NEW YORK (360Dx) – Many physicians and healthcare providers are questioning a study recently published in the New England Journal of Medicine that said that the use of procalcitonin testing for lower respiratory tract infections did little to reduce antibiotic use in hospitals.
In the study, called the Procalcitonin Antibiotic Consensus Trial (ProACT), published in May and involving 14 hospitals in the US and 1,656 patients, the authors concluded that providing procalcitonin assay results, when compared with usual care, did not reduce antibiotic use.
"The conclusion of the study is not congruent with my many years of experience of using PCT in the emergency department," Michael Pulia, medical director of the emergency medicine antimicrobial stewardship program at BerbeeWalsh Department of Emergency Medicine at the University of Wisconsin School of Medicine and Public Health, said in an interview. He advises diagnostic industry executives about the use of procalcitonin testing, but is an independent researcher and not financially connected with any companies.
The value of procalcitonin as a marker of infections becomes apparent when a physician struggles to decide whether a patient needs an antibiotic, but that scenario is specific to clinical conditions and circumstances, he said. "The ProACT study," he said, "mixes conditions for which procalcitonin has little value in prescribing with other conditions where it has tremendous value, and the devil is in the details.
"Applied to every patient walking in the door with a lower respiratory condition, its value is going to be watered down," Pulia said.
The study's results contradict the results of other published studies, as well as the perspectives of clinicians directing emergency department antimicrobial stewardship programs and regulatory actions that suggest PCT assays could stem the overuse of antibiotics. Last May, for example, the US Food and Drug Administration expanded the use of BioMérieux's Vidas BRAHMS PCT automated assay to help clinicians make decisions regarding the optimal use of antibiotics in lower respiratory tract infections and sepsis.
The overuse of antibiotics remains a serious global issue with severe clinical consequences. In recent years, the FDA has cleared several assays for use in treating medical conditions sometimes related to sepsis. They include a PCT-based assay from Thermo Fisher Scientific that it acquired in 2009 when it purchased BRAHMS. Thermo Fisher has licensed the technology to BioMérieux, Roche, and other companies.
Most recently the FDA cleared Fujirebio Diagnostics' PCT assay for the Lumipulse G1200 immunoassay platform last week.
In ProACT, researchers from University of Pittsburgh School of Medicine provided guidance for clinicians about national clinical practice recommendations for the treatment of lower respiratory tract infections and the interpretation of procalcitonin assays. They noted that national authorities and medical societies in Europe have reached varying conclusions about procalcitonin-guided antibiotic prescription in suspected lower respiratory tract infection, and few studies were conducted in the US where there is lower adoption of PCT in prescribing antibiotics than in Europe.
Looking at patients enrolled between November 2014 and May 2017, ProACT's final analysis group of 1,656 patients — 826 in a procalcitonin group and 830 in the usual-care group — the researchers reported that providing procalcitonin test results did not result in less use of antibiotics, compared to usual care of patients with suspected lower respiratory tract infection
The treating clinician received procalcitonin assay results for almost 96 percent of patients in the procalcitonin group and for just over 2 percent of the patients in the usual-care group.
Since the publication of the study, some clinicians are voicing doubts about the results.
"It was a very unexpected result and that is partly why it is getting so much attention," Duke University Researcher and VA Health Care System physician Ephraim Tsalik said in an interview. "When this type of trial was done multiple times and in different hospitals and populations in Europe, you always saw a reduction in antibiotic use without any negative impact on patient outcomes."
Looking into the details of the ProACT study in the US, however, a lack of the adherence by some physicians with an algorithm recommending withholding or prescribing antibiotics was seen, he said. "When the algorithm said to withhold antibiotics, they prescribed them anyway."
Tsalik, who was not involved in the ProACT study, is the principal investigator of another study to determine whether low levels of procalcitonin can reliably reveal whether a person’s lower respiratory tract infection will improve with antibiotics.
The design of that trial, which is sponsored by the National Institute of Allergy and Infectious Diseases, is "fundamentally different" from the design of the ProACT trial, he said. The investigators in the NIAID study are holding discussions with physicians prior to their patients' being enrolled. If the physician believes she needs to administer an antibiotic regardless of whether the procalcitonin level is low or high, the patient is not included in the study, Tsalik said.
He noted that he and his colleagues will continue to recruit patients for the trial during the next year and a half, and its design would mitigate the issue of physicians' ignoring guidance to withhold antibiotics based on a procalcitonin score.
Why change didn't happen
There are probably many reasons for the ProACT study results, according to David Huang, a principal investigator of that trial and an associate professor at the University of Pittsburgh.
"Among them, PCT levels generally tracked with illness severity and clinicians' estimate of the likelihood of bacterial etiology," he said in an interview. As a result, tracking PCT levels could have provided "minimal incremental value over clinical judgment…and decisions to initially withhold antibiotics on the basis of procalcitonin level were subsequently overruled in the outpatient setting."
The researchers noted that in the usual-care group, clinicians prescribed antibiotics less frequently to patients with lower procalcitonin levels than to those with higher levels even when they didn't know the procalcitonin assay result.
"Patients with lower procalcitonin levels also had fewer clinical features of infection, and in that context, procalcitonin probably provided a modest amount of additional information to guide decisions," they wrote.
In another US study that contradicts the ProACT trial, Five Rivers Medical Center Pharmacy Director Mike Broyles reported in October 2017 that the addition of PCT in a facility with an established stewardship program "resulted in a significant reduction in antibiotic exposure and adverse outcomes."
The retrospective cohort study compared data collected from more than 2,100 patients who had been treated at the medical center. In the study, the investigators evaluated data from four years prior to and after a PCT algorithm was used to guide initiation and the discontinuation of antibiotics, as well as to decrease antibiotic exposure without increasing adverse clinical outcomes.
"Our approach to the management of the patient population was significantly different from that used in the ProACT trial," Broyles said in an interview. "This is in no way a reflection of the trial," because the ProACT researchers were evaluating a broader population than his study, he said. "The medical conditions in their patient population was less severe than ours. They included patients with respiratory asthma and acute bronchitis exacerbations that we did not include."
BerbeeWalsh's Pulia noted that in the ProACT trial, a significant percentage of the people had asthma as a final diagnosis, and procalcitonin testing has almost no value in asthma diagnostics. Further, the trial included a significant population of patients with COPD, and there's sparse evidence to indicate that PCT should be used to guide a decision about antibiotics for that condition. "It's interesting in my mind that 70 percent of the trial is made up of people for whom I would rarely order a procalcitonin test, and very little data suggest it holds value in those conditions," Pulia said.
Also, for patients with pneumonia, clinicians in the trial largely ignored the procalcitonin algorithm, he said, and prescribed antibiotics as a precaution even though evidence indicates that a large percentage of pneumonia cases are viral and don't require an antibiotic.
"There's a lot of work to be done to convince physicians that it is okay to not treat a clearly diagnosed pneumonia with antibiotics," he said. "But this was the most important group of patients from the viewpoint of where the use of procalcitonin could have had an impact on prescribing antibiotics. It had the smallest part in the trial, and the highest rate of non-compliance among physicians."
Pulia said that he has used PCT testing since 2014 on hundreds of patients, and over the years, he has grown to understand how it can be used in "clinical sweet spots."
There are a lot of factors to be considered to make sure it's the safe thing to do, he noted, because it is not a perfect assay. "But it should give you an extra piece of effective data that reassures you to do the right thing and say, 'Let's hold this antibiotic and get close follow-up … for the patient.' That's a groundbreaking paradigm shift for stewardship, and I am not surprised that physicians did not buy into it in the study."
UPitt's Huang said that the ProACT trial researchers found some evidence of lack of trust in PCT. "Of the 466 time points at which antibiotics were prescribed despite low procalcitonin levels, the most common reasons were clinician belief that a bacterial infection was present," he said.