community-acquired pneumonia: doxycycline works well

 It is generally problematic finding studies comparing newer antibiotics to older antibiotics (or studies including pretty much any other types of older medications), given that studies are largely paid for by drug companies, and it could turn out that the old medications are as good as, or perhaps better than, the new ones the drug companies are trying to promote. An analysis, based on older studies, found that doxycycline seems to be as efficacious as macrolides or fluoroquinolones in treating mild-to-moderate community acquired pneumonia (see pneumonia CAP doxycycline helps CID2023 in dropbox, or doi.org/10.1093/cid/ciac615) 

 

Details: 

-- a systematic review/meta-analysis of 6 randomized controlled trials with 834 patients performed between 1984 and 2004 that compared doxycycline to 3 macrolides (roxithromycin, spiramycin, and erythromycin) and 3 fluoroquinolones (ofloxacillin, fleroxacin, and levofloxacin) 

   -- of note, the currently used macrolides of azithromycin and clarithromycin were not available at that time, and there are no (ie, zero) studies with direct comparisons of these meds with doxycycline

   -- the 6 RCTs were found after an initial search finding 1979 articles (ie, not a lot of relevant articles to assess….) 

        -- 5 studies were done in Europe and 1 in the US 

        -- 5 studies were multi-center, 5 were double-blinded 

        -- 2 trials enrolled only outpatients with mild community-acquired pneumonia (CAP), a total of 215 participants: 

            -- a Norwegian study reported in 1988, double-blind multi-center RCT with 191 outpatients with mild CAP, mean age 52, randomized to doxycycline (200mg on day 1, then 100mg/d for 8 more days) vs spiramycin 1 g po  TID on day 1, then 1 g po BID for another 4.5 days) 

            -- a German study reported in 1993, double-blind multi-center RCT with 24 outpatients with mild CAP, mean age 42, randomized to doxycycline 100mg po BID for 7-14 days vs erythromycin 400 mg po BID for 7-14 days 

            -- there was also a multi-country study reported in 1997 with a mix of outpatients and those on a general medical ward (no breakdown of how many were outpatients, or the criteria for hospitalization), with mild-to-moderate CAP: 414 patients median age 32, randomized to doxycycline 100mg po BID for 10 days vs fleroxacin 400mg daily for 10 days [note: this is a fluoroquinolone, not a macrolide]

                -- in this study, they did have results of the CAP microbiology in 66% of the patients: 133 were caused by mycoplasma/chlamydia/legionella; 30 were viral; 9 had pneumococcal or H flu infection; and 93 had mixed infections, some of which included S. pneumonia (see https://doi.org/10.1093/jac/39.4.499 ) 

 

-- primary outcome: clinical cure rate of doxycycline vs the comparator antibiotics noted above, defined as resolution of clinical signs and symptoms of continued pulmonary infection at the end of therapy, without new symptom onset, complications, or need for further antibiotic therapy 

-- secondary outcomes: all-cause mortality, length of hospital stay, cost of the antibiotics, microbiological eradication rate, and adverse events  

  

Results: 

-- clinical cure rate for CAP: 

    -- doxycycline: 87.1% (381 of 437 individuals) 

    -- comparator: 82.6% (328 or 397 individuals) 

        -- odds ratio 1.29 (0.73-2.28), p=0.39, not statistically significant 

    -- no studies reported a microbiological eradication rate 

 

-- secondary endpoints: 

    -- median time to respond to theearpy: 2.0 days for doxycycline, 3.0 days for comparator group, p=0.001 

    -- median length of hospitalization: 3.0 vs 5.0 days, p=0.04 

    -- median cost of hospitalization: $5126 vs $6528, p<0.001 

    -- median cost of antimicrobial agent: $33 vs $171, p<0.001 

 

-- subgroup analysis: 

    -- 4 of the 6 studies evaluated for risk of bias were found to be “high risk” 

    -- for the 2 studies with the lowest risk of bias: 

        -- clinical cure rates for CAP: 

            -- doxycycline: 87.1% (196 of 225 individuals) 

            -- comparator: 77.8% (165 of 212 individuals) 

                -- odds ratio 1.92 (1.15-3.21), p=0.01 

 

-- adverse event rates: 

    -- no mortality events were reported 

    -- overall adverse events rates were 82 reported in 297 patients (27.6%) on doxycycline versus 94 events in 283 patients (33.2%) on comparator meds, not statistically significant 

 

Commentary:

-- current guidelines for community-acquired pneumonia (CAP), including in the US, recommend a beta-lactam antibiotic in conjunction with a macrolide, with doxycycline as an "alternative"

-- there are ongoing concerns about macrolide antibiotics, given that there is increasing resistance internationally (and, per the guidelines, they should be avoided in areas with >25% pneumococcal resistance), and they may be associated with prolonged QTc, ventricular arrhythmias, and aortic aneurysms, as noted in this paper

-- there also are concerns about the adverse effects of macrolides on the microbiome, where the evidence suggests that doxycycline is a much more healthy alternative: https://gmodestmedblogs.blogspot.com/2024/01/doxycycline-has-fewer-c-diff-infections.html

-- doxycycline is a long-acting tetracycline with documented efficacy against many of the organisms associated with CAP, and has high bioavailability, high penetration of lung tissue, low cost and good safety profile

 

-- this systematic review/meta-analysis above basically found no difference in clinical cure rate or adverse event rate between doxycycline and the various comparator groups in those adults with non-severe CAP

-- one concern here, pointed up by the researchers, is that doxycycline remains very effective against H. influenzae, Moraxella catarrhalis, Mycoplasma pneumoniae, Chlamydia pneumophila, and Staphylococcus aureus; but, there is concern about pneumococcal resistance. The SENTRY trial including Europe, Asia, and Latin America (https://doi.org/10.1016/j.ijid.2018.10.004) reported in 2018 that 30.2% of  S. pneumoniae were resistant to tetracycline, though 32.4% were resistant to azithromycin, 23.4% were resistant to clindamycin, 18.3% were resistant to penicillin, and 6.7% were resistant to ceftriaxone (though, resistance rates tended to be in the low 20% range in Western Europe, and the 40 to 50% range in the Asian/Pacific region)

    -- i.e., tetracycline was at least as good as azithromycin; and the combination with ceftriaxone would seem to be reasonably effective

    -- the issue of microbial resistance to available antimicrobials is hugely concerning:

        -- the WHO reported very high levels of microbial resistance in 2014 and again in 2017, including some with resistance to all meds: https://gmodestmedblogs.blogspot.com/2017/03/antibiotic-resistant-bacteria-of-concern.html  (this blog has reference to many other studies)

        -- a more recent article was on antibiotic-resistant gonorrhea in the US, along with reference to many prior blogs on overuse of antibiotics in people and animals, more studies on the effects of antibiotics on the microbiome (including increase in colorectal adenomas and perhaps colon cancer), and the effect of hospital stewardship programs to limit the use of antibiotics): https://gmodestmedblogs.blogspot.com/2023/01/antibiotic-resistant-gonorrhea-in-us.html

-- a Canadian study (J Antimicrob Chemother 2018; 73 Suppl 7: vii5–vii11 doi:10.1093/jac/dky156) reported on antimicrobial susceptibility of invasive isolates of Streptococcus pneumoniae from 2011-2015, finding that sensitivity to antibiotics were:

    -- penicillin: in 2011, 97.3%, increasing to 99% in 2015

    -- ceftriaxone: 98.6% sensitive in 2011, increasing to 99.7% in 2015

    -- clarithromycin: 76.8% sensitive in 2011, decreasing to 74.9% in 2015

    -- levofloxacin: 99.6% sensitive in 2011, 99.7% in 2015

    -- doxycycline: 88.5% in 2011, increasing to 90.2% in 2015

-- an analysis of the SENTRY global surveillance program found an interesting outcome in terms of doxycycline resistance associated with decreased sensitivity to penicillin:

    -- isolates sensitive to penicillin: 93.6% sensitive to doxycylcine, 86.5% to erythromycin, 99.5% sensitive to ceftriaxone

    -- isolates with intermediate sensitivity to penicillin: 63.6% sensitive to doxycycline, 37.6% sensitive to erythromycin, 89.9% sensitive to ceftriaxone

    -- isolates resistant to penicillin: 34.6% sensitive to doxycycline, 8.5% sensitive to erythromycin, 5.8% sensitive to ceftriaxone

        -- they also tested the tetracycline derivative omadacycline (brand-name Nuzyra), which worked despite doxycycline resistance, with no cross-resistance to tetracycline or doxycycline

        -- this study would suggest that doxycycline is a useful anti-pneumococcal agent in those without penicillin resistance, raising a plausible approach of using doxycycline to treat those with CAP in areas where penicillin resistance is low. Also ceftriaxone does not work well with penicillin resistance

-- a small US study was done finding that hospitalized patients with CAP but not severely ill who were randomized to levofloxacin 500 mg IV daily (30 patients) vs doxycycline 100mg BID IV (35 patients), finding no significant difference in terms of antibiotic efficacy, but did find that the doxycycline group had shorter length of stays (4.0 vs 5.7 days) and lower antibiotic cost ($64.98 vs $122.07): see doi: 10.1111/j.1365-2710.2009.01073.x 

 

Limitations: 

-- the studies in the RCTs cite above were unfortunately old ones, given that no drug company is comparing their newer drugs to the generically available doxycycline. So no information directly relevant to azithromycin, for example, and there were only 3 studies using the older macrolides with a total of 353 patients. Also, the resistance patterns of the bacteria may have changed since these studies were done 

    -- the treatments performed are pretty different from our current approach: a single agent (doxycycline vs comparator) vs the current use of a b-lactam along with a macrolide, or the "alternate" of doxycycline 

    -- the study’s main outcome was efficacy of doxycycline against “comparators” for CAP, either macrolides or fluroquinolones, and combined them. But these are very different drugs with different actions on different bacteria and with likely very different evolution of antibiotic resistance globally; combining them in this study is questionable (of course, the combination does increase the numbers of patients assessed, which helps statistically, if not clinically, in getting more valid conclusions)

-- this study assessed clinical improvement (certainly a valid endpoint), though as a result we do not know the actual causative agents for the CAP: for example, macrolides are clearly preferable if the organism were some species legionella, and the causative microbes of CAP overall might be so different in different localities that comparisons of different antibiotics might well yield very different conclusions in different areas

    -- it was interesting that in the German study mentioned above, of 258 cases where the microbial agent was identified, only 9 had strep pneumonia and H influenza infections. 133 had “atypical” pneumonias

        -- which really brings up a limitation of this and many other antibiotic studies: the causative agents may well be different in different parts of the world, in different climates, in different times of the year, in different social conditions, and in different health care systems/access to health care. And the antibiotic of choice may well vary based on these issues 

    -- and this all may limit the generalizability of the results (as well as the applicability of guideline-recommended care)  

 

So, though there are no really good comparative trials of the currently prescribed antibiotics for CAP, this information, along with the information in the prior blog on microbiome changes (see https://gmodestmedblogs.blogspot.com/2024/01/doxycycline-has-fewer-c-diff-infections.html), would suggest a few things:

-- doxycycline seems to be a very useful antibiotic for CAP, and likely better than macrolides in terms of effectiveness (macrolides seem to have rapidly ascending resistance rates, as well as bad effects on the all-important microbiome)

-- so it seems that doxycycline should be given equal billing in the guidelines with the macrolides (except in the case of suspected Legionella, where macrolides are especially effective)

-- it seems that drug resistance patterns are changing pretty quickly, and these need to be monitored continually in different localities in different countries, with up-to-date information available to the clinicians/patients on the microbial composition of CAP and drug resistance patterns. that way the most effective therapy with the fewest serious adverse effects can be utilized

 

-- it really is unconscionable that drug companies are not required to perform direct drug comparisons between their newest drugs to the older ones. Indirect comparisons (eg showing that the new drug is better than historical data on older ones) is clearly insufficient: these studies were done in different times, on patients with different underlying conditions, at times of different drug resistance patterns, etc.

    -- it is certainly in the public interest (though perhaps not the drug company’s) to have the best medications (including assessment of potential adverse events such as the effects on the microbiome, as noted above) with the fewest serious adverse effects and the lowest cost. And, my feeling is that if the new meds were only marginally better, there should be a bias to using the older meds where there is a known track record vs the newer ones that may have unsuspected adverse effects that may be identified well after the initial studies were done

 

geoff

-----------------------------------

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

  

to get access to all of the blogs:

 

 go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

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

  -- 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

 

if you would like to see the articles in this blog, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list