Prevalence of diabetes mellitus is increasing rapidly in South Korea and is a major risk factor for lower extremity amputations. From 2003 to 2021, the incidence of lower extremity amputations in the general population more than doubled, with the contribution of diabetes mellitus to these amputations also increasing from 47% in 2003 to 70% in 2021. This indicates that a significant proportion of lower extremity amputations in South Korea is due to diabetic foot complications.¹⁾ Diabetic foot infection (DFI) stands as the primary and most catastrophic complication associated with diabetic foot conditions.²⁾

Research on Western populations consistently shows that Gram-positive aerobes, particularly Staphylococcus aureus, are the predominant causative organisms in DFI, with detection rates often around 20%∼50%.^3-6) Additionally, methicillin-resistant Staphylococcus aureus (MRSA) cases have been increasing, indicating a growing prevalence of resistant strains over time.⁷⁾ In contrast, studies from the Middle East, Asia, and Africa report a higher isolation rate of Gram-negative rods, such as Pseudomonas aeruginosa, compared to Western populations.^8-11) For instance, research in South India found Pseudomonas spp. as a significant pathogen, and a comprehensive study in Kuwait reported Enterobacteriaceae and Pseudomonas accounting for a combined prevalence of nearly 45%.^12-14)

Similarly, South Korean studies have shown a predominance of Gram-positive aerobes, with S. aureus detected in up to 39.8% of cases, while Gram-negative aerobes, including Pseudomonas spp., are also commonly isolated.^15-17) These regional patterns highlight the need for tailored infection control strategies based on local pathogen profiles.

Previous studies have mainly focused on identifying pathogens present before surgery, often overlooking the types of pathogens that can appear after the procedure. Additionally, there is a notable lack of research comparing the microbiological profiles of patients undergoing single versus revision amputations. Most existing studies in this area have focused more on analyzing patient histories and various clinical parameters rather than examining the specific bacteria involved.^18-20) Given the increasing number of DFIs and the complications associated with them, including the need for repeated surgeries, there is a critical need for more comprehensive data. Therefore, the goal of our study is to analyze the differences in pathogen detection and eradication in DFI patients who undergo single versus revision amputations. This study hypothesizes that pathogen diversity and pathogen persistence will differ between single and revision amputation cases, which could provide insight into tailored infection management strategies for high-risk patients.

We examined the medical records of 168 diabetic foot ulcer patients who received surgical treatment in the Orthopedic Surgery Department at Wonju Severance Christian Hospital from November 2020 to January 2023. A thorough medical history was compiled for each patient. The analysis covered participants’ ages, sexes, identified pathogens, C-reactive protein (CRP) and surgical details, including the types of surgeries performed.

DFIs were diagnosed according to the criteria established by the International Working Group on the Diabetic Foot (IWGDF) 2019 update guideline, using the Site, Ischemia, Neuropathy, Bacterial infection, Area and Depth (SINBAD) classification system, supplemented by magnetic resonance imaging findings to enhance diagnostic accuracy. All patients were diagnosed and managed by a single surgeon to ensure consistency in clinical assessments.

We categorized 168 patients who underwent lower extremity amputation surgery due to diabetic foot ulcers into two groups: the single amputation group and the revisional amputation group. The single amputation group included cases with no prior surgeries on the affected area, while the revision amputation group consisted of cases where the infection worsened in the same anatomical location, necessitating a secondary surgery. In the single amputation group, there were 113 cases, whereas the revisional amputation group consisted of 55 patients who underwent a total of 132 cases (Fig. 1).

Figure 1. Flow diagram.

For bacterial culture, deep tissue samples were collected during preoperative and intraoperative periods, while swab samples were taken postoperatively. Deep tissue samples were obtained according to established protocols, which included tissue biopsies or bone specimens stored in aseptic tubes and sent immediately to the microbiology laboratory for culture.^21,22) For both groups, we analyzed the pathogens detected in the wound before surgery and those identified in the wound 3∼5 days post-surgery. Swab samples were primarily used postoperatively and are acknowledged for being less reliable than tissue samples, particularly in identifying Gram-negative bacteria such as Escherichia coli and Citrobacter.²³⁾

Due to the diversity of pathogens, it was impractical to present the full scientific names of all species in each table and graph. Therefore, we assigned codes to each pathogen based on its type to streamline the analysis and presentation. Specifically, Gram-positive bacteria were coded as “1,” Gram-negative bacteria as “2,” and fungal species as “3.”

For postoperative pathogen survival analysis, we defined the ‘survival rate’ as the frequency of pathogens that persisted postoperatively divided by the frequency of pathogens identified preoperatively. This metric was used to assess the persistence of specific pathogens after surgical intervention, allowing for a comparative analysis between single and revision amputation groups.

The CRP levels were analyzed based on blood tests conducted around the same time as specimen collection. Following the approach used by Bravo-Molina et al.,²⁴⁾ which demonstrated that elevated CRP levels can predict the risk of major amputation, we collected and included CRP values in our analysis to evaluate their association with surgical outcomes.

All analyses were conducted using the R statistical software (version 4.1.0; R Foundation for Statistical Computing). Categorical variables are expressed as frequencies and percentages. The diversity and frequency of bacterial species identified before and after surgical procedures were evaluated using data aggregation and transformation techniques suitable for non-normally distributed variables. Statistical significance for differences in bacterial frequencies was assessed using the chi-squared test for categorical variables and the Mann^–Whitney U-test for continuous variables related to bacterial counts. All statistical tests were two-sided, and significance was set at p<0.05.

This study protocol was approved by the Institutional Review Board of Yonsei University Wonju Severance Christian Hospital (Approval No: CR323082). All study procedures were performed in accordance with the ethical standards of the institutional and/or national research committee. Data were collected from patients who consented to participate, excluding those who expressed a clear intention to withdraw or whose data were insufficient.

The patient demographics, as summarized in Table 1, show the distribution and characteristics of participants in the single and revision amputation groups. Among the 168 patients with diabetic foot ulcers, a total of 34 different pathogens were identified and coded by type, as shown in Table 2.

Table 1 . Patient Demographics

	Total	Single	Revision	p-value
No. of patient	168	113	55
Age (yr)	67±13	65±14	69±10	0.048
Height (cm)	164.5±9.5	162.0±11.0	169.6±5.2	>0.05
Weight (kg)	63.6±12.6	61.2±12.7	68.5±12.4	0.022
Male/female ratio	4.25:1	3.70:1	5.88:1	>0.05
Male	136	89	47
Female	32	24	8

Values are presented as mean±standard deviation or number only.

Table 2 . Identified Pathogens

Pathogen	Code
Gram-positive aerobes
Staphylococcus
MSSA	1-1s
MRSA	1-1r
Staphylococcus epidermidis	1-1b
Staphylococcus lugdunensis	1-1c
Staphylococcus haemolyticus	1-1d
Staphylococcus caprae	1-1e
Streptococcus
Streptococcus oralis	1-2a
Streptococcus agalactiae	1-2b
Streptococcus dysgalactiae	1-2c
Streptococcus anginosus	1-2d
Streptococcus constellatus	1-2e
Enterococcus
Enterococcus faecalis	1-3a
Enterococcus faecium	1-3b
Enterococcus avium	1-3c
Corynebacterium
Corynebacterium striatum	1-4a
Corynebacterium jeikeium	1-4b
Microbacterium
Microbacterium oxydans	1-5a
Dermabacter
Dermabacter hominis	1-6a
Bacillus	1-7a
Gram-negative aerobes
Serratia
Serratia marcescens	2-1a
Serratia liquifaciens	2-1b
Pseudomonas
Pseudomonas aeruginosa	2-2a
Pseudomonas monteilii	2-2b
Klebsiella	2-3a
Enterobacter
Enterobacter cloacae	2-4a
E.coli	2-5a
Proteus
Proteus mirabilis	2-6a
Proteus vulgaris	2-6b
Aeromonas
Aeromonas hydrophila	2-7a
Acinetobacter
Acinetobacter baumannii	2-8a
Morganella
Morganella morganii	2-9a
Stenotrophomonas
Stenotrophomonas maltophilia	2-10a
Burkholderia
Burkholderia cenocepacia	2-11a
Fungal
Candida albicans	3-1a

MSSA: methicillin-sensitive Staphylococcus aureus, MRSA: methicillin-resistant Staphylococcus aureus.

Comparing the variety of pathogens detected in the wound samples of each patient prior to their first surgical procedure revealed a statistically significant difference between the two groups (p=0.0029). Patients in the single amputation group had an average of 1.45±0.58 different pathogens identified, whereas patients in the revision amputation group had an average of 1.93±1.07 different pathogens detected.

1. Preoperative pathogen detection and postoperative pathogen survival rates

In the single amputation group, which included 113 patients, preoperative pathogen detection frequencies are presented in Table 3. A total of 111 organisms (73.0%) were identified as Gram-positive, while 34 organisms (22.3%) were Gram-negative. Among the Gram-positive aerobes, Enterococcus faecalis was the most commonly isolated organism (13.2%), followed by Streptococcus agalactiae (10.5%), Staphylococcus epidermidis (9.9%), MRSA (9.9%), and methicillin-sensitive Staphylococcus aureus (MSSA) (6.6%). For Gram-negative organisms, P. aeruginosa was the most frequently isolated (8.6%).

Table 3 . Single Amputation Group Pathogen

Pathogen	Pre-op frequency	Survival rate (%)
Gram-positive aerobes	111 (73.0)
Enterococcus faecalis	20 (13.2)	20
Streptococcus agalactiae	16 (10.5)	12.5
Staphylococcus epidermidis	15 (9.9)	20
MRSA	15 (9.9)	13.3
MSSA	10 (6.6)	10
Others	35 (23.0)
Gram-negative aerobes	34 (22.4)
Pseudomonas aeruginosa	13 (8.6)	23.1
Others	21 (13.8)
Fungal
Candida albicans	7 (4.6)	14.3
Total	152

Values are presented as number (%).

MRSA: methicillin-resistant Staphylococcus aureus, MSSA: methicillin-sensitive Staphylococcus aureus.

In terms of postoperative survival rates, among the Gram-positive organisms, E. faecalis and S. epidermidis both had a survival rate of 20.0%, followed by MRSA at 13.3%. For the Gram-negative organisms, P. aeruginosa showed a survival rate of 23.1% (Fig. 2).

Figure 2. Detailed pathogen profiles: single amputation group.Pre-Op: preoperative.

In the revision amputation group, which included 55 patients, preoperative pathogen detection frequencies are presented in Table 4. Here, 225 organisms (57.0%) were identified as Gram-positive, and 155 organisms (39.2%) were Gram-negative. Among the Gram-positive aerobes, S. epidermidis was the most frequently detected organism (9.9%), followed by E. faecalis (9.1%), MSSA (8.4%), and MRSA (6.6%). Among the Gram-negative organisms, Proteus mirabilis was the most commonly isolated (9.9%), followed by P. aeruginosa (5.3%), Morganella morganii (4.1%), and E. coli (3.5%).

Table 4 . Revision Amputation Group Pathogen

Pathogen	Pre-op frequency	Survival rate (%)
Gram-positive aerobes	225 (57.0)
Staphylococcus epidermidis	39 (9.9)	28.2
Enterococcus faecalis	36 (9.1)	33.3
MSSA	33 (8.4)	30.3
MRSA	26 (6.6)	38.5
Streptococcus agalactiae	20 (5.1)	30
Corynebacterium striatum	16 (4.1)	31.3
Others	55 (13.9)
Gram-negative aerobes	155 (39.2)
Proteus mirabilis	39 (9.9)	23.1
Pseudomonas aeruginosa	21 (5.3)	23.8
Morganella morganii	16 (4.1)	12.5
Escherichia coli	14 (3.5)	21.4
Others	65 (16.5)
Fungal
Candida albicans	15 (3.8)	26.7
Total	395

Values are presented as number (%).

MSSA: methicillin-sensitive Staphylococcus aureus, MRSA: methicillin-resistant Staphylococcus aureus.

For postoperative survival rates in the revision group, MRSA showed the highest survival rate among the Gram-positive organisms at 38.5%, followed by E. faecalis at 33.3%, MSSA at 30.3%, and S. epidermidis at 28.2%. Among the Gram-negative organisms, P. aeruginosa had the highest survival rate at 23.8%, followed by P. mirabilis at 23.1%, and E. coli at 21.4% (Fig. 3).

Figure 3. Detailed pathogen profiles: revision amputation group.Pre-Op: preoperative.

The seven most frequently detected pathogens in both groups^—MRSA, MSSA, S. epidermidis, E. faecalis, S. agalactiae, P. aeruginosa, and Candida albicans^—were analyzed to compare postoperative survival rates between the single and revision amputation groups. Although no statistically significant differences were observed in the survival rates for these pathogens between the two groups (p>0.05 for all pathogens), a trend toward higher survival rates in the revision group was noted.

2. Changes in CRP levels following the eradication of preoperative detected pathogens

In the single amputation group, postoperative CRP levels showed inconsistent changes following the eradication of preoperatively detected pathogens. Although CRP levels tended to decrease postoperatively in most cases, there was significant variation, and the changes were not uniform (Fig. 4).

Figure 4. C-reactive protein (CRP) change after pathogen removal: single amputation group.

In the revision amputation group, postoperative CRP levels generally decreased when preoperatively detected pathogens were eradicated. Despite some variability, most pathogens were associated with an average reduction in CRP levels following their removal (Fig. 5).

Figure 5. C-reactive protein (CRP) change after pathogen removal: revision amputation group.

All detected pathogens were analyzed to assess postoperative CRP changes between the single and revision amputation groups. No statistically significant differences in CRP reduction were observed between the two groups for any specific pathogen (p>0.05 for all pathogens), though a trend toward higher CRP reduction was noted in the revision group.

Our study examined differences in pathogen detection and survival rates between single and revision amputation cases for DFIs, focusing on pathogen persistence post-amputation within a South Korean cohort. While prior research has largely focused on preoperative pathogen profiles,⁴⁾ this study provides unique insights into postoperative pathogen survival, particularly in revision surgeries where such data are limited.¹⁸⁾

Factors influencing the observed differences include the complexity and chronicity of infections in revision cases, leading to greater microbial diversity and higher resistance, as evidenced by a higher average pathogen count per patient.^12,18) In contrast, single amputation cases showed reduced pathogen diversity and survival, suggesting initial surgery may be more effective in early-stage infections.^18,20)

Our findings highlight a higher frequency of pathogen diversity in revision amputation cases, with a notable prevalence of Gram-positive organisms, particularly S. epidermidis, E. faecalis and MRSA. This aligns with studies in Western populations reporting Gram-positive aerobes as primary DFI pathogens.^3,7,17,18) However, our data uniquely show a higher survival rate of MRSA and other Gram-negative organisms such as P. aeruginosa in revision amputation cases, suggesting that repeated surgical interventions may favor the survival of more resistant or difficult-to-eradicate strains. This distribution is consistent with studies from Asia, where Gram-negative pathogens such as P. aeruginosa are frequently observed in DFI cases.^8-14) This is particularly significant as it implies the necessity for tailored antibiotic strategies and infection management protocols in revision cases to address these persistent pathogens effectively.^17,18)

Interestingly, our study also examined CRP levels as an inflammatory marker, following the method of Bravo-Molina et al.²⁴⁾ While CRP levels generally decreased postoperatively, indicating infection control, the variability in CRP response was more pronounced in revision cases, suggesting that pathogen eradication does not uniformly correlate with inflammation reduction in these patients. This variability contrasts with previous studies that demonstrated a clear association between CRP reduction and infection resolution.²⁴⁾ These results suggest that CRP might be less reliable as a sole marker of infection control in complex cases involving multiple surgical interventions.

Our study contributes to the current understanding by offering insights into pathogen behavior post-amputation and the differing infection risks associated with single versus repeated surgeries. This evidence underscores the importance of consistent pathogen surveillance and individualized postoperative care plans to manage persistent infections in diabetic patients requiring revision amputations. Future research should focus on refining diagnostic tools and developing targeted interventions for managing high-risk, recurrent infections in DFI patients.

This study has several limitations. First, due to limited sample sizes for some pathogens, p-values did not consistently demonstrate statistical significance, and our findings are therefore presented with an emphasis on observed trends rather than definitive statistical outcomes. Additionally, the postoperative wound swab method used to monitor pathogen persistence in closed wounds carries a risk of contamination, potentially affecting identification accuracy. Although specimens were carefully collected to minimize contamination, this method has limitations in reliably detecting certain pathogens. Furthermore, as a single-center, retrospective study, these findings may not be fully generalizable to other populations or healthcare settings. Finally, factors such as comorbidities and diabetes duration, which could affect infection complexity, were not specifically analyzed, limiting the assessment of their impact on outcomes. Future studies should consider these variables for a more comprehensive evaluation.

Our study reveals significant differences in pathogen detection and survival between single and revision amputation cases for DFIs. Revision amputation cases demonstrated greater pathogen diversity and persistence, notably with higher survival rates among resistant strains such as MRSA, Gram-negative organisms like P. aeruginosa, and C. albicans, highlighting the challenges of managing recurrent infections.

This research was financially supported by a grant from the Korea Health Technology R&D Project (grant number: HI23C0476) conducted by the Korea Health Industry Development Institute (KHIDI) and funded by the Ministry of Health and Welfare, Korea. It was also supported by the “Regional Innovation Cluster Development Program (R&D, grant number: P0025898, project number: 1415188707)” conducted by the Korea Institute for Advancement of Technology (KIAT) and funded by the Ministry of Trade, Industry and Energy, Korea.

None.