Activated protein C: do more survive?

Excerpt

Sir: There are a number of important general principles raised by Dr. MacKenzie's [1] contribution. The first concerns unbiased and even discussion of the subject matter. Dr. MacKenzie's concluding paragraph is an excessively biased and unreasonable claim supported by selective data presentation. Many of his criticisms of the PROWESS trial have already been discussed in previous publications to which he refers [2, 3, 4]. Dr. MacKenzie cites the ENHANCE trial as evidence of the risks of bleeding but fails to provide data on the background incidence of intra-cranial haemorrhage in critically ill patients in general. Dr. Mackenzie also omits to mention that the 28-day mortality of the 2,375 patients enrolled in ENHANCE using similar inclusion criteria to the PROWESS trial mirrored the mortality of the intervention arm of the PROWESS trial (25.3% vs. 24.7%, respectively) [5]. Claiming that there are inconsistent guidelines is unfair when the data upon which the European Medicines Agency based its decision in 2005 was simply not available to the National Institute of Clinical Excellence in 2004. Basic statistical principles dictate that long-term outcome studies require large samples of patients, and looking for survival differences at 30 months in a trial designed to explore 28-day mortality differences is therefore unsound. Furthermore, it is important to remember that the PROWESS trial was stopped early by the independent data monitoring committee, and thus arguing that significance of results changes following reallocation of seven deaths is flawed. …