I must admit I had felt a little smug about my discovery of Elsa Bruegger’s faulty walker. It really seemed like a very logical explanation to her walking into walls all the time. As it happened, her new walker didn’t quite solve the problem. She continued to be off balance and sometimes did seem a bit unfocused, even downright sedated.
Looking back through her record, I came across a mildly elevated ammonia level a few months ago. I remember speaking with her psychiatrist back then about Elsa and a couple of other patients we share, whose routine ammonia levels were mildly elevated. Elsa had a standing order from the psychiatrist for ammonia levels every three months because of her valproic acid (Depakote) prescription for her mood disorder.
All my research has led me to the conclusion that ammonia levels are of little or no value in predicting whether patients on valproic acid are headed for trouble due to the drug’s unpredictable tendency to cause ammonia to build up within the central nervous system. I have come to understand that ammonia levels are only slightly helpful even in assessing a patient with coma or near coma; the correlation between brain levels and peripheral blood levels of the toxic ammonia relate poorly to each other because of how the blood-brain barrier works to keep the chemistries inside and outside the central nervous system separate. Many experts recommend against routine measurements of ammonia levels for this reason.
Watching Elsa fumble her way down the hall, I decided to order an ammonia level “just in case”. It came back elevated – twice as high as it had been ten weeks ago. Her liver function tests were normal.
I ordered her valproic acid stopped and made sure her psychiatrist got a copy of the lab report and my notes.
This week, Elsa is finally walking straight. She is attending her day program, says “good morning”, makes good eye contact and smiles. She also shows more of a temper, but nothing inappropriate.
Maybe this time I finally got it right.
It takes great humility to reassess a diagnosis. Thanks for the reminder that a diagnosis is never absolute or finite.