Favorable Response to Long-Term Azithromycin Therapy in Bronchiectasis Patients with Chronic Airflow Obstruction Compared to Chronic Obstructive Pulmonary Disease Patients without Bronchiectasis

Introduction

Frequent exacerbations of chronic obstructive pulmonary disease (COPD) are a major cause of morbidity and mortality, leading to a substantial healthcare burden.^1–4 To reduce exacerbations, the current COPD guidelines recommend pharmacological treatment with anti-inflammatory agents, including inhaled corticosteroids (ICS), roflumilast, and long-term macrolides.⁵ Among macrolides, long-term azithromycin added to the usual treatment reduced the frequency of exacerbations and improved quality of life,^6,7 particularly in those who are not current smokers.⁸ As this long-term macrolide therapy significantly reduces the incidence of non-cystic fibrosis bronchiectasis (NCFB) exacerbations,^9–11 NCFB management guidelines also recommend long-term macrolide therapy for bronchiectasis patients with frequent exacerbations,^12,13 regardless of the presence of chronic airflow obstruction (CAO).

With the wide availability of chest computed tomography (CT), previously unrecognized bronchiectasis is increasingly identified in patients with COPD. Indeed, bronchiectasis has been reported in 29–57% of patients with COPD, and the presence of bronchiectasis is associated with frequent exacerbations, COPD severity, and even increased mortality.^14–17 COPD is also a common comorbidity in patients with bronchiectasis, reporting up to 58% of prevalence of COPD diagnosis in patients with bronchiectasis using health-insurance claim data.^18,19 Another data from US bronchiectasis research registry also showed that airflow limitation was present in more than half patients of bronchiectasis patients.²⁰

Despite this frequent coexistence of COPD and bronchiectasis, previous studies on the benefits of long-term macrolides in patients with COPD did not assess the presence of bronchiectasis; thus, there is no available study to show the treatment efficacy according to the presence of bronchiectasis in COPD.^6–8 Similarly, it remains unclear whether coexisting CAO in bronchiectasis patients are associated with a greater benefit of long-term macrolide treatment than COPD without bronchiectasis.^9–11 Therefore, this study aimed to compare the effect of long-term azithromycin between bronchiectasis patients with CAO and COPD patients without bronchiectasis.

Methods

Study Population

Between July 2014 and October 2018, 82 patients who 1) received azithromycin for more than 12 weeks, and 2) were free of nontuberculous mycobacterial (NTM) pulmonary disease were retrospectively identified at a single referral hospital in South Korea. After excluding 21 patients with post-bronchodilator forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) ≥ 0.7 and two patients without post-bronchodilator spirometry, 59 patients with CAO, defined as post-bronchodilator FEV₁/FVC < 0.7 were included in this study. The diagnosis of bronchiectasis was based on the CT (bronchial dilatation, lack of bronchial tapering, or identification of bronchi within 1 cm of the pleura)²¹ and compatible symptoms. In this study, the group with bronchiectasis and CAO was named as bronchiectasis with CAO rather than COPD with bronchiectasis. Up to 39% of COPD patients are never smokers²² and they are diagnosed based on CAO. Nevertheless, so far, as there is no solid definition of COPD and bronchiectasis overlap, bronchiectasis with CAO would be more appropriate nomenclature.

This study was approved by the Institutional Review Board of the Samsung Medical Center (IRB No. SMC 2019–10-096), and the requirement for informed consent was waived due to the retrospective nature of the study. All patients’ data were anonymized and deidentified before the analysis, and this study was conducted according to the Declaration of Helsinki.

Data Collection and Clinical Assessment

Clinical characteristics, including demographics, symptoms, history of exacerbations, medications before initiating azithromycin, pulmonary function, laboratory findings, sputum culture, and adverse events of azithromycin therapy, were obtained from the medical records. Symptomatic burden and quality of life were assessed using the modified Medical Research Council (mMRC) and COPD Assessment Test (CAT).²³ The severity of bronchiectasis was evaluated using the Bronchiectasis Severity Index (BSI).²⁴

Spirometry was performed using Vmax 22 (SensorMedics, Yorba Linda, CA, USA) according to the ATS/European Respiratory Society criteria.²⁵ Absolute values of FVC and FEV₁ were obtained, and the percentage predicted values (%pred) were calculated using data obtained from a representative Korean sample.²⁶ A positive bronchodilator response was defined as a post-bronchodilator increase in FEV₁ of at least 12% and 200 mL from the baseline value.²⁷

The primary outcome of this study was the development of exacerbations. Moderate exacerbation was defined as an outpatient clinic visit, while severe exacerbation was defined as hospitalization or an emergency room visit owing to one or more of the following: worsening of dyspnea, increased sputum volume, and purulent sputum. Exacerbations were treated with short-acting bronchodilators and antibiotics with or without systemic corticosteroids. The secondary outcome was symptom improvement during azithromycin treatment. An asymptomatic responder was defined as a patient who achieved a minimum clinically important difference (MCID) for CAT, which is a ≥2 point decrement from the initial CAT score.²⁸ We used the CAT score measured at 3, 6, and 12 months after azithromycin initiation.

Statistical Analysis

Categorical variables were analyzed using Pearson’s χ² test or Fisher’s exact test. Continuous variables were analyzed using the Mann–Whitney U-test. Univariable and multivariable analyses with logistic regression models were performed to identify factors associated with the development of exacerbations during azithromycin treatment. In multivariable analyses, variables with P < 0.20 on univariate analyses and clinically relevant variables were included. With 95% confidence intervals (CIs), P values of less than 0.05 were considered significant. All analyses were performed using IBM SPSS 25.0 (IBM Corp., Armonk, NY).

Results

Baseline Characteristics

Fifty-nine patients with CAO were categorized into two groups: bronchiectasis with CAO group (N = 43) vs COPD without bronchiectasis group (N = 16). The median age was 64 years, and patients in bronchiectasis with CAO group were younger (P = 0.023), more likely to be female (P = 0.017), and never smokers (P < 0.001) than those in COPD without bronchiectasis group (Table 1). While no patients had allergic bronchopulmonary mycosis, six patients (14.0%) and a patient (6.3%) had co-existing bronchial asthma in bronchiectasis with CAO group and COPD without bronchiectasis group, respectively (P = 0.661). Most patients reported dyspnea with mMRC grade ≥ 2 (83.1%) and CAT score ≥ 10 points (98.0%), which did not differ between the two groups. About 90% of patients had a history of moderate or severe exacerbations in the previous year, which were similar between the two groups.

Table 1 Baseline Characteristics at the Time of Azithromycin Treatment Initiation

Regarding pulmonary function, FVC of patients in bronchiectasis with CAO group was lower than those in COPD without bronchiectasis group, whereas there were no significant differences in FEV₁. Compared to patients in COPD without bronchiectasis group, those in bronchiectasis with CAO group had higher eosinophil counts (P = 0.019), which was persistent even after excluding 7 patients with asthma. Although it did not reach statistical significance, patients with bronchiectasis had a higher rate of positive sputum culture (10/40, 25.0%) than those without bronchiectasis (0/10, 0.0%). In addition, among patients with bronchiectasis, the median number of lobes involved in bronchiectasis was 4, and 26 patients (60.5%) had BSI ≥9 points (data not shown).

Outcomes of Long-Term Azithromycin Treatment

The median dose of azithromycin was 1750 mg per week, and the median duration of treatment was 15 (IQR 8.0–25.0) months. Most of the patients in COPD without bronchiectasis group (93.8%) were treated with ICS before initiating azithromycin, while only 58.1% of bronchiectasis with CAO patients did so (P = 0.010) (Table 2). During the 12-month follow-up, the development of ≥2 moderate or ≥1 severe exacerbation was significantly lower in bronchiectasis with CAO group than in COPD without bronchiectasis group (46.5% vs 87.5%, P = 0.005). However, there was no difference in the development of severe exacerbations between the two groups (Table 3). The proportion of patients who achieved MCID for CAT score was significantly higher in bronchiectasis with CAO group than in COPD without bronchiectasis group at 3 months (75.9% vs 42.9%), 6 months (74.2% vs 40.0%), and 12 months (68.2% vs 16.7%) of azithromycin treatment (Figure 1).

Table 2 Baseline Characteristics of Given Treatment

Table 3 Outcomes of Long-Term Azithromycin Treatment at 12 Months Follow-Up

Figure 1 The proportion of symptomatic responders* after long-term azithromycin treatment. *Defined as patients with ≥2 points decrement from the initial COPD Assessment Test (CAT) score. CAT was measured in 43, 47, and 35 patients at 3, 6, and 12 months after initiation of azithromycin treatment. Abbreviations: CAO, chronic airflow obstruction; COPD, chronic obstructive pulmonary disease.

Adverse events were reported in three patients: lip swelling, abdominal discomfort, and general weakness. In terms of hearing loss, audiometry was performed in 25 patients at baseline and 14 patients repeated audiometry following azithromycin treatment. No hearing impairment was observed in these patients, and there was no subjective hearing loss following azithromycin treatment in all patients.

Risk Factors Associated with Acute Exacerbation During Azithromycin Treatment

Factors associated with the development of ≥2 moderate or ≥1 severe exacerbation during the 12-month follow-up were evaluated using logistic regression analysis. In univariable analysis, male sex (unadjusted odds ratio [OR] = 3.14, 95% CI = 1.07–9.19), smoking history (unadjusted OR = 3.04, 95% CI = 1.03–8.97) were associated with a higher risk of exacerbation, while the bronchiectasis with CAO group had a lower risk of exacerbation compared with COPD without bronchiectasis group (unadjusted OR = 0.12, 95% CI = 0.03–0.61). In multivariable analysis, patients with bronchiectasis with CAO had a significantly lower risk for exacerbation after long-term azithromycin therapy, compared with COPD patients without bronchiectasis (adjusted OR = 0.15, 95% CI = 0.03–0.87) (Table 4).

Table 4 Risk Factors for Acute Exacerbationsa During 12 Months of Follow-Up

Discussion

In the present study, bronchiectasis patients with CAO had a significantly lower risk of exacerbation than COPD patients without bronchiectasis during azithromycin therapy with a median follow-up of 15 months. It is noteworthy that there was no difference in the previous exacerbation history or the degree of airflow limitation between the two groups. In addition, the proportion of CAT responders indicating symptomatic improvement was markedly higher in bronchiectasis with CAO group than in COPD without bronchiectasis group. Given that the diagnosis of COPD is based on CAO and the delineation between bronchiectasis with CAO and COPD-bronchiectasis overlap is not always clear in practice,²⁹ our study would provide important therapeutic implication of coexisting bronchiectasis among COPD patients with frequent exacerbations.

In addition to their antimicrobial action, macrolides are well known for their immunomodulatory effects by mitigating neutrophil-mediated lung damage and promoting mucociliary clearance, for which the effect occurs below the minimum inhibitory concentration for bacterial infection.^30,31 Several studies have proven the effect of long-term macrolide treatment for exacerbation prevention in COPD patients.^6,7,32–34 A recent meta-analysis including 2151 patients showed that long-term macrolide treatment reduced COPD exacerbation compared with the control (OR = 0.40; 95% CI = 0.24–0.65).³⁵ Long-term macrolide treatment also improved the quality of life measured with the St George’s Respiratory Questionnaire score, although it did not exceed the MCID. However, there are notable adverse effects associated with macrolide therapy, including gastrointestinal disturbance, hearing impairment, and risk of antibiotic resistance, which were more evident with azithromycin, among other macrolides.³⁵ Therefore, it is important to identify patients who are most likely to benefit from long-term azithromycin therapy. Landmark trials reported that daily azithromycin treatment for 12 months decreased the frequency of COPD exacerbation,⁶ and the post hoc analysis found that older (age > 65) patients, those with milder airflow limitation (FEV₁ > 50% pred), and ex-smokers were subgroups with greater efficacy of azithromycin therapy.⁸ However, the presence of bronchiectasis was not assessed, although it was not an exclusion criterion of the trial. In another study of azithromycin therapy, patients with bronchiectasis were excluded from the beginning.⁷ In the aspect of bronchiectasis, the reported efficacy of long-term azithromycin use in bronchiectasis is approximately halving the exacerbation rate (adjusted incidence rate ratio 0.49, 95% CI = 0.36–0.66) in a recent individual participant data meta-analysis, which was consistently observed in all subgroups.³⁶ However, no study has addressed the question regarding the different efficacy of long-term azithromycin between these two frequently overlapping diseases since most studies on azithromycin treatment in bronchiectasis did not exclude patients with coexisting COPD or spirometric CAO.^9–11 Correspondingly, 65% of bronchiectasis patients in azithromycin trials were using ICS,³⁶ similar to our study (approximately 60%). Thus, our study compared the efficacy of azithromycin treatment between bronchiectasis with CAO and COPD patients without bronchiectasis, and showed the lower exacerbation risk and better symptomatic improvement in bronchiectasis with CAO patients than in COPD patients without bronchiectasis during long-term azithromycin treatment, even using much less ICS. Our study suggests that azithromycin is a more effective anti-inflammatory treatment in bronchiectasis with CAO and this is also aligned with the fact that bronchiectasis patients taking macrolide monotherapy were less likely to be hospitalized for a respiratory infection or exacerbated compared with ICS.³⁷

The microbiological burden of coexisting bronchiectasis likely has a beneficial effect on azithromycin treatment. Several studies consistently demonstrated that coexisting bronchiectasis in COPD is associated with chronic bronchial infection (CBI) by potentially pathogenic microorganisms and greater bacterial load,^38–40 with more than 50% of sputum samples yielding a positive result.⁴⁰ Our study also showed that patients with bronchiectasis were more likely to have a positive sputum culture than those without bronchiectasis. This CBI perpetuates inflammation and contributes to COPD progression and frequent exacerbations.^41,42 Indeed, some previous trials in patients with COPD showed a nonsignificant reduction in airway bacterial load with three months of azithromycin treatment.^43,44 Taken together, the presence of bronchiectasis might indicate the CBI phenotype of COPD, which is associated with a better response to long-term azithromycin treatment for preventing exacerbation.

We acknowledge several limitations of this study. First, it was a retrospective study at a single referral hospital in Korea, including a small number of patients, limiting its generalizability. Notably, long-term azithromycin as an anti-inflammatory agent for bronchiectasis or COPD has not yet been covered by national health insurance in South Korea. Second, comprehensive microbiological data were not available due to the retrospective nature of this study. Although we speculate that CBI might have played a major role in the favorable outcomes of azithromycin treatment, we did not obtain data regarding the change in bacterial load or the emergence of antibiotic resistance following azithromycin treatment.⁴⁵ We excluded patients with NTM pulmonary disease, in which concerns about inducing macrolide resistance are critical to macrolide monotherapy. Further studies with a larger number of patients and prospectively collected microbiologic data are warranted to prove that coexisting bronchiectasis in patients with COPD is a predictive factor for the treatment response to long-term azithromycin. Finally, this study reported a markedly lower occurrence of adverse events than the previous trials,^6,7,9 of which the prevalence ranged from 36–42%, probably owing to its retrospective nature or partly because we only selected those who had already tolerated azithromycin for more than 12 weeks. It is noteworthy that subjective hearing loss was not reported in all patients, while audiometry was performed in 42% and showed no hearing impairment. A further prospective study with sufficiently long follow-up is required to clarify the optimal duration of treatment to balance the benefits and risks.

Conclusion

In conclusion, this study generated a hypothesis suggesting that bronchiectasis patients with CAO could benefit more from long-term azithromycin treatment in terms of exacerbation prevention and quality of life improvement than COPD patients without bronchiectasis. Further prospective studies are necessary to confirm our findings.