Fluoroquinolones

Withdrawal of fluoroquinolones

Owing to adverse effects and reactions (including severe anaphylaxis, QT interval prolongation, and potential cardiotoxicity), several fluoroquinolones have had to be withdrawn (for example, temafloxacin and grepafloxacin) or strictly limited in their uses (for example trovafloxacin) after marketing. A serious idiosyncratic reaction profile is possibly related to the immunologically reactive 1-difluorophenyl substituent that characterizes temafloxacin, tosufloxacin, and trovafloxacin [ ].

The withdrawal of temafloxacin in 1992, only 6 months after its introduction, followed the observation of serious adverse events that were labeled the “temafloxacin syndrome” [ ]. Adverse effects, including hemolysis, renal dysfunction, coagulopathy, and hepatic dysfunction, were estimated to occur in one in 3500 patients. For comparison, incidence rates for these adverse events were about one in 17 000 patients treated with ciprofloxacin and one in 33 000 patients treated with ofloxacin.

In 1999 trovafloxacin was withdrawn after reports of lethal hepatic damage. Before this development, several studies had been published describing impressive clinical efficacy coupled with an almost flawless safety profile [ , ].

Grepafloxacin was withdrawn from the market in 1999 because of its adverse cardiovascular events, which included dysrhythmias [ ].

Clinafloxacin has been withdrawn because of phototoxicity and hypoglycemia and sparfloxacin because of phototoxicity [ ].

These experiences demonstrate the need to be vigilant about any untoward events associated with large-scale use of new fluoroquinolones.

General adverse effects and adverse reactions

In spite of widely publicized negative experiences, most fluoroquinolones are safe, and adverse events are, compared with other groups of antimicrobial agents, relatively rare [ ]. The rates of overall adverse events were similar in several comparisons of individual fluoroquinolones among themselves or with other antimicrobial agents. In one multicenter, double-blind, randomized study drug-related adverse events were reported by 8.9% of those taking levofloxacin and 8.2% of those taking ciprofloxacin for 7–10 days [ ]. A meta-analysis of data from 20 phase II and III studies in 4926 patients treated with moxifloxacin showed that adverse events led to withdrawal of treatment in 3.8% of patients [ ]. In a comparison of grepafloxacin and ciprofloxacin for exacerbations of chronic bronchitis withdrawal was precipitated by adverse events in 3% of 624 patients; there was no difference between ciprofloxacin and grepafloxacin [ ]. The most frequent causes of withdrawal were nausea, vomiting, dysgeusia, dizziness, and diarrhea. In smaller studies, adverse events are occasionally observed more often. In a prospective, double-blind comparison of ciprofloxacin, ofloxacin, and co-trimoxazole for short-course treatment of acute urinary tract infections, there were drug-related adverse events in 26% of patients treated with ciprofloxacin and 34% of patients treated with ofloxacin [ ].

If analysis of the frequency of adverse events is based on prescription event monitoring (PEM), the rates of adverse events are usually substantially lower. In one study, over 11 000 patients taking each antibiotic were monitored [ ]. Among the fluoroquinolones, ciprofloxacin, norfloxacin, and ofloxacin were used. Adverse events resulted in withdrawal of norfloxacin or ofloxacin in under 1% of patients [ ].

The main systems affected by adverse reactions to the fluoroquinolones are the skin, liver, and nervous system. The best-known adverse reaction is phototoxicity, the risk of which varies markedly among the fluoroquinolones; lomefloxacin and sparfloxacin carry a particularly high risk. The development of phototoxicity is based on an interaction between light and the drug. Neurotoxicity also occurs, with marked variation of incidence between the various compounds. Hypersensitivity reactions to fluoroquinolones are rare, and include anaphylactic shock and non-IgE-related anaphylactic (anaphylactoid) reactions. Organ-specific reactions attributed to hypersensitivity involve the liver and kidneys. If hypersensitivity reactions occur, switching from one quinolone compound to another is probably not advisable, since there is cross-reactivity.

The risk of neoplastic disease is minimal, even during long-term use, but may be increased by exposure to UVA light.

Using electron spin resonance spectroscopy and spin trapping, ciprofloxacin has been shown to cause free radical production in a dose- and time-dependent manner; the authors suggested that this effect may contribute to drug-related adverse effects, including phototoxicity and cartilage defects [ ].

The most common drug interactions include malabsorption interactions associated with multivalent cations and CYP450 interactions [ ].

Pharmacoeconomics

The pharmacoeconomic impact of adverse reactions to antimicrobial drugs is enormous. Antibacterial drug reactions account for about 25% of adverse drug reactions. The adverse reactions profile of an antimicrobial agent can contribute significantly to its overall direct costs (monitoring costs, prolonged hospitalization due to complications or treatment failures) and indirect costs (quality of life, loss of productivity, time spent by families and patients receiving medical care). In one study an adverse event in a hospitalized patient was associated on average with an excess of 1.9 days in the length of stay, extra costs of $US2262 (1990–93 values), and an almost two-fold increase in the risk of death. In the outpatient setting, adverse drug reactions result in 2–6% of hospitalizations, and most of them were thought to be avoidable if appropriate interventions had been taken. In a review, economic aspects of antibacterial therapy with fluoroquinolones have been summarized and critically evaluated [ ].

Comparative studies

In a crossover, randomized study, 16 fasted volunteers (eight men, eight women) took single oral doses of gemifloxacin 320 mg and ofloxacin 400 mg on two separate occasions, in order to assess urinary excretion [ ]. Urine concentrations of ofloxacin were higher than those of gemifloxacin. There were no adverse reactions.

Fluoroquinolones have been compared retrospectively in 11 controlled trials in acute cystitis in 7535 women [ ]. Photosensensitivity reactions were more common with sparfloxacin than ofloxacin. Adverse cutaneous events and photosensitivity causing withdrawal were more common with lomefloxacin than ofloxacin. Nervous system adverse events and insomnia were reported more frequently with rufloxacin and enoxacin than with pefloxacin or ciprofloxacin.

Systematic reviews

Fluoroquinolones and beta-lactams have been compared in a meta-analysis of 11 randomized controlled trials [ ]. Adverse events did not differ significantly between the groups; nausea and diarrhea were the major adverse events.

Cardiovascular

Some fluoroquinolones can prolong the QT interval, with a risk of cardiac dysrhythmias. In an in vitro study in isolated canine cardiac Purkinje fibers the rank order of potency in prolonging action potential duration was sparfloxacin > grepafloxacin = moxifloxacin > ciprofloxacin [ ]. In guinea-pig ventricular myocardium sparfloxacin prolonged the action potential duration by about 8% at 10 μmol/l and 41% at 100 μmol/l [ ]. Gatifloxacin, grepafloxacin, and moxifloxacin were less potent, but prolonged the action potential duration at 100 μmol/l by about 13%, 24%, and 25% respectively. In contrast, ciprofloxacin, gemifloxacin, levofloxacin, sitafloxacin, tosufloxacin, and trovafloxacin had little or no effect on the action potential at concentrations as high as 100 μmol/l.

Preclinical and clinical trial data and data from phase IV studies have shown that levofloxacin, moxifloxacin, and gatifloxacin cause prolongation of the QT interval, but that the potential for torsade de pointes is rare and is influenced by several independent variables (for example concurrent administration of class Ia and III antidysrhythmic agents) [ ]. There is a moderate increase in the QT interval associated with sparfloxacin, averaging 3%, and the few serious adverse cardiovascular events that have been reported during postmarketing surveillance all occurred in patients with underlying heart disease [ ].

In patients taking fluoroquinolones (ciprofloxacin 11 477, enoxacin 2790, ofloxacin 11 033, and norfloxacin 11 110; mean ages 49–57 years) there was no evidence of drug-induced dysrhythmias associated with enoxacin within 42 days of drug administration [ ]. Of the other fluoroquinolones, atrial fibrillation was reported most often within 42 days of ciprofloxacin administration, with no change in event rate over that time. The crude rate of palpitation did not change significantly with ciprofloxacin, norfloxacin, or ofloxacin. Syncope and tachycardia were also reported with ciprofloxacin and ofloxacin. There was no evidence of drug-induced hepatic dysfunction within 42 days of drug administration with any of the fluoroquinolones used.

In a retrospective database analysis 25 cases of torsade de pointes associated with ciprofloxacin (n = 2), ofloxacin (n = 2), levofloxacin (n = 13), and gatifloxacin (n = 8) were identified in the USA [ ]. Ciprofloxacin was associated with a significantly lower rate of torsade de pointes (0.3 cases per 10 million prescriptions) than levofloxacin (5.4 per 10 million) or gatifloxacin (27 per 10 million). When the analysis was limited to the first 16 months after initial approval of the drug, the rates for levofloxacin (16 per 10 million) and gatifloxacin (27 per 10 million) were similar.

Dysrhythmias associated with fluoroquinolones therapy have been reviewed. Moxifloxacin was associated with the greatest risk of QT interval prolongation, with lower risks for gemifloxacin, levofloxacin, and ofloxacin; the lowest risk was attributed to ciprofloxacin [ ].

Two cases of QT interval prolongation with torsade de pointes have been reported.

• A 16-year-old boy, who had received intravenous ciprofloxacin 400 mg bd and metronidazole 500 mg 6-hourly for acute Crohn’s disease, developed reversible QT interval prolongation to 486 ms and a bradycardia [ ].

• A 66-year-old man with an implanted cardioverter-defibrillator had QT interval prolongation and multiple episodes of torsade de pointes when ciprofloxacin and azimilide were co-administrated [ ].

Nervous system

The main central nervous system adverse reactions of the fluoroquinolones include dizziness, convulsions, psychosis, and insomnia; levofloxacin, moxifloxacin, and ofloxacin are reportedly least likely to cause these reactions, based on a study of European and international data from about 130 million prescriptions [ ].

A review has suggested that fluoroquinolone-associated peripheral nervous system events are mild and short-term [ ]. Among 60 courses of fluoroquinolones in 45 patients (levofloxacin 33 courses, ciprofloxacin 11 courses, ofloxacin six courses, lomefloxacin one course, trovafloxacin one course; in eight cases the same antibiotic was prescribed twice) there were 36 severe events that typically involved multiple organ systems. The symptoms lasted more than 3 months in 71% of cases and more than 1 year in 58%. The onset of the adverse events in the 45 patients was usually rapid: 15 events began within 24 hours of the start of treatment, 26 within 72 hours, and 38 within 1 week.

Dizziness is not rare in patients taking fluoroquinolones, and was observed in 2.8% of patients taking moxifloxacin [ ], in 8% and 9% of patients taking grepafloxacin 400 mg/day and 600 mg/day, and in 6% of patients taking ciprofloxacin [ ]. Prescription event monitoring found markedly lower rates of this adverse event during treatment with ciprofloxacin, norfloxacin, or ofloxacin [ ].

Headache was recorded in 8% of patients taking ciprofloxacin and 9% taking ofloxacin during short-course treatment of urinary tract infections [ ]. Similar rates are reported for grepafloxacin [ ].

In two patients with myasthenia gravis, ciprofloxacin and norfloxacin exacerbated the symptoms [ ].

Encephalopathy with unconsciousness has been reported in a 48-year-old woman with Machado–Joseph disease after the administration of fleroxacin (200 mg/day) for 3 days [ ]. She recovered after withdrawal. The serum and cerebrospinal fluid concentrations of fleroxacin were within normal limits.

In a review of drug-induced mania, fluoroquinolones were listed among the drugs that are most frequently implicated [ ].

Sensory systems

Psychiatric

Isolated cases of depression [ ] and psychosis have been described in temporal association with fluoroquinolones [ ]. In a retrospective study the data on fluoroquinolones and other antibacterial drugs, rufloxacin was associated with a reporting rate of 221 reports per daily defined dose per 1000 inhabitants per day, and the most frequent were psychiatric disorders [ ].

Metabolism

There was a higher rate of hyperglycemia with gatifloxacin or levofloxacin compared with ceftriaxone in a retrospective chart review of 17 000 patients [ ]. Sulfonylurea therapy was identified as an independent risk factor for hypoglycemia.

Hematologic

In over 33 000 patients treated with ciprofloxacin, norfloxacin, or ofloxacin, no case of hemolytic anemia was discovered by prescription event monitoring [ ]. However, ciprofloxacin has been associated with hemolysis in combination with a severe skin reaction in a young adult [ ].

Leukopenia has been observed in 0.1–0.7% of patients given fluoroquinolones in Japan and eosinophilia was observed in 0.5–2.2% of these patients [ ]. Leukopenia was generally mild and reversible after dosage reduction or withdrawal [ , ].

Anemia, thrombocytopenia, and thrombocytosis have only rarely been reported [ ].

Gastrointestinal

Adverse gastrointestinal events are not uncommon during treatment with fluoroquinolones. There may be some dose-dependency, since with grepafloxacin 600 mg/day the rates of the following adverse events were noticeably higher than with 400 mg/day: nausea (15% versus 11%), vomiting (6% versus 1%), and diarrhea (4% versus 3%) [ ]. In a randomized, double-blind comparison of prulifloxacin 600 mg/day and ciprofloxacin 500 mg bd in 235 patients with acute exacerbations of chronic bronchitis, the most common treatment-related adverse event was gastric pain of mild or moderate intensity, reported in 8.5% of the patients taking prulifloxacin and 6.8% of those taking ciprofloxacin [ ].

Based on a literature review it has been concluded that fluoroquinolones seems to predispose patients to diarrhea associated with Clostridium difficile [ ].

Liver

Transient rises in serum aminotransferase and serum alkaline phosphatase activities have been observed with all fluoroquinolones. This occurred in 0.9–4.3% of patients in Japan. In the vast majority of the cases this alteration was self-limited and reversible without withdrawal of the drug [ ].