Thursday, September 14, 2017

Are blood ammonia levels helpful?

Here are their recommendations:

HE is a diagnosis of exclusion and is made on clinical grounds.

Do not check serum ammonia levels in patients with CLD to diagnose HE, to assess the severity of HE, or to determine whether HE is resolving.

Use your clinical evaluation to determine the severity and course of HE.

Treatment should be tailored according to clinical findings, not ammonia levels.

I get a little nervous about absolute recommendations to stop a widely accepted, physiologically rational, low tech, low harm practice just because there’s a lack of “high level” supporting evidence. An unfounded assumption here is that a lack of high level evidence equates to evidence against.

Another questionable assumption is that the test inherently is bad. Certainly at the cutoff of 55, as cited in the article, the test characteristics are poor. At extreme values, however, it may be more helpful. In a comatose patient presenting to ER whose baseline level is known to be 40, for example, a triple digit ammonia may add greatly to the diagnostic confidence and obviate an MRI and LP pending a therapeutic trial directed at HE. In such a case it might even save resources. This article strikes me as a case of black and white thinking. How about a more nuanced approach in which the ordering threshold is proportional to the cost and potential harm of the test?

No comments: