Procalcitonin-guided therapy for acute respiratory illnesses

in February 2017, the US FDA approved using procalcitonin as a marker to guide antibiotic therapy in patients with acute respiratory infections, indicating that this assay could be used to determine if "antibiotic treatment should be started or stopped in patients with lower respiratory tract infections, such as community-acquired pneumonia, and stopped in patients with sepsis".  (see https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm543160.htm ). a recent meta-analysis confirmed that procalcitonin-guided therapy reduced antibiotic exposure and adverse events, and improved survival in hospitalized patients (see doi.org/10.1016/ S1473-3099(17)30592-3).

Details:

--26 randomized trials, 6708 patients from 12 countries

    --mostly from Europe, but some from China, Brazil, US, Australia

    --mean age 61, 57% men, 49% in ED/37% ICU/15% primary care

    --primary diagnosis: 

        --URI 9%

        --lower respiratory tract infection 92%: 44% community acquired pneumonia, 19% COPD exacerbation, 9% acute bronchitis, 8% hospital-acquired pneumonia, 6% ventilator-associated pneumonia.

    ​--procalcitonin levels: <0.1 mg/L in 35%, 0.1-0.25 mg/L in 20%, 0.25-0.5 mg/L in 12%, 0.5-2.0 mg/L in 15% and >2 mg/L in 20%

    --most of the studies used cutpoint of 0.25 mg/L, though some used 0.1 mg/L, 0.5 mg/L or 1.0 mg/L

--primary endpoints: 30-day mortality, and setting-specific treatment failure. secondary endpoints: antibiotic use, length of stay, antibiotic adverse events

Results:

-- 30-day mortality: 286 (9% of 3336 patients) in group with procalcitonin-guided therapy; 336 (10% of 3372 patients) in controls: a 17% risk reduction, OR 0.83 (0.70-0.99), p=0.037

    --benefit was across subgroups by setting and type of infection, though mortality was very low in outpatient setting (only 1 patient) and in patients with acute bronchitis (2 patients).

--antibiotic use was 2.4 days less in the procalcitonin-guided therapy group ( 5.7 vs 8.1 days, p<0.0001)

--antibiotic-related adverse events was 34% lower in the procalcitonin-guided therapy group, OR 0.68 (0.57-0.82), p<0.0001

--treatment failures were non-significantly lower in the procalcitonin group (23.0% vs 24.9%, 10% lower but p=0.068)

--initiation of antibiotics:  lower in the procalcitonin group in each location (overall: 71% in the procalcitonin group and 86% in controls):

    --outpatient care: 23% vs 63%, with total days of exposure to antibiotics 1.6 vs 4.6

    --ED: 69% vs 83%, with total days of exposure to antibiotics 5.2 vs 8.2

    --ICU: 92% vs 99%, with total days of exposure to antibiotics 8.1 vs 49.5

--disease-specific outcomes:

    ​--there were significant decreases in antibiotic initiation for community-acquired pneumonia (CAP) 93% vs 99%, COPD exacerbations 43% vs 72%, and acute bronchitis (26% vs 66%)

Commentary:

--procalcitonin, a calcitonin-related gene product, has been used for years to assess the likelihood of bacterial pneumonia/systemic infections in hospitalized patients. it is expressed by human epithelial cells in response to bacterial infections but downregulated in viral infections. Procalcitonin levels increase within 6-12 hours of a bacterial infection, correlate with disease severity/outcomes, is more specific for bacterial infection than other inflammatory markers (eg CRP, WBC, ESR) and falls rapidly during recovery from these infections.

--acute respiratory illnesses are both a common cause for hospital admissions/mortality but also the most common cause of antibiotic overuse (>70% of the 100 million outpatient visits for this lead to inappropriate antibiotic prescriptions: see http://gmodestmedblogs.blogspot.com/2016/01/antibiotic-overprescribing-and-acute.html ).

--but, there were only 2 trials that enrolled primary care patients (n=1008)

--limitations of the meta-analysis: there were limited number of trials/patients in outpatient setting (where the majority of antibiotic overuse happens), it assessed procalcitonin use only in acute respiratory infections (not sepsis, other bacterial infections), there was no control over the structure of the trials as a meta-analysis ​(ie, inclusion/exclusion criteria, and variable adherence to the algorithms for continuation of antibiotics for the differing procalcitonin cutpoints, ranging from 44-100%), and there were no blinded outcomes in half the trials.  

--this meta-analysis and the FDA approval elevate the validity of procalcitonin use and suggest large benefits if its use were more generalized.​

so,

--this article is very important given the increasingly emergence of antibiotic-resistant bacteria.  There is an international priority to decrease antibiotic use as a result. see below for a slew of prior blogs addressing this quite concerning development.

--of note in this meta-analysis (and not surprisingly), the most dramatic decrease in the initiation of antibiotics and the total antibiotic exposure was greatest (by far) in the outpatient care setting.  [ie, most of the really sick patients in the ICU will get antibiotics no matter what the procalcitonin level is...]

--the cutpoint for procalcitonin used in the 2 outpatient care studies was <0.25 mg/L. 

--given the lack of some specific data (eg utility of procalcitonin in immunocompromised population, or even outpatients with lots of comorbidities), I would be hesitant to generalize these results too broadly. But for healthy people with URIs, bronchitis, or even possible community-acquired pneumonia who are stable, this test, with its quick turnaround (our hospital-based lab gets results within 2 hours), may be very useful to reassure patients (and us) if negative, assuming there can be good followup to make sure the patient is doing okay. [on subgroup analysis, however, there was clear benefit for procalcitonin-guided therapy for community-acquired pneumonia and COPD exacerbations. but this should probably be tested directly in the outpatient setting]

--but, we really do need more RCTs using prespecified procalcitonin cutpoints/assessing outcomes to know for sure that this intervention is reliable, especially in the outpatient setting​ 

--And, i do think that patient expectations and provider approaches, at least from my limited observations at our health center, have changed over the past couple of decades. fewer patients seem to be insisting on antibiotics, and have a much easier acceptance (with the example of rapid strep tests) of "and, the good news is that you do not have a bacterial infection and do not need antibiotics...".  when appropriate, low procalcitonin levels could probably produce similar results for acute bronchitis and other respiratory viral illnesses, the main culprits for antibiotic overprescribing as noted above.

the blog noted above (http://gmodestmedblogs.blogspot.com/2016/01/antibiotic-overprescribing-and-acute.html  ) has the article on antibiotic overprescribing, but also several others showing some improvement in decreasing antibiotic prescriptions, and references to several blogs on the daunting increase in antibiotic-resistant bugs worldwide

see http://gmodestmedblogs.blogspot.com/2017/07/take-full-course-of-antibiotics.html​ for review of really interesting article challenging the widely-held concept that we should encourage patients to "take the full course of antibiotics", which may in fact lead to much more antibiotic resistance in some of the most dangerous pathogens