surgical antibiotic prophylaxis and adverse events

A recent large VA study looked at antibiotic duration after surgery and its association with surgical site infections, kidney injury, and C. difficile infections (see antibiotics surg prophy adverse events jamasurg2019 in dropbox, or doi:10.1001/jamasurg.2019.0569)

Details:

-- multicenter, national Veterans Affairs health care system cohort study of patients who had cardiac, orthopedic total joint replacement, colorectal, and vascular procedures, from 2008-2013

-- 79,058 patients were included, 96% men, mean age 65, 78% white/14% black, diabetes 27%, active smokers 29%

    -- 21,396 cardiac procedures

    -- 38,675 orthopedic procedures

    -- 10,810 colorectal procedures

    -- 8177 vascular procedures

--antibiotics:

    -- beta-lactam: 69%

    -- vancomycin: 25%

    -- aminoglycoside: 2%

    -- vancomycin plus aminoglycoside: 1%

-- median duration of surgical prophylaxis: 18.5 hours after skin closure

-- duration of anti-microbial prophylaxis was defined as <24 hours (in 73%), 24-48 hours (in 22%), 48-72 hours (4%), and >72 hours (1%)

-- Main outcomes: surgical site infection (SSI), acute kidney injury (AKI), and C. difficile infection

Results:

-- results were adjusted for type of surgery, age, sex, race, diabetes, smoking, American Society of Anesthesiologists score >2 (having at least severe systemic disease), MRSA, mupirocin, type of prophylaxis.

-- adjusting for a priori SSI risk factors: no decrease in SSIs overall if antimicrobials >24 hours

-- in those receiving <24 hours of antimicrobial prophylaxis, baseline risk for AKI was 14.7% and C. difficile was 0.8%

-- AKI for cardiac surgery: aOR, compared to <24 hours of antibiotic prophylaxis:

    -- 1.03 (0.95-1.12) for 24-48 hours

    -- 1.22 (1.08-1.39) for 48-72 hours

    -- 1.82 (1.54-2.16) for > 72 hours

-- AKI for non- cardiac surgery: aOR, compared to <24 hours of antibiotic prophylaxis:

    -- 1.31 (1.21-1.42) for 24-48 hours

    -- 1.72 (1.47-2.01) for 48-72 hours

    -- 1.79 (1.27-2.53) for > 72 hours

-- the overall unadjusted number needed to harm for AKI was 9 after 24-48 hours, 6 after 48-72 hours, and 4 after >72 hours

-- C. difficile infection: aOR, compared to <24 hours of antibiotic prophylaxis:

    -- 1.08 (0.89-1.31) for 24-48 hours

    -- 2.43 (1.80 3.27) for 48-72 hours

    -- 3.65 (2.40-5.55) for >72 hours

-- the overall unadjusted number needed to harm for C. difficile was 2000 for 24-48 hours, 90 for 48-72 hours, and 5 for >72 hours

-- of note, AKI was more common in those on vancomycin therapy; C. difficile in those on the combo of vancomycin plus aminoglycoside (though the latter did not reach statistical significance)

-- stratification by MRSA status: the magnitude of the association between vancomycin and postoperative AKI was similar whether that patients were MRSA positive or negative

-- subgroup analysis suggested that in cardiac patients, AKI risk increased with each additional day of antibiotic exposure and with receipt of combination vs single drug regimen: the absolute increase in AKI was 6%, vs an absolute reduction in SSI of 0.6%

    -- the number needed to harm for one episode of AKI was 18.7; for those who likely required hemodialysis, the number needed to treat was 116; and the number needed to prevent one SSI was 167

Commentary:

--a systematic review of RCTs found no difference between a single antibiotic dose within 2 hours before the incision vs multiple doses for either less than or more than 24 hours after surgery (see https://watermark.silverchair.com/ajhp0195.pdf?token=AQECAHi208BE49Ooan9kkhW_Ercy7Dm3ZL_9Cf3qfKAc485ysgAAAj4wggI6BgkqhkiG9w0BBwagggIrMIICJwIBADCCAiAGCSqGSIb3DQEHATAeBglghkgBZQMEAS4wEQQMqntJyd68O5fUM_wNAgEQgIIB8fT3IMfNAsj7B1D-nfz94Jdh8RFmC2h_wboIudcp1DoBOrYiZipNVPq_0_U6fGnwQoEw5mabaQXezEnPdcOOPEUbG8ZCY3tIiLCuDgKgehpvIURN1zD_tLhcA-9D0KjOCUyGcxEJvH_yn8cP9v72O_yAfrbCerM_Jt33-gbiM_vCCd68kmboXSuRpIFY1KCAcZueaiTXWXOl5qffTke2P9eV6_v5NFO4nFPqRRuIP2TTRAa7q44W7feINcRT_7EAOLebmrEPDM8Fi2LAPGbkrrep4qDeh022HR8MD0ZocunqNfxeQrKe9XBEcEZa11IY6C184SRdiQRt6UO9BklDaGMkfCAqirNswaVVpWBC9YE9SeiDrSZ7iw71ml6cjr4vVTfdy8w_4c8U2TMFlFeCyWs3SIsSW2YoSMR4A6BtvbEAxPkJw_yLA_1eb9B0-S5wiKv4iQJytbQwOhZpKJ2nrsyMbHIaPXUkmBXUvtKzPe1VMdyvA2dWJ89Rz_eyW0tYsgF807lzZurngWNOSaX8ReOZkfqOn1Z_8bw8wfdLgXmRlU7hPZ6ppGgU2L-FT5Ds_0dYj4hC_PSVVDXnpn-6_HsTDVUoI5TwgMuF_NFZisTJkjCAoC0P17GkGvMxkKYMyrChZAih2VMGPcXKWEyJ01sU )

-- Most current surgical prophylaxis guidelines recommend that antimicrobials be stopped within 24 hours after incision (multi-society guidelines permit durations of up to 48 hours following cardiac surgical procedures, though this is not based on any evidence demonstrating a benefit of additional doses following skin closure). One perspective on why this VA study is important is that it provides high-quality data on a large number of patients regarding real clinical outcomes associated with length of therapy for common surgical procedures. As noted in a recent blog, in cardiology guidelines about one half of the recommendations rely exclusively on expert opinion, and this number has not changed over the past 10 years: see http://gmodestmedblogs.blogspot.com/2019/04/guidelines-lacking-evidence-based.html . This VA study should help reinforce the evidence-basis of stronger recommendations to limit the duration of antibiotics.

-- This article follows a recent article/critique on urologists’ overprescribing antibiotics, per the American Urological Association guidelines of 2008: see http://gmodestmedblogs.blogspot.com/2019/01/antibiotic-overprescribing-by-urologists.html

-- A tangentially related article challenged the whole concept of taking the full course of antibiotics for many infections, noting a general shift in bacteria involved in infections (more commensal flora) and that longer courses of antibiotics tend to breed more antibiotic resistance: see http://gmodestmedblogs.blogspot.com/2017/07/take-full-course-of-antibiotics.html

-- And, of course, the context here: the overall overprescribing of antibiotics (the CDC estimates that more than 30% of prescriptions are unnecessary), antibiotics are among the most common causes of emergency room visits for drug-related adverse events, and the development of antimicrobial resistance through widespread use of antibiotics in patients, and even more so in farming (see prior blogs noted below for more details)

-- This VA study is a large data-mining study, and its conclusions are limited to associations and not definitive causality. They did try to control for patients who are more likely to be sicker by showing that there was no difference in those who were MRSA positive, a group more likely to have more medical comorbidities. But overall these patients were likely at increased risk of surgical complications, given that there was a higher rate of diabetes vs the general population (27% vs about 10%) and smoking (29% vs about 15%), which might have some bearing on the outcome. Also, as a VA study, one major limitation is the small number of female patients.

    -- the trend in groups of patients of increasing antibiotic duration and increasing adverse effects, with no evident effect on SSIs, is pretty powerful. But, they did not have any granular data as to why longer courses of antibiotics were prescribed on an individual basis. It would be useful to have randomized controlled trials of different subgroups of patients to see if longer courses of antibiotics were useful

-- Of note, in the cardiac surgery group, there seemed to be some overall benefit from the vancomycin containing prophylaxis regimens. The researchers raise the point that vancomycin might be targeted specifically for those who are MRSA positive, given the adverse overall effects of vancomycin based therapies (esp AKI)

for access to a pretty large number of relevant studies on antibiotic overprescribing, antibiotic resistance, and changes in the microbiome, see https://gmodestmedblogs.blogspot.com/2019/01/antibiotic-overprescribing-2-more.html 

so, this study does reinforce:

-- in general antibiotics are useful in most surgeries typically until skin closure, but <24 hours post-op.

-- for unclear reasons, cardiac surgery has not endorsed these goals. There is generally a higher rate of SSIs after cardiac surgery, but the guidelines are still inconsistent and suggest courses ranging from 24 to 96 hours despite lack of supportive data

-- this study adds to the large numbers of studies reinforcing minimizing antibiotic usage as much as possible, to avoid adverse effects, antibiotic-resistance, and protect the poor little microbiome….

-- There should be general systems in place, perhaps more annoying flags in the computer systems, to minimize antibiotic usage after appropriate surgical cases.

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