Six studies (n=653) were deemed to be of sufficient quality for inclusion in the review: two RCTs (n=170); two retrospective cohort studies (n=267); one prospective cohort study (n=135); and one matched case-control study (n=81). Allocation concealment and comparability and definition of groups were scored positively in the two RCTs. Criteria for other study designs were generally were scored positively.
Five of the six studies (including the two RCTs) found no difference in pre-term delivery when ultrasound-predicated and history-predicated cerclage groups were compared. The remaining prospective cohort study found that pre-term delivery below 30 weeks was lower in the ultrasound group, but this was not statistically significant.
In three of four studies (including one RCT) that reported pregnancy loss before 24 weeks, none showed a statistically significant difference between groups; there appeared to be a discrepancy in reporting of the fourth study. From the data presented, it was unclear whether a lower pregnancy loss or no pregnancy loss was achieved in the ultrasound-predicated cerclage group. The proportion of women who did not undergo cerclage following ultrasound management ranged from 40% to 68%.
Only the case-control study reported on cerclage-related complications. It concluded that there was no significant difference between study groups.