This month’s Editor’s Research Choice is a study from the Community-Acquired Pneumonia – Study on the Initial Treatment with Antibiotics of Lower Respiratory Tract Infections (CAP-START) Study Group based in the Netherlands. The authors conclude that monotherapy with a beta-lactam antimicrobial alone is as good as various combinations that also include macrolides, ciprofloxacin, moxifloxacin and levofloxacin. This is important since the possibility of resistance and Clostridium difficile infection rises with increased antibiotic use.

Pneumonia is one of a group of diagnoses collectively labelled ‘lower respiratory tract infection’ or ‘LRTI’. This umbrella grouping is designed to separate LRTI, which is usually bacterial infection, from upper respiratory tract infection (URTI), which is usually viral, and therefore does not benefit from antimicrobial therapy. 

There are essentially three entities that comprise LRTI:

• Acute bronchitis, which is clinically similar to URTI except for the production of purulent sputum due to bacterial infection in the trachea and bronchi. Antimicrobial treatment is appropriate but may not be necessary unless there are co-morbidities
• Infective exacerbations of chronic obstructive pulmonary disease, which are easy to identify due to pre-existing COPD and benefits from antimicrobial therapy although non-infective exacerbations are also recognised and may be difficult to distinguish
• Pneumonia which is a clinical and radiological diagnosis.

Given these three quite distinct diagnoses the widespread use of ‘LRTI’ as a definitive diagnosis is perhaps surprising. The problem is one of precision, or the lack of it, in diagnosis. Many of the patients labelled ‘LRTI’ are elderly with multiple co-morbidities, 

The symptoms and signs of infection may be vague. The risk of failing to recognise and treat infection may be significant. The expectation of the patient and their carers is therapeutic intervention. Faced with this pressure to treat, the instinct is to administer an antimicrobial. However, the evidence to support a firm diagnosis may be thin. Chest x-rays may be difficult to interpret in older people and new changes may be hard to differentiate from chronic ones. Slight elevation of the white cell count or CRP does not necessarily indicate bacterial infection.

The clinician whose instinct is to administer antimicrobials is challenged to justify this decision. Pneumonia is a precise diagnosis. ‘LRTI’ seems somehow less rigorous but will still allow antimicrobial therapy. So the diagnosis is a fudge – ‘I’m going to treat this patient with an antibiotic even though I cannot demonstrate criteria for a diagnosis of pneumonia’. 

This is bad medicine and runs contrary to evidence-based practice. It is worth remembering that scoring systems for severity of pneumonia are only applicable in patients with pneumonia. Diagnosis must come first.