BMJ  2005;330 (26 February), doi:10.1136/bmj.330.7489.0-g

Editor's choice

Affairs of the thorax

The BMJ strives to help doctors in their clinical practice. An important difference between the BMJ and the Lancet, as one Lancet editor described it, is that the BMJ publishes articles focusing on the point of delivery in health care, where doctor meets patient and policy maker meets policy. Some, possibly too many, readers believe that our vision has drifted woefully from the point of delivery to an obsession with health policy and the "softer" social and political issues that are mere eyewash to dedicated clinicians (pp 474, 478). Our view is that a good journal incorporates all these elements and that diversity is the strength of the BMJ. Even so, we spend a great deal of time trying to find ways to publish more papers of clinical relevance,

This week we cut through the soft underbelly of social medicine to report findings that should interest our readers who spend their days marching around hospital wards and locked in patient consultations. Any doctor with a day's clinical experience will have faced the challenge of narrowing down the possible causes of chest pain. We've all been through it, straining to elicit textbook descriptions of angina or pleuritic chest pain, followed by a prod of the patient's chest, searching for the reassurance of a less life threatening diagnosis. In the heat of an examination the usefulness of a diagnostic test is less bothersome than the thought of awkward questions on the morning ward round and the vision of the lead physician rubbishing your diagnosis of pulmonary embolism with the patient's cry of aguish as he reproduces the chest pain by palpation. Well, forget all that. Gregoire Le Gal and fellow investigators studied the records of 965 people with suspected pulmonary embolism to discover that reproducing chest pain by prodding and pummelling a patient's chest is not associated with a lower prevalence of pulmonary embolism (p 452).

Doctors from New Zealand reveal another useful message for clinicians treating people with non-severe community acquired pneumonia. They investigate the effectiveness of lactams compared with antibiotics with specific activity against atypical pathogens and find that lactams should be the antibiotics of choice in all patients except the few with legonella related pneumonia (p 456). In an accompanying commentary, Mark Woodhead and Theo Verheij describe this meta-analysis as a valuable contribution to a topic that causes fierce discussion among doctors (p 460).

Continuing with affairs of the thorax, a multinational observational study addresses the dilemma of where to send patients with acute coronary syndrome. Should they be dispatched to the nearest cardiac catheterisation laboratory? Will the nearest hospital do even if it does not have interventional facilities? Frans Van de Werk and others discover that availability of a catheterisation laboratory does not confer survival benefit and may even increase the risk of major bleeding and stroke in hospital (p 441). That's as relevant to the point of delivery as you can get.