Latest / Behind The Knife: The Surgery Podcast / Journal Review in Emergency General Surgery: STOP-IT - What a Decade of Data Has Taught Us
Transcript
- Emergency General Surgery Journal Review === [00:00:00] All right, guys, welcome back to Behind the Knife with the EGS team here in Tiger Country at the University of Missouri. I'm Rushab Dev, and I'm one of the acute care surgeons here at Mizzou. Today, we're gonna be tackling something every acute care surgeon manages almost every day, antibiotics for complicated intra-abdominal infection. And you know we have to start with the paper that everybody knows, even if some of us haven't read it since residency. That paper is STOP IT We all know the story. The patient has a perforated colon, then we operate it. We think we controlled the source. The patient's improving, but somebody has to ask, "How long are we gonna be treating?" The senior, of course, automatically answers, "Four days. Stop it." Everyone nods and the plan goes into the note, rounds move on. But Stop It was published in 2015, so today we must not just ask what Stop It showed. What we're actually asking is what a decade of additional evidence has taught us, and where that famous four-day answer begins, and where it begins to bend, I guess. Does sepsis change duration? What about obesity? What about diabetes, percutaneous drainage, bacteremia? Or is a patient [00:01:00] still intubated and on dopaminephrin? What do the updates from the SIS and IDSA actually say, and where does double AST's critical committee stand with all this? And also, in the era of ESBL, AMP-C, CRE, all the alphabet soup and difficult to treat pseudomonas, how should we think about the drugs? Do we just do a stop date or do we need to do more? Most importantly though, when the patient is not getting better, is the answer really more antibiotics or do we fail to control the source? We have an outstanding team here at Tiger Country to help us go through this topic. Dr. Skye Schultz, Dr. Kelly Wray, Dr. Desiree Fletcher, and Dr. Bryson Ratcliff. Team, welcome back. All right, Kelly, take us back to 2015. What exactly did Stop It test? So Stop It was a pragmatic, multicenter randomized trial of 518 patients with complicated intra-abdominal infection who had undergone what the treating team considered adequate source control. Patients were randomized to one of two strategies. The experimental group received fixed four-day course, technically four plus or minus one calendar day, [00:02:00] and the control group continued antibiotics until two days after resolution of fever, leukocytosis, and ileus, with a maximum duration of 10 days. In practice, antibiotic exposure was about four days versus eight days. Yeah. So what did the outcomes show and what should everybody know about those outcomes? The composite of surgical site infection, recurrent intra-abdominal infection, or death occurred in 21.8% of the short course group and 22.3% of the traditional duration group. The difference is essentially zero. Shorter therapy reduced antibiotic exposure without worsening outcomes. Yeah, so the headline is legitimate. After adequate source control, four days performed roughly like eight days. Ex- exactly, with three words doing a tremendous amount of work, adequate source control. Stop It did not include patients with an uncontrolled leak, dead bowel left behind, or an undrained abscess. It asked about the duration after the source was controlled. That's why I think surgeons sometimes quote the right number, but for the wrong reason. The trial was not saying antibiotics are unimportant. They're saying that antibiotics cannot compensate [00:03:00] for a source that remains in the abdomen. Yeah, so let's appraise, um, this whole thing like a journal review. What makes Stop It strong? Randomization, multiple centers, a clinically relevant strategy comparison, an outcomes-based assessment, and the protocol that was easy to implement. The trial directly challenged common behavior. Treating until every inflammatory abnormality resolved. Yeah, I think it was also practice-changing because it's simple. Guidelines can be multiple pages, multiple paragraphs, but this is very simple, very straightforward. Four days after source control. One sentence, one order set, one quick conversation on rounds. Yeah, so let's, uh, dive into the limitations. Sure. It was based on clinical judgment. Adequacy of source control was up to the clinicians rather than a universally measurable state. Uh, the population was heterogeneous, which helps with generalizability but complicates disease-specific inferences. The trial was meant to be generalized rather than to identify every individual component and their implications. Some of the patients we worry about [00:04:00] most, profound shock, severe immunocompromise, or unresolved source control, were not represented well enough to claim certainty. And the trial should not be changed into every sick abdominal patient gets four days. It applies when the diagnosis is complicated intra-abdominal infection and source control has been achieved. Critical illness, bacteremia, difficult pathogens, and uncertainty, lack of clarity about the source create additional questions that weren't answered by STOP-IT. Yeah. And there seems to be another number in STOP-IT that deserves more attention. Roughly one in five patients still experience an SSI, recurrent abscess, or death. Longer antibiotics did not prevent it. Exactly. So when a patient fails, your first thought should be anatomy. An antibiotic cannot drain an abscess, can't take out dead bowel, can't close an ongoing perforation, can't repair an anastomotic leak. Also, you have to think about things like pneumonia, line infection, C. diff, or the numerous other sources that can cause an [00:05:00] infection. Yeah, so that gives us our first clinical truth. Antibiotics treat susceptible organisms. Surgeons treat the source. Don't confuse the two. So, Desiree, STOP-IT was not the end of the story. Which patients did we initially worry might still need longer therapy? Septic ones. Uh, so specifically patients with obesity or diabetes, higher severity of illness scores, and then those patients treated with PERC drains rather than taken to the operating room for source control. These are the type of patients with bedside anxiety tends to extend the stop date of the antibiotics. Uh, so a post-hoc analysis of the STOP-IT data set, they were actually reassuring. Uh, 112 protocol-adherent patients who met the study's sepsis criteria, both short and longer courses produced similar rates of surgical site infections when looked at. Uh, recurrent intra-abdominal and extra-abdominal infections, hospital days, and mortality were also factors. The presence of systemic illness did not mandate longer treatment when source control was definitively obtained. Yeah, so one [00:06:00] important nuance, though, is that those were the study's older sepsis criteria, so, you know, back in, I think, Surviving Sepsis II 2016. Temperature and white count were not necessarily contemporary Surviving Sepsis 2026 organ dysfunction criteria, so it's just something that we should consider moving forward. Yeah, that's correct, and I think it's important that that distinction is actually said out loud. Another post analysis found that there was no signal that obesity, diabetes, or higher APACHE II scores actually required longer treatment. Patients whose source control was achieved with PERC drainage also did not benefit from a longer course of antibiotics. So the 2024 SIS guidelines translated those findings into recommendations. So a strong recommendation, no more than four days after adequate source control is a Grade 1A. Overall, uh, and re- recommends four days in septic patients, patient with obesity or diabetes, higher illness scores, and those with perk drainage supported by moderate quality evidence in those subgroups. Sweet. And those subgroup [00:07:00] papers were post-hoc, correct? Correct. Uh, they're clinically useful, consistent, uh, with the original trial, but they were not independently powered or randomized. So the direction of evidence is coherent. However, the power within each subgroup is lower. This matters because the subgroup did not show a difference is not the same as proving absolute equivalence for every diabetic patient in shock. We should not falsify certainty the study design did not provide. Yeah. So how well did this practice change after STOPIT? Not nearly as well as the fame of the trial would suggest. Population-based interrupted time series analysis of about 4,400 patients found that the median therapy for non-appendiceal, non-biliary complicated intra-abdominal infection remained about seven days and actually did not meaningfully fall or become clinically relevant after the STOPIT trial. These single center studies have similarly reported durations beyond the guideline targets. However, on the other hand, a 32-hospital stewardship intervention achieved high acceptance of [00:08:00] recommendations and decreased the proportion of courses exceeding seven days. This is also an implementation lesson. Evidence did not shorten a course by itself. You need source control documentation, a planned stop date, a pharmacist, and stewardship partnership, and a reassessment when the patient deviates from the expected course. Nice. That's a lot. Why don't we shift a little bit to Bryson? The newer literature also tightens the timing around source control. What do our listeners need to know about that? Yeah. So the 2024 Surgical Infection Society guideline recommends source control within 12 hours for lower-risk patients and within six hours for higher-risk patients with septic shock. These are both Grade 1B recommendations. The 2026 Surviving Sepsis Campaign similarly suggests early source control, ideally within six hours of the diagnosis of sepsis or septic shock when a source control procedure is required, although the certainty there is pretty low. Yeah. So the antibiotic order is not, you know, the completion of sepsis care, right? Exactly. Antibiotics begin treatment. Source control changes the trajectory. For a [00:09:00] perforation, ischemic segment, infected collection, obstructed biliary system, or uncontrolled anastomotic leak, every hour you spend debating the perfect drug can distract you from your definitive intervention. Source control is not synonymous with a procedure happened. The operation or drain has to accomplish the goal: remove the infection, stop the ongoing contamination, restore drainage when it's required, and create a path to recovery. Sweet. So- What makes you question whether source control was actually inadequate? Uh, there's a lot of things that should make you question this. Progressive organ dysfunction, persistent or recurrent shock, worsening abdominal findings, failure of lactate or inflammatory physiology, um, trending in the direction that you expect them to trend, deviation from the expected clinical course new bacteremia or a pattern that simply doesn't fit recovery. None of those automatically proves failure, but they should trigger a deliberate reassessment. Yeah, and the automatic reassessment shouldn't just be to broaden the antibiotics. [00:10:00] It's repeat examination, review the operative findings review the functionality of the drains, cultures, susceptibilities, and timely imaging when the patient can safely undergo it. Yeah, so the safe question is not how many more days. More safe question or the better thing to ask is what are we actually missing? So now let's make everybody uncomfortable. Not that you already aren't. It's post-op day four after a perforated colon. You believe the source is well controlled. The patient is still ventilated, hypothetically on dopa, and the white count is 18. Nobody in the ICU feels good about stopping therapy, right? Skye, how does this work? Yeah. Well, first, critically ill is not a microbiologic diagnosis. Ventilation, vasopressors, ileus, leukocytosis, renal failure, and delirium can persist because severe inflammation and organ injury outlast viable bacterial infection. So I would not extend antibiotics only because the patient still looks like an ICU patient. Second, I would reassess source control, culture data, [00:11:00] drug exposure, and reassessment of any alternative infections. Is there a residual collection? Is the drain functioning? Was there ischemic bowel? Is the organism susceptible to the drugs we have already been given? Has the antibiotic trough been at therapeutic levels, even in the setting of renal replacement, obesity, or a large volume of distribution? Maybe at that point it's time to talk with your ICU pharmacist. Yeah, but then where does the SIS eight-day recommendations fit in all this? Uh, the 2024 SIS guidelines suggest limiting therapy to eight days in critically ill patients after adequate source control. That is a grade 2B recommendation. As a weak recommendation based on moderate quality evidence, mainly an open label randomized trial and a meta-analysis totaling 489 critically ill patients with postoperative intra-abdominal infection. Shorter regimens, eight versus 15 days and five versus 10 days in the additional cohort, had similar mortality, lengths of stay, MDR [00:12:00] emergence, and reoperation, with more antibiotic free days in the shorter group. Yeah, so up to eight is not every ICU patient gets eight, correct? That's exactly right. Four days remains the strong default after adequate source control. Eight days is a ceiling to consider in select critically ill patients when clinical uncertainty remains despite reassessment. The recommendation does not justify 14 days because the white count is being stubborn. This is where recommendation strength matters. Four days after source control is grade 1A. Up to eight in the critically ill patient is grade 2B. Those statements shouldn't carry equal weight when discussing them on rounds. And modern ICU stewardship includes how we administer therapy. The 2026 Surviving Sepsis Campaign recommends a loading dose followed by prolonged beta-lactam infusion, such as in Zosyn, rather than conventional bolus maintenance in sepsis or septic shock, with a strong recommendation [00:13:00] therein. In selected patients, therapeutic drug monitoring may be considered where available. Better exposure is not the same as longer duration. Sweet. That's a great distinction Optimize the dose and take into consideration the patient's physiology that can have implications on pharmacokinetics and pharmacodynamics before just reflexively extending the days. Bryce, walk us through the major sources and their scopes. Yeah, so the 2024 Surgical Infection Society document is the most comprehensive adult intra-abdominal infection treatment guideline in this discussion. They used a systematic research from February 2024 and grade style ratings. It addresses drug selection, timing, duration, de-escalation, source control, specific pathogens, and stewardship. Whereas the 2024 IDSA complicated intra-abdominal infection update is different. It's a part one of a multi-phase update on the older joint SIS IDSA guideline. It focuses on risk assessment, imaging, and microbiological [00:14:00] evaluation. It's not the source for the four-day duration recommendation. It should not be cited as though it solved every treatment question. Yeah. So what's the most useful IDSA point for surgeons to take home then? So for adults and kids with complicated intra-abdominal infection who are undergoing a source control procedure, the IDSA suggests obtaining intra-abdominal fluid cultures to guide therapy. It's a conditioned recommendation with moderate certainty. Fluid inoculation is preferred for uncomplicated appendicitis undergoing appendectomy. Routine cultures are not recommended. Blood cultures are most useful when sepsis, hospitalization, or resistance risks makes the likely result to change management. The AAST Critical Care Committee consensus published in Trauma Surgery and Acute Care open journal is a pragmatic ICU document about fever and infection It reinforces short courses after source control and emphasizes that persistent illness should, should prompt investigation rather than indefinite therapy. But it's also a cl- clinical consistent document, not a new randomized trial or a dedicated systematic IAI [00:15:00] guideline. The 2026 Surviving Sepsis Campaign is even more broad. It covers sepsis and septic shock from all sources. For abdominal sepsis, its most relevant statements are early source control, MDR risk-based empiric coverage, prolonged beta-lactam infusion, de-escalation, shorter therapy after sh- adequate source control, and procalcitonin plus clinical evaluation when the optimal stop date is unclear. Man, so the evidence hierarchy is part of the content, right? A randomized trial, systemic guideline, a focused diagnostic update, and the expert consensus are not interchangeable. Kelly, let's shift gears. We have spent probably a, a large part of this episode on duration. Stewardship also means choosing the right drug. A stable community-acquired infection comes into the ED. Think about it. What's the framework when we approach that? Yeah. So you have to start off with the infection setting and the severity. Community-acquired lower risk disease generally needs enterococcal and anaerobic coverage. Higher risk or healthcare-associated disease requires broader assessment. Shock, prior [00:16:00] antibiotics, recent hospitalization, prior colonization or infection, uh, local resistance patterns, immunocompromise, end organ dysfunction, and the likelihood of enterococcus, MRSA, candida, or other gram-negative resistant organisms. Yeah, so the regimen isn't just the anatomy. It's anatomy plus ecology. I like that word. Exactly. A ceftriaxone plus metronidazole may be perfectly reasonable for a lower risk community-acquired infection in susceptible ecology. Zosyn or cefepime plus metronidazole may be appropriate in other settings. A carbapenem may be necessary for severe infection with credible ESBL risk, but the local antibiogram and an individual patient's prior cultures matter more than the national podcast naming one universal drug. And do not just add Flagyl to a drug that already provides reliable antibiotic coverage just because the add-on feels scary. Double antibiotic coverage usually just adds toxicity without adding meaningful activity. In that same vein, routine empiric MRSA, [00:17:00] enterococcal, or antifungal therapy is not appropriate for every intra-abdominal infection. Add those in the appropriate clinical setting and patient-specific risk factors justify them, but remove them when clinical data allows you to. Yeah, and to possibly be a dead horse about duration again, for the patient in shock, timely effective empiric therapy matters, but broad immediately and broad indefinitely are not the same. The initial regimen should be, uh, reassessed every day and as soon as cultures return. All right. Well, let me, let me change up things a bit. I'm gonna talk about the alphabet soup that's changed. ESBL, AmpC, KPC, NDM, CRE, and DTR Pseudomonas mean nothing to me. So what I'd rather us do is give the responsible surgeon's approach without turning this into a pharmacy board review. So first you wanna identify whether the concern is empiric or targeted. In the 2026 IDSA AMR guidance advises against spending newer agents on routine ESBL or susceptible AmpC infections when established options are [00:18:00] effective. For invasive ESBL infection outside the u- urinary tract, uh, carbapenem remains a preferred foundation Cefepime is preferred option for organisms at moderate risk of inducible AmpC when the isolate is susceptible or susceptible dose-dependent. Third gen cephalosporins and zosyn are generally not recommended for serious invasive AmpC infections. For the carbapenem-resistant Enterobacter, mechanism matters. So Klebsiella pneumonia carbapenemase producers may be treated with agents such as meropenem. For metallo-beta-lactamase-producing bugs, aztreonam-avibactam is now an important option with ceftazidime-avibactam plus aztreonam as another mechanism-based strategy. Kelly, keep going. Difficult-to-treat Pseudomonas is also susceptibility driven. The newer anti-pseudomonal beta-lactams are not interchangeable, and institutional susceptibility patterns matter. The moment to involve infectious disease and antimicrobial stewardship pharmacist is early. [00:19:00] And remember the SITE. New does not just mean universally broader or automatically better. Yeah, so the SAFE answer is mechanism, susceptibility, source, and stewardship, not just what the newest drug is. We want changes that provide some form of good management. Yep. So in complicated intra-abdominal infection undergoing a source control procedure, collecting intra-abdominal fluid properly matters. The IDSA specifically prefers fluid inoculation, um, avoiding a superficial swab. Blood cultures should be obtained when sepsis, shock, hospitalization, or resistance risk makes bacteremia clinically plausible and the result could al-alter your therapy. We shouldn't routinely culture straightforward uncomplicated appendicitis after an appendectomy. Then we use the result. De-escalation means stopping unnecessary agents, narrowing the spectrum, or converting combination therapy to an appropriate monotherapy. The 2024 Surgical Infection Society guideline suggests de-escalation in ICU patients with healthcare-associated intra-abdominal infection while maintaining anaerobic [00:20:00] coverage when the chosen narrowed regimen lacks it The 2026 SSC recommends de-escalation when a pathogen and susceptibility profile are available and also suggests de-escalation when final cultures are negative if the clinical course supports it. A culture that never changes therapy is not stewardship. It's documentation. The value comes from connecting microbiology to a stop, narrow, and redirect decision. Sweet. So Skye, let's shift gears a little bit. Procalcitonin, it keeps showing up. Yeah. So ADAPT-Sepsis randomized 2,760 critically ill adults with suspected sepsis across 41 UK ICUs to daily procalcitonin-guided advice, daily CRP-guided advice, or standard care. Procalcitonin guidance reduced antibiotic exposure by 0.88 days versus 9.8 and then respectively 10.7 days and met the study's pre-specified non-inferiority criteria for 28-day mortality. CRP [00:21:00] guidance did not reduce the total duration and its mortality comparison was inconclusive. Uh, sweet. That sounds useful but modest. Exactly. This was not an intra-abdominal infection-specific trial. The reduction was less than one day and the mortality non-inferiority margin was 5.4 percentage points. It supports procalcitonin as an adjunct when the optimal duration is genuinely unclear. It does not overturn a disease-specific four-day course after adequate control, um, ade-adequate source control specifically, and it does not rescue failed source control. The 2026 Surviving Sepsis Campaign mirrors that nuance. It suggests clinical evaluation alone rather than procalcitonin to decide whether to start antibiotics. But when the patient has sepsis or septic shock, source control is adequate and the optimal stop date remains unclear, it suggests procalcitonin plus clinical evaluation to support discontinuation. Conditional low certainty evidence. If [00:22:00] the CT shows an undrained abscess, I don't care what the procalcitonin says. The patient needs source control. So Bryson, let's go ahead and bring this all together. When you compare the era before STOP IT with where we are now, the biggest change is not simply that antibiotic courses became shorter. We changed how we interpret persistent physiology. For years, surgeons treated the fever, the white count and the ileus. STOP IT forced us to separate ongoing inflammation from ongoing infection. The literature since has largely reinforced that distinction. I see four major changes. First, source control is the center of microbiological strategy. Second, shorter therapy is the default after adequate source control. Third, the critically ill patient requires more thoughtful reassessment, not automatic prolonged therapy. And fourth, stewardship now includes resistance risks, cultures, PK/PD, de-escalation, and preserving new agents for the times that we really need them. The implementation gap remains real. We can quote STOP IT perfectly and still prescribe seven, nine, or 11 days [00:23:00] The practical solution is when to write the, to the stop date, document the source control event and force a reassessment when the expected clinical choice changes. The first antibiotic decision and the final antibiotic decision are not the same. Empiric therapy should be timely and appropriately broad for the patient's risk. Definitive therapy should become narrower, smarter and shorter as we accumulate more information. And in the ICU, optimize exposure. Use appropriate loading, prolonged infusion when indicated, renal and extracorporeal circuit adjustments and monitoring when available. Do not use extra days to compensate for a poorly dosed drug. Finally, name the document you are quoting. IDSA 2024 part one informs imaging and cultures. IDSA AMR guidance is mechanism specific. AASD provides pragmatic ICU consensus. Surviving Sepsis applies broader sepsis principles. Their scopes are certainly are different[00:24:00] All right, so where does that leave us here in Tiger Country and across acute care surgery? I think we can distill the decade after STOP IT into five clinical truths. Truth number one: source control starts the clock. Four days does not mean four days after admission or far after the first dose. It means approximately four days after source control. Truth number two: for most source control complicated IAIs, four days is enough. Sepsis alone, obesity, diabetes, higher physiologic risk, or percutaneous drainage does not automatically buy a longer course. Truth number three: critical illness changes the conversation, not the fundamental rule. Selected critically ill patients may be considered for up to eight days, but deterioration should trigger reassessment of source control, pathogen exposure, and alternative diagnoses.[00:25:00] Truth number four: stewardship means choosing smart, dosing correctly, culturing intelligently, and narrowing early. Truth number five: persistent illness should make you question source control before you question the STOP date. More antibiotics are not source control. So then how should acute care surgeons integrate the decade of literature since STOP IT into practice? Control the source. Start the clock. Treat most source-controlled patients for four days. Recognize when critical illness deserves additional consideration. Culture with purpose. Dose well. De-escalate aggressively. And when the patient is not getting better, don't just keep treating with calendar extensions of the antibiotic. Go find the problem. Hit the reset button. STOP IT has become the paper every acute care surgeon quotes because it gave us a wonderful, simple answer: four days. More than a decade later, though, the deeper lesson is that duration matters far less than whether we control the disease and [00:26:00] what we're treating. So with that being said, from Tiger Country at Mizzou, this is Behind the Knife acute care surgery team signing off. We want you to remember to control the source, respect the bugs, stop the antibiotics, and when the job is done, until next time, stay sharp and dominate the day