ID & CV II (April 2016)
Review: Singer et al. Third Intnl Consensus Definition Sepsis & Septic Shock. JAMA 2016
Research: Talan et al. TMP-SMX vs Placebo for Uncomplicated Skin Abscess. NEJM 2016
AIR: CV (II)
Review: Singer et al. Third Intnl Consensus Definition Sepsis & Septic Shock. JAMA 2016
Research: Talan et al. TMP-SMX vs Placebo for Uncomplicated Skin Abscess. NEJM 2016
AIR: CV (II)
13 comments
NEJM Bactrim for uncomplicated abscesses:
.
The outcomes of study definitely go against classic teaching that abx do not improve outcomes in uncomplicated abscesses.
.
In this study, most participants had a total lesion size, including associated erythema, of more than 5 cm, and many met other guideline criteria for antibiotic treatment already. So it’s really hard to tell if this changes our practice.
.
Secondary outcomes like recurrence at new site and infection in family members are really interesting and adds to the benefit of bactrim.
.
In terms of adverse events, some GI sx are not really the biggest issue. This study is likely not big enough to detect significant but rare adverse events like c.diff or SJS. That’s what I really want to know. My dad just recently got over his own c.diff. It’s everywhere! Not to mention the concern for abx resistance.
.
I think we’ll need more large RCT’s to set this issue to rest.
.
Discuss!
NEJM: Bactrim for uncomplicated abscesses
One strengh of this article is that it is appropriately powered. As noted in the article: previous studies looking at this question only considered pediatric populations and employed far smaller sample sizes.
One limitation of this study is that, as Alex mentioned above, its unclear if this changes our practice. Presently large abscesses usually get antibiotic coverage as well as drainage. I suppose the benefit here is that it encourages the use and that is now backed up by original research whereas before the addition of antibiotics for the treatment of an abscess was simply a recommendation by “experts within the field”
The outcomes of this research indicate that the addition of bactrim can be of benefit by reducing recurrence rates and complications. Serious complications from the use of Bactrim were not detected in this study and only minor GI upset was listed as an adverse effect.
At this point I don’t know that this article will do anything to change my practice. The benefit of having read this article lies in the fact that now I can point to some original research as rationale for prescribing antibiotics for larger abscesses with surrounding cellulitis rather than simply relying on consensus opinion.
JAMA Sepsis and Septic Shock
I don’t know that this alteration in definitions will affect our management in the ED of patients presenting with sepsis or with septic shock. I feel like the ED is good at identifying these patients and resuscitating them appropriately. This change will likely benefit patients on inpatient floors as they may be identified as decompensating more quickly by nursing staff.
I always had a suspicion that the severe sepsis label would go the way of the Dodo at some point. I’m glad that this committee have debunked its existence. Clearly you either have sepsis (by the old definition or the new one) or you have septic shock.
I was not familiar with the SOFA score previous to reading this article. One concern that I have with its employment as a risk stratification tool is that its not always feasible to obtain a PaO2 in all patients. I suppose the abbreviated SOFA score can be employed in that case.
Will it change my practice? I dont think that this shift will directly change my practice in the ED for the reasons listed above. I’m sure as the change in definition is more widely accepted and this information is disseminated we will be heading more about the SOFA scores.
JAMA sepsis definitions
.
I have always thought “severe sepsis” was redundant as well, as if “sepsis” itself was nothing to worry about. I’m glad that “severe sepsis” was removed.
.
The focus on organ dysfunction does make more sense, and appears to be based in evidence (10% increase in mortality). I could go with that. The SOFA score seems reasonable and not too cumbersome. I have had some prior sepsis research experience using it before. However, it’s more widely established in the critical care realm, not the ED, as significant lab data is required to make an assessment.
.
But the qSOFA of SBP 22, as a quick way to assess for sepsis appears easy enough in the ED. But aren’t these findings very concerning for a patient moving to septic shock? The clinician might have missed the boat already if a patient is having AMS and hypotension. At least, that’s what we have been led to believe with the use of SIRS criteria in the past.
.
As the article says, there’s no need to throw away SIRS just yet. Perhaps a combination of SIRS and qSOFA may help in clinical gestalt and then proceed to work up using SOFA if there’s any concern. That’s probably what I will do for now.
.
JAMA Sepsis Definitions
1. How has reading the article changed your practice? This article was published about 1 week before my MICU rotation and was a topic of discussion during my time in there. These new guidelines created some confusion among the attendings, fellows, and residents there. In terms of my own practice, this article will not change it significantly beyond simply knowing about the new qSOFA and having another tool to use if I have diagnostic uncertainty.
2. What information had you believed in previously that were debunked by reading this article? I thought that the SIRS criteria were sensitive but not specific for sepsis. But according to this article, 1/8 of patients admitted with sepsis do not have the minimum of 2 SIRS criteria.
3. What new information did you learn from reading this article? Patients with a SOFA score of 2 or more have a mortality risk that is greater than the mortality risk for STEMI.
4. What are current areas of uncertainty on this topic that can be potential areas for research? I think the SOFA score has a lot of potential limitations that should be studied. For example, how helpful are these new definitions in the ED setting? No emergency physicians were on the task force and the qSOFA is not well studied. How good is the qSOFA? While the SIRS criteria are flawed, they do seem to aid in early recognition and qSOFA may be less useful in that regard. In addition, the multiple lab values needed for calculation make the SOFA score cumbersome and less feasible in low resource settings.
HOT NEW SEPSIS
This article hasn’t changed my practice but it’s interesting to get an update in the area of sepsis. I had no idea it had been ~20 years since SIRS was introduced! I like qSOFA score as a quick “sepsis” screen and it’s easier to teach medical students and gauge your clinical gestalt on if a patient is sick or getting sicker (alterations in mental status or RR may prompt further w/u).
I agree with Andrew, I thought SIRS was a pretty good screen (not specific but fairly sensitive). 1/8 of patients admitted with sepsis do not have the minimum of 2 SIRS criteria. Crazy. The article says that SIRS is still a meaningful screening test which, from the ED perspective, is one of the most important aspects of care!
I had only peripherally heard of the SOFA score prior to this study, though this is only useful in the ICU. I also love lactate and per the article, “..lactate measurement offered no meaningful
change in the predictive validity beyond 2 or more qSOFA criteria in
the identification of patients likely to be septic..”
There will be tons of qSOFA studies in the ED +/- lactate changes. I would love to see an ED based comparison study between qSOFA and SIRS.
Final thought, which was mentioned previously and asked by Scott Weingart in his interview with the lead author, why weren’t any ED providers included on this paper!?!?
Link to podcast: http://emcrit.org/podcasts/sepsis-3/
* How has reading the article changed your practice?
This article won’t completely change my practice, but I think having the qSOFA in my arsenal will allow for quick screening criteria for sepsis, as well as a way to continuously reevaluate their clinical progression.
* What information had you believed in previously that were debunked by reading this article?
I actually had even less faith in the SIRS criteria than most of you prior to reading the article, especially in patients that are not able to mount a robust inflammatory response, and was surprised that it wasn’t more than 1/8 of patients that had sepsis that didn’t meet SIRS criteria.
* What new information did you learn from reading this article?
One big thing that I found interesting from this article was that lactate is often not meaningful in identifying patient that may be septic. What was also interesting, however, was that in their flowsheet, it was a requirement for diagnosis of septic shock. They talk about how lactate can be an identifier of “cryptic shock” in a patient that may not manifest hypotension, so I found it interesting that they require BOTH vasopressor dependence and elevated lactate to define septic shock. I think that will provide a good specificity but may lose some sensitivity.
* What are current areas of uncertainty on this topic that can be potential areas for research?
The question of the lactate is what I find really interesting and would like to see explored further. And, as everyone has already said, let’s get some ED providers in on this!
NEJM: Bactrim for abscesses
* What are some strengths of the study?
This study was powered strongly, and had really detailed and reproducible methods, so in this particular data set, I would say the results could be viewed as valid.
* What are the limitations?
One thing I found to be curious, was that providers could choose to not enroll patients they thought might not be good candidates, and a fair number of the patients were excluded due to being lost to follow-up. They do address this in the paper, but I think it would be interesting to look closely at what sort of effect this may have had. Also, their primary outcome seemed a little subjecive (see below).
* What are the main outcomes of the study?
The main outcome is clinical cure, which they defined as “abscence of clinical failure” at or before the time of the final visit. Clinical failure was just determined by the physician’s exam at the time, and they did not have real objective quantified criteria for this.
* Does reading this article change your practice? If so, how?
This article, while being interesting, does not change my practice. While this article may show that Bactrim was beneficial in this particular population, I’m not sure it would be generalizable to out patient population. I’d like to see more studies that demonstrate the same findings before I start prescribing antibiotics to all my abscesses and creating a whole mess of resistant strains.
1. This article did not really change my practice. I would still use SIRS as a screen for sepsis. I do like qSOFA but feel like the signs are already used inherently by me to decide if someone is sick as they add to my gestalt that the patient is doing poorly. It’s nice that these criteria were introduced in the paper to make me think about what signs and other clinical indicators I use personally to decide who is sick.
2. I did not know how that a significant number of patients who were fairly sick did not meet SIRS criteria. I always was taught that SIRS was sensitive.
3. New information I got was the significant mortality of someone with >2 SOFA score. I also liked that the experts are starting to reinforce how the definition of severe sepsis does not add much once suspicion for septic state is confirmed.
4. Uncertainty exists in the fact that for us ED providers it is difficult to integrate the SOFA score given its required lab results. it would be useful to see studies of outcomes in the ED and ICU of SIRS and SOFA criteria as well as more research done in emergency academia.
1. Strengths multiple sites and this was reportedly by this article the first to provide a large enough study as prior studies which showed no difference were small.
2. I give abs for signs of cellulitis, which most of these study participants already had which does not really change what I do. ‘
3. Primary outcome was clinical cure of the abscess at 7-14 days after abs course. The secondary outcomes were also interesting: less recurrent procedures, new infections, and family member infections.
4. I don’t like how the paper glosses over the fact the broadening the coverage of antibiotics leads to med interactions, resistance, allergic reaction etc which this study needs more participants to catch these events.
interesting results and I think the NIH is doing a similar study but I would stick which my current practice until more data are availalble.
Bactrim and Abcess
– Strengths: Large multicenter, double-blind, randomized trial. Per authors, it had to be a large study since cure rates with drainage only are already high.
– Limitations: A bit confused by the study – 95% of participants had a hx of MRSA and it was found in only 45% of participants. Could there have been a significant amount of strep cellulitis, for which Bactrim has not so good coverage? Additionally, there was some degree of non-adherence. Probably some operator bias as some abscesses may not have been drained as well as others (although to their credit, the authors conducted a standardized training). Lastly, some bias in not enrolling candidates who were perceived as being high risk.
– Outcomes: Using Bactrim for uncomplicated abscess had a higher cure rate vs placebo. It also was a/w lower rates of subsequent drainage procedures, lower rate of new and household infections.
– Changed practice? Prior to reading this I was pretty much a stickler with tx choices for uncomplicated cellulitis (Keflex), uncomplicated abscess (I/D), abscess with surrounding cellulitis (Bactrim). Check out the following for a good podcast on skin infections: http://foamcast.libsyn.com/episode-25-skin-and-skin-structure-infections. However, the study makes an interesting point for using Bactrim i/s/o of MRSA prevalent communities and how it may affect the noted secondary outcomes. However, I probably won’t be changing anything as I’d be more worried about abx resistance and the fact that you still get a pretty comparable cure rates with just I/D. There’s also the risk of side effects (ie Stevens Johnson) although adverse effects in the study were unremarkable (probably too small). Additionally, the authors used a dose of 320/1600 (ie, double dose) vs 160/800 which has been shown to be adequate in treating MRSA. Again, I don’t think I’ll be double dosing anytime soon without literature to back it up.
Sepsis Definitions
– How has reading the article changed your practice?
Don’t know how much it’ll change my practice at this moment as many of us are still stuck using the same language. It’s going to take a bit more awareness to start using the definition but I think in the long run it will prove helpful. The lack of specificity when it comes to the old definitions was always a bit annoying (“that guys is septic…he’s looks better than half the staff!). Now with this new definition, septic = SICK….and as a result, much more appropriate.
– What information had you believed in previously that were debunked by reading this article?
Old SIRS definition was worse than I thought. We all kind of knew that pretty much anyone walking into the ED would be SIRS or sepsis but I didn’t realize how bad it was. And that’s on both ends of the spectrum (ie, SIRS definition even missed up to 1/8th of ICU pts in one study).
– What new information did you learn from reading this article?
The definition now boils down to sepsis and septic shock (no more SIRS, severe sepsis). Sepsis is now suspected infection w qSOFA of 2/3 (AMS, hypoTN, tachypnea). Septic shock is now sepsis + vasopressors need + lactate > 2. It seems as if they authors did not like the ambiguity of previous definitions and were hoping to make ‘sepsis’ really mean something (ie, correlated w/ a mortality rate of 10%!).
– What are current areas of uncertainty on this topic that can be potential areas for research?
Applying qSOFA to ED setting. Important to note that ED providers were not included in the new definitions and therefore ACEP/SAEM did not endorse. Particularly concerning when 2/3rds of sepsis management is initiated in the ED. Additionally, CMS is not adapting to this definition after releasing a new algorithm…which potentially may force the hand of providers to use the CMS guidelines and NOT the definitions as stated in the paper. Another thing that would be interesting to look at is to apply the new definitions to seminal sepsis papers (ie, early gold directed therapy, etc.) and see how they match up.
Sepsis & SOFA
How has reading the article changed your practice?
I don’t believe the qSOFA score has changed my practice. I think SIRS was helpful with diagnosing an undifferentiated patient w/ lactate elevation, leukocytosis, etc. If a patient is altered and hypotensive, then your suspicion for sepsis is clearly high and the qSOFA score is rendered useless/obvious. Although SIRS is less specific for sepsis, it helps me consider a complicated infection in patients that don’t exactly fit the septic picture.
What information had you believed in previously that were debunked by reading this article?
Like Travis, I was surprised that 1/8 pts admitted to ICU, don’t meet 2+ SIRS criteria. I was also surprised by the lactate controversy. Although I knew it existed, I feel like it is commonplace to get a lactate in suspected sepsis at BMC and Lahey. In fact, the docs at Lahey told me they get “dinged” by the hospital if they don’t order a lactate in suspected sepsis.
What new information did you learn from reading this article?
I learned more about the SOFA score. I never learned about the SOFA score in the ICU. Everyone used the sepsis, severe sepsis, septic shock definitions last year.
What are current areas of uncertainty on this topic that can be potential areas for research?
Evaluating qSOFA criteria in the ED. Also I believe that the SIRS criteria are built into EPIC to highlight potentially sick patients. I don’t know that qSOFA will be as efficient since AMS is a red flag in itself.