Abstract
An assessment of the efficacy and satisfaction of women in active labour having digital cervical stretching compared to women who did not have this intervention. Ours was a randomised controlled trial at a tertiary centre in India. Low-risk women at term with vertex presentation in active labour with ruptured membranes and cervical dilation of 4–6 cm were included. Stretching to delivery interval was 247.5 ± 158.2 min in the intervention group and 265.5 ± 158.4 in the control group. The mode of delivery, incidence of cervical tear, and maternal, fetal and neonatal complications were similar in both groups. The Labour and Delivery Satisfaction Index (LADSI) was similar in both groups. While no significant discomfort was perceived with stretching, it does not appear to expedite labour.
Keywords
Introduction
Interventions that could shorten the duration of active phase of labour should be assessed for efficacy. Shortening the duration of active labour is useful especially when the fetal heart rate tracing is abnormal. Digital stretching of the cervix to achieve early vaginal delivery is an intervention practised by many midwives and obstetricians as it is thought to expedite delivery. However, it is not recommended by some caregivers asserting that it is associated with patient discomfort. Others argue that a shorter period of labour may decrease the duration of pain and earlier delivery of the fetus. There are no studies to prove its efficacy to date.
Ferguson’s reflex 1 describes a complex neuroendocrine positive bio-feedback phenomenon wherein pressure on the cervix and vaginal wall during labour causes the pituitary release of oxytocin, which in turn increases pressure at the cervix, leading to a self- sustaining cycle that should ultimately result in fetal delivery. Oxytocin then acts on myometrial receptors leading to myometrial contraction and further positive feedback on the reflex. Stretching a well effaced cervix may also mechanically dilate it and thereby shorten the time taken till full dilation thus facilitating descent of the presenting part. Since the main apprehension concerning this intervention is of patient discomfort, evaluating its benefit is paramount. Several questionnaires 2 exist to evaluate satisfaction of care during labour and birth: the Likert score 3 is simple and quick, and the Labour and Delivery Satisfaction Index (LADSI) 4 is a more detailed 38-item questionnaire.
Materials and methods
Ours was an unmasked randomised controlled trial conducted at a tertiary care teaching hospital in South India which has approximately 14,000 deliveries per year. The study was approved by the Institutional Review Board and Ethics committee of the institution under CTRI/2014/08/004899. Low-risk women at term (37–41 weeks) with vertex presentation and no antenatal risk factors in active labour with ruptured membranes were recruited between 1 July 2014 and 30 June 2015 after informed consent. Thus, women with one or more previous Caesarean deliveries or any uterine surgery, multiple pregnancy, fetal growth restriction, an abnormal fetal heart rate, antepartum haemorrhage, preterm labour or a non-vertex presentation were excluded.
Permuted-block randomisation, with block sizes of 2, 4 and 6, was performed using SAS version 9.3.1 (SAS Institute, Cary, NC, USA). Serially numbered, opaque, sealed envelopes containing the group allocation were opened in a central research office. The sample size was calculated for a power of 80% and 5% significance, assuming a mean difference of 60 min between the duration of labour with or without digital stretching.
Rupture of membranes, active labour and cervical dilation of 4–6 cm were confirmed by the principal investigator or co-investigators. The latter examinations were standardised using a cardboard box with circular cut-outs, each with diameters of 4, 5 and 6 cm. The index and middle fingers of the hand not performing the vaginal examination were placed in the cut-out during assessment. If randomised to the interventional group, digital cervical stretching was carried out lasting two whole contractions; otherwise the patient was left alone. The group to which the woman was randomised was not revealed to the labour room team of doctors/nurses responsible for her care. The details of the delivery, pain and satisfaction scores were collected by the research officer, who was also masked as to the allocated group, on the second post-natal day. A detailed assessment of the patient’s satisfaction was done using the LADSI 4 and Likert score. 3 We also used nine questions from the original complete LADSI questionnaire which were relevant to the study to obtain an abridged version and this was compared between the groups. Adverse events and data were monitored by the principal investigator and the Data and Safety Monitoring Board (DSMB) of the institution.
The primary outcome studied was intervention to delivery interval.
The secondary outcomes were mode of delivery, traumatic postpartum haemorrhage due to cervical tear, need for oxytocin, uterine hyperstimulation causing fetal distress, trace abnormality, chorioamnionitis, Apgar score <7 at 5 min, need for resuscitation, admission to neonatal intensive care, need for ventilation, need for antibiotics in the neonate and duration of hospitalisation.
Statistical analysis was done using SPSS version 17 (SPSS Inc., Chicago, IL, USA). A P value <0.05 was considered significant for all analyses. Intention to treat analyses were done. Descriptive measures such as mean, median and standard deviation were used for continuous variables, Fisher’s exact test was used for comparing categorical variables and Mann–Whitney U test was used for comparison of continuous variables.
Allocation sequence was concealed from researchers before analysis. OMIT Cronbach’s α was calculated to assess reliability of the original and the abridged questionnaire.
Results
A total of 264 women were assessed for eligibility (Figure 1) and 196 women agreed to participate. Ninety-eight women were randomised for stretching and 98 as controls. The mean age, numbers of primigravid, numbers in low income group, gestational age at randomisation, body mass index, in spontaneous labour, cervical dilation, occipito-posterior presentation, presence of caput, moulding, meconium staining of liquor, duration from rupture of membranes to randomisation, use of amnioinfusion,
5
use of epidural and neonatal birth weight were similar in both groups (Table 1).
Participant flow diagram. Baseline characteristics. BMI, body mass index; GA, gestational age; NNT/NNH, numbers needed to treat/numbers needed to harm; R, randomisation; SD, standard deviation.
The station of the fetal head in the intervention group was higher than in the control group and this was statistically significant. However, statistical analysis on the outcomes adjusting for this difference did not show any difference in the two groups.
The primary outcome Table 2, namely, the intervention to delivery interval, as well as the number of women who underwent Caesarean section, instrument and normal vaginal delivery, oxytocin indication, uterine hyperstimulation causing fetal distress, trace abnormalities, chorioamnionitis, Apgar score <7 at 5 min, need for resuscitation, admission to neonatal intensive care, need for ventilation, need for antibiotics in the neonate and mean duration of hospitalisation were all similar in both groups. None of the women in the trial had postpartum haemorrhage from cervical trauma.
Outcomes.
Patient and caregiver’s satisfaction.
BMI, body mass index; GA, gestational age; LADSI, Labour and Delivery Satisfaction Index; NICU, Neonatal Intensive Care Unit; PPH, postpartum haemorrhage; R, randomisation; SD, standard deviation.
Discussion
There is very little information on cervical dilation to expedite labour. Mechanical dilation using instruments to shorten the active phase of labour was assessed almost three decades ago6,7 and was thought to be beneficial. However, no further studies have been done after this. None of the studies examined digital cervical dilation.
Our small study showed an 18-min difference between the interventional group and the control, which was not statistically significant. The major limitation of our study was that it was small and so not adequately powered to show a significant difference in the two groups. In the presence of a potentially abnormal fetal heart trace, fetal scalp pH assessment is done at 20-min intervals to look for a drop in pH. Even a shortening of 18 min in the active phase may therefore have a role in managing labour in low-resource settings, where facilities for scalp pH or lactate are not available.
The strength of our study was that it was a randomised controlled trial ruling out any concern of bias. The most important concern with the use of this intervention was discomfort to the woman. Hence, we used the elaborate LADSI and the abridged score in addition to the Likert score. LADSI and Likert scores were similar in both groups. LADSI measured the technical and caring components of patient satisfaction. Though this tool is reliable, it has sufficient internal validity (Cronbach’s α of 0.5) only with the use of the total score. When sub-components of the tool are used, this value falls to 0.3. When using this tool in our study, we found that with the total scores there was no difference in the two groups. The content validity of only nine questions truly reflected our intervention. Hence, we also used the abridged version of this questionnaire. When we used the scores for these sub-components, we found insufficient internal validity as predicted by the authors of LADSI. Cronbach’s α for the complete questionnaire in our study was 0.4 and 0.2 for the abridged version. However, the scores were similar in both groups.
We speculate that there was no difference in efficacy in the two groups because the intervention was performed when the cervix was less pliable. A larger study could be done when the cervix is dilated to >6 cm to show if digital dilation would then be more effective.
Footnotes
Acknowledgements
The authors thank the Research Officers Nayana John and M. Nirmala for their valuable efforts, the mothers and children who participated in the study, and the midwives, registrars, obstetricians and neonatologists who cared for the women and babies.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.