Ofloxacin

Study objective: Comparison of efficacy and safety of sparfloxacin vs ofloxacin for treatment of acute bacterial exacerbations of chronic bronchitis (ABECB).

Design: Multicenter, double-blind, randomized study.

Setting: Sixty-eight private offices and outpatient clinics in the United States and Canada.

Patients: Seven hundred ninety-eight adults with ABECB, as confirmed by the acute onset of new (or worsened from the immediate premorbid state) cough and sputum production.

Interventions: Randomization 1:1 to sparfloxacin, 400 mg on day 1, then 200 mg once daily, or ofloxacin, 400 mg twice daily, with matching comparator placebos, given concurrently for 10 consecutive days.

Results: The primary efficacy parameter was overall response in the bacteriologically evaluable population. Overall success rates in this population were 85.3% and 89.3% for sparfloxacin and ofloxacin, respectively. The two-sided 95% confidence interval was -9.9, 1.9, indicating that sparfloxacin was statistically equivalent to ofloxacin. The all-treated population analysis was similar to that in the evaluable population. Bacterial eradication rates were similar in both treatment groups for Haemophilus influenzae, Streptococcus pneumoniae, Moraxella catarrhalis, Chlamydia pneumoniae, Haemophilus parainfluenzae, Klebsiella pneumoniae, Enterobacter cloacae, and Staphylococcus aureus. The frequency of adverse events overall was comparable in the two treatment groups. The sparfloxacin group had a lower frequency of digestive and nervous system adverse events, but a higher frequency of photosensitivity reactions than the ofloxacin group.

Conclusions: Once-daily oral treatment with 200 mg sparfloxacin (after initial 400 mg dose) is as effective as twice-daily treatment with 400 mg ofloxacin in patients with ABECB.

(CHEST 1998; 114:120-130)

Key words: acute bacterial exacerbations of chronic bronchitis; drug therapy; ofloxacin; sparfloxacin

Abbreviations: ABECB=acute bacterial exacerbations of chronic bronchitis; MIC=minimum inhibitory concentration; QTc=corrected QT interval

Chronic bronchitis, which is characterized by a history of cough with excessive sputum production occurring on most days for at least 3 months of the year for 2 successive years, is exhibited by as many as 25% of adults in the United States.[1,2] The etiologies of this condition include cigarette smoking, air pollution, bronchopulmonary infections, and other toxic stimuli. Abnormalities in the bronchial system resulting from chronic bronchitis, such as obstructed bronchioles, impaired mucociliary clearance, and chronic bacterial infections of bronchial epithelium, may predispose the patient to acute bacterial infection.[3] Acute bacterial exacerbations of chronic bronchitis (ABECB) are manifested by marked increase in cough and sputum production, often accompanied by increased sputum purulence. Patients may also experience fever, wheezing, and shortness of breath. Causative' pathogens most frequently implicated in ABECB are Haemophilus influenzae, Haemophilus parainfluenzae, Streptococcus pneumoniae, Moraxella catarrhalis, Neisseria spp, and, to a lesser extent, Klebsiella spp, and Pseudomonas spp.[3-6] Recent data suggest that Chlamydia pneumoniae could play a role in some patients.[7,5]

Controversy about the efficacy of antibiotic therapy for ABECB has resulted from conflicting clinical trial results, due partially to study design differences and modest success rates. Despite the controversy, treatment with antibiotics is an accepted therapy that may also reduce the number of future bacterial exacerbations and ultimately improve pulmonary function.[3,8-10] Ampicillin and tetracycline were first-line treatment options in the past; current first-line therapies include trimethoprim/sulfamethoxazole, amoxicillin clavulanate, cephalosporins, and macrolides. However, increasing resistance of organisms to these commonly utilized antibiotics presents a challenge in the treatment of ABECB. Fluoroquinolones are an alternative since they are not destroyed by [Beta]-lactamase-producing organisms and are therefore effective against Gram-negative pathogens such as H influenzae and M catarrhalis. Some quinolones also have activity against S pneumoniae, including multidrug-resistant strains.

Sparfloxacin is a synthetic, broad-spectrum, fluoroquinolone antibacterial agent for oral administration that is more active in vitro against common respiratory pathogens than other available quinolones.[7,11,12] Sparfloxacin is a potent inhibitor of bacterial DNA gyrase and is bactericidal at or near the minimum inhibitory concentration (MIC). Important advantages of sparfloxacin include excellent in vitro activity against Gram-positive and Gram-negative aerobic and atypical organisms causing ABECB (ie, S pneumoniae [[MIC.sub.90], 0.5 [micro]g/mL], including penicillin-resistant species, M catarrhalis [[MIC.sub.90], 0.03 [micro]g/mL], H influenzae [[MIC.sub.90], 0.015 [micro]g/mL], H parainfluenzae [[MIC.sub.90], 0.06 [micro]g/mL], and C pneumoniae [[MIC.sub.90], 0.6 to 0.25 [micro]g/mL])7 and extensive respiratory tract tissue penetration.[13,14] In addition, sparfloxacin has a prolonged half-life of 16 to 20 h, which allows for once daily administration.[15] Sparfloxacin also exhibits no interaction with theophylline,[16] digoxin,[17] warfarin,[18] cimetidine,[19] or probenecid.[20] Ofloxacin, a synthetic, broad-spectrum, oral antibacterial agent of the fiuoroquinolone class, is indicated for the treatment of ABECB caused by susceptible strains of designated organisms at a recommended dose of 400 mg every 12 h. This study was initiated to compare the efficacy and safety of sparfloxacin with the efficacy and safety of ofloxacin in the treatment of ABECB.

MATERIALS AND METHODS

Study Design

This was a double-blind, randomized, multicenter, comparative study involving approximately 800 patients with ABECB randomized 1:1 to receive either sparfloxacin (400 mg loading dose followed by 200 mg once daily) or ofloxacin (400 mg twice daily) for 10 days. A double-dummy system was used such that matching comparator placebos were given concurrently with the active drug during the treatment period. Sixty-eight investigative sites in the United States and Canada enrolled patients. This study was conducted in compliance with institutional review board regulations and was monitored according to good clinical practices. An informed consent was obtained from each patient. The first patient entered the study in November 1992 and the last patient completed the study in October 1994.

Study Population

Adult patients ([is greater than or equal to] 18 years of age) presenting with ABECB could be enrolled in the study. The diagnosis of ABECB required the presence of chronic bronchitis (ie, history of cough with production of sputum occurring on most days for at least 3 months of the year for at least 2 successive years) and an ABECB as confirmed by the acute (within 14 days) onset of new (or worsened from the immediate premorbid state) cough and sputum production. If the immediate premorbid state (ie, the patient's stable baseline condition in the 60 days prior to the onset of the current exacerbation of symptoms) included sputum production, the patient must have had both an increased volume and an increased purulence of the sputum produced. Gram's stain of' a purulent or mucopurulent sputum specimen obtained within 48 h prior to starting treatment with the study medication had to show [is greater than] 25 neutrophils and [is less than] 10 squamous epithelial cells per low-power field. Patients may also have had fever (temperature[is greater than] 38.0 [degrees] C by the oral route), wheezing, and shortness of breath. A chest radiograph confirming the absence of pulmonary findings consistent with pneumonia was required.

Patients with the following conditions were excluded by protocol: pneumonia; baseline pathogen(s) known to be resistant to either study drug prior to randomization; and hypersensitivity to or prior photosensitivity reaction with quinolones. Patients with the following conditions were also excluded: pulmonary disease(s) that precluded evaluation of the therapeutic response; significant hepatic disease or obstruction of the biliary tract, including patients with baseline bilirubin or hepatic enzyme levels (aspartate transaminase, alanine transaminase) that were greater than two times the upper limit of normal; baseline known or estimated creatinine clearance of [is less than or equal to] 50 mL/min; epilepsy, history of convulsive disorders (other than febrile seizures of childhood), or underlying conditions that could have predisposed to seizures; GI tract abnormalities that would have prevented absorption of oral medication; known HIV infection, neutropenia (WBC count [is less than] 3,000/[mm.sup.3]), or current use of immunosuppressive or chemotherapeutic agents for hematologic, oncologic, or organ transplantation purposes; a rapidly fatal illness; or a life expectancy of [is less than or equal to] 1 month.

Female patients of childbearing potential were not eligible for study participation unless they were using reliable contraception. Further exclusion criteria included the following: clinical findings that, in the investigator's opinion, warranted parenteral antibiotic therapy; concomitant infections requiring systemic antibacterial therapy; systemic antimicrobial treatment within 3 days prior to study drug administration; anticipated need for treatment during the study with antacids, other antibiotics, or any other concomitant systemic medication that may have interfered with interpretation of response; use of any other investigational compound within 28 days of starting treatment with the study medication or during study; and prior participation in this study.

Because a small increase in corrected QT interval (QTc) interval is caused by sparfloxacin treatment,[21] patients were excluded for congenital prolonged QT syndrome, concurrent use of antiarrhythmic agents or other medications known to cause QT prolongation, or baseline QTc interval [is greater than] 500 ms. Patients with the inability to minimize unprotected exposure to sunlight were 'also excluded from the study because of the photosensitizing effect of sparfloxacin documented in early clinical studies. The protocol was amended in October 1993 to provide new information regarding sparfloxacin-associated photosensitivity and additional precautions regarding sun exposure.

Study Conduct

Patients were evaluated at baseline (visit 1), during treatment (visit 2), 10 days posttreatment (visit 3 [test-of-cure]), and 28 days posttreatment (visit 4 [follow-up]). Baseline visit procedures included verification of inclusion/exclusion criteria, informed consent, medical history, physical examination (including vital signs), height, weight, clinical evaluation of signs and symptoms, pregnancy test (if appropriate), ECG, chest radiograph, pulmonary function testing (FVC and [FEV.sub.1]), samples for laboratory analyses (hematology, chemistry, urinalysis, and theophylline and digoxin concentrations [if appropriate]), and microbiology testing (see below). Physical examination, clinical evaluation of signs and symptoms, laboratory and microbiological analyses, ECG, and FVC/[FEV.sub.1] were repeated during treatment, at test-of-cure, and at follow-up. Patient weight was determined at test-of-cure and chest radiograph was repeated at premature discontinuation or at follow-up if pneumonia was suspected. Chest radiographs were read and interpreted by a radiologist and documented in a signed and dated written report. Comparison of baseline with premorbid chest radiographs was made when possible. An assessment of the patient's response to study drug treatment was made at test-of-cure (or premature discontinuation before test-of-cure); recurrence of infection was determined at follow-up (or premature discontinuation after test-of-cure).

To ensure optimal absorption of the study drugs, patients were told to take the study medication 2 h before or after taking ferrous sulfate, zinc supplements, or multivitamins with zinc and at least 1/2 h, and preferably 2 h, before or after a meal. Investigators reviewed each patient's concomitant medications to identify potential drug interactions. Patient compliance with the study drug regimen was assessed during therapy and at the test-of-cure assessment, and safety was monitored during treatment, at test-of-cure assessment, and at follow-up. Patients were advised to avoid exposure to direct or indirect sunlight and artificial ultraviolet light (eg, sunlamps) during treatment with study drug and for several days after therapy. Safety parameters included adverse clinical and laboratory events, and changes in ECG, vital sign, and physical examination findings. The QTc interval was determined either manually by the investigator or ECG-machine calculated. Adverse events were recorded and classified by the investigator as to severity and relationship to study drug medication. Patients could withdraw from the study at any time at their discretion or could be discontinued by the investigator for medical or administrative reasons. A final evaluation (physical examination, signs and symptoms, cultures, laboratory testing, ECG, FVC, and [FEV.sub.1], efficacy, and safety) was to be performed prior to withdrawal from the study.

Microbiology

A sputum specimen[22] for Gram's stain, routine culture, and susceptibility was obtained at all visits and at premature discontinuation. Sputum was Gram stained at the site to determine patient eligibility for enrollment. Sputum samples (including Gram-stained and unstained slides of sputum) were sent to a central laboratory (SciCor Laboratories; Indianapolis) for review of the Gram's stain and identification of etiologic pathogens and susceptibility testing. All bacteria considered to be pathogenic were identified to genus and species and tested for susceptibility to sparfloxacin and ofloxacin according to the National Committee for Clinical Laboratory Standards guidelines.[23,24] Susceptibility was performed by the broth dilution (MIC) method for ofloxacin and by MIC and disk (Kirby-Bauer) for sparfloxacin. Isolation and identification of C pneumoniae by accepted laboratory methods was attempted at baseline and posttreatment visits, and at premature discontinuation. Each case of C pneumoniae infection was diagnosed on the basis of a fourfold or greater increase in IgG titer or on the basis of appearance of IgM. A progressive fourfold decrease in C pneumoniae IgG titer was accepted as evidence of acute infection in the absence of a pathogen identified by culture. A single C pneumoniae IgG [is greater than or equal to] 1:512 also was accepted as evidence of acute infection with C pneumoniae in the absence of a pathogen identified by culture.[22]

Efficacy Analysis

Efficacy analyses were performed on the all-treated and evaluable populations. Clinical response was assessed by the investigator as cure (resolution of all signs and symptoms with no new signs or symptoms associated with the original infection or return of the given finding to the premorbid state), improvement (in patients not cured, resolution or reduction of the majority of signs and symptoms with no new signs or symptoms associated with the original infection), failure (no resolution and no reduction of a majority of signs and symptoms; worsening of one or more signs or symptoms, new signs or symptoms associated with the original infection or a new infection; or patient required other antimicrobial therapy for this episode of ABECB prior to or at test-of-cure), or indeterminate (inability to assess the patient's signs and symptoms due to "lost to follow-up" or no information) at the test-of-cure visit or on discontinuation prior to test-of-cure. Wheezing was not evaluated as part of the clinical response. Clinical success was defined as cure or improvement. The investigator also evaluated clinical recurrence (development of new or worsened signs and symptoms) at follow-up for those patients demonstrating clinical success at the test-of-cure assessment.

Bacteriologic response was determined for each patient in the bacteriologically evaluable and all-treated with a baseline pathogen populations at the test-of-cure assessment. The following categories were used for assessing the bacteriologic response to baseline pathogens from sputum cultures: eradication (absence of a pathogen at a postbaseline visit that was present at baseline), presumed eradication (no material available for culture at a postbaseline visit, for a pathogen that was present at baseline, with a corresponding clinical response of cured or improved), persistence (presence of a baseline pathogen at a postbaseline visit), presumed persistence (no results from a culture at a postbaseline visit for a pathogen, that was present at baseline, with a corresponding clinical response of failure), superinfection (emergence of a new organism that was not identified as a causative pathogen at baseline, accompanied by signs or symptoms of ABECB), multiple pathogens with partial eradication (eradication of at least one baseline pathogen, with persistence of at least one baseline pathogen), and indeterminate (no information on any baseline pathogen). Bacteriologic success was defined as eradication or presumed eradication. Superinfection, relapse/ reinfection (eradication of a baseline pathogen at test-of-cure and reappearance of the same pathogen [relapse] or reappearance of a superinfecting pathogen [reinfection] after test-of-cure, accompanied by signs or symptoms of ABECB), and colonization (appearance of a new organism, during or after therapy, not accompanied by signs or symptoms of ABECB) were evaluated at follow-up for patients demonstrating bacteriologic success at test-of-cure.

The primary efficacy variable was overall response in the bacteriologically evaluable patient population. Any patient who had a clinical response of cure or improvement (as determined by the investigator at the test-of-cure visit) and a bacteriologic response of either eradicated or presumed eradicated was deemed a success. Any patient with a clinical response of failure or a bacteriologic response of persistence, presumed persistence, superinfection, or multiple pathogens with partial eradication was considered a failure. All other outcomes were deemed indeterminate.

All randomized patients who received at least one dose of study medication constituted the all-treated population. The all-treated "with a baseline pathogen population consisted of a subset of the all-treated population who had at least one baseline pathogen. Patients who met the inclusion/exclusion criteria, had no substantive protocol deviation, and were treated with [is greater than]80% of the protocol-defined treatment course (at least 9 tablets of sparfloxacin or 16 tablets of ofloxacin) were included in the clinically evaluable population (for successes). Patients who discontinued before test-of-cure were evaluated as failures only if they received 2 full days of study medication. The bacteriologically evaluable population was a subset of the clinically evaluable population who had an identified lower respiratory tract pathogen.

Safety Analysis

All randomized patients who received at least one dose of study medication were assessed for safety.

Statistics

Sample size calculations were based on a two-sided 95% confidence interval method of establishing equivalence of two proportions with an assumed overall response success rate (cured plus improved) of 90% in the ofloxacin group. It was determined that 190 patients per treatment group would be required to have all 80% probability of showing that sparfloxacin did not differ from ofloxacin by [is greater than]10% in either direction. Given an expected attrition rate of approximately 10%, and an expectation that 50% of enrolled patients would have an identifiable respiratory tract pathogen, total study enrollment was estimated a priori at 850 patients.

Differences between treatment groups in the distribution of demographic variables and baseline characteristics were tested using two-way analysis of variance for continuous variables and the Cochran-Mantel-Haenszel test for categorical variables. The Cochran-Mantel-Haenszel test was also used to analyze selected adverse events. The two-sided 95% confidence interval for the difference in treatment response rates was used to determine equivalence between sparfloxacin and ofloxacin for efficacy variables. In addition, logistic regression analyses were performed to explore the effect of variables, in addition to treatment group, that were likely to be related to response: investigator, age, gender, race, pulmonary function status (maximum ratio of [FEV.sub.1]/FVC at baseline), underlying condition, type and number of baseline pathogens, baseline WBC count, baseline fever, and medication compliance.

RESULTS

Patient Population

A total of 798 patients were enrolled in the study. A higher percentage of patients (approximately 80%) had an identifiable respiratory pathogen than estimated a priori (50%). The number and percentage of patients in each population for both treatment groups are presented in Table 1. Overall, 78.4% of the enrolled patients completed the study: 80.0% in the sparfloxacin group and 76.9% in the ofloxacin group. Eighty-one percent of the patients were clinically evaluable and 61.2% of the patients were bacteriologically evaluable. As shown in Table 2, most of the all-treated patient population in each treatment group were white, while the percentages of male and female patients were similar. There were no statistically significant differences in demographic characteristics between the sparfloxacin and ofloxacin groups with the exception of a lower mean body weight in the sparfloxacin group as compared with the ofloxacin group. The sparfloxacin and ofloxacin groups were comparable with respect to smoking status, alcohol/illicit drug use, recent hospitalization, renal and pulmonary function, laboratory characteristics, underlying disease conditions, and signs and symptoms of infection at baseline. Demographic and other baseline characteristics of clinically evaluable patients were similar to those of the all-treated patients.

(*) p<0.020 by a two-way analysis of variance on ranks between the two treatment groups.

([dagger]) A single erroneous value from one patient in the ofloxacin group was omitted in this analysis.

Pathogens Isolated at Baseline

Pathogens identified in [is greater than or equal to] -10 patients are summarized in Table 3. Comparable numbers of isolates of each organism were obtained from patients in both treatment groups, with H parainfluenzae being the most frequently isolated pathogen (approximately 29%).

(*) S=sensitive; I=intermediate; R=resistant. Other identified pathogens included A anitratus, A lwoffi, Chlamydia spp, C psittaci, Comamonas acidovorans, E cloacae, Klebsiella oxytoca, N meningitidis, Streptococcus marcescens, Streptococcus group G, Streptococcus agalactiae, and Streptococcus pyogenes.

([dagger])Results obtained by serology. All other pathogens isolated from sputum.

Efficacy at the Test-of-Cure Assessment

Statistical equivalence of sparfloxacin 200 mg once daily (after initial 400 mg dose) to ofloxacin 400 mg every 12 h for 10 days was demonstrated for the primary efficacy parameter, overall success in the bacteriologically evaluable population, and for clinical success in the clinically evaluable population. The results of the all-treated population analyses were similar to those of the evaluable populations for overall and clinical success. For the all-treated population, bacteriologic success was equivalent between the treatment groups, and bacteriologic success in the bacteriologically evaluable population approached statistical equivalence; the 95% confidence interval was slightly below the equivalence range (Table 4).

(*) 95% CI=confidence interval of the difference in treatment response rates.

([dagger]) Excluding patients with indeterminate response.

([double dagger]) Primary efficacy variable.

Clinical Efficacy: Eighty-seven percent of the all-treated patients in the sparfloxacin group and 87.6% of the all-treated patients in the ofloxacin group were designated as clinical successes at the test-of-cure assessment. Of the clinically evaluable patients, 86.4% in the sparfloxacin group and 88.0% in the ofloxacin group were designated as clinical successes. The 95% confidence interval for the difference between the two treatment groups in clinical success rate supports equivalence of the two treatments for both the all-treated and evaluable populations.

Clinical success rates for both treatment groups did not vary as a function of gender, race, age, medical history (cigarette, alcohol, and illicit-drug use), recent hospitalization, underlying condition (COPD, or other lung disease, coronary artery disease, or diabetes mellitus), or bacteriologic status at entry (number of pathogens at study entry). Logistic regression analysis of clinical response indicated that, regardless of treatment group, patients who had a higher [FEV.sub.1]/FVC were more likely to have a response of cured or improved. Specifically, for each percentage point increase in the [FEV.sub.1]/FVC ratio, a patient's odds of having a clinical response of cured or improved increased by a factor of 1.03 (p=0.0025).

Bacteriologic Efficacy: Success rates for the most frequently isolated pathogens are shown in Table 5 for the all-treated patients with a baseline pathogen population and in Table 6 for the bacteriologically evaluable population. Results for each pathogen in the all-treated patients with a baseline pathogen population were similar to those of the evaluable population. Of note is that five of five S pneumoniae strains in the sparfloxacin group and three of three S pneumoniae strains in the ofloxacin group exhibiting decreased susceptibility to penicillin were eradicated or presumed eradicated after treatment with study medication. Bacteriologic success rates did not vary substantially for monomicrobic vs polymicrobic infection. In the sparfloxacin group, success rates were 88.8% for pathogens isolated in monomicrobic infections and 90.2% for pathogens isolated in polymicrobic infections. In the ofloxacin group, success rates were 94.4% for pathogens isolated in monomicrobic infections and 91.5% for pathogens isolated in polymicrobic infections.

Table 5--Bacteriologic Success Rates at Test-of-Cure by Most Frequently Isolated Baseline Pathogen (All-Treated with a Pathogen Population)(*)

(*) See Table 3 for explanation of abbreviations.

([dagger]) Results obtained by serology. All other pathogens isolated from sputum.

Table 6--Bacteriologic Success Rates at Test-of-Cure by Most Frequently Isolated Baseline Pathogen (Bacteriologically Evaluable Population)(*)

(*) See Table 3 for explanation of abbreviations.

([dagger]) Results obtained by serology. All other pathogens isolated from sputum.

Of the all-treated patients with a baseline pathogen population, bacteriologic success was demonstrated in 87.3% and 92.0% of patients who received sparfloxacin and ofloxacin, respectively. Approximately 87% of bacteriologically evaluable patients in the sparfloxacin group and 92.6% of bacteriologically evaluable patients in the ofloxacin group were considered bacteriologic successes. The 95% confidence interval for the difference between the two treatment groups in bacteriologic success rate supports equivalence of the two treatments for the all-treated patients with baseline pathogen populations; bacteriologic success approached equivalence in the bacteriologically evaluable populations. Logistic regression analyses for the bacteriologically evaluable population indicated that, regardless of treatment group, patients who had a higher [FEV.sub.1]/FVC were more likely to have a bacteriologic response of eradicated or presumed eradicated. For each percentage point increase in the [FEV.sub.1]/FVC ratio, a patient's odds of bacteriologic success increased by a factor of 1.03 (p=0.012). Regardless of treatment group, with each 1,000 cells per cubic millimeter increase in WBCs or in the presence of C pneumoniae infection, there was a decrease in the probability of bacteriologic success by a factor of 0.87 (p=0.026) and 0.35 (p=0.014), respectively. Of patients with a single Gram-negative aerobe, those who received ofloxacin were more likely to have bacteriologic success than those who received sparfloxacin (p=0.0009, odds ratio=9.32).

Overall Efficacy: The overall response rate (clinical response of cured or improved and bacteriologic response of eradicated or presumed eradicated) among the all-treated patients was 85.4% for the sparfloxacin group and 87.8% for the ofloxacin group at the test-of-cure assessment. Among bacteriologically evaluable patients, the overall response was 85.3% for the sparfloxacin group and 89.3% for the ofloxacin group at the test-of-cure assessment. The 95% confidence interval for the difference between the two treatment groups in overall response rate supports equivalence of the two treatments in both the all-treated and evaluable populations. Logistic regression analysis of overall response indicated that, regardless of treatment group, patients who had a higher [FEV.sub.1]/FVC were more likely to have an overall response of success; for each percentage point increase in the [FEV.sub.1]/FVC ratio, the patient's odds of having a clinical response of cured or improved increased by a factor of 1.03 (p=0.0006). Of patients with a single Gram-negative aerobe, those who received ofloxacin were more likely to have bacteriologic success than those who received sparfloxacin (p=0.007, odds ratio=3.43).

Efficacy at Follow-up

Efficacy results at follow-up are presented in Table 7. The clinical recurrence rate was comparable in the two treatment groups. Three patients who received sparfloxacin and three patients who received ofloxacin developed a superinfection. In the sparfloxacin group, superinfecting organisms were S pneumoniae (two) and Pseudomonas aeruginosa (two). In the ofloxacin group, superinfecting organisms were Acinetobacter lwoffi (one), Enterobacter cloacae (one), Stenotrophomonas maltophilia (one), S pneumoniae (one), and Klebsiella pneumoniae (one). Seven patients in the sparfloxacin group and nine patients in the ofloxacin group also developed colonization. The rate of bacterial relapse/reinfection was comparable in the two treatment groups. The pathogens most frequently associated with relapse in the sparfloxacin group were H influenzae and H parainfluenzae. H influenzae and H parainfluenzae were also the most common in the ofloxacin group. Other isolated pathogens associated with bacterial relapse were in catarrhalis, S pneumoniae, and K pneumoniae in both treatment groups, Acinetobacter anitratus in the sparfloxacin group, and Serratia marcescens, Neisseria meningitidis, and P aeruginosa in the ofloxacin group.

Safety

The percentage of patients reporting an adverse event was comparable between the two treatment groups. For all patients treated, the percentage of patients reporting an adverse event in the sparfloxacin group was 54.4% (215/395) as compared with 55.6% (224/403) in the ofloxacin group.

The percentage of patients reporting adverse events considered by the investigator to be related (possible or probable) to study medication was 27.8% in the sparfloxacin group and 28.5% in the ofloxacin group. A summary of the most frequently reported related adverse events (at least 2% of patients in a treatment group) is presented in Table 8. Overall, the most common related adverse events were insomnia, nausea, and photosensitivity reaction. The percentages of patients who experienced related nausea and insomnia were lower in the sparfloxacin group as compared with the ofloxacin group. The percentage of patients who experienced a related photosensitivity reaction was higher in the sparfloxacin group as compared with the ofloxacin group. A single sparfloxacin patient experienced a severe photosensitivity rash to the hands, neck, face, and left arm that was considered by the investigator to be probably related to study medication. The patient did not discontinue from the study and did recover. This adverse event occurred before the October 1993 protocol amendment that further enhanced instructions regarding sparfloxacin-related photosensitivity. The protocol amendment heightened warnings about sun exposure; the frequency of photosensitivity reaction in the sparfloxacin group was reduced slightly in patients enrolled after the protocol amendment (7.1%) as compared with patients enrolled prior to the amendment (7.8%).

(*) Possibly or probably related to study medication as assessed by the investigator.

Most adverse events reported for patients in both treatment groups were of mild or moderate severity. Twenty-seven patients, 12 (3.0%) in the sparfloxacin group and 15 (3.7%) in the ofloxacin group, experienced a related adverse event that led to discontinuation from the study. In the sparfloxacin group, related adverse events leading to discontinuation were dyspnea, asthma, prolonged QT interval (one patient; mild severity), abdominal pain, photosensitivity (one patient; moderate severity), tremor, pneumonia, headache, dizziness, asthenia, angioedema, pruritis, atrial fibrillation, cyanosis, agitation, and hallucination. Related adverse events leading to discontinuation in the ofloxacin group were rash, diarrhea, nausea, dyspepsia, vomiting, insomnia, agitation, abnormal dreams, hyperkinesia, and urticaria. No adverse event leading to discontinuation occurred at an incidence of [is greater than or equal to] 2% in either treatment group. One patient in the ofloxacin group died due to arteriosclerotic and hypertensive cardiovascular disease that was considered remotely related to ofloxacin. There was no death in the sparfloxacin group.

Changes in vital signs, physical examination, and laboratory parameters were similar in the two groups. The frequency and type of adverse laboratory events were comparable for the two treatment groups; the most common was elevated WBC count (sparfloxacin, 1.8%; ofloxacin, 3.2%). Sparfloxacin administration appeared to have no effect on serum digoxin or serum theophylline concentrations; the same was true for ofloxacin, although the small number of patients who received digoxin precluded treatment group comparison.

QTc interval

A greater mean change from baseline to maximum on-treatment value for QTc interval was observed in the sparfloxacin group (0.012 s) as compared with the ofloxacin group (0.001 s) (p [is less than] 0.001; 95% confidence interval, 0.007, 0.014). There were no cardiovascular adverse events attributed to the increase in QTc interval. No significant differences were observed between the two treatment groups for the mean change from baseline to maximum off-treatment values for QTc interval.

DISCUSSION

Administration of once-daily sparfloxacin for 10 days was shown to be effective for the treatment of ABECB and equivalent to ofloxacin administered twice daily for 10 days. Equivalence was demonstrated for the primary efficacy parameter of overall response in the bacteriologically evaluable patient population, as well as for clinical success in the clinically evaluable population. Bacteriologic success in the bacteriologically evaluable population approached statistical equivalence. The results of all-treated patient population analyses were similar to those of the evaluable patient populations; additionally, statistical equivalence was demonstrated for bacteriologic success in this population.

Emerging bacterial resistance, particularly in strains of S pneumoniae, is a concern in the treatment of patients with ABECB. Currently, approximately 25% of S pneumoniae strains have intermediate or high level resistance to penicillin, with a higher rate anticipated by the next decade,[12] All [Beta]-lactam antimicrobials have decreased activity against this pathogen. In addition, a recent survey revealed the emergence of S pneumoniae resistance to chloramphenicol, tetracycline, trimethoprimsulfamethoxazole, the macrolides, and ofloxacin.[12,25] Inconsistent in vitro activity against S pneumoniae by ciprofloxacin, ofloxacin, lomefloxacin, enoxacin, pefloxacin, and fleroxacin suggests that they may not always be clinically effective against this pathogen.[7,10,26]

Increased rates of [Beta]-lactamase-mediated resistance have also been observed in other causative pathogens, such as H influenzae (approximately 45 to 50%) and M catarrhalis (approximately 100% of clinical isolates). Sparfloxacin has demonstrated superior in vitro activity against S pneumoniae, S aureus, H influenzae, M catarrhalis, C pneumoniae, and Mycoplasma pneumoniae when compared to ofloxacin and other available fluoroquinolones.[7]

In the current study, pathogen eradication rates for sparfloxacin were consistent with in vitro activity. The eradication rate of pathogens at test-of-cure was 86.9% for sparfloxacin and 92.6% for ofloxacin in bacteriologically evaluable patients. The eradication rate for S pneumoniae (88.2% for sparfloxacin and 90.9% for ofloxacin) was comparable in the two treatment groups. All five strains of S pneumoniae with decreased susceptibility to penicillin were eradicated or presumed eradicated by sparfloxacin treatment. The bacteriologic success rate for S pneumoniae in the sparfloxacin group is similar to that documented in other studies (approximately 90 to 96%),[27-29] but in the ofloxacin group, it was somewhat higher in this study than previously reported (approximately 40 to 79%).[30-32] Eradication rates for commonly occurring Gram-negative organisms in patients with ABECB (H parainfluenzae, H influenzae, and M catarrhalis) were also similar in the two treatment groups.

There was a high incidence of positive serology for C pneumoniae from patients in both treatment groups; C pneumoniae was the sole etiologic organism for 14 patients in the sparfloxacin group and 13 patients in the ofloxacin group. A similar bacteriologic success rate for C pneumoniae was observed in the sparfloxacin group (81.4%) and the ofloxacin group (89.7%). In general, C pneumoniae infection was rarely diagnosed in most previous studies of patients with ABECB since serology for atypical pathogens was usually not performed. The current study suggests that C pneumoniae could be a pathogen in ABECB, paralleling the relatively high prevalence of C pneumoniae in community-acquired pneumonia reported in more recent studies in which serologic methods for diagnosis were employed,[8,29,33-36]

Sparfloxacin was generally well tolerated. Adverse events were reported in 54.4% of patients in the sparfloxacin group and 55.6% of patients in the ofloxacin group. Approximately one half of the adverse events reported were considered by the investigator to be possibly or probably related to study medication. A small percentage of patients experienced adverse events that led to discontinuation from the study by the investigator. Approximately one half of the adverse events that led to discontinuation in either treatment group were considered related to study medication by the investigator (3.0% sparfloxacin; 3.7% ofloxacin). Sparfloxacin had significantly fewer adverse events associated with the digestive and nervous systems, particularly nausea (4.3% sparfloxacin; 10.2% ofloxacin) and insomnia (1.5% sparfloxacin; 14.9% ofloxacin), than ofloxacin.

Photosensitivity reaction was observed in both treatment groups, but at a higher frequency in the sparfloxacin group (8.1% sparfloxacin; 0.7% ofloxacin). All photosensitivity reactions in patients who received sparfloxacin were of mild or moderate severity, except for one patient treated before the protocol amendment that further enhanced instructions regarding sparfloxacin-related photosensitivity (see "Results" section). Photosensitivity dermatitis, a class effect of quinolones, is primarily induced by UV-A present in sunlight, artificial sunlamps, etc.[37]

Modest QTc prolongation was observed in the sparfloxacin group but was not associated with any cardiovascular adverse event. The QTc effect of sparfloxacin has been described previously.[21] The major consideration in prescribing is avoidance of coprescription of sparfloxacin with other QTc-prolonging agents, especially antiarrhythmics such as amiodarone.

An advantage of sparfloxacin therapy is the once-daily dosing regimen compared with the twice-daily dosing regimen of ofloxacin. Sparfloxacin, which has a long plasma elimination half-life (approximately 20 h) allowing for maintenance of plasma and tissue concentrations for the entire dosing interval, is widely distributed into most body fluids and tissues, even with once-daily administration.[38] Concentrations of sparfloxacin higher than those in serum (0.6 [+ or -] 0.2 mg/L) have been measured in bronchial mucosa (1.6 [+ or -] 0.5 mg/kg), pulmonary epithelial lining fluid (10.2 [+ or -] 4.5 mg/L), and alveolar macrophages (36.3 [+ or -] 10.9 mg/kg), which have persisted at these levels for at least 24 h.[13,14] The distribution of sparfloxacin in alveolar macrophages is of potential importance because atypical pathogens, such as C pneumoniae, multiply within these cells. The concentration of sparfloxacin in alveolar macrophages has been shown to greatly exceed the concentrations necessary to inhibit atypical pathogens.[13] Fluoroquinolones accumulate throughout alveolar macrophages, in contrast to macrolides, which have been shown to accumulate in the lysosomes of alveolar macrophages.[39] Aminoglycosides and [Beta]-lactams do not accumulate within alveolar macrophages; rather they are slowly endocytosed into lysosomes where the low pH diminishes their activity.[39] The relative penetration of sparfloxacin into bronchial mucosa is approximately twice that of ciprofloxacin, temafloxacin, and rufloxacin; penetration of ofloxacin and levofloxacin into healthy and diseased lung tissue is similar to sparfloxacin penetration into bronchial mucosa.[14,26,40,41] Relative sparfloxacin penetration into pulmonary epithelial lining fluid is four to six times greater than ciprofloxacin, lomefloxacin, and temafloxacin, and into alveolar macrophages it is two to four times greater than ofloxacin, ciprofloxacin, lomefloxacin, temafloxacin, and rufloxacin.[14,42]

In conclusion, oral treatment with sparfloxacin is effective, and comparable to ofloxacin, for the treatment of ABECB. Sparfloxacin is effective against the major causative pathogens, including H influenzae, S pneumoniae, M catarrhalis, C pneumoniae, H parainfluenzae, K pneumoniae, E cloacae, and S aureus. In addition to its efficacy, the advantages of a once-a-day regimen and high bronchial penetration make sparfloxacin an alternative for the treatment of ABECB in those patients who are not at risk for photosensitivity reactions or cardiovascular complications of a QTc interval increase.

ACKNOWLEDGMENTS: We thank D. Magner for statistical assistance, D. Matour for the preparation of the manuscript, and L. Lamb for secretarial support.

REFERENCES

[1] American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am J Respir Crit Care Med 1995; 152:S77-120

[2] Woolcock AJ. Epidemiology of chronic airways disease. Chest 1989; 96(suppl):302S-06S

[3] Chodosh S. Treatment of acute bacterial exacerbations of chronic bronchitis: state of the art. Am J Med 1991; 91(suppl 6A):87-92

[4] Reynolds HY. Chronic bronchitis and acute infectious exacerbations. In: Mandell GL, Douglas RG Jr, Bennett JE, eds. Principles and practice of infectious diseases. 3rd ed. New York: John Wiley, 1990; 529-31

[5] Rodnick JE, Gude JK. The use of antibiotics in acute bronchitis and acute bacterial exacerbations of chronic bronchitis. West J Med 1988; 149:347-51

[6] Ellner JJ. Management of acute and chronic respiratory tract infections. Am J Med 1988; 85(suppl 3A):2-5

[7] Baquero F, Canton R. In vitro activity of sparfloxacin in comparison with currently available antimicrobials against respiratory tract pathogens. J Antimicrob Chemother 1996; 37(suppl A):1-18

[8] Isada CM. Pro: antibiotics for chronic bronchitis with exacerbations. Semin Respir Infect 1993; 8:243-53

[9] Ball P, Tilltoson G, Wilson R. Chemotherapy for chronic bronchitis: controversies. Presse Med 1995; 24:189-94

[10] Cunha B. Newer antimicrobial agents for use in acute bacterial exacerbations in chronic bronchitis. Adv Ther 1996; 13:313-23

[11] Canton R, Peman J, Jimenez, et al. In vitro activity of sparfloxacin compared with those of five other quinolones. Antimicrob Agents Chemother 1992; 36:558-65

[12] Doern GV. Trends in antimicrobial susceptibility of bacterial pathogens of the respiratory tract. Am J Med 1995; 99(suppl 6B):3-7

[13] Honeybourne D, Greaves I, Baldwin DR. The concentration of sparfloxacin in lung tissues after single and multiple doses. Int J Antimicrob Agents 1994; 4:151-55

[14] Wise R, Honeybourne D. A review of the penetration of sparfloxacin into the lower respiratory tract and sinuses. J Antimicrob Chemother 1996; 37(suppl):57-63

[15] Nakashima M, Uematsu T, Takiguchi Y. Phase I study of pyridone carboxylic acid antibacterial agent, sparfloxacin. Rinsho-Yakuri 1991; 7:1639-84

[16] Mizuki Y, Fujiwara l, Yamaguchi T. Pharmacokinetic interactions related to the chemical structures of fluoroquinolones. J Antimicrob Chemother 1996; 37(suppl):A41-55

[17] Johnson RD, Wilson J, Talbot G, et al. The effect of sparfloxacin on the steady-state pharmacokinetics of digoxin in healthy male volunteers. Pharm Res 1994; 11(10 suppl):S429

[18] Jonkman J, Crijns H, Surtees M. Phase I, open study of the effect of sparfloxacin (400 mg for 1 day followed by 200 mg for a further 6 days) on warfarin pharmacodynamics in 10 healthy volunteers. (Protocol AT-113). July 2, 1993; unpublished data; Rhone-Poulenc Rorer

[19] Gries JM, Honorato J, Taburet AM, et al. Cimetidine does not alter sparfloxacin pharmacokinetics. Int J Clin Pharmacol Ther 1995; 33:585-87

[20] Shimada J, Saito A, Shiba K, et al. Pharmacokinetics and clinical studies on sparfloxacin. Chemotherapy 1991; 39: 234-44

[21] Jaillon P, Morganroth J, Brumpt l, et al. Overview of electrocardiographic and cardiovascular safety data for sparfloxacin. J Antimicrob Chemother 1996; 37(suppl): 161-67

[22] Woods GL, Washington JA. The clinician and the microbiology laboratory. In: Mandell GL, Douglas RG Jr, Bennett JE, eds. Principles and practice of infectious diseases, 4th ed. New York: Churchill Livingstone, 1995; 169-99

[23] Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. 3rd ed; approved standard. NCCLS Document M7-A3 (vol 13), No. 25, 1993

[24] Performance standards for antimicrobial disk susceptibility tests. 5th ed; approved standard. NCCLS Document M2-A5 (vol 13), No. 24, 1993

[25] Butler JC, Hofmann J, Cetron MS, et al. Thc continued emergence of drug-resistant Streptococcus pneumoniae in the United States: an update from the Centers for Disease Control and Prevention's Pneumococcal Sentinel Surveillance System. J Infect Dis 1996; 174:986-93

[26] Chodosh S. Use of quinolones for the treatment of acute bacterial exacerbations of chronic bronchitis. Am J Med 1991; 91(suppl 6A):93-100

[27] Hara K, Kobayashi H, Tanimoto H, et al. A multicenter double-blind comparative study of sparfloxacin and ofloxacin in the treatment of chronic respiratory tract infections [abstract 878]. 31st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, Sept 29-Oct 2, 1991; 245

[28] Soejima R, Hara K. Clinical response of sparfloxacin in respiratory tract infections. Drugs 1993; 45(suppl 3):405-06

[29] Ortqvist A, Valtonen M, Cars O, et al. Oral empiric treatment of community-acquired pneumonia: a multicenter, double-blind, randomized study comparing sparfloxacin with roxithromycin. Chest 1996; 110:1499-1506

[30] Saito A, Katsu M, Saito A, et al. Ofloxacin in respiratory tract infections: a review of the results of clinical trials in Japan. Drugs 1987; 34(suppl 1):83-89

[31] Grassi C, Grassi G, Mangiarotti P. Clinical efficacy of ofloxacin in lower respiratory tract infections. Drugs 1987; 34(suppl 1):80-82

[32] Rademaker CMA, Sips AP, Beumer HM, et al. A double-blind comparison of low dose ofloxacin and amoxycillin/ clavulanic acid in acute exacerbations of chronic bronchitis. J Antimicrob Chemother 1990; 26(suppl D):75-81

[33] Soejima R, Niki Y, Kishomoto T, et al. Antichlamydial activities of newly developed fluoroquinolones and their potential future role in the treatment of chlamydial respiratory infections. Drugs 1995; 49(suppl 2):257-59

[34] Carbon C. Quinolones in the treatment of lower respiratory tract infections in adult patients. Drugs 1993; 45(suppl 3): 91-97

[35] Isada CM, Stoller JK. Chronic bronchitis: role of antibiotics. In: Niederman MS, Sarosi GA, Glassroth J, eds. Respiratory infections. Philadelphia: WB Saunders, 1994; 621-31

[36] Marrie TJ, Peeling RW, Fine MJ. Ambulatory patients with community-acquired pneumonia: the frequency of atypical agents and clinical course. Am J Med 1996; 101:508-15

[37] Lietman PS. Fluoroquinolone toxicities: an update. Drugs 1995; 42(suppl 2):159-63

[38] Shimada J, Nogita T, Ishibashi Y. Clinical pharmacokinetics of sparfloxacin. Clin Pharmacokinet 1993; 25:358-69

[39] Tulkins P. Intracellular distribution and activity of antibiotics. Eur J Clin Microbiol Infect Dis 1991; 10:100-06

[40] Couraud L, Fourtillan JB, Saux MC, et al. Diffusion of ofloxacin into human lung tissue. Drugs 1987; 34(suppl 1):37-38

[41] Levofloxacin package insert. Raritan, NJ: Ortho-McNeil Pharmaceutical, 1997

[42] Perea EJ. Ofloxacin concentrations in tissues involved in respiratory tract infections. J Antimicrob Chemother 1990; 26(suppl D):55-60

APPENDIX

Members of the Multicenter ABECB Study Group were as follows: Mohamed Al-Ibrahim, MD, Baltimore; Huong Bach, MD, Sacramento, Calif; Charles Banov, MD, Charleston, SC; Wayne Beauford, MD, High Point, NC; George Bensch, MD, Stockton, Calif; Ross Black II, MD, Cuyohoga Falls, Ohio; Steven Bowman, MD, Clearwater, Fla; Robert Broker, MD, Greer, SC; Walter Brzezinski, MD, Charleston, SC; Christopher Chappel, MD, Kissimmee, Fla; Thomas Chick, MD, Albuquerque; Sanford Chodosh, MD, Boston; Bya Clecner, MD, St. Jerome, Quebec, Canada; Richard Coalson, MD, Dayton, Ohio; David Daniel, MD, Wenatchee, Wash; Diana Dark, MD, Kansas City, Mo; C. Andrew DeAbate, MD, Metaire, La; Margaret Drehobl, MD, San Diego; Vilma Drelichman, MD, Southfield, Mich; N. Fellers, MD, Greely, Colo; Robert Fiddes, MD, Whittier, Calif; Jonathan Flescher, MD, Raleigh, NC; Larry Gilderman, MD, Pembroke, Fla; Gurmitt Gill, MD, Garden City, NY; Gregory Gordon, MD, Ambler, Pa; Richard Greenberg, MD, Lexington, Ky; Charles Hanna, MD, Spartansburg, SC; H. Freeman Harris, MD, Lake Oswego, Ore; Thomas Hauch, MD, Charlotte, NC; Dan Henry, MD, Salt Lake City, Utah; Ronald Hubbard, MI), Shreveport, La; Jonathan Ilowite, MD, Mineola, NY; Kirk Jacobson, MD, Eugene, Ore; Areal Jubran, MD, Hines, Ill; Rashid Khairi, MD, Indianapolis; C.M. Khurana, MD, Evanston, Ill; Harold Kimmerling, MD, Dallas; Charles Kish, DO, Lansdale, Pa; Jacques LaForge, MD, Quebec, Canada; J. Lampasso, MD, Buffalo, NY; Leonard Lazarus, MD, La Jolla, Calif; Jack LcFrock, MD, Sarasota, Fla; J. Lewis, RN, Salt Lake City, Utah; Benjamin Lipsky, MD, Seattle; Nola Mahoney, DO, Vorhees, NJ; J. Foster Manning, MD, Portland, Ore; Dennis McCluskey, MD, Mogadore, Ohio; Phillip McElvaine, MD, El Paso, Tex; David Miller, DO, Morrisville, Pa; Paul Montner, MD, Albequerque; Javier Morales-Ramirez, MD, Candado, Santurce, PR; Michael Nelson, MD, Kansas City, Mo; Thomas Nolen, MD, Columbiana, Ala; John Ondrejicka, MD, Jacksonville Beach, Fla; Michael Opipari, DO, Warren, Mich; Meenakshi Patel, MD, Kettering, Ohio; Russel Platt, MD, Cuyahoga Falls, Ohio; Julio Ramirez, MD, Louisville, Ky; Alan Rogers, MD, Tucson, Ariz; Clint Sanford, MD, Portland, Ore; Edwin Schachter, MD, New York; John Schoenberger, MD, Scottsdale, Ariz; M. Shankman, MD, Watertown, Mass; and Selwyn Spangenthal, MD, Charlotte, NC; Steven Springmeyer, MD, Seattle; William Stein, MD, Rochester, NY; Michael Taylor, MD, Richmond, Va; Raymond Tidman, MD, Smyrna, Ga; and Kumjad Unnoppet, MD, Birmingham, Ala.

C. Andrew DeAbate, MD; Dan Henry, MD; George Bensch, MD; Amal Jubran, MD; Sanford Chodosh, MD; Lisa Harper, BS, MT, ASCP; Diane Tipping, MS; and George H. Talbot, MD; For the Sparfloxacin Multicenter ABECB Study Group([dagger])

(*) From the Medical Research Center (Dr. DeAbate), New Orleans; Foothill Family Clinic (Dr. Henry), Salt Lake City, Utah; Allergy Immunology Asthma, Inc (Dr. Bensch), Stockton, Calif; Edward Hines, Jr. Veterans Affairs (VA) Hospital/Loyola University of Chicago (Dr. Jubran), Hines, Ill; VA Outpatient Clinic (Dr. Chodosh), Boston; and Rhone-Poulenc Rorer Research and Development (Mss. Harper and Tipping and Dr. Talbot), Collegeville, Pa. The study was presented, in part, at the 36th Interscience Conference on Antimicrobial Agents and Chemotherapy, New Orleans, September 15-18, 1996.

([dagger]) Members of the Sparfloxacin Multicenter ABECB Study Group are listed in the Appendix.

This was a multicenter study initiated and supported by RhonePoulenc Rorer. All investigator sites received a fee per included patient to cover the expenses of the procedures. Ms. Harper, Ms. Tipping, and Dr. Talbot are currently employees of Rhone-Poulenc Rorer. None of the other authors has any financial interest in Rhone-Poulenc Rorer.

Manuscript received May 28, 1997; revision accepted December 2, 1997.

COPYRIGHT 1998 American College of Chest Physicians
COPYRIGHT 2000 Gale Group