Editorial

Donald Low, MD, FRCPC, Microbiologist-in-Chief, Department of Microbiology, Mount Sinai Hospital; Professor, Department of Laboratory Medicine and Pathobiology, University of Toronto, Canada

The emergence of antimicrobial resistance to penicillin, tetracycline, trimethoprim/sulfamethoxazole, and the macrolide antibiotics in respiratory pathogens, especially Streptococcus pneumoniae, has raised concerns regarding the use of these agents for the empiric treatment of respiratory tract infections.¹ As a result, fluoroquinolones (such as levofloxacin, gatifloxacin, moxifloxacin, and gemifloxacin), with exquisite activity against Haemophilus influenzae, Moraxella catarrhalis, Mycoplasma pneumoniae, Chlamydia pneumoniae, and Legionella species and with increased activity against S. pneumoniae, have been developed and recommended for the treatment of patients with community-acquired pneumonia (CAP), acute exacerbations of chronic bronchitis (AECB), and acute bacterial sinusitis.¹ Abstracts and oral presentations from experts from around the world presented at the 7th International Symposium on New Quinolones reconfirmed the safety, cost-effectiveness, and usefulness of this class of antibiotics. Of particular importance was the evidence presented for the evolving role of the newer respiratory quinolones for the treatment of respiratory tract infections, especially lower-tract infections. There is mounting evidence that the fluoroquinolones may provide enhanced efficacy over other nonfluoroquinolone antimicrobials. When levofloxacin was compared to the cephalosporins for the treatment of CAP, clinical success was higher in the levofloxacin group (96%) than in the patients who received ceftriaxone/cefuroxime (90%). Gemifloxacin was compared to clarithromycin for the treatment of severe AECB (the GLOBE study). By day 6 of the study, no patient treated with gemifloxacin had a positive sputum culture for Haemophilus influenzae, but one third of clarithromycin patients did. In addition, the researchers concluded that 5 days of gemifloxacin treatment for AECB compared with 7 days of clarithromycin resulted in significantly more patients remaining free of AECB recurrence at long-term follow-up. Additional secondary benefits included fewer hospitalizations, fewer hospital days, and less time lost from work and other patient activities.

One area of special concern raised at the meeting was the potential for the development and dissemination of fluoroquinolone resistance in S. pneumoniae. Therefore, the use of a fluoroquinolone with marginal activity against S. pneumoniae, i.e., a concentration of a fluoroquinolone at the site of infection that is not adequate to kill first- and second-step mutants, will allow these isolates to survive, multiply, replace the previously susceptible flora, and in some instances, result in clinical failure.²

Data were presented demonstrating that fluoroquinolone resistance has in fact been documented in several countries; albeit, at extremely low levels. Efforts to control the evolution and dissemination of such resistance in S. pneumoniae includes discouraging inappropriate use of antibiotics and, when warranted, using appropriate doses of highly active antimicrobials.

This Forum Report presents exciting new information about this continuously evolving class of antimicrobials, especially their growing role in the treatment of respiratory tract infections.

The Development of Fluoroquinolones

Worldwide, increasing antibiotic resistance to conventional antibiotic agents puts patients at risk of ineffective treatment. An exciting focus of antibiotic research is the quinolone group, now in its fourth generation of development. Nalidixic acid (NegGram, Sanofi-Synthelabo U.S.), the first antibiotic from the quinolone group, became available in 1962. Because of its excellent activity against gram-negative urinary tract pathogens, but poor serum and tissue concentrations, its use has been primarily limited to the treatment of urinary tract infections.

Quinolones have a dual-ring chemical structure with either a quinolone or a naphthyridone nucleus. The second generation of quinolones added a fluorine atom to the C-6 position and a piperazine group to the C-7 position, which markedly improved potency and conferred advantageous pharmacokinetics. These advances produced drugs such as norfloxacin (Noroxin, Merck), lomefloxacin (Maxaquin, VistaPharm), enoxacin (Penetrex, Aventis), ofloxacin (Floxin, Ortho-McNeil), and ciprofloxacin (Cipro, Bayer). Introduced in 1987, ciprofloxacin has systemic bioavailability effective against gram-negative pathogens (including Pseudomonas aeruginosa) and some gram-positive pathogens. This expanded spectrum of activity allowed treatment of not only urinary tract infections (e.g., pyelonephritis, prostatitis), but also systemic infections (e.g., sexually transmitted diseases, skin and soft-tissue infections). Ciprofloxacin's good bone penetration makes it useful for treating osteomyelitis. Because of its high intracellular penetration and excellent oral bioavailability, it can be a substitute for intravenously administered antibiotics in many patients.

Peter Ball, MD, FRCP, Consulting Physician at St. Andrews Hospital and Chairman of the 7th International Symposium on New Quinolones (Edinburgh, June 10-12, 2001), observed "what a huge difference the [second-generation] quinolones made in 1985 when they came along." He noted that they had substantially increased potency, novel modes of action and resistance mechanisms, good tissue penetration, pharmacodynamics suitable for severe gram-negative rod infections, flexible dosing regimens (intravenous and oral), intravenous/oral switch potential, and an excellent adverse reaction profile.³ The success of these agents is exemplified by ciprofloxacin, whose broad spectrum of activity and availability in oral, intravenous, and ophthalmic routes of administration have contributed to making it one of the most frequently prescribed antibiotics in the world.⁴

Chemical modifications and structural alterations have produced third-generation quinolones with different antimicrobial activities and side-effect profiles. These "respiratory quinolones" such as levofloxacin (Levaquin, Ortho-McNeil), sparfloxacin (Zagam, Mylan), gatifloxacin (Tequin, Bristol-Myers Squibb), and moxifloxacin (Avelox, Bayer) are widely used for AECB and CAP because of their expanded coverage against gram-positive bacteria (including penicillin-resistant S. pneumoniae and Staphylococcus aureus) and atypical pathogens (such as Chlamydia, Mycoplasma, and Legionella species) and some activity against tuberculous and nontuberculous mycobacteria.

A fluoroquinolone in late-stage clinical development is gemifloxacin (Factive, GlaxoSmithKline), an investigational, enhanced-affinity fluoroquinolone. Other fourth-generation quinolones, including D61-1113, PGE-9262932, and T-3811ME/BMS-284756 (a 6-desfluoroquinolone),5 are currently under clinical development in investigational trials.

Does In Vitro Activity Correlate With High Clinical Efficacy?

The fluoroquinolone microbiologic spectrum for respiratory pathogens--in particular, S. pneumoniae--varies by agent. For instance, ciprofloxacin, which has the best gram-negative and antipseudomonal activity, has marginal activity against S. pneumoniae. Ofloxacin has only slightly improved gram-positive coverage compared with ciprofloxacin. Levofloxacin has increased activity against S. pneumoniae compared with ciprofloxacin and ofloxacin. In addition to antipneumococcal activity, levofloxacin, gatifloxacin, moxifloxacin, and trovafloxacin retain gram-negative activity against H. influenzae and Moraxella catarrhalis and have activity against atypical respiratory pathogens, including Mycoplasma pneumoniae, Chlamydia pneumoniae, and Legionella species. Gemifloxacin has enhanced activity against S. pneumoniae⁶ and excellent activity against atypical pathogens. It is the most potent fluoroquinolone against S. pneumoniae and maintains good activity against ciprofloxacin-resistant strains.6 Their demonstrated efficacy against multidrug-resistant S. pneumoniae is the reason the medical community has turned to the fluoroquinolones.

The newer fluoroquinolones exhibit good tissue penetration, particularly in renal, lung, prostate, bronchial, nasal, gall bladder, bile, and genital-tract tissues.^7-9 Long elimination half-lives (gemifloxacin, 8.5 hours; gatifloxacin, 7.8 hours; levofloxacin, 6.3 hours; and moxifloxacin, 10.4 hours) allow for once-a-day dosing. The quinolones vary with respect to the relative contribution of renal and nonrenal pathways for their elimination. Ofloxacin and levofloxacin are exclusively eliminated by the kidney¹⁰ and require dosage adjustments in patients with renal impairment. Renal and nonrenal (gastrointestinal or hepatic) mechanisms are responsible for the elimination of nalidixic acid, cinoxacin, norfloxacin, ciprofloxacin, enoxacin, lomefloxacin, gatifloxacin, moxifloxacin, sparfloxacin, and gemifloxacin.

Researchers compared the activity of gemifloxacin to several other quinolones against 147 isolates of S. pneumoniae with documented reduced sensitivity to ciprofloxacin (minimum inhibitory concentrations [MICs] of ≥4 mg/L).¹¹ The Canadian Bacterial Surveillance Network had identified these samples from 40 different laboratories in eight Canadian provinces from 1996 to 2000.

Gemifloxacin had the lowest MIC₉₀ (concentration inhibiting growth in 90% of organisms) (0.5 mg/L) against ciprofloxacin-resistant S. pneumoniae compared with ciprofloxacin (32 mg/L), levofloxacin (16 mg/L), gatifloxacin (4 mg/L), and moxifloxacin (4 mg/L). Of the 147 clinical isolates, 91 (62%) had a gemifloxacin MIC of ≤0.06 mg/L while 9 (6%) isolates had a moxifloxacin MIC of ≤0.06 mg/L; all remaining MIC values were higher.

Blondeau et al. tested the susceptibility of H. influenzae, Moraxella catarrhalis, and S. pneumoniae isolates to seven fluoroquinolones. The rank order of potency based on MIC₉₀ values was as follows: gemifloxacin (0.031-0.063 mg/L), trovafloxacin (0.125 mg/L), moxifloxacin (0.125-0.25 mg/L), grepafloxacin (0.125-0.25 mg/L), gatifloxacin (0.5 mg/L), levofloxacin (1 mg/L), and ciprofloxacin (2 mg/L).¹² Regardless of methodology, all fluoroquinolones were highly active against the isolates. The clinical importance of these findings becomes more focused when considering pharmacological treatment of CAP due to penicillin-resistant S. pneumoniae, H. influenzae, and Moraxella catarrhalis.

Research during the past decade has identified a variety of pharmacokinetic and pharmacodynamic factors that are major determinants of in vivo antimicrobial activity and may be useful in predicting the development of resistance. Pharmacokinetic/pharmacodynamic parameters correlating with efficacy (bactericidal activity) may be either time- or concentration-dependent. For time-dependent agents (e.g., beta-lactams, macrolides, vancomycin, clindamycin), time above the MIC is critical in determining efficacy. For concentration-dependent agents (e.g., azalides, tetracyclines, aminoglycosides), it is important to maximize the peak drug concentration (Cmax):MIC and area under the curve (AUC):MIC. The fluoroquinolones have concentration-dependent bactericidal activity, and the goal of therapy is to maximize achievable drug concentrations in relation to the MIC of the pathogen. The Cmax:MIC and AUC:MIC ratios correlate with clinical and microbiologic outcomes for the fluoroquinolones.^13-15 Though the fluoroquinolones differ in their pharmacokinetic profiles and in their in vitro potency, the minimal threshold for effective eradication of S. pneumoniae is an AUC:MIC value of approximately 30.^14,16 The AUC:MIC values of levofloxacin, trovafloxacin, gatifloxacin, moxifloxacin, and gemifloxacin against S. pneumoniae all exceed 30; however, the AUC:MIC value for ciprofloxacin does not exceed 30 for S. pneumoniae. Therefore, ciprofloxacin is a poor choice for the treatment of pneumococcal infections.

The pharmacokinetic parameters of the newer fluoroquinolones (e.g., gemifloxacin, gatifloxacin, levofloxacin, moxifloxacin) are improved compared with the original quinolones: the drugs are rapidly and almost completely absorbed from the gastrointestinal tract. Peak serum concentrations after oral administration are very near to those achieved with intravenous administration.¹⁷ These agents can be taken with food but should not be administered concomitantly with medications or supplements that contain multivalent metal cations (aluminum, magnesium, calcium, iron, or zinc) because of decreased absorption. Administering products containing the cations 4 hours before or 2 hours after fluoroquinolone administration can overcome this problem.

Emerging Quinolone Resistance of Streptococcus Pneumoniae

According to Dr. Axel Dalhoff (a scientist at Bayer AG, Wuppertal, Germany), an AUC:MIC ratio of 30-50 is sufficient for most community-acquired infections, whereas in hospitalized patients with gram-negative infections, an AUC:MIC of 125 or greater is a predictor for rapid clearance of infection.

Although fluoroquinolones are highly effective against S. pneumoniae, resistant strains are beginning to emerge (in 1998, 1.7% of S. pneumoniae had reduced susceptibility to ciprofloxacin, a second-generation quinolone).¹⁸ Quinolones target the bacterial enzymes DNA gyrase and topoisomerase IV. These enzymes are critical for bacterial proliferation, because DNA gyrase introduces negative supercoils into DNA, while topoisomerase IV is responsible for the separation of daughter chromosomes after chromosome division. The potent antipneumococcal activity of gemifloxacin is associated with dual targeting of gyrase and topoisomerase IV, an in vivo target preference for gyrase, and enhanced stabilization of cleavable complexes in vitro.¹⁹ Mutations in subunits of DNA gyrase and topoisomerase IV produce quinolone-resistant bacteria.

Community-Acquired Pneumonia and the Problem of Penicillin-Resistant Streptococcus pneumoniae

According to Professor Javier Garau (an infectious disease specialist at the Hospital Mutua de Terrassa, Barcelona, Spain), Streptococcus pneumoniae is the most common identified pathogen in CAP, accounting for 25% of cases. Other reports suggest a range of 20%-60%.²⁰ Viruses are responsible for 12.2%, followed by infections with H. influenzae (7.3%) and Legionella species (5.7%). Dr. Garau pointed out that in the majority of cases (39%), physicians cannot identify the cause. Because of the multiplicity of organisms, he said, "the treatment of CAP remains an empirical exercise, and in this respect, the quinolones have a clear advantage compared to other agents."

In addition, S. pneumoniae causes one third of cases of acute maxillary sinusitis,²¹ 15% of bacterial etiologies of AECB,²² and 9%-55% of hospitalizations for CAP.²³ As many as 50.4% streptococcal isolates in the United States are not susceptible to penicillin.^24,25

Respiratory tract isolates tend to be more resistant than isolates from blood, sputum, or other sites. Penicillin-resistant pneumococci are often less susceptible to other antibiotics such as other beta-lactams, chloramphenicol, macrolides, tetracyclines, and trimethoprim/sulfamethoxazole. Pathogen resistance to penicillin, however, does not affect susceptibility to fluoroquinolones. Consequently, physicians can choose fluoroquinolones as empiric therapy for bacterial respiratory infections when penicillin-resistant pneumococci may challenge conventional treatment.²⁶ Dr. Garau presented the results of a double-blind, randomized, parallel, three-armed trial of sparfloxacin, amoxicillin/clavulanic acid, and erythromycin for CAP that included more than 800 patients.²⁷ One third of the patients were outpatients, and two thirds were hospitalized for more severe disease. The overall success rates for sparfloxacin (87%), amoxicillin/clavulanic acid (80%), and erythromycin (85%) were similar in evaluable patients.

In another double-blind, randomized, parallel study, researchers randomized 264 patients to grepafloxacin versus amoxicillin (Amoxil, GlaxoSmithKline) for a 7- to 10-day course of treatment.²⁸ All of these patients were ambulatory; 207 (78%) completed the study. Clinical response at follow-up was similar in the two groups (grepafloxacin, 76% [87/114]; amoxicillin, 74% [85/111]; 95% confidence interval [CI], -12% to 10%). Microbiological eradication with grepafloxacin was statistically superior to that with amoxicillin in the evaluable population; the success rate was 89% (32/36) in the grepafloxacin group compared with 71% (32/45) in the amoxicillin group (95% CI, 2%-37%).

A third study was an open-label, randomized study of levofloxacin versus ceftriaxone and/or cefuroxime axetil in the treatment of adults with CAP.²⁹ Levofloxacin could be administered either orally (500 mg once a day) or intravenously; ceftriaxone was administered parenterally (1-2 g once or twice a day) with or without cefuroxime axetil administered orally (500 mg twice a day). Fifty-three percent of patients were ambulatory, and 47% required hospitalization. There was 3.4% mortality overall in the study (5% in the ceftriaxone/cefuroxime group vs. 2% in the levofloxacin group). Clinical success at 5 to 7 days after therapy was superior in the levofloxacin group (96%) compared with the ceftriaxone and/or cefuroxime axetil group (90%) (95% CI, -10.7 to -1.3). The overall bacteriologic eradication rates were superior for the levofloxacin group (98%) compared with the ceftriaxone and/or cefuroxime axetil group (85%) (95% CI, -21.6 to -4.8). Levofloxacin eradicated 100% of reported H. influenzae and S. pneumoniae isolates.

Because of gemifloxacin's potent activity against S. pneumoniae, Dr. Hartmut Lode (Professor of Medicine at City Hospital Zehlendorf, Germany, and Chief, Department for Pulmonary and Infectious Diseases) and the Gemifloxacin Study Group undertook an efficacy comparison of oral gemifloxacin versus intravenous ceftriaxone followed by cefuroxime in hospitalized patients with moderate to severe CAP.³⁰ Patients received gemifloxacin 320 mg daily for 7-14 days or intravenous ceftriaxone 2 g for 1-7 days with or without a macrolide, followed by oral cefuroxime 500 mg twice a day for 1-13 days (total treatment time of up to 14 days). The screening visit occurred on day 1-0, visit 2 occurred on day 1-5, visit 3 on day 0-6 after therapy, and visit 4 on day 19-41 after therapy. The subjects provided sputum or other respiratory samples for gram stain, culture, and sensitivity. Aerobic and anaerobic blood cultures were also taken.

Among the eligibility requirements were hospitalization due to a clinical and radiologic episode of CAP and at least two additional signs of disease. Patients were excluded for significant underlying disease, hypersensitivity to treatment agents, prior antimicrobial therapy for this episode of CAP, or pregnancy. Demographics and key pathogens at screening were similar in the two treatment groups. S. pneumoniae was the most common pathogen.

Of the 341 patients who entered the study, 116 completed the study in the gemifloxacin group and 121 in the ceftriaxone/cefuroxime group. Both treatment groups had high clinical success rates (gemifloxacin, 107 of 116 [92.2%]; ceftriaxone/cefuroxime with or without macrolide, 113 of 121 [93.4%]). Both groups also had high bacteriological success (gemifloxacin, 58 of 64 [90.6%]; ceftriaxone/cefuroxime with or without macrolide, 55 of 63 [87.3%]).

Radiologic success and time to discharge from the hospital were also similar in the two groups. The bacteriological success rate at follow-up was 100% (12/12) for S. pneumoniae bacteremic patients in the gemifloxacin group.

The authors concluded that the efficacy of oral gemifloxacin was "at least as good as" the comparator in hospitalized patients with moderate to severe CAP.

Gemifloxacin Versus Clarithromycin in AECB

In the United States, AECB afflicts approximately 14 million people (54 per 1000 population), who require frequent treatment for acute infections. Clinical recurrence of AECB remains a vexing problem. Repeated infections result in damage to delicate lung tissue and progressive reduction of pulmonary function. Because of the recurrent nature of AECB, these patients require frequent health care utilization. In addition, AECB has a significant impact on quality of life.

H. influenzae, Moraxella catarrhalis, and S. pneumoniae are responsible for as many as 95% of cases of bacterial exacerbations.³¹ S. pneumoniae continues to develop increased resistance to penicillin and macrolides, while H. influenzae and Moraxella catarrhalis can also produce beta-lactamase to inactivate penicillins and cephalosporins.

Researchers explored the use of gemifloxacin versus that of clarithromycin (Biaxin, Abbott)³² in patients with AECB. Eligibility criteria for the study included age ≥40 years, a history of chronic bronchitis for at least 2 consecutive years, and an acute bronchitis exacerbation based on purulent sputum, cough, and dyspnea (Anthonisen class 1). Patients with significant complicating medical illnesses, hypersensitivity to gemifloxacin or clarithromycin, previous antibiotic treatment within 1 week of study entry, or predisposal to drug reactions with the study drugs were excluded.

Seven hundred nine patients participated in this multicenter, randomized, double-blind, controlled trial. Patients in the gemifloxacin group (n = 351) received 5 days of oral gemifloxacin 320 mg and 7 days of placebo. Patients in the clarithromycin group (n = 358) received 7 days of oral clarithromycin 500 mg twice a day and 5 days of placebo. The two groups were well-matched regarding age, sex, race, duration of chronic bronchitis, and number of AECBs in the last 12 months.

Patients had a total of 5 visits: screening (day 0), therapy (days 2-4), end of therapy (days 8-12), follow-up (days 13-24), and long-term follow-up (days 25-38). Enrolled patients from the United States and Canada had the option of participating in the Gemifloxacin Long-term Outcomes in Bronchitis Exacerbations (GLOBE) study (see below) to assess long-term clinical and economic outcomes.

Sputum cultures most often grew H. influenzae (gemifloxacin group, 20 of 57 [35.1%]; clarithromycin group, 18 of 66 [27.3%]). The other key pathogens identified were H. parainfluenzae, Moraxella catarrhalis, S. pneumoniae, and Staphylococcus aureus.

After 1 day of treatment, gemifloxacin eradicated H. influenzae from sputum cultures in 7 of 7 patients (100%), clarithromycin in only 3 of 9 (33.3%). By day 6 of the study, no patient treated with gemifloxacin had a positive sputum culture for H. influenzae, but one third of clarithromycin patients did. A bacteriological analysis revealed that gemifloxacin had a significantly faster time to eradication of H. influenzae (p = 0.02).

Clinical success rates were similar in the two groups, 91.3% for gemifloxacin and 91.4% for clarithromycin. Bacteriologic success, defined as eradication (or presumed eradication) of initial pathogens without superinfection, new infection, or recurrence, was also similar but showed a trend favoring gemifloxacin over clarithromycin, 93.6% vs. 81.5%, respectively (95% CI -0.4 to 24.6).

At long-term follow-up (days 25-38), clinical success rates had decreased but remained similar in the two groups, 79.6% for gemifloxacin vs. 78.5% for clarithromycin. However, gemifloxacin demonstrated a clear statistical superiority for bacteriologic success, 81.8% vs. 62% (95% CI 2.2-37.5, p < 0.05).

The most common adverse events were diarrhea and headache. Diarrhea occurred in 23 gemifloxacin patients (6.6%) and 30 clarithromycin patients (8.4%). In the gemifloxacin group, 22 (6.3%) complained of headache compared with 31 (8.7%) in the clarithromycin group. Only clarithromycin patients reported taste perversion (18 [5%], p < 0.01).

The authors concluded that a 5-day course of gemifloxacin was well tolerated and a "highly effective therapy for AECB," with faster and more complete bacteriologic eradication of H. influenzae than 7 days of twice-a-day clarithromycin treatment.

Long-Term Clinical Outcomes and Cost-Effectiveness of AECB Therapy—The GLOBE Study

As a follow-up to the gemifloxacin versus clarithromycin acute study (the 068 Clinical Study Group), GLOBE compared the long-term clinical benefits and cost-effectiveness of gemifloxacin versus oral clarithromycin in the treatment of AECB in a 26-week, double-blind, prospective, multicenter, parallel-group study of 438 patients (United States, 386; Canada, 52).³³ There were 214 patients (108 males, mean age 58.5 years) in the gemifloxacin group and 224 patients (123 males, mean age 57.6) in the clarithromycin group. Smoking and history of steroid use were similar in the two groups.

At 26 weeks of follow-up, significantly more gemifloxacin patients than clarithromycin patients remained recurrence free (120 of 169 [71%] vs. 100 of 171 [58.5%]; 95% CI 2.5-22.6, p = 0.016). Furthermore, the number of patients hospitalized for RTI-related events over the study period was less in the gemifloxacin group (5 of 214 [2.34%] vs. 14 of 224 [6.25%]; p = 0.059).

The significant reduction of AECB recurrence by gemifloxacin at long-term follow-up and evidence of fewer patient hospitalizations demonstrate the potential cost-effectiveness of fluoroquinolone (e.g., gemifloxacin) therapy. Current pharmacoeconomic analysis considers both the overall costs and clinical outcomes involved in pharmacologic treatment.³⁴ Determination of overall costs considers not only the initial cost of the treatment but also intermediate costs, such as the length of hospitalization, laboratory tests, respiratory therapy, monitoring and nursing administration costs, costs due to adverse effects and drug interactions, and cost of treatment failures, which often require additional medical consultation. Although the fluoroquinolones often have higher acquisition costs, their often shorter duration of therapy, oral/intravenous bioequivalence, and clinical efficacy as demonstrated by the outcomes in the GLOBE study suggest a favorable economic profile compared with alternative oral/parenteral treatments.

Smokers and Ex-Smokers Do Better With Gemifloxacin

Most of the subjects in the GLOBE study were smokers or ex-smokers (364 of 438 [83%]).³⁵ These subjects took the St. George's Respiratory Questionnaire (SGRQ) several times during the study to assess their health status (baseline and at 4, 12, and 26 weeks). The SGRQ has three component scores (symptoms, activity, and impacts) and a total score. A difference of 4 points in the total score is clinically meaningful (lower scores are better).

Results were reported for patients who completed the entire 26-week study. Gemifloxacin resulted in a significant decrease in symptom score (47.6, from baseline 73.9), as did clarithromycin (54.5, from baseline 74.5); gemifloxacin's decrease was significantly higher that that of clarithromycin (p = 0.03). Activity, impacts, and total scores were also lower for gemifloxacin than clarithromycin at 26 weeks, but these differences did not reach statistical significance.

With respect to quality of life, relief from respiratory symptoms appeared to be the biggest benefit for smokers and ex-smokers.

Nonantibacterial Effects of Antimicrobials

Dr. Lowell Young, Director of the Kuzell Institute for Arthritis and Infectious Diseases and Chief of Infectious Diseases at California Pacific Medical Center (San Francisco), discussed the possibility that antibiotics of several classes, including macrolides, penicillins, and fluoroquinolones, may have potentially beneficial nonantibacterial benefits in the treatment of infection. These mechanisms include decreased bacterial adherence and colonization, boosts of cellular or humoral immunity, cytokine modulation, virulence factor elaboration, and effects on nitric oxide (endothelial relaxing factor).

Dr. Young observed that fluoroquinolones reduce adherence to various epithelial cells and may interfere with the early phase of infection. One study of norfloxacin in cirrhotic patients demonstrated an improvement in blood flow, suggesting an antagonism of the vasodilatory effects of nitric oxide. In another study, fluoroquinolone treatment caused an increase in cytokines. "I think there are major limitations on interpreting this type of data," he said. "In all of these animal studies, a big difference may result from the choice of the dose of drug and the timing of administration." In addition, the clinical significance of modest endotoxin-neutralizing abilities remains unclear. To further explore these reported nonantibacterial effects, Dr. Young recommended comprehensive human antibiotic trials that examine outcomes of blood pressure, acute respiratory distress syndrome (ARDS), disseminated intravascular coagulation (DIC), renal failure, survival, and measurement of microbial products in addition to conventional treatment end points.

Fluoroquinolones May Provide Best Response After Biological Terrorism

Dr. Ethan Rubinstein, Professor of Medicine in the Infectious Disease Unit of Sheba Medical Center, Tel Hashomer, Israel, reviewed the history of biological warfare and presented sobering statistics regarding the modern potential for its threat.

Currently, Dr. Rubinstein noted, 54 nations sell anthrax, "the deadliest biological weapon." Anthrax expresses three toxins and "causes death in 87% to 95% of persons exposed to as little as 10,000 spores." Dr. Rubinstein calculated that a mere 50 kg of anthrax could cause 100,000 deaths in a city of 500,000 people (20% mortality). Although a vaccine for anthrax exists, it is associated with frequent side effects, requires a complex immunization schedule, and is not widely available. Anthrax has achieved resistance to penicillin and doxycycline through plasmid transfer. Consequently, he said, "fluoroquinolones remain ... the only oral alternative for mass treatment." Current guidelines for prophylaxis after an anthrax attack recommend ciprofloxacin 500 mg orally every 12 hours for 60 days.³⁶ The exposed population would also have to receive anthrax vaccination to prevent chronic reinfection from an environment contaminated with anthrax spores.

Another 18 nations sell Yersinia pestis, the bacteria responsible for bubonic plague (also called the Black Death), an ancient cause of human decimation. Although Y. pestis is generally susceptible to multiple antibiotics, a case report from Madagascar³⁷ revealed significant antibiotic resistance to aminoglycosides due to a plasmid carrier, which, Dr. Rubinstein said, "leaves the fluoroquinolones as the only available option for the treatment of such infections."

As a response to these threats, many countries have developed biological warfare response plans and stockpiled fluoroquinolones. Dr. Rubinstein pointed out that the ecological impact of prolonged treatment of thousands of people with fluoroquinolones is unknown. Because of the continued threat of biological warfare, he recommended more research into the mass use of fluoroquinolones.

Fluoroquinolones in Pediatric and Geriatric Populations

Pediatric
David Spencer, MD, FRCPCH, Consultant in Respiratory Paediatrics at Newcastle upon Tyne Hospitals NHS Trust (Newcastle upon Tyne, UK), discussed the use of fluoroquinolones in children. These drugs are most often used to eliminate infection with Pseudomonas aeruginosa in children with cystic fibrosis.³⁸ Immunoglobulin G (IgG) antibodies in early Pseudomonas infection may lag 6-25 months after the onset of infection. By that time, most patients are chronically colonized and more difficult to treat. Dr. Spencer said that using Pseudomonas antibodies alone for diagnosis of infection in this population may result in "missing the boat."

Epidemiologic studies reveal that chronic Pseudomonas infection in cystic fibrosis patients is associated with respiratory function decline and early death. But it is not clear if Pseudomonas actually hastens these end points. Dr. Spencer noted that although many people refer to Pseudomonas "colonization," even patients without clinical symptoms may have evidence of infection, such as neutrophil influx and inflammatory markers and mediators (leukotriene B4 [LTB4] and interleukin-8 [IL-8]). Pseudomonas produces elastase, which causes tissue degradation, potentiates effects of IL-8, and increases bronchial responsiveness, among other effects.

Dr. Spencer observed that at least one third of patients with cystic fibrosis receive treatment with quinolones. Long-term nebulized antibiotics appear to benefit children colonized with Pseudomonas, but the precise indications for intermittent systemic antibiotics remain unclear. Further studies are necessary to clarify the role of quinolone therapy in these patients.

Dr. Urs Schaad, Professor and Chairman of Pediatrics and Head of the Division of Infectious Diseases at University Children's Hospital, Basel, Switzerland, reviewed the problem of quinolone-induced arthropathy, which has been observed in juvenile animals. The pathogenesis of this phenomenon is unclear, although it may be related to magnesium chelation by fluoroquinolones and disturbance of chondrocyte endocrine receptors. All quinolones tested have been associated with this unusual side effect. Consequently, regulatory agencies have not approved quinolones for children, growing adolescents, or pregnant women. One study of 15 children used magnetic resonance imaging and found no cartilage pathology.³⁹ Similarly, no skeletal abnormalities have been found in the small number of fetal exposures to this class of antibiotics.⁴⁰ Because of compassionate-use programs and the necessity to treat Pseudomonas and other bacteria resistant to conventional antibiotics, Dr. Schaad estimated that more than 1,000,000 children have received fluoroquinolones, particularly ciprofloxacin. To date, there have been no reports of quinolone-related arthropathy in these children. Given the number of pediatric exposures, Dr. Schaad calculated that the risk of cartilage toxicity occurring in this population was approximately 0.0003%. Despite the potential toxicity of these drugs and the risk of increased bacterial resistance, Dr. Schaad recommended "not restriction, but controlled use" of fluoroquinolones in children.

Geriatric
Dr. Andrew Davies, MBBS, MRCP, Consultant Geriatrician, Sunderland Royal Hospital (Sunderland, UK), reviewed the role of quinolones in the geriatric population. He observed that infection is a common problem in the elderly and that "about 25% of [geriatric practice] consultations are respiratory related, of which 80% are due to infection."

S. pneumoniae is still the most common cause of pneumonia in this age group, but gram-negative bacteria, anaerobes, and Staphylococcus aureus are more common than in younger patients. Escherichia coli is the most common cause of urinary tract infection, but proportionally less common than in younger patients. Bacteriuria in chronically catheterized nursing home patients is often polymicrobial and may be due to Proteus species.

Dr. Davies commented that "old age in itself is a recognized risk factor for infection." He outlined some of the many age-related physiologic changes that predispose patients to infection, including the breakdown of mucus membrane integrity, reduced gastric acidity and increased gastric reflux increasing gram-negative chest infections, urinary obstruction and stasis due to prostate enlargement and bladder-wall changes that promote urinary tract infection, decline of thymic function with consequent reduced levels of cell-mediated immunity, increased percentage of memory T cells and reduced levels of naive T cells, reduced lymphocyte proliferation, reduced IL-2 production, decreased natural killer (NK)-cell activity, decreased B-cell function, lower levels of circulating antibodies, and impaired chemotactic response of polymorphonuclear cells.

In addition, the diagnosis of infection can be more difficult in this population. An infection may present as confusion or delirium or have other atypical presentations. The propensity for treatment-related drug interactions is also increased--physicians prescribe an average of 15 new medications per year for patients >65 years old (compared with 5 per year for patients <65 years old).

Fluoroquinolones represent good antibiotic choices for the treatment of a variety of infections in the elderly for a number of reasons. Quinolones are potent against gram-negative pathogens, H. influenzae, Pseudomonas aeruginosa, and Legionella species and can arrest these potentially severe infections early before culture and sensitivity information becomes available. Fluoroquinolones are well absorbed and can be given orally, sparing the patient intravenous therapy. Dose adjustment is required only in patients with severe renal or hepatic disease. Compliance may be enhanced with the simple once- or twice-a-day regimen employed in quinolone therapy.

Single-dose treatment produces bacteriologic cure in 80% of patients with a simple urinary tract infection; 3 days of treatment increases the effectiveness. In complicated urinary tract infections, quinolones produce cure in 85%-95% of cases. Quinolones are also useful in acute prostatitis. They are effective in lower respiratory tract infections, particularly in patients with chronic obstructive pulmonary disease.

Broad-spectrum cephalosporins, amoxicillin, and ampicillin can lead to Clostridium difficile diarrhea, which in 1997 was responsible for ward bed closures in 16% of hospitals in the UK (13,000 cases).⁴¹ Therapy with quinolones is rarely associated with this type of diarrhea.

Dr. Spencer offered several cautions concerning the use of quinolones in elderly individuals. Oral absorption may be decreased, he said, when quinolones are taken with magnesium, zinc, or iron, and quinolones may decrease theophylline clearance and interact with warfarin. In addition, they may produce seizures in combination with nonsteroidal anti-inflammatory drugs (NSAIDs). He advised that quinolones are a good first-line therapy in patients with urine or respiratory infection, the bulk of infections in the elderly.

Dr. Lindsay Nicolle, Consultant in Infectious Diseases at the Health Sciences Centre and Professor of the Section of Infectious Diseases at the University of Manitoba (Winnipeg, Manitoba, Canada), also treats many elderly patients. She commented that the favorable qualities of the fluoroquinolones, such as broad-spectrum activity, oral and infrequent dosing, and excellent safety profile, "have certainly led to their overuse."

In one study, 41% of long-term-care residents in Nashville, Tennessee, who presented to the emergency room with urinary tract infection had fluoroquinolone resistance.⁴² Dr. Nicolle noted that "there is an intense use of fluoroquinolones in long-term-care facilities." She agreed that quinolones are generally better than other drugs for urinary tract infections, but added that they are not always significantly better.

To address the problem of fluoroquinolone overuse, Dr. Nicolle suggested clinical trials in patients in long-term-care facilities to determine the best antimicrobial therapies, as well as guidelines and reviews of antimicrobial usage. "These are valuable antimicrobials," she said. "I think we need to be much more cautious about our usage of them."

Adverse Effects and Safety Issues

The selection of the best antibiotic for a particular infection depends not only on its bacterial sensitivity, but also on its side-effect profile and cost. The prescribing physician must consider individual patient factors as well, such as preferred route of administration, drug allergies, levels of hepatic and renal function, and potential for interactions with other concurrently administered medications.

While many of the fluoroquinolones are considered safe and well tolerated, several have been associated with specific adverse effects.⁴³ The most common adverse effects involve the gastrointestinal tract, skin, and central nervous system (CNS). As with many antibiotics, gastrointestinal effects include gastric irritation, nausea, vomiting, and diarrhea, occurring in 3%-6% of patients.⁹ The CNS effects, occurring in approximately 1% of patients, include headache, dizziness, and confusion. Skin reactions such as rash, pruritis, flushing, and phototoxicity (explored more fully below) can also occur. Attention has been focused on tendon rupture associated with even relatively short courses of fluoroquinolones. Tendonitis and tendon rupture (usually of the Achilles tendon) are rare class side effects of fluoroquinolones. A retrospective analysis by van der Linden et al.⁴⁴ revealed that ofloxacin was the most frequently implicated and that fluoroquinolone-associated tendon disorders are more common in patients more than 60 years of age. Tendon disorders after treatment with pefloxacin (not commercially available in the US) have also been frequently reported,⁴⁵ with an estimated incidence of 15-20 cases per 100,000 patients. Predisposing factors appear to be concomitant use of steroids, end-stage renal disease, age, and mechanical stress on the tendon.

The nature of the chemical structure of certain fluoroquinolones can be associated with specific side effects. For example, sparfloxacin (currently back-ordered with no release date per Bertek Pharmaceuticals [6/27/01]), clinafloxacin (no longer in development), and lomefloxacin have a halogen attached to the C-8 position of the 4-quinolone nucleus and can produce phototoxicity. A difluorophenyl group in the R1 position common to temafloxacin (Omniflox, Abbott; withdrawn from market) and trovafloxacin appears responsible for hepatotoxicity. Grepafloxacin and sparfloxacin can prolong the QT interval (explored more fully below). The potential for severe cardiovascular adverse reactions resulted in the manufacturer voluntarily withdrawing grepafloxacin from the market in October 1999. In general, it is advisable to avoid using fluoroquinolones in combination with other drugs that may prolong the QT interval (class IA antiarrhythmics--quinidine, procainamide; class III antiarrhythmics--sotalol, amiodarone; tricyclic antidepressants; phenothiazines; cisapride; pentamidine; and erythromycin) and in patients known to have prolonged QT intervals or uncorrected hypokalemia. Newer quinolones will be developed with an eye to lowering articular, photo-, and neurotoxicities.

Central Nervous System
Dr. S. Ragnar Norrby, Director General of the Swedish Institute for Infectious Disease Control (Solna, Sweden), reviewed the incidence of CNS side effects. Patients have experienced agitation, dizziness, headache, insomnia, hallucinations and seizures. In his review of the clinical trial data, Dr. Norrby noted that grepafloxacin and levofloxacin had an incidence of CNS side effects of <1%, trovafloxacin 3%, and sparfloxacin 5%. However, when trovafloxacin was given at a higher dose (200 mg), the incidence rose to 11%. Postmarketing, levofloxacin had 4 in 100,000 adverse CNS reactions, while trovafloxacin had 26 in 100,000. Dr. Norrby observed that although ofloxacin caused insomnia and dysphoria, these symptoms are not seen with levofloxacin, its levo-isomer. He suggested that better understanding of the three-dimensional configuration of these molecules would lend insight into the pathophysiologic basis of adverse reactions.

Dr. Norrby postulated that substitutions at R7 may be important for binding at the gamma-aminobutyric acid (GABA) receptor and responsible for CNS side effects. Increased lipophilicity is another drug property that could contribute to toxicity by increasing CNS drug penetration. He hypothesized that some quinolones may cause seizures by interfering with the magnesium component of the N-methyl-d-aspartate receptor. However, "CNS toxicity varies between quinolones, with no specific culprit for serious reactions," he said.

Drug interactions between fluoroquinolones and NSAIDs, theophylline, or foscarnet (Foscavir, AstraZeneca) can also increase the risk of seizures. Consequently, Dr. Norrby recommended that "combinations of quinolones and NSAIDs should be used carefully, as should combinations with foscarnet." Dr. Norrby concluded his talk by noting that severe CNS side effects such as seizures are rare. "CNS toxicity of quinolones is not a major problem with any of the frequently used derivatives," he said.

Phototoxicity
Professor James Ferguson, from the Photobiology Unit at Ninewells Hospital and Medical School, Dundee, Scotland, reviewed the problem of fluoroquinolone photosensitivity. He pointed out that in addition to the quinolones, there are "many groups of drugs that produce phototoxicity." Phototoxicity can be drug and dose related and is due mostly to ultraviolet A (UVA) exposure. UVA is present all year in high latitudes; penetrates glass, water, and thin cotton clothing; and increases with reflection and altitude. Sunscreen absorbs UVA, but not as well as it absorbs UVB.

UV light acts on the drug or metabolite, producing free radicals or toxic photoproducts. Phototoxicity can cause acute and chronic skin damage, systemic toxicity due to circulating photoproducts, and drug photodegradation with loss of effect. Chronic effects of phototoxicity include skin pigmentation, photoaging, and photocarcinogenesis.

Dr. Ferguson pointed out that slight structural changes in the fluoroquinolone molecule could dramatically change phototoxic properties. Addition of a halide ion to the C-8 position (absent in gemifloxacin) is thought to be responsible for phototoxicity. Consequently, he said, "fluoroquinolone phototoxicity severity varies greatly from one drug to another." (Some fluoroquinolones are actually photoprotective.) Sparfloxacin and lomefloxacin have relatively high phototoxicity, moxifloxacin's and gatifloxacin's are similar to controls, and that of gemifloxacin is mild (similar to that of ciprofloxacin) and dose-dependent.^46,47

With the use of a monochromator, Dr. Ferguson has tested many fluoroquinolones to determine their level of phototoxicity. He recommends evaluating drugs during phase I development, with testing at all wavelengths. Although certain fluoroquinolones are very phototoxic, Dr. Ferguson emphasized that for drugs with low phototoxicity, "simple light avoidance" is usually sufficient to protect patients.

Repolarization Abnormalities
Dr. John Camm, a cardiologist at St. George's Hospital Medical School, London, UK, explained that many classes of drugs affect cardiac repolarization, causing an increase in the QT interval. These drugs include chloroquine, cisapride, clarithromycin, class I antiarrhythmics, erythromycin, halofantrine, pentamidine, phenothiazines, and tricyclic antidepressants. Typically, these drugs block the cardiac repolarizing potassium current (I_Kr), which leads to prolonged action potential and prolonged QT interval that can be measured on electrocardiogram. A prolonged action potential can cause early after-depolarizations that result in a reentrant mechanism. Extra ventricular beats and ventricular tachycardia (torsades de pointes) can ensue.

Patients with histories of bradycardia, ventricular tachycardia, congestive heart failure, ventricular hypertrophy, metabolic abnormalities, and underlying QT prolongation are at risk for adverse cardiac events. For reasons that are not fully understood, women and the elderly are more susceptible. The risk of QT prolongation is greatest with sparfloxacin and grepafloxacin. Gemifloxacin appears to have a low potential for producing cardiotoxicity.⁴⁶

Dr. Paul Iannini, Clinical Professor of Medicine, Yale University School of Medicine (New Haven, CT), explained that very few cases of torsades de pointes have been reported in association with the use of fluoroquinolones. For example, three cases have been reported worldwide for norfloxacin, of 6.3 million patients. Similarly, there have been five cases reported for gatifloxacin, of 4 million patients. Although the exact incidence of torsades de pointes in the general population is not known, Dr. Iannini commented that it appeared to be of a similar magnitude. However, given the uncertainty of the figures and the potential lethality of this rare side effect, "avoidance of quinolone use in any patient with predisposing factors should be the rule; avoidance of quinolone usage with any other drugs with similar cardiotoxicity is advisable," he said. "The more severe the [heart] failure, the more likely there will be an interaction between the pathophysiologic state and the I_Kr." He concluded that "the public health risk of acquired long QT syndrome due to fluoroquinolones is not great."

References

1. Bartlett JG et al. Practice guidelines for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clin Infect Dis 2000;31:347

2. Weiss et al. Clin Infect Dis, in press

3. Ball P. Quinolone generations: natural history or natural selection? J Antimicrob Chemother 2000;46(suppl T1):17

4. Acar JF, Goldstein FW. Trends in bacterial resistance to fluoroquinolones. Clin Infect Dis 1997;24(suppl 1):S67

5. Takahata M et al. In vitro and in vivo efficacies of T-3811ME (BMS-284756) against Mycoplasma pneumoniae. Antimicrob Agents Chemother 2001;45:312

6. Heaton VJ et al. Activity of gemifloxacin against penicillin- and ciprofloxacin-resistant Streptococcus pneumoniae displaying topoisomerase- and efflux-mediated resistance mechanisms. Antimicrob Agents Chemother 1999;43:2998

7. Garey KW, Amsden GW. Trovafloxacin: an overview [review]. Pharmacotherapy 1999;19:21

8. Stein GE. Pharmacokinetics and pharmacodynamics of newer fluoroquinolones [review]. Clin Infect Dis 1996;23(suppl 1):S19

9. Fitton A. The quinolones: an overview of their pharmacology [review]. Clin Pharmacokinet 1992;22(suppl 1):1

10. Hardy D et al. Comparative in vitro activities of ciprofloxacin, gemifloxacin, grepafloxacin, moxifloxacin, ofloxacin, sparfloxacin, trovafloxacin, and other antimicrobial agents against bloodstream isolates of gram-positive cocci. Antimicrob Agents Chemother 2000;44:802

11. De Azavedo et al. Activity of gemifloxacin against ciprofloxacin-resistant Streptococcus pneumoniae. Abstract 85, 7th International Symposium on New Quinolones, June 10-12, 2001, Edinburgh

12. Blondeau JM et al. Susceptibility of Canadian isolates of Haemophilus influenzae, Moraxella catarrhalis and Streptococcus pneumoniae to oral antimicrobial agents. Int J Antimicrob Agents 2001;17:457

13. Forrest A et al. Pharmacodynamics of intravenous ciprofloxacin in seriously ill patients. Antimicrob Agents Chemother 1993;37:1073

14. Craig WA. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men [review]. Clin Infect Dis 1998;26:1

15. Preston SL et al. Pharmacodynamics of levofloxacin: a new paradigm for early clinical trials. JAMA 1998;279:125

16. Lacy MK et al. Pharmacodynamic comparisons of levofloxacin, ciprofloxacin, and ampicillin against Streptococcus pneumoniae in an in vitro model of infection. Antimicrob Agents Chemother 1999;43:672

17. Pickerill KE et al. Comparison of the fluoroquinolones based on pharmacokinetic and pharmacodynamic parameters [review]. Pharmacotherapy 2000;20:417

18. Chen DK et al. Decreased susceptibility of Streptococcus pneumoniae to fluoroquinolones in Canada. Canadian Bacterial Surveillance Network. N Engl J Med 1999;34:233

19. Heaton VJ et al. Potent antipneumococcal activity of gemifloxacin is associated with dual targeting of gyrase and topoisomerase IV, an in vivo target preference for gyrase, and enhanced stabilization of cleavable complexes in vitro. Antimicrob Agents Chemother 2000;44:3112

20. Bernstein JM. Treatment of community-acquired pneumonia: IDSA guidelines. Infectious Diseases Society of America. Chest 1999;115(3 suppl):9S

21. Gwaltney JM Jr. Acute community-acquired sinusitis [review]. Clin Infect Dis 1996;23:1209

22. Ball P, Make B. Acute exacerbations of chronic bronchitis: an international comparison. Chest 1998;113:199S

23. Marrie TJ. Community-acquired pneumonia: epidemiology, etiology, treatment [review]. Infect Dis Clin North Am 1998;12:723

24. Jacobs MR et al. Susceptibilities of Streptococcus pneumoniae and Haemophilus influenzae to 10 oral antimicrobial agents based on pharmacodynamic parameters: 1997 U.S. surveillance study. Antimicrob Agents Chemother 1999;43:1901

25. Susceptibility of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis to oral agents: results of a 1998 US outpatient surveillance study [abstract]. In: Program and abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco. Washington, DC: American Society for Microbiology, 1999

26. Bartlett JG et al. Community-acquired pneumonia in adults: guidelines for management. The Infectious Diseases Society of America. Clin Infect Dis 1998;26:811

27. Lode H et al. Treatment of community-acquired pneumonia: a randomized comparison of sparfloxacin, amoxycillin-clavulanic acid and erythromycin. Eur Respir J 1995;8:1999

28. O'Doherty B et al. Randomized, double-blind, comparative study of grepafloxacin and amoxycillin in the treatment of patients with community-acquired pneumonia. J Antimicrob Chemother 1997;40(suppl A):73

29. File TM Jr et al. A multicenter, randomized study comparing the efficacy and safety of intravenous and/or oral levofloxacin versus ceftriaxone and/or cefuroxime axetil in treatment of adults with community-acquired pneumonia. Antimicrob Agents Chemother 1997;41:1965

30. Efficacy of oral gemifloxacin (7-14 days) compared with intravenous ceftriaxone (1-7 days) followed by oral cefuroxime (1-13 days, пїЅ a macrolide) in the treatment of CAP. Abstract 117, 7th International Symposium on New Quinolones, June 10-12, 2001, Edinburgh

31. Ball P. Epidemiology and treatment of chronic bronchitis and its exacerbations [review]. Chest 1995;108(2 suppl):43S

32. Wilson et al. and the 068 Clinical Study Group. Efficacy of once-daily gemifloxacin compared with twice-daily clarithromycin for 7 days in the treatment of AECB. Abstract 116, 7th International Symposium on New Quinolones, June 10-12, 2001, Edinburgh

33. Wilson et al. and the GLOBE study group. Gemifloxacin long-term outcomes in bronchitis exacerbations (GLOBE) study-An assessment of health outcome benefits in AECB patients following 5 days gemifloxacin (GEMI) therapy. Abstract 108, 7th International Symposium on New Quinolones, June 10-12, 2001, Edinburgh

34. Shreve JL et al. The use of pharmacoeconomic research by MCOs: an actuary's view. Drug Benefits Trends 2000;12:45

35. Jones et al. Greater Improvement in health status of smokers and ex-smokers treated for AECB with gemifloxacin versus clarithromycin: the GLOBE study. Abstract 106, 7th International Symposium on New Quinolones, June 10-12, 2001, Edinburgh

36. Inglesby TV et al. Anthrax as a biological weapon: medical and public health management [review]. Working Group on Civilian Biodefense. JAMA 1999;281:1735

37. Guiyoule A et al. Transferable plasmid-mediated resistance to streptomycin in a clinical isolate of Yersinia pestis. Emerg Infect Dis 2001;7:43

38. Fredericksen B et al. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 1997;23:330

39. Redmond A et al. Oral ciprofloxacin in the treatment of pseudomonas exacerbations of paediatric cystic fibrosis: clinical efficacy and safety evaluation using magnetic resonance image scanning. J Int Med Res 1998;26:304

40. Berkovitch M et al. Safety of the new quinolones in pregnancy. Obstet Gynecol 1994;84:535

41. Zadik PM, Moore AP. Antimicrobial associations of an outbreak of diarrhoea due to Clostridium difficile. J Hosp Infect 1998;39:189

42. Wright SW et al. Prevalence and risk factors for multidrug resistant uropathogens in ED patients. Am J Emerg Med 2000;18:143

43. Mandell LA. Antimicrobial safety and tolerability: differences and dilemmas [review]. Clin Infect Dis 2001;32(suppl 1):S72

44. van der Linden PD et al. Tendon disorders attributed to fluoroquinolones: a study on 42 spontaneous reports in the period 1988 to 1998. Arthritis Rheum 2001;45:235

45. Ball P et al. Comparative tolerability of the newer fluoroquinolone antibacterials [review]. Drug Saf 1999;21:407

46. Lowe MN, Lamb HM. Gemifloxacin. Drugs 2000;59:1137

47. Vousden M et al. Evaluation of phototoxic potential of gemifloxacin in healthy volunteers compared with ciprofloxacin. Chemotherapy 1999;45:512