All Content

HM13 Session Analysis: Pneumonia Update


Scott Flanders, MD, SFHM, director of the hospitalist program at the University of Michigan in Ann Arbor set out to answer three key questions in “rapid-fire” format during the “Pneumonia Update” at HM13.

  1. Is procalcitonin ready for “prime time” in determining community-acquired pneumonia (CAP) treatment duration?
  2. What is the utility of diagnostic testing in CAP patients?
  3. How do you decide which pneumonia patients need broad-spectrum antibiotic coverage?

It turns out that lots of other countries follow procalcitonin levels as a marker of inflammation during CAP treatment. A 2012 Archives of Internal Medicine article notes that U.S. compliance with measuring procalcitonin levels is less than 40%, and monitoring these levels can help diagnosis and guide treatment and the duration of treatment. Procalcitonin is released in the blood in response to bacterial infection. It rises within four hours of infection (earlier than other markers such as CRP and ESR.) The degree and rate of rise is associated with severity; the rate of decline is associated with resolution. Numerous studies show that when providers correlate CAP treatment with procalcitonin levels there is a safe reduction in antibiotic days.

Dr. Flanders also examined the utility of diagnostic tests. In general, CAP outcomes are unchanged and management rarely is impacted by sputum collection. Within the ICU setting it is recommended to get sputum cultures, as it may have a role in healthcare-associated pneumonia, especially if a patient has a history of drug-resistant organisms.

With regard to blood culture analysis, only 4% to 7% of blood cultures are positive in CAP, with many of them being false positives. False positive cultures lead to a 50% increase in charges, and increase length of stay by 65%. A 2004 American Journal of Respiratory and Critical Care Medicine article recommends targeted blood culture screen that correctly detect 90% of bacteremia with 40% fewer cultures. It recommends that patients who are at risk for bacteremia (those with prior antibiotics, WBC count greater than 20, systolic BP less than 90, history of liver disease, temperature greater than 40 degrees or less than 35 degrees celsius, elevated BUN greater than 30, sodium less than 130, pulse greater than 125) be given a point for each risk factor. Those with no risk and no prior antibiotics were deemed safe to forgo cultures. Those with one risk factor, with prior antibiotics were recommended to get one set of cultures. Those with more than one risk factor were recommended to receive two sets of cultures.

Pneumococcal urinary antigen was evaluated. It is noted to have great specificity, but lousy sensitivity. Patients with bacteremia might have false negative results. In general, the antigen might be appropriate in non-severe cases if it will help you narrow therapy. But it shouldn’t be ordered if it is not going to change therapy.

Dr. Flanders also noted that urinary legionella antigen is 80% sensitive for legionella.

Answering the question about the need for broad-spectrum antibiotics, it was thought that any patient receiving home care or home wound care, goes to a dialysis center, lives in a NH or LTC facility would need broad-spectrum antibiotics for HCAP. But Dr. Flanders states it may be a case of doing too much too fast. He recommends patients that reside in nursing homes or who receive home care be treated as a CAP, as the risk of drug-resistant organisms isn’t actually that high in that group. But if a patient had previous admission to the hospital, he recommends treatment for HCAP.

Strong risk factors for resistant organisms include prior hospitalization in past 90 days, LTAC/SNF patients if they have had prior antibiotics and have poor functional status, critically-ill patients, or those with prior MRSA/pseudomonal infections. The data for nursing home patients, home health or home wound care or dialysis patients is less clear. TH

Tracy Cardin is a nurse practitioner in the section of hospital medicine at University of Chicago.

Next Article:

   Comments ()