Swine flu and antibiotics--authors' response

Sir, We thank Michael Eisenhut for his helpful comments on our article. We agree that it is difficult to clinically distinguish viral from bacterial lower respiratory tract infection (LRTI), but this fact emphasizes the importance of careful severity (prognostic) assessment in LRTI. Macfarlane et al. previously showed that in patients with LRTI managed in the community, irrespective of whether the infection was viral or bacterial, outcome was unaffected by receipt of an antibiotic. We acknowledge the excellent work of Schuetz et al. on the role of procalcitonin in both severity assessment and guiding the need for antibiotic therapy in LRTI, and agree that it may have a useful role in hospitals that have experience in using it. However, procalcitonin is not widely available or used in the UK and so it is unlikely that it will contribute greatly to guiding antibiotic therapy during the current pandemic. It is also debatable as to whether increasing the complexity of care for LRTI patients at short notice is desirable during what is likely to be a very busy winter period. Procalcitonin is most likely to be helpful in patients with mild to moderate illness; overall mortality in the study by Schuetz et al. was relatively low at 5% (intensive care unit admission1⁄47.5%). During the coming winter, patients with severe LRTI are likely to form a high proportion of patients admitted to hospital. It is unlikely that clinicians in the UK will be willing to restrict antibiotic use in these patients, regardless of the procalcitonin level.