Antimicrobial resistance of Streptococcus uberis isolated from bovine mastitis: Systematic review and meta-analysis

MIOTTI, CAMILA; CICOTELLO, JOAQUÍN; SUAREZ ARCHILLA, GUILLERMO; NEDER, VERÓNICA; ALVARADO LUCERO, WANDA; CALVINHO, LUIS; SIGNORINI, MARCELO; CAMUSSONE, CECILIA; ZBRUN, MARÍA VIRGINIA; MOLINERI, ANA INÉS

RESEARCH IN VETERINARY SCIENCE

ELSEVIER SCI LTD

Año: 2023 vol. 164

0034-5288

Streptococcus uberis is one of the most common pathogens associated with bovine mastitis, commonly treated with antimicrobials (AM), favoring the appearance of antimicrobial resistance (AMR). The objective of this work was to determine the proportion of phenotypic AMR among S. uberis isolated worldwide from bovine intramammary infections between the years 1983–2022, and to assess the variables associated by means of a systematic review and metanalysis. Sixty articles were eligible for quantitative review. Ninety-four independent studies were obtained. The antimicrobials evaluated in more S. uberis strains were penicillin (21,987 strains), oxacillin (21,727 strains), erythromycin (20,013 strains), and ampicillin (19,354 strains). Most of the studies included in this meta-analysis were from Europe (44), followed by America (25), Africa (10), Asia (10), and Oceania (5). Among the included articles, 22 were published from 1983 to 2006, 23 from 2007 to 2012, 25 from 2013 to 2015, and the remaining 24 after 2016. Penicillin, erythromycin, and tetracycline were the antimicrobials with >25 studies. Therefore, the following analyses were performed only for these antimicrobials, presenting a high heterogeneity index (I2). The variability observed for penicillin and tetracycline was only explained, partially, by continent of origin. The variability observed for erythromycin was not explained by any of the potential explanatory variables included in this study. The S. uberis proportion of resistance to antimicrobials is highly variable and probably influenced by many factors other than those studied in this meta-analysis, where it was not possible to inform a unique average proportion of resistance.