Efficacy and Safety of a Novel, Single-dose Azithromycin Microsphere Formulation Versus 10 Days of Levofloxacin for the Treatment of Acute Bacterial Sinusitis in Adults

Otolaryngology–Head and Neck Surgery (2005) 133, 194-201

Efficacy and Safety of a Novel, Single-dose Azithromycin Microsphere Formulation Versus 10 Days of Levofloxacin for the Treatment of Acute Bacterial Sinusitis in Adults John J. Murray, MD, PhD, Paz Emparanza, MD, Eugenijus Lesinskas, MD, PhD, Margaret Tawadrous, MD, MS, and Jeanne D. Breen, MD, Nashville, Tennesee, Santiago, Chile, Vilnius, Lithuania and London, Connecticut

A

OBJECTIVE: To compare the efficacy and safety of a single 2.0-g dose of a novel azithromycin microsphere formulation with that of 10 days of levofloxacin, 500 mg/d, when used to treat adults with uncomplicated acute bacterial maxillary sinusitis (ABS). STUDY DESIGN AND SETTING: An international, multicenter, randomized, double-blind, double-dummy trial. Eligible outpatients ⱖ18 years of age with clinical and radiographic evidence of ABS underwent maxillary sinus aspiration before randomization. Primary endpoint was clinical efficacy at the test-of-cure visit (day 17-24). RESULTS: Clinical success rates were 94.5% (242/256) in azithromycin-microspheres–treated patients and 92.8% (233/251) in the levofloxacin group. In patients with documented Streptococcus pneumoniae, Haemophilus influenzae, or Moraxella catarrhalis, clinical cure rates were 97.3% (36/37), 96.3% (26/27), and 100% (8/8), respectively, for the azithromycin group and 92.3% (36/39), 100% (30/30), and 90.9% (10/11), respectively, for the levofloxacin group. CONCLUSIONS: Single-dose azithromycin microspheres provided clinical and bacteriologic efficacy and safety comparable to 10 days of levofloxacin. SIGNIFICANCE: A novel microsphere formulation of azithromycin given as a single dose was safe and effective for the treatment of ABS. © 2005 American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. All rights reserved.

cute bacterial sinusitis (ABS) is among the most common diagnoses made in primary care and is the 5th most common diagnosis for which antibiotics are prescribed.1 Its prevalence contributes to high healthcare costs, which are estimated to approach $6 billion a year in the United States alone.2 ABS often is preceded by an acute viral upper respiratory tract infection (URTI) or allergic rhinitis1 and is characterized by inflammation of the paranasal sinuses arising from bacterial overgrowth in the sinus cavities. Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis are the predominant pathogens isolated from the maxillary sinuses of patients with ABS, with Staphylococcus aureus and Streptococcus pyogenes less frequently isolated.1,3 Sinus aspiration is the most definitive means of documenting the causative pathogens of ABS; however, this procedure is rarely performed because of its invasive nature and because of the delay, up to 72 hours, of identifying pathogens by culture. Diagnosis and treatment therefore are most commonly empiric, based on the presence of clinical signs and symptoms such as purulent nasal discharge, nasal congestion, and fever accompanied by persistent facial pain and tenderness of 7 to 28 days’ duration. Although ⱕ40% of

From Vanderbilt Medical School, Nashville, Tennessee (Dr Murray); Otomed-Servicio Otorrinolaringologia, Santiago, Chile (Dr Emparanza); Vilnius University, Vilnius, Lithuania (Dr Lesinskas); and Pfizer Inc, New London, Connecticut (Drs Tawadrous and Breen). This clinical trial was funded by Pfizer Inc. Data from this study were the focus of an oral presentation by Dr

Murray at 44th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), October 30-November 2, 2004, Washington, DC. Reprint requests: Margaret Tawadrous, MD, MS, Pfizer Inc, Clinical Research and Development PGRD, 50 Pequot Avenue, New London, CT 06320. E-mail address: [email protected].

0194-5998/$30.00 © 2005 American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. All rights reserved. doi:10.1016/j.otohns.2005.04.020

Murray et al

Efficacy and Safety of a Novel, Single-dose. . .

sinusitis patients may recover spontaneously, numerous reports have concluded that antimicrobial therapy is beneficial in properly diagnosed cases of acute sinusitis.1 Treatment with antibiotics results in more rapid resolution of symptoms; the greatest benefit has been demonstrated in patients with clinical and radiologic evidence of ABS.1 Empirical therapy for ABS traditionally has included the use of ␤-lactams or trimethoprim–sulfamethoxazole as 1stline agents, usually administered 2 to 3 times daily for periods of 7 to 14 days, sometimes longer.1 Growing evidence suggests that shortened courses of antibiotic therapy for upper respiratory tract infections, including ABS, are at least as effective clinically as are standard courses and may confer benefits such as improved compliance with therapy, decreased adverse events (AEs), lower rates of treatment failure, and reduced costs.4-6 Yet, before approval of 3-day azithromycin for the treatment of ABS in 2004, the shortest antibiotic treatment regimen approved by the FDA was 10 days. Azithromycin, an azalide antibiotic, is widely used to treat a variety of respiratory tract infections, including ABS. Its efficacy in the treatment of ABS has been established in a series of clinical trials in which once-daily treatment over 3 or 6 days achieved clinical success rates of 87% to 98%, comparable to those of amoxicillin, amoxicillin– clavulanate, and erythromycin.2,7-10 Azithromycin accumulates within a variety of cells, including fibroblasts, epithelial cells, macrophages, monocytes, and neutrophils.11 Its ability to penetrate and concentrate in cells is responsible for its distinctive pharmacokinetic profile, which features an extremely long elimination half-life of approximately 60 hours and high and sustained tissue concentrations.11 These properties allow for much shorter dosing regimens than are possible with other antibiotics that have short elimination half-lives. A novel microsphere formulation of azithromycin recently has been developed that makes it possible to administer a higher oral dose of azithromycin as a single-dose regimen while maintaining tolerability. The peak serum concentration (Cmax) and 24-hour area under the curve (AUC0-24) after a single 2.0-g dose of azithromycin microspheres are 2 and 3 times higher, respectively, than those achieved with 1.5 g of conventional, immediate-release azithromycin, given over 3 or 5 days. This so-called frontloading of the dose early in the course of infection maximizes drug exposure when the bacterial burden is highest. Front loading also increases the AUC0-24/MIC ratio, the pharmacokinetic/pharmacodynamic parameter that best predicts efficacy of azithromycin.12 Data from animal infection models suggest that this approach is effective in a variety of respiratory tract infections.13-15 Administration of a single, large dose of azithromycin has been shown to achieve more rapid bacterial eradication and enhanced survival than the same dose divided over several days in preclinical infection models of otitis media, murine pneumonia, and septicemia.14,15

195

The present study was designed to test the hypothesis that a single 2.0-g dose of azithromycin microspheres is clinically noninferior to 10 days of levofloxacin, 500 mg/d, when used to treat adults diagnosed with uncomplicated acute bacterial maxillary sinusitis.

MATERIALS AND METHODS Study Design and Ethics This was a multicenter, randomized, double-blind, doubledummy, comparative trial conducted in 81 outpatient centers in the United States, India, Europe, and Latin America. The study was approved by the relevant institutional review boards in accordance with local legislation and good clinical practices. Written informed consent was obtained from all patients before study enrollment.

Patients Male and female outpatients, 18 years of age or older, were eligible for enrollment, provided that they had cardinal signs and symptoms of at least 7 days’ duration; these included facial pain, pressure, and/or tightness over 1 or both maxillary sinuses combined with evidence of purulent discharge from the nose or the maxillary sinus orifice and/or purulent drainage in the posterior pharynx. In addition, at least 2 of the following signs and symptoms were required: fever, leukocytosis, frequent coughing, headache, nasal congestion, or postnasal drainage. The diagnosis of sinusitis was confirmed by the presence of either complete or partial opacification or an air–fluid level on Water’s view roentgenograph. All patients underwent maxillary sinus puncture and aspiration at baseline. Key exclusion criteria included the following: symptoms lasting for more than 28 days, recurrent sinusitis (⬎4 episodes in the preceding 12 months), nasal or sinus surgery in the preceding 3 months, complicated sinusitis, and nosocomial sinusitis.

Treatment Eligible patients were randomized in a 1:1 ratio to receive either a single 2.0-g dose of azithromycin microspheres, administered as an oral suspension, plus levofloxacin placebo (once daily for 10 days), or oral levofloxacin (500 mg once daily for 10 days) plus a single dose of azithromycin microspheres placebo. Study regimens were administered in a double-blind, double-dummy fashion. Patients were observed for 30 minutes after administration of active azithromycin microspheres or azithromycin placebo. Subjects who vomited within 30 minutes of dosing were to have a blood sample drawn at 2 hours after dose for determination of azithromycin levels.

Study Procedures Patients were assessed at baseline (day 1), during treatment (day 3-5), at the end of treatment (EOT; day 11-13), and at

196

Otolaryngology–Head and Neck Surgery, Vol 133, No 2, August 2005

posttreatment test of cure (TOC) 17-24 days after the 1st dose. The baseline visit included a detailed medical history, a physical examination, clinical assessment of signs and symptoms of ABS, as well as vital signs, sinus radiography, and maxillary sinus aspiration. Women of childbearing potential had a pregnancy test performed. At the day 3-5 visit, patients were assessed for signs and symptoms of infection as compared with baseline and were queried regarding compliance with study medication and the occurrence of any AEs. These assessments were repeated in a telephone follow-up at the EOT visit and in the clinic at the TOC visit. Maxillary sinus puncture could be repeated at any time during the study if patients had persistent or worsened cardinal signs and symptoms of ABS. All sinus aspirates were sent for routine culture and antibiotic susceptibility testing.

Efficacy Outcome Measures The primary efficacy endpoint in this study was clinical response in the clinical-per-protocol (CPP) population at TOC. This population included all patients who met the inclusion criteria, received 8 days of study medication (including active plus placebo doses), were assessed at relevant timepoints, and did not receive any concomitant systemic antibiotic therapy with efficacy against ABS pathogens. Clinical response was assessed as cure or failure, where cure was defined as resolution of signs and symptoms related to the acute infection or clinical improvement such that no additional antibiotics were deemed necessary. Failure was defined as the persistence or worsening of signs and symptoms of acute infection requiring the use of additional antibiotics or the development of new signs and symptoms of ABS requiring the use of additional antibiotics. Secondary endpoints included bacteriologic response and clinical response by baseline pathogen in the bacteriologic-per-protocol (BPP) population at TOC. The BPP population included all CPP patients with a bacterial pathogen isolated at baseline. For each pathogen isolated at baseline, a bacteriologic response of eradication, presumed eradication, persistence, or presumed persistence was assigned. Assessment of bacteriologic eradication was based on posttreatment culture results; if a sinus aspirate specimen was not available subsequent to the baseline visit, the bacteriologic response was based on the clinical assessment at the TOC visit.

Safety and Compliance With Study Medication Safety was assessed in all patients who took at least 1 dose of study medication. The nature, frequency, and severity of all AEs as well as their relationship to study therapy were recorded at each visit or contact. Compliance with study medication also was assessed at each visit or via telephone contact.

Table 1 Baseline demographics and clinical history (all treated population) Parameter

Azithromycin Levofloxacin (n ⫽ 270) (n ⫽ 268)

Gender (M/F) 126/144 Mean age (y) ⫾ SD 38.4 ⫾ 14.4 Age range (y) 18-88 Race, n (%) White 180 (66.7) Black 9 (3.3) Asian 37 (13.7) Hispanic 44 (16.3) Mean duration of current episode (d) 13.2 Baseline signs and symptoms, n (%) Facial pain, pressure or tenderness 270 (100) Purulent nasal discharge 270 (100) Headache 238 (88.1) Nasal congestion 265 (98.1) Postnasal drainage 240 (88.9) Radiology findings, n (%) Air–fluid level only 46 (17.0) Opacification only 172 (63.7) Both 52 (19.3)

99/169 39.4 ⫾ 14.1 18-81 181 5 37 43

(67.5) (1.9) (13.8) (16.0)

13.3

268 (100) 266 248 260 245

(99.3) (92.5) (97.0) (91.4)

51 (19.0) 160 (59.7) 56 (20.9)

Statistical Methods Azithromycin was considered to be noninferior to levofloxacin if the lower boundary of the 95% confidence interval (CI) around the difference in cure rates between the 2 treatments was greater than –10%. A target sample size of 504 subjects was established on the basis of the following assumptions: 80% power to show noninferiority at the 2-sided level of significance, an assumed clinical cure rate of 85% for both treatment groups, and a nonevaluability rate of 20%.

RESULTS Patients A total of 541 patients with a clinical diagnosis of ABS were randomized to treatment, of whom 538 received at least 1 dose of study medication and were included in the safety analysis. Of these, 507 (94.2%) met the criteria for inclusion in the CPP population: 256 and 251 in the azithromycin- and levofloxacin-treatment arms, respectively. Lack of a TOC visit (21 patients, 4%) and receipt of other antibiotics (6 patients, 1%) were the most common reasons for exclusion from the CPP population at TOC. Overall baseline demographic and clinical characteristics were similar in the 2 treatment arms (Table 1). Slightly more female patients were randomized to treatment with levofloxacin than azithromycin (63% versus 53%). About

Murray et al

Efficacy and Safety of a Novel, Single-dose. . .

197

Table 2 Clinical response at test of cure* by population

Population

n

All treated Clinically eligible Clinical per protocol Bacteriologic per protocol

538 534 507 213

Azithromycin microspheres n (%) 246 246 242 99

Levofloxacin n (%)

(91.1) (91.1) (94.5) (97.1)

239 235 233 102

95% confidence interval

(89.2) (89.0) (92.8) (91.9)

ND ND (⫺2.5, 5.9) ND

ND, not done. *Day 17-24.

one third of patients suffered from allergic rhinitis, and a similar proportion were current or ex-smokers. Pathogens were recovered from baseline sinus aspirates in 42% (228/ 538) of patients, including 108 (40%) in the azithromycin arm and 120 (44%) in the levofloxacin arm. The percentage of patients with a single pathogen in the azithromycin and levofloxacin groups was 36% and 38%, respectively. The percentage of patients with multiple pathogens in the azithromycin and levofloxacin groups was 4% and 6%, respectively.

tient in the azithromycin arm had S. pneumoniae highly resistant to azithromycin isolated at baseline (MIC ⬎ 256 ␮g/mL). Repeat sinus aspiration, however, documented eradication of the original pneumococcus; Escherichia coli was recovered from the same specimen. With the single exception of the patient described above, all patients in the azithromycin treatment group who were infected with strains of S. pneumoniae that were nonsusceptible to azithromycin (MIC ⬎ 2 ␮g/mL) were clinically cured (Table 4).

Efficacy

Compliance With Study Medication

A single dose of azithromycin microspheres was as effective in providing a clinical cure at TOC in the CPP population, as was 10 days of levofloxacin (94.5% versus 92.8%, respectively; 95% CI ⫽ ⫺2.5%, 5.9%, Table 2). Clinical cure rates for azithromycin microspheres and levofloxacin were also similar in the all-treated, clinically eligible, and BPP populations (Table 2). Clinical cure rates by baseline prognostic factors were similar between the 2 groups, with the exception of subjects with bilateral disease; in this group, 95.6% (109/114) of patients in the azithromycin group and 90.7% (98/108) in the levofloxacin group were considered to be clinical cures in the primary population. Among patients who were culture positive at baseline, clinical cure rates were consistent with those observed in the CPP population at TOC. Both azithromycin and levofloxacin were highly effective in patients with S. pneumoniae, H. influenzae, or M. catarrhalis (Table 3). Because follow-up sinus taps were not required, 229 of the 232 eradicated pathogens were assigned a bacteriologic response of presumed eradication. Among 81 S. pneumoniae isolated at baseline, the rates of penicillin susceptible, intermediate, and resistant isolates were 57%, 27%, and 16%, respectively. The rates of azithromycinsusceptible, -intermediate, and -resistant isolates were 85%, 0, and 15%, respectively. All pneumococcal isolates were susceptible to levofloxacin. Interestingly, 2 of the 3 patients with documented eradication of baseline pathogens were considered to be clinical failures. One patient was in the levofloxacin treatment arm and had E. sakazakii isolated at baseline, whereas the follow-up culture yielded no growth. A 2nd pa-

Of all patients randomized to azithromycin, 99.6% (270/ 271) complied with the prescribed regimen, compared with 96.3% (260/271) of those in the levofloxacin treatment arm. Of those patients fully compliant with prescribed study treatments, rates of clinical cure were 91.1% (246/270) and 90.0% (234/260) for azithromycin and levofloxacin, respectively.

Safety Treatment-related AEs occurred in 23.3% (63/270) and 15.3% (41/268) of patients in the azithromycin and levofloxacin groups, respectively. Of the treatment-related AEs occurring in ⬎1% of subjects, gastrointestinal disturbances were the most common in both treatment groups and, in

Table 3 Clinical cure by pathogen in the BPP population at test of cure* Parameter

Azithromycin

Levofloxacin

Total number of patients with baseline pathogens Pathogen, n (%) S. pneumoniae H. influenzae M. catarrhalis

102

111

36/37 (97.3) 26/27 (96.3) 8/8 (100)

36/39 (92.3) 30/30 (100) 10/11 (90.9)

*Day 17-24.

198

Otolaryngology–Head and Neck Surgery, Vol 133, No 2, August 2005

Table 4 Clinical and bacteriological outcome in patients infected with S. pneumoniae isolates resistant to azithromycin (MIC > 0.5 ␮g/mL) Patient

Country

1

US

2 3 4 5

Chile Chile Poland Lithuania

Azithromycin MIC ⬎256

4 8 ⬎4 4

Penicillin MIC

Clindamycin Macrolide Clinical KB genotype response

2 (resistant)

Resistant

mef/ ermTR

Failure

1 (intermediate) 2 (resistant) 1 (intermediate) ⬍0.03 (sensitive)

Sensitive Resistant Intermediate Sensitive

mef ermB NA NA

Cure Cure Cure Cure

Bacteriologic response Documented eradication; E. coli and viridans streptococci isolated on repeat culture Presumed eradication Presumed eradication Presumed eradication Presumed eradication

All patients were in azithromycin treatment group. NA, not available. KB, Kirby-Bauer interpretation.

most instances, occurred with similar frequency (nausea: 4.4% versus 3.4%; abdominal pain: 2.6% versus 0.4%). The incidence of diarrhea and loose stools was higher among azithromycin-treated patients (11.1% versus 1.9%). However, the majority of cases (15/27; 56%) resolved in 1-2 days. Two patients in the azithromycin group (0.7%) and 1 patient in the levofloxacin group (0.4%) experienced vomiting that was considered to be treatment related. All cases were mild-to-moderate in severity. One azithromycin subject experienced vomiting on day 1, which occurred after the 30-minute observation period; the 2nd azithromycin subject vomited on day 6, while on levofloxacin placebo. In the levofloxacin group, 1 patient vomited on day 10. One patient (0.4%) in each treatment group experienced a serious AE; neither of these was considered treatment related. Four patients discontinued treatment because of treatment-related AEs: 1 in the azithromycin arm and 3 in the levofloxacin arm. The discontinuation in the azithromycin arm was attributed to stomach cramps on day 2. The discontinuations in the levofloxacin arm were attributed to an allergic reaction on day 1, a cutaneous rash on day 8, and worsening sinusitis on day 12, for which the patient received additional antibiotic therapy. One additional patient in the azithromycin arm discontinued therapy for a single day on day 5 because of severe diarrhea, which subsequently resolved on day 7. Because this patient was treated with azithromycin on day 1, the discontinued treatment was levofloxacin placebo.

DISCUSSION In this international, multicenter, randomized, double-blind, double-dummy study, a novel azithromycin microsphere formulation, given as a single 2.0-g dose, was as clinically effective as a 10-day course of levofloxacin, 500 mg/d, for the treatment of adults with clinical and radiologic evidence of ABS.

Pathogens were isolated in 42% of patients, similar to isolation rates achieved in other studies in which sinus aspiration has been performed.16 As expected, S. pneumoniae, H. influenzae, and M. catarrhalis were the primary causative organisms, and clinical cure rates of ⱖ96% were achieved in azithromycin-treated patients with 1 of these key pathogens isolated at baseline. These findings are consistent with results from other studies in which azithromycin has demonstrated efficacy in the treatment of ABS when administered for periods of up to 6 days.2,7-10 In this study, 15% of all pneumococcal isolates were resistant to azithromycin, a rate lower than that reported in other studies.2 Of the 5 azithromycin-treated patients who had azithromycin-resistant pneumococci, only 1 was deemed a clinical failure. Of note, this patient’s follow-up sinus aspirate documented eradication of the pathogen. High-dose antibiotic therapy has been advocated for difficult-to-treat URTIs, in which infection may be caused by isolates with reduced antimicrobial susceptibility.17 With this single-dose formulation of azithromycin, the 24-hour AUC is nearly 4 times that achieved with conventional 3and 5-day dosing regimens. This front loading of the dose increases the AUC0-24/MIC, which is thought to be the best predictor of efficacy for azithromycin.12 Additionally, targeted delivery of the drug by white blood cells contributes to local concentrations in sinus fluid and mucosa for several days after dosing, which may enhance efficacy in ABS.18 Patients in both treatment arms were required to take study medication (active drug or placebo) for 10 days. Compliance with prescribed therapy was excellent in this study, higher than might be expected under routine conditions. Indeed, there is substantial evidence to suggest that compliance with medication declines in relation to the frequency and duration of antibiotic therapy,19 with patients rarely completing a full course of 7-10 days of therapy if symptoms begin to resolve within the first few days of treatment. A meta-analysis of several studies on bacterial URTIs, which included acute otitis media, ABS, and streptococcal pharyngitis, concluded that shorter courses of an-

Murray et al

Efficacy and Safety of a Novel, Single-dose. . .

tibiotics have the potential to increase patient compliance with therapy, decrease AEs (by decreasing drug exposure), decrease the emergence of resistant strains, and reduce cost.5,6 The importance of compliance is reflected in the recent FDA labeling for systemic antibiotics, which advises that skipping doses or not completing the full course of therapy may decrease the effectiveness of the treatment and increase the likelihood that bacteria will develop resistance.20 Single-dose antibiotic therapy helps to maximize patient compliance and therefore may reduce the likelihood of treatment failures and the emergence of resistant pathogens. Another compliance-related advantage of single-dose therapy with azithromycin microspheres is the potential for use as directly observed therapy. This option may be valuable in the clinic or emergency department, where compliance may be in doubt or barriers to filling prescriptions may exist. The microsphere regimen delivers a higher total dose than traditional 3- and 5-day azithromycin regimens, and the safety of administering a macrolide agent as a single, large dose might be of concern. However, azithromycin’s lack of interaction with the cytochrome P-450 system distinguishes it from other macrolides, which have been associated with drug– drug interactions and serious cardiac effects.21,22 In this trial, the microsphere formulation demonstrated an excellent safety profile, with no treatmentrelated serious AEs reported. Only 1 patient in each treatment group discontinued therapy prematurely because of AEs in this study, reflecting good toleration of study drugs. Patients in the azithromycin treatment arm experienced a higher incidence of diarrhea and loose stools compared with those receiving levofloxacin, but in the majority of patients, this resolved within 1 to 2 days of treatment. The incidence of diarrhea in the present study is lower than rates (ⱕ21%) observed in an earlier ABS study in which azithromycin was dosed for up to 6 days.2 In this study, in which stringent clinical and radiographic diagnostic criteria were used to confirm the diagnosis of ABS, a single 2.0-g dose of azithromycin microspheres was as effective as 10 days of levofloxacin, 500 mg/d, when used to treat patients with uncomplicated ABS. Among azithromycin-treated patients with S. pneumoniae, H. influenzae, and M. catarrhalis identified at baseline, cure rates were ⱖ96%. Treatment with azithromycin was associated with a higher incidence of diarrhea or loose stools; however, the majority of cases resolved within 2 days. As this study has demonstrated, a novel microsphere formulation of azithromycin is an effective alternative to fluoroquinolones for treatment of ABS and offers the added benefit of single-dose therapy, which maximizes compliance. The authors acknowledge the following statisticians for their contributions to study planning and analysis: Jennifer Garner,

199

MS; Rick Whaley, PhD; Linda Shurzinske, MS; and Harry Haber, MPH.

REFERENCES 1. Brook I. Antimicrobial management of acute sinusitis: a review of therapeutic recommendations. Infect Med 2002;19:231–7. (Grade B). 2. Henry DE, Riffer W, Sokol N, et al. Randomized double-blind study comparing 3- and 6-day regimens of azithromycin with a 10-day amoxicillin-clavulanate regimen for treatment of acute bacterial sinusitis. Antimicrob Agents Chemother 2003;47:2770 – 4. (Grade A). 3. Ahuja GS, Thompson J. What role for antibiotics in otitis media and sinusitis? Postgrad Med 1998;104:93–99, 103– 4. (Grade D). 4. Snow V, Mottur-Pilson C, Hickner JM. Principles of appropriate antibiotic use for acute sinusitis in adults. Ann Intern Med 2001;134: 495–7. (Grade B). 5. Pichichero ME. Short course of antibiotic in acute otitis media and sinusitis infections. J Int Med Res 2000;28:25A–36A. (Grade D). 6. Pichichero ME. Short course antibiotic therapy for respiratory infections: a review of the evidence. Pediatr Infect Dis J 2000;19:929 –37. (Grade D). 7. Klapan I, et al. Azithromycin versus amoxicillin/clavulanate in the treatment of acute sinusitis. Am J Otolaryngol 1999;341:233–9. (Grade A). 8. Clement PA, de Gandt JB. A comparison of the efficacy, tolerability and safety of azithromycin and co-amoxiclav in the treatment of sinusitis in adults. J Int Med Res 1998;26:66 –75. (Grade A). 9. Felstead SJ, Daniel R. Short-course treatment of sinusitis and other upper respiratory tract infections with azithromycin: a comparison with erythromycin and amoxycillin. European Azithromycin Study Group. J Int Med Res 1991;19:363–72. (Grade A). 10. Haye R, et al. Azithromycin versus placebo in acute infectious rhinitis with clinical symptoms but without radiological signs of maxillary sinusitis. Eur J Clin Microbiol Infect Dis 1998;17:309 –12. (Grade A). 11. Rothermel CD. Single-dose azithromycin for acute otitis media: a pharmacokinetic/pharmacodynamic rationale. Curr Ther Res 2003; 64(Suppl 1):4 –15. (Grade B). 12. Craig WA. Postantibiotic effects and the dosing of macrolides, azalides, and streptogramins. In: Zinner SH, Young LS, Acar JF, Neu HC, editors. Expanding indications for the new macrolides, azalides and streptogramins. New York: Marcel Dekker; 1997. p. 27–38.(Grade B). 13. Babl FE, Pelton SI, Li Z. Experimental acute otitis media due to non-typeable Haemophilus influenzae: comparison of high and low azithromycin doses with placebo. Antimicrob Agents Chemother 2002;46:2194 –2199. (Grade D). 14. Girard D, Finegan SM, Clmochowski CR, et al. Accelerated dosing of azithromycin in preclinical infection models. 102nd American Society for Microbiology General Meeting, Orlando, FL. American Society for Microbiology, Washington, DC, Abstract. #A-57., 2002. (Grade D). 15. Kamicker BJ, Bertsche CD, Medina IA. In rodent models, a single dose of azithromycin was more effective than dosing split over 3 or 5 days. In: 44th Annual meeting of Interscience Conference on Antimicrobial Agents and Chemotherapy, American Society for Microbiology, Washington DC. Abstract B-1181, 2004. (Grade D). 16. Hansen JG, Schmidt H, Rosborg J, et al. Predicting acute maxillary sinusitis in a general practice population. BMJ 1995;311:233– 6. (Grade B). 17. Arrieta A, Arguedas A, Fernadez P, et al. High-dose azithromycin versus high-dose amoxicillin-clavulanate for treatment of children with recurrent or persistent acute otitis media. Antimicrob Agents Chemother 2003;47:3179 – 86. (Grade A). 18. Karma P, Pukander J, Penttila M. Azithromycin concentrations in sinus fluid and mucosa after oral administration. Eur J Clin Microbiol Infect Dis 1991;10:856 –9. (Grade C).

200

Otolaryngology–Head and Neck Surgery, Vol 133, No 2, August 2005

19. Sclar DA, Tartaglione TA, Fine MJ. Overview of issues related to medical compliance with implications for the outpatient management of infectious diseases. J Inst Med Res 1994;28:25A–36A. (Grade D). 20. FDA Labeling Requirements for Systemic Antibacterial Drug Products. Available at http://www.fda.gov/oc/opacom/hottopics/anti_resist.html. Accessed on June 28, 2004. 21. Ray WA, Murray KT, Meredith S, et al. Oral erythromycin and the risk of sudden death from cardiac causes. N Engl J Med 2004;351:1089 – 96. (Grade B). 22. Rubenstein E. Comparative safety of the different macrolides. Int J Antimicrob Agents 2001;18:S71– 6. (Grade D).

Editorial Commentary: Dilemma in Trial Design: Do Current Study Designs Adequately Evaluate Effectiveness Antibiotic in ABRS? Acute bacterial rhinosinusitis (ABRS) is estimated to be one of the most common health problems in the United States and remains a leading reason for the use of antibiotics.1 It is, therefore, not surprising that interest regarding appropriate antibiotic selection is high. The article “Efficacy and safety of a novel, single-dose azithromycin microsphere formulation versus 10 days of levofloxacin for the treatment of acute bacterial sinusitis in adults,” by Murray et al, offers an example of a study designed to address the issue of antibiotic efficacy. The ultimate outcome of this study suggests both azithromycin microspheres and levofloxacin as administered are capable of resulting in resolution of clinical ABRS in ⬎90% of cases. At initial glance, efficacy rates of this magnitude appear to reinforce confidence in the 2 comparator drugs as reasonable, if not superior, treatment choices for ABRS. Closer examination of the design of this study, however, should elicit a certain degree of appropriate concern. Foremost among the questions that arise is whether this study actually provided a fair assessment of the impact of the drugs that were used. This exercise can be thought of simply as a form of cognitive quality control, which constitutes an essential component of the peer review process. Ample literature exists supporting the positive effect of antibiotics in the clinical treatment of ABRS. Perhaps one of the better current sources of information is the 2002 Cochrane Database Review assessing the effect of antibiotics in the treatment of acute maxillary sinusitis.2 This metaanalysis analyzed the results of 49 well-designed comparator-controlled or placebo-controlled antibiotic trials. Overall, data from this review supported the use of antibiotics over placebo in the treatment of ABRS, and noted that clinical success was high (84% clinical resolution), regardless of the antibiotic that was used. In summary, the Cochrane review suggested that although antibiotics can help relieve uncomplicated sinus infections, they do not appear to significantly impact clinical cures. These observations were supported by the findings of the Agency for Health Care Policy and Research (AHCPR), which performed an equally rigorous analysis of available high-grade literature that yielded no evidence of clinical superiority of one antibiotic over any other.3,4

Variations in trial design can potentially impact the apparent outcome of such studies and should be carefully considered. The AHCPR report noted that the majority of antibiotic trials that were used in their analysis measured outcomes in terms of the “extent of persisting symptoms at the end of treatment, which typically occurred about 10 days after the onset of treatment”.3,4 The importance of choosing an appropriate time to measure treatment outcome becomes even more important in disease states that enjoy a high rate of spontaneous resolution. In the case of ABRS, placebocontrolled studies suggest a rate of spontaneous resolution at 10 days of as high as 72%.3,5 Presumably this rate climbs even higher as the time elapses beyond the 10-day point. This realization led the AHCPR to state, “given that rhinosinusitis usually resolves spontaneously within weeks, inferring efficacy from one or two time points (“snapshots”) is insufficient to quantify the benefits of treatment. It would be better to follow the rate of resolution of symptoms (e.g., with a Kaplan-Meier plot) so that the benefits, if any, of a treatment could be expressed as a reduction in symptomdays.”3 Although Murray et al should be congratulated for strict adherence to the trial design recommended by the US Food and Drug Administration for evaluation of antibiotics in the treatment of acute maxillary sinusitis,6 their study appears to offer an excellent example of the concerns expressed by the AHCPR. Their study was designed as a randomized, double-blind, double-dummy drug comparator trial intended to demonstrate noninferiority when comparing a single dose of a microsphere formulation of azithromycin with that of a 10-day course of levofloxacin in patients suffering from ABRS. Patients were enrolled after suffering symptoms of ABRS for at least 7 but not greater than 28 days. The primary end point then was chosen as clinical resolution of disease (test of cure) at 17 to 24 days after initiation of therapy. This resulted in the test of cure assessment occurring no earlier than 24 days, but as late as 52 days, from the initial onset of symptoms. Given the recognized spontaneous resolution rate of ABRS, it becomes clear that any potential impact of antibiotics in this study was likely obscured by the natural history of the underlying disease process. This dilemma of improper study design as applied to evaluating the impact of antibiotics in the management of ABRS has been previously raised. Systematic reviews of literature, such as the Cochrane Review and AHCPR report, highlight the need for improvement in the quality of studies evaluating the treatment options for ABRS.2,3 The AHCPR attempted to address this problem by offering several suggestions for future research that focus upon the translation of evidence into useful clinical information. Study populations should be rigorously and objectively selected to insure internal validity and allow for fair data analysis. Clinical outcomes should be explicitly defined and scored. Further, given information that suggests the major effect offered by antibiotics is a decrease in the duration of symptoms rather