Surveillance of antimicrobial resistance (AMR) is crucial for identifying trends in resistance and developing strategies for prevention and treatment of infections. Globally, AMR surveillance systems differ in terms of organizational principles, comprehensiveness, accessibility, and usability of data presentation. Until recently, the data on AMR in Russia were scarcely available, especially to international community, despite the fact that the large prospective multicenter surveillance in Russia was conducted and data were accumulated for over 20 years. We describe the source of data, structure, and functionality of a new-generation web platform, called AMRmap (https://amrmap.net/), for analysis of AMR surveillance data in Russia. The developed platform currently comprises susceptibility data of >40,000 clinical isolates, and the data on abundance of key resistance determinants, including acquired carbapenemases in gram-negatives, are updated annually with information on >5,000 new isolates. The AMRmap allows smart data filtration by multiple parameters and provides interactive data analysis and visualization tools: MIC and S/I/R distribution plots, time-trends and regression plots, associated resistance plots, prevalence maps, statistical significance graphs, and tables.

Статья подготовлена при финансовой поддержке компании Пфайзер. В статье выражена позиция авторов, которая может отличаться от позиции компании Пфайзер.

No abstract

Смоленский государственный медицинский университет, Смоленск, Российская ФедерацияБолезнь Легга-Кальве-Пертеса -форма

Objective. To review the basic principles and functionality of the AMRmap online resource. Materials and Methods. The AMRmap platform was developed using the R programming language and various downloadable modules – packages. The current annually updated version of EUCAST clinical breakpoints was applied for determination of categories of susceptibility to antimicrobial agents. Descriptive analysis includes calculation of absolute and relative frequencies, median values, and confidence intervals using the Wilson method. Categorical variables are compared using Fisher’s exact test and Holm correction method for multiple comparisons. The algorithms are used to visualize multiple comparisons, kernel regression for trend analysis, and algorithms for finding associative rules. Results. The developed surveillance system includes modules for filtering, analyzing and visualizing antibiotic resistance data. The filters allow creating a sample of data with a specific list of parameters. A tab-based separation of analysis and visualization options ensure efficient stepwise evaluation of results. Data saving and sharing functions are also provided. Conclusions. This web-based informatics system provides a convenient way to AMR data from prospective microbiological surveillance studies in Russia. AMRmap can be accessed at https://amrmap.ru.