Network meta-analysis: a technique to gather evidence from direct and indirect comparisons

Vitamins are essential micronutrients with antioxidant potential that may provide a complementary treatment for patients with chronic diseases. Our aim was to assess the effect of vitamin supplementation on the antioxidant status and glycemic index of type 2 diabetes mellitus patients. We performed a systematic review with meta-analyses. Electronic searches were conducted in PubMed, Scopus, and Web of Science (December 2017). Randomized controlled trials evaluating the effect of any vitamin or vitamin complex supplementation on antioxidant status as primary outcome were included. The outcomes considered were: reduction of malondialdehyde (MDA); augmentation of glutathione peroxidase (GPx); changes in total antioxidant capacity (TAC), enhance in superoxide dismutase enzyme—SOD, and thiobarbituric acid reactive substances (TBARS). Outcomes of glycemic control were also evaluated. Pairwise meta-analyses were performed using software Review Manager 5.3. Thirty trials fulfilled the inclusion criteria, but only 12 could be included in the meta-analyses of antioxidant outcomes. The most commonly studied vitamins were B, C, D and E. Vitamin E was related to significant reduction of blood glucose as well as glycated hemoglobin compared to placebo, while both vitamins C and E were mainly associated with reducing MDA and TBARS and elevating GPx, SOD and TAC, compared to placebo. However, outcome reports in this field are still inconsistent (e.g. because of a lack of standard measures). Supplementation of vitamin E may be a valuable strategy for controlling diabetes complications and enhancing antioxidant capacity. The effects of other micronutrients should be further investigated in larger and well-designed trials to properly place these complementary therapies in clinical practice.

The conduction and report of network meta-analysis (NMA), including the presentation of the network-plot, should be transparent. We aimed to propose metrics adapted from graph theory and social network-analysis literature to numerically describe NMA geometry. A previous systematic review of NMAs of pharmacological interventions was performed. Data on the graph’s presentation were collected. Network-plots were reproduced using Gephi 0.9.1. Eleven geometric metrics were tested. The Spearman test for non-parametric correlation analyses and the Bland-Altman and Lin’s Concordance tests were performed (IBM SPSS Statistics 24.0). From the 477 identified NMAs only 167 graphs could be reproduced because they provided enough information on the plot characteristics. The median nodes and edges were 8 (IQR 6–11) and 10 (IQR 6–16), respectively, with 22 included studies (IQR 13–35). Metrics such as density (median 0.39, ranged 0.07–1.00), median thickness (2.0, IQR 1.0–3.0), percentages of common comparators (median 68%), and strong edges (median 53%) were found to contribute to the description of NMA geometry. Mean thickness, average weighted degree and average path length produced similar results than other metrics, but they can lead to misleading conclusions. We suggest the incorporation of seven simple metrics to report NMA geometry. Editors and peer-reviews should ensure that guidelines for NMA report are strictly followed before publication.

We aimed to determine the methodological quality of network meta-analyses (NMAs) and their compliance with reporting guidelines. A systematic review of NMAs comparing any pharmacological interventions was performed (searches in Medline and Scopus). The characteristics of NMAs were collected by two independent reviewers. We applied R-AMSTAR to all NMAs, generating a methodological quality score that could range from 11 to 44 points. PRISMA and PRISMA-NMA reporting checklists were converted into quantitative scores (maximum values of 27 and 32 points). To normalize the values between these two checklists, a third score (PRISMA-SCORE) of 0–1 was created. The correlation of the scores with NMA publication year, journal impact factor and most productive countries were calculated using non-parametric tests. We identified 477 NMAs. Only 36.1% of studies reported having followed PRISMA statements. The medians of R-AMSTAR, PRISMA and PRISMA-NMA scores were 28 (IQR 25–31), 21 (IQR 19–23) and 23 (IQR 19–26), respectively. Several problems were noted in NMAs (e.g. lack of study protocol, issues in literature searches, lack of raw data). NMAs from the most productive countries (USA and China) have similar methodological quality. Correlation analyses between R-AMSTAR and normalized PRISMA-SCORE revealed a strong positive correlation (Spearman’s ρ = 0.776; P <0.001). A weak but positive correlation was found for PRISMA-SCORE and journal impact factor (0.193; P <0.001). The important growth of NMA publication rate during the past 5 years is not associated with better methodological and reporting quality. Editors, peer reviewers, researchers and funding agencies should ensure that methodological and reporting standards are met before publication.

Medication reconciliation at patient admission: a randomized controlled trial