Antimicrobial susceptibility testing of invasive isolates of Streptococcus pneumoniae from Canadian patients: the SAVE study, 2011–2020

Abstract

Abstract Objectives To assess the antimicrobial susceptibility of 14 138 invasive Streptococcus pneumoniae isolates collected in Canada from 2011 to 2020. Methods Antimicrobial susceptibility testing was performed using the CLSI M07 broth microdilution reference method. MICs were interpreted using 2022 CLSI M100 breakpoints. Results In 2020, 90.1% and 98.6% of invasive pneumococci were penicillin-susceptible when MICs were interpreted using CLSI meningitis or oral and non-meningitis breakpoints, respectively; 96.9% (meningitis breakpoint) and 99.5% (non-meningitis breakpoint) of isolates were ceftriaxone-susceptible, and 99.9% were levofloxacin-susceptible. Numerically small, non-temporal, but statistically significant differences (P < 0.05) in the annual percentage of isolates susceptible to four of the 13 agents tested was observed across the 10-year study: chloramphenicol (4.4% difference), trimethoprim-sulfamethoxazole (3.9%), penicillin (non-meningitis breakpoint, 2.7%) and ceftriaxone (meningitis breakpoint, 2.7%; non-meningitis breakpoint, 1.2%). During the same period, annual differences in percent susceptible values for penicillin (meningitis and oral breakpoints) and all other agents did not achieve statistical significance. The percentage of isolates with an MDR phenotype (resistance to ≥3 antimicrobial classes) in 2011 and 2020 (8.5% and 9.4%) was not significantly different (P = 0.109), although there was a significant interim decrease observed between 2011 and 2015 (P < 0.001) followed by a significant increase between 2016 and 2020 (P < 0.001). Statistically significant associations were observed between resistance rates to most antimicrobial agents included in the MDR analysis (penicillin, clarithromycin, clindamycin, doxycycline, trimethoprim/sulfamethoxazole and chloramphenicol) and patient age, specimen source, geographic location in Canada or concurrent resistance to penicillin or clarithromycin, but not biological sex of patients. Given the large isolate collection studied, statistical significance did not necessarily imply clinical or public health significance in some analyses. Conclusions Invasive pneumococcal isolates collected in Canada from 2011 to 2020 generally exhibited consistent in vitro susceptibility to commonly tested antimicrobial agents.