from their clinical experience that side
effects appeared no more frequent in castor oil-induced women, consistent with the
meta-analysis by Moradi et al. [24].
Conclusions
Our study findings present compelling evidence supporting the safety and efficacy
of employing a castor oil cocktail for labor induction in multiparous women at term
within a tertiary care setting. A critical prerequisite for utilizing such a cocktail
is the adoption of a standardized preparation comprising exclusively quality-tested
castor oil. Nonetheless, additional research is imperative before considering the
incorporation of this method into current guidelines. Our data not only establish
a solid foundation for future clinical trials but also underscore the necessity of
strict adherence to a precisely defined protocol, as evidenced by the consistently
safe outcomes observed.
