Accuracy of rapid microcapillary direct antibiotic susceptibility testing for urine samples collected with bacteriostatic boric acid from patients with suspected urinary tract infection

Abstract Objectives Antibiotic susceptibility testing (AST) results are needed more rapidly to support antimicrobial stewardship and improve patient outcomes. Diagnostic microbiology laboratories receive hundreds of urine samples daily from patients with suspected urinary tract infection (UTI) in both community and inpatient settings. Bacteriostatic boric acid preserves microbial contents during transport but may interfere with rapid AST methods. This study aimed to assess the accuracy of rapid microcapillary direct-from-urine (RMD) AST with suspected UTI patient urine, and to determine whether RMD AST is affected by boric acid. Methods The overall accuracy of RMD AST was assessed with 352 diagnostic remnant urine samples collected with boric acid, for seven first-line antibiotics (ampicillin, amoxicillin/clavulanic acid, trimethoprim, nitrofurantoin, ciprofloxacin, cefalexin and cefoxitin). A further 90 urine samples were tested in duplicate with or without addition of bacteriostatic. Results RMD AST showed a concordance with the reference method of 572/590 bacteria/antibiotic combinations (96.95%) for urine samples containing a single organism. The mean time to AST result was 5.85 h. When duplicate samples with or without boric acid were directly compared there was a categorical agreement of 158/160 (98.75%). Conclusions The overall high accuracy of RMD AST for determining antimicrobial susceptibility to seven first-line antibiotics for UTI shows this method can deliver rapid results—without requiring additional processing—for urine samples routinely collected with boric acid from suspected UTI patients. The close agreement between duplicates with or without boric acid confirms this rapid direct method is unaffected by bacteriostatic collection.