Characteristics and Microbiological Profile of Patients with Diabetic Foot Infections in Kuantan, Pahang

Introduction: The number of people suffering from diabetic foot infection (DFI) has increased precipitously over the years in Malaysia, owing to increased population, urbanisation, the surge of number of people with obesity and physical inactivity. As one of the most dreaded complications of diabetes mellitus, DFI is associated with high morbidity and mortality. We aim to study the microbiological profile of patients with DFI at a university hospital in Kuantan, Pahang. Materials and methods: This retrospective study was carried out at at Sultan Ahmad Shah Medical Centre @IIUM (SASMEC @IIUM) from 1 January 2018 to 30 April 2019. Patients’ demographic data, types of infection and surgical intervention, and the microbiological profile were obtained from the medical records. Results: A total of 142 causative pathogens were cultured from 130 tissue samples, with an average of 1.09 pathogens per lesion. Majority of the pathogens were gram-negative pathogens (52.8%). Staphylococcus sp. was the most common pathogen isolated (22.5%). This was followed by Streptococcus sp. (10.6%), Pseudomonas sp. (9.2%), Morganella sp. (5.6%), Klebsiella sp. (4.9%), Enterobacter sp. (4.9%), and others. Among the 142 pathogens, there were 9 multidrug-resistant strains observed. Most of the antibiotics were effective against the gram-positive pathogens except benzylpenicillin, tetracyclin, fusidic acid and ciprofloxacin. Meanwhile, cefotaxime, amoxicillin and ampicillin-sulbactam were also not suitable against gramnegative pathogens. Oxacillin and sulfamethoxazole/ trimethoprim can be used as empirical antibiotics against gram-positive pathogens, while vancomycin should be reserved for patients with septic shock or suspected multidrug resistant strain infection. Piperacillin/tazobactam and ceftazidime can be used as empirical antibiotics against gram-negative pathogens. Conclusion: Early initiation of empirical antibiotic(s) is paramount to stymie the infection from getting worse while waiting for the identification of causative pathogens in the management of DFI. This study provides a guide for treating physicians to initiate the most appropriate empirical antibiotic in DFI.


INTRODUCTION
Diabetes mellitus, a major non-communicable disease in the world, has imposed a significant burden on the health care system 1 . In the US, financial burden of diabetes mellitus is estimated to cost $327 billion ($237 billion from direct medical cost and $90 billion from reduced productivity) 1 . Adding insult to that, the number of people with diabetes will continue to rise, owing to increased population, urbanisation, the surge of number of people with obesity and physical inactivity 2 . In 2004, Wild et al predicted the prevalence of diabetes will be 4.4% in 2030 with a total of 366 million people living with diabetes 2 . In Malaysia, the prevalence of diabetes mellitus has far exceeded the estimation by Wild et al. Based on the first National Health and Morbidity Survey (NHMS) in 1986 and the subsequent NHMSs, the prevalence of diabetes in Malaysia has shown a steadfast increment, from 6.3% in 1986 to 17.5% in 2015 (8.3% -1996; 11.6% -2006; 15.2% -2011) 3,4 . diabetic foot ulcers and 0.9% of them have previous amputations 6 . This is lower than the estimated 15-25% of diabetic patients having diabetic foot ulcers and 4.3% of diabetic patients having had lower leg amputation, possibly due to recruitment of patients only from Health Clinics (Klinik Kesihatan) by the Registry and not from hospitals where most of the patients present with more severe complications [7][8][9][10] . In fact, up to 30% of diabetics will develop a foot ulcer, and there is an average of one leg amputation every 20 seconds, causing diabetic foot infection being ranked as one of the top ten diseases that pose heavy burden globally [11][12][13][14] .
For the treatment of diabetic foot infection (DFI), aside from surgical procedures, it is paramount to initiate empirical antibiotic(s) early prior to isolation of the causative pathogens, to stymie the infection from getting worse 15 . The empirical antibiotics should be broad-spectrum and target majority of the predicted pathogens in DFI, based on the local microbiological profile 15 . Previous study by Kow et al demonstrated that the bacteriology profile of patients in Southeast Asia countries differs starkly compared to other developed countries which show a predominance of grampositive pathogens 5,16 . Even within Malaysia, the microbiological profile of patients with DFI varies among centres, in terms of percentage of gram-staining pathogens and the percentage of culture with growth 5 . With this context in mind, it is important to obtain the local microbiological profile in DFI. We, therefore, aim to describe the characteristics and microbiological profile of patients with DFI at a newly established university hospital in Kuantan, Pahang.

MATERIALS AND METHODS
This retrospective study was carried out at at Sultan Ahmad Shah Medical Centre @IIUM (SASMEC @IIUM), a 350bed teaching hospital in Kuantan, Pahang. This study was part of the SANDI project (Septic Arthritis and Diabeticrelated Infection) initiated by the authors. Ethical approval was obtained from the Kulliyah of Medicine Research Committee (Ref: IIUM/305/20/4/1/7). Medical records of patients who were admitted from 1 January 2018 to 30 April 2019 to IIUM MC were reviewed for suitability of recruitment into the study. All patients with diabetic foot infections who had received surgical interventions were included. For patients with history of multiple admissions, only the first admission was included. Patients with incomplete data were excluded from the study.
Demographic data such as age, gender, precipitating factors, duration of illness prior to hospital presentation, types of infection, and types of surgical intervention were extracted from patients' medical records. The precipitating factors included scratches, insect bite, improper shoe wear, burn, and trauma to the foot such as stepping on a nail or an object falling onto the foot etc. Types of infection were broadly classified into four categories, namely: abscess, infected wound, necrotising fasciitis and gangrene (both wet and dry gangrene), similar as the previous study 17 . On the other hand, types of surgical procedure were classified based on the level of amputation (if amputation was performed). There were also wound debridement, ray amputation or disarticulation of toe, mid-or hindfoot amputation, below-knee amputation, and above-knee amputation. There was no through-knee amputation or hip disarticulation during the study period.
The microbiological profile of DFI was also retrieved from the medical records, including type of specimen, causative pathogens, and the sensitivity of the pathogens to antibiotics tested. Only deep tissue or bone samples obtained during the surgery were included. In accordance with other studies, the samples obtained were incubated at the hospital microbiology laboratory for two days at 37 o C in blood agar, MacConkey agar and Chocolate agar 5,18 . Conventional method was performed to identify the cultured pathogen and the antibiotic susceptibility testing was carried out based on the National Committee for Clinical Laboratory Standards (CLSI) using the disk diffusion method on Mueller-Hinton agar plates 19 . Likewise, types and dose of antibiotics in the susceptibility testing were similar to previous studies 5, 18 . Descriptive data was used for the presentation of results.

RESULTS
Among the 130 patients included in this study, majority of them were males (n=77, 59.2%), with a male-to-female ratio of 1.45:1 ( Table I). Majority of patients were of the elderly age group, with a mean of 62 years (range 31 to 82 years). Most of the patients (67.7%) did not have any precipitating factors prior to developing diabetic foot infections and they normally presented to the hospital within 2 weeks after the development of initial symptoms, with a mean of 11.7 days (range 1 to 90 days) (Table I).

DISCUSSION
Consistent with other studies in Malaysia, patients with DFI in our study show a male predominance, in which male patients outnumber female patients by more than 40% 5,17,18,20 . More than two-third of the patients do not have any identifiable precipitating factor that initiate the infection. This is probably due to diabetic neuropathy, a common microvascular complication in diabetic patients, in which they have lost the protective sensation at the foot, rendering them vulnerable to repeated microtrauma to the foot and subsequently infections 5 . This is further compounded by other complications of diabetes mellitus such as immunopathy (dysfunction immune system), vasculopathy (inadequate blood supply) and autonomic dysfunction (dry and crack skin) which contribute to diabetic foot infection 5 .
Similar to the finding of a study done at a rural area in Pahang, patients from the urban city of Pahang also tend to present late to the tertiary hospital. Kow et al previously demonstrated that the duration of illness is a predictive factor of major amputation in patients with DFI, hence patient education should be emphasised for patients to seek medical attention early to prevent further morbidity and mortality 17 .
Most of the patients at the rural area present with infected wound which is also a finding recognised in our study at urban city 17 . The rate of major lower limb amputation for diabetic foot infection is 16.1% in this study which is lower than that reported in other countries, such as Singapore (27.2%) and Hong Kong (30.3%) 21,22 . In Malaysia, the rate of major amputation ranges from 14.2% to 20% 20,23 . A lower rate of amputation observed in Malaysia is likely due to refusal for amputation among the local population, especially major limb amputation, despite detailed counselling on its indication 24 .
In this cohort, we examine the difference between blood culture and intra-operative tissue culture in patients with DFI. In the management of sepsis, blood culture is considered the clinical gold standard to identify the causative pathogen 25 18 . This indicates that the treating physician in urban area need to take gram-positive pathogens into consideration before initiating the empirical antibiotic(s). Meanwhile, physicians can confidently initiate gram-negative-targeting antimicrobial therapy in rural areas of Malaysia, as majority of pathogens tend to be gramnegative in nature (67-95%) 5 .
Consistent with previous studies, although Staphylococcus sp. is the most common pathogen isolated in DFI, the percentage is much lower compared to that in western countries where it can be present in up to 50% of the diabetic foot wounds 5,16 . As for the pathogens' antimicrobial susceptibility, gram-positive pathogens are not sensitive to penicillin (50%) and fusidic acid (36.36%). This trend is also shared among patients from rural area, revealing the possibility of resistant strains due to over-prescribed antibiotics or over-the-counter self-medication 5  Our study has several limitations. First of all, the retrospective nature of this study precludes the inclusion of several information such as HbA1c, albumin level and type of antibiotic(s) taken by patients prior to admission. In addition, our centre does not routinely perform anaerobic or fungal sampling for patients with DFI. Despite the limitations mentioned, we manage to colligate the microbiological profile of patients with DFI in this newlyestablished tertiary centre, and it can serve as a guide to dictate the type of antimicrobial therapy for patients with DFI. Nevertheless, each hospital is recommended to obtain its own microbiologic profile of patients with diabetic foot infections and treat their patients with the most appropriate empirical antibiotic(s).

CONCLUSION
The microbiological profile of patients with DFI in Kuantan mimics those reported in Kuala Lumpur of Malaysia. Although Staphylococcus sp. is the most common causative pathogen, Streptococcus sp, Pseudomonas sp, and Morganella sp. are identified as significant causative organisms in our centre. This microbiological profile serves as a guide for treating physicians to initiate the most appropriate empirical antibiotic(s), based on the severity of disease.