Use of Tigecycline for Diabetic Foot Infections

Objective. The aim of this study is to evaluate tigecycline for diabetic foot infections (DFIs). Materials and Methods. In this prospective observational study, the investigators included patients who had consultation with the Diabetic Foot Council of Ege University Faculty of Medicine (Izmir, Turkey) between March 2013 and July 2015 and who used tigecycline during their treatment. Treatment success was assessed by design-specific criteria for each evaluation. Results. The study included 105 cases. Of those, 37 (35.2%) were women (mean [± standard deviation] age, 61.9 ± 11.9 years). The success rate of tigecycline treatment was 93.3% in mild infections, 56.2% in moderate, 57.7% in severe, and 61.9% in all cases. The authors found a 9-fold decrease (P = .046) in the success of tigecycline treatment among those who developed moderate or severe DFIs and a 6.4-fold decrease (P < .0001) among those who had arterial stenosis. For 33 (71.7%) of 46 (43.8%) patients who experienced a side effect, tigecycline treatment was continued as it could be tolerated. Conclusions. If tigecycline is to be the treatment choice, extra attention must be paid to patients with arterial stenosis, severe DFIs, and side effects. The common disadvantage is the high side effect rate, especially nausea. However, it is generally not necessary to discontinue the medication in cases with nausea. Therefore, tigecycline may be used as a choice of therapy in mild DFIs. 

Glycylcyclines are a novel class of antibacterial drugs, developed a decade ago as a new version of tetracycline.¹ Molecule and 30S ribosomal subunit interaction prevent the binding of aminoacyl-transfer ribonucleic acid and inhibit protein synthesis.¹ Different species of bacteria can cause diabetic foot infections (DFIs)^2,3; broad-spectrum antibiotics could solve this problem. Tigecycline (9-t-butylglycylamido-minocycline; GAR-936; Wyeth Laboratories, Collegeville, PA), the first type of glycylcycline antibiotics introduced, is derived from minocycline and developed in response to the emergence of many kinds of resistant organisms such as antibiotic-susceptible and/or antibiotic-resistant Gram-positive bacteria (eg, vancomycin-resistant enterococci, methicillin-resistant Staphylococcus aureus, and penicillin-resistant Streptococcus pneumoniae), anaerobes, and multidrug-resistant Gram-negatives.^4-7 In contrast, tigecycline is not suitable for treating Pseudomonas spp and Protea.³

Treatment of DFIs includes many disciplines and modalities. Severe infection and osteomyelitis are important issues that need to be overcome.^8,9 Previous studies^9,10 reported that tigecycline therapy has some disadvantages with the treatment of osteomyelitis and severe infections. The purpose of this study is to evaluate the success of tigecycline treatment for DFIs and determine the risk factors that contribute to the failure of this treatment

In this prospective observational study, the investigators included patients who had consultation with the Diabetic Foot Council of Ege University, Faculty of Medicine, Izmir ,Turkey, between March 2013 and July 2015 and who used tigecycline during their treatment. Patients were followed prospectively, and ethical approval was obtained from the Scientific Research Ethics Committee of Ege University, Faculty of Medicine, Izmir, Turkey.

Management protocol in brief
The Diabetic Foot Council met on a weekly basis to assess patients. Patients with diabetic foot ulcers who were to be treated with tigecycline were hospitalized. Patient medical history was recorded during admission. General laboratory tests, including those for blood glucose and other inflammatory markers, were performed at the beginning of tigecycline treatment. To evaluate the vascularity of the diabetic foot, Doppler ultrasound was used. Intravenous antibiotics were administered empirically or according to the results of the culture and sensitivity tests. An initial dose of 100 mg tigecycline, followed by 50 mg every 12 hours, was administered intravenously over about 30 to 60 minutes.¹¹ Osteomyelitis was diagnosed by magnetic resonance imaging and histopathologic examinations of the bone and bone biopsy cultures. The severity of infection was determined via the classification of Infectious Diseases Society of America (IDSA; eTable 1).^9,12 All patients were provided information about the study, and consent was obtained. Three patients did not consent to the study; thus, they were not included (Figure).

Patient demographics, neuropathy, severity of infection, treatment success, side effects, treatment success to pathogen, and treatment period were recorded. If needed, consults from infectious diseases and clinical microbiology specialists assessed the antibiotic treatment and these data were recorded daily. All related data were collected observationally. 

Adverse events were collected prospectively and divided into 2 groups according to continuation of the treatment.

Minor criteria of the evaluation of antibiotic treatment:

• Acute-phase response. The response is defined by leukocytosis count > 11 000/mm³, neutrophilia > 70%, C-reactive protein (CRP) > 1 mg/dL, or erythrocyte sedimentation rate > 20 mm/hour as well as providing an improvement in at least 1 of these laboratory findings during the antibiotic treatment. Deterioration of at least 1 of these laboratory findings is considered an aggravating factor. 
• Fever response. Patients with a body temperature > 38ºC before treatment should have a highest body temperature of 37ºC within 72 hours after initiation of treatment. Hypothermia is a worsening criterion of fever response.
• Microbiological response. Microbiological response is defined by eradication of the microorganism in the control cultures isolated from deep tissue, bone, abscess, and blood cultures. Addition of new microorganisms or persistence of the same bacteria in patients is considered deterioration.
• Wound response. Wound response is defined as any reduction in wound size (length, width, depth) or decrease of purulent discharge, bad odor, edema, redness, and warmth of the wound or wound area with antibiotherapy. 
• Sepsis response. Regression of septic condition in a patient who has sepsis, severe sepsis, or septic shock is defined as sepsis response. 
• The main evaluation criteria of antibiotic treatment for DFIs is used to assess the success of the treatment (eTable 2). 

Statistical analysis in algorithm
In this study, the authors planned to investigate the independent variables affecting the results of tigecycline success with their coefficient effects. The average standard deviation (SD) was used for qualitative data, and the number with percentage was used for categorical data. After separation of parametric and nonparametric data, the differences of average and SD between groups were investigated using Student’s t test for parametric data and Mann-Whitney U test for nonparametric data. Cross tables were analyzed with chi-square, and Fisher’s exact test was used for the discussion of categorical data. Pearson and Spearman tests were performed for correlation analysis between data. One of the highest significance levels of the findings identified a high correlation, and data derived from the categorized qualitative variables were used in the next statistical analyses. Univariate analysis was performed in the next phase and stepwise logistic regression analysis in the final. Significant data were saved as the independent risk factor. Odds ratio for the determined independent risk factors and 95% confidence interval and P values were calculated; P < .05 was used for statistically significant difference in all tests.

Of 105 cases, 37 (35.2%) were women (mean [± SD] age, 61.9 ± 11.9 years). The duration of tigecycline treatment was 16 ± 11.4 days. Tigecycline was the first-line therapy in 69 (65.7%) cases, second-line in 32 (30.5%), and third-line in 4 (3.8%). The mean ± SD for total antibiotic duration was 38.2 ± 16.9 days. Demographics and general properties of the patients are presented in eTable 3. At the end of treatment, 65 (61.9%) cases were successfully treated with tigecycline; tigecycline success rates are presented in eTable 4. 

The correlation and the type of organism could not be statistically analyzed due to the small number of cases (eTable 5). There were no correlations between success rates and wound locations (eTable 6). 

Five variables had significant differences between groups of tigecycline success and failure: arterial stenosis, moderate or severe infection, finger deformity, dorsal foot ulcer, and altered mental status (eTable 5). At the end of univariate analysis, 4 variables had significant differences (arterial stenosis, moderate or severe infection, finger deformity, and altered mental status [eTable 5]). There were 2 independent risk factors for tigecycline success. In the presence of arterial stenosis, the success of tigecycline was reduced by about 6.4 times. Tigecycline success rates were reduced by about 9 times more than in patients with moderate or severe infections (eTable 7). 

Tigecycline is an antibiotic used in skin and soft-tissue infections. One type of complicated skin and soft-tissue infections is a DFI; however, the choice of antibiotics in the treatment of DFIs is quite different from the standard skin and soft-tissue infections.⁹ This study aims to evaluate the use of tigecycline in DFIs. In this scope of study, side effects were encountered in almost half of the tigecycline-treated patients; thus, tigecycline treatment was discontinued in nearly 30% of those cases. Nausea and vomiting were the most common side effects. There were 5 variables associated with the failure of tigecycline treatment: arterial stenosis, moderate or severe infections, finger deformity, dorsal localized foot ulcer, and altered mental status. Independent risk factors leading to the failure of tigecycline treatment included the presence of a moderate or severe infection (lowering the success by 9 times) and arterial stenosis (lowering the success by 6.4 times).

A phase 3 study¹⁰ comparing tigecycline and ertapenem in patients with DFIs showed an overall success rate of tigecycline between 71.4% and 77.9%. The success of tigecycline treatment in patients with osteomyelitis was reported to be 36%.¹⁰ In the present study, the overall success of tigecycline was about 62%; however, the presence of osteomyelitis did not affect tigecycline success.

In addition, the phase 3 study¹⁰ of patients using the tigecycline treatment could not be continued due to the development of nausea in about 40% of patients; therefore, it has been identified that tigecycline does not meet criteria for noninferiority. Nausea and vomiting rates were found to be similar in the present study. However, the discontinuation rates due to nausea and vomiting were much lower as the present patients were able to tolerate more. Keeping infusion times longer helped to alleviate tigecycline-related nausea and vomiting. 

In a meta-analysis conducted on the treatment of tigecycline¹³ (considering high mortality and low clinical and microbiological success), behaviors of the clinicians in the use of tigecycline were evaluated. It was emphasized that in serious infections, physicians prefer tigecycline in combination with other drugs to monotherapy. Similar to the comments in the meta-analysis,¹³ the findings obtained in the study herein showed a decrease in the success of tigecycline treatment for moderate and severe DFIs.

It is more beneficial to follow patients with DFIs with multidisciplinary diabetic foot councils. Patients with diabetic foot ulcers are advised to seek medical care at tertiary care hospitals to prevent poor prognosis.¹⁴ Tertiary care health centers (ie, Ege University Faculty of Medicine) have diabetic foot councils so the symptoms and findings are followed efficiently, and even difficult therapies (tigecycline) can be utilized.

In terms of examining the use of tigecycline in patients with a DFI with osteomyelitis, this study was not completely conclusive because the surgical interventions could not be standardized between comparison groups. Thus, the effect of osteomyelitis on the success of tigecycline treatment could not be reported. The number of patients receiving tigecycline treatment with known bacterial factor was low. Another limitation was the lack of an assessment regarding relapse and reoccurrence. 

Furthermore, the success of tigecycline appeared to decrease with the increasing severity of infection defined by IDSA classification. The presence of arterial stenosis was another independent significant reason for failure. If tigecycline is to be a treatment choice for patients with DFIs, extra attention must be especially paid to cases with side effects, as it is generally not necessary to discontinue medication in most cases. Likewise, the most common side effects (nausea and vomiting) can be handled by extending intravenous infusion duration of tigecycline. This scenario should be followed carefully.

Affiliations: Department of Infectious Diseases and Clinical Microbiology, Ege University, Izmir, Turkey; Buca Seyfi Demirsoy State Hospital, Izmir, Turkey; Department of Endocrinology and Metabolism, Ege University; Department of Orthopedics and Traumatology, Ege University; Department of Dermatology, Ege University; and Department of Internal Medicine, Ege University

Correspondence:
Serhat Uysal, MD
Özmen Street, No. 145
Buca Seyfi Demirsoy State Hospital
35200 Buca/Izmir
Turkey 
drserhatuysal@gmail.com

Disclosure: The authors disclose no financial or other conflicts of interest.