Chronic rhinosinusitis (CRS) is a complex disorder that has previously described conditions ranging from isolated odontogenic sinusitis to broader chronic airway inflammation. The current definition of primary CRS includes widespread inflammation of all paranasal sinuses for at least 12 weeks [1,2]. CRS has traditionally been classified into two phenotypes: with nasal polyps (CRSwNP) or without nasal polyps (CRSsNP) . However, these two subtypes of CRS can experience overlap in their inflammatory profiles, clinical presentation, and effects of treatment .
The standard treatment for primary CRS requires a combination of medical and surgical treatment, including functional endoscopic sinus surgery (FESS) and postoperative saline irrigations . While systemic anti-inflammatory medications are used in primary medical management in CRS, systemic and topical corticosteroids are commonly used in postoperative care and long-term management of the condition [4,5▪▪,6▪]. Typically, patients with Th2-mediated eosinophilic CRS respond well to local corticosteroid therapy with reduction in polyp size, nasal congestion, total nasal symptom score, and nasal resistance. However, there is a group of patients with noneosinophilic CRS who do not respond to corticosteroids and have been found to benefit from long-term, low-dose macrolide treatment [5▪▪,7].
Macrolide antibiotics are known for their anti-inflammatory and immunomodulatory effects. The immunomodulatory effects lead to inhibition of proinflammatory cytokines, including IL-8 and TNF-α, inhibition of neutrophil adhesion and migration, and changes to mucus synthesis and secretion [1,4,5▪▪].
Currently, macrolide therapy in the management of CRS is controversial. The evidence supporting macrolides in CRS is mixed, the studies are heterogenous, and the types (14-membered and 15-membered lactone ring macrolides), dosages (half dose and very low dose), and duration of treatment vary. The European Position paper on Rhinosinusitis and Nasal Polyps 2012  recommends the use of long-term, low-dose macrolide therapy for CRSsNP, while the International Consensus Statement on Allergy and Rhinology  recommends macrolides for both CRSsNP and CRSwNP subtypes.
ANTI-INFLAMMATORY EFFECT OF MACROLIDES
The inhibitory effects of EM900, derivative of erythromycin, has recently been compared with those of clarithromycin (CAM) on inflammatory cytokine production in human nasal epithelial cells. IL-8 and vascular endothelial growth factor (VEGF), that is produced by the inflammatory epithelial cells and fibroblasts, were measured using real-time-PCR and an ELISA. IL-8 and VEGF play important roles in the pathogenesis of paranasal inflammation, and their effects can be suppressed with steroids. IL-8 is a representative inflammatory cytokine that particularly activates neutrophils, and VEGF is believed to participate in the initial steps of nasal polyp formation by the outgrowth of the sinus edematous mucosa. An ELISA and real-time-PCR confirmed that TNF-α stimulation significantly increased the production of IL-8 and VEGF. However, both EM900 and CAM significantly inhibited IL-8 production, but did not suppress the increased VEGF production. Therefore, EM900 can demonstrate promising antineutrophil inflammatory effects in CRS that are similar to those observed using CAM.
MACROLIDES IN CHRONIC RHINOSINUSITIS WITH NASAL POLYPS AND CHRONIC RHINOSINUSITIS WITHOUT NASAL POLYPS
Macrolides in noncorticosteroid responders
A recently conducted case–control study aimed at defining a CRS phenotype suitable to macrolide therapy [5▪▪]. CRSwNP and CRSsNP patients were placed on a 3-month low-dose CAM 250 mg daily therapy after failing at least 3 months of corticosteroid irrigation therapy post-FESS. Macrolide responses were evaluated with endoscopic and nasal symptom scores. Tissue eosinophilia (>10/high powered field) was inversely associated with macrolide response (17.7 vs. 62.5%, P = 0.02) and serum eosinophil level was lower in responders (0.16 ± 0.11 vs. 0.39 ± 0.36, P = 0.03). Tissue neutrophilia (presence of focal or diffuse neutrophilic infiltrate) was not significantly different (58.8 vs. 37.5%, P = 1.0) and neutrophil level was similar between groups (4.3 ± 2.4 vs. 3.8 ± 2.2, P = 0.67). Squamous metaplasia was overexpressed in nonresponders (0 vs. 37.5%, P = 0.01). Therefore, low-tissue and serum eosinophilia and absence of tissue squamous metaplasia could indicate a CRS phenotype suitable for a trial of long-term macrolide therapy when surgery and topical corticosteroid therapy have failed.
Macrolides vs. topical corticosteroid
A randomized controlled trial (RCT) stratified 187 Chinese patients with CRS in CRSsNP, eosinophilic CRSwNP, and noneosinophilic CRSwNP groups and randomized them to receive fluticasone propionate nasal spray (200 μg) or CAM (250 mg daily) for 3 months following FESS . All patients used nasal saline irrigation (250 ml twice a day) along with oral prednisone rescue therapy if polyps and/or symptoms of CRS recurred. Patients were assessed before surgery and 1, 3, 6, and 12 months after initiation of treatment. The primary outcome measure was the total visual analogue score (VAS) symptom score and secondary outcomes included the individual symptom VAS scores, overall symptom burden scores, total endoscopic scores, and individual endoscopic domain scores. There was no significant difference in the baseline total VAS symptom scores, individual symptom VAS scores, overall burden of symptoms, or total endoscopic scores between fluticasone propionate and CAM group at 1, 3, 6, and 12-month follow-ups. While this RCT randomized patients in all three groups, there was no subgroup analysis on macrolide response in those corticosteroid nonresponders/minimal responders. The use of systemic corticosteroids ad hoc in the postoperative period also confounds the results.
Another study compared the effects of long-term, low-dose erythromycin (250 mg) twice a day with the use of mometasone furoate nasal spray for 12 weeks in postoperative CRS patients with persistent rhinosinusitis . Nasal saline irrigation was performed daily but no other medication was prescribed. Patients were assessed with the Taiwanese version of the 22-item Sino-Nasal Outcome Test (TWSNOT-22) questionnaire, endoscopic examination [Lund–Mackay score (LMS)], acoustic rhinometry measuring the second minimal cross-sectional area (MCA2) of the nasal cavity, smell test using smell threshold test (Sensonics, Inc., Hadden Heights, New Jersey, USA), and saccharine transit test before and after 12 weeks of treatment. The endoscopic score was significantly higher in the erythromycin group preoperatively and before treatment (P = 0.002 and 0.001), but displayed no significant difference between groups after treatment. In the erythromycin group, no significant difference was observed in the TWSNOT-22 scores, mean MCA2, University of Pennsylvania Smell Identification Test - Traditional Chinese (UPSIT-TC) scores, or bacterial culture rates after treatment; however, the endoscopic scores and smell thresholds significantly decreased. In the intranasal steroid group, no significant difference was observed in the TWSNOT-22 scores, smell thresholds, UPSIT-TC scores, or bacterial culture rates after treatment. However, the endoscopic scores significantly decreased, while the mean MCA2 significantly increased. The improvement rate of saccharine transit time was significantly higher in the erythromycin group (64.9 vs. 35.1%, P < 0.01). Therefore, erythromycin treatment was beneficial in the management of postoperative persistent rhinosinusitis; however, this study did not compare outcomes between the two groups and cannot determine superiority between treatment arms.
Macrolides + corticosteroid vs. corticosteroids
Deng et al. performed an RCT to evaluate CAM response in CRSwNP and CRSsNP patients without FESS. One group was treated with CAM 0.25 g/day (clarithromycin + budesonide (BUD) group; Klacid, Abbott, Lake Bluff, Illinois, USA) and topical budesonide 256 μg/day (BUD group; Rhinocort, AstraZeneca, Cambridge, UK) while the control group received only budesonide 256 μg/day for 3 months. The primary outcome was VAS for five major symptoms and a general nasal symptom score. Secondary outcomes included SNOT-22, LMS, and Lund–Kennedy score. Nasal secretions were analyzed before and after the treatment to evaluate the concentration of cytokines IL-1β, IL-2, IL-4, IL-5, IL-6, IL-8, and TNF-α. VAS scores improved significantly in both groups, without significant difference between groups. Although the change in SNOT-22 was greater in the CAM /budesonide group, the results were NS when compared with the budesonide group. Similarly, there was no significant difference in LMS or Lund–Kennedy score between groups. IL-6 and IL-8 were higher in patients who responded to macrolide therapy and were markedly reduced posttreatment in the CAM and budesonide group, suggesting that macrolides may be more effective in patients with increased IL-6 and IL-8.
Meta-analysis of macrolides in chronic rhinosinusitis
Recent meta-analysis with 10 studies assessed prognostic factors of macrolide therapy that may predict favorable clinical outcomes of low-dose macrolides in CRS. The primary outcome was SNOT-22. Secondary outcomes included CRS subtypes, serum IgE level, membered lactone ring of macrolides, concurrent FESS, and dosage and duration of the macrolides [6▪]. Macrolides and placebo were not different in SNOT improvement, symptom score, and endoscopy score. There was no difference in the SNOT improvement between macrolides and standard treatment compared with standard treatment. Similarly, no difference was found between macrolides and intranasal steroid spray in the improvement of symptom score and endoscopy score. Subgroup analyses showed that the effects favored macrolides over placebo in SNOT improvement in CRSsNP patients but not in CRSwNP patients [standardized mean difference (SMD) = −0.64, 95% confidence interval (CI): −1.01 to −0.27 vs. SMD = 0.18, 95% CI: −0.19–0.55, P = 0.009]. Similarly, symptom scores improved in CRSsNP patients but not in CRSwNP patients (SMD = −0.89, 95% CI: −1.41 to −0.37 vs. SMD = 0.31, 95% CI: −0.21–0.83, P = 0.001). Compared with placebo, macrolides showed greater symptom improvement when administered to patients without FESS over patients with FESS (MD = −0.89, 95% CI: −1.41 to −0.37 vs. mean difference = 0.31, 95% CI: −0.21–0.83, P = 0.001). SNOT and endoscopy score improved in patients with FESS but not in patients without FESS (SMD = −1.68, 95% CI: −2.40 to −0.95 vs. SMD = −0.01, 95% CI: −0.63–0.61, P < 0.001, and SMD = −3.79, 95% CI: −4.85 to −2.73 vs. mean difference = 0.02, 95% CI: −0.46–0.51, P < 0.001, respectively). Regarding the dose of macrolides, improvement in SNOT was greater in patients who received a half dose of macrolides than those who received a less than half dose (SMD = −0.64, 95% CI: −1.01 to −0.27 vs. SMD = 0.18, 95% CI: −0.19–0.55; P = 0.002). A half dose of macrolides also had better effects on symptom score (SMD = −0.89, 95% CI: −1.41 to −0.37 vs. mean difference = 0.31, 95% CI: −0.21–0.83; P = 0.001). As for duration, SNOT improved in patients who received 24-week macrolides compared with those who received them in 12 and 8-week durations (SMD = −1.68, 95% CI: −2.40 to −0.95 vs. SMD = −0.28, 95% CI: −0.77–0.21 vs. mean difference = 0.36, 95% CI: −0.33–1.04, respectively; P = 0.002). Symptom improvement also favored patients receiving 24-week treatments than 12 or 8-week ones (SMD = −1.65, 95% CI: −2.37 to −0.93 vs. mean difference = −0.10, 95% CI: −0.59–0.38 vs. mean difference = −0.29, 95% CI: −0.73–0.15, respectively; P = 0.001) and better results were observed on endoscopic score in 24-week treatments than in 12-week ones (SMD = −3.79, 95% CI: −4.85 to −2.73 vs. mean difference = 0.02, 95% CI: −0.46–0.51; P < 0.001). There was no difference between the 14-membered and 15-membered ring macrolides. The assessment of concurrent FESS had mixed results and serum IgE level could not be assessed. The findings suggested that macrolide therapy provided clinical effectiveness to CRSsNP patients and, a half-dose for at least 24 weeks duration is recommended.
A meta-analysis including 17 RCTs assessed the safety and efficacy of oral CAM for CRS treatment. Primary outcomes included total nasal symptom scores, VAS, the Lund–Kennedy endoscopic score (LKES), and LMS. Secondary outcomes were inflammatory mediators and adverse events. Length of outcome measurement was defined as follows: short-term (<1 month), medium-term (1–3 months), and long-term (>3 months) . When CAM was added to nasal corticosteroid, it improved clinical symptoms in the medium term (SMD = −0.85; 95% CI: −1.42 to −0.29, P < 0.01). For patients post-FESS, combined therapy showed greater clinical symptoms improvement in the long term. In short-term, medium-term, and long-term LKES improved when CAM was added to corticosteroids (SMD = −1.09; 95% CI: −1.34 to −0.84 vs. SMD = −0.80; 95% CI: −1.07 to −0.52 vs. SMD = −3.93; 95% CI: −5.03 to −2.83). Combined therapy also improved LMS in the short-term (MD = −1.05; 95% CI: 1.27 to −0.82) and medium-term (SMD = −1.87; 95% CI: −2.27 to −1.47). When comparing CAM alone with corticosteroid, no significant difference was identified between groups regarding symptoms, LKES, and LMS. High-dose oral CAM (500 mg, twice daily) was significantly better than low-dose (250 mg, once daily) for the short-term as for clinical symptoms and LKES evaluation in CRSsNP patients without FESS (P < 0.025). However, short-term antibiotic therapy is typically used to treat an acute sinusitis or acute exacerbation of chronic sinusitis. This analysis showed that adding oral CAM to nasal corticosteroid spray may achieve better results than nasal corticosteroid spray alone for CRS patients.
Another meta-analysis analyzed seven RCTs and four cohort trials. The outcomes analyzed included SNOT-22, endoscopic and computed tomography (CT) scores. The subgroup of cohort trials indicated a significant difference in SNOT scores between the macrolide-treated and control groups at the 8-week follow-up (SMD = −5.50; 95% CI: −9.60, −1.40; P = 0.009). SNOT scores at 12-weeks of macrolide treatment demonstrated a significant improvement in the subgroup of trials in Asian patients (SMD = −0.51; 95% CI: −0.96, −0.02; P = 0.04), but not in non-Asians. No significant difference in SNOT was noted at the 12 or 24-week follow-ups compared with the control group, in both RCTs and cohort trials subgroups. However, the LKES was significantly improved when macrolide was compared with placebo in the subgroup of non-RCT studies after 8 weeks (SMD = −0.77; 95% CI: −1.07, −0.46; P < 0.01) and 12 weeks (SMD = −1.40; 95% CI: −1.97, −0.82; P < 0.01). Similarly, the LMS showed significant improvement compared with the baseline after the 12-week treatment (SMD = −5.81; 95% CI: −8.10, −3.52; P < 0.01) in the cohort trials . Their findings support the claim that macrolides may be beneficial in improving endoscopic and CT scores in CRS patients.
MACROLIDES IN CHRONIC RHINOSINUSITIS WITH NASAL POLYPS
Sireci et al. recently suggested that a prolonged treatment with low-dose CAM is effective for patients with recalcitrant CRS associated with asthma and/or aspirin-exacerbated respiratory disease and/or atopy in those who did not respond well to FESS and postoperative topical steroid therapy and nasal irrigation. In their study, 10 patients with CRSwNP who failed topical steroid therapy and nasal irrigation were treated with CAM 500 mg/daily for 3 days a week for 1 month. The primary outcomes included SNOT-22 and endoscopic appearance score (EAS) at 1-month post treatment. SNOT-22 showed improvement in blowing the nose (2.52 ± 0.87 vs. 1.3 ± 0.67, P < 0.01), sneezing (2.1 ± 0.99 vs. 0.7 ± 0.94, P < 0.01), hyposmia (3.1 ± 1.4 vs. 1.7 ± 1.49, P = 0.04), and dense mucous discharge (3.5 ± 1.17 vs. 1.5 ± 0.97, P < 0.01), while EAS had an improvement in secretions (1.6 ± 0.69 vs. 0.3 ± 0.48, P < 0.01) and edema (1.3 ± 0.48 vs. 0.2 ± 0.42, P < 0.01). Patients reported no worsening of symptoms 6 months after treatment, and the SNOT 22 and EAS showed no significant changes compared with the end of therapy.
An in-vitro study aimed at investigating the effects of erythromycin on cell proliferation, apoptosis, and the expression of mitogen-activated protein kinase (MAPK) and extracellular signal-regulated kinase (ERK) as well as the activation of the ERK/MAPK signaling pathway on cultured nasal polyp cells (NPC) compared cells from nasal polyp with cells derived from normal nasal tissue (NNC) . Cells were divided into four groups: the control, the erythromycin-treated (100 μmol/l); the selumetinib-treated (2 nmol/l); and erythromycin + selumetinib-treated group. Ki-67 expression in NPC was significantly increased compared with NNC, thereby suggesting that increased cell proliferation might be involved in the pathogenesis of nasal polyps. Apoptotic rate in NPC was significantly lower than that of NNC. The expression of B cell lymphoma-2 mRNA was increased, while the expression of BAX mRNA was decreased in NPC. The expressions of phosphorylated MEK1 (p-MEK1) and p-ERK1 were increased in NPC compared with NNC, thereby indicating that the activity of ERK/MAPK pathway was enhanced in nasal polyps. NPC in the erythromycin-treated group was significantly lower when compared with the untreated group and was dose-dependent. Erythromycin treatment was significantly associated with the apoptosis of NPC, and downregulated the expression of p-MEK1 and p-ERK1 in NPC. Furthermore, the MEK1 inhibitor selumetinib significantly inhibited the expression of p-MEK1 and p-ERK1 and exhibited a synergistic effect with erythromycin in reducing the activity of the ERK/MAPK signaling pathway. Selumetinib treatment had a synergistic effect with erythromycin to reduce the expression of p-MEK1 and p-ERK1, reduce cell proliferation, and increase cell apoptosis. This finding indicates that erythromycin may play a role in the management of patients with CRSwNP; however, more in-vivo studies are required to confirm this hypothesis. This study did not assess the tissue or serum eosinophilia of the nasal polyps, therefore further subtype analysis of these patients was not available.
Latest evidence of macrolide therapy in CRS has demonstrated that macrolides play an important role in a select group of patients, further highlighting the importance of appropriate patient selection for success of treatment. Patients with low-tissue and serum eosinophilia and unresponsive to corticosteroids (with or without polyps) have had improved outcomes with initiation of macrolide therapy. Treatment with a half dose of macrolides (i.e. CAM 250 mg daily) for at least 12–24 weeks has showed significant control of the underlying condition in CRS. The recommended tapering plans and success of these was not the focus of these articles. Our current practice is to initiate daily treatment for 3 months for noncorticosteroid responders, if successful, the macrolide is decreased to three times per week until 12 months of treatment.
Financial support and sponsorship
Conflicts of interest
Prof R.S. is a consultant for Medtronic and is on a speaker bureau for Meda Pharmaceuticals. All other authors have no financial disclosures or conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Huang Z, Zhou B. Clarithromycin for the treatment of adult chronic rhinosinusitis
: a systematic review and meta-analysis. Int Forum Allergy Rhinol 2019; 9:545–555.
2. Fokkens WJ, Lund VJ, Mullol J, et al. European position paper on rhinosinusitis and nasal polyps 2012. Rhinol Suppl 2012; 23:1–298.
3. Zeng M, Wang H, Long XB, et al. Comparison of efficacy of fluticasone propionate versus clarithromycin for postoperative treatment of different phenotypic chronic rhinosinusitis
: a randomized controlled trial. Rhinology 2019; 57:101–109.
4. Wu SH, Hsu SH, Liang KL, Jiang RS. The effects of erythromycin towards the treatment of persistent rhinosinusitis after functional endoscopic sinus surgery: a randomized, active comparator-controlled study. J Chin Med Assoc 2019; 82:322–327.
5▪▪. Oakley GM, Christensen JM, Sacks R, et al. Characteristics of macrolide responders in persistent postsurgical rhinosinusitis. Rhinology 2018; 56:111–117.
The study aimed at defining a chronic rhinosinusitis (CRS) phenotype suitable to macrolide therapy. Their results indicate a CRS phenotype suitable for a trial of long-term macrolide therapy when surgery and topical corticosteroid therapy have failed.
6▪. Seresirikachorn K, Suwanparin N, Srisunthornphanich C, et al. Factors of success of low-dose macrolides
in chronic sinusitis
: systematic review and meta-analysis. Laryngoscope 2019; 129:1510–1519.
The meta-analysis assessed prognostic factors of macrolide therapy that may predict favorable clinical outcomes of low-dose macrolides in CRS. Their results demonstrate favorable outcomes in patients with CRS without polyps.
7. Wakayama N, Matsune S, Takahara E, et al. Anti-inflammatory effects of EM900 on cultured human nasal epithelial cells. J Nippon Med Sch 2018; 85:265–270.
8. Orlandi RR, Kingdom TT, Hwang PH, et al. International consensus statement on allergy and rhinology: rhinosinusitis. Int Forum Allergy Rhinol 2016; 6: (Suppl 1): S22–S209.
9. Deng J, Chen F, Lai YY, et al. Lack of additional effects of long-term, low-dose clarithromycin combined treatment compared with topical steroids alone for chronic rhinosinusitis
in China: a randomized, controlled trial. Int Forum Allergy Rhinol 2018; 8:8–14.
10. Shen S, Lou H, Wang C, Zhang L. Macrolide antibiotics in the treatment of chronic rhinosinusitis
: evidence from a meta-analysis. J Thorac Dis 2018; 10:5913–5923.
11. Sireci F, Speciale R, Gallina S, et al. Clarithromycin in the management of chronic rhinosinusitis
: preliminary results of a possible its new use. Indian J Otolaryngol Head Neck Surg 2018; 70:87–91.
12. Liu X, Wang X, Chen L, et al. Effects of erythromycin on the proliferation and apoptosis of cultured nasal polyp-derived cells and the extracellular signal-regulated kinase (ERK)/mitogen-activated protein kinase (MAPK) signaling pathway. Med Sci Monit 2018; 24:8048–8055.