Our findings indicate that guidance with the measurement of procalcitonin levels reduces the exposure of patients to antibiotics after presentation to the emergency department for exacerbations of COPD. This initial difference in antibiotic exposure is not followed by increased antimicrobial usage after hospitalization for up to 6 months. Thereby, the clinical outcome, including exacerbation rate and time to the next exacerbation, was not compromised.
The absolute risk reduction of 31.5% in antibiotic exposure implies that for one in every four patients who were admitted to the hospital due to an ECOPD, one course of antibiotic therapy can be prevented (number-needed-to-treat, 3.2; 95% CI, 2.3 to 5.3).
Given the prevalence of COPD and the duration of illness, a reduction in antibiotic prescriptions for the treatment of exacerbations could have a tremendous impact on the overall economic burden of the disease under current budget constraints. In addition, the controlled prescription of antibiotics decreases selective pressure for the emergence of bacterial resistance.
A metaanalysis of placebo-controlled trials concluded that there was a small but significant benefit from the treatment of exacerbations of COPD with antibiotics in terms of overall recovery. A more recent extensive review of the literature suggested that antibiotic therapy significantly decreased mortality and lack of response to treatment in patients experiencing exacerbations of COPD. Moreover, it has been proposed that antibiotic therapy may reduce further antibiotic prescriptions following the presenting exacerbation and may increase the time until the next exacerbation in a selected population of patients. Several characteristics have been suggested to identify patients who are at a greater risk for severe exacerbation, including the presence or severity of underlying obstructive disease, comor-bid conditions, frequency of exacerbation, and severity of symptoms at presentation. Most of these proposed criteria have been analyzed in different retrospective study designs. However, none has been validated by a prospective randomized trial.
Similarly, a wide variety of surrogate markers of the inflammatory process have been measured in patients in the stable state and during and ECOPD. Most of them provide laboratory confirmation supporting the diagnosis of exacerbation. Unfortunately, their role in patient management is far from certain, as prospective studies under longterm follow-up are not available.
In this prospective, interventional study, we randomized unselected, consecutive COPD patients to receive antibiotics based on serum procalcitonin levels at hospital admission. The vast majority of admitted patients took part in the study, assuring the applicability of the proposed approach under “real-life” conditions. The acquisition of data from family physicians provided unbiased long-term follow-up information.
Although procalcitonin guidance reduced antibiotic prescription by 44%, its use did not result in an increase in the relapse of COPD, a decrease in the length time before the next exacerbation, or a more rapid decline in lung function. Patients who were assigned to the procalcitonin group, who received antibiotics, indeed had greater improvement in FEV1 compared to the patients who received antibiotics in the standard-therapy group. Thus, it is tempting to speculate that procalcitonin levels at hospital admission identifies patients who present with more severe or tissue-invasive bacterial infection and hence would most likely to benefit from antibiotic therapy.
Sputum microbiology is considered to be of limited value in investigating exacerbations of COPD. Conversely, the acquisition of new strains of microorganisms might indicate an impending bacterial exacerbation., In our study, there was no correlation between procalcitonin levels and sputum bacteriology. However, we could not analyze whether high procalcitonin levels at hospital admission were associated with the acquisition of new bacterial strains, as molecular genotyping has not been performed.
This study might have a potential limitation in regard to its generalizability. As previous experience with procalcitonin was available at the study site, physician adherence to the study protocol was facil-itated. Furthermore, only patients with an ECOPD requiring hospitalization were included in the study. Thereby, procalcitonin measurement results were available within 1 h after hospital admission, allowing an immediate treatment decision. Thus, the applicability of these findings to mild exacerbations treated in the outpatient setting still has to be confirmed in multiinstitutional, international studies. Furthermore, we cannot exclude that some of the patients who were randomized to the procalcitonin group, who received antibiotics because of elevated circulating procalcitonin levels, would have recovered without antibiotic therapy. However, taking into account the well-established association between elevated procalcitonin levels and clinically relevant bacterial infection (ie, sepsis), we considered the inclusion of a third study arm, which would withhold antibiotic treatment, to be in all cases unethical.
In conclusion, our results suggest that procalcitonin could be a suitable biomarker of exacerbations of COPD, which may be used to target management for each patient and episode more specifically, allowing a sustained reduction in antibiotic use for the treatment of COPD both at short-term and longterm follow-up.