Current Antibiotic Susceptibility Test Underestimates Minority Resistance: Implications for High-Risk Infections

Abstract

Antibiotic susceptibility testing (AST) reports classify isolates as “susceptible” despite potential undetected resistant subpopulations—a phenomenon termed susceptibility heterogeneity (SH). Found in 15–97% of clinical isolates of Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, and Klebsiella pneumoniae, SH arises from heteroresistance or polyclonal diversity and may evade standard low-inoculum protocols. Clinically, this can lead to treatment failure, particularly in high-risk cases including immunocompromised patients, bloodstream infections, transplant recipients, and situations where minor resistant subpopulations significantly affect outcome. We argue that ethical principles of non-maleficence, transparency, and equity now compel laboratories to acknowledge this limitation. A simple annotation—“Limited susceptibility possible; resistant subpopulations may not be detected”—should accompany “susceptible” results in immunocompromised patients. High-risk cases warrant enhanced testing. This commentary calls for zone inspection, staff training, and Clinical and Laboratory Standards Institute (CLSI)/European Committee on Antimicrobial Susceptibility Testing (EUCAST) guideline updates to reflect SH. Transparency enhances clinical decision-making without implying diagnostic fault.

Susceptibility heterogeneity (SH) refers to the presence of subpopulations within a bacterial sample that exhibit differing antibiotic susceptibility. It encompasses both clonal heteroresistance—resistant subpopulations within a genetically identical group—and/or polyclonal diversity, where distinct strains coexist in a specimen [1,2,3,4,5].

Across pathogens such as E. coli, S. aureus, K. pneumoniae, and P. aeruginosa, SH prevalence ranges widely—reported between 15% and 97%—depending on species and detection methods [1,2,3,4,5]. Studies using high-inoculum or multi-colony AST, population analysis profiling, and zone inspection report significantly higher detection rates than routine single-colony protocols [1,6,7].

Susceptibility heterogeneity has demonstrated clinical consequences: heteroresistant subpopulations may expand under therapy, resulting in breakthrough growth, treatment failure, or evolution to full resistance, as shown in colistin-heteroresistant Acinetobacter baumannii and other pathogens [8]. In immunocompromised patients, susceptibility heterogeneity is not just a theoretical concern: vancomycin-heteroresistant coagulase-negative staphylococcal bloodstream infections in children with leukemia, as well as echinocandin-heteroresistant Candida parapsilosis fungemia in allogeneic hematopoietic cell transplant recipients, have both been directly associated with treatment or prophylaxis failure [9,10].

The clinical relevance of SH extends far beyond the well-documented examples in Enterobacterales, S. aureus, and A. baumannii. In Mycobacterium tuberculosis infections, mixed-strain (polyclonal) infections are common and frequently lead to discordant resistance profiles within the same patient sample. It was repeatedly shown that diagnostic sputum often contains multiple genetic variants at resistance-associated loci, whereas standard single-colony culture captures only one subpopulation, potentially underestimating resistance and yielding a falsely susceptible MIC [11,12]. Similarly, Nathavitharana and colleagues demonstrated that distinct M. tuberculosis genotypes co-existing in the same specimen can carry different resistance mutations, rendering routine single-colony AST unreliable [13].

In non-tuberculous settings, Weiss and co-workers provided definitive experimental evidence that minority resistant subpopulations in Gram-negative pathogens (frequencies as low as 10⁻⁶–10⁻⁸ [6]) are clinically relevant and directly cause treatment failure despite a susceptible MIC [14]. Recent genomic epidemiology studies demonstrate that Pseudomonas aeruginosa can develop ceftolozane–tazobactam resistance through multiple, convergent evolutionary pathways, even when initial AST results indicate full susceptibility [15]. Finally, genomic epidemiology studies of Streptococcus reveal that what is reported as “one isolate” frequently represents a mixture of closely related lineages with differing susceptibility and tolerance phenotypes [16].

These independent lines of evidence—from TB, Gram-negative nosocomial pathogens, cystic fibrosis, transplant medicine, and community carriage—converge on the same conclusion: the traditional paradigm of “one patient → one isolate → one MIC” is microbiologically obsolete and clinically unsafe in an increasing proportion of high-risk infections.

Ethically, the imperative to disclose SH in AST reports stems from non-maleficence (avoiding preventable harm), transparency (acknowledging test limitations), and equity (avoiding information gaps between labs with differing resources) [17]. These principles align with global stewardship priorities and are supported by WHO’s Global Action Plan on AMR [18] and the GBD AMR report [19].

For high-risk infections, universal, low-cost annotation stating “Limited susceptibility possible; resistant subpopulations may not be detected by standard methods” should be added to “susceptible” results. This is analogous to disclaimers already used for ESBL, inducible clindamycin resistance, and certain agent–organism combinations in EUCAST guidance and has been supported by methodological studies and expert guidance on heteroresistance and AST method modification [1,20,21]. For high-risk cases, laboratories could consider zone inspection, retesting with higher inoculum, or orthogonal methods.

Practical Recommendations

To balance feasibility and clinical impact, we divide these general recommendations (applied in all high-risk infections) into universal measures and selective ones (which could be under the supervision of Microbiologists).

Recommendations (for all high-risk cases):

• Add the annotation, “Limited susceptibility possible; resistant subpopulations may not be detected.”
• Inspect inhibition zones for internal colonies during standard AST.
• Train staff to recognize SH patterns.
• Advocate for CLSI/EUCAST guidelines to address SH formally.

Selective recommendations (if required by Microbiologist):

• Retest using multiple colonies, higher inoculum, or orthogonal methods.