COPD: CRP-guided antibiotic prescribing

A few recent articles found that too many antibiotics were being prescribed for pulmonary conditions.  will review an article on using CRP levels to guide antibiotic prescribing (see copd CRP guided abx nejm2019 in dropbox, or DOI: 10.1056/NEJMoa1803185). will do a subsequent one on overprescribing antibiotics in patients in the hospital with pneumonia

Details:

--653 patients in 86 general medical practices in UK who presented with an acute exacerbation of COPD were randomized to usual care guided by CRP point-of-care testing vs usual care alone

--mean age 68, 50% male, 7 days of symptoms prior to seeing clinician, Clinical COPD Questionnaire score 3.2 (a 10-item scale with scores from 0 reflecting good COPD status to 6 reflecting very poor status), FEV1/FVC=0.6, FEV1=59% of predicted, COPD severity: mild 11%/moderate (GOLD stage 2) 54%/severe (GOLD stage 3) 30%/very severe (GOLD stage 4) 5%, prior antibiotics in past 12 months 67%, use of regular inhalers 96%

--comorbidities: heart failure 5%, CHD 17%, diabetes 15%, CKD 8%, hypertension 38%

--smoking 34% current, 59% former 

--clinicians made their own decisions about whether to start antibiotics based on a comprehensive assessment of the likely risks and benefits of antibiotics, though for the CRP group there was guidance that CRP <20 mg/L made antibiotics unlikely to be beneficial, levels from 20-40 that antibiotics were possibly beneficial, and >40 mg/L that antibiotics were likely beneficial

--primary outcomes: patient-reported use of antibiotics for the acute COPD exacerbation within 4 weeks of randomization, COPD-related health status at 2 weeks using Clinical COPD Questionnaire

--seconday outcomes included prevalence of potentially pathogenic and resistant pathogens in sputum and commensal organisms in the throat

--mean CRP value in the CRP-guided group was 6 mg/dL, with 76% having values <20 mg/L, 12% with CRP 20-40mg/L, and 12% CRP >40 mg/L

--potential organisms from sputum samples at baseline (of the total of 216 potential bacterial pathogens): 53 H. flu, 50 M. catarrhalis, 35 S. pneumoniae, 15 P. aeruginosa, 13 S. maltophilia, 8 S. aureus, 5 K. pneumonia and the rest <5 cases found

Results:

--antibiotic use: 57.0% in CRP-guided antibiotic prescribing group vs 77.4% in usual care, adjusted odds ratio: 0.31 (0.20-0.47)

    --a total of 158 antibiotic prescription written for the CRP-guided group vs 234 in usual care, about 80-85% were for 7 days

        --of the antibiotics: 63% were for amoxicillin, 22% for doxycycline, 12% clarithromycin, rest <2%

--antibiotic prescribing at initial visit:  47.7% vs 69.7%, aOR 0.31 (0.21-0.45)

    --stratified by CRP status in the CRP-guided group, percent given antibiotics:

        --CRP <20: 33%

        --CRP 20-40: 84%

        --CRP >40: 95%

--antibiotic prescribing in first 4 weeks: 59.1% vs 79.7%, aOR 0.30 (0.20-0.46) [ie, vast majority of antibiotics writter

--no different between the CRP-guided vs usual care group in other treatments, such as steroid use; or the use of specialist consultants

--no difference in diagnosis of pneumonia in the first 4 weeks of follow-up, or those that had sputum samples containing antibiotic-resistant bacteria

--in the 6-month follow-up: 3% in the CRP-guided group and 4% in usual care had a diagnosis of pneumonia, OR 0.73 (0.29-1.82), not statistically significant

--adverse effects from antibiotics: no difference between the 2 groups

--Clinical COPD Questionnaire at 2 weeks: -0.19 points in favor of CRP-guided treatment (minimally clinically importance difference being 0.4)

--and no difference in health care-seeking measures at 6-months of follow-up

Commentary:

--background:

    --COPD was the 3^rd leading cause of death in the US in 2014

    --6.4% of Americans have a diagnosis of COPD; 2% in the UK

    --1/2 of patients with COPD have at least 1 acute exacerbation annually, leading to antibiotics, steroids or both

        --1/4 have 2 or more exacerbations per yr

    --more than 80% of patients with COPD exacerbations are prescribed antibiotics in the US and Europe

--the current guidelines promote the use of antibiotics, based on the Anthonisen criteria (increase dyspnea, increased sputum volume, and increased sputum purulence; though these criteria are derived from older studies, and increased sputum purulence by itself may be sufficient from newer studies)

--this study found much less antibiotic usage in those in the CRP-guided care group, with no difference in clinical effects as per the Clinical COPD Questionnaire or long-term outcomes (with trends for both favoring the CRP-guided therapy)

    --the vast majority of those receiving antbiotics got them in their initial clinical encounter (ie, those not getting them initially were unlikely to get them subsequently)

    --it is notable that 76% of the patients in the CRP-guided group had a low CRP level, confirming that most COPD patients with exacerbations are unlikely to benefit from antibiotics. we do not have CRP data for the usual care group (CRP levels could have been done but not disclosed to the clinician) or granular data on the relationship between specific Anthonisen criteria symptoms and antibiotic prescribing, but this study does suggest that we might be able to decrease antibiotic prescribing by about 3/4, with careful patient followup....

--these results of CRP-guided antibiotic prescribing are consistent with several prior studies

--the fact that S. pneumoniae comprised the 3^rd most common bacterium of the pathogens found (with 35 cases) also further supports the potential advantage of immunizing adults with PCV-13 (seehttp://gmodestmedblogs.blogspot.com/2019/07/acip-new-recommendations-for-hpv-and.html). though this is not strong support, given that we do not know if the PCV-13 vaccine would have covered these serotypes, or even that these were truly involved in causing the infection and not just commensals.

--one of the major concerns with antibiotic overprescribing is its effect on the respiratory (and GI) microbiomes, leading to antibiotic resistance (and the array of morbidities related to potential adverse microbiome changes, such as obesity, diabetes, heart disease, CNS problems, as outlined in a large number of blogs: http://gmodestmedblogs.blogspot.com/search?q=microbiome )

    -- eg see pneumonia abx and later resistance Infect2018 in dropbox, or doi.org/10.1007/s15010-018-1147-z : a recent case-control study for predictors of bacterial pneumonia (a bit different from COPD exacerbations) due to b-lactam-susceptible vs resistant pseudomonas, finding that recent exposure to b-lactam antibiotics (both those targeting pseudomonas and not) was associated with subsequent increased risk of multi-drug resistant bacterial airway colonization by both b-lactam-susceptible and resistant pneumonas infections 

    -- it is notable in this study that more microbiome-friendly antibiotics were used than what we usually prescribe (eg, they used lots of amoxacillin, then doxycycline, vs our microbiome-trashing use of azithro/clarithromycin, or broader-spectrum cephalosporins, or even fluorquinolones). And that the course of therapy was only 7 days. this shorter-term use of narrow-spectrum antbiotics might explain why there was no difference in adverse events between the 2 groups (certainly many other studies have shown increased C difficile colitis with the more broad-spectrum antibiotics)

--and, one of the most pressing global medical issues now is the emergence of drug-resistant bacteria, some of which are resistant to all of our current medications and are spreading worldwide (eg see http://gmodestmedblogs.blogspot.com/2014/05/who-report-on-antimicrobial-resistance.html and http://gmodestmedblogs.blogspot.com/2017/03/antibiotic-resistant-bacteria-of-concern.html), which also references many other blogs on such things as e coli superbugs, antibiotic-resistant sexually-transmitted diseases, longterm effects of antibiotics on the microbiome, overprescribing antibiotics, even the passing of drug resistant bugs from mothers to kids. And http://gmodestmedblogs.blogspot.com/2017/06/decreasing-antibiotic-resistance-by.html describes some effective antibiotic stewardship programs

--CRP point-of-care analyzers are available in the US

--study limitations include that there was no sham CRP testing in the usual care group (perhaps those patients in the CRP group perceived more intensive work-ups)

--another study found that evaluating procalcitonin in patients with acute respiratory illnesses (which included COPD exacerbations) led to decreased antibiotic prescribing, fewer antibiotic-related adverse events, decreased 30-day mortality, and fewer treatment failures (seehttp://gmodestmedblogs.blogspot.com/2017/10/procalcitonin-guided-therapy-for-acute.html )
-- an intriguing analysis argued that taking the full course of antibiotics as prescribed may lead to worse outcomes, especially in light of the currently evolving changes in the bugs causing infections to more being caused by commensal organisms: http://gmodestmedblogs.blogspot.com/2017/07/take-full-course-of-antibiotics.html

So, this study is important in that it found that using a simple point-of-care test for CRP led to a dramatic decrease in antibiotic prescribing, with no evidence of adverse consequences.  Many articles and blogs have warned against the emergence of antibiotic resistant bacteria world-wide, much related to overuse of antibiotics (and, as noted in more detail in prior blogs, the vast majority of antibiotics are being used in industrial animal farming, independent of animal infections). Antibiotic resistance is becoming an increasingly scary issue, moreso in lower income countries where there are much more limited public health and medical resources. And, also as per prior blogs, there is less interest in drug companies to make new antibiotics (with many rescinding prior governmental pledges to work hard on this), given the relatively poor financial return on meds that are used only occasionally and for short term (much better for them to have a new diabetes med which will be prescribed to huge and increasing numbers of people for decades....  eg see http://gmodestmedblogs.blogspot.com/2018/07/novartis-dumps-antibiotic-research.html)

geoff

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list