Tuesday, March 23, 2010

Infectious disease pearls for hospitalists

---are provided in a recent review in the Journal of Hospital Medicine.

Points of interest:

Don't chase your tail with antibiotics or, as the authors put it, avoid spiraling empiricism. Although it may at times be necessary to escalate antibiotics in a non-responding patient don't do it willy-nilly. Think and, if clinically appropriate, re-evaluate before you do.

In bacteremic patients do serial blood cultures (every 24-48 hours until clear) especially in infections with staph, enterococcus and yeast. The results impact decisions on duration of treatment and source evaluation.

Remove lines from patients with candidemia to optimize the chance for a good outcome.

Candida colonization is common and usually dismissed but be suspicious if you grow it from multiple sites.

Don't treat asymptomatic bacturiuria except in pregnant patients and those about to undergo GU manipulation.

A similar open access article from The Hospitalist is dated but still relevant.

No comments: