X2.2.3 Antibiotics for treatment of exacerbations

Exacerbations with clinical features of infection (increased volume and change in colour of sputum and/or fever) benefit from antibiotic therapy [evidence level II, strong recommendation]

Bacterial infection may have either a primary or secondary role in about 50% of exacerbations of COPD (Macfarlane 1993, Wilson 1998, Miravitlles 1999, Patel 2002). Haemophilus influenzae, Streptococcous pneumoniae and Moraxella catarrhalis are most commonly involved (Macfarlane 1993, Soler 1998, Murphy 1999). Mycoplasma pneumoniae and Chlamydia pneumoniae have also been reported (Macfarlane 1993, Mogulkoc 1999). As lung function deteriorates (FEV₁ < 35%), Pseudomonas aeruginosa and Staphylococcus aureus may be encountered (Macfarlane 1993, Soler 1998, Miravitlles 1999). Multi drug resistant Ps. aeruginosa is associated with 6 fold increased risk of death (Montero 2009) [evidence level III-2].

Nonetheless, sputum colour was shown to have limited value as a stand-alone test in diagnosing bacterial infection in a systematic review and meta-analysis of 13 studies by Spies et al (Spies 2023) [evidence level I).

A re-examination of data from the placebo arm of a Spanish antibiotic trial that recruited patients with mild to moderate COPD from primary care confirmed that sputum purulence increased the likelihood of treatment failure 6-fold.  A CRP elevated greater than 40 mg/L was also independently associated with a 13 fold increase in the risk of treatment failure (Miravitlles 2013) [evidence level III-2].

A study of 220 patients hospitalised with exacerbations of COPD with clinical features of infection, randomised to CRP-guided antibiotic therapy (antibiotics if CRP ≥ 50mg/L) or GOLD criteria based antibiotic treatment found a significant reduction in antibiotic use in the CRP guided group, with an absolute reduction in antibiotic use of 14.5% (Prins 2019) [evidence level II]. An open label RCT (n=653) of patients in the UK showed that in patients with COPD exacerbations treated in primary care, use of point-of-care CRP testing to guide prescribing of antibiotics lowered patient-reported antibiotic use (OR 0.31, 95% CI 0.20 to 0.47) (Butler 2019) [evidence level II]. The judicious use of CRP testing in primary or tertiary care may assist in determining the need for antibiotics for exacerbation management.

El Moussaoui et al (El Moussaoui 2008) conducted a systematic review of 21 randomised controlled trials of antibiotics in exacerbations of chronic bronchitis and COPD. There were similar rates of clinical or bacteriological cure with short courses (≤ 5 days) and longer courses of antibiotics [evidence level I]. A related systematic review (Falagas 2008) found that patients receiving short courses experienced fewer adverse effects than those receiving longer courses. It would be necessary to treat 26 (95% CI 15 to 134) patients with short course antibiotics to prevent one adverse effect. However, the antibiotics evaluated were late generation cephalosporins, macrolides and fluoroquinolones, which are not those recommended in Australia.

Procalcitonin is an acute phase reactant. Procalcitonin levels increase in bacterial infections but do not increase in viral infections or auto-immune inflammation (Gilbert 2011). Procalcitonin has been proposed as a measure to determine if patients with an exacerbation of COPD require oral antibiotics.  In most clinical trials, use of antibiotics was discouraged if procalcitonin was 0.1ng/ml or lower and encouraged if procalcitonin was above 0.25ng/ml.

A meta-analysis of eight randomised or quasi-randomised trials, evaluating 1,062 patients, compared procalcitonin-based protocols to initiate or discontinue antibiotics, versus standard care in COPD exacerbation (Mathioudakis 2017). Procalcitonin-based protocols decreased antibiotic prescription (relative risk (RR) 0.56, 95% CI 0.43–0.73) without affecting clinical outcomes such as rate of treatment failure, length of hospitalisation, exacerbation recurrence rate or mortality (low to moderate quality evidence). Since the publication of this meta-analysis, a further trial has also reported that procalcitonin-based protocols reduce antibiotic use without increasing complications (Wang 2016a).

A meta-analysis of RCTs and observational studies investigating the impact of a procalcitonin-based protocol on antibiotic prescription and clinical outcomes in patients with COPD exacerbations, found that the use of procalcitonin-based protocols significantly reduced the length of antibiotic treatment in COPD exacerbation (MD = -2.01 days, 95% CI -3.89 to -0.14 days, p=0.04, moderate quality, and MD = -1.64 days, 95% CI -2.91 to -0.36 days, p=0.01, very low quality for RCTs and observational study, respectively), while no apparent effects were found on length of hospital stay, treatment failure and all-cause mortality. The effect of procalcitonin on antibiotic duration was no longer significant (MD = -1.88 days, 95% CI -3.95 to 0.19 days, p=0.08, and MD = -1.72 days, 95% CI -4.28 to 0.83 days, p=0.19, respectively), when studies with high risk of bias were excluded. Procalcitonin has limited value in guiding antibiotic use in COPD exacerbation (Chen 2020) [evidence level I].

It is important to note that patients with pneumonia were excluded from these trials. Based on the evidence from these trials, it may be possible to withhold antibiotic therapy in patients presenting to the emergency department with an exacerbation of COPD, who are afebrile, have no pneumonia on chest imaging, and have a serum procalcitonin level of <0.1ng/ml. This test is not currently funded by Medicare in Australia and is only available in some centres.  Despite promising data from multiple clinical trials, cross-sectional and longitudinal analysis of over 200,000 COPD admissions from 505 US hospitals did not show a change in antibiotic prescribing rates or duration of use in hospitals that had begun using procalcitonin testing (Lindenauer 2017). The authors conclude that further implementation research is required.

Therapeutic guidelines: antibiotic (Therapeutic Guidelines Limited 2019) recommend the use of oral agents such as amoxycillin or doxycycline.

A retrospective cohort study from the Danish registry of COPD by Bagge et al (2021) examined outcomes following patients redeeming prescriptions for amoxycillin (AMX) or amoxycillin clavulanic acid (AMC) for presumed community exacerbations of COPD. They found pneumonia hospitalisation or death by all cause after 30 days was decreased with AMX compared to AMC (adjusted HR 0.6, 95% CI 0.5-0.7, p<.0001). This was also observed for all cause hospitalisation or death (aHR 0.8, 95% CI 0.8-0.9, p<0.0001). Although confounding by severity is not excluded, the findings of this study support the recommendation broad -spectrum antibiotics such as AMC should not be the drug of first choice for outpatient exacerbations of COPD (Bagge 2021) [evidence level III-2].

If pneumonia, Pseudomonas or staphylococci are suspected, appropriate antibiotics should be used.

Typically, a course of antibiotics should be five days. A systematic review and meta-analysis by Llor et al (2022) including only patients with spirometrically-proven COPD (n=eight trials) concluded that there were no significant differences in clinical cure rates or bacterial eradication rates of short courses of antibiotics (≤5 days) compared with longer courses (≥6 days). Nonetheless, the majority of studies included fluroquinolones as first line therapy, which is not common practice in Australia, raising questions about the face validity of this study [evidence level I] (Llor 2022). A historical population-based cohort study found that co-treatment of an exacerbation with oral corticosteroids and oral antibiotics significantly increased the time to subsequent exacerbations (median 312 versus 418 days, p<0.001 to next compared to oral corticosteroids alone) (Roede 2008)[evidence level III-2].

Two Australian retrospective case series of hospitalised COPD patients have found that antibiotic treatment was guideline concordant in less than 15% of cases (Brownridge 2017, Fanning 2014). This was due to over-use of intravenous antibiotics and prescription of dual antibiotics. Further efforts are needed to increase adherence to the use of oral antibiotics in patients hospitalised with exacerbations of COPD, where appropriate.

Radiologically proven pneumonia in patients with COPD, especially in those who have been frequently hospitalised, may not be restricted to the above organisms. Gram-negative organisms, Legionella and even anaerobic organisms may be responsible. Initial empiric antibiotic therapy should be tailored according to clinical and radio­graphic criteria.