CardioExchange Archive

CardioExchange is pleased to reprint selections from Dr. Richard Lehman’s weekly journal review blog at BMJ.com. Selected summaries are relevant to our audience, but we encourage members to engage with the entire blog.

JAMA  6 June 2012  Vol 307

Troponin and Mortality in Non-Cardiac Surgery (pg. 2295): If you want to get a paper in a major journal, go for something about prognosis. It couldn’t be easier. Take some bloods – or better still, use some that have already been taken and stored – and measure something that is bound to be associated with overall mortality: B-type natriuretic peptide, rise in creatinine, cystatin C, copeptin, or, as in this case, troponin T (TnT). Pick a clinical scenario: in this case non-cardiac surgery. Then shock horror: overall predictor of mortality predicts mortality in this population! A million people die every year within 30 days of surgery: measuring the peak fourth-generation troponin T (TnT) measurement in the first 3 days after noncardiac surgery will allow you to identify some of those at highest risk. Compared with what other clinical characteristics or biochemical markers? And what are you going to do with this information?

Ann Intern Med  5 June 2012  Vol 156

Predictors of HF Mortality (pg. 767): If in the year 2012 you sent a leading medical journal a case series describing the results of treatment for pneumonia using sulfanilamide, you might expect rejection – followed by an immediate visit from a disciplinary officer, perhaps accompanied by a psychiatrist. For therapy, we require that the most effective modern treatment be used, and that any new treatment should have a good chance of being shown to be more effective in a blinded randomized controlled trial. And taken alone, observational data are seldom enough. But as I’ve already hinted, it is so very different when the study is about prognosis. Here, it seems, the best is never the gold standard: any available combination of data will do. This Canadian study of mortality following acute heart failure crunches data relating to various emergency room measurements such as heart rate, creatinine, BP, oxygen saturation and troponin, and finds that you can derive a multivariate index which is somewhat predictive for mortality. Of course you can: but why ignore the best marker for mortality in heart failure, which is B-type natriuretic peptide? There are few clinical uses for BNP in real life, but the one that has been firmly established for 15 years is predicting death.