Fluoroquinolone antibiotics as a class have the potential to cause various adverse effects. Limited evidence, including animal data, case reports, and one small postmarketing trial, suggests that gatifloxacin (Tequin) can cause hypoglycemia and hyperglycemia, although the proposed mechanism of action is not well understood.

Park-Wyllie and colleagues conducted two retrospective, case-control studies of patients who were prescribed broad-spectrum antibiotics to determine whether antibiotic use was associated with emergency care or hospitalization for hyperglycemia or hypoglycemia. They reviewed data from hospitalization and pharmacy claims databases for 1.4 million adults who were 66 years or older in Ontario, Canada.

Case patients were those who received emergency or hospital care for hyperglycemia or hypoglycemia within 30 days of a prescription for a macrolide, an oral second-generation cephalosporin, or a fluoroquinolone (ciprofloxacin [Cipro], gatifloxacin, levofloxacin [Levaquin], or moxifloxacin [Avelox]). Control patients were those who were prescribed these antibiotics during the same period but who did not receive hospital care for hyperglycemia or hypoglycemia. Case and control patients were matched by age, sex, and diabetes status.

Hypoglycemic events (n = 788) occurred primarily in patients with existing diabetes (91.9 percent). The average age of these patients was 78, and more than one half (53.6 percent) were released after evaluation in the emergency department. The median time from antibiotic prescription to presentation was six days. Among those admitted to the hospital, the median length of stay was seven days, and 8.1 percent died. Patients with hypoglycemia were four times more likely to have received gatifloxacin than macrolides (odds ratio [OR] = 4.3). A less prominent effect was found with levofloxacin (OR = 1.5). No increased risk was found with moxifloxacin or ciprofloxacin.

Hyperglycemic events were less common (n = 470). The average age of patients with hyperglycemia was 77. More than one third of patients treated for hyperglycemia had not been treated for diabetes in the previous 180 days. Median time from antibiotic prescription to presentation was five days. One half of the patients were treated and released from the emergency department. Among those admitted, the median length of stay was nine days, and 16.5 percent died. There was a 17-fold likelihood of recent treatment with gatifloxacin compared with macrolides in the hospitalized patients with hyperglycemia (OR = 16.7). No other fluoroquinolones or second-generation cephalosporins were associated with an increased risk of hyperglycemia.

In total, there were 16,697 gatifloxacin prescriptions, which were associated with 178 hospital visits (1.1 percent). This rate was higher than the rates for other fluoroquinolones studied, including those for ciprofloxacin (0.3 percent), levofloxacin (0.3 percent), and moxifloxacin (0.2 percent). Comparison antibiotic classes also had lower rates: 0.2 percent for second-generation cephalosporins and 0.1 percent for macrolides.

The authors conclude that outpatient gatifloxacin treatment in older adults presents an increased risk of hospitalization for dysglycemia compared with other broad-spectrum antibiotics.