In the ongoing battle against antibiotic overuse in hospitals, some experts are seeing one biomarker test as a potential game-changer.
Many physicians still default to a conservative approach when prescribing antibiotics, said Michael Pulia, MD, MS, assistant professor and medical director of the emergency medicine antimicrobial stewardship program at the University of Wisconsin School of Medicine and Public Health in Madison. “When there's uncertainty, especially around the diagnosis of pneumonia, the physicians will prescribe. This effect is compounded if they're worried about patient follow-up or liability,” he said. “The reality is people are still doing it, and they're not going to stop doing it unless they have objective data to guide their decision making.”
A procalcitonin assay may be the test to provide this type of support, giving cautious clinicians the green light to avoid antibiotics when a bacterial cause of infection is unlikely. In 2017, the FDA approved use of a procalcitonin test for guiding antibiotic use in patients with respiratory infections or sepsis.
However, the test's uptake in the U.S. has been slow compared to Europe, where it's been routinely ordered for more than 15 years, said Ephraim Tsalik, MD, PhD, an infectious diseases physician at the Durham VA Health Care System in North Carolina.
“It certainly is increasing at a faster pace as clinicians become more familiar with it and more comfortable with it, and I suspect that the new FDA indication for use will only further grow clinicians' interest in it,” he said.
For such clinicians, experts explained the promise of procalcitonin and some caveats and barriers to its use.
The test's promise
Procalcitonin is an infection biomarker, and the higher its level in the blood, the higher the likelihood of bacterial infection.
Last year's expanded FDA indication for the first procalcitonin test, the Vidas Brahms PCT Assay, applies to two scenarios: starting or stopping antibiotic treatment in patients with lower respiratory tract infections, and stopping antibiotic treatment in patients with sepsis. The test was originally approved in 2007 to aid in risk assessment for sepsis and septic shock on the first day after admission to an ICU. It also received approval in June 2016 to help assess a patient's risk of dying of severe sepsis.
Clinicians who test the procalcitonin level of a patient with a suspected or confirmed lower respiratory tract infection can use a clinical algorithm to guide antibiotic use. The algorithm, included in the test's instructional booklet and based on previous trial protocols, discourages antibiotic therapy if these patients have procalcitonin levels less than or equal to 0.25 ng/mL and recommends it at levels greater than 0.25 ng/mL. For ICU patients with sepsis, antibiotics are discouraged at procalcitonin levels less than 0.5 ng/mL and encouraged at levels greater than or equal to 0.5 ng/mL.
Using procalcitonin to guide antibiotic therapy can help decrease mortality rates and exposure to antibiotics, according to an October 2017 Cochrane review of 26 randomized trials that included a total of 6,708 participants with acute respiratory infections.
Patients who received procalcitonin-guided therapy had significantly lower rates of 30-day mortality than control patients (8.6% vs. 10.0%), and the procalcitonin group also had a 2.4-day reduction in exposure to antibiotics (5.7 days vs. 8.1 days) and a lower risk of antibiotic-related side effects.
Adopting procalcitonin testing can also save hospitals money, mainly by decreasing unnecessary antibiotic use, according to a modeling study published in March 2015 by Clinical Chemistry and Laboratory Medicine. Procalcitonin-guided care resulted in a net savings of about $700,000 compared to usual care in 2014, according to a cost-impact model based on a cohort of 1 million members within a typical U.S. integrated delivery network.
The test may not be all that costly to implement, either. It's not a particularly expensive assay, experts said, as the CMS reimbursement rate is about $25 to $40 per test. And at Duke Regional Hospital in Durham, N.C., which implemented the test in 2016, the machine that runs it cost $15,000, said John Boreyko, PharmD, an infectious diseases pharmacist and co-director of the antimicrobial stewardship program. “When you look at all these $500,000, $600,000 instruments, that's pretty inexpensive,” he said.
At the University of Wisconsin School of Medicine and Public Health, Dr. Pulia said the first time he heard about procalcitonin was in 2014 when he was admitting a patient with suspected pneumonia. “One of our hospitalists said, ‘Hey, can you add a procalcitonin?’ and I was like, ‘Never heard of it.’”
That was shortly after the institution had set up a platform to run the test in real time, and it was years before the expanded FDA approval. “We submitted a letter about our experience to the FDA when they were reviewing the approval. . . . They basically looked at us as an institution that had a tremendous amount of real-world experience with procalcitonin for antibiotic stewardship,” said Dr. Pulia.
These days, he frequently orders the test in cooperation with the hospital medicine group as they admit patients with respiratory conditions. On the inpatient side, it's mostly used for de-escalation of antibiotic therapy, whereas in the ED it's more of a diagnostic aid, Dr. Pulia said.
“It's cases of diagnostic uncertainty, particularly involving pneumonia, where it adds tremendous value,” he said. “I've had cases where . . . I thought it was a benign viral lower respiratory infection, and the procalcitonin would come back in the near-sepsis range. In this scenario, it could potentially prevent you from sending somebody home with unrecognized bacterial pneumonia or developing sepsis.”
For best results, physicians should monitor procalcitonin levels on a serial basis when the initial level is elevated, said Philipp D. Schuetz, MD, lead author of the Cochrane review. Hospitalist Bartho Caponi III, MD, FACP, an associate professor at University of Wisconsin School of Medicine and Public Health, said he routinely orders the test both on admission and at the 48-hour mark.
“I often get that initial procalcitonin if I'm thinking lower respiratory tract infection to give me a better idea of what I've got to do, and the 48-hour procalcitonin is another place where I've gotten a lot of mileage in terms of being able to de-escalate sooner,” he said.
Regardless of the setting or testing strategy, it is important to remember that procalcitonin “only works in conjunction with clinical judgment,” Dr. Schuetz said. Dr. Caponi agreed, saying that while it shouldn't be the sole factor in treatment decisions, it offers another data point to help hospitalists be good stewards of antibiotics.
“I think procalcitonin is a very useful tool to help me determine how to care for my patients in the best and safest manner possible,” he said.
Along with the benefits of the procalcitonin test, there are also caveats to remember, said hospitalist Bryan J. Huang, MD, ACP Member, associate professor of medicine at the University of California, San Diego.
“There probably aren't serious downsides to using the procalcitonin test, per se, but the test can be used incorrectly,” he said.
After his medical center introduced a rapid-turnaround version of the test in April 2017, clinicians used it for indications lacking established evidence, such as cellulitis, urinary tract infections, and gastrointestinal infections, Dr. Huang said. His group's findings were presented in an abstract at Hospital Medicine 2018.
“We also found that antibiotic use was often discordant with the procalcitonin value (i.e., continuation of antibiotics despite a low procalcitonin). . . . Thus, it's important that providers receive proper education regarding the use of the procalcitonin test when it becomes available,” said Dr. Huang.
Before ordering the test, clinicians should consider whether the results will change the care plan, said Dr. Tsalik, who is also an associate professor of medicine at Duke University School of Medicine in Durham, N.C.
“When I teach my students about ordering tests, I ask them, ‘What are you going to do with the result if it's normal?’ . . . If a test result is not going to change what you do, then it's not worth ordering. These are the caveats that come with any biomarker,” he said.
Another concern is that procalcitonin levels can be low early in infection, Dr. Tsalik said. “If a patient just became symptomatic within the past day or so, it's very possible that procalcitonin will not be elevated,” he said. “So thinking about the timing is important, and if you repeat it the next day, you may find that now it's high.”
Levels may also be low despite the presence of infection in immunocompromised patients or in those taking steroids, said Dr. Tsalik. “But it generally is a fairly robust biomarker, even in patients who are immunocompromised, because it's not produced by the immune system; it's produced by the tissues that are reacting to the infection.”
False positives can happen too, but they are fairly rare, said Dr. Pulia. Based on his experience and some scientific articles, “One particular case you've got to watch out for is people on dialysis, as they tend to have elevated procalcitonin values.” People with severe trauma injuries, burns, or thyroid cancer can also have falsely elevated procalcitonin levels, he said.
Clinicians should also remember that procalcitonin is not a perfect test, said Dr. Tsalik.
“They're all hoping for a troponin of infectious diseases—that if it's high, you had a heart attack, and if it's low, you didn't. It doesn't do that,” he said. “It's an important piece of the puzzle that up until now hasn't existed, but it's still just a piece of the puzzle. You have to take into account the rest of the picture when you're trying to figure out what it is that you need to do.”
Mixed study results
Adding procalcitonin testing to hospital practice has shown benefits for some facilities but not for others in recent studies.
After Duke Regional Hospital implemented a procalcitonin-guided protocol, researchers assessed antibiotic use in a cohort of about 500 patients, most of whom tested negative for procalcitonin during four months when testing was available, compared to a similar preintervention cohort.
The researchers found that, after adjustment, 53% of the postintervention cohort received antibiotics, compared to 87% of the preintervention cohort, a 34% reduction, according to results presented in October 2017 at IDWeek.
Another study of about 2,500 patients is under way, said Dr. Boreyko, who is one of the investigators. “I actually expect that 34% to be much higher now that physicians are trusting the protocol,” he said.
Researchers at Allegheny General Hospital and the Western Pennsylvania Hospital in Pittsburgh conducted a similar retrospective study in patients with pneumonia, which was also presented at IDWeek 2017 and later published in February 2018 by the American Journal of Medicine.
Compared to a preintervention cohort of about 150 patients, mean duration of antibiotic therapy decreased from 9.9 days to 6.0 days in the postintervention group of about 230 patients. Average length of stay also decreased, from 4.9 days to 3.5 days, while readmission rates were unaffected.
In contrast, the Procalcitonin Antibiotic Consensus Trial (ProACT) produced lackluster results, which were published in May 2018 by the New England Journal of Medicine. The study took place at 14 primarily urban academic U.S. hospitals, none of which had used procalcitonin in routine care.
Researchers randomly assigned about 1,650 patients who presented to the ED with an initial diagnosis of acute lower respiratory tract infection but with an uncertain need for antibiotics to either a procalcitonin group, where clinicians received initial (and serial, if applicable) test results, or to usual care. In both treatment groups, nearly 50% of patients were admitted, so the study included both emergency physicians and hospitalists. Clinicians in both groups retained autonomy in clinical decision making. By the end of the trial, there was no significant overall difference in the duration of antibiotic therapy (4.2 days in the procalcitonin group vs. 4.3 days in the control group) or adverse outcomes within 30 days.
Lead author David T. Huang, MD, MPH, an associate professor of emergency and critical care medicine at the University of Pittsburgh School of Medicine, said he was surprised by the results. He added that the implications are, “at least in hospitals and EDs similar to the 14 in ProACT, to not expect much, if any, antibiotics reduction for lower respiratory tract infections” with procalcitonin use.
However, the results were not surprising to Dr. Pulia because the study population primarily consisted of conditions for which antibiotics are not typically considered or recommended (e.g., asthma and bronchitis). The pneumonia group was underrepresented in the trial and had less than 40% clinician adherence to the procalcitonin-guided antibiotic recommendations, he said.
“Basically, the clinicians did not trust the procalcitonin and use it to guide therapy when it has the most impact, based on my experience. Suspected bacterial infection was the most common reason for deviating from procalcitonin guidance. Results that are ignored [are] equal to usual care,” Dr. Pulia said. “This seems mostly to have proven that U.S.-based emergency physicians do not trust procalcitonin to guide therapy when it comes to suspected pneumonia, but not the main question of whether or not it is a safe and effective approach for these patients.”
The trial did, however, implement the procalcitonin protocol using quality-improvement principles, including extensive use of education, prompts, and feedback, noted Dr. Huang, its principal investigator. “This design mimics a best-case scenario for deployment of a new intervention in the U.S. health care setting,” he said.
Ultimately, clinicians need to trust the test for it to make an impact, said Dr. Boreyko. “That [trust] only comes with adjudicating hundreds if not thousands of results and correlating it with patient outcomes. Providing a clinician with a number and guidance does not replace . . . trust in the number.”
At Washington University School of Medicine in St. Louis, associate professor of emergency medicine Christopher R. Carpenter, MD, MSc, has advocated for use of procalcitonin for respiratory infections but hasn't gotten much buy-in.
“There has been some reluctance to accept this lab test for a variety of reasons, because there are precedents where lab tests that looked very promising weren't proven to be very effective or were misused downstream,” he said.
A second reason for the reluctance is that most of the studies on procalcitonin were done in Europe. “For better or for worse, U.S. physicians tend to think that only studies that were done in U.S. citizens are believable for our population,” said Dr. Carpenter.
Finally, there's the fact that most of the investigators in previous procalcitonin trials had disclosures related to industry sponsorship, he noted. “It's something just to keep in the back of our minds that there may be some conflicts there, but I think that we still need to pursue this idea,” Dr. Carpenter said.
He's planning to gather clinicians at his facility from infectious diseases, lab medicine, pulmonary medicine, emergency medicine, and primary care for a journal club looking at the data on procalcitonin testing in patients with respiratory infections. “I think that'll be a catalyst to precipitate change at our institution to make it available to us.”
But convincing all the necessary stakeholders can be a lengthy process. Dr. Boreyko said it took him almost 18 months to receive approval and trust from both the medical staff and hospital administration to implement the procalcitonin test at Duke Regional Hospital.
For hospitals that do have the test, yet another barrier may be the inability to run it in real time. This is especially obstructive for emergency physicians, as send-out tests prohibit timely use in the ED, said Dr. Pulia, whose own institution can run the test STAT and provide results in an hour or so.
“Maybe that send-out's going to help [on the wards] the next day or something, but I'm really pushing [clinicians] to try to change initial antibiotic decision making in the ED, and it's impossible when they're send-out tests,” he said. The assay can now be run on more platforms than in the past, which should increase accessibility, Dr. Pulia noted.
Introducing procalcitonin testing also requires recommendations and guidance from an antimicrobial stewardship program, or it's not likely to make much of a difference in clinicians' prescribing, Dr. Tsalik said. At a minimum, small hospitals without stewardship personnel should incorporate the guidance that comes with the test into the test results, he suggested.
“If you make this test available, but no one's there to provide the guidance necessary to know how to use it, then it didn't really do anything other than waste dollars,” said Dr. Tsalik.