SLR - September 2014 - James McKee
Reference: Chou, HW, Wang, JL, Chang, CH et al. Risk of Severe Dysglycemia Among Diabetic Patients Receiving Levofloxacin, Ciprofloxacin, or Moxifloxacin in Taiwan. Clinical Infectious Disease. 2013; 57 (7): 971-980Scientific Literature Review
Reviewed By: James McKee, DPM
Residency Program: VA Puget Sound Health Care System, Seattle Washington
Podiatric Relevance: Fluoroquinolones are commonly used to treat gram-negative infections in podiatric practice. They are also an attractive antibiotic choice in our patient population, secondary to their QD or BID dosing, which can increase patient compliance with antibiotic usage. The article is significant to podiatric physicians, in that it details the potential for increased dysglycemic (hyperglycemic or hypoglycemic) reactions in the diabetic patient we treat with these commonly used antibiotic medications.
Methods: The authors conducted a population-based inception cohort study of diabetic patients that had been given new prescriptions of either levofloxacin, ciprofloxacin, moxifloxacin, second-generation cephalosporin’s (cefuroxime, cefaclor, cefprozil), or macrolides (azithromycin, clarithromycin) orally. Macrolides were selected as the reference group in this study, because they had not previously been reported to be involved in dysglycemia events. The study aimed to evaluate dysglycemic events while taking the medications utilizing visit to emergency department or hospitalization for dysglycemia within 30 days of initiation of antibiotic therapy. Study data was obtained from the insurance claims made to the Taiwan National Health Insurance, which currently provides coverage for 99.48 percent of the population and includes the records for patient care covering outpatient visits, hospital admissions, prescriptions, diagnosis and procedures performed for all beneficiaries. Exclusion criteria included patients that were administered any of the study medications within six months of the index date, were prescribed refills for chronic illness, were treated with intravenous or intramuscular injections routes, or were treated with multiple study medications on the index date. Dysglycemia was defined as patients that had emergency department visits or were hospitalized within 30 days of administration of a study medication AND had an ICD-9 code of 250.1-250.3 (hyperglycemia) that cross-referenced with prescriptions of insulin or had an ICD-9 code of 251.0 – 251.2 (hypoglycemia) that cross-referenced to prescriptions for glucose water. ICD-9 code 250.8 was not included, unless a co-diagnosis of osteomyelitis or other non-macrolide treated condition. Patients admitted to the study could only be admitted once.
Results: The study population included 78,433 patients. Of the dysglycemic events that occurred in patients enrolled in the study, 215 were hyperglycemic and 425 were hypoglycemic. Patients that used fluoroquinolones had the highest number of dysglycemic events, compared to macrolides. Adjusted Odds Ratio (AOR) for hyperglycemia were 2.48 for moxifloxacin, 1.75 for levofloxacin, and 1.87 for ciprofloxacin. AOR for hypoglycemia were 2.13 for moxifloxacin, 1.79 for levofloxacin, and 1.45 for ciprofloxacin. This was in comparison to there being no significant association between cephalosporin’s and dysglycemic events. In addition, a significant increase in hypoglycemia was also observed in patients taking moxifloxacin with insulin, with an AOR 2.28.
Conclusions: Fluoroquinolones are a go-to antibiotic for many providers when gram-negative infection is suspected or known in the diabetic population. Moxifloxacin is a commonly prescribed fluoroquinolone as it has gram-negative coverage, and also because of its once daily dosing allowing for ease of use and higher patient compliance. Prescription of fluoroquinolones, and moxifloxacin specifically, must be considered carefully due to the possibility of dysglycemia.