Empirical Antibiotic Treatment in Diabetic Foot Infection: A Study Focusing on the Culture and Antibiotic Sensitivity in a Population From Southern China

Abstract

Diabetic foot and subsequent diabetic ulcer infections are the most devastating complication of diabetes. This study was conducted to explore the bacterial spectrum, sensitivity of microbials, and analysis of the empirical antibiotic regimens in our health center. The study included patients with diabetic foot ulcer infection (DFI) seen from 2009 to 2014. The patients included had all information covering the physical examination, laboratory tests, and image examinations. We sent appropriately obtained specimens for culture prior to starting empirical antibiotic therapy in all participants. A total of 312 patients were included: 52, 112, 95 and 53 patients within uninfected, mild, moderate, and severe infection groups. The total percentages of Gram-positive cocci (GPCs) and Gram-negative rods (GNRs) were 54% and 48.8% (P = 0.63). The most common GPC was Staphylococcus aureus (22.4%) and GNR was Pseudomonas aeruginosa (11.9%). Methicillin-resistant Staphylococcus aureus was isolated from 21 patients (6.7%). Even in the mild infection group, there was no significant difference between GPC and GNR infection, irrespective of recent antibiotic use (P = 0.053). The most frequently used empirical antibiotics in our center were second-/third-generation cephalosporin ± clindamycin, both in the mild and moderate/severe infection groups. In our center, the amoxicillin/clavulanate or ampicillin/sulbactam (β-L-ase 1) and second-/third-generation cephalosporins were highly resistant to the common GNR (30%-60%). The ticarcillin/clavulanate, piperacillin/tazuobactam (β-L-ase 2), fluoroquinolone, and group 2 carbapenem had good sensitivity. This study presents a comprehensive microbiological survey of diabetic foot ulcers in inpatients and provides reliable evidence of the local microbial epidemiology and sensitivity of antibiotics, which may help us improve clinical outcomes in DFI patients.

Keywords

diabetic foot infection culture sensitivity antibiotics

The prevalence of diabetes mellitus has grown rapidly in recent decades. It is estimated that it will increase to nearly 600 million by 2035 around the world, and 80% of those with diabetes will live in developing countries.¹ In China, diabetes has become a public health problem, with a high prevalence of 11.6%, and the prevalence of prediabetes is 50.1%.² Diabetic foot, one of the most devastating complications of diabetes, can cause diabetic disability and death. Nearly 80% of diabetes-related lower-extremity amputation is caused by a foot ulcer, and 15% to 25% of those with diabetes probably develop diabetic foot ulcers during their lifetime.^3-5 Diabetic foot ulcer infection (DFI) is a frequent (40%-80%) complication in these patients⁶ and results in substantial morbidity, such as reduced physical and mental quality of life,⁷ need for health care provider visits, antimicrobial treatment, and surgical resections or amputation.^8-12 What is more, DFI was the most common reason for hospitalization and a huge economic burden on patients and their families.^6,10

Antibiotic therapy is the primary treatment for diabetic ulcer infection, besides surgical intervention. Timely and effective antibiotic therapy is often associated with better clinical outcomes. The choice of antibiotics should be based on the likely or proven causative pathogens, their antibiotic susceptibilities, the clinical severity of the infection, and evidence of antibiotic efficacy for DFI and expense. Because of lagging culture results, it is usual for most patients to be given an antibiotic regimen empirically. Hence, it is crucial to obtain the following important information: patients’ comorbidities, previous antibiotic use, the local microbial epidemiology, and the sensitivity of antibiotics.

This study was a cross-sectional, retrospective study from a southern city in China. The study analyzed retrospectively the clinical characteristics of patients according to the different severities of DFI and aimed to explore the microbial spectrum and antimicrobial susceptibilities in the local region. Also, we analyzed retrospectively the empirical antibiotic regimens in our health center. We conducted the study with the hope of improving the clinical outcome of patients with DFI in our health care center.

Patients and Methods

We included all patients with diabetic foot ulcer admitted to the department of endocrinology and metabolism of the Sun Yat-sen Memorial Hospital of Sun Yat-sen University from the year 2009 to 2014. We obtained information on prior history of ulceration, amputation, angioplasty or vascular surgery, microvascular disease (retinopathy, renal disease), macrovascular disease (coronary heart disease and cerebrovascular disease), and recent antibiotic use. The patients included had all the information covering the physical examination, laboratory tests, and imaging examinations. The physical examination included details of blood pressure, height, weight, and pedal pulses; the peripheral neuropathy exam involved 10-g nylon monofilament, pinprick, and temperature and vibratory sensations. Laboratory tests included blood routine test, liver function, kidney function, lipid test, glucose, hemoglobin A1c, plasma albumin, uric acid, high-sensitivity C-reactive protein, urine albumin excretion rate, and estimated glomerular filtration rate. We acquired every patient’s important foot ulcer signs, assessed the depth and extent of the ulcer, and recorded the secondary signs (including nonpurulent secretions, friable or discolored granulation tissue, undermining of wound edges, and foul odor).¹³ For patients with suspicious DFI, we obtained cultures as soon as possible after cleaning and debridement of wounds. We collected specimens from deep tissue with biopsy or curettage. All the specimens were obtained in the operating theater and sent to the laboratory immediately for more reliable results. The culture was obtained before prescribing of empirical antibiotics. All patients underwent ankle-brachial index measurement, color Doppler ultrasound of the lower limb or computed tomography of the lower limb artery and peripheral nerve conduction velocity test, electrocardiogram, and eye examination.

The criteria for diagnosing peripheral arterial disease were as follows: a prior history of angioplasty or vascular surgery; symptoms and signs that included rest pain, intermittent claudication, and abnormal pedal pulses (weak or absent); ankle-brachial index measurement ≤0.9; and stenosis confirmed by color Doppler ultrasound or computed tomography of the lower-limb artery. The criteria for peripheral neuropathy were as follows: clinical symptoms of peripheral neuropathy (eg, abnormal sensation or neuropathic pain); abnormal results in a 10-g nylon monofilament; abnormal pinprick, temperature, and vibratory sensations; and exclusion of other related disease.

The severity of any DFI was assessed and graded into 4 groups according to the Infectious Diseases Society of America (IDSA) criteria (2004)¹²: grade 1, uninfected; grade 2, mild infection; grade 3, moderate infection; and grade 4, severe infection. Empirical antibiotic therapy was defined as effective if clinical symptoms and signs improved—for example, remission of pain and improvement in swelling, erythema, warmth and purulent discharge, and systemic inflammatory response syndrome. Otherwise, it was considered to be ineffective. We defined response (effective or ineffective) to empirical antibiotic therapy at the time of the results of culture specimens coming out and before we estimated whether change was needed in the initial antibiotics. The patient was diagnosed with osteomyelitis according to the combination results of the probe-to-bone, serum inflammatory markers, plain X-ray, and magnetic resonance image.

We selected a particular empirical antibiotic based on the IDSA recommendation.^12,14 The factors we took into consideration were as follows: age, diabetic complications, combined diseases, history of therapy, possible microbiology, the economic condition of the patients, and the nature of the drugs.

This study was approved by the ethics committee of the Sun-yat sen Memorial hospital affiliated to Sun Yat-sen University (2015 No. 51, 25-12-2015) and was in accordance with the principle of the Helsinki Declaration II.

Statistical Analysis

SPSS18.0 was used for statistical analysis. If the numerical variables were nonnormally distributed, we described them as medians (interquartile ranges) and compared them using the nonparametric test (2 independent-samples Mann-Whitney U test). Otherwise, the data were described as means (±SD) and compared using the t test. Categorical variables were described in terms of frequency and compared using the χ² test. P values ≤0.05 were identified as statistically significant.

Results

Patients

A total of 312 patients were included: 52, 112, 95, and 53 patients in the uninfected, mild infection, moderate infection, and severe infection groups, respectively. There were 264 patients with positive cultures. In this study, the mean diabetes duration was about 10 years. More than 85% of patients had peripheral artery disease, and 70% had neuropathy; the diabetic foot ulcers of 73.4% of patients were chronic wounds, and 71% had previous antibiotic use history before hospitalization. Compared with the uninfected and mildly infected patients, the moderately and severely infected patients had poorer nutritional status, such as lower total cholesterol, hemoglobin, plasma albumin, and uric acid. The patients in the latter group needed longer hospital stays and higher expense in the use of antibiotics (shown in Table 1).

Table 1.

Comparison of Clinical and Biochemical Characteristics Between the Uninfected and Mildly Infected, and Moderately and Severely Infected Groups.^a

	IDSA Classification of Infection		P Value
	Uninfected and Mildly Infected (164 Patients)	Moderately and Severely Infected (148 Patients)	P Value
SBP (mm Hg)	145 (130-160)	139 (123-157)	0.200
DBP (mm Hg)	79 (72-87)	78 (70-80)	0.333
HbA1c (%)	8.61 (6.95-10.10)	8.86 (7.00-10.73)	0.507
TC (mmol/L)	6.71 (3.79-5.21)	4.28 (3.20 ± 4.89)	0.001
HDL-C (mmol/L)	1.09 (0.91-1.26)	0.99 (0.75-1.21)	0.009
Hb (g/L)	117.72 (107.0-130)	106.50 (92.0-120.0)	0.000
Albumin (g/L)	37.11 (34.70-39.70)	34.50 (30.6-37.78)	0.000
Uric acid (µmol/L)	381.0 (284.5-466.5)	334.1 (236.3-428.5)	0.001
eGFR (mL/min)	49.5 (37.1-60.7)	51.8 (36.1-62.5)	0.281
hsCRP (mg/L)	38.09 (3.22-49.03)	79.71 (12.33-104.6)	0.000
WBC (×10⁹)	8.39 (6.66-9.59)	10.76 (7.29-13.18)	0.000
Neu (%)	67.02 (61.55-73.45)	70.89 (64.00-79.40)	0.000
Major amputation^b	0	3	0.106
Minor amputation^c	2	2	1.000
Peripheral artery disease	142 (86.6%)	138 (93.2%)	0.062
Peripheral neuropathy	121 (73.8%)	117 (79.1%)	0.289
Hospital stays (days)	19.46 (9.5-22.5)	24.62 (13.0-30.5)	0.001
Cost of antibiotics (yuan)	3328.3 (45.0-4648.1)	5454.9 (1418.9-7450.5)	0.000

Abbreviations: IDSA, Infectious Diseases Society of America; SBP, systolic blood pressure; DBP, diastolic blood pressure; HbA1c, hemoglobinA1c; TC, total cholesterol; HDL-C, high density lipoprotein cholesterol; Hb, hemoglobin; eGFR, estimated glomerular filtration rate; hsCRP, high-sensitivity C-reactive protein; WBC, white blood cell count; Neu, neutrophil percentage.

The number was presentation as median (liner-quartile), except the data of major and minor amputation(showed as number);the data of peripheral artery and neuropathy were showed as number(percentage).

Amputation above the ankle.

Amputation below the ankle.

Pathogens

In our study, the most common Gram-positive coccus (GPC) was Staphylococcus aureus (22.4%). Pseudomonas aeruginosa (11.9% of patients) was the most frequently isolated Gram-negative rod (GNR; Figure 1). The total percentages of GPC and GNR were 54% and 48.8%, respectively (P = 0.063, shown in Table 2). Methicillin-resistant Staphylococcus aureus (MRSA) was isolated from 21 patients (6.7%), which accounted for 30% of Staphylococcus aureus. In the mild infection group, there were 28 (60.9%) GPCs and 18 (39.1%) GNRs in patients with no recent history of antibiotic use, whereas it was 26 (41.3%) and 37 (58.7%) in patients with recent history of antibiotic use (P = 0.053).

Table 2.

Spectrum of Microbes Isolated and Recent Antibiotic Therapy History in the 4 Groups.^a

	Uninfected	Mild Infection	Moderate Infection	Severe Infection	Total, n (%)	P Value
Recent antibiotic use	66.7	66.7	80.0	70.0	71.5	0.180
Spectrum of microbes isolated						0.063
Culture negative	16 (30.8)	19 (17.1)	12 (12.6)	6 (11.3)	53 (17.0)
Gram-positive cocci	14 (26.9)	36 (32.4)	38 (40.0)	18 (34.0)	106 (34.1)
Gram-negative rod	13 (25.0)	37 (33.3)	28 (29.5)	12 (22.6)	90 (28.9)
Both	9 (17.3)	19 (17.1)	17 (17.9)	17 (32.1)	62 (19.9)

Recent antibiotic use is shown as percentage; spectrum of microbes isolated is shown as number (%).

Antibiotic Susceptibilities

Methicillin-susceptible Staphylococcus aureus (MSSA) and Streptococcus agalactiae were highly resistant to clindamycin (30%-50%). MRSA and coagulase-negative staphylococci, which were almost methicillin-resistant coagulase negative Staphylococcus (MRCNS), were all sensitive to linezolid and vancomycin. The common GNR sensitivities to third-generation cephalosporins were not so good, with high resistance of about 30% to 60%. P aeruginosa was sensitive to the aminoglycoside, fluoroquinolone, piperacillin/tazobactam, and imipenem but more than 30% resistant to ceftazidime and cefoperazone-sulbactam (shown in Tables 3 and 4).

Table 3.

Analysis of Antibiotic Susceptibilities (Gram-Positive Cocci).

Antibiotics	MSSA	MRSA	Enterococcus faecalis	Staphylococcus haemolyticus	Streptococcus agalactiae	Staphylococcus epidermidis
β-L-ase 1	95.3	0	100	0	100	16.7
Clindamycin	67.4	16.7	—^a	37.5	56.3	16.7
Sulfamethoxazole	88.4	72.2	—^a	56.3	—^a	33.3
Ceftriaxone	95.3	0	—^a	0	100	16.7
FQ	93	44.4	96.9	12.5	87.5	33.3
Group 2 carbapenem	95.3	0	100	0	100	16.7
Vancomycin	100	100	100	100	100	100
Linezolid	100	100	100	100	100	100

Abbreviations: MSSA, methicillin-susceptible Staphylococcus aureus; MRSA, methicillin-resistant Staphylococcus aureus; β-L-ase 1, amoxicillin/clavulanate or ampicillin/sulbactam; FQ, fluoroquinolone with good activity against aerobic Gram-positive cocci (eg, levofloxacin or moxifloxacin); Group 2 carbapenem, imipenem, meropenem.

The Enterococcus faecalis was naturally resistant to cephalosporin, clindamycin, and trimethoprim/sulfamethoxazole.

Table 4.

Analysis of Antibiotic Susceptibilities (Gram-Negative Rod).

Antibiotics	Pseudomonas aeruginosa	Klebsiella pneumoniae	Escherichia coli	Morganella morganii	Proteus vulgaris
β-L-ase 1	—^a	60	40	0	85
Pip/tazo	80	100	100	85.7	100
Ceftriaxone	—^a	57.1	40	100	92.3
ceftazidime	66.7	76.2	55	100	100
Cef/sul	63.3	66.7	60	78.6	76.9
Amikacin	96.7	100	100	100	100
Sulfamethoxazole	—^a	56	48	72	62
FQ	96.7	90.5	50	85.7	92.3
Group 2 carbapenem	80	100	100	100	100

Abbreviations: Pip/tazo, piperacillin/tazobactam; Cef/sul, cefperazone-sulbactam; Group 2 carbapenem, imipenem, meropenem; FQ, fluoroquinolone with good activity against aerobic Gram-positive cocci (eg, levofloxacin or moxifloxacin); β-L-ase 1,amoxicillin/clavulanate or ampicillin/sulbactam; β-L-ase 2, ticarcillin/clavulanate,piperacillin/tazobactam.

Pseudomonas aeruginosa was naturally resistant to β-L-ase 1, sulfamethoxazole, and ceftriaxone.

The Empirical Antibiotic Therapy in Our Health Center

In all, 248 patients received empirical antibiotic treatment, and 180 patients (73%) responded successfully to the empirical antibiotic therapy; 91% of them did not need a change in the initial antibiotic regimens during the hospitalization. Of 248 patients, 199 had proven antibiotic susceptibilities. The empirical antibiotic regimen covered the bacterial spectrum of 74.4% of patients. Of 248 patients, 155 (62.5%) received empirical antibiotic regimens against GPCs, GNRs, and anaerobes simultaneously; 71 (28.6%) against P aeruginosa; and 5 (2.0%) against MRSA. The most common GPCs resistant to the empirical antibiotics were MRSA, Enterococcus faecalis, and MRCNS. The most common GNR was P aeruginosa.

In patients with foot ulcer infection in our center, the most frequent empirical antibiotic regimens selected were second-/third-generation cephalosporins in monotherapy and third-generation cephalosporins + clindamycin/metronidazole in combination therapy (Figure 2). Among patients with grade 1 infection or those who were uninfected, there were 40 (77%) patients who received antibiotic therapy in our center.

We divided the patients with foot ulcer infection who received empirical antibiotic treatment (208 patients) into effective and ineffective groups according to the clinical response to initial antibiotic regimens. The patients in the ineffective group were older and had higher inflammatory response (higher white blood cell count). The patients in the ineffective group had higher percentage of GNR, or GNR and GPC infection. Also, polymicrobial infection was more common in this group. Empirical antibiotics resistant to or not covering MRSA and P aeruginosa could result in failure, whereas this was not true for other microbes (Enterococcus faecalis, Staphylococcus haemolyticus, Staphylococcus epidermidis, and Klebsiella pneumoniae; Table 5).

Table 5.

Comparison of Clinical and Culture Characteristics Between Effective and Ineffective Groups in Patients With Infection Who Received Empirical Antibiotics Therapy.^a

	Effective (149)	Ineffective (59)	P Value
Clinical characteristics
Age (years)	64 ± 11	68 ± 12	0.017
Diabetes duration (years)	11 ± 10	10 ± 6	0.515
SBP (mm Hg)	140 ± 26	144 ± 23	0.365
DBP (mm Hg)	78 ± 16	80 ± 13	0.640
HbA1c (%)	9.0 ± 2.5	8.5 ± 2.2	0.176
TC (mmol/L)	4.57 ± 3.70	4.32 ± 1.38	0.618
TG (mmol/L)	1.50 ± 0.81	1.59 ± 0.90	0.492
LDL-C (mmol/L)	2.56 ± 0.93	2.68 ± 1.06	0.434
HDL-C (mmol/l)	1.06 ± 0.35	0.98 ± 0.26	0.116
WBC (×10⁹)	9.72 ± 4.26	11.39 ± 5.21	0.018
Hemoglobin (g/L)	111.9 ± 22.97	108.2 ± 16.11	0.189
Urea (µmol/L)	351.46 ± 127.5	337.42 ± 121.70	0.477
Peripheral arterial disease	137 (91.9)	56 (94.9)	0.564
Peripheral neuropathy	111 (74.5)	48 (81.4)	0.366
Mild infection	62 (41.6)	22 (37.3)	0.179
Moderate infection	59 (39.6)	19 (32.2)
Severe infection	28 (18.8)	18 (30.5)
Pathogen
Empirical antibiotic covering pathogen	97 (74.4)	17 (25.8)	0.005
GPC	57 (47.1)	16 (28.1)	0.027
GNR	40 (33.1)	21 (36.8)
GPC and GNR	24 (19.8)	20 (35.1)
One bacterium	78 (64.5)	21 (36.8)	0.002
Two bacteria	32 (26.4)	26 (45.6)
Polymicrobial (3 and more)	11 (9.1)	10 (17.5)
Antibiotic-resistant or uncovered strains
MRSA	3 (2.6)	11 (19.3)	0.000
Pseudomonas aeruginosa	2 (1.8)	7 (12.3)	0.007
Enterococcus faecalis	5 (4.4)	7 (12.3)	0.059
Staphylococcus haemolyticus	5 (3.5)	7 (14.0)	0.059
Staphylococcus epidermidis	4 (3.5)	3 (5.3)	0.429
Klebsiella pneumoniae	1 (0.9)	3 (5.3)	0.108

Abbreviations: SBP, systolic blood pressure; DBP, diastolic blood pressure; HbA1c, hemoglobinA1c; TC, total cholesterol; TG, triglycerides; LDL-C, low density lipoprotein cholesterol; HDL-C, high density lipoprotein cholesterol; WBC, white blood cell count; GPC, Gram-positive cocci; GNR, Gram-negative rod; MRSA, methicillin-resistant Staphylococcus aureus.

The numerical variables are shown as mean (±SD); otherwise the data are given as number (percentage).

Discussion

Our study was an open and retrospective study. In the present study, the participants had long diabetes duration, and most of them had diabetic complications. These comorbid conditions were considered to be risk factors for diabetic foot ulcer and infection.^10,15-18 Compared with the uninfected patients with mild diabetic foot, the patients with moderate-severe foot ulcer infection usually suffered from malnutrition. This was also seen in previous studies by Jiang et al¹⁹ and Chaturvedi et al.²⁰ This suggests that malnutrition is one of the signs of severe DFI. Therefore, it is important to identify patients with severe DFI with malnutrition and deal with them in a timely manner.

In our center, the most common GPC was Staphylococcus aureus, followed by Staphylococcus haemolyticus, Streptococcus agalactiae, and Staphylococcus epidermidis. The most frequent GNR was P aeruginosa, and then Klebsiella spp, Escherichia coli, Morganella morganii, and Proteus spp. This result was similar to that in a study from India, which explored microbiological profiles of pathogenic bacteria in DFIs.²¹ A numbers of studies have suggested that GPCs are the predominant organisms isolated from DFIs,^22-24 although some other studies showed contrary results. Gadepalli et al²⁵ reported that in183 isolates from 80 ulcer specimens, 28.7% were Gram negative and only 13.8% Gram positive. Shankar et al²⁶ also suggested that 51.4% were Gram-negative aerobes, followed by Gram-positive aerobes (33.3%). In our study, the percentages of Gram-positive and Gram-negative aerobes were 54% and 48.8%, respectively, with no significant difference. In the mild infection group, there was no significant difference between patients with or without antibiotics use, which was not consistent with the guidelines. Therefore, it was reasonable to select broad spectrum antibiotics initially in our center, even in patients with mild infection and no recent antibiotic use. This was not consistent with the guideline recommendations.

MRSA infections prolong wound healing times and hospitalization stays, increase the need for surgical procedures, and result in treatment failure.²⁷ Some studies reported that MRSA was isolated in almost 30% of DFIs.^22,28 A review, containing 20 studies from 1993 to 2007, found that the isolates of MRSA range from 5% to 30%.²⁷ In the present study, the prevalence of MRSA infection was 6.7%, a relatively lower level compared with previous studies. Previous studies reported that the probable risks for MRSA infection are prior long-term or inappropriate use of antibiotics, previous hospitalization, long foot wound duration, the presence of osteomyelitis, and history of MRSA infection, which has been recognized as the most reliable predictor.^27,29-31 The IDSA guideline suggests that the empirical antibiotics regimen should cover MRSA in patients with previous history infection of MRSA, high local prevalence of MRSA, and very severe infection.¹⁴ Our study also showed that empirical antibiotics that did not cover or were resistant to MRSA would result in ineffective therapy response. So it is vital to screen for and identify risk factors for MRSA infection. Our center was not a high infection region of MRSA. It was unnecessary to cover MRSA regularly, unless there was high risk of MRSA.

A number of studies of complicated skin infections in developed countries reported that P aeruginosa was isolated in <10% of wounds. In the present study, P aeruginosa was isolated in 11.9% of wounds, which was a relatively high local prevalence. The probable reasons for this were that our hospital was located in Guangdong province, a city in southern China, where the climate is always warm and humid.^32-34 The 2015 International Working Group on the Diabetic Foot (IWGDF) guideline recommended to cover P aeruginosa empirically in moderately and severely infected patients with macerated ulcer or from warm climates.³⁵ As a high prevalence region of P aeruginosa, we should pay more attention to this organism, and increase the choice of empirical antibiotics that cover the organism in moderate and severe infection.

Although there was a high frequency of isolates, Enterococcus faecalis and coagulase-negative staphylococci were considered to be usually of secondary clinical importance and likely colonizers.³⁶ Our study also showed that not covering the above 2 kinds of organisms did not affect the clinical outcome of empirical antimicrobial treatment. So we suggest that it is usually unnecessary to target antibiotic treatment against the above organisms, unless the patients had a failure response to the selected antibiotics that did not cover the above organisms.

In the present study, 77% of grade 1 infected/uninfected patients assessed by IDSA criteria also received the empirical antibiotic treatment. The possible reason could be that most patients in this study had peripheral vascular neuropathy complications and chronic wound duration, and it was sometimes difficult to identify infection. In our center, the patients with atypical but clinically suspicious signs of infection also received the empirical antibiotics treatment. There may have been some excessive medication treatment in these patients. Anyway, for uninfected wounds, taking into account antimicrobial resistance, financial cost, and drug-related adverse effects, empirical antibiotic therapy was not recommended. We may need to change our empirical treatment policy later in this population.

In our center, infections with both GPCs and GNRs were common, no matter the severity of infection. Hence, in our study, most of the empirical antibiotic agents initially selected covered both GPCs and GNRs. This was consistent with the spectrum of microbes in the present study, although not consistent with the guideline recommendations.³⁵

Nevertheless, the effectiveness of empirical antibiotic treatment in this study was only about 73%, a little lower than the result of 85% reported by Balakrishnan et al.³⁷ The possible reasons were as follows. First, the most frequent antibiotic regimens selected in our study were third-generation cephalosporins or third-generation cephalosporins + clindamycin/metronidazole. But antibiotic susceptibilities showed that this empirical regimen was not enough against the common GNR. Antibiotics that have relatively high sensitivity to GNRs, such as piperacillin/tazobactam, group 2 carbapenem, and fluoroquinolone, were selected rarely. Second, this region had a high prevalence of P aeruginosa. The guideline recommends covering it specifically with β-Lase-2, semisynthetic penicillinase-resistant penicillin (S-S pen) + ceftazidime/antipseudomonal fluoroquinolone (cipro), and group 2 carbapenem, whereas in the present study, the selection of empirical antibiotics covering the organism was no more than 30%. Third, most patients in our study had peripheral vascular disease, and ischemic wounds were very common, whereas the empirical antibiotic regimen simultaneously covering GPC + GNR + Anaerobes was only about 60%. The above results demonstrated that the optimal empirical antibiotic regimens should be broader-spectrum agents, especially antibiotics with high sensitivity to GNRs (including P aeruginosa), such as piperacillin/tazobactam, group 2 carbapenem, and fluoroquinolone. What is more, we needed to improve the coverage for anaerobes in patients with ischemic foot wounds. In older patients and those with more severe infection, we need to be more attentive to GNRs and polymicrobial infection. What is more, MRSA and P aeruginosa were more common in these patients. Of course, the selection of antibiotics always needs to be made taking the patients and drug factors into consideration.³⁵

This study was a cross-sectional study from a southern city in China. It presented a comprehensive microbiological survey of diabetic foot ulcers in inpatients. The study provided reliable evidence of the local microbial epidemiology and sensitivity to antibiotics, which may help in improving clinical outcomes of DFI patients. The study also had some limitations. For example, the sample size was relatively small, and we did not do a follow-up study of the participants. Also, this was an open, retrospective study, not a controlled and randomized study, and all the participants came from a single center. In future, we need to enlarge the sample and conduct a multicenter study. We will further improve our study design and carry out a prospective study to explore more effective therapeutic regimens in the latter population.

Footnotes

Authors’ Note

Wen-Xia Wu and Dan Liu contributed equally to the study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Sun Yat-sen University Clinical Research 5010 Program (2010010).

References

Guariguata

Whiting

Hambleton

Beagley

Linnenkamp

Shaw

JE.

Global estimates of diabetes prevalence for 2013 and projections for 2035. Diabetes Res Clin Pract. 2014;103:137-149.

Wang

et al . Prevalence and control of diabetes in Chinese adults. JAMA. 2013;310:948-959.

Singh

Armstrong

Lipsky

BA.

Preventing foot ulcers in patients with diabetes. JAMA. 2005;293:217-228.

Boulton

Vileikyte

Ragnarson-Tennvall

Apelqvist

The global burden of diabetic foot disease. Lancet. 2005;366:1719-1724.

American Diabetes Association. Consensus Development Conference on Diabetic Foot Wound Care: 7-8 April 1999, Boston, Massachusetts. American Diabetes Association. Diabetes Care. 1999;22:1354-1360.

Prompers

Huijberts

Schaper

et al . Resource utilisation and costs associated with the treatment of diabetic foot ulcers: prospective data from the Eurodiale study. Diabetologia. 2008;51:1826-1834.

Raspovic

Wukich

DK.

Self-reported quality of life and diabetic foot infections. J Foot Ankle Surg. 2014;53:716-719.

Jeffcoate

Lipsky

Berendt

et al . Unresolved issues in the management of ulcers of the foot in diabetes. Diabet Med. 2008;25:1380-1389.

Lavery

Armstrong

Murdoch

Peters

Lipsky

BA.

Validation of the Infectious Diseases Society of America’s diabetic foot infection classification system. Clin Infect Dis. 2007;44:562-565.

10.

Lavery

Armstrong

Wunderlich

Mohler

Wendel

Lipsky

BA.

Risk factors for foot infections in individuals with diabetes. Diabetes Care. 2006;29:1288-1293.

11.

Lavery

Armstrong

Wunderlich

Tredwell

Boulton

AJ.

Diabetic foot syndrome: evaluating the prevalence and incidence of foot pathology in Mexican Americans and non-Hispanic whites from a diabetes disease management cohort. Diabetes Care. 2003;26:1435-1438.

12.

Lipsky

Berendt

Deery

et al . Diagnosis and treatment of diabetic foot infections. Clin Infect Dis. 2004;39:885-910.

13.

Gardner

Hillis

Frantz

RA.

Clinical signs of infection in diabetic foot ulcers with high microbial load. Biol Res Nurs. 2009;11:119-128.

14.

Lipsky

Berendt

Cornia

et al . 2012 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. Clin Infect Dis. 2012;54:e132-e173.

15.

Baba

Davis

TM.

A longitudinal study of foot ulceration and its risk factors in community-based patients with type 2 diabetes: the Fremantle Diabetes Study. Diabetes Res Clin Pract. 2014;106:42-49.

16.

Shahbazian

Yazdanpanah

Latifi

SM.

Risk assessment of patients with diabetes for foot ulcers according to risk classification consensus of International Working Group on Diabetic Foot (IWGDF). Pak J Med Sci. 2013;29:730-734.

17.

Hao

Zhang

Feng

Chai

Contribution of infection and peripheral artery disease to severity of diabetic foot ulcers in Chinese patients. Int J Clin Pract. 2014;68:1161-1164.

18.

Al-Rubeaan

Al Derwish

Ouizi

et al . Diabetic foot complications and their risk factors from a large retrospective cohort study. PLoS One. 2015;10:e0124446.

19.

Jiang

Ran

Jia

et al . Epidemiology of type 2 diabetic foot problems and predictive factors for amputation in China. Int J Low Extrem Wounds. 2015;14:19-27.

20.

Chaturvedi

Stevens

Fuller

Lee

Risk factors, ethnic differences and mortality associated with lower-extremity gangrene and amputation in diabetes. The WHO Multinational Study of Vascular Disease in Diabetes. Diabetologia. 2001;44(suppl 2):S65-S71.

21.

Ramakant

Verma

Misra

et al . Changing microbiological profile of pathogenic bacteria in diabetic foot infections: time for a rethink on which empirical therapy to choose? Diabetologia. 2011;54:58-64.

22.

Dang

Prasad

Boulton

Jude

EB.

Methicillin-resistant Staphylococcus aureus in the diabetic foot clinic: a worsening problem. Diabet Med. 2003;20:159-161.

23.

Citron

Goldstein

Merriam

Lipsky

Abramson

MA.

Bacteriology of moderate-to-severe diabetic foot infections and in vitro activity of antimicrobial agents. J Clin Microbiol. 2007;45:2819-2828.

24.

Goldstein

Citron

Nesbit

CA.

Diabetic foot infections: bacteriology and activity of 10 oral antimicrobial agents against bacteria isolated from consecutive cases. Diabetes Care. 1996;19:638-641.

25.

Gadepalli

Dhawan

Sreenivas

Kapil

Ammini

Chaudhry

A clinico-microbiological study of diabetic foot ulcers in an Indian tertiary care hospital. Diabetes Care. 2006;29:1727-1732.

26.

Shankar

Mohan

Premalatha

Srinivasan

Usha

AR.

Bacterial etiology of diabetic foot infections in South India. Eur J Intern Med. 2005;16:567-570.

27.

Eleftheriadou

Tentolouris

Argiana

Jude

Boulton

AJ.

Methicillin-resistant Staphylococcus aureus in diabetic foot infections. Drugs. 2010;70:1785-1797.

28.

Tentolouris

Petrikkos

Vallianou

et al . Prevalence of methicillin-resistant Staphylococcus aureus in infected and uninfected diabetic foot ulcers. Clin Microbiol Infect. 2006;12:186-189.

29.

Yates

May

Hale

et al . Wound chronicity, inpatient care, and chronic kidney disease predispose to MRSA infection in diabetic foot ulcers. Diabetes Care. 2009;32:1907-1909.

30.

Feng

Chu

Wang

Jun

Min

XM.

Risk factors and gene type for infections of MRSA in diabetic foot patients in Tianjin, China. Int J Low Extrem Wounds. 2013;12:106-112.

31.

Liu

Bayer

Cosgrove

et al . Clinical practice guidelines by the Infectious Diseases Society of America for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children. Clin Infect Dis. 2011;52:e18-e55.

32.

Martinez-Gomez Dde

Ramirez-Almagro

Campillo-Soto

Morales-Cuenca

Pagan-Ortiz

Aguayo-Albasini

. Diabetic foot infections: prevalence and antibiotic sensitivity of the causative microorganisms [in Spanish]. Enferm Infecc Microbiol Clin. 2009;27:317-321.

33.

Bansal

Garg

Bhatia

Attri

Chander

Spectrum of microbial flora in diabetic foot ulcers. Indian J Pathol Microbiol. 2008;51:204-208.

34.

Shakil

Khan

AU.

Infected foot ulcers in male and female diabetic patients: a clinico-bioinformative study. Ann Clin Microbiol Antimicrob. 2010;9:2.

35.

Bakker

Apelqvist

Lipsky

Van Netten

Schaper

NC.

The 2015 IWGDF guidance on the prevention and management of foot problems in diabetes. Int Wound J. 2016;13:1072.

36.

Kopp

Silberstein

Derweesh

IH.

Laparo-endoscopic single-site (LESS) radical nephrectomy with renal vein thrombectomy: initial report. BMC Urol. 2010;10:8.

37.

Balakrishnan

Shahid

Fairuz

Ramdhan

“Does the National Antibiotic Guideline- 2008 remain applicable for treating diabetic foot infection?” A new evidence-based regional study on culture and sensitivity patterns in Terengganu population. Malays Orthop J. 2014;8:42-44.