What is the role of long‐term macrolide therapy in the treatment of recalcitrant chronic rhinosinusitis?

Long-duration therapy with macrolide antibiotics has been advocated for the treatment of recalcitrant chronic rhinosinusitis (CRS). The rationale for this treatment is based on the unique anti-inflammatory properties of macrolides rather than a strict antibacterial effect. However, uncertainty exists as to which patients will respond to such treatment, the degree of benefit likely to be obtained, and the relevant risks to the patient and community at large. There is substantial in vitro and in vivo evidence that 14- and 15-membered ring macrolide antibiotics (azithromycin, clarithromycin, erythromycin, roxithromycin) have intrinsic anti-inflammatory properties in both the upper and lower airways.1 The majority of this effect involves the inhibition of neutrophilic inflammation. Studies have shown that macrolides interfere with the neutrophil's ability to migrate toward and adhere to sites of inflammation. Macrolides have also been shown to reduce the neutrophil's oxidative burst and increase the rate of programmed cell death. The anti-inflammatory effect appears to extend beyond simply neutrophils, as macrolides also inhibit a wide range of inflammatory cytokines including interleukin-1b, interleukin-8, nuclear factor kB, transforming growth factor B, and granulocyte-macrophage colony stimulating factor. Studies have also suggested that macrolides may inhibit formation of pseudomonal biofilms, decrease airway mucus secretion, and even have a reparative effect on inflamed airway mucosa. Early clinical reports of macrolide use in CRS come predominantly from the Japanese literature. These studies consistently demonstrate improvements in a wide variety of sinus-related symptoms, endoscopic findings, and computed tomography (CT) scores after prolonged courses of macrolides (range, 2 weeks–12 months). The weakness of these studies is that they were small (10–45 subjects), open-label, and uncontrolled, which makes it difficult to know whether the improvements seen were a result of placebo effect and/or spontaneous disease resolution. The first English-language publication came from Hashiba and Baba in the mid-1990s.2 They treated 45 patients with 400 mg of clarithromycin per day for 8 to 12 weeks and reported findings at several time points. After 12 weeks of therapy, 51% had reduced nasal obstruction, 62% reported improved postnasal drainage, and 64% had thinning of nasal secretions. Interestingly, overall improvements were incremental over time, with 5% reporting improvement at 2 weeks, 48% at 4 weeks, and 71% at 12 weeks. Suzuki et al. investigated the prognostic factors for those undergoing low dose macrolide therapy.3 In 2000, they treated 16 CRS patients with either 200 mg of clarithromycin or 150 mg of roxithromycin daily for 8 to 12 weeks. At baseline, they assessed serum IgE, peripheral blood leukocyte count, nasal cytology, and CT scores. Patients with elevated levels of IgE (>250 U/mL) showed significantly less symptomatic improvement than those with low serum IgE (4.9 ± 3.1% vs. 42.1 ±11.2; P = .046). The symptomatic improvement rate also inversely correlated with the eosinophil count in the peripheral blood (r = −0.51; P = .02), nasal smear (r = −0.54; P = .045), and sinus mucosa (r = −0.5; P = .02). No other factors correlated with outcomes. The most informative clinical study of macrolide use in CRS is the double-blind, placebo-controlled trial performed by Wallwork et al. in 2006.4 In this study, either 150 mg of roxithromycin or placebo was administered daily to 29 and 35 subjects, respectively, over a 2- to 3-month period. Outcome measures included peak inspiratory flow, saccharin transit time (a test of mucociliary clearance), nasal lavage assays (IL-8, a2-macroglobulin, fucose), olfactory function, endoscopy scores, and the sinonasal outcome test-20 (SNOT-20). At the end of treatment, macrolide-treated subjects had significant improvements in IL-8 level, saccharin transit time (mucociliary clearance), and nasal endoscopy. There was no improvement on any outcome measure in the placebo-treated group. SNOT-20 scores were evaluated after 6 weeks of treatment, after 12 weeks of treatment, and then 12 weeks after completion of treatment. Interestingly, a significant improvement in SNOT-20 score was seen at 12 weeks (end of treatment), but not after 6 weeks of treatment. The level of improvement seemed to fade 12 weeks after completion of treatment, with the degree of improvement no longer reaching statistical significance. Of note, a subgroup analysis was done comparing improvement in SNOT-20 scores in those with low serum IgE (<200 μg/L) and those with high serum IgE (>200 μg/L). The low IgE group had a more robust improvement, which was statistically significant at 12 weeks and appeared durable 12 weeks after treatment completion (although this did not quite reach significance with a P value of .06). The high-IgE group did not have improvements in SNOT-20 score at any point in time. As a global measure of improvement, the authors asked subjects to rate their overall response to treatment. Placebo-treated subjects responded that they were unchanged, whereas, macrolide-treated subjects responded that they were slightly improved, with the subgroup with low IgE values reporting that they were slightly-to-much improved. The use of low-dose antibiotics for prolonged time periods has theoretical risks including the development of bacterial resistance both in the individual patient and in the community at large. Several studies, although small, have evaluated sinonasal culture data before and after prolonged macrolide use and have not documented the development of resistant strains in the individual patient.5 Prolonged macrolide therapy has not been studied on a population level and one must assume that widespread use would lead to increased resistance in the community at large. Low-dose, long-duration macrolide therapy is a viable option for patients with CRS refractory to standard medical and surgical therapies. Based on the limited evidence available, daily clarithromycin (250 mg), azithromycin (250 mg), or roxithromycin (150 mg) should be continued for at least 12 weeks to achieve measurable results. Patients should be selected carefully, as those with atopy and eosinophilic inflammation are unlikely to respond. Patients should understand that improvements are likely to be modest, and symptoms will likely not be fully controlled. At this point in time, further studies are needed to determine whether treatment longer than 12 weeks is beneficial. Similarly, future studies are needed to document the long-term outcomes after cessation of therapy. Available evidence suggests that low-dose macrolide therapy is safe for the individual patient, but clinicians should remain cognizant of the potential for bacterial resistance in the population at large. In this review, one level 1 study (Randomized Controlled Trial) and several level 4 studies (cohort studies without comparison groups) were reviewed.