short course penicillin for strep?

A recent Swedish study found that a 5-day course of penicillin for strep pharyngotonsillitis is not inferior to a standard 10-day course (see strep shortcourse pcn bmj2019 in dropbox or doi.org/10.1136/bmj.l5337)

Details:

--433 patients > 6yo with pharyngotonsillitis from 17 primary healthcare centers in Sweden, from 2015-18

    --all had infection from Group A streptococci documented by a positive rapid antigen plus 3 or 4 Centor criteria (fever >38.5C, tender lymph nodes, coatings of the tonsils or inflamed tonsils in children, absence of cough)

--63% female; median age 30 (17% <12yo/8% 13-17yo; 75% >17yo); Centor criteria 3 in 50%/4 in 50%; fever >38.5C in 75%/tender nodes 92%/tonsillar coating 87%/absence of cough 95%; median duration of symptoms: 3 days; severity per patient assessment: 59% severe/38% moderate/4% mild; effect on patient overall: mild 31%, moderate 68%

--randomized to Pen V 800mg 4x/d for 5 days or 1000mg 3x/d for 10 days (a total penicillin difference of 14g higher in the 10-day course); for children up to 40kg, the dose was 250mg per dose in those 10-25kg, and 500 mg/dose in those 20-40kg

--primary outcome: clinical cure 5-7 days after end of treatment (non-inferiority margin prespecified as 10 percentage points); secondary outcomes: bacteriological eradication, time to relief of symptoms, frequency of relapses, complications and new tonsillitis, and adverse events

Results:

--clinical cure, per protocol:

    --5-day group: 89.6% (181/202)

    --10-day group: 93.3% (182/195)

        --therefore: non-inferiority of the 5-day course

--bacteriological eradication:

    --5-day group: 80.4% (156/194)

    --10-day group: 90.7% (165/182)

--comparing 5- vs 10-day courses:

    --relapses: 8 vs 7 patients (though 12 of these 15 patients had bacteriological eradication at the test-of-cure)

    --complications:  0 vs 4 patients (3 were peritonsilliltis and one psoriasis felt to be provoked by streptococci)

    --new tonsilitis: 6 vs 13 patients

    --time to relief of symptoms: no difference at 4 days

    --adverse events, patient-reported, comparing 5- vs 10-day groups: mainly diarrhea (26% vs 35%, mostly lasting 2 days), nausea (26 vs 32%), vulvovaginal symtoms (10 vs 16%), rash (6% vs 8%); all higher in the 10-day group. no serious adverse events occurred

--self-reported clinical cure, per patient diaries:

    --at 5 days: 86.6% in 5-day regimen vs 67.1% in the 10-day group

    --at 7 days: 94.6% in 5-day regimen vs 84.5% in the 10-day group

    --at 9 days: 89.7% in 5-day regimen vs 91.0% in the 10-day group (though there was more missing data as time went on)

--subgroup analyses: no significant difference by sex, age <18 vs >18, Centor score of 3 or 4, or any of the above secondary endpoints other than bacteriological eradication at test of cure (10.2 percentage point difference)

--adherence in both groups was very high, per patient diaries, but higher in the 5 day group

--63% of all patients would prefer a 5-day regimen with 4x/d dosing vs 10 days with 3x/d dosing

Commentary:

--Sweden has overall low use of antibiotics, though 11% are for sore throat

--Group A strep is the major identified pathogen, found in 33%

--the main reasons to treat are: to avoid serious complications (acute rheumatic fever/glomerulonephritis), speed resolution of symptoms, and to prevent the rare complications of peritonsillitis, impetigo, cellulitis, otitis media, and sinusitis.

--this study used either TID or QID penicillin dosing, noting that other studies have found similar medication adherence rates with TID vs QID, and that there is more consistent MIC (minimum inhibitory concentrations) with the 3-4x/day dosing vs less frequent dosing

    --though, based on studies a few years ago, i have been treating it with penicillin V 500 mg bid for 10 days. this has become a standard approach, though symptom resolution is typically within the first 5 days

--this study did find that symptom relief was slightly shorter with the 5-day regimen, though bacterial eradication was lower. other findings favoring the 5-day regimen included fewer new pharyngotonsillitis cases or complications. number of relapses 1 month later was essentially the same (and was not related to bacteriological persistence at the test-of-cure)

And, should we be treating strep throat anyway???

--the major argument to treat strep throat is to decrease the incidence of the "nonsuppurative" complications, specifically rheumatic heart disease and acute glomerulonephritis

--but the incidence of these complications has decreased dramatically in high-income countries

    --there is a strikingly higher incidence in more resource-poor countries, mostly in Oceania, sub-Saharan Africa, South Asia; and the 5 with the highest number of cases (India, China, Pakistan, Indonesia and Democratic Republic of the Congo) having 73% of global cases: see Watkins DA N Engl J Med. 2017; 377: 713-722.

    --but no clear explanation at to why. it is  not because of consistent efforts to decrease strep: many people do not come in for care (eg they think it is a URI and it does get better on its own, they do not have access to care because of distance/insurance/copays), and a major reservoir of strep is in asymptomatic carriers (see below)

    --for glomerulonephritis, see https://www.ncbi.nlm.nih.gov/books/NBK538255/

        --and it is not clear that treating strep throat affects the development of glomerulonephritis: see cochrane review: https://pure.bond.edu.au/ws/portalfiles/portal/33009394/fulltext.pdf

--treatment of strep throat basically leads to decrease in local symptoms, but only by 1-2 days (symptoms spontaneously resolve in 3-5 days, decreasing to 1-3 days with treatment). but, in the above study, the median duration of symptoms prior to taking penicillin was 3 days.  so, not treating would probably not have had much symptomatic benefit anticipated if symptoms only last 3-5 days without treatment

--and, as per a recent blog, penicillins may be one of the most microbiome-toxic meds, and may be associated wtih colorectal cancer (see https://gmodestmedblogs.blogspot.com/2019/10/antibiotics-increased-colon-cancer-risk.html )

--the other big issue with antibiotic overuse is microbial resistance to penicillin (eg: pneumococcal or gonorrhea being resistant to penicillin): eg see https://gmodestmedblogs.blogspot.com/2018/04/antibiotic-overuse-in-animals-and.html 

--one other difficult issue with strep throat is directly clinical. does the sore throat and positive rapid strep test mean that the strep is causing the pharyngitis??

    --strep infections happen concurrently at the time there are lots of other potential causes of pharyngitis: eg viruses

    --our accepted clinical correlations between "strep" symptoms and actually finding strep to be the culprit is not great: see https://gmodestmedblogs.blogspot.com/2015/02/pharyngitis-and-fusobacterium.html, which provides data on the poor predictive value of the Centor criteria (eg, about 1/2 of those with 3 Centor criteria do not have strep)

    --and, at the time pharyngitis-causing nonstrep microbes occur, there is a really large percentage of people (on the order of 25%, see https://www.ncbi.nlm.nih.gov/books/NBK374206/) who are carriers of strep in their pharynx (so will likely have a positive strep culture or rapid strep test) and they may well have Centor criteria in the 3-4 range (as per above), and in fact do not have strep causing their clinical symptoms....

    --so, we are undoubtedly treating way more "strep throats" than are real (one could do serologic testing to assess whether the strep is causative, though we mostly never do), our general clinical assessment is inaccurate, we are medicalizing patients and families, we may be unnecessarily disrupting our microbiomes (and ?? leading to more colon cancer), we may be creating more penicillin-resistant microbes, and ?? perhaps the risks outweigh the pretty marginal benefits in many people???

so, a few points:

--there are likely advantages to the shorter term of therapy: fewer adverse events, more likely increased medication adherence, probably less likely to get microbiome changes or bacterial resistance

--in high income countries where the incidence of acute rheumatic fever or glomerulonephritis is very rare, perhaps we should not have a knee-jerk antibiotic response to treating strep (defined typically as above: high Centor criteria and positive rapid strep, or even just high Centor criteria). Perhaps we should stratify patients for treatment: those who are really symptomatic and especially come in for care very soon (?1-2 days) after getting symptoms??? or maybe using a 5-day treatment with penicillin as above, since symptomatic relief was the same as the 10-day course. or maybe lower dose penicillin, such as 500mg bid for 5 days since symptom resolution usually happens by day 5 (though this all should be tested in an RCT, which i think should include no treatment, 5-day and 10-day treatments)????  we might then be able to significantly reduce penicillin use overall, since strep throat is such a common cause of penicillin prescribing.... 

--And, we need more data on different therapeutic approaches to kids with strep, since only 17% in the above study were less than 12yo, and this group is such a rich repository of strep pharyngitides.

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