Septum resection for women of reproductive age with a septate uterus

Abstract

Rationale: Women with a septate uterus are at increased risk for subfertility, recurrent miscarriage, and preterm birth. Restoration of the anatomy of the uterus by hysteroscopic septum resection is an established intervention. This treatment has been assessed mainly in retrospective cohort studies, which suggested a positive effect on pregnancy outcomes. The major flaw in these studies is the before-after design, which will always favour the tested intervention.

Objectives: To evaluate the benefits and harms of septum resection in women of reproductive age with a septate uterus compared to expectant management.

Search methods: We searched the Cochrane Gynaecology and Fertility Group Specialised Register, CENTRAL, MEDLINE, Embase, and PsycINFO databases on 22 September 2025. We also searched trial registries and reference lists from relevant papers and contacted experts in the field for any additional studies. This is an update of a Cochrane review first published in 2011 and previously updated in 2017.

Eligibility criteria: We included randomised controlled trials (RCTs) and prospective and retrospective observational studies assessing the effect of septum resection versus no septum resection on reproductive outcomes and safety in women of reproductive age with a septate uterus. We excluded studies if they lacked a control population or had a before-after study design.

Outcomes: Critical outcomes were live birth rate and surgical complications. Important outcomes included ongoing pregnancy, clinical pregnancy, and miscarriage.

Risk of bias: We used the Cochrane RoB 1 tool to assess risk of bias in RCTs and the ROBINS-I tool to assess risk of bias in non-randomised studies.

Synthesis methods: We combined data using a fixed-effect model if studies were sufficiently similar. We did not combine RCTs with non-randomised studies. For dichotomous outcomes, we used the numbers of events in the control and intervention groups of each study to calculate Mantel-Haenszel odds ratios (ORs). For continuous outcomes, we used means and standard deviations to calculate mean differences (MD). We presented 95% confidence intervals (CI) for all outcomes using RevMan for statistical analysis. We used GRADE to assess the certainty of evidence for live birth, surgical complications, ongoing pregnancy, clinical pregnancy, and miscarriage.

Included studies: We included 13 studies, of which one was an RCT and 12 were non-randomised studies, all of which compared septum resection versus expectant management. In the RCT, 39 women received septum resection and 40 women received expectant management. In the 12 non-randomised studies, 1134 women received septum resection and 692 women received expectant management.

Synthesis of results: Results from the RCT suggest that septum resection may result in little to no difference in live birth compared to expectant management (OR 0.83, 95% CI 0.32 to 2.11; 1 study, 79 participants; low-certainty evidence). This suggests that if the observed average live birth per woman following expectant management is 35%, the chance of live birth with septum resection is between 15% and 53%. Based on results from non-randomised studies, we are uncertain whether septum resection increases live birth compared to expectant management (OR 1.15, 95% CI 0.90 to 1.48; 8 studies, 1383 participants; very low-certainty evidence). Among the 39 women who underwent septum resection in the RCT, two complications were reported: one uterine perforation and one residual septum. Of six non-randomised studies that reported on surgical complications, three described no complications, and three reported mainly uterine perforation, bleeding, or residual septum requiring repeat surgery. The remaining studies did not report on complications. Results from the RCT suggest there may be little or no difference in ongoing pregnancy between the septum resection group and the expectant management group (OR 0.93, 95% CI 0.37 to 2.35; 1 study, 79 participants; low-certainty evidence). Based on results from non-randomised studies, we are uncertain whether septum resection increases ongoing pregnancy rates compared to expectant management (OR 0.47, 95% CI 0.27 to 0.81; 1 study, 257 participants; very low-certainty evidence). Results from the RCT suggest there may be little or no difference in clinical pregnancy between the septum resection group and the expectant management group (OR 1.43, 95% CI 0.59 to 3.47; 1 study, 79 participants; low-certainty evidence). Based on results from non-randomised studies, we are uncertain whether septum resection increases the likelihood of clinical pregnancy compared to expectant management (OR 1.37, 95% CI 0.96 to 1.97; I² = 49%; 5 studies, 724 participants; very low-certainty evidence). Septum resection may increase the odds of miscarriage; however, the CI included the possibility of no effect (OR 2.75, 95% CI 0.86 to 8.84; 1 study, 79 participants; low-certainty evidence). Based on results from non-randomised studies, we are uncertain if septum resection decreases the odds of miscarriage compared to expectant management (OR 0.74, 95% CI 0.53 to 1.04; 7 studies, 1259 participants; very low-certainty evidence). We downgraded the certainty of the RCT evidence to low due to very serious imprecision, and the non-randomised studies evidence to very low due to very serious risk of bias and inconsistency.

Authors' conclusions: Based on the currently available RCT evidence, there may be little or no difference in live birth, ongoing pregnancy, and clinical pregnancy between septum resection and expectant management. Miscarriage may be slightly increased in the septum resection group, although the CI included the possibility of no effect. Based on results from non-randomised studies, we are uncertain whether septum resection affects pregnancy outcomes, as the certainty of the evidence was very low.

Funding: This review had no dedicated funding.

Registration: Protocol (2010): https://doi.org/10.1002/14651858.CD008576.pub2 Review (2011): https://doi.org/10.1002/14651858.CD008576.pub3 Review update (2017): https://doi.org/10.1002/14651858.CD008576.pub4.