We Since multiple valve trials have already
We followed the preferred reporting items for systematic review and meta-analyses (PRISMA) guidelines for this meta-analysis 5.
Search StrategyLiterature search was performed using PubMed, COCHRANE database, OVID, and Google Scholar using the following search terms: bronchoscopic lung volume reduction, lung volume reduction coils, coils, lung volume reduction valves, endobronchial valves, one-way valves, lung volume reduction sealants, bronchial thermal vapor ablation, lung sealants, biological lung volume reduction, BioLVR, endobronchial occluder, and emphysema. We also manually searched references of the recently published relevant articles. The last literature search was performed on 06/15/2016.Study Eligibility and Exclusion CriteriaAll published human BLVR trials in English were considered for inclusion. Case series, prospective cohort studies, retrospective cohort studies, and randomized controlled trials were included. The outcomes of interest were forced expiratory volume in 1 s (FEV1), 6-min walk test (6MWT), and St.
George’s Respiratory Questionnaire (SGRQ). Since multiple valve trials have already shown disappointing results in patients without intact interlobar fissures 6, 7, 8, and current studies are only including patients with intact fissures, we analyzed the valve data only from those patients with intact fissures.Case reports and studies with follow-up duration less than 6 months were excluded. Smaller publications from individual centers of a multicenter study were excluded. If two publications shared study subjects, the article with the smaller number of subjects or with smaller duration of follow-up was excluded to avoid duplication. We also excluded studies where the BLVR modalities were used for diseases other than COPD (e.
g., valves for giant emphysematous bullae and persistent air leak).Outcomes of InterestOur primary outcome of interest was an absolute change in FEV1 (liters). Secondary outcomes were changes in 6MWT (meters) and SGRQ (points).
The minimal clinically important difference (MCID) used to ascertain a clinically significant change was predetermined as follows: FEV1 ? 0.10 L or ?12% 9, 6MWT ? 26 m 10, and SGRQ ? 4 points 11. We broadly reviewed the major complications of each BLVR modality. A detailed systematic review of the complications is beyond the scope of this study.
Data AbstractionWe abstracted data for FEV1, 6MWT, and SGRQ. Before and after intervention, absolute mean difference ± SD for each variable was recorded at 6 and 12 months (when available). Only the intervention arm data were abstracted for the trials that had between group comparisons with intervention and control arms. The number of MCID responders in each study was also abstracted if data was reported in the full text article or supplementary material. For the meta-analysis, in studies where the standard deviation of the mean difference was not provided, we imputed the mean value of the standard deviation for that variable within the same BLVR modality 12. Two authors (AK and MJM) independently performed literature search and data abstraction. There were no between-reviewer disagreements on study selection and final data synthesis.Assessment of the Risk of BiasThe Cochrane risk of bias assessment tool was used to assess the quality of reporting of the included studies 12.
Data AnalysisPooled estimates of absolute change from baseline and their 95% confidence intervals (CI) were derived for each variable of interest at 6-months interval when data from ?3 studies was available. We used OpenMetaAnalyst software to pool data. Owing to significant variability between studies, we used the random effects model.