Chapter 4.1 56 To assess the risk of bias (ROB) for non-randomised studies (NRS), the ROBINS-I (Risk Of Bias In Non-randomized Studies) tool was used.114 For randomised controlled trials (RCT), we used a revised tool for assessing the risk of bias in randomised trials (RoB 2).115 Reviewers (ASP and GP) independently rated for each study any domain for overall risk of bias as low, moderate, serious, or critical in ROBINS-I and as low risk, some concerns, or high risk in RoB 2. The quality of evidence was assessed using the GRADE framework (Grading of Recommendations, Assessment, Development and Evaluations).116 Data collection and analysis Two reviewers (ASP and GP) independently extracted all the details, as displayed in the collection form. A Microsoft Excel® data collection sheet (Microsoft, Redmond, Washington, USA) was developed to record the data.117 Reviewers were not blinded to author or publication source. The data were compared between the two reviewers. Discrepancies required agreement between the reviewers, and if necessary, a third reviewer (JDP) adjudicated. In case of overlapping study populations, the largest and most complete datasets were included. Patient and study characteristics and the risk of bias were reported. All comparisons and outcomes were described in detail in accordance with established recommendations. 118,119 A quantitative meta-analysis was not performed considering the design of most of the included studies (single-arm) and the heterogeneity of outcome measurements. This systematic review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and SWiM (Synthesis without meta-analysis) guidelines for systematic reviews.118,120,121 The protocol was prospectively registered on PROSPERO (CRD42021267941). RESULTS Search results A total of 11.953 citations were identified by database searching and 10 potentially relevant articles by searching the grey literature (Figure 1). After removing duplicates, 7130 articles were screened by title and abstract. Full-text evaluation was performed on 46 publications and eventually 9 were included.37,106-110,122-124 Study characteristics Table 1 summarises the characteristics of the 9 included studies. One study was a randomised trial.106 The other reports were all case series, ranging in size from 10122,123 to 224109 participants. Follow-up ranged from 4 weeks to two years after the index procedure (mean 9.7 months ± 7.3
RkJQdWJsaXNoZXIy MTk4NDMw