BMJ Journals research highlights—24 February 2014

Research questionsBMJ Journals research highlights is a regular round-up of research papers appearing in the BMJ Journals.


If clotters fibrose what do bleeders do?
Evolution has ensured that numerous genes associated with a profibrotic state have survived through the generations. These genes might lower the risk of peri-partum bleeding, but they increase the risk of serious thromboembolic disease in later life, well after the selfish gene has replicated. A high quality case-control study suggests that an additional consequence might be a higher risk of idiopathic pulmonary fibrosis (IPF). The association between clotting abnormalities and IPF was large and it is biologically plausible, making it unlikely that Berkson’s fallacy is at play. This association cannot establish chicken and egg and any potential mechanism is likely to be complex. One immediate thought is if clotters fibrose, what do bleeders do? Inflame? Research in the bleeding obvious? Go into politics? The usual Thorax prize of immense prestige but less than zero monetary or aesthetic value (in this case a signed photograph of the editors) for the most entertaining answer.

When a clot is a lot
How’s this for a classic conundrum? A patient presents with a large proximal pulmonary embolus, systolic blood pressure hovering around 100 mm Hg and CT evidence of right ventricular dysfunction. Often recent surgery or bleeding will be thrown into the mix just to make decision making really tough. My colleagues and I debate this issue endlessly and split roughly 50:50 into those who would and wouldn’t give thrombolytic therapy. What would you do? Jiménez et al help us by providing guidance on the prognostic significance of a CT showing evidence of right ventricular dysfunction. In short, it identifies patients who are more likely to have echo evidence of right ventricular dysfunction and a raised brain naturetic peptide level but not those with a higher chance of death or a complicated recovery. The question of when and in whom to use thrombolytic therapy in acute submassive pulmonary embolus remains a vexed one

Coming in from the cold
Lung transplantation represents the last hope for many patients with end-stage lung disease, but given the scarcity of organs and the worse results of re-do transplantation, the transplanted lung must be protected from damage as carefully as the reputations of our egregious politicians. Respiratory viruses are already causally implicated in the lung attacks which cause so much long term damage in asthma, COPD, cystic fibrosis (CF), and other diseases. Bridevaux et al report a long-term prospective study of the prevalence of common respiratory viruses in lung transplant recipients. They found that viruses were common, almost invariably associated with symptoms, but they could not show any association with rejection.

Evidence Based Medicine

Pollock and Price make a strong case that changes in the NHS resulting from the Health and Social Care Act of 2012 will lead to worse health resulting from a loss of critical data sources. Geographical and population based data are critical to determining how a health system should distribute access to care and resources. They are also key to health and health services research. The changes in the act make it such that data be collected by clinical commissioning groups, that will collect data only on those for whom they are responsible, and private healthcare delivery entities who may provide little or no data, leading to absence of population based data much like occurs in the fragmented delivery system in the US.

Newcombe and Bender provide a simple method, via a freely available spreadsheet, to calculate confidence intervals for a risk difference obtained from independent estimates of the baseline risk and the relative risk. The method is an advance over prior methods.  The importance of precision around the risk difference is that risk differences are key to understanding the effectiveness of treatments.