The prevalence of diabetes mellitus is increasing annually across the globe. According to the International Diabetes Federation (IDF) Diabetes Atlas 2025, approximately 589 million adults between 20–79 years of age are currently living with diabetes worldwide, representing nearly one in nine adults (11.1%) in this age group (1–3). As a result of multiple pathophysiologic changes, these patients commonly get affected by foot-related disorders, particularly infections, ulcers, and gangrene (4–6). Diabetic foot infections (DFIs) are among the most severe complications faced by individuals with diabetes mellitus, contributing significantly to morbidity, hospitalization rates, and healthcare costs globally (6–10). If not managed promptly, these ulcers can develop into soft tissue infection and osteomyelitis, dramatically increasing the risk of gangrene and limb amputation (11). The management of DFIs is complicated by the increasing prevalence of multidrug-resistant (MDR) organisms, which limits the use of narrow spectrum and cost-effective antibiotic choices and underscores the importance of culture-guided therapy (12).

International clinical practice guidelines, such as those developed by the Infectious Diseases Society of America (IDSA) and the International Working Group on the Diabetic Foot (IWGDF), provide structured, evidence-based approaches for the diagnosis and management of DFIs (13). These guidelines emphasize early recognition of infection, severity classification, and the use of empiric antimicrobial regimens tailored to common pathogens in high income countries (13–15).Furthermore, regional differences in antibiotic prescribing guidelines highlight the challenges of standardizing treatment approaches across different regions and have significant implications for future strategies to reduce antimicrobial resistance (AMR) (16).

In contrast, Pakistan, a low- and middle-income country (LMIC) with limited resources and diverse socioeconomic standards faces unique challenges in managing DFIs. The diabetes incidence in Pakistan is relatively high, so it ranks among the top 10 nations worldwide. Diabetes affected 5.2 million Pakistanis in 2000, but will likely affect approximately 13.9 million in 2030 (17). Despite efforts such as the establishment of the Pakistan Working Group on Diabetic Foot (PWGDF) and national programs to improve diabetic care, the absence of locally tailored, evidence based guidelines hinders effective infection control (18–20). As a result, empirical treatment often relies on clinician experience rather than microbiological evidence, increasing the risk of treatment failure and resistance development (21–23). In Pakistan, infections frequently involve MDR bacteria and access to advanced imaging or vascular interventions may be limited (24). With such a burden, evidence-based guidelines are essential for standardizing DFI management. The development of evidence-based, locally adapted treatment guidelines requires the integration of real-world clinical data and antimicrobial susceptibility patterns to support informed and context-specific decision-making. Therefore, it is important to integrate clinical data and local antimicrobial susceptibility patterns to formulate context-appropriate treatment recommendations.

The World Health Organization (WHO) introduced the Access, Watch, and Reserve (AWaRe) classification of antibiotics as a global antimicrobial stewardship framework to promote rational prescribing (25–27). The system classifies antibiotics as Access agents, which are narrow-spectrum first-line options with a low potential for resistance; Watch agents, which are broad-spectrum drugs with high resistance potential requiring careful monitoring; and Reserve agents, which are last treatment option for MDR infections and must be used under strict stewardship protocols (28–30). Although AWaRe provides a useful framework to optimize antibiotic use, its effective implementation in LMICs such as Pakistan is challenging because of distinct microbial landscapes and limited stewardship infrastructure (31–34). Therefore, integrating local microbiological data and antimicrobial susceptibility information into the AWaRe framework is essential to ensure context-appropriate guidelines.

Given the increased burden of DFIs and the high prevalence of AMR, there is an urgent need to develop evidence-based, locally adapted treatment guidelines for DFIs that incorporate real-world clinical data and antimicrobial sensitivity patterns. This study aims to address this gap by analyzing the demographic characteristics, microbial isolates, and antibiotic susceptibility profiles of patients with DFIs admitted to a tertiary care hospital in Pakistan. The findings are intended to inform context specific empirical antibiotic protocols that align with regional resistance trends and healthcare realities, ultimately supporting more effective and sustainable DFI management strategies. Adapting empirical treatment to reflect regional microbial trends and resistance profiles is essential for improving therapeutic outcomes, minimizing treatment failure, and promoting antimicrobial stewardship.

The aim of this study was to develop context-specific, evidence-based empirical treatment recommendations for diabetic foot infections (DFIs) by combining local microbiological data with established international guideline frameworks. Specifically, the study sought to describe the microbial profile of wound cultures obtained from DFI patients at our hospital, assess the antimicrobial susceptibility patterns of the isolated pathogens according to CLSI standards. Moreover, compare these local trends with recommendations from the IDSA/IWGDF guidelines and relevant global evidence. In addition, the study aimed to formulate severity-stratified empirical antibiotic recommendations (mild, moderate, and severe) using a structured multidisciplinary Antimicrobial Stewardship (AMS) approach and to grade these recommendations based on the Strength of Recommendation (SoR) and Quality of Evidence (QoE) criteria.

Out of the 130 collected wound samples from hospitalized patients with DFIs, 118 (90.8%) yielded positive microbial cultures and were included in the final analysis. Among these patients, the majority were male (70%), while females comprised 30% of the cohort. Most individuals (87%) were adults aged under 65 years, and 13% were categorized as geriatric. Patient demographics and sample characteristics are summarized in Table 1.

In addition to demographic characteristics, the clinical spectrum of DFIs was also examined. The infections were classified into wound types; namely infected ulcers, abscesses, cellulitis, gangrene, and necrotizing fasciitis to assess both the severity and depth of infection. Infected ulcers were the most common, ranging from superficial to deep tissue involvement. Abscesses and cellulitis represented localized infections, whereas gangrene and necrotizing fasciitis reflected severe, limb-threatening conditions often associated with systemic signs. This classification not only reflected the clinical burden of infection but also provided the framework for interpreting microbiological findings and informing empirical treatment recommendations.

Microbiological analysis revealed a predominance of Gram-negative organisms, accounting for 62% of all the isolates, whereas Gram-positive organisms made up the remaining 38%. The most frequently isolated microorganism was E.coli (25%), followed closely by methicillin-resistant Staphylococcus aureus (MRSA) at 23%. Other common isolates included Proteus mirabilis (14%), methicillin-sensitive Staphylococcus aureus (MSSA) and Pseudomonas aeruginosa (10.3% each), Streptococcus species (5%), Enterobacter species (4.2%), Klebsiella pneumoniae and Citrobacter species (3.3% each), and Acinetobacter baumannii (1.6%). The frequency and distribution of microbial isolates from culture-positive DFI patients are presented in Table 2.

Antibiotic susceptibility patterns vary across pathogens. E. coli exhibited the highest sensitivity to tigecycline (86%), amikacin (83%), and doripenem (79%), while demonstrating complete resistance to ampicillin and co-amoxiclav. Carbapenems such as imipenem, meropenem, and ertapenem showed relatively good efficacy (70–76%) against E. coli, although decreased sensitivity was noted in Pseudomonas aeruginosa (58–66%). Proteus mirabilis demonstrated excellent susceptibility to carbapenems (94–100%) and piperacillin-tazobactam (100%), but limited response to penicillins and certain cephalosporins. Pseudomonas aeruginosa showed moderate susceptibility to carbapenems and aminoglycosides, although resistance to third-generation cephalosporins and fluoroquinolones was evident.

Among the Gram-positive isolates, MRSA was highly susceptible to vancomycin, and linezolid. MSSA also showed high sensitivity to glycopeptides and oxazolidinones, in addition to aminoglycosides such as gentamicin (100%). Tetracyclines, particularly minocycline and tigecycline, showed good efficacy against both MRSA and MSSA. However, resistance to beta-lactams was common across Gram-positive isolates, notably with ampicillin and co-amoxiclav and this represents an alarming situation.

In terms of Class, penicillins and early-generation cephalosporins were largely ineffective against both Gram-positive and Gram-negative organisms. Carbapenems remained among the most potent options, although emerging resistance was noted in non-fermenting Gram-negative bacilli. Aminoglycosides and polymyxins demonstrated variable but generally acceptable sensitivity patterns, whereas glycopeptides and oxazolidinones were uniformly effective against Gram-positive organisms. Antimicrobial susceptibility patterns of major bacterial isolates are presented in Table 3 and Figure 3.

Antimicrobial susceptibility patterns stratified by WHO AWaRe classification are summarized in Table 4. Access agents showed limited effectiveness, Watch group drugs provided the mainstay for moderate to severe infections, whereas Reserve agents retained excellent activity but were restricted for MDR cases in line with stewardship principles.

On the basis of local resistance profiles and infection severity classifications (mild, moderate, and severe), empirical antibiotic guidelines were formulated to reflect the regional microbial landscape, current susceptibility trends, and antibiotic availability within the healthcare setting. The framework was designed to support clinicians in selecting appropriate empirical therapies while awaiting definitive culture and sensitivity results, thereby promoting timely, effective, and rational antibiotic use in DFI management. For mild infections, co-amoxiclav combined with clindamycin was recommended, with ciprofloxacin or fusidic acid as alternatives. For moderate infections, vancomycin plus piperacillin and azobactam was suggested as the first-line option, with linezolid, clindamycin, or fluoroquinolones considered suitable alternatives. For severe infections with systemic involvement, vancomycin in combination with carbapenems or piperacillin–tazobactam was preferred, whereas alternative regimens including clindamycin, aminoglycosides, or fluoroquinolones were selected according to renal function and local resistance patterns. Severity-stratified empirical antibiotic regimens recommended for DFIs are presented in Table 5.

This study provides critical observations regarding the microbiological profile and antibiotic resistance patterns associated with DFIs in a tertiary care setting in Pakistan. These findings highlight the predominance of Gram-negative organisms, particularly E. coli, which contrasts with many Western studies where Gram-positive organisms, especially S. aureus, tend to dominate. However, our results are consistent with those of regional studies in South Asia, where Gram-negative pathogens are increasingly involved in DFIs due to environmental exposure, poor wound hygiene, and widespread antibiotic misuse (39). Gram-negative bacteria account for 76.27% of the isolates from Pakistani DFIs, with E. coli (15.72%), P. aeruginosa, and P. mirabilis as key pathogens (40, 41). This contrasts markedly with Western studies, where Gram-positive bacteria such as S. aureus dominated 40-60% of cases (42).Although E.Coli and MRSA were among the most frequent pathogens overall, the majority of mild- severity infections in our study were superficial ulcers in which Gram-positive cocci particularly streptococci and MSSA remained the predominant organisms. Accordingly, our empiric recommendations for mild DFIs prioritize narrow spectrum agents targeting typical Gram positive pathogens, while broader Gram negative coverage is reserved for moderate and severe infections. This stratified approach is consistent with IDSA/IWGDF principles and aligns with antimicrobial stewardship goals by minimizing unnecessary exposure to broad spectrum antibiotics in clinically mild presentations.

The high prevalence of multidrug-resistant organisms, including MRSA and ESBL producing Enterobacteriaceae, reflects a serious threat to empirical treatment strategies (43–46). The degree of resistance to commonly used antibiotics such as ampicillin, co-amoxiclav, and cephalosporins is alarmingly high, indicating that traditional empirical regimens may no longer be effective in this setting. The relatively preserved activity of carbapenems against E. coli and P. mirabilis, along with the efficacy of amikacin and tigecycline, underscore the need to reserve these agents for confirmed or highly suspected resistant infections, especially given concerns around toxicity, cost, and the risk of promoting further resistance (20, 24, 47, 48).

Importantly, glycopeptides and oxazolidinones maintained strong efficacy against MRSA and MSSA isolates, supporting their continued use in severe infections caused by Gram-positive pathogens. Nonetheless, the moderate resistance of MRSA to agents such as clindamycin and co-trimoxazole reinforces the importance of culture-directed therapy. These findings also underscore a significant loss of susceptibility to narrow-spectrum and cost-effective antimicrobial agents (48).The sustained efficacy of glycopeptides such as vancomycin and oxazolidinones such as linezolid against MRSA and MSSA isolates in DFIs highlights their critical role in managing severe Gram-positive infections, particularly in regions with high MRSA prevalence. Vancomycin remains universally effective against MRSA, despite emerging concerns about increasing minimum inhibitory concentrations (MICs), whereas linezolid offers strong clinical efficacy and oral bioavailability, making it valuable for outpatient care (49, 50).This approach not only optimizes treatment success but also helps preserve last-line antibiotics such as carbapenems and polymyxins for confirmed resistant infections, mitigating further resistance development.

These findings also have significant implications in the context of the WHO AWaRe classification. In our study, many of the antibiotics that demonstrated the highest activity such as carbapenems, colistin, and glycopeptideswere included within the Watch and Reserve groups. This means that Access antibiotics, which the WHO recommends as first-line therapy, are largely ineffective in our setting. The reliance on Watch and Reserve antibiotics for empirical therapy undermines global stewardship goals and highlights the urgent need for region-specific AWaRe adaptations. The incorporation of local antibiogram data into the AWaRe framework may help balance stewardship principles with the clinical imperative of ensuring effective empirical coverage in resource-constrained healthcare systems.

Our findings highlight the urgent need for local antibiograms to guide empirical therapy and highlight the limitations of applying international guidelines such as those from the IDSA or IWGDF without contextual adaptation. For example, broad-spectrum cephalosporins commonly recommended in international protocols have shown limited effectiveness in our patient population, further justifying the creation of locally tailored empirical guidelines (51–53). Moreover, The preserved efficacy of colistin in DFIs offers a crucial last-line defense against MDR Gram-negative pathogens such as Pseudomonas aeruginosa, particularly in regions with high carbapenem resistance. However, its use is fraught with significant challenges, including a well-documented toxicity profile-ranging from nephrotoxicity in 10–50% of patients to neurotoxicity which complicates its administration in diabetic populations that are often burdened by preexisting renal impairment (54).

Therefore, the development of empirical treatment protocols based on the basis of local sensitivity patterns stratified by infection severity represents a practical step toward improving treatment outcomes. These guidelines, if adopted and regularly updated, can serve as a cornerstone for antimicrobial stewardship programs, particularly in resource-limited settings where culture facilities and infectious disease expertise are often scarce.

This study systematically integrated international evidence, local microbiological surveillance, and multidisciplinary expertise to develop an evidence-based antimicrobial guideline for diabetic foot infections. The findings demonstrated that Access group antibiotics have limited empirical utility in this setting, while Watch group agents remain the cornerstone for moderate to severe infections, and Reserve agents retain high efficacy but must be preserved for MDR cases. By generating hospital-wide antibiograms and translating them into context-specific treatment recommendations, this work addresses a critical gap in standardized local practice. The resulting guideline promotes rational antibiotic use, supports antimicrobial stewardship, and is expected to improve both clinical outcomes and resource utilization. Periodic updates will be essential to align with evolving resistance trends and ensure sustained clinical relevance.