Abstract :
[en] This study, focusing on bloodstream infections and conducted in a rural setting in Mali, aimed to assess bacterial resistance using a two-step diagnostic approach. The first step consisted of preliminary on-site detection of bacteria using manually prepared CHROMagar media, providing a simple, rapid, and cost-effective method. The second step involved phenotypic and genotypic confirmation of isolates in a reference laboratory using advanced techniques such as MALDI-TOF MS, VITEK® 2, and conventional PCR to validate resistance profiles.
Among the 508 blood cultures analyzed, 29.1% (148/508; 95% CI [25–33]) were positive, identifying 16.9% (86/508; 95% CI [13.7–20.2]) resistant strains. Of these, 75.6% (65/86; 95% CI [66.6–84.6]) were Enterobacteriaceae resistant to β-lactams and carbapenems, 13.9% (12/86; 95% CI [6.5–21.3]) corresponded to methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus (VRE), and 10.5% (9/86; 95% CI [4–16.9]) to other bacterial species.
The most frequently detected resistance genes included CTX-M in 39% (49/126; 95% CI [30.3–47.5]), TEM/SHV in 26% (32/123; 95% CI [18.2–33.8]), genes encoding carbapenemases (KPC, VIM, NDM, OXA-48) in 15% (19/127; 95% CI [8.7–21.3]), and genes encoding other β-lactamases (OXA-1, bla-BEL, bla-ADCb, FOX-1) in 9% (11/122; 95% CI [4–14]). Among Staphylococcus and Enterococcus isolates, the mecA and vanA/vanB genes were detected in 3% (4/133; 95% CI [0.1–5.9]) and 4% (5/125; 95% CI [0.6–7.4]) of strains, respectively.
CHROMagar media showed a sensitivity of 80% (69/86; 95% CI [71.2–88.8]), a specificity of 85% (73/86; 95% CI [75–91.9]), a positive predictive value of 90% (77/86; 95% CI [82.4–95.1]), and a negative predictive value of 72% (62/86; 95% CI [61.6–80.9]). These findings confirm the potential of this simplified approach to strengthen bacterial resistance surveillance in resource-limited settings, while highlighting the need for reference laboratory validation and protocol optimization to improve its integration into decentralized health systems.
