Study objectives: To compare the safety and efficacy of oral azithromycin and levofloxacin in the treatment of outpatients with acute bacterial exacerbations of chronic bronchitis (ABECB).
Design: Randomized, double-blinded, double-dummy, multicenter trial with 1:1 treatment allocation.
Setting: Outpatient treatment setting.
Patients: Two hundred thirty-five male or female outpatients between the ages of 35 and 75 years who had received a clinical diagnosis of ABECB.
Interventions: Blinded treatment with either oral azithromycin, 500 mg on day 1 and 250 mg per day for days 2 to 5, or, oral levofloxacin, 500 mg q24h for 7 days.
Results: Both treatments were well-tolerated, with the majority of adverse events being GI in nature. Favorable clinical outcomes in clinically evaluable patients were demonstrated in 89% of patients receiving azithromycin and in 92% of patients receiving levofloxacin by day 4 of therapy. At day 24, the posttherapy visit, favorable responses were approximately 82% and 86%, respectively, for patients in the two treatment groups. The bacterial eradication rates of respiratory pathogens were 96% for azithromycin and 85% for levofloxacin.
Conclusions: Despite increasing concerns over macrolide resistance and a higher incidence of Gram-negative pathogens, a standard 5-day course of oral azithromycin was clinically and bacteriologically equivalent to a 7-day course of oral levofloxacin in the treatment of patients with ABECB. (CHEST 2003; 123:772-777)
Key words: acute bacterial exacerbations of chronic bronchitis; azithromycin; chronic bronchitis; levofloxacin
Abbreviations: ABECB = acute bacterial exacerbations of chronic bronchitis; AECB = acute exacerbations of chronic bronchitis; MIC = minimum inhibitory concentration
Approximately 14 million Americans have COPD in varying stages of severity. Most patients have American Thoracic Society stage I disease, which is defined as an FE[V.sub.1] of [greater than or equal to] 50% of the predicted value and causes minimal impact on health-related quality of life. Patients with stage II and III disease experience significant impacts on their health-related quality of life and have FE[V.sub.1] values of 35 to 49% of predicted and < 35% of predicted, respectively. (1) Although stage I patients are generally cared for by general practitioners and represent only modest amounts of per capita health-care expenditures, those with stage II and III disease are usually relegated to specialty care and consume a much higher amount of health-care resources. As the majority of these patients experience COPD due to the presence of chronic bronchitis (89%), a fair amount of the per capita expenditures for all these patients are the acute exacerbations of chronic bronchitis (AECB) that are common to their underlying disease, no matter the overall severity of their condition.
The treatment of AECB remains controversial to this day. Some think that the evidence in the literature does not bear out the bacterial etiology of AECB and that, therefore, treatment is not warranted. Conflicting reports in the literature show an increase in the incidence of positive culture results and bacterial loads as well as changes in leukocyte influx during exacerbations compared with remission periods. (2) Although the role of noninfectious causes for AECB episodes is well-recognized, a thorough review of the literature does not support withholding antibiotic treatment. Rather, it encourages the identification and treatment of the appropriate subgroups of patients that will benefit the most from antibiotic intervention. For example, a study by Anthonisen and colleagues, (3) demonstrated that AECB patients with at least two cardinal AECB symptoms (ie, increased dyspnea, increased sputum volume, and increased sputum purulence) had a higher response rate to antibiotic therapy than to placebo, as well as a significantly shorter course of illness and a more rapid improvement in peak flow rates. The difference was even more striking for those patients with all three cardinal symptoms. (3) As demonstrated by Destache and colleagues, (4) antibiotic selection is as important as identifying patients for whom antibiotic therapy is indicated. In that study, patients who received second-line agents (eg, amoxicillin/clavulanate, ciprofloxacin, or azithromycin) had fewer hospitalizations and less frequent exacerbations than those who received older first-line agents (eg, amoxicillin, co-trimoxazole, or erythromycin). (4)
As evidenced by the study of Destache et al, (4) advanced-generation macrolide agents and newer fluoroquinolone agents provide a broader spectrum of activity than did the older generic agents and improve outcomes in patients with AECB of bacterial origin (ABECB). Although both azithromycin and levofloxacin are widely utilized for the treatment of ABECB, there is growing concern over increasing rates of macrolide resistance among patients with Streptococcus pneumoniae. (5) To address this issue, this study was conducted to compare the clinical and bacteriologic outcomes associated with using a standard 5-day course of azithromycin vs a 7-day course of levofloxacin for the outpatient treatment of patients with ABECB.
MATERIALS AND METHODS
Study Population and Design
Two hundred thirty-five outpatients with ABECB were enrolled into this randomized, double-blinded, double-dummy study at approximately 21 centers that had all provided evidence of institutional review board approval. Male or female outpatients between the ages of 35 and 75 years who provided written informed consent were included in the study if they had received a clinical diagnosis of ABECB based on the following: a history of chronic bronchitis (ie, the presence of persistent productive cough for at least 3 months of 2 consecutive years); purulent sputum that was positive for a bacterial pathogen by Gram stain or culture; and at least two of three other criteria (ie, increased sputum purulence, increased dyspnea, and increased volume of sputum). Subjects had to be appropriate for outpatient antibiotic therapy with a baseline FE[V.sub.1] > 35% of predicted normal values in an acute exacerbation-free period and a chest radiograph that was negative for pneumonia. Major exclusion criteria included the following: hospitalization; acquired infection while in a hospital or skilled nursing facility; acute bronchitis or diagnosed types of COPD other than chronic bronchitis; significant comorbidities that may interfere with the evaluation of the clinical response; alcoholism or drug dependence; hypersensitivity to macrolide or fluoroquinolone agents; the presence of an infection known or suspected to be caused by a pathogen that was resistant to the antibiotics used in the study; a history of other antimicrobial or investigational drug therapy in the previous 4 weeks; the inability to give informed consent; previous participation in the study or another clinical trial utilizing the same study drugs; and, pregnancy, nursing, or unsatisfactory birth control methods.
Prior to enrollment, each subject's clinical findings, pulse oximetry values, or arterial blood gas measurements, as well as FE[V.sub.1] results were reviewed by an investigator. After the clinical diagnosis was made and it was determined that the subject was eligible for study enrollment, the subject signed an informed consent form and was randomized in a 1:1 ratio to study therapy. Depending on the assignment, patients received either 250-mg tablets of azithromycin and matching levofloxacin placebo, or 250-mg tablets of levofloxacin (in two capsules) and matching azithromycin placebo tablets. Subjects were told that they could take their medication at any time of the day. Subjects who were randomized to azithromycin took two tablets together on the first day, followed by one tablet daily for 4 more days (ie, days 2 to 5). Subjects who were randomized to levofloxacin took two capsules once daily for 7 days (ie, days 1 to 7). The matching placebo for the medication to which the subject was not randomized was taken concurrently with the active medication. The use of other concurrent systemic medications was limited during the study to those necessary for the well-being of the patient, with the exception of other antimicrobial agents.
Clinical/Bacteriologic Assessments and Definitions
At baseline, freshly expectorated sputum specimens were obtained, sent to a central laboratory, and examined microscopically for quality assessment. Routine cultures were performed on specimens only if > 25 WBCs and < 10 epithelial cells were present per low-power field. In vitro susceptibility tests for the study drugs were performed on all clinically significant isolates by disk diffusion and broth microdilution methods following the National Committee for Clinical Laboratory Standards guidelines. (6) National Committee for Clinical Laboratory Standards break points were applied to interpret the results. (6) During the study, patients returned for two follow-up visits. The first visit, on day 4 of therapy, included the following: an interval history of symptoms, clinical response and concurrent therapy; an interval physical examination and pulse oximetry/arterial blood gas check; a safety evaluation including a review of any adverse experiences, a Gram stain, and a culture of freshly expectorated sputum, if available, and repeat laboratory safety testing for any abnormalities noted at baseline; an assessment of the clinical and bacteriologic responses; and an assessment of drug compliance. The posttherapy assessment occurred on day 24 after screening and included all the testing and observations involved in the day 4 visit. At the end of the study, patients were deemed to be either evaluable or nonevaluable. To be considered evaluable, patients had to have finished at least 2 days of azithromycin therapy or 4 days of levofloxacin therapy, had to have been assessed at the day 24 post-therapy visit, and could not have received any other systemic antibiotics.
The clinical response was classified for each evaluable subject at both follow-up visits. Patients with favorable clinical outcomes were cured (ie, their condition had resolved with a return to baseline characteristics) or improved (ie, ABECB had subsided but with an incomplete return to baseline values). Patients with unfavorable clinical outcomes either did not respond to therapy or had a recurrence that was defined as a worsening of infection at the day 24 visit subsequent to a "favorable" outcome classification at the day 4 visit. For those evaluable patients who had a causative pathogen identified before treatment, the bacteriologic response was defined for each pathogen as follows: eradication, elimination of the pathogen from the culture at the day 4 visit, and for those patients with the inability to produce a sputum specimen, no residual symptoms of pulmonary infection; persistence, the presence of the initial causative pathogen in a sputum culture at either follow-up visit; or recurrence, the presence of the initial causative pathogen in a sputum culture from the day 24 follow-up visit in a patient who had been classified as "eradication" at the day 4 visit.
Efficacy and Statistical Analysis
A clinical response of cure or improvement was anticipated to occur in 85% of the chronic bronchitis patients in the levofloxacin group. Based on an assumption of equivalence between the azithromycin and levofloxacin groups, this study required 90 evaluable subjects per treatment group. With the assumption of a 70% evaluability rate in both groups, 130 chronic bronchitis patients were treated with azithromycin and the same number were treated with levofloxacin. This sample size ensured at least an 80% probability that the lower bound of the 95% confidence interval around the difference in the success rate between the two treatment groups in the evaluable population did not go below -15%. The primary measures of the effectiveness of the treatments were the clinical and bacteriologic responses at the day 4 and day 24 follow-up assessments. Nonevaluable patients were excluded from these analyses. Analyses of the efficacy data and the baseline characteristics were conducted for both the standard efficacy evaluable and intention-to-treat (ie, nonevaluable) patients.
Baseline demographic characteristics for the two treatment groups were compared using the two-sample t test for continuous variables and the [chi square] test or Fisher exact test, as appropriate, for categoric data. The distributions of the clinical responses by treatment group were compared separately at the day 4 and day 24 visits using the [chi square] test or Fisher exact test, as appropriate. The day 24 visit was considered to be the primary efficacy end point. Two-sided 95% confidence intervals, which were based on the normal approximation to the binomial distribution, were calculated for the difference of the proportion of satisfactory clinical responses (ie, cured or improved/total) between the two treatment groups. The treatment-by-center interaction was tested at a level of 0.15 significance in the clinically evaluable population using logistic regression models. Centers with fewer than five subjects per treatment arm were pooled for the analysis. Multivariate statistical procedures (eg, logistic regression) were used to identify prognostic baseline variables, such as age, weight, sex, and race, and other variables on the primary efficacy variables as an exploratory analysis. Where applicable, interactions with treatment were included in the model.
From August 1999 to May 2000, 235 patients were randomized to participate in the trial, with 118 randomized to receive azithromycin and 117 randomized to receive levofloxacin. Although all randomized subjects were included in the modified intent-to-treat analyses, 108 of the patients treated with azithromycin (91.5%) and 104 of the patients treated with levofloxacin (88.9%) were considered to be clinically evaluable for the study. Those who were not evaluable were excluded from analysis for the following reasons: received concurrent systemic antibiotics; did not receive at least 50% of the antibiotic regimen; or were lost to follow-up. Actuarial demographic characteristics of the clinically evaluable patients were not significantly different between the two treatment groups and are summarized in Table 1. Approximately half of the patients were current smokers (azithromycin group, 58 patients; levofloxacin group, 61 patients) and had smoked approximately one pack of cigarettes per day for about the last 31 years. Other types of tobacco use were negligible. Most nonsmokers were ex-smokers (azithromycin group, 47 patients; levofloxacin group, 42 patients) who had smoked for approximately 30 years but had stopped within the past 10 to 15 years. The average length of time for which chronic bronchitis had been diagnosed in both groups was approximately 10 to 12 years. Other comorbidities that were documented at baseline were typical for the age group being studied and were not significantly different between the two treatment groups. Those comorbidities with the highest incidence included the following: arthropathies; dorsopathies; GI disorders (ie, reflux or ulcer disease); cardiovascular disorders (ie, hypertension or ischemic heart disease); cataracts/glaucoma; diabetes mellitus; anxiety/ depression; and hypothyroidism. Although during the study patients in both groups increased their use of symptomatic therapies such as cough and cold preparations, corticosteroids, and inhalers, the majority of concomitant medications were used to treat underlying comorbidities. Overall, both the azithromycin and levofloxacin regimens were well-tolerated by patients, with 18% of patients (21 of 118 patients) and 20% (23 of 117 patient), respectively, reporting mild-to-moderate treatment-related adverse events. As would be expected, the majority of the adverse events that were reported were GI in nature for both antibiotics. In reviewing these patients, those receiving azithromycin were about twice as likely (8.5% vs 4.3%, respectively) to report diarrhea or loose stools, while those receiving levofloxacin reported nausea six times more often (6% vs 0.8%, respectively).
An analysis of the clinical patient outcome of evaluable patients is delineated in Table 2. Overall, on the day 4 visit while patients were receiving therapy, 89% of azithromycin-treated patients and 92% of levofloxacin-treated patients (p = 0.2252) had favorable responses to their therapy. When reassessed at the day 24 posttherapy visit, there was still no significant difference between the favorable response rates for the two treatment groups (82% vs 86%, respectively; p = 0.5266). The response rates for the modified intent-to-treat group did not vary from those of the clinically evaluable group. In the azithromycin arm, 27% of clinically evaluable patients (29 of 108 patients) had positive bacterial culture findings compared with 22% of patients in the levofloxacin arm (23 of 104 patients). The eradication rates for Haemophilus influenzae, Moraxella catarrhalis and S pneumoniae are listed in Table 3. A limited number of Staphylococcus aureus and Gram-negative isolates other than H influenzae and M catarrhalis were recovered. Eradication rates for these pathogens are listed in Table 4.
All respiratory pathogens were susceptible to study drugs at baseline. None developed resistance during therapy. When bacteriologic response rates from the day 4 visit while patients were receiving therapy were analyzed, azithromycin had eradicated 97% of all pathogens identified and 100% (Table 5) of identified respiratory pathogens. For the levofloxacin-treated patients, 92% of all pathogens were eradicated, as were 90% of identified respiratory pathogens. At the final follow-up visit, azithromycin and levofloxacin had eradicated 93% and 76%, respectively, of all baseline pathogens. These numbers included respective 96% and 85% eradication rates of respiratory pathogens. Eradication rates for both follow-up visits did not differ significantly (p = 0.2955) between the two treatment groups.
A review by Sethi (7) of the diagnosis and management of infectious exacerbations of chronic bronchitis states that fluoroquinolones are the preferred first-line treatment in chronic bronchitis patients in whom the disease is complicated by comorbid illness, severe obstruction (ie, FE[V.sub.1], < 50%), old age (ie, > 65 years), and/or have recurrent exacerbations. (7) Based on the demographic data presented about the patients in this study, it is evident that the majority of the patients met the criteria described by Sethi (7) that would put a patient at higher risk for failing nonfluoroquinolone therapy. Despite these findings, the patients in this study who were treated with a standard course of azithromycin had both equal clinical and bacteriologic outcomes compared to the levofloxacin-treated patients. In many ways, the results of this study mirror the results of a recently published study by DeAbate et al (8) that compared a 5-day course of azithromycin to a 5-day course of another fluoroquinolone, moxifloxacin. In that study too, both the clinical efficacy and bacteriologic eradication rates were equal between the two treatments, including the eradication rate of S pneumoniae.
The number of cultures that were positive for respiratory pathogens in this study was less than that anticipated, with single or multiple bacterial pathogens identified in only 25% of the study population. Recent clinical trials (9) looking at quinolone agents for the treatment of AECB have reported that between 35% and 52% of patients had positive culture findings. (9) A respiratory surveillance study (10) conducted in 1999 to 2000, which included patients with a culturable focus of infection, recovered pathogens from 30% of enrolled patients (1,468 of 4,779 patients) in whom AECB had been diagnosed. The poor recovery of resistant pathogens from evaluable subjects in clinical trials has confounded attempts to correlate in vitro resistance with clinical outcome.
Despite concerns over increasing macrolide resistance, the present study has demonstrated that a standard course of azithromycin is as efficacious as a course of levofloxacin for the treatment of patients with ABECB. Over the past 5 years, in vitro surveillance studies have reported that the rate of macrolide resistance among patients with S pneumoniae has increased steadily. This and other recent studies, together with the lack of corresponding clinical failures that have been reported in the literature, brings into question the need for avoiding macrolide agents in the treatment of patients with community-acquired respiratory infections. (5,11) In the case of azithromycin, the reason for the disparity between the rates of in vitro resistance and clinical failures has been hypothesized to be due to high concentrations of the drug within the WBCs (peak, [greater than or equal to] 80 mg/L; 12 days after the start of a 5-day regimen, > 10 mg/L) that bacteria are exposed to on phagocytosis, both in the blood and at the infection site. (11,12) As it is necessary to maintain the concentrations of the drug above the minimum inhibitory concentration (MIC) at the clearance site for as long as possible to optimize the pharmacodynamic effects of the drug, it becomes obvious that although the macrolides may retain their clinical activity in infections caused by pneumococcal isolates that are moderately resistant (MIC, 1 to 32 mg/L), they most likely wig not in those infections that are caused by isolates with high-level resistance (MIC, > 32 mg/L). (11) This is reflected in the most recent treatment guidelines for community-acquired pneumonia by the Infectious Diseases Society of America, (13) which continues to recommend the use of macrolide agents. It is important to consider that pneumococcal resistance varies geographically on national, regional, and community levels. In North America, the most common mechanism of pneumococcal resistance to macrolides is due to an efflux mechanism that confers low-level resistance. (5) In areas such as southern Europe, the majority of cases of macrolide resistance are due to target site modification resulting in MICs of > 64 mg/L. In these regions, alternative agents may have to be utilized for appropriate pneumococcal coverage. (14)
In conclusion, based on the results of this study it is evident that a standard 5-day regimen of azithromycin and a once per day, 7-day regimen of levofloxacin are equally efficacious and safe for patients with ABECB. Despite concerns, it appears as though macrolide agents remain a viable treatment option for patients with ABECB, as well as for those with other community-acquired respiratory tract infections.
(1) American Thoracic Society. Definitions, epidemiology, pathophysiology, diagnosis, and staging. Am J Respir Crit Care Med 1995; 152:S78-S83
(2) Hirschmann JV. Do bacteria cause exacerbations of COPD? Chest 2000; 118:193-203
(3) Anthonisen NR, Manfreda J, Warren CPW, et al. Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 1987; 106:196-204
(4) Destache CJ, Dewan N, O'Donahue WJ, et al. Clinical and economic considerations in the treatment of acute exacerbations of chronic bronchitis. J Antimicrob Chemother 1999; 43(suppl): 107-113
(5) Doern GV, Brueggemann A, Holley HP, et al. Antimicrobial resistance of Streptococcus pneumoniae recovered from outpatients in the United States during the winter months of 1994 and 1995: results of a 30-center national surveillance study. Antimicrob Agents Chemother 1996; 40:1208-1213
(6) National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing: ninth informational supplement, M100-S9. Wayne, PA: National Committee for Clinical Laboratory Standards, 1999
(7) Sethi S. Infectious exacerbations of chronic bronchitis: diagnosis and management. J Antimicrob Chemother 1999; 43(suppl):97-105
(8) DeAbate CA, Mathew CP, Warner JH, et al. The safety and efficacy of short course (5-day) moxifloxacin vs azithromycin in the treatment of patients with acute exacerbation of chronic bronchitis. Respir Med 2000; 94:1029-1037
(9) Guthrie R. Community-acquired lower respiratory tract infections. Chest 2001; 120:2021-2134
(10) Pfaller MA, Ehrhardt AF, Jones RN. Frequency of pathogen occurrence and antimicrobial susceptibility among community acquired respiratory tract infections in the respiratory surveillance program study: microbiology from the medical office practice environment. Am J Med 2001; 111(suppl): 4S-12S
(11) Amsden GW. Pneumococcal macrolide resistance: myth or reality? J Antimicrob Chemother 1999; 44:1-6
(12) Amsden GW, Nafziger AN, Foulds G. Pharmacokinetics in serum and leukocyte exposures of oral azithromycin, 1,500 milligrams, given over a 3- or 5-day period in healthy subjects. Antimicrob Agents Chemother 1999; 43:163-165
(13) Bartlett JG, Dowell SF, Mandell LA, et al. Practice guidelines for the management of community-acquired pneumonia in adults: guidelines from the Infectious Diseases Society of America. Clin Infect Dis 2000; 31:811-838
(14) Pantosti A, S'Ambrosio F, Tarasi A, et al. Antibiotic susceptibility and serotype distribution of Streptococcus pneumoniae causing meningitis in Italy, 1997-1999. Clin Infect Dis 2000; 31:1373-1379
* From The Clinical Pharmacology Research Center and Department of Adult and Pediatric Medicine (Dr. Amsden), Bassett Healthcare, Cooperstown, NY; Riverside Methodist Hospital (Dr. Baird), Columbus, OH; Southeast Research Associates (Dr. Simon), Austell, GA; and Pfizer (Dr. Treadway), New York, NY. This study was funded by a grant from Pfizer, Inc. Dr. Amsden consults for, does research for, and speaks for Pfizer, Inc. Dr. Treadway is employed by Pfizer, Inc. Drs. Baird and Simon both have conducted research for Pfizer in the past.
Manuscript received January 9, 2002; revision accepted August 1, 2002.
Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: firstname.lastname@example.org).
Correspondence to: Guy W. Amsden, PharmD, Clinical Pharmacology Research Center, Bassett Healthcare, One Atwell Rd, Cooperstown, NY 13326; e-mail: email@example.com
COPYRIGHT 2003 American College of Chest Physicians
COPYRIGHT 2003 Gale Group