The controversies associated with managing women with a short cervix.
Preterm birth is the leading cause of neonatal mortality and morbidity. Prevention of preterm birth by optimal management of the woman with a short cervix is subsequently a high priority for clinicians. The various causes of preterm birth culminate in a final common pathway of uterine activity and cervical shortening. Consequently, the use of cervix length measurements to aid in prediction of preterm birth has become established. However, aspects of screening, diagnosis and management are debated. We present some of the common dilemmas facing clinicians.
To screen or not to screen?
The Australasian Society of Ultrasound in Medicine advises the cervical length should be assessed at the time of the mid trimester ultrasound.1 A short cervix diagnosed in asymptomatic women at the time of routine screening at 20–24 weeks is associated with an increased risk of preterm birth.2 Recent data appear to support the use of vaginal progesterone in preventing preterm birth in asymptomatic women with a mid-trimester ultrasound diagnosis of a short cervix. Hassan et al demonstrated, in their randomised controlled trial (RCT), vaginal progesterone reduced preterm birth before 28, 33 and 35 weeks, and ‘any neonatal morbidity or mortality event’.3 Controversy exists owing to the resource demands of universal screening and difficulties in standardising cervical length measurement.
How do you measure the length of the cervix?
Cervical length is typically measured by transvaginal ultrasound, after the bladder is emptied, which may deter some sonographers, and pregnant women, from performing the test. There is a lack of agreement among Australian specialists and sonographers on reporting on the cervical length, especially when confusion exists in the presence of prominent cervical mucous or funnelling.4 Transabdominal cervical measurement could potentially alleviate the need for transvaginal measurement, but is inaccurate and visualisation of the cervix can be difficult if there is a fetal part overlying the cervix.5 The cervical length can be dynamic and change with uterine contraction, fundal or probe pressure, leading to further inaccuracies in the measurement.
When should you place a rescue cerclage?
The decision to place an emergency or ‘rescue’ cerclage for an open cervix before 24 weeks should be made by an experienced obstetrician after considering the individual circumstances of the woman. In their guideline, the Royal College of Obstetricians and Gynaecologists (RCOG) cites evidence that rescue cerclage may delay delivery by a further five weeks, compared to expectant management or bed rest alone, and may be associated with a two-fold reduction in preterm delivery before 34 weeks.6 Concerns arise regarding the use of a rescue cerclage when faced with symptomatic women. While it is widely accepted that active labour, ongoing vaginal bleeding, premature rupture of membranes and chorioamnionitis are contraindications to cerclage6, it can be difficult to be reassured, in the presence of vague symptoms, that placement of cerclage is of more benefit than harm. Furthermore, although studies have shown rescue cerclage prolongs the pregnancy, there is limited evidence that this translates into an improvement in neonatal morbidity or mortality.6 Clinicians must also consider if a rescue cerclage keeps a baby in an unfavourable uterine environment or converts a miscarriage to an extremely preterm birth.
What dose of progesterone should be prescribed?
There is a lack of evidence to dictate the best dose and route of administration of progesterone for women with a short cervix. The intramuscular preparation of 17-alpha-hydroxyprogesterone commonly used in the US is not available in Australia. Vaginal progesterone has the perceived advantage of high uterine bioavailability and doses from 90–400mg have been used in studies showing their effectiveness in prevention of preterm birth. One meta-analysis showed no difference in efficacy in preventing preterm birth in studies that used 90–100mg or 200mg doses of progesterone (micronised progesterone or progesterone pessaries).7 In the absence of evidence that any dose is superior, local availability of supply and cost should be taken into account when prescribing.
Can a twin pregnancy be managed the same way?
It is tempting to extrapolate that the benefits of cerclage and progesterone recognised in singleton pregnancies complicated by a short cervix will be seen in multiple pregnancies. Contrary to this expectation, a recent Cochrane review demonstrated in a subgroup analysis (with substantial heterogeneity) that women with multiple pregnancy and a short cervix, who underwent an ‘ultrasound-indicated cerclage’, had an increased risk of low and very low birthweight babies and increased respiratory distress syndrome.8 In their guideline, the RCOG cites a number of small studies, including one RCT, that fail to show a benefit for cerclage in twin pregnancies, when making their recommendation against the use of cerclage in twin pregnancy.6
The use of progesterone in twin pregnancies complicated by a short cervix is even more controversial. In one meta-analysis, progesterone use in twin pregnancies did not significantly reduce preterm birth compared to placebo, but did reduce composite neonatal morbidity/mortality.7 Two further RCTs did not show any benefit of progesterone use compared with placebo in preventing preterm birth in twin pregnancies or twin pregnancies complicated by a short cervix.9,10 Given the lack of evidence for an effective treatment to prevent preterm birth in women with a multiple pregnancy and a short cervix, the value of measuring the cervical length at ultrasound in multiple pregnancies must be questioned. While the knowledge may guide decision-making about the appropriate centre for management or assist in the timing of steroid loading, it may also create an added layer of anxiety for women with an already high-risk pregnancy.
Can cervical pessaries help?
A 2013 Cochrane review of the evidence for the use of cervical pessaries in women with a short cervix included only one RCT of 385 women with a cervix length of 25mm or less who were randomised to cervical pessary or expectant management. The pessary group had a significantly lower rate of spontaneous preterm birth (<37 weeks) compared with expectant management.11 Further research would assist in supporting this finding and guiding clinicians.
Should I prescribe tocolytics?
Much interest exists in tocolytics, in the hope that suppression of uterine activity will prolong pregnancy and prevent the morbidity and mortality associated with preterm birth. In a general population at risk of preterm birth, there is no clear evidence that tocolytic drugs have a significant effect on preterm birth or perinatal or neonatal morbidity.12 Tocolytics are most widely employed to attempt to delay delivery for the purpose of in-utero transfer or steroid loading.
What about bed rest?
It is tempting to restrict women with a short cervix to bed rest in the hope that reduced activity and favourable gravitational forces will prolong the pregnancy. Unfortunately, evidence is lacking to support this notion and, in fact, some evidence exists to the contrary. One retrospective cohort study of women with a short cervix demonstrated 40 per cent of patients in the cohort were advised activity restriction and this same group had a higher number of preterm births compared to the cohort who were not advised to restrict activity.13 It is recommended to consider not only the woman’s individual circumstances, but also the potentially detrimental social, medical and economical impacts such a management strategy may have.6
Inpatient versus outpatient management
Further controversy exists about whether a woman with a short cervix should be allowed to be managed at home, when at risk of potential preterm birth. While bed rest is not recommended, it is possible that inpatient admission, with mobilisation privileges, could lead to better outcomes through immediate access to obstetric and neonatal care. Contrary to this hypothesis, a retrospective cohort study on hospitalisation in women with a short cervix showed that hospitalisation was an independent risk factor for cervical shortening, was associated with increased risk of earlier delivery and a shorter time from diagnosis to delivery.14 Further research is required to guide clinicians. It is also necessary to consider the woman’s gestation, comorbidities, resources available at the closest hospital and ability for timely access (which can be influenced by transport, distance from hospital, support people at home) when arranging the location of care.
Is it okay to have sex?
The effect of intercourse in women with a short cervix has not been rigorously studied with RCTs. In term patients, there is evidence that intercourse does not increase the rate of spontaneous onset of labour, as demonstrated by an RCT of women awaiting induction of labour at term.15 It is difficult to infer similar outcomes in preterm women with a short cervix, so advice should be individualised in the absence of guiding evidence, accounting for patient factors and preferences.
Clearly, there are many areas of uncertainty in the management of women with a short cervix. Hopefully, continuing research in this area of interest will help to guide future practice. In the meantime, it is important to individualise the care for the woman, taking into account her personal history and circumstances.