Comparative Effectiveness of and Maternal Acceptability with Cervical Ripening using Cervical Massage Versus Membrane Sweep (CEASELESS): A Randomized Controlled Trial

Comparative Effectiveness of Acceptability with Cervical Ripening using Cervical Massage Versus Membrane Sweep (CEASELESS): A Controlled Trial. 6 Abstract Methods: This randomized controlled non-blinded clinical trial conducted at a leading tertiary obstetric care unit in Colombo, Sri Lanka, compared the maternal acceptability and effectiveness of membrane sweep (MS) and cervical massage (CM) in ripening uterine cervix to avoid formal induction of labour. MS, CM, and control (C) groups had 104, 106 and 102 women after randomization. Primiparous women underwent either MS or CM at 39 th and 40 th weeks while multiparous women underwent at 40 th week of gestation only. All were followed up until 24 hours postpartum. 100, 102 and 101 women in MS, CM and C groups respectively were included in the analysis. Results: Overall, MS significantly improved Modified Bishop’s Score (MBS) to ≥7 compared to C ( p= 0.0310) but not compared to CM ( p =0.2639). There was no significant improvement of MBS after CM compared to C ( p= 0.2795). Among primiparous ( p= 0.047) and multiparous ( p= 0.038) women separately, mean survival times without going into labour (MBS≥7) were significantly shorter after MS compared to C labour, emergency cesarean section rate in labour, postpartum hemorrhage, maternal fever, and APGAR score at 5 minutes. Maternal acceptability was assessed using a validated questionnaire consisting of 4 Likert scale items (S-CVI = 0.875; Cronbach’s Alfa= 0.876). Maternal acceptability score of CM was significantly higher than that of MS during overall analysis as well as during subgroup analysis according to parity ( p= 0.0011). Conclusions : MS is an effective adjunct to induction which ripens cervix, prevents formal induction of labour (NNT = 7), significantly reduces the duration of pregnancy, and shortens hospital stay for delivery overall. Although CM is more acceptable than MS and shortened hospital stay for delivery overall, it is not an effective method to ripen the uterine cervix or to prevent formal induction labour regardless of parity. This trial was self-funded. Trial Registration No - SLCTR/2020/003, Date of registration – 22/01/2020, Universal Trial Number - U 1111-1244-8026. effectiveness prevent Size of – MS group, n=61 group; multiparous group, in group) regardless unable hasten in effect of of gestation to had effects of a intervention compared to repeated interventions and the ideal interval between two are until the moment ≥7 –


Introduction
WHO Global Survey on maternal and perinatal health reports that 9.6% of all deliveries start with an induction worldwide, and Sri Lanka tops the list at 35.5% [1]. Induction of labour is the artificial stimulation of the uterus to start labour and consists of initiation of uterine contractions and cervical ripening [2]. The phrase 'induction of labour' is sometimes used to denote the artificial initiation of contractions on a favorable cervix. And the phrase 'cervical ripening' is used to denote a separate process of achieving a Modified Bishop's Score (MBS) of ≥7. In accordance to accepted definition of labour, it is logical to consider artificially initiating uterine contractions and cervical ripening as two components of process of induction of labour.
Methods of induction of labour are classified into mechanical and pharmacological categories. Mechanical methods of induction are membrane sweeping, cervical massage and balloon catheter. Principle of these mechanical methods of induction is to potentiate a local release of endogenous prostaglandin by physically stimulating the cervix.
Pharmacological methods of induction introduce exogenous prostaglandin achieving high local concentrations at the cervix. Prostaglandin leads to dilatation, shortening and softening of cervix through complex interactions with cervical cells and matrix [3]. In addition, Prostaglandin acts on myometrial cellular receptors triggering rhythmic uterine contractions which are augmented by oxytocin [4,5]. Balloon catheter, while not recommended for routine use by NICE, is widely utilized in Sri Lanka. Membrane sweep and cervical massage are considered as adjuncts to induction and not as formal methods of induction of labour in NICE guidance (CG70, 2008) 6 . NICE recommends offering membrane sweep to nulliparous women at 40 th and 41 st weeks of gestation, to parous women at 41 st week of gestation [6] and additional membrane sweeps offered as needed. NICE (CG70) also suggests that if the cervical OS is closed to admitting the examining finger, massaging around the cervix may achieve a similar effect. Boulvain et al concludes on the utility of membrane sweeping towards avoiding formal induction of labour (number needed to treat -8) [7]. But there are hardly any studies comparing membrane sweep to cervical massage.
The theoretical framework of acceptability of a clinical intervention in seven-fold. It consists of, affective attitude, burden, perceived effectiveness, ethicality, intervention coherence, opportunity costs, and self-efficacy. Affective attitude is the emotional reaction to the intervention, for example "I feel scared when I think about the intervention".
Burden refers to the effort or toll taken by the health care system for the implementation of an intervention.
Perceived effectiveness refers to the impression patients have about the effectiveness of an intervention. Ethicality means ensuring beneficence, non-maleficence, justice, and autonomy. Coherence of an intervention is the quality of being logical and consistent. Opportunity cost means the benefits lost elsewhere due to implementation of one intervention. Self-efficacy of an intervention refers to the ability of an intervention to achieve its aims on its own without and external compounding factors [8]. This randomized controlled non-blinded clinical trial compares the effectiveness and acceptability of cervical massage vs membrane sweep at cervical ripening.

Methods
This study was a non-blinded randomized controlled trial conducted in De Soysa Hospital for Women, Colombo 08, Sri Lanka from 20 th February 2020 to 04 th July 2020. Ethics approval was granted by the Ethics Review Committee of Faculty of Medicine, University of Colombo on 17 th October 2019 for application no Protocol-19-064 [9]. Trial was in the control group, a 15% difference was considered of clinical importance [11]. The minimum sample size required to detect such a difference at an 80% statistical power and at a 95% confidence interval was 100 per group.
Interventions compared were membrane sweep and cervical massage. Cervical Massage (CM) consisted of three circumferential passes of around the cervix massaging the cervix with examiner's examining fingers for 15-30 seconds. Membrane Sweep (MS) involved insertion of the operator's finger through the internal OS of cervix followed by three circumferential passes separating the chorio-amnion from the lower segment of the uterus 12 .
Ability to insert at least a part of one examining finger is a prerequisite for MS which was checked during screening to ensure all 3 interventions were possible in all women recruited. Principle investigator himself performed these interventions to ensure standardization throughout the study. MBS was confirmed to be ≤6 before initial intervention.
Pregnant women between 38 -40 weeks of gestation attending antenatal clinic of Professorial Obstetrics and Gynaecology Unit of Faculty of Medicine of University of Colombo were given sequential pink numbers as they were registered. A set of random numbers between 1-500 were computer generated, sorted in ascending order, and compared with the given pink numbers to select a simple random sample. Throughout the screening process, 349 nulliparous and parous women in total were assessed for eligibility according to predetermined eligibility criteria (Table 1). Thirty-seven women were excluded according to exclusion criteria and 312 women were recruited and randomized into the three study groups. Membrane sweeping cervical massage and no intervention (C) was allocated to three computer generated sets of random numbers. Equal sized red colored cards were prepared with the random number and its allocated method of induction printed overleaf. These cards were kept locked in an opaque container in the ward. After a study participant signed the consent form, the next available sequential number was extracted, and the method mentioned was administered as the method of induction of labour. vi. Had latex allergies (Latex gloves were worn during certain procedures during the study). The allocated method was performed once at 39 th and once at 40 th weeks of gestation for primiparous women and once at 40 th week of gestation for multiparous women. Study participants were followed up as outpatients regularly till 40+6 weeks where an MBS ≤6 was considered a failure and a formal induction method was administered as per unit protocol. If participants went into labour (MBS ≥7) they were followed up throughout labour and up to 24 hours postpartum. There were no changes made to pre-defined outcome measures ( Table 2).          Overall and among multiparous women, both MS and CM significantly reduced the total duration of hospital stay for delivery (MS vs C -p=0.0015, 95%CI 0.1752-0.7166; CM vs C -p=0.0197,95%CI 0.0603-0.6757) compared to C.

MS
Neither MS nor CM was able to reduce the hospital stay for delivery among primiparous women. Between MS, CM and C groups overall there were no significant differences with regards to synthetic oxytocin use, uterine hyperstimulation during labour, emergency cesarean section rate in labour, postpartum hemorrhage, maternal fever, and APGAR score at 5 minutes (Table 6).

Factor
Comparison MS CM C  Cronbach's Alfa = 0.876). Mean overall acceptability scores of MS and CM were 11.4400 and 13.2642 respectively (Table 7). Maternal acceptability score of CM was significantly higher than that of MS during overall analysis as well as during subgroup analysis according to parity (p=0.0011, 95% CI -0.7349 to 2.9145). comparing the effects of a single intervention compared to repeated interventions and the ideal interval between two such interventions are required. We measured survival time from intervention until the first moment an MBS ≥7 was detected. Cervical dilatation at that point ranged between 3cm -10cm. Therefore, it was obvious that the speed of cervical dilatation differed among women. We did not study how fast each method dilated the uterine cervix in this study. Designing such a study would be difficult because it requires prolonged admission and repeated vaginal examinations inconveniencing women and increasing the risk of infections, discomfort, withdrawal of consent, dropout from the study, emergency cesarean sections, fetal and maternal distress. Both MS and CM reduced the duration hospital stay for delivery compared to C among multiparous women only.

Conclusions
MS is an effective adjunct to induction which ripens uterine cervix, prevents formal induction of labour (NNT = 7) and significantly reduces the duration of pregnancy regardless of parity. It shortens hospital stay of multiparous women admitted for delivery as an added advantage. Although CM is more acceptable to pregnant women than MS and shortens hospital stay of multiparous women admitted for delivery it is not an effective method to ripen the uterine cervix or to prevent formal induction labour regardless of parity.

Declarations Ethics Approval and consent to participate
Ethics approval was granted by the Ethics Review Committee of Faculty of Medicine, University of Colombo on 17th October 2019 for application no Protocol-19-064. The study was performed in accordance with all the relevant guidelines and regulations adhering to the approved study protocol. Informed consent was obtained from all study