Systematic review and network meta‐analysis of hydrosalpinx treatment before in vitro fertilization

ABSTRACT Objectives To compare different methods to treat hydrosalpinx, based on both ablative and non‐ablative approaches, in infertile patients before undergoing IVF‐ET. Methods Systematic review and network meta‐analysis (NMA) of comparisons between different treatments of hydrosalpinx in infertile patients undergoing IVF. Data sources: structured searches in common citation databases. Study inclusion criteria: peer‐reviewed randomized trials (RCT) or cohort studies comparing effects of salpingectomy, laparoscopic proximal tubal occlusion (LTO), insertion of intratubal device (ITD), sclerotherapy, ultrasound‐guided aspiration and no treatment, on live birth, ongoing pregnancy, clinical pregnancy as main outcomes, considering also miscarriage, ectopic pregnancy and complications as secondary outcomes. Principal NMA included RCT, and aggregated NMA of RCT and observational studies was carried out. Pooled effects have been estimated by Odds Ratio (OR) and its 95% confidence interval (CI) for direct and indirect‐mixed comparisons, derived from random‐effects models. Imprecision and heterogeneity of NMA estimations was assessed by comparison of its 95% CI with predefined intervals for clinically relevant size of effect (OR <0.9 or >1.1). Surface under the cumulative ranking curve (SUCRA) were used to predict treatment rankings for each outcome. Results Nine RCT were included in main analysis, plus 17 additional observational studies in additional analysis. NMA of RCT did not identify significant differences in the effect of compared treatments on live birth rate, and LTO was the option with the highest value of SUCRA (0.92, mean rank: 1.2). Salpingectomy and US‐aspiration associated to a significant increase of ongoing pregnancy rate compared to no treatment, according to NMA results (NMA OR: 4.35; 95% CI: 1.7, 11.14 and 2.8; 95% CI: 1.03, 7.58 respectively). Salpingectomy had the highest SUCRA value (0.88, mean rank: 1.4). NMA estimated significant increase of clinical pregnancy rate for salpingectomy compared with no treatment (NMA OR: 2.24; 95% CI: 1.3, 3.86) as well as for LTO versus no treatment (NMA OR: 2.55; 95% CI: 1.2, 5.41). Both comparisons were affected by a high grade of heterogeneity. For clinical pregnancy, LTO was the intervention with highest SUCRA (0.85; mean rank: 1.6). Regarding secondary outcomes, feasible NMA estimates did not support significant differences between treatments effects. According to aggregated NMA including randomized and observational studies, sclerotherapy showed significant beneficial effects on live birth rate compared to no treatment (NMA (OR: 4.6; 95% CI: 1.21, 17.46). Compared with untreated patients, the aggregated NMA estimates a higher ongoing pregnancy rate in patients treated with salpingectomy (NMA OR: 3.35; 95% CI: 2.12, 5.12), US‐aspiration (NMA OR: 2.16; 95% CI: 1.28, 3.65) and LTO (NMA OR: 2.46; 95% CI: 1.11, 5.43). Salpingectomy and LTO produced a higher beneficial effect compared to ITD, based on both direct and indirect comparisons. Salpingectomy obtained the highest SUCRA value in rank of effects on ongoing pregnancy (0.94; mean rank: 1.2). NMA found significant effects on clinical pregnancy for comparisons between the different active management procedures compared with no treatments, with the exception of ITD insertion. LTO had more increasing effect on clinical pregnancy rate compared with US‐aspiration (NMA OR: 2.04; 95% CI: 1.05, 3.97), while for the rest of the comparisons between procedures no significant differences were identified. NMA ranked LTO as the treatment with a highest SUCRA value (0.91; mean rank: 1.5). NMA prediction models identified LTO as best intervention to reduce miscarriage (SUCRA value: 0.84; mean rank: 1.8), as sclerotherapy as safer option in terms of ovarian response to IVF stimulation. Conclusions The present NMA fails to support the effectiveness of any option to treat hydrosalpinx before IVF in order to improve live birth rates, although the beneficial effect of salpingectomy and US aspirations on ongoing pregnancy rates and of both salpingectomy and LTO on clinical pregnancy rates emerges from our analysis, which reinforces current recommendations. Based on the aggregated analyses, sclerotherapy could be a promising alternative to conventional laparoscopic techniques, combined with a favorable safety profile. This article is protected by copyright. All rights reserved.