Management of infection in exacerbations of chronic obstructive pulmonary disease

INTRODUCTION — Most exacerbations of chronic obstructive pulmonary disease (COPD) are caused by respiratory tract infections. Empiric antibiotic therapy is indicated for patients who are most likely to have a bacterial infection causing the exacerbation and for those who are most ill.

The role of antibiotic therapy in exacerbations of COPD will be reviewed here. The evaluation for infection in exacerbations of COPD and other aspects of management (eg, bronchodilators, glucocorticoids, oxygen, and mechanical ventilation) are discussed separately. (See "Evaluation for infection in exacerbations of chronic obstructive pulmonary disease" and "COPD exacerbations: Management".)

DEFINITIONS — The Global Initiative for Chronic Obstructive Lung Disease guidelines define an exacerbation of COPD as an event characterized by dyspnea and/or cough and sputum that worsens over ≤14 days, which may be accompanied by tachypnea and/or tachycardia, and is often associated with increased local and systemic inflammation caused by airway infection, pollution, or other insult to the airways [1,2].

The three cardinal symptoms that characterize an exacerbation of COPD are [1,3]:

●Increased dyspnea

●Increased sputum volume and/or viscosity

●Increased sputum purulence

Other findings that may accompany the cardinal symptoms of an exacerbation of COPD (eg, tachypnea, chest discomfort, fatigue, sleep disturbance, decline in pulmonary function) are discussed separately. (See "COPD exacerbations: Management".)

INDICATIONS FOR ANTIBACTERIAL THERAPY

Our approach — Empiric antibiotic therapy is indicated for patients who are most likely to have a bacterial infection causing the exacerbation and for those who are most ill [1,4]. In general, we determine the need for antibiotics based on the number of cardinal symptoms present, and the need for hospitalization and/or ventilatory support [1,5-7].

●We suggest empiric antibiotic treatment in patients with a COPD exacerbation and ≥2 of 3 cardinal symptoms: increased dyspnea, increased sputum volume/viscosity, or increased sputum purulence or a COPD exacerbation requiring hospitalization and/or ventilatory support (either invasive or noninvasive).

●We do not initiate antibiotic therapy in patients with a COPD exacerbation and only 1 of 3 cardinal symptoms who do not require hospitalization or ventilatory support. New onset of increased wheezing may serve as an additional negative predictor for bacterial infection; therefore, if it is a prominent finding, it steers us away from antibiotic use [8].

Antibiotics are also indicated for patients with concurrent pneumonia (ie, those with fever, signs of consolidation on chest examination and/or chest imaging). Selection of an antibiotic regimen for patients with pneumonia differs from COPD exacerbations and is discussed separately. (See "Overview of community-acquired pneumonia in adults" and "Treatment of hospital-acquired and ventilator-associated pneumonia in adults".)

Rationale — Our approach is based upon randomized trials and large cohort studies demonstrating that prompt, appropriate antibiotic use improves clinical outcomes for selected patients with COPD exacerbations [3,9-16]. The benefit is greatest in severely ill patients and those with a greater number of symptoms but appears to diminish as the severity of illness declines. In a meta-analysis of five randomized trials evaluating 803 patients hospitalized with an acute COPD exacerbation, antibiotic use was associated with reduced treatment failure when compared with placebo (risk ratio [RR] 0.76, 95% CI 0.58-1.0) [9]. The benefits of antibiotic treatment were most prominent in a single trial evaluating 93 intensive care unit (ICU) patients, which showed a reduction in all-cause mortality (4 versus 22 percent with placebo) in addition to reductions in treatment failure, duration of mechanical ventilation, length of ICU stay, and need for additional courses of antibiotics [10]. In a cohort study evaluating >84,000 hospitalized patients with COPD exacerbations, the risk of treatment failure was lower when antibiotics were given in the first two hospital days compared with later treatment or no treatment (odds ratio [OR] 0.87, 95% CI 0.82-0.92) [12]. Multivariate analysis of this cohort demonstrated that antibiotic treatment was associated with decreased risk of in-hospital mortality (OR 0.60, 95% CI 0.50-0.73) and a substantial reduction in the risk of 30-day readmission for COPD (OR 0.87, 95% CI 0.79-0.96) [14].

Although meta-analyses of randomized trials and large cohort studies have also found reduced treatment failure rates with antibiotic use in outpatients [9,11], the benefit appears to be greatest in those with a higher number of cardinal symptoms. This finding is best supported by the Anthonisen trial, one of the largest and more rigorous randomized trials evaluating the efficacy for antibiotics for COPD exacerbations to date [3]. The trial evaluated 173 patients and a total of 362 COPD exacerbations over a three and a half year period. Antibiotic therapy was associated with increased clinical improvement (defined as resolution of symptoms without additional intervention) compared with placebo (68 versus 55 percent). The greatest effect was observed in patients who presented with increased dyspnea, sputum production, and sputum purulence when compared with placebo (63 versus 43 percent); benefit was least evident in patients with only one of these three symptoms (75 versus 70 percent). This observation serves as the foundation for the antibiotic treatment indications outlined above.

Based upon these studies, most clinical practice guidelines recommend antibiotic treatment of moderate to severe exacerbations for outpatients and patients who require hospitalization but not routinely for those with mild exacerbations in the outpatient setting [1]. Our approach is similar to but varies slightly from the strategy outlined by the Global Initiative for Chronic Obstructive Lung Disease (GOLD), which recommends antibiotic therapy for patients who have the following features: a severe exacerbation requiring mechanical ventilation (noninvasive or invasive), an exacerbation with all three cardinal symptoms, or an exacerbation with two of these three symptoms if sputum purulence is one of the symptoms [1,3,17]. While some studies suggest that sputum purulence is associated with an increased likelihood of infection [18,19], this finding is not consistent across studies [20,21], and we do not consider this finding alone to be clearly predictive of need for antibiotic treatment [22,23].

CRP and procalcitonin — Multiple studies have addressed use of serum biomarkers, such as C-reactive protein (CRP) and procalcitonin, to help determine the need for antibiotic treatment in patients with COPD exacerbations [24-29]. However, study results do not clearly and consistently demonstrate that use of either assay adds value to clinical judgment alone; we generally do not use them to guide treatment decisions. (See "Evaluation for infection in exacerbations of chronic obstructive pulmonary disease", section on 'Procalcitonin and C-reactive protein' and "Procalcitonin use in lower respiratory tract infections", section on 'Acute exacerbations of chronic obstructive pulmonary disease'.)

EMPIRIC ANTIBACTERIAL TREATMENT

Risk stratification — We use a "risk stratification" approach when selecting initial empiric antibiotic therapy for the treatment of acute exacerbations of COPD [30,31]. We categorize patients based on treatment setting (eg, outpatient versus inpatient), risk for poor clinical outcomes, and risk for infection with Pseudomonas (algorithm 1A-B).

●Risk for poor outcomes – Patients with greater underlying COPD severity are at higher risk for poor outcomes if initial antibiotic therapy is inadequate. We thus use a broader empiric regimen for such patients. Risk factors for poor outcomes include [32-34]:

•Comorbid conditions (especially heart failure or ischemic heart disease)

•Severe underlying COPD (forced expiratory volume in one second [FEV₁] <50 percent)

•Frequent exacerbations of COPD (ie, ≥2 exacerbations per year)

•Hospitalization for an exacerbation within the past three months

•Receipt of continuous supplemental oxygen

Older age (eg, age ≥65 years) is also associated with poorer outcomes and/or risk of infection with drug-resistant pathogens. While not a strict indication for broadening antibiotic therapy, we consider older age as additive to the risk factors listed above.

●Risk for pseudomonal infection – Patients with greater underlying COPD severity are also at risk for infection with Pseudomonas. Specific factors associated with an increased risk of Pseudomonas infection include [35-38]:

•Chronic colonization or previous isolation of Pseudomonas aeruginosa from sputum

•Very severe COPD (FEV₁ <30 percent predicted)

•Bronchiectasis on chest imaging (eg, chest radiograph, computed tomography)

•Broad-spectrum antibiotic use within the past three months

•Chronic systemic glucocorticoid use

This approach to risk stratification promotes judicious antibiotic use by reserving broader spectrum empiric regimens for patients with greater underlying COPD severity [32,39,40]. Although it has not been validated in clinical trials, similar approaches are used for the treatment of other infectious diseases (eg, community-acquired pneumonia, acute rhinosinusitis) [41,42].

Antibiotic selection — Empiric antibiotic regimens are designed to target the most likely infecting pathogens (table 1). Specific antibiotic selection and need for sputum Gram stain and culture vary based on risk for poor clinical outcomes, risk for infection with Pseudomonas, and treatment setting (algorithm 1A-B) [1,34,37]. Determining need for hospitalization in patients with COPD exacerbations is discussed elsewhere. (See "COPD exacerbations: Management", section on 'Triage to home or hospital'.) (Related Pathway(s): Chronic obstructive pulmonary disease: Identifying patients with an acute exacerbation who warrant hospitalization.)

Outpatients — Antibiotic therapy is indicated for outpatients with COPD exacerbations and an increase in ≥2 of 3 cardinal symptoms: dyspnea, sputum volume/viscosity, or sputum purulence (algorithm 1A) [34,37]. Patients with only one cardinal symptom generally do not require antibiotic therapy. (See 'Indications for antibacterial therapy' above.) (Related Pathway(s): Chronic obstructive pulmonary disease: Empiric antimicrobial therapy for outpatients with acute exacerbations.)

●For outpatients who do not have risk factors for poor outcomes or Pseudomonas infection, we target Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis and select among the following options: a macrolide (ie, azithromycin, clarithromycin) or a second- or third-generation cephalosporin (eg, cefuroxime, cefpodoxime, cefdinir). Trimethoprim-sulfamethoxazole is a reasonable alternative to these agents but not first line because trial data suggest it may be less effective [43]. We no longer use doxycycline based on a randomized trial showing minimal benefit [44].

●For outpatients who have risk factors for poor outcomes (but no increased risk for Pseudomonas infection) (table 2), we broaden the initial regimen to include treatment for macrolide-resistant S. pneumoniae and to enhance eradication of H. influenzae. For these patients, we select either amoxicillin-clavulanate or a respiratory fluoroquinolone (ie, levofloxacin or moxifloxacin).

●For outpatients who have risk factors for poor outcomes and a risk for Pseudomonas infection (table 3), we generally treat with ciprofloxacin. Because fluoroquinolone resistance is prevalent among P. aeruginosa strains, we also obtain a sputum Gram stain and culture with susceptibility testing for these patients to help guide subsequent management decisions (eg, change in therapy based on susceptibility testing for those who do not respond to empiric treatment). (See 'Follow-up' below.)

Some experts add amoxicillin or another agent with better activity against S. pneumoniae to ciprofloxacin for broader empiric treatment.

Levofloxacin is a reasonable alternative for patients who have risk factors for Pseudomonas infection but no prior history of positive cultures for Pseudomonas. Although levofloxacin is less potent than ciprofloxacin for the treatment of Pseudomonas, it has comparatively greater activity against other common pathogens (S. pneumoniae and M. catarrhalis). Moxifloxacin is not recommended for patients with risk factors for Pseudomonas, as it has little activity against this pathogen.

In most cases, several treatment options are available. We select among antibiotic options based on the patient's previous response to that agent, patient allergies and intolerances, the drug's adverse event profile, drug interactions, susceptibility pattern of organisms isolated in recent sputum cultures (if available), and local antimicrobial resistance patterns. In some instances, local antibiotic resistance rates can be determined by obtaining an antibiogram from a local hospital. However, this information is not always available or readily accessible.

In general, we also avoid using the same antibiotic class more than once within a three-month period. Antibiotic exposure in the previous three months is one of the best predictors of pathogens resistant to that drug class in an individual patient. For example, if a patient has responded well and tolerated a certain antibiotic and that antibiotic was not used within the past three months, then that agent (or one in the same class) is a good choice. However, if that agent was used within the past three months, then an agent of a different class should be selected.

The antibiotic selection outlined above is based on the susceptibility profiles and evolving resistance patterns of the most common pathogens. For example, amoxicillin, which was favored in the past, is no longer a recommended agent because it is inactivated by many nontypeable H. influenzae and most strains of M. catarrhalis. Doxycycline resistance among S. pneumoniae is prevalent in some regions, and it does not appear to improve clinical outcomes when compared with systemic glucocorticoids alone [44,45]; thus, it is also no longer first line. This approach is further supported by a meta-analysis of 12 randomized trials evaluating >2100 patients, which showed that amoxicillin-clavulanic acid, macrolides, second- or third-generation cephalosporins, and fluoroquinolones were more effective than amoxicillin, ampicillin, pivampicillin, trimethoprim-sulfamethoxazole, and doxycycline for the treatment of COPD exacerbations (odds ratio [OR] 0.51, 95% CI 0.34-0.75) [46]. Treatment success was defined as resolution or improvement of symptoms.

Several studies have compared the efficacy of macrolides, fluoroquinolones, and amoxicillin-clavulanate with one another [43,47,48]. One meta-analysis of 19 trials, evaluating >7400 patients with COPD exacerbations, compared the effectiveness of macrolides (ie, azithromycin, clarithromycin), fluoroquinolones (ie, levofloxacin, moxifloxacin, ciprofloxacin), and amoxicillin-clavulanate [43]. Antibiotic selection did not affect short-term treatment success, defined as resolution or improvement of symptoms. However, among patients who had a pathogen isolated from sputum at presentation, treatment success was lower for macrolides when compared with fluoroquinolones (OR 0.47, 95% CI 0.31-0.69). Recurrence, in the 26 weeks following therapy, was also less frequent in patients who were treated with fluoroquinolones compared with macrolides. The reduction in treatment success observed with macrolides may be related to their limited ability to eradicate H. influenzae. Amoxicillin-clavulanate had similar efficacy when compared with other agents but was associated with more adverse effects (mostly diarrhea) than the other drugs. In a subsequent randomized trial evaluating >500 patients with exacerbations of COPD, moxifloxacin and amoxicillin-clavulanate appeared to have similar overall efficacy [47]. However, clinical failure rates were lower with moxifloxacin when the analysis was limited to patients with a pathogen isolated from sputum (19 versus 25 percent). Adverse effects associated with fluoroquinolone use are discussed separately. (See "Fluoroquinolones" and "Fluoroquinolones", section on 'Adverse effects'.)

Hospitalized patients — For patients hospitalized for treatment of a COPD exacerbation (who lack clinical and radiographic suspicion for pneumonia), we primarily base empiric antibiotic selection on the risk for Pseudomonas (algorithm 1B). We select among appropriate options based on the patient's previous response to that agent, patient allergies and intolerances, the drug's adverse event profile, drug interactions, local antimicrobial resistance patterns, and the susceptibility pattern of organisms isolated in recent sputum cultures (if available).

●For most inpatients without risk factors for Pseudomonas infection, we select either a respiratory fluoroquinolone (ie, levofloxacin 500 mg orally or intravenously [IV] once daily or moxifloxacin 400 mg orally or IV once daily) or a third-generation cephalosporin (eg, ceftriaxone or cefotaxime).

●For most inpatients with risk factors for Pseudomonas infection, we select one of the following: cefepime, ceftazidime, or piperacillin-tazobactam (4.5 g IV every six hours). For those who cannot tolerate these agents, alternatives include ciprofloxacin, aztreonam, certain carbapenems (eg, meropenem, imipenem), and aminoglycosides. Each comes with relative disadvantages when compared with the antipseudomonal beta-lactams (table 4). We generally select among them based on local epidemiology, prior susceptibility testing results, drug interactions, and patient comorbidities or intolerances. Two agents are often needed for empiric treatment. (See "Principles of antimicrobial therapy of Pseudomonas aeruginosa infections".)

For all hospitalized patients who are able to produce a good-quality sputum sample, we obtain a Gram stain and culture to help guide management. For most others, we do not obtain microbiologic testing on sputum because it has limited diagnostic accuracy and results are unlikely to change management. (See "Evaluation for infection in exacerbations of chronic obstructive pulmonary disease", section on 'When to obtain sputum studies'.)

Patients with clinical concern for concurrent pneumonia should be treated with empiric antibiotic regimens based on the suspected pathogens, severity of illness, and type of pneumonia. (See "Treatment of community-acquired pneumonia in adults who require hospitalization" and "Treatment of hospital-acquired and ventilator-associated pneumonia in adults".)

Additional detail on the antibiotic selection for patients with known or suspected pseudomonal infections are provided separately. (See "Principles of antimicrobial therapy of Pseudomonas aeruginosa infections".)

Duration — The duration of therapy for patients who are clinically improving is generally five days for outpatients and five to seven days for hospitalized patients. However, azithromycin can be given for as few as three days when administered at a dose of 500 mg orally daily because of its long half-life. Patients who are initially started on parenteral antibiotics should be switched to an oral regimen when able to take medications orally.

A meta-analysis that compared five days with seven or more days of antimicrobial therapy (fluoroquinolones, cefixime, or clarithromycin) for exacerbations of COPD found no difference in outcome between the two groups, although there were fewer adverse events among patients who received a five-day course [49].

Some experts use procalcitonin, a biomarker that rises in response to bacterial infections, to guide antibiotic duration. However, clinical utility and safety of using procalcitonin to guide therapy in patients with COPD exacerbations is not firmly established and its use is controversial. (See "Procalcitonin use in lower respiratory tract infections", section on 'Acute exacerbations of chronic obstructive pulmonary disease'.)

RESPIRATORY VIRUS TREATMENT

Influenza — Antiviral therapy is usually indicated for patients whose acute exacerbations of COPD have been triggered by influenza virus. However, the benefits of antiviral therapy diminish with time; thus, for patients presenting ≥72 hours after illness onset, we take the patient's clinical trajectory (eg, worsening/improving) into account when deciding to prescribe. For patients with influenza who also meet criteria for antibacterial treatment (ie, ≥2 cardinal symptoms, need for hospitalization and/or ventilatory support), we treat with both antiviral and antibacterial therapy.

Inhaled zanamivir is contraindicated in this patient population due to the risk of airway reactivity. Other agents, such as oral oseltamivir, oral baloxavir, or, in certain situations, intravenous peramivir or zanamivir (not available in the United States) can be used. (See "Seasonal influenza in nonpregnant adults: Treatment".)

Coronavirus disease 2019 — COPD increases the morbidity and mortality associated with coronavirus disease 2019 (COVID-19; caused by severe acute respiratory syndrome coronavirus 2 [SARS-CoV-2]) and we test all COPD patients with symptoms of an exacerbation for COVID-19.

Standard treatment of the exacerbation with bronchodilators, systemic corticosteroids, and antibiotics should not be altered in patients with both COPD and COVID-19 [50]. However, these patients need to be observed closely for deterioration. Given their age and comorbidity, patients with COPD and COVID-19 are likely to be candidates for COVID-19 specific therapy. (See "COVID-19: Management of adults with acute illness in the outpatient setting" and "COVID-19: Management in hospitalized adults".)

FOLLOW-UP — Most patients should demonstrate some improvement in 48 to 72 hours of starting antibiotic therapy. For those who fail to improve, we generally obtain a sputum culture (if not already obtained) to help guide any subsequent changes in antibiotic treatment, reevaluate our approach to other aspects of care (eg, bronchodilators, need for mechanical ventilation), consider potential contributing comorbidities, and broaden our differential diagnosis to include other cardiopulmonary disorders (eg, pneumonia, heart failure, lung cancer). (See "COPD exacerbations: Management", section on 'Adjusting therapy for poor response' and "Approach to the patient with dyspnea".)

PREVENTION

Vaccination — Patients with COPD should be vaccinated against seasonal influenza virus (annually), SARS CoV-2, and pneumococcus (table 5).  

The evidence supporting vaccination of patients with COPD includes the following:

●The utility of influenza vaccination is well established for reducing both the rate and severity of symptoms due to influenza, including respiratory symptoms. A meta-analysis of 11 trials, including 6 specifically performed in patients with COPD, found a significant reduction in the number of exacerbations per patient compared with placebo [51]. (See "Seasonal influenza vaccination in adults".)

●For pneumococcal vaccination, a meta-analysis of 12 randomized trials evaluated the efficacy of PPSV23 on 2171 patients with COPD [52]. Analysis of five trials showed a reduction in the rates of community-acquired pneumonia with vaccination (odds ratio [OR] 0.61, 95% CI 0.42-0.89), and four trials showed a reduction in rates of COPD exacerbation (OR 0.60, 95% CI 0.39-0.93). There were no significant differences in all-cause or cardiorespiratory mortality, although event rates were low. (See "Pneumococcal vaccination in adults".)

Patients with COPD should also receive other vaccines according to the schedule summarized in the following figure (figure 1); in particular, vaccine uptake for pertussis is low and incidence is rising. (See "Pertussis infection in adolescents and adults: Treatment and prevention", section on 'Vaccination'.)

All patients who lack contraindications should be vaccinated against SARS-CoV-2. (See "COVID-19: Vaccines".)

Prophylactic macrolides — We do not routinely use prophylactic macrolides or other antibiotics for the long-term care of patients with COPD. For most patients, the benefits of long-term antibiotic use do not outweigh the risks. However, for selected patients with severe COPD with frequent exacerbations (≥2 per year) despite optimal medical management (long-acting bronchodilators, inhaled glucocorticoids, pulmonary rehabilitation, smoking cessation), macrolide prophylaxis may be advantageous. The risks associated with the development of antibiotic resistance should be taken into account when deciding to use long-term prophylaxis. When considering prescription, we carefully weigh the potential benefit against the risk of long-term macrolide use (eg, QT prolongation, other cardiovascular events, Clostridioides difficile infection). (See "Azithromycin and clarithromycin", section on 'Adverse reactions'.)

When prescribing a macrolide for long-term prophylaxis, we typically use azithromycin, which can be given as 250 mg daily [53] or at a lower dose of 250 to 500 mg three times per week [54-57]. We often use 250 mg three times per week to reduce adverse effects, although this dose is less well studied. Erythromycin (500 mg two times per day) is an alternative [1]. While the optimal duration of therapy is not known, a 12-month course is commonly used [1]. If there is suspicion for Mycobacterium avium complex (MAC) infection based on symptoms or radiological signs (bronchiectasis, tree in bud, chronic opacities on chest radiograph or computed tomography), sputum or bronchoscopic cultures for MAC should be obtained prior to initiation of macrolides; their use is not recommended if the cultures are positive. (See "Overview of nontuberculous mycobacterial infections", section on 'Clinical manifestations'.)

The benefit of macrolides is attributed to their immunomodulatory effects in addition to their potential to prevent infection [58,59]. In a systematic review of 14 randomized trials evaluating 3932 patients with moderate-to-severe COPD, the proportion of patients experiencing ≥1 exacerbation was reduced when comparing prophylactic antibiotic use (primarily macrolides) with placebo (OR 0.57, 95% CI 0.42-0.78) [58]. A marginal improvement in quality-of-life measures was detected; trends toward improvement in the number of hospital admissions, change in forced expiratory volume in one second (FEV₁), serious adverse events, and all-cause mortality were observed but were not statistically significant.

The benefits and risks of long-term macrolide use are well illustrated in one of the randomized trials included in the meta-analysis [53,58]. In this trial, 1142 patients with COPD were randomly assigned to receive azithromycin 250 mg orally daily or placebo for one year in addition to their usual COPD regimen [53]. The following findings were observed:

●The median time to first COPD exacerbation was significantly longer among the patients who received azithromycin compared with those who received placebo (266 versus 174 days).

●Patients who received azithromycin had a significant (but modest) reduction in the frequency of COPD exacerbations compared with those who received placebo (1.48 versus 1.83 exacerbations per patient-year; hazard ratio 0.73, 95% CI 0.63-0.84).

●Patients who received azithromycin had a significantly higher rate of nasopharyngeal colonization with macrolide-resistant bacteria (Staphylococcus aureus, S. pneumoniae, Haemophilus spp, Moraxella spp) than those who received placebo (81 versus 41 percent). Lower airway and enteric microbiology were not monitored, so the emergence of macrolide-resistant strains in these relevant sites was not assessed.

●Hearing decrements (assessed by audiometry) were more common in the azithromycin group than the placebo group (25 versus 20 percent). However, hearing loss associated with azithromycin usually results from long-term use and is reversible.

Patients with resting tachycardia, hearing impairment, and those with or at risk for a prolonged QT interval were excluded from participating. Thus, the risks of long-term use may be higher when used in the general COPD population.

Other antibiotics, particularly moxifloxacin, also have demonstrated some efficacy for preventing COPD exacerbations [17]. However, we typically reserve their use for the treatment of infections in order to avoid fluoroquinolone-related adverse effects and the selection of fluoroquinolone-resistant bacteria [60]. In one trial comparing pulsed moxifloxacin (400 mg orally daily for five days every eight weeks for six cycles [total duration of 48 weeks]) in 1157 patients with COPD at high risk for recurrent exacerbations, the risk of COPD exacerbation was lower with moxifloxacin, both in the per-protocol analysis (OR 0.75, 95% CI 0.565-0.994) and in the intent-to-treat analysis (OR 0.81, 95% CI 0.645-1.008) [17]. A larger benefit was observed in those with purulent or mucopurulent sputum production (OR 0.55, 95% CI 0.36-0.84) in post-hoc analysis. Gastrointestinal adverse effects were more frequent with moxifloxacin; however, C. difficile infections were not observed. Sustained emergence of moxifloxacin-resistant strains was not observed in sputum or in enteric flora.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic obstructive pulmonary disease".)

SUMMARY AND RECOMMENDATIONS

●Background – Most exacerbations of chronic obstructive pulmonary disease (COPD) are caused by respiratory tract infections. Empiric antibiotic therapy is indicated for patients who are most likely to have a bacterial infection causing the exacerbation and for those who are most ill. (See 'Introduction' above.)

●Clinical characteristics – An exacerbation of COPD is an event characterized by dyspnea and/or cough and sputum that worsens over ≤14 days and is often associated with increased airway or systemic inflammation. Most exacerbations of COPD are due to respiratory infection. Cardinal symptoms of a COPD exacerbation include (see 'Introduction' above and 'Definitions' above):

•Increase in dyspnea

•Increase in sputum volume and/or viscosity

•Increase in sputum purulence

●When to treat with antibiotics – We typically determine the need for antibiotics based on the number of cardinal symptoms present and the need for hospitalization and/or ventilatory support. The benefit of empiric antibiotic treatment is greatest in severely ill patients and those with a greater number of symptoms but appears to diminish as the severity of illness declines. (See 'Indications for antibacterial therapy' above.)

•We suggest empiric antibiotic treatment in patients with a COPD exacerbation and ≥2 of 3 cardinal symptoms: increased dyspnea, increased sputum volume/viscosity, or increased sputum purulence or a COPD exacerbation requiring hospitalization and/or ventilatory support (either invasive or noninvasive) (Grade 2C).

•We do not initiate antibiotic therapy in patients with a COPD exacerbation and only one of three cardinal symptoms who do not require hospitalization or ventilatory support. New onset of increased wheezing may serve as an additional negative predictor for bacterial infection; therefore, if it is a prominent finding, it steers us away from antibiotic use.

●Risk stratification to guide antibiotic selection – We use a "risk stratification" approach when selecting initial empiric antibiotic therapy for the treatment of acute exacerbations of COPD. We categorize patients based on treatment setting (eg, outpatient versus inpatient), risk for poor clinical outcomes (table 2), and risk for infection with Pseudomonas (table 3). (See 'Risk stratification' above.)

●Empiric antibiotic regimens – Empiric antibiotic regimens are designed to target the most likely infecting pathogens (table 1); specific selection varies based on the patient's risk status (algorithm 1A-B). When selecting an antibiotic, we take into account the patient's prior antibiotic exposure, prior clinical response to specific antibiotics, allergies and intolerances, drug interactions, the drug's adverse event profile, and the susceptibility pattern of organisms isolated in recent sputum cultures (if available). (See 'Antibiotic selection' above.)

●Duration of antibiotics – The duration of therapy is generally five days for outpatients and five to seven days for most hospitalized patients. However, azithromycin can be given for as few as three days when administered at a dose of 500 mg orally daily because of its long half-life. (See 'Duration' above.)

●Antiviral treatment – Patients with COPD are at increased risk for complications of influenza, so antiviral therapy (eg, oral oseltamivir or an intravenous agent) may be appropriate for exacerbations triggered by influenza virus, depending on timing and susceptibility patterns. Inhaled zanamivir is contraindicated in this patient population due to the risk of airway reactivity. Similarly, patients with COPD and COVID-19 are likely to be candidates for COVID-19-specific therapy. (See 'Respiratory virus treatment' above.)

●Vaccination – To prevent future exacerbations, patients with COPD should be vaccinated against influenza and pneumococcus, according to the following schedule (figure 1). All patients who lack contraindications should also be vaccinated against SARS-CoV-2. (See 'Vaccination' above.)

●Macrolide prophylaxis for selected patient with severe COPD – We do not routinely use prophylactic macrolides or other antibiotics for the long-term care of patients with COPD. For most patients, the benefits of long-term antibiotic use do not outweigh the risks. However, for selected patients with severe COPD and frequent exacerbations (≥2 per year) despite optimal medical management (bronchodilators, inhaled glucocorticoids, pulmonary rehabilitation, smoking cessation), macrolide prophylaxis may be advantageous. Azithromycin can be given as 250 mg daily or at a lower dose of 250 to 500 mg three times per week. We often use 250 mg three times per week to reduce adverse effects. (See 'Prophylactic macrolides' above.)

ACKNOWLEDGMENT — UpToDate gratefully acknowledges John G Bartlett, MD (deceased), who contributed on earlier versions of this topic and was a founding Editor-in-Chief for UpToDate in Infectious Diseases.