Abstract
The purpose of this study was to evaluate intravenous (i.v.) azithromycin followed by oral azithromycin as a monotherapeutic regimen for community-acquired pneumonia (CAP). Two trials of i.v. azithromycin used as initial monotherapy in hospitalized CAP patients are summarized. Clinical efficacy is reported from an open-label randomized trial of azithromycin compared to cefuroxime with or without erythromycin. Bacteriologic and clinical efficacy results are also presented from a noncomparative trial of i.v. azithromycin that was designed to give additional clinical experience with a larger number of pathogens. Azithromycin was administered to 414 patients: 202 and 212 in the comparative and noncomparative trials, respectively. The comparator regimen was used as treatment for 201 patients; 105 were treated with cefuroxime alone and 96 were given cefuroxime plus erythromycin. In the comparative trial, clinical outcome data were available for 268 evaluable patients with confirmed CAP at the 10- to 14-day visit, with 106 (77%) of the azithromycin patients cured or improved and 97 (74%) of the comparator patients cured or improved. Mean i.v. treatment duration and mean total treatment duration (i.v. and oral) for the clinically evaluable patients were significantly (P < 0.05) shorter for the azithromycin group (3.6 days for the i.v. group and 8.6 days for the i.v. and oral group) than for the evaluable patients given cefuroxime plus erythromycin (4.0 days for the i.v. group and 10.3 days for the i.v. and oral group). The present comparative study demonstrates that initial therapy with i.v. azithromycin for hospitalized patients with CAP is associated with fewer side effects and is equal in efficacy to a 1993 American Thoracic Society-suggested regimen of cefuroxime plus erythromycin when the erythromycin is deemed necessary by clinicians.
Community-acquired pneumonia (CAP) is a serious health care problem in the United States, with an incidence estimated at 4 million cases annually (5). Approximately 600,000 hospitalizations for CAP occur per year at an annual cost of $23 billion (20, 21), with mortality rates in the most severely affected patients approaching 40% (11). Causative pathogens are identified in fewer than 50% of cases in good microbiology studies, and diagnostic tests are not often performed in outpatient settings (6, 9). Accordingly, empiric antibiotic regimens are commonly chosen on the basis of the results of clinical trials. Empiric therapy of CAP is usually directed against Streptococcus pneumoniae and Haemophilus influenzae, the most common pathogens identified in hospitalized adults with CAP. Coverage of atypical pathogens such as Mycoplasma pneumoniae, Chlamydia pneumoniae, and Legionella pneumophila also has been recommended.
The 1993 American Thoracic Society (ATS) recommendations for empiric therapy of CAP are based on an understanding of the microbial etiology of CAP and the type of pathogen likely to be implicated with specific patient types and pneumonia severity (24). ATS recommends that hospitalized patients with CAP receive therapy with an agent active against the typical organisms such as a beta-lactam–beta-lactamase inhibitor or an expanded- or broad-spectrum cephalosporin plus a macrolide when the atypical pathogens Legionella, Chlamydia, and Mycoplasma are suspected. A combination of two drug classes provides coverage for the atypical pneumonia pathogens, which are not susceptible in vitro to beta-lactams, and for organisms such as S. pneumoniae and H. influenzae.
Azithromycin is a macrolide antibacterial agent with a chemical designation as an azalide because of a methyl-substituted nitrogen atom in the lactone ring. This substitution confers upon azithromycin unique pharmacokinetic and antimicrobial properties (7, 19, 26) that have resulted in improved activity against a variety of pathogens that cause pneumonia (1–3, 8, 29). This paper summarizes two trials of initial therapy with intravenous (i.v.) azithromycin used as a single agent to treat hospitalized CAP patients. Clinical efficacy is reported from a randomized trial of azithromycin compared to cefuroxime with or without erythromycin. Bacteriologic and clinical efficacy results are also presented from a noncomparative trial of i.v. azithromycin that was designed to give additional clinical experience with a larger number of pathogens.
MATERIALS AND METHODS
Patients.
Male and female patients who were at least 16 years of age and who had clinical and radiographic documentation of CAP that necessitated hospitalization and initial i.v. antibiotic therapy were eligible for the study. Radiographic diagnosis required the appearance within 48 h of the beginning of therapy of a new pulmonary infiltrate(s) that was not attributable to another etiology. A clinical diagnosis of CAP was supported by at least one of the following: physical signs or symptoms (chest pain, cough, rales, rhonchi, and/or signs of consolidation) suggestive of bacterial pneumonia, an oral temperature of >38°C (100.4°F), or leukocytosis (leukocyte count, >10,000/mm3, or >15% band forms).
The following conditions or situations were exclusion criteria for the study: pregnancy or lactation in women; known hypersensitivity or intolerance to macrolide antibiotics (and β-lactams in the comparative study); active peptic ulcers; gastrectomy or other conditions that affect drug absorption; known drug or alcohol dependence; evidence or history of significant hematologic, renal, cardiovascular, or hepatic disease; AIDS or human immunodeficiency virus infection; metastatic tumor; chronic bronchitis, bronchiectasis or chronic obstructive pulmonary disease without acute pneumonia; evidence of infection with gram-negative organisms known to be resistant to the study medications; aspiration or postobstructive pneumonia; suspected septic shock; hospitalization within the past 14 days; rapidly progressive underlying disease that precluded evaluation of therapy; cystic fibrosis; coma; need for mechanical ventilation; current chemotherapy or immunosuppressive therapy; infection at enrollment potentially requiring treatment with antibiotics other than the study drugs; treatment with any systemic antibiotic within 24 h of enrollment; treatment with any other investigational drug within the past 4 weeks; and use of terfenadine or loratadine concurrently with the study drugs or within the preceding 14 days or the use of astemizole concomitantly or within the preceding 30 days. Written informed consent was obtained from all patients, and the institutional review board at each participating medical center approved the study.
Treatment.
Azithromycin was administered to 414 hospitalized adults with CAP who participated in the two phase III trials described here. The azithromycin regimen in each study consisted of 500 mg given i.v. daily for 2 to 5 days, followed by 500 mg given orally once daily for a total of 7 to 10 days of therapy. The comparative study was a multicenter, parallel-group, randomized, open-label trial of 202 patients receiving azithromycin compared to 201 patients treated with cefuroxime at 750 mg i.v. every 8 h for 2 to 7 days, followed by cefuroxime axetil at 500 mg orally every 12 h for a total of 7 to 10 days of therapy. For cefuroxime recipients with suspected pneumonia due to Mycoplasma, Legionella, or Chlamydia, erythromycin given at 500 mg orally four times a day or erythromycin lactobionate given at 500 mg to 1 g i.v. every 6 h could be added for up to 21 days. The noncomparative trial was a multicenter, open-label study of i.v. azithromycin with sequential enrollment (n = 212) and the same dosing regimen used for the comparative trial. In both trials, investigators made decisions about the time of switch to oral therapy and the total treatment duration on the basis of the patient's clinical response.
Clinical assessments.
Clinical evaluation of response was based on resolution or improvement of clinical and laboratory signs of infection, such as defervescence, normalization of leukocytosis, disappearance of or diminution in the level of purulent sputum production, stabilization of general physical condition, and radiographic resolution of lung infiltrates. Patients were assessed at the baseline, day 3, every 5 to 7 days during treatment, 10 to 14 days posttherapy, and 4 to 6 weeks posttherapy. A chest X ray was obtained at the baseline and at the main clinical evaluation time points. The location and extent of lung involvement (e.g., segmental or lobar) and the presence of pleural effusion were recorded.
Response criteria.
The primary efficacy measure was clinical outcome, which was determined 10 to 14 days posttherapy. The clinical responses at this time were classified as follows: cure, complete resolution of all signs and symptoms of pneumonia and improvement or resolution of chest X-ray findings; improvement, improvement or resolution of all radiographic findings with incomplete resolution of all signs and symptoms of pneumonia; and failure, persistence or progression of signs and symptoms after 3 days of therapy, development of new findings consistent with active infection, persistence or progression of radiographic findings, death due to pneumonia, or inability to complete the study because of pneumonia-related adverse events. At 4 to 6 weeks posttherapy, response was classified as cure or failure.
Sample size considerations.
Accounting for an initial projection of a 20 to 25% clinically nonevaluable rate, the comparative study protocol called for enrollment of a sufficient number of patients into the study to obtain 300 clinically evaluable patients for analysis purposes. Three hundred evaluable patients would yield at least an 80% power to detect a difference between treatment groups of 15% with respect to clinical cure rates at the 0.05 level of significance, assuming an expected cure rate of 75% for the comparator group.
Patients evaluable for efficacy.
In the comparative trial, inferences about clinical cure rates were based primarily on the group of clinically evaluable patients. A patient evaluable for clinical outcome was defined as a patient who had confirmed diagnosis of pneumonia, whose clinical outcome was determined at either the 10- to 14-day or the 4- to 6-week visit, and who had received at least 5 days of treatment. Patients with clinical failures were considered evaluable after only 3 days of treatment regardless of treatment group. An intent-to-treat analysis was also performed for patients who received one dose of medication and for whom clinical evaluations were performed at either 10 to 14 days or 4 to 6 weeks.
Microbiologic assessments.
Cultures of respiratory tract secretions were done at the baseline and, if samples were obtainable, on day 3, every 5 to 7 days during treatment, and 10 to 14 days and 4 to 6 weeks posttherapy, as well as when clinically indicated. Only adequate sputum specimens as determined by Gram staining were cultured. The absence of adequate sputum or other appropriate culturable material was considered equivalent to a negative result. Blood cultures, if initially positive, were repeated daily until two consecutive sterile cultures were obtained.
Routine cultures of respiratory specimens were performed in local laboratories as well as the reference laboratory of Arthur Barry, Clinical Microbiology Institute, Tualatin, Oreg. Evidence for M. pneumoniae infection was based on sputum culture or by serologic determination of an immunoglobulin M (IgM) titer of ≥10 or a fourfold rise in IgG titer from the acute- to the convalescent-phase sera in the laboratory of Gail Cassell, University of Alabama, Birmingham. C. pneumoniae infection was identified serologically by the presence of an IgM titer of ≥20, a single IgG titer of >512, or a fourfold rise in the IgG titer from the acute- to the convalescent-phase sera by a microimmunofluorescence assay in Thomas Grayston's laboratory at the University of Washington. L. pneumophila infection was confirmed by culture or a fourfold rise in the serum IgG titer from the acute- to the convalescent-phase sera in Paul Edelstein's laboratory at the University of Pennsylvania.
The microbiologic response at the last available follow-up was used as a secondary measure of treatment efficacy. Patients were included in the summary of microbiologic outcome if they were clinically evaluable, if they had at least one baseline pathogen susceptible to a study drug(s), and if follow-up cultures were available (or the patients were not producing sputum or their clinical condition otherwise obviated microbiologic investigation).
Safety assessments.
All observed or volunteered adverse events were recorded, including intercurrent illnesses and exacerbations of preexisting illnesses. Investigator assessments of severity, seriousness, and causality were also documented. Hematologic and serum chemistries and urinalyses were performed at the baseline, day 3, every 5 to 7 days during treatment, and 10 to 14 days and 4 to 6 weeks posttherapy.
Severity assessments.
Pneumonia Severity Index (PSI) scores were calculated for each patient by using information collected during study participation (12). The data were used to calculate the PSI retrospectively since the methodology for calculating the PSI was published after the two trials described in this paper were completed.
RESULTS
A total of 615 hospitalized patients with CAP were treated in the two trials. Azithromycin was administered to 414 patients: 202 and 212 in the comparative and noncomparative trials, respectively. The comparator regimen was used to treat 201 patients; 105 were treated with cefuroxime alone, and 96 were given cefuroxime plus erythromycin.
The a priori definition of a clinically evaluable patient included one who had received therapy for 5 or more days, one who met the inclusion or exclusion criteria for the trial, and one who had had a clinical evaluation either at the 10- to 14-day period or at the 4- to 6-week period. Criteria were satisfied for 268 patients in the comparative trial at 10 to 14 days. The five most common reasons for exclusion from analysis were similar in each arm of the trial and included early withdrawal for protocol violations or other reasons (18 in the azithromycin group, 23 in the comparator group), noncompliance with dosing (5 in the azithromycin group, 16 in the comparator group), loss to follow-up (6 in the azithromycin group, 7 in the comparator group), lack of receipt of study drug (4 in the azithromycin group, 6 in the comparator group), and an inappropriate primary diagnosis (7 in the azithromycin group, 3 in the comparator group).
Clinical response.
The characteristics of the patients in these studies are summarized in Table 1. Baseline demographic characteristics and specific factors associated with a complicated clinical course were comparable for the patients in the azithromycin and comparator arms. The PSI classes and mean scores were similar in both arms of the randomized trial and in the azithromycin noncomparative open-label trial.
TABLE 1.
Baseline characteristics of patientsa
| Characteristic | Comparative trial
|
Noncomparative trial, azithromycin (n = 212) | All azithromycin (n = 414) | |
|---|---|---|---|---|
| Azithromycin (n = 202) | Cefuroxime ± erythromycin (n = 201) | |||
| Sex | ||||
| Male | 125 (62) | 127 (63) | 136 (64) | 261 (63) |
| Female | 77 (38) | 74 (37) | 76 (36) | 153 (37) |
| Age (yr) | ||||
| 16–44 | 51 (25) | 51 (25) | 60 (28) | 111 (27) |
| 45–64 | 59 (29) | 58 (29) | 56 (26) | 115 (28) |
| >64 | 92 (46) | 92 (46) | 96 (45) | 188 (45) |
| Comorbidity | ||||
| Tobacco use | 68 (34) | 65 (32) | 76 (36) | 144 (35) |
| Chronic airway obstruction | 65 (32) | 74 (37) | 51 (24) | 116 (28) |
| Diabetes mellitus | 39 (19) | 40 (20) | 35 (17) | 74 (18) |
| PSI scoresb and PaO2c for clinically evaluable patients | ||||
| PSI class | ||||
| 1 | 29 (21) | 22 (17) | 36 (21) | |
| 2 | 36 (26) | 37 (28) | 50 (28) | |
| 3 | 44 (32) | 44 (34) | 46 (26) | |
| 4 | 26 (19) | 27 (21) | 41 (23) | |
| 5 | 2 (2) | 1 (1) | 3 (2) | |
| Total no. of evaluable patients | 137 | 131 | 176 | |
| PSI (mean) | 70 | 72 | 71 | |
| PaO2, <60 mm Hgd | 41/97 (42) | 38/89 (43) | 23/46 (50) | |
Unless indicated otherwise, data are number (percent) of patients.
Calculation based on Fine et al. (12).
PaO2, partial arterial oxygen pressure.
Score and predictors are the number of patients with signs, symptoms, or laboratory findings/number of evaluable patients (percent).
Table 2 displays the clinical responses at the early and late follow-up time periods for the patients evaluable for efficacy. In the comparative trial, clinical outcome data were available for 268 evaluable patients with confirmed CAP at the 10- to 14-day visit, with 106 of 137 (77%) azithromycin patients cured or improved and 97 of 131 (74%) of the comparator patients cured or improved. Results were similar at the 4- to 6-week posttherapy visit in the evaluable patient analysis. In the intent-to-treat analysis, 322 patients received at least one dose of medication, and demonstrated cure or improvement was found for 125 of 163 (77%) azithromycin-treated patients and 124 of 159 (78%) comparator patients at days 10 to 14 and for 111 of 150 (74%) azithromycin-treated and 110 of 146 (75%) comparator patients at 4 to 6 weeks. The mean duration of i.v. therapy and mean duration of total therapy in clinically evaluable patients were significantly shorter (P < 0.05) for the azithromycin group (3.6 days for the i.v. group and 8.6 days for the i.v. and oral group) than for evaluable patients given cefuroxime plus erythromycin (4.0 days for the i.v. group and 10.3 days for the i.v. and oral group).
TABLE 2.
Clinical responses of evaluable patients with CAP
| Assessment | Comparative trial
|
Noncomparative trial, azithromycin | All azithromycin | |
|---|---|---|---|---|
| Azithromycin | Cefuroxime ± erythromycin | |||
| 10 to 14 days posttherapy | ||||
| No. of evaluable patients | 137 | 131 | 84 | 221 |
| No. (%) of patients with treatment: | ||||
| Cure or improvement | 106 (77) | 97 (74) | 74 (89) | 180 (81) |
| Failure | 31 (23) | 34 (26) | 10 (12) | 41 (19) |
| 4 to 6 weeks posttherapy | ||||
| No. of evaluable patients | 130 | 122 | 85 | 215 |
| No. (%) of patients with treatment: | ||||
| Cure | 98 (75) | 87 (71) | 73 (86) | 171 (80) |
| Failure | 32 (25) | 35 (29) | 12 (14) | 44 (20) |
Bacteriologic results.
Results from the comparative and the noncomparative trials are summarized in Table 3. Clinical success rates for patients receiving azithromycin are pooled according to the baseline pathogen for the two protocols and at the 10- to 14-day evaluation period are 54 of 61 (89%) patients with S. pneumoniae infection, 32 of 38 (84%) patients with H. influenzae infection, 7 of 9 (78%) patients with Staphylococcus aureus infection and 8 of 9 (89%) patients with Moraxella catarrhalis infection. Success rates stratified by pathogen in the comparative trial were similar for the two regimens, notably for patients with S. pneumoniae infection, and clinical failure rates were 7 of 27 (26%) among patients treated with the comparator regimen and 4 of 30 (13%) among patients receiving azithromycin (P was not significant). Table 4 shows the comparable results for patients with S. pneumoniae bacteremia. In the comparative trial, the three clinical failure patients in the comparator arm had a negative blood culture before clinical failure was determined. Of the four azithromycin clinical failures, two patients had a negative follow-up blood culture and two patients did not have repeat cultures performed.
TABLE 3.
Clinical response by pathogens most frequently isolated at baseline from evaluable patientsa
| Assessment and pathogen | No. of patients responding/no. of patients with pathogen at baseline (%)
|
|||
|---|---|---|---|---|
| Comparative trial
|
Noncomparative trial, azithromycin | All azithromycin | ||
| Azithromycin | Cefuroxime ± erythromycin | |||
| 10 to 14 days posttherapy | ||||
| S. pneumoniae | 26/30 (87) | 20/27 (74) | 28/31 (90) | 54/61 (89) |
| H. influenzae | 12/15 (80) | 8/10 (80) | 20/23 (87) | 32/38 (84) |
| S. aureus | 4/5 (80) | 1/3 (33) | 3/4 (75) | 7/9 (78) |
| M. catarrhalis | 2/3 (67) | 1/3 (33) | 6/6 (100) | 8/9 (89) |
| M. pneumoniae | 7/9 (78) | 4/5 (80) | 9/9 (100) | 16/18 (89) |
| C. pneumoniae | 9/12 (75) | 14/18 (78) | 19/21 (90) | 28/33 (85) |
| Legionella | 3/4 (75) | 4/5 (80) | 10/12 (83) | 13/16 (81) |
| 4 to 6 weeks posttherapy | ||||
| S. pneumoniae | 19/23 (83) | 18/26 (69) | 27/32 (84) | 46/55 (84) |
| H. influenzae | 11/14 (79) | 8/10 (80) | 18/22 (82) | 29/36 (81) |
| S. aureus | 4/5 (80) | 2/4 (50) | 3/4 (75) | 7/9 (78) |
| M. catarrhalis | 1/2 (50) | 1/3 (33) | 5/7 (71) | 6/9 (67) |
| M. pneumoniae | 8/10 (80) | 5/6 (83) | 10/10 (100) | 18/20 (90) |
| C. pneumoniae | 8/11 (73) | 14/18 (78) | 20/22 (91) | 28/33 (85) |
| Legionella | 4/5 (80) | 3/4 (75) | 12/14 (86) | 16/19 (84) |
Some patients were infected with more than one pathogen at the baseline.
TABLE 4.
Results for patients with S. pneumoniae bacteremia at the baseline
| Study and treatment | No. of patients infected with S. pneumoniae isolates (all valid sources) | No. of patients infected at baseline | No. (%) of patients with clinical success, 10 to 14 days posttherapy |
|---|---|---|---|
| Comparative trial | |||
| Azithromycin | 30 | 15 | 11 (73) |
| Cefuroxime ± erythromycin | 27 | 13 | 10 (77) |
| Noncomparative trial, azithromycin | 31 | 9 | 8 (89) |
| Total, all azithromycin | 61 | 24 | 19 (79) |
A microbiologically evaluable population was defined as patients for whom actual eradication was documented by culture or patients with presumed eradication and with a favorable clinical response and no sputum available for culture. The number of patients in this group is larger than the clinical evaluable subset because of the wider inclusion criteria. Rates of eradication by azithromycin for the pathogens most commonly isolated at the baseline were 64 of 67 (96%) for S. pneumoniae, 41 of 43 (96%) for H. influenzae, 9 of 10 (90%) for S. aureus, and 9 of 10 (90%) for M. catarrhalis. In the comparative trial, rates of H. influenzae eradication were 15 of 16 (94%) in the azithromycin group and 8 of 11 (73%) in the comparative group (P was not significant).
At 4 to 6 weeks clinical success for evaluable patients with serologic or other evidence of atypical pneumonia treated with azithromycin was seen for 18 of 20 (90%) patients with M. pneumoniae infection and 28 of 33 (85%) patients with C. pneumoniae infection. Diagnosis of M. pneumoniae infection was done by culture and PCR for 5 patients, by detection of an IgM titer of >10 for 13 patients, and by detection of a fourfold rise in IgG titer for 2 patients. The diagnosis of C. pneumoniae infection in 30 patients was made on the basis of an IgG titer of >512 or a fourfold rise in titer, and the diagnosis in 3 patients was made on the basis of an IgM titer of >20. Nineteen patients in the two azithromycin trials and four patients in the comparative group had evidence of infection with Legionella that was most often diagnosed by detection of an IgG titer with a fourfold or greater elevation from that at the baseline. Clinical success rates for the azithromycin-treated patients at 4 to 6 weeks were 16 of 19 (84%), whereas they were 3 of 4 (75%) for patients treated with the comparator regimen (P was not significant).
Responses of penicillin-insensitive and macrolide-resistant pneumococci.
Of 89 S. pneumoniae isolates tested at a reference laboratory, 13 (15%) had reduced susceptibility to penicillin (MICs ≥0.12 μg/ml) and 4 (5%) had reduced susceptibility to azithromycin (MICs >0.5 μg/ml). Six patients were treated with an antibiotic to which the S. pneumoniae isolate had reduced susceptibility. Two patients infected with S. pneumoniae with reduced penicillin susceptibility and treated with cefuroxime monotherapy were considered clinical cures. The penicillin MIC for the isolate from one patient was 0.5 μg/ml, and the disk diffusion zone around a cefuroxime disk was 43 mm for this isolate. For the isolate from the second patient, the disk zone size was 11 mm, but no broth microdilution MIC was determined. The disk diffusion zone sizes of the isolates from the four patients treated with azithromycin and infected with S. pneumoniae with reduced susceptibility were 6, 10, 14, and 14 mm, respectively. Clinical success was observed in two patients, and organisms were eradicated from three of four (75%) patients receiving azithromycin. An S. pneumoniae isolate from one patient treated with azithromycin for 5 days i.v. and for 5 days orally had a disk diffusion size of 14 mm, and the MIC for the isolate was 8 μg/ml. The patient was judged to be cured by the investigator, with eventual eradication of the organism.
Safety and tolerability.
Overall, a slightly greater proportion of patients in the comparative group than azithromycin-treated patients in the comparative trial reported treatment-related adverse events (49 of 201 [24.4%] versus 39 of 202 [19.3%], respectively). However, as seen in Table 5, the incidence of nausea among patients receiving the comparator regimen (8.0%) was significantly higher than that among patients receiving azithromycin (2.0%) (P < 0.05). The treatment-related adverse events reported by >2% of the 202 patients in the azithromycin treatment group in the comparative trial were diarrhea and insertion site pain (each 5.4%; 11 of 202 patients) and insertion site infection or inflammation (3.5%; 7 of 202). Among the 201 patients in the comparative group, the treatment-related adverse events reported by >2% of patients were diarrhea and nausea (each 8.0%; 16 of 201), insertion site infection or inflammation (7.0%; 14 of 201), and insertion site pain (6.0%; 12 of 201). In the azithromycin treatment group there was a single report of severe treatment-related oral moniliasis, and in the comparative agent treatment group there were two severe cases of treatment-related insertion site pain and single reports each of severe treatment-related diarrhea, nausea, and vomiting.
TABLE 5.
Treatment-related adverse events in patients with CAP
| Eventa | No. (%) of patients
|
|||
|---|---|---|---|---|
| Comparative trial
|
Noncomparative trial, azithromycin (n = 212) | All azithromycin (n = 414) | ||
| Azithromycin (n = 202) | Cefuroxime ± erythromycin (n = 201) | |||
| Gastrointestinal | ||||
| Diarrhea | 11 (5.4) | 16 (8.0) | 7 (3.3) | 18 (4.3) |
| Nausea | 4 (2.0)b | 16 (8.0) | 12 (5.7) | 16 (3.9) |
| Abdominal pain | 2 (1.0) | 2 (1.0) | 9 (4.2) | 11 (2.7) |
| Vomiting | 0 (0.0) | 4 (2.0) | 6 (2.8) | 6 (1.4) |
| Insertion site | ||||
| Pain | 11 (5.4) | 12 (6.0) | 16 (7.5) | 27 (6.5) |
| Infection or inflammation | 7 (3.5) | 14 (7.0) | 6 (2.8) | 13 (3.1) |
Incidence of >2% in any group.
Significant difference between azithromycin and comparator group (P = 0.006).
Treatment-related side effects caused discontinuation of therapy for five (1.2%) azithromycin recipients (two with rash and one each with nausea and vomiting, abdominal pain, and somnolence) and six (3.0%) patients receiving the comparator regimen (three with diarrhea and two with rash in the group receiving cefuroxime plus erythromycin and one with colitis in the group receiving cefuroxime alone). Most clinically important laboratory abnormalities were associated with underlying conditions or a resolving acute infection. In the comparative trial, 2 of 202 (1%) patients in the azithromycin group were discontinued from active therapy and from the study due to laboratory abnormalities that were considered related to the study drug, and one other patient was discontinued from active therapy due to abnormal transaminase values but completed the study. In the comparator arm 2 of 201 (1%) patients were discontinued from active therapy and from the study due to laboratory abnormalities that were judged by the investigator to be unrelated to study drug.
The rate of mortality at 35 days posttherapy in these trials was 3.5% (14 patients) for the azithromycin group and 4.0% (8 patients) for the comparator group. The comparative trial had 51 of 268 (19%) patients in PSI class 1, and patients in this group have previously been shown to have a low (0.1%) mortality risk; therefore, on this basis it is suggested that outpatient treatment might be considered for these patients (12). The low PSI scores for 51 patients were due to the low mean age of 37 years. Signs that may have led to hospitalization in this group include arterial blood gas oxygen pressure of <60 mm/Hg (n = 8 patients), a respiratory rate of ≥28 breaths per minute (n = 28), heart rates of ≥110 beats per min (n = 14), and pleural effusions on chest radiographs (n = 4).
DISCUSSION
The comparative study described in this report demonstrates that i.v. azithromycin therapy for hospitalized patients with CAP is equal in efficacy to a 1993 ATS-suggested regimen of cefuroxime with the addition of erythromycin when clinicians believe that it is necessary. Gastrointestinal tolerance of the azithromycin regimen was better than that of the ATS-suggested regimen, and azithromycin therapy was associated with a significantly reduced length (in days) of i.v. and total therapy compared to that for the ATS-suggested regimen. A decreased length of i.v. therapy and a switch to oral therapy have been correlated with shorter lengths of stay and decreased costs of hospitalization (4). In the case of azithromycin, the same medication can be given orally once a day, thus simplifying outpatient treatment regimens.
Approximately 20% of patients with CAP require hospitalization during their illness, often because they are too ill to take an orally administered antimicrobial agent or they require careful hemodynamic monitoring. The ATS guidelines for therapy of hospitalized patients with CAP recommend addition of a macrolide generally for 7 to 10 days but for up to 21 days when Legionella is suspected or identified. Many practitioners choose to use cephalosporin monotherapy, as was the case in this study, with the view that the clinical features of the patient can give clues as to when an atypical pathogen is present. When a clinician treats a patient infected with Mycoplasma or Chlamydia with a beta-lactam or cephalosporin monotherapy, failure to respond clinically will usually prompt added coverage for an atypical organism. However, patients with CAP caused by Legionella may suffer worsening of their condition and increased mortality when not treated with specific therapy against this organism in a timely manner (17).
Azithromycin has been shown in vitro and in vivo to concentrate in phagocytic cells, a phenomenon possibly related to its amphiphilic cationic properties. The uptake of azithromycin into leukocytes results in delivery of the drug to the site of infection, without impairment of phagocytic cell functions (15). This phenomenon may also be responsible for the activity of azithromycin against pathogens that are relatively resistant to intracellular killing within macrophages such as S. aureus (22). In addition, the high and sustained concentrations of azithromycin in tissue have been correlated with efficacy against intracellular pathogens in laboratory models of infections due to Legionella species (13) and C. pneumoniae (18).
The limited data on pneumococcal resistance from the two studies reported here would indicate that clinical success still occurs with cephalosporins and macrolides when modest increases of MICs in vitro are recorded. Some investigators have suggested that pharmacodynamics must be considered individually for drugs and that in vitro breakpoints do not take into consideration in vivo levels of antimicrobial agents in tissues (14, 25). The number of patients infected with resistant organisms was small in the two trials reported on here, and decisions on the treatment regimens to be used for patients with a substantial mortality risk need to take into consideration local susceptibility patterns. Reports have recently examined additional properties of macrolides that may explain their benefit beyond the antimicrobial activity observed in vitro (27). Macrolide therapy has also been associated with improved clinical outcomes in elderly patients with comorbidities compared to the outcomes for outpatients with CAP who receive guideline recommendations of a beta-lactam or a sulfa antibiotic (16). Treatment regimens that include a macrolide have also been associated with lower fatality rates in CAP patients with bacteremic pneumococcal pneumonia compared to those for penicillin or cephalosporin regimens (23).
The PSI was developed as a prediction tool for determination of whether patients should be admitted to the hospital on the basis of an objective method of stratifying the mortality risk for patients. The predicted mortality rates range from 0.1% for patients in class 1 to 31.1% for patients in class 5 from the studies conducted by the pneumonia patient outcome research team. The PSI was calculated retrospectively for this trial and is likely to be an underestimate because of missing data that were not obtained systematically. The mean PSI scores in the comparative and noncomparative trials indicate that, on average, patients had similar severities of illness and needed hospitalization. Even the younger patients with a PSI score that placed them in class 1 and a low probability of dying had physical findings, laboratory results, or a pleural effusion that prompted the hospital admission. This differs from data reported from trials in which outpatients are combined in the analysis with inpatients to report overall high rates of clinical cure (10) since outpatients with less severe disease have less of a chance of complications.
The 1997 Infectious Disease Society of America CAP guidelines support the use of azithromycin in hospitalized CAP patients as monotherapy or as the macrolide component of a combination regimen (4). In summary, the two clinical trials described here demonstrate that an i.v.-oral regimen of azithromycin is effective and well tolerated as monotherapy in adult patients admitted to a general medical ward with CAP requiring initial i.v. treatment.
Appendix
Investigators who contributed to these studies included Vincent Acampora, William A. Alto, Guy W. Amsden, Paul M. Arnow, Neil Barg, Jack M. Bernstein, Karl R. Beutner, Frederick Bode, Leigh Ann Callahan, Kerry O. Cleveland, Paul Peniston Cook, Larry Danziger, Carlos A. Dela Garza, Naresh A. Dewan, Mark Doner, Charles D. Ericcson, James V. Felicetta, James I. Fidelholtz, Mitchell P. Fink, J. F. Foss, Henry Fraimow, Harry A. Gallis, John A. Gezon, Kevin Gleeson, Kalpaletha K. Guntupalli, Marilyn T. Haupt, William J. Holloway, Judith M. Hyatt, Amy A. Imm, Keith I. Ironside, Jr., Kirk D. Jacobson, Stephen G. Jenkinson, Monroe Karetzky, Karen E. Kirkham, Tim Kotschwar, Robert J. Lapidus, David M. Letzer, Nazir A. Memon, Burt Meyers, Dennis J. Mikolich, Michael Nelson, Cheryl L. Newman, M. Brooke Nicotra, Donald S. North, Richard F. O'Brien, Col. Charles N. Oster, Joseph A. Paladino, Darwin L. Palmer, Alexander J. Pareigis, Gregory E. Peterson, Rick Player, Julio A. Ramirez, Jay Redington, Richard K. Root, Robert Salata, W. Michael Scheld, Douglas B. Schwartz, Bruce W. Sherman, Robert Siegel, Charles Sigmund, Jr., David L. Smith, Rodney M. Snow, Stephen Storfer, Byungse Suh, Eric S. Swann, Charles van Hook, Andrew G. Villanueva, David K. Wagner, Winkler Weinberg, Warren L. Whitlock, Randall Willis, Sandra K. Willsie, Rodney M. Wishnow, and Marcus Zervos.
REFERENCES
- 1.Agacfidan A, Moncada J, Schachter J. In-vitro activity of azithromycin (CP-62,993) against Chlamydia trachomatis and Chlamydia pneumoniae. Antimicrob Agents Chemother. 1993;37:1746–1748. doi: 10.1128/aac.37.9.1746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barry A L, Fuchs P C. In vitro activities of a streptogramin (RP59500), three macrolides, and an azalide against four respiratory tract pathogens. Antimicrob Agents Chemother. 1995;39:238–240. doi: 10.1128/aac.39.1.238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Barry A L, Pfaller M A, Fuchs P C, Packer R R. In vitro activities of 12 orally administered antimicrobial agents against four species of bacterial respiratory pathogens from U.S. medical centers in 1992 and 1993. Antimicrob Agents Chemother. 1994;38:2419–2425. doi: 10.1128/aac.38.10.2419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bartlett J G, Breiman R F, Mandell L A, File T M., Jr Community-acquired pneumonia in adults: guidelines for management. Clin Infect Dis. 1998;26:811–838. doi: 10.1086/513953. [DOI] [PubMed] [Google Scholar]
- 5.Bartlett J G, Mundy L M. Community-acquired pneumonia. N Engl J Med. 1995;333:1618–1624. doi: 10.1056/NEJM199512143332408. [DOI] [PubMed] [Google Scholar]
- 6.Bates J H, Campbell G D, Barron A L, McCracken G A, Morgan P N, Moses E B, Davis C M. Microbial etiology of acute pneumonia in hospitalized patients. Chest. 1992;101:1005–1012. doi: 10.1378/chest.101.4.1005. [DOI] [PubMed] [Google Scholar]
- 7.Donowitz G R. Tissue-directed antibiotics and intracellular parasites: complex interaction of phagocytes, pathogens, and drugs. Clin Infect Dis. 1994;19:926–930. doi: 10.1093/clinids/19.5.926. [DOI] [PubMed] [Google Scholar]
- 8.Edelstein P H, Edelstein M A C. In vitro activity of azithromycin against clinical isolates of Legionella species. Antimicrob Agents Chemother. 1991;35:180–181. doi: 10.1128/aac.35.1.180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fang G D, Fine M, Orloff J, et al. New and emerging etiologies for community acquired pneumonia with implications for therapy—prospective multicenter study of 359 cases. Medicine. 1990;69:307–316. doi: 10.1097/00005792-199009000-00004. [DOI] [PubMed] [Google Scholar]
- 10.File T M, Segreti J, Jr, Dunbar L, Player R, Kohler R, Williams R R, Kojak C, Rubin A. 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;41:1965–1972. doi: 10.1128/aac.41.9.1965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fine M J, Smith M A, Carson C A, Mutna S S, Sankey S S, Weissfeld L A, Kapoor W N. Prognosis and outcomes of patients with community-acquired pneumonia. A meta-analysis. JAMA. 1996;275:134–141. [PubMed] [Google Scholar]
- 12.Fine M J, Auble T E, Yealy D M, Hanusa B H, Weissfeld L A, Singer D E, Coley C M, Marrie T J, Kapoor W N. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336:243–250. doi: 10.1056/NEJM199701233360402. [DOI] [PubMed] [Google Scholar]
- 13.Fitzgeorge R B, Lever S, Baskerville A. A comparison of the efficacy of azithromycin and clarithromycin in oral therapy of experimental airborne Legionnaires' disease. J Antimicrob Chemother. 1993;31(Suppl. E):171–176. doi: 10.1093/jac/31.suppl_e.171. [DOI] [PubMed] [Google Scholar]
- 14.Garey K W, Amsden G W. Intravenous azithromycin. Ann Pharmacother. 1999;33:218–228. doi: 10.1345/aph.18046. [DOI] [PubMed] [Google Scholar]
- 15.Gladue R P, Bright G M, Isaacson R E, Newborg M F. In vitro and in vivo uptake of azithromycin by phagocytic cells: possible mechanism of delivery and release at sites of infection. Antimicrob Agents Chemother. 1989;33:277–282. doi: 10.1128/aac.33.3.277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gleason P P, Kapoor W N, Stone R A, Lare J R, Obrosky D S, Schulz R, Singer D E, Coley C M, Marrie T J, Fine M J. Medical outcomes and antimicrobial costs with the use of the American Thoracic Society guidelines for outpatients with community-acquired pneumonia. JAMA. 1997;278:32–39. [PubMed] [Google Scholar]
- 17.Heath C H, Grove D I, Looke D F. Delay in appropriate therapy of Legionella pneumonia associated with increased mortality. Eur J Clin Microbiol Infect Dis. 1996;15:286–290. doi: 10.1007/BF01695659. [DOI] [PubMed] [Google Scholar]
- 18.Malinverni R, Cho-Chou K, Campbell L A, Lee A, Grayston J T. Effects of two antibiotic regimens on the course and persistence of experimental Chlamydia pneumoniae TWAR pneumonitis. Antimicrob Agents Chemother. 1995;39:45–49. doi: 10.1128/aac.39.1.45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mandell G L. Delivery of antibiotics by phagocytes. Clin Infect Dis. 1994;19:922–925. doi: 10.1093/clinids/19.5.922. [DOI] [PubMed] [Google Scholar]
- 20.Marrie T J, Durant H, Yates L. Community-acquired pneumonia requiring hospitalization: 5-year prospective study. Rev Infect Dis. 1989;11:586–599. doi: 10.1093/clinids/11.4.586. [DOI] [PubMed] [Google Scholar]
- 21.Marrie T J. Community-acquired pneumonia. Clin Infect Dis. 1994;18:501–515. doi: 10.1093/clinids/18.4.501. [DOI] [PubMed] [Google Scholar]
- 22.Meyer A P, Bril-Bazuin C, Mattie H, van den Broek P J. Uptake of azithromycin by human monocytes and enhanced intracellular antibacterial activity against Staphylococcus aureus. Antimicrob Agents Chemother. 1993;37:2318–2322. doi: 10.1128/aac.37.11.2318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mufson M A, Stanek R J. Bacteremic pneumococcal pneumonia in one American city: a 20-year longitudinal study, 1978–1997. Am J Med. 1999;107:34S–43S. doi: 10.1016/s0002-9343(99)00098-4. [DOI] [PubMed] [Google Scholar]
- 24.Niederman M S, Bass J B, Jr, Campbell G D, Fine A M, Grossman R F, Mandel L A, Marrie T J, Sarosi G A, Torres A, Yu V L. Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy. Am Rev Respir Dis. 1993;148:1418–1426. doi: 10.1164/ajrccm/148.5.1418. [DOI] [PubMed] [Google Scholar]
- 25.Pallares R, Linares J, Vadillo M, Cabellos C, Manresa F, Viladrich P F, Martin R, Gudiol F. Resistance to penicillin and cephalosporin and mortality from severe pneumococcal pneumonia in Barcelona, Spain. N Engl J Med. 1995;333:474–480. doi: 10.1056/NEJM199508243330802. [DOI] [PubMed] [Google Scholar]
- 26.Rapp R P. Pharmacokinetics and pharmacodynamics of intravenous and oral azithromycin: enhanced tissue activity and minimal drug interactions. Ann Pharmacother. 1998;32:785–793. doi: 10.1345/aph.17299. [DOI] [PubMed] [Google Scholar]
- 27.Shryock T R, Mortensen J E, Baumholtz M. The effects of macrolides on the expression of bacterial virulence mechanisms. J Antimicrob Chemother. 1998;41:505–512. doi: 10.1093/jac/41.5.505. [DOI] [PubMed] [Google Scholar]