Antibiotic stewardship: Medical Grand Rounds at the University of Arizona

The term antibiotic stewardship has been tossed around and misused to the point of becoming nebulous. But there are some robust, evidence based processes that can be applied as brought out in this Grand Rounds presentation. Here are some of the key issues:


Overuse of antibiotics has consequences and it's not just about money

We're getting a handle on most health care associated infections but Clostridium difficile continues to rise at an alarming rate. There is a strong consensus that antibiotic overuse is driving the increase.


Prior antibiotic approval policies are not recommended

In dealing with really sick patients time is of the essence concerning the initiation of antibiotics. Moreover, the autonomy of the clinician is a key element in evidence based medicine. These objectives would be undermined by prior approval requirements and the grand rounds speakers do not recommend them, favoring other process improvements.


Sometimes it's just a matter of making the right diagnosis

Whether it's pseudocellulitis, pseudopneumonia or pseudoUTI antibiotic stewardship goes out the window if you don't make the right diagnosis. Over diagnosis of several infections is increasingly recognized and drives inappropriate antibiotic use.


Staph aureus bacteremia

There's a checklist of processes necessary for the best chance of a good outcome. These include:

Treating long enough (less is not necessarily more in this case).

Getting repeat blood cultures during treatment. The clock for duration of therapy starts on the day of the first negative culture. Contrast this with gram negative bacteremia, where it's not usually necessary.

Appropriate imaging for deep sites of infection.

Switching to a beta lactam if it turns out to be MSSA.

Considering alternatives to vancomycin for MRSA when the MIC is high even though below the breakpoint.

Appropriately addressing and removing lines.

Getting an ID consult. (Read here).


Candiduria

Candiduria is common but true candida UTI is uncommon. Candiduria does not usually warrant antifungal treatgment and when it does, special considerations apply.

The speaker gave some recommendations and drew from this article. From the article:

Treatment is indicated only for the following groups with asymptomatic candiduria: very low birth weight infants, patients undergoing urinary tract procedures, and neutropenic patients. The vast majority of patients should not be treated.

Patients who have symptoms of urinary tract infection should be treated. The treatment of choice is oral fluconazole. Amphotericin B and fluctyosine are less desirable alternatives, and there is little role for amphotericin B bladder irrigation.

Other antifungal agents, such as voriconazole, posaconazole, and the echinocandins, cannot be recommended for Candida urinary tract infections because very little active drug is found in the urine.

According to the speaker, if amphoterecin is used for candiduria it needs to be plain ampho, as lipid based amphoterecin does not get into the urinary tract. If candiduria is felt to represent candidemia or invasive disease elsewhere in the body, agents such as echinocandins might be justified if the goal is to treat the invasive disease in those areas, realizing that they will not clear the urinary tract.

Again, from the article, the approach to asymptomatic funguria would be as follows:

Repeat clean-catch urine culture to be sure not a contaminant

If cannot obtain clean-catch urine, obtain urine by catheterization

Candida grows on repeat culture: correct predisposing factors (stop antibiotics, remove catheter)

Repeat urine culture after correct predisposing factors

Culture remains positive: obtain ultrasound to look for obstruction

No obstruction: observe and do not treat

Obstruction present: urology consultation

Treat with fluconazole if procedure performed to correct obstruction

When to deescalate

In the case of gram negative infections it is largely a matter of clinical judgment. It can be guided by sensitivity results with the caveat that acquired gram negative resistance is a known issue in certain situations, eg infections with AmpC producing organisms. In the case of empiric MRSA coverage there is a pearl: if initial cultures do not grow staph within 48-72 hours the coverage can be discontinued. Staph aureus grows hardily. It declares itself or, in the words of the speaker, “doesn’t hide out.”


Duration of treatment: less may be more

The notable exception to this is Staph aureus as addressed above. For other types of infections we tend to treat too long. In pneumonia, for example, 5 days is enough for CAP, 8 days for HCAP.