Clinical, Pathological and Microbiological Evaluation of Diabetic Foot Syndrome

Background and objectives: Diabetic foot ulcer (DFU) is one of the serious complications of diabetes, being related to frequent and long-term hospitalisation, reduced quality of life of the patient, amputations, a high rate of morbidity and mortality. The bacterial aetiology is complex, sometimes involving more than one pathogen, playing a major role in the infection prognosis and development of microbial resistance. This study evaluated the current state of the aetiology, clinical and pathological characteristics of DFU in a single diabetes centre in order to provide some specific measures to prevent it. Materials and Methods: This retrospective study was conducted on patients with diabetes mellitus (252 individuals diagnosed with DFU) between January 2018–December 2019. All participants were assessed based on their clinical characteristics, including complications of diabetes and pathological and microbiological evaluations. Results: The present research revealed that diabetic foot ulcer prevalence was higher in males than in females and higher in type 2 diabetic patients than in type 1 diabetic patients. The patients with diabetic foot ulcer were older, had a higher body mass index (BMI), longer diabetic duration and had more diabetic complications, such as retinopathy, diabetic polyneuropathy and diabetic kidney disease, than patients without diabetic foot ulceration. Conclusions: Taking into account all factors involved, including the aetiology and the antibiotic susceptibility pattern of these isolates, planning the suitable treatment options of patients is possible.


Introduction
The incidence of diabetes is expected to increase rapidly-from 58 million cases in 2017 to a predicted value of 67 million cases by 2045 in Europe, especially in low-and middle-income countries, having an extremely strong cost impact on all public health systems [1]. There are many explanations, such as unhealthy diets, urbanisation, increasingly sedentary lifestyle and, at the same time, inadequate resources for preventive or medical care for populations, followed by higher rates of obesity and diabetes in many countries. The World Health Organization defined diabetes foot ulcer (DFU) as an

Patient Selection
The retrospective hospital-based study was performed for two years (between January 2018 and December 2019) for inpatients of the Diabetes Clinic of Emergency County Clinical Hospital of Oradea (located in Oradea, North-western Romania) diagnosed with DFU. Demographic and clinical characteristics, associated factors and laboratory data were extracted from the hospital computer system and medical records of patients. The medical records included type, duration and treatment of diabetes and comorbidities/complications (peripheral neuropathy and arterial disease, amputations and ulcerations). Out of 2992 diabetic patients, 252 patients diagnosed with DFU were included in the study group. A control group was selected from the rest of the diabetic patients to explore the risk factors associated with DFU. Their inclusion was done by representative sampling; every 10th record of the 2740 patients without DFU was considered, resulting in a group of 274 patients evaluated as a control group.
The ulcers were graded consistent to the Wagner-Meggitt's classification as follows: 0-intact skin, pain only; 1-superficial and localised ulcer of skin or subcutaneous tissue; 2-deep ulcers to tendon, bone or capsule; 3-deep ulcer with bone involvement or abscess; 4-gangrene of toes or forefoot and 5-midfoot or hindfoot gangrene (full-foot gangrene) [11,12]. In order to assess the vascular component (ischemia) and the neurological impairment, respectively, to establish the type of lesion, the ankle brachial index (ABI) was used; if the ABI is <0.9, the patient is diagnosed with peripheral arterial disease, and the lesion was considered "ischemic-neuropathic"; otherwise, if the ABI ≥0.9, the lesion was considered "neuro-ischemic".

Sample Collection and Laboratory Tests
Laboratory tests included, for all patients, glycated haemoglobin (HbA1c), complete blood count (CBC), C-reactive protein (CRP), creatinine and microalbuminuria and a culture of samples collected from DFUs, according with internal protocols. Swabs were used to collect the pus sample from deeper portions of the ulcer by making a rotatory movement with the swab after using aseptic techniques to avoid contamination. The samples were transported to the laboratory within the shortest time, and they were processed by inoculation on culture media as follows: Columbia blood agar base, Levine agar (eosin methylene blue) and anaerobes culture media (all from Merck Romania SRL, Bucharest, Romania) and incubated at 37 • C for 24-48 h. The identification of isolates was performed using Maldi Tof mass spectrometry and antibiotic sensitivity by Vitek-2 Compact Systems (Biomerieux, Paris, France) and the Kirby-Bauer disk diffusion method. The isolate was classified as susceptible, intermediate or resistant based on the Clinical Laboratory Standards Institute (CLSI) criteria [13]. The antimicrobial discs used in this study included the antibiotics presented in Table 1. Extended spectrum of beta lactamase (ESBL) confirmatory test was performed by using ceftazidime (30 µg), ceftazidime/clavulanate (30/10 µg), cefotaxime and cefotaxime/clavulanate (30/10 µg) discs placed on a Muller Hinton Agar plate on which a 0.5 McFarland test organism was inoculated. After 16-18 h and at 35 ± 2 • C, the organism was considered as an ESBL producer if there was a 5-mm increase for either antimicrobial agent tested in combination with clavulanate versus the zone diameter of the agent tested alone. For isolates of Enterobacterales suspicious for carbapenemase production was performed the modified carbapenem inactivation method (mCIM test). Methicillin resistances of the Staphylococcus strains were evaluated using cefoxitin (30 µg) disc strains inoculated on a Mueller Hinton agar plate and incubated at 33-35 • C for 16-18 h. MDR pathogens were defined as acquired nonsusceptibility for at least one agent in three or more antimicrobial categories.
E. coli ATCC 25922, S. aureus ATCC 29213, P. aeruginosa ATCC 27853 and E. faecalis ATCC 29212 were used as quality control strains to check the quality of the culture media and antimicrobial cards and disks.

Ethical Statement
The study was approved by the Ethics Committee of the Emergency County Clinical Hospital of Oradea (no. 25,677/24.10.2019) and respected the principles of Good Clinical Practice and the World Medical Association Declaration of Helsinki-Ethical Principles for Medical Research Involving Human Subjects [14]. All patients included signed written informed consent prior to any study-related activities.

Statistical Analysis
The statistical analysis was performed with Excel software, using descriptive statistics as mean ± standard deviation (SD), chi square and an ANOVA test. p-values < 0.05 were considered significant.

Discussion
DFU is a frequent complication of diabetes, but a good control of the disease can reduce the risk of amputation and, also, improve the overall quality of life [15]. The prevalence of DFU in this research was 8.42%, close to the global prevalence of diabetic foot ulceration of 6.3% [2].
The results revealed by the current study are in accordance with other studies, as there were more males than females, and most patients were in the age group of 60-79 years [16,17]. The results revealed by the current research are consistent with those of other studies, which confirm the existence of more men than women affected by this disease, most patients being in the age group 60-79 years [2,16,17]. These data also show that the factors associated with the development of DFU are different in women than in men, as revealed in the study published by Navarro-Peternella et al. [18]. The glycaemic control was poor in more than 80% of patients. The same patients were overweight, and the duration of diabetes was >10 years. The contribution of obesity to the risk of DFU is questionable. There are studies revealing that it is associated with DFUs, while other studies show that BMI has no significant correlation [18][19][20][21]. The analysis revealed that patients with DFU had higher BMIs range (from 25 to 46 kg/m 2 ), and the DFU prevalence was higher in type 2 diabetes mellitus than in type 1 diabetes mellitus patients, which was consistent with the results of previous studies, but the mechanisms explaining it were not completely elucidated [20,[22][23][24].
The risk factors identified for the presence of DFUs were very similar to those identified in previous literature data, such as older age, poor glucose control, a high prevalence of diabetic neuropathy and a high prevalence of peripheral artery disease [22][23][24][25][26]. Coronary artery disease (CAD) had a significantly higher prevalence in patients with foot ulcerations, a fact frequently reported in literature, due to an aggregation of cardiovascular risk factors. It is estimated that cardiovascular-related mortality and morbidity is two to four times higher in DM patients diagnosed with diabetic foot ulceration compared with those without ulcerations [25,26]. The relationship is bidirectional; firstly, hyperglycaemia, hypertension and dyslipidaemia contribute both to the development of CAD and diabetic foot ulceration; additionally, the presence of diabetic foot ulceration and the infection is associated with an inflammatory reaction, a subclinical inflammation characterised by high levels of circulating cytokines such as tumour necrosis factor (TNF) or interleukin 6 (IL-6), resistin, macrophage inflammatory protein-1β and low circulating levels of adiponectin [26,27]; these cytokines have a negative role on the normal function of the endothelium, thus increasing the risk for subsequent cardiovascular events [28].
Careful evaluation of diabetic foot syndrome is essential due to the proinflammatory nature of diabetes. The investigation of any inflammatory syndrome is performed by using many markers (or combinations of them), but none has been completely validated for an accurate diagnosis of infections, including sepsis [29]. The most commonly used biomarkers of inflammation were CRP and leucocytes, while, for sepsis screening, besides these two, serum procalcitonin and presepsin are useful. Despite divergent data from various studies, CRP is used to assess acute exacerbations of chronic obstructive pulmonary disease or infectious complications of malignancies in order to reduce the use of antibiotics. Moreover, CRP has been included in point-of-care diagnostic devices [30,31].
Serum procalcitonin is used in many hospitals as a biomarker of sepsis and could differentiate the most common aetiologies. In cases of bacterial and fungal sepsis, the level of procalcitonin is elevated, while, in the case of viral infections, its level remains normal, or it is only slightly elevated. In the same context, procalcitonin is considered a useful marker to indicate guided antibiotics therapy and predict the mortality for patients with sepsis [32,33] The Wagner-Meggitt classification is used for grading DFUs in many settings. According to this classification, patients that were included in this research-most of them in stage II, III and IV-had almost the same results from other studies, meaning lesions from deep ulcers to tendon, bone or capsule to deep ulcers with bone involving the abscess and gangrene of toes or forefeet [12,34,35]. Evidence of infection was seen in the majority of patients, and many DFUs are polymicrobial. The present study revealed a higher percentage of monomicrobial infections, findings similar to other authors [36,37]. Gram-negative organisms were isolated in greater numbers than Gram-positive organisms, and Escherichia coli was the predominant bacterial etiological agent reported, which is not correlated with other papers (where P. aeruginosa was described as predominant) [38,39].
S. aureus was the predominant isolate in this research, as other findings show as well [37,[40][41][42], and the methicillin-resistant (MRSA) was reported in more than 50% cases, which has consequences in terms of therapeutic options [43]. According to the susceptibility rates, in the treatment of DFUs caused by S. aureus, quinolones (levofloxacin) and amoxicillin/clavulanic acid could be used and glycol-peptides and oxazolidinone antibiotics (linezolid) in the case of MRSA. Linezolid appears to be more effective than vancomycin in the treatment of diabetic foot [44]. The same therapeutic options (glycol-peptides, oxazolidinone and quinolones) are also effective for enterococci and streptococci, but, in this case, ampicillin and macrolides are therapeutically effective as a first option.
The present analysis showed that the most common Gram-negative bacilli types among patients with foot syndrome were Escherichia coli, Proteus spp., Pseudomonas aeruginosa and Klebsiella spp., which accounted more than 40% of the total strains. Additionally, the Enterobacterales family showed the highest susceptibility to amikacin, amoxicillin/clavulanic acid and carbapenems and, partly, to cephalosporins. The susceptibility of Escherichia coli to ceftriaxone was only 25%, and only half of them were sensitive to ciprofloxacin. The susceptibilities to levofloxacin were better than for ciprofloxacin in the case of the Enterobacterales family. In contrast, Pseudomonas aeruginosa was less susceptible to all antimicrobials tested, except for ceftazidime and meropenem. Pseudomonas aeruginosa was less susceptible to aminoglycoside antibiotics (especially to gentamycin).
Carriages or infections with MDR pathogens result in less treatment options and high mortality rates in patients. Skin and soft tissue infections, including DFUs, have increased every year, the frequency of antimicrobial MDR organisms-not only MRSA but, also, vancomycin-resistant enterococci (VRE)-extended spectrum beta-lactamase-producing and carbapenemase-producing Gram-negative organisms [45,46]. Obtained data showed 15 strains of ESBL that accounted for 8.02% of the Gram-negative bacilli, and most of them were Escherichia coli; these strains presented the highest susceptibility to carbapenems, followed by amikacin, levofloxacin and piperacillin-tazobactam.
A number of 6 out of 10 strains were MDR Acinetobacter baumannii. This strain was the most commonly isolated, especially from the respiratory tract in the ICU, and resistant to all the β-lactam antibiotics, including carbapenems [47,48]. The same characteristics were present in the strains isolated from diabetic foot lesions, which raised treatment difficulties [49,50].
In the healthcare field, it is well-known that "prevention is better than the treatment of the disease". Diabetic patients should be informed in a manner that they understand the issues related to their disease and how to prevent DFU and other complications. Patients usually need to be hospitalised for a long time or repeatedly in a specific period of time because of the symptom exacerbations; additionally, they need to take long-term medications for diabetes and complications. The treatment of infections in DFU should be done after identification of the pathogen and analysing the susceptibility to antibiotics, and a correct diagnostic is established [51,52]. Costs related to microbiological testing are variable with the method used, confirmation of a strain with a resistance phenotype, therapeutic options, etc.
The Kirby-Bauer disk diffusion method (that uses a number of eight antibiotics, from different classes, for testing the susceptibility of the pathogen to antimicrobials) has a lower cost than the one using Vitek-2 Compact Systems (as is specified in Section 2, it brings sufficient therapeutic options in the case of a patient with DFU onset) [45,46]. Infections with MDR pathogens or relapse of the disease need additional testing to confirm the resistance phenotype and more expensive therapeutic options.
This study has some limitations. Firstly, the patients included were from a single hospital. Secondly, they were evaluated inpatients for two years, resulting in a relatively small number of specimens for some pathogens. Anaerobic bacteria were not isolated, probably because of poor handling techniques/preservation methods for anaerobic organisms. However, the positivity rates of samples were 93.55%, and all pathogens were tested for susceptibility to antimicrobials.

Conclusions
DFU is a major complication of diabetic patients, resulting in amputation and determining a higher morbidity and mortality. These are reasons to screen, prevent and control the prevalence of diabetic foot ulceration. S. aureus and E. coli were found to be the most predominant isolated strains from DFUs. Clinicians, in a multidisciplinary approach, must assess risk factors and microorganisms involved to provide early diagnosis and proper therapy applied for minor injuries to avoid amputation. For example, fluoroquinolone and ceftriaxone, which are frequently used in the hospital where the study was performed, showed a significantly reduced susceptibility to Gram-negative bacilli, suggesting that drug susceptibility testing should be performed to select susceptible antibiotics for treatment. Knowing the aetiology and the antibiotic susceptibility pattern of these isolates is important for planning the appropriate treatment options of patients by selecting the appropriate antimicrobial and good glycaemic control and proper foot care.