Tuesday, September 2, 2008
Annals of Internal Medicine has a narrative review of the use of furosemide for hypercalcemia, a syndrome which is unfortunately not uncommon in advanced cancer patients. Its conclusion is straight-forward: it has no role in the treatment of hypercalcemia (which should instead be saline hydration, bisphosphonates, +/- calcitonin). Towards the end of it there are some philosophical remarks about evidence:
We show that despite more than 20 years of careful phase I to III research supporting saline hydration with bisphosphonates as the preferred first-line therapy in hypercalcemia and the lack of significant supporting evidence for furosemide, this medication continues to be routinely recommended for emergency management. As recently as 2005, a review recommended 20 to 40 mg of furosemide after adequate hydration; the author acknowledged that this was based on "historical precedent and common practice." Reliance on historical precedent is not limited solely to the management of hypercalcemia; many therapies have become habit, often without supporting evidence. We may not know where such a habit came from and may struggle to find its origin; however, that search may be informative and practice-changing. Such a search may have several possible outcomes: No evidence is found to support the current recommendations, leading to a new standard of care; limited but supporting evidence is found that argues for additional studies; or adequate evidence is found to support the existing precedent. We argue that our search of furosemide studies is an example of the first outcome.
To clarify the 'no evidence' language here - what is actually being talked about is that there is some evidence directly supporting the idea that furosemide is not very effective (as opposed to there just being an absence of data/research) plus solid evidence for an effective alternate therapy. This is the perennial Cochrane lament (I imagine the plaintive moaning of a loon) a la this post, and why it's important to clarify, at least for the nerds writing this blog, whether one is talking about an absence of research data or research findings with negative conclusions.
Annals also has a brief, chatty, practical look at navigating language barriers in difficult clinical circumstances. It is, perhaps, a little too brief, and does not specifically address end of life concerns, but would be an OK one for the teaching file, especially for medical students and interns.