PropertyValue
is nif:broaderContext of
nif:broaderContext
is schema:hasPart of
schema:isPartOf
nif:isString
  • Data were collected from women attending specialist prematurity antenatal clinics in two UK tertiary referral centres between 2001 and 2012. Women were referred for trans-vaginal (TV) USS cervical surveillance when they had a history of PTB (defined as delivery at 24+0 to 36+6 weeks gestation), mid-trimester loss (defined as miscarriage at 14+0 to 23+6 weeks gestation), or cervical treatment (knife cone biopsy or large loop excision of the transformation zone following abnormal cervical cytology). A retrospective cohort study of women with a singleton pregnancy undergoing USS-indicated cerclage was performed. USS-indicated cerclage was defined as a suture being placed in the index pregnancy when CL was ≤ 25 mm and the external os remained closed. Women received either nylon or mersilene sutures. All women were re-scanned within 14 days of cerclage placement. Demographic data, CL before and after insertion of cerclage and distance of cerclage from external cervical os, were collated. Outcome measures were delivery prior to 34 and 37 completed weeks gestation. This study was exempt from requiring ethical approval under the UK Health and Social Care Act 2011 which states that research involving anonymised clinical data is excluded from research ethics committee review. Data were anonymised by JC and SC (at Queen Charlotte’s and Chelsea Hospital) and MC and SC (at St. Thomas’ Hospital). The authors interacted with the patients when they performed the ultrasound scans or inserted cervical cerclages. The standardised method of obtaining CL as described by Berghella and recommended by the Cervical Length Education and Review (CLEAR) programme (Perinatal Quality Foundation, OK, USA) was used [7, 8]. The woman was asked to empty her bladder and then the probe was inserted into the anterior fornix of the vagina to obtain a sagittal long-axis view of the cervical canal. Care was taken not to exert pressure on the cervix and inadvertently elongate it. The CL was measured from the internal to the external os and the distance between the cervical suture (if inserted) and the internal os was also measured. At least three measurements were obtained for each dimension and the shortest was used in the analysis. One of two senior obstetricians (AS, PB) inserted all cerclages using either the McDonald or modified Shirodkar technique where the bladder is dissected from the anterior cervix and the suture ‘buried’ underneath. The latter technique was used where examination under anaesthetic suggested that bladder dissection would be needed to ensure satisfactory placement of the stitch. The cohort was initially categorically divided according to distance of cerclage in millimetres (mm) from the external os (cerclage height) at first USS after suture insertion. The relative risks (RR) of PTB at cerclage heights of <10mm and <15mm, relative to the remainder of the cohort, were calculated. Cerclage height was also analysed as a continuous variable and association with subsequent delivery at <34 and <37 weeks gestation was assessed using logistic regression. A decision tree analysis [9] was also performed to select the optimal cerclage height threshold to predict PTB <34 weeks and <37 weeks gestation. Partition tree analysis was completed using R package “rpart”: function “rpart.” Examining absolute cerclage height does not take account of the range of CLs in the cohort, such that poorer outcomes in women with sutures placed <10 mm from the external os may merely reflect the presence of an already shorter cervix, and thus an increased risk of preterm delivery due to reduced initial total CL. In order to take account of the different CLs within the cohort, we examined the gestation at delivery according to cerclage height as a percentage of total CL, as observed at the first USS (<14 days) after suture insertion. The RR of preterm delivery was therefore calculated in those with a cerclage in the distal half of the cervix compared with those with sutures in the upper half of the cervix. Cerclage height as a percentage of CL was also assessed as a continuous variable using logistic regression. We further hypothesised that women with more distally placed sutures may have a greater RR of preterm delivery because a shorter cerclage height reflects a smaller initial CL. We therefore assessed gestational age at delivery in women relative to their CL at the time that a decision for cervical cerclage was made, and compared outcomes in those with initial CLs ≤10 mm and 11–20 mm with those women with the ‘longest’ CLs in our cohort (21–25 mm). Initial CL was also assessed as a continuous variable using logistic regression. Furthermore, in a previous observational study undertaken by our group, we demonstrated that CL increases following cerclage insertion (mean increase 11.5 mm two weeks post-insertion, 95% CI 5.9–17.1; p<0.001) [10]. We therefore examined CL at the first TV USS after cerclage insertion and compared outcomes in those whose CL remained ≤25 mm with those whose CL ‘normalised’ to >25mm following cerclage, we also assessed post-stitch CL as a continuous variable. We then compared outcomes in women whose CL did not change or shortened following cerclage with those whose CL increased by any amount, and also assessed if ‘change in CL’ as a continuous variable was associated with PTB. Statistical analysis was carried out using STATA v11 (Texas, USA) unless otherwise stated.
rdf:type