Transcervical Balloon Cervical Ripening and Risk of Uterine Rupture After Previous Cesarean Delivery: A Systematic Review and Meta-analysis.

OBJECTIVE

To estimate comparator-specific associations between transcervical balloon cervical ripening and complete uterine rupture during trial of labor after cesarean.

DATA SOURCES

PubMed/MEDLINE, Embase, CINAHL, Cochrane CENTRAL, Web of Science, and Google Scholar were searched from inception through June 2026 without language restriction, plus reference lists and trial registries (PROSPERO CRD420261444368).

STUDY ELIGIBILITY CRITERIA

Studies of trial of labor after a previous low-transverse cesarean delivery that used a Foley, single-balloon, or double-balloon catheter for cervical ripening were included. Comparative observational studies with spontaneous labor, oxytocin, or prostaglandins as comparators were included in the quantitative synthesis. Amniotomy alone was a prespecified eligible comparator; however, no study reported a separable amniotomy-only arm, precluding quantitative pooling of this comparison. Single-arm cohorts were summarized descriptively, while randomized trials were synthesized qualitatively. Primary analyses were restricted to studies reporting complete uterine rupture separately. Because balloon placement was followed by amniotomy, oxytocin, or both in nearly every protocol, the exposure is designated a balloon-based induction strategy rather than the catheter in isolation.

STUDY APPRAISAL AND SYNTHESIS METHODS

Risk of bias was assessed with ROBINS-I, the Newcastle-Ottawa Scale, and Cochrane RoB 2; certainty was rated with GRADE. Comparator-specific odds ratios (ORs) and absolute risk differences were estimated with Mantel-Haenszel methods, with Peto and DerSimonian-Laird models as sensitivity analyses; potentially overlapping Swedish cohorts were not pooled.

RESULTS

Eighteen nonrandomized studies and two randomized trials contributed 28,035 study observations, including 5,516 balloon exposures. Relative to prostaglandin induction, balloon ripening showed lower odds of complete uterine rupture across three cohorts (Mantel-Haenszel OR, 0.43; 95% confidence interval [CI], 0.26-0.72; risk difference, -19 per 1,000 [95% CI, -31 to -7] across four cohorts); a potentially overlapping Swedish cohort and a cohort reporting a combined rupture/dehiscence endpoint gave concordant estimates, and one trial recorded no ruptures. No difference was detected versus oxytocin alone (OR, 0.24; 95% CI, 0.05-1.18) or spontaneous labor (seven cohorts; OR, 1.29; 95% CI, 0.90-1.84; OR, 1.09 [95% CI, 0.73-1.62] excluding the most influential cohort). Across all pharmacologic comparators the OR was 0.40 (95% CI, 0.25-0.66), whereas dehiscence was reported more often after balloon-based induction (five studies; OR, 1.86; 95% CI, 1.06-3.29), with concordant estimates against oxytocin-based induction (OR, 1.80; 95% CI, 0.83-3.92), spontaneous labor (OR, 2.24; 95% CI, 0.91-5.52), and prostaglandins (OR, 1.09; 95% CI, 0.15-7.84; P = .80 for subgroup difference). Certainty was very low throughout.

CONCLUSIONS

A balloon-based induction strategy was not associated with an increase in complete uterine rupture compared with spontaneous labor or oxytocin alone and showed lower odds than prostaglandin induction. Because amniotomy, oxytocin, or both almost always followed catheter placement, these estimates cannot be attributed to the catheter alone. Mechanical ripening therefore remains reasonable in carefully selected candidates when induction is indicated, with counseling, continuous monitoring, and immediate access to emergency cesarean delivery.