Sepsis continues to cause substantial mortality worldwide, with an estimated 5.3 million deaths annually [1]. Sepsis is a leading cause of in-hospital mortality and contributes substantially to overall inpatient deaths [2,3]. This challenge is expected to worsen as antimicrobial resistance (AMR) continues to rise. The spread of resistant bacteria diminishes the effectiveness of available treatments and impairs timely clinical intervention in patients with sepsis [4,5]. The increasing prevalence of AMR, coupled with high mortality associated with sepsis, presents a looming crisis. If AMR is not addressed urgently, a resurgence in sepsis-related deaths and major setbacks in global infectious disease control efforts are imminent [4].

Notably, recent Korean national cohort data indicate that, although the incidence of sepsis has increased in Korea over the past decade, in-hospital mortality rates have gradually declined with improvements in care; however, the burden remains substantial [6]. This trend aligns with growing concerns regarding AMR in Korea, as studies have shown increasing rates of multidrug-resistant (MDR) pathogens complicating sepsis treatment, particularly in long-term care and hospital-acquired settings [7]. Korean pediatric data also note rising resistance rates, underscoring the urgency of stewardship among vulnerable populations [8,9]. Given the increasing complexity of sepsis management in the context of AMR, both the recent advancements and ongoing challenges in sepsis treatment should be assessed.

Antimicrobial stewardship programs (ASPs) are systematic, multidisciplinary strategies designed to optimize the use of antimicrobial agents, ensuring that patients receive the appropriate antibiotic at the correct time, dose, route, and duration, based on evidence and local epidemiology. The overarching goals of ASPs are twofold: (1) to improve individual patient outcomes, such as survival, adverse event reduction, and timely infection resolution; and (2) to combat the wider threat of AMR by minimizing unnecessary or inappropriate antibiotic exposure at the healthcare system level [10–15]. ASPs incorporate prospective audits, prescriber feedback, formulary restrictions, and protocolized guidance, all supervised by core teams that typically include infectious disease specialists, pharmacists, and clinical microbiologists, are supported by hospital leadership [11,13]. The significance of ASPs is increasingly recognized globally, given the escalation of AMR and lag in the development of new antibiotic agents; thus, stewardship is now considered an essential component of modern infectious disease management and a pillar of patient safety and quality improvement initiatives [10,16].

This review addresses the global threat of AMR, focusing on its impact on clinical management and patient outcomes. It discusses the principles of the Surviving Sepsis Campaign while acknowledging its limitations in addressing the evolving challenges posed by AMR. The core principles of ASPs in sepsis, emphasizing timely antibiotic initiation, the importance of de-escalation, and the vital role of infectious disease specialists in optimizing therapy, have also been explored. This review also evaluated the effect of ASPs on sepsis outcomes, underscoring their role in improving patient care and curbing the spread of resistant pathogens. The review concludes by discussing the limitations of the current ASP strategies for sepsis treatment and highlighting future opportunities for strengthening stewardship practices in response to the AMR crisis.

The global threat posed by AMR has reached unprecedented levels, fundamentally challenging effective management of sepsis across healthcare systems. Before the COVID-19 pandemic, AMR was among the most serious global public health concerns, undermining the use of existing antibiotics and contributing to increased mortality, extended hospitalization, and increased healthcare expenditures for patients [5,11,17]. The pandemic has further complicated this picture: although some initial infection prevention measures (e.g., improved hand hygiene and reduced interpersonal contact) may have temporarily curbed certain transmission pathways, disruptions in standard care and the widespread empirical use of antibiotics for viral illnesses with low rates of bacterial co-infection, such as COVID-19, have accelerated the overuse and misuse of antimicrobials [17–20]. This surge in inappropriate prescriptions is now widely recognized as a driver of the emergence and dissemination of MDR pathogens in critical care and hospital settings [5,17,18,21].

Decades of evidence have established a clear and direct relationship between indiscriminate antibiotic use, including overuse (unnecessary initiation or prolonged duration) and misuse (suboptimal selection or dosing), and the development of resistance. In intensive care units (ICUs), where sepsis often requires immediate empiric therapy, the frequent administration of broad-spectrum antibiotics in the absence of adequate diagnostic procedures paradoxically increases the selection pressure for resistant organisms [5,21,22]. This dynamic is further exacerbated when delays in pathogen identification prompt clinicians to extend or escalate antimicrobial coverage [5,22]. As AMR rates increase, clinicians are forced into a cycle of escalating empirical regimens, which further promotes resistance and reduces the future effectiveness of first-line agents [5,11,21]. Multicenter data from Korea have demonstrated that the prevalence of MDR organisms in sepsis is particularly high in long-term care hospitals and nursing home-acquired sepsis cases, even exceeding that in hospital-acquired cases. Drug-bug mismatches in empiric antibiotics are common, highlighting the pressing need for stewardship in these settings [7].

Recognizing this destructive cycle, ASPs have become a cornerstone strategy for curbing resistance, particularly in resource-intensive environments such as sepsis treatment [11,21,22]. ASPs employ multifaceted interventions, including prescriber education, formulary restrictions, protocolized therapy, and audit feedback mechanisms, to ensure that antibiotics are used only when clinically indicated, promptly de-escalated based on microbiological data, and optimized for both spectrum and duration [11,21–25]. In Korea, national surveys of ASPs have revealed an increasing adoption of stewardship practices but have continued limitations in specialist personnel and programmatic focus, with a substantial dependency on pre-authorization measures and ongoing opportunities for improvement [24,26]. Despite these limitations, the stepwise implementation of core stewardship elements at a Korean tertiary hospital has been associated with reductions in antibiotic consumption and more favorable trends in resistance rates than nationwide patterns, underscoring the potential of long-term, multidisciplinary stewardship measures to positively affect AMR burdens, even in resource-limited settings [25].

Recent studies have shown that effective stewardship programming improves patient outcomes and reduces the incidence of healthcare-associated infections caused by MDR organisms [18,22]. Even during the COVID-19 pandemic, the integration of stewardship strategies, such as biomarker- guided discontinuation and rapid diagnostic pathways, has limited unnecessary antibiotic exposure and mitigated the long-term risk of resistance amplification [17,18]. When ASPs are embedded in sepsis protocols and actively supported by infectious disease specialists, pharmacists, clinical microbiologists, and microbiology laboratories, they represent the most effective defense against the relentless progression of AMR in the context of both pandemic and endemic infectious threats [11,21–23,25,27].

In recent years, the Surviving Sepsis Campaign and other international guidelines have underscored the importance of early initiation of antimicrobial therapy in patients with suspected sepsis, emphasizing that delays in antibiotic administration can significantly increase mortality risk, particularly in cases of severe sepsis and septic shock [18,21,22,28]. Multiple large-scale studies and performance improvement programs have demonstrated a clear and near-linear association, in which each hour of delay in administering appropriate empiric antibiotics after the recognition of sepsis correlates with measurable increases in in-hospital mortality, with the effect being most pronounced in patients with severe conditions, including shock and multi-organ failure [29–31]. For instance, a retrospective study of more than 28,000 patients with severe sepsis or septic shock found that mortality increased steadily with every hourly delay in the first antibiotic administration, even after adjusting for sepsis severity and other confounders [29]. Similar findings were observed in emergency department settings, where each hourly delay in antibiotic administration corresponded to increased odds of in-hospital mortality, particularly among those with septic shock [30,31]. Data from Korean studies support these findings. A nationwide analysis from 2007 to 2016 showed that while the incidence of sepsis increased, improved management, including prompt antibiotic initiation, was associated with decreased in-hospital mortality, hospital stay, and ICU admissions for patients with sepsis [6].

Beyond timing, the adequacy of empiric antibiotic therapy, which is the extent to which the initially chosen regimen is active against the identified or suspected pathogen( s), plays a critical role in patient outcomes. Inadequate empiric therapy was significantly associated with increased mortality among patients with sepsis, even after adjusting for comorbidities and illness severity. One meta-analysis of 70 prospective studies quantified this relationship, demonstrating a pooled adjusted odds ratio (OR) for mortality of 1.60 (95% confidence interval [CI], 1.37–1.86) for inappropriate empiric antibiotics compared with appropriate ones; this effect was particularly strong in patients with septic shock [32]. Furthermore, delays in the administration of appropriate antibiotics were independently associated with prolonged ICU and length of hospital stays in patients with sepsis, highlighting the importance of early and appropriate therapy selection [33]. Multicenter studies in Korea have emphasized that the risk of drug-bug mismatch is highest in patients with long-term care hospital-acquired sepsis, and careful initial antibiotic selection tailored to the onset location is recommended [7].

However, the strong push for immediate broad-spectrum antibiotic coverage in all suspected sepsis cases has prompted a significant debate, particularly concerning the potential risks of indiscriminate antibiotic use. Broad-spectrum agents are frequently employed empirically to ensure early and adequate coverage in the face of increasing rates of MDR organisms and evolving resistance patterns [5,34]. Although this strategy may improve survival in cases of true infections, it also carries the risk of promoting AMR, increased drug-associated toxicity, development of Clostridioides difficile infections, and unnecessary selective pressure on local microbial ecology [5,34]. Empirical overuse of broad-spectrum antibiotics is of particular concern in patient populations with a low risk of MDR, particularly in settings where rapid diagnostics or validated risk stratification tools are available to support targeted therapy [5].

Indeed, the current research underscores the need for individualized risk assessment and stewardship. Kalin et al. have highlighted that identifying MDR risk factors and using rapid molecular diagnostics may facilitate targeted therapy, reduce unnecessary broad-spectrum use, and mitigate collateral damage [5]. Additionally, large real-world analyses of the most recent Surviving Sepsis Campaign guidelines advocate a risk- and probability-stratified approach to antibiotic timing [35]. Specifically, immediate antibiotic therapy within 1 h is recommended primarily for septic shock or strongly probable infection, whereas a 3-hour window is recommended for patients with possible but less likely infections and no shock [35]. This stratification aligns with real-world data: the mortality benefit of rapid antibiotics is substantial in patients with shock or high-probability sepsis but marginal in those without shock and with only a possible infection, for whom undue haste may more often result in overtreatment than enhanced survival [35,36].

Moreover, critical analyses of the evidence on universal “time-to-antibiotic” performance targets highlight significant methodological heterogeneity and bias in the literature [36,37]. Weinberger et al. have argued that the majority of supporting studies have been observational, have frequently focused on patients in extremely severe conditions (e.g., those with shock or requiring ICU care), and have failed to fully adjust for confounding factors such as presenting symptoms, comorbidities, or illness trajectories [36]. Because many patients with suspected sepsis ultimately have non-infectious or self-limiting etiologies, routine administration of broad-spectrum antibiotics in rigid timeframes risks exposing these individuals to unnecessary harm without a clear survival advantage [36]. Some studies have indicated that the mortality differences between immediate and early antibiotic administration (within 3 h) may not be statistically significant in a broader population of patients with sepsis without shock [38].

Therefore, these insights have shaped consensus on the latest practice recommendations. Although early empirical antibiotics are undeniably lifesaving for high-risk patient groups (notably those with shock, high severity scores, or definitive infection), the indiscriminate or unrefined application of broad-spectrum therapy for all suspected cases carries substantial risks to individual and public health. Accumulated evidence strongly supports a balanced, patient-specific approach that incorporates infection probability, severity, local resistance patterns, and the growing threat of MDR pathogens to maximize survival benefits while minimizing antimicrobial overuse and its repercussions [5,34–36].

ASPs are fundamental for optimizing sepsis treatment in an era marked by an increase in AMR and MDR pathogens, both of which threaten the effectiveness of standard therapeutic approaches. In sepsis treatment, the core principles of ASPs involve a multifaceted strategy that balances the need for timely and effective empirical therapy by minimizing unnecessary antibiotic exposure and subsequent resistance selection.

ASPs have emerged as essential frameworks for sepsis management. The aim of ASPs is to achieve a balance between aggressive early empiric treatment and avoid antibiotic overuse, which has profound repercussions across the clinical, economic, and public health spectra. The following sections elaborate on the effects of ASPs on various sepsis outcomes, with supporting evidence (Table 1, Fig. 1).

In summary, the persistent global threat of sepsis, compounded by the proliferation of MDR organisms, demands a multifaceted and forward-thinking approach to antimicrobial therapy. The Surviving Sepsis Campaign provides a crucial framework for the early recognition and management of sepsis. However, limitations persist, particularly regarding the balance between prompt empirical therapy and appropriate antimicrobial selection in this era of increasing resistance. This review underscores the core tenets of ASPs, appropriate empirical antibiotic use, timely de-escalation, and active involvement of infectious disease experts, which are indispensable for optimizing clinical outcomes and preserving antimicrobial efficacy. Evidence shows that ASPs reduce mortality, hospital LOS, and the incidence of adverse events while curbing the development of resistance and associated costs in sepsis treatment. Nevertheless, significant challenges remain, including delays in diagnosis, gaps in rapid resistance detection, pharmacokinetic variability in critical illnesses, human resource limitations, and the inconsistent global adoption of stewardship strategies. The integration of rapid molecular diagnostics, individualized dosing approaches, and robust interdisciplinary collaboration holds significant promise for further refinement of stewardship efforts. Sustained investment in human resources, infrastructure, and capacity-building for ASPs, as well as support for innovation and research, is essential. In parallel, high-quality evidence from well-designed studies is urgently needed to strengthen the foundation for effective implementation of ASPs in sepsis treatment. These efforts are critical for bridging the current gaps and ensuring optimal outcomes in patients with sepsis amid the evolving threat of AMR.