Published in the October 2015 issue of Today’s Hospitalist
Editor’s note: This is the second in a two-part series on pneumonia care
WHAT DIAGNOSTIC TESTS should hospitalists use “and avoid “when treating pneumonia patients? At this year’s Society of Hospital Medicine meeting, Scott A. Flanders, MD, the hospitalist program director at the University of Michigan in Ann Arbor, discussed the pros and cons of various diagnostic tests.
In general, Dr. Flanders urged hospitalists to refrain from “routinely” ordering diagnostic testing for pneumonia patients. Instead, he said, physicians are better off using tests “strategically to target treatment,” particularly to help make decisions about stopping, changing or deescalating antimicrobial therapy.
“There are a lot of hospitalized patients you don’t have to send blood cultures for.”
~ Scott A. Flanders, MD, University of Michigan
“We have not yet seen that routine diagnostic testing can improve outcomes for most patients with pneumonia,” he explained.
Here are the tests Dr. Flanders discussed:
Serum procalcitonin, said Dr. Flanders, is “very appealing” and can help physicians make deci sions on antibiotic regimens by ruling out bac terial infections. He proposed that hospitalists follow an algorithm derived from an article in the April 2012 issue of Chest for determining what levels of serum procalcitonin would warrant starting or withholding antibiotics. (See “Using procalcitonin to rule out bacterial infections.”)
He noted that the study had what he called an “opt-out clause.” Researchers said that even if the procalcitonin was low and the patient looked sick, Dr. Flanders explained, physicians should feel free to give antimicrobials.
But he was quick to offer several caveats. “It is not a perfect test,” he pointed out, so it shouldn’t replace clinical judgment. “If you are convinced the patient needs antibiotics, don’t send this test. You shouldn’t change your decision based on the test in and of itself.”
He offered the example of a woman he saw recently who had classic bacterial pneumonia and possible shock. Ordering a serum procalcitonin would be “another test to increase the cost of care of pneumonia patients without having any impact on her overall outcome.”
An example on the other side, Dr. Flanders said, involved a patient with a new diagnosis of aortic stenosis and respiratory distress. Her chest X-ray cleared up quickly after she was aggressively diuresed, and she was transferred to the cardiology service. There, after a procalcitonin sent by the emergency department came back as abnormal, the cardiologists unnecessarily restarted her on the antibiotics that the hospitalist had stopped.
“It was obviously flash pulmonary edema,” Dr. Flanders said, “but because the procalcitonin was high, they started antibiotics. That’s not the way you want to use procalcitonin.”
Pneumococcal urinary antigen
The pneumococcal urinary antigen (Binax NOW) “warrants a closer look,” Dr. Flanders said, because its sensitivity is “less than perfect”: It misses patients who have pneumococcal infection. “Just because it is negative doesn’t mean you can avoid covering for pneumococcus,” he noted. “Its utility is in its specificity. If the test is positive, it is highly suggestive that your patient has a pneumococcal infection.”
Moreover, he added, a small and worrisome study was published online by Thorax in August 2009. It found that patients whose antibiotics were narrowed based on the antigen had more problems with relapses than those who didn’t.
“It may be a reasonable test for understanding when a patient’s therapy can be narrowed,” Dr. Flanders said, but again, it may be overused. “If you would never use narrow spectrum treatment for a patient, don’t send it. All it does it add to costs.”
Hospitalists order “a lot” of blood cultures to screen patients for bloodstream infections. But those cultures come back positive in pneumonia patients only between 4% and 7% of the time. This means, Dr. Flanders said, that the cultures “generate a lot of excess resource utilization.” Physicians need “some sort of diagnostic algorithm for predicting bacteremia to figure out the threshold to send it or to feel comfortable that they need to send it.”
The best of these, he said, was presented in an Aug. 1, 2012, article in the Journal of the American Medical Association. That article concluded that the only patients who need to have blood cultures sent are ones who have at least two systemic inflammatory response syndrome (SIRS) criteria. Otherwise, patients will have “a very low chance” of a positive blood culture and bacteremia. “That means there are a lot of hospitalized patients you don’t have to send blood cultures for.”
Viral PCR tests
Viral PCR tests, meanwhile, can be useful, but only with several subsets of patients. The problem is that one-fifth of CAP patients with bacterial pneumonia will also have viruses, “so it makes it hard to figure out what to do with a positive test,” Dr. Flanders said. The test can be useful, however, for patients suspected of having a viral pneumonia who were started on empiric oseltamivir. If the viral PRC comes back negative, you can stop the oseltamivir.
The test is also used for patients who don’t have the usual signs and symptoms of bacterial pneumonia. If the PCR is positive for virus, Dr. Flanders said, “that is an appropriate situation when you can try a patient on something other than antimicrobials and see how they do.”
Sputum tests tend not to lead to improved outcomes, Dr. Flanders said, but they are recommended for ICU patients and for patients at risk for multidrug resistant organisms. “You can send it in those situations, but you should not send it routinely,” Dr. Flanders said.
The same is true for urinary legionella. “For critically ill patients and patients you have a suspicion about, it’s a reasonable test to send out.”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.
THE FOLLOWING ALGORITHM, which was published in the April 2012 issue of Chest, helps guide doctors in using serum procalcitonin levels to decide when to start or withhold antibiotics in patients with community-acquired pneumonia in the ED or ICU or on the wards.