The Historical Practice of “Membrane Sweep” to Initiate Labour: Does it Have a Role in Contemporary Obstetric Practice?

Membrane stripping or sweeping involves a vaginal examination during which a finger is introduced into the cervical os to separate the membranes from the lower uterine segment by a circular movement. Membrane sweeping was first suggested by James Hamilton in 1810 [2] for labor induction. Mc Colgin [3] confirmed that the membrane sweep was associated with an increased activity of prostaglandin F2 and phospholipase A. A Bishop’s score [4] to assess the cervical length, dilatation, effacement, consistency position of cervix and fetal descent is used to predict the prognosis of the membrane sweep and the induction process.


History and Practice of "Membrane sweeping" (Artificial Separation of Membranes or "ASOM")
Membrane stripping or sweeping involves a vaginal examination during which a finger is introduced into the cervical os to separate the membranes from the lower uterine segment by a circular movement. Membrane sweeping was first suggested by James Hamilton in 1810 [2] for labor induction. Mc Colgin [3] confirmed that the membrane sweep was associated with an increased activity of prostaglandin F2 and phospholipase A. A Bishop's score [4] to assess the cervical length, dilatation, effacement, consistency position of cervix and fetal descent is used to predict the prognosis of the membrane sweep and the induction process.
What are the Potential Risks of the "Membrane Sweeping"?
The mucosa of the vagina is continuous with the skin of the thigh, perineum and has an anatomical proximity to the anal canal and the rectum. The pregnant uterus is anatomically connected to the vagina through the cervix. This makes it more susceptible to colonization and infection by bacteria that normally reside in the vagina and rectum during pregnancy and in labor. The function of the cervix is to provide mechanical strength to retain the growing fetus until term. In addition, a long and a closed uterine cervix along with the mucous plug within the cervical canal provides a mechanical barrier which prevents any ascending infection from the vagina into the amniotic cavity [5]. However, recent evidence has shown that the endocervical epithelium and the cervical mucous plug have significant innate and adaptive immune functions [6]. Any disturbance to the closed cervix can lead to dislodgement of the protective mucous plug and thereby promoting the microbial ascent through the cervix into the amniotic cavity ( Figure 1). Approximately 40 years ago, it was shown that bacteria could cause intra-amniotic infection even in the presence of "intact" amniotic membranes [7]. It has also been reported that markers of inflammation found within the amniotic cavity in approximately Global Journal of Reproductive Medicine 26% of suspected cases of intra amniotic infection following amniocentesis in women without the rupture of membranes [8]. It has also been shown that the degree of severity of intra-amniotic infection correlates to poor neonatal outcomes [9]. Therefore, the earlier assumption that the likelihood of intra-amniotic infection was low due to the "mechanical barrier" provided by the amniotic membranes when the "Membrane Sweep" was introduced into clinical practice 200 years ago, can no longer be accepted in the light of current scientific evidence. Therefore, one should not fall prey to the false sense of reassurance by assuming that the ASOM is safe, because the amniotic membranes are intact. Several studies have confirmed that the presence of inflammatory mediators within the amniotic cavity is associated with perinatal brain injury [10][11][12][13]. During the ASOM, it is likely that many organisms may be introduced by the clinician's finger, through the cervical canal, adjacent to amniotic membranes. Therefore, the "bacterial load" that would initiate the inflammation and subsequent infection would be much higher following the "Membrane Sweep". It is well known that in addition to the virulence of the organism, the bacterial load is an independent risk factor which increases the severity of any infection in humans. Moreover, studies show that Prostaglandin F 2 Alpha was elevated in 40.2% of women who had preterm labor with intact membrane and was an independent risk factor for intra amniotic inflammation and infection [14]. As ASOM has been shown to release prostaglandins in term pregnancies, it may increase the likelihood of intra amniotic inflammation and infection.

Scientific Evidence Regarding the Effectiveness of "Membrane Sweeping"
A recent Cochrane review [15] suggested that the women who had membrane sweep were more likely to have spontaneous onset of labor but found no clear difference in unassisted vaginal births. There were no clear differences between the groups for caesarean section, instrumental vaginal births or serious illness or death of the mother or baby. NICE guidelines recommend that the women should be offered a membrane sweep as an adjunct to formal induction of labor [16]. The STRIP-G study showed an adverse neonatal outcome in 5.9% of cases in women with vaginal colonization of Group B Streptococcus (GBS) identified on a high vaginal swab [17]. It is important to appreciate that the recent Cochrane Review [15] which compared membrane sweep with no or sham treatment concluded that the likelihood of spontaneous labour following the membrane sweep was only 21%. There was no statistically significant difference in the rate of caesarean sections or perinatal outcomes. A recent randomized controlled study [18] has concluded that the difference in onset of spontaneous labor in those who underwent a "membrane sweep" was only 15%.

What do Observational Data Show?
In the United Kingdom, the ASOM is widely practiced, despite the Cochrane Systematic Review concluding in 2001 [1] that "Routine use of sweeping of membranes from 38 weeks of pregnancy onwards does not seem to produce clinically important benefits". The "Early Notification Report" from the NHS Resolution has highlighted in 2019 that perinatal infection was responsible for 15.6% of all admissions to the neonatal unit in the UK [19]. We conducted an observational study in 57 cases of confirmed acute chorioamnionitis, by placental histopathology results over a period of 6 years (Unpublished Data). Out of the 57 women, 50 (87.7%) did have a membrane sweep between 24-48 hours of birth and 54 women (94.7%) had a caesarean section with their babies being admitted to the neonatal unit for suspected or confirmed sepsis receiving antibiotics. 47 women (87%) were readmitted with wound infection. Because the "Membrane Sweep" is widely practiced in the UK, the data from the NHS Resolution, and our retrospective observational study (Unpublished Data) suggest the need to re-visit this historical practice. Whilst it may continue to have a place in obstetric practice in low-resource settings, where prostaglandins are not freely available for induction of labour due to costs due to the potential detrimental effects (Table 1), the continued use of the ASOM in well-resourced settings need to be questioned. Most of the current guidelines on CTG Interpretation have been designed to detect intrapartum hypoxia and not fetal sepsis, the CTG abnormalities observed in chorioamnionitis such as absence of cycling and the "Zig Zag" Patterns [20][21] may be easily missed. if guidelines such as the NICE which are based on "pattern recognition"with arbitrary parameters are used. Therefore, fetuses with ongoing inflammatory damage or chorioamnionitis following the "Membrane Sweep" may be missed at the time of admission or during labor.

Action Detrimental Effects Possible Harm
Clinician's finger is inserted through the cervical canal to reach the amniotic membranes Disruption of cervical mucous Introduction of the bacteria from the maternal vagina into the space above the internal cervical os "Sweeping Action" of the clinician's finger, separating the membranes from the cervix" Blood is a good culture medium for bacteria, and it also provides neutrophils and phagocytes, at the site of amniotic membranes, which can initiate an inflammatory response.

c) A "large dose" of the bacteria is pushed above the internal cervical os
Possibility of over-riding fetal host response, and causing inflammation and infection Delayed time interval between the "Membrane Sweep" and the onset of labourd time interval between the "Membrane Sweep" and the onset of labour A longer incubation period which facilitates the colonisation and growth of organisms, when the woman is sent home, and not monitored.
Possible increase in the risks of maternal and fetal infection.

Conclusion
Membrane sweep was introduced into clinical practice in the 17th century before the availability of chemical or mechanical methods for cervical ripening. However, in modern obstetric practice, these are freely available in well-resourced settings. Therefore, the continued historical practice of the "Membrane sweep" which could not only lead to interruption in the barrier mechanism provided by the uterine cervix, but can also cause other detrimental effects, should be questioned. The magnitude of benefit of initiating a spontaneous onset of labour (approximately 21%), should be weighed against potential harmful effects (Table  1) of the "Membrane Sweep", and women should be counselled accordingly. Evidence-based clinical practice should always trump historical obstetric practices with unproven benefits and potential risks to ensure "first do no harm" to our patients.