Complicated intra-abdominal infection

Personal study digest for a new ID fellow. The IDSA 2024 update covers ONLY risk assessment, diagnostic imaging, and microbiology — antimicrobial therapy and source control are forthcoming parts, so the therapy/dose/duration content here is drawn from our and Michigan CPGs (anchored on the 2010 IDSA/SIS guideline + STOP-IT). The "What's changed" section is reviewer synthesis of newer evidence, each claim cited. Not a substitute for the full guideline.

In one line

cIAI = infection that has breached the hollow viscus into the peritoneum (peritonitis ± abscess). Source control is the treatment; antibiotics are adjunctive — and after adequate source control, ~4 days of antibiotics is enough (STOP-IT).

◆When to suspect / how IDSA 2024 says to image

Complicated = extends beyond the viscus into a normally sterile space (peritonitis ± abscess). The label is about anatomy, not severity. Scope: appendicitis, cholecystitis, cholangitis, diverticulitis, abscess, secondary perforation, necrotizing pancreatitis.
Imaging branch point (new, GRADE-graded): CT first in non-pregnant adults for suspected appendicitis, diverticulitis, and intra-abdominal abscess. US first in children and in pregnant people — escalate to MRI (or CT) if equivocal and suspicion persists.
Biliary is the exception: suspected cholecystitis/cholangitis → US first even in non-pregnant adults; if equivocal → CT, then MRI/MRCP or HIDA.
Risk-stratify within 24 h: if you use a score, IDSA suggests APACHE II (WSES Sepsis Severity Score is an acceptable cIAI-specific alternative). No validated pediatric score.

◆Cultures the fellow drives (IDSA 2024)

Blood cultures only when they'll change management: obtain if elevated temperature AND hypotension and/or tachypnea and/or delirium, OR there's concern for a resistant organism (regional resistance, prior 90-day colonization/infection or antibiotic exposure, elderly/immunocompromised, healthcare-associated). Well-appearing without these → don't routinely draw them.
Intra-abdominal cultures whenever there's a source-control procedure in cIAI (moderate certainty — the strongest rec in Part 1). Fluid inoculation is the preferred collection method. Skip cultures for uncomplicated appendicitis getting routine appendectomy.

◆Empiric therapy & dose (local CPGs — IDSA therapy part pending)

Scenario	Regimen / dose / route	Notes
Community-acquired, mild–moderate	Ceftriaxone 1–2 g IV q24h + metronidazole 500 mg IV q8h (our institution); or cefuroxime 1.5 g IV q8h + metronidazole (Michigan)	Cover enteric GNR + anaerobes for distal small-bowel/appendiceal/colonic source
Severe sepsis/shock or MDR/HCA risk	Piperacillin-tazobactam 4.5 g IV q6h; or cefepime 2 g IV q8h + metronidazole 500 mg q8h; add vancomycin for severe sepsis / known MRSA	Pip-tazo 3.375 g q6h baseline, ↑ to 4.5 g if Pseudomonas suspected (our institution)
Severe β-lactam allergy	Aztreonam 2 g IV q8h + metronidazole 500 mg q8h ± vancomycin	Aztreonam OK unless prior reaction to ceftazidime/aztreonam
Oral step-down	Amox-clav 875 mg PO BID (or cefuroxime 500 mg BID) + metronidazole 500 mg BID; FQ-allergy path: cipro 750 mg BID + metronidazole	When tolerating PO and susceptibilities don't show resistance

Enterococcus: cover in healthcare-associated/postoperative or critically ill with risk factors (recent cephalosporin/quinolone, prosthetic valve, immunocompromise). Empiric VRE coverage only in liver-transplant, prior VRE IAI, or colonized septic patients.
Candida: cover for gastroduodenal/esophageal perforation, recurrent (tertiary) peritonitis, immunocompromised — treat if recovered; not empirically for community-acquired disease. Fluconazole 800 mg ×1 then 400 mg daily; micafungin 100 mg daily if septic/candidemic.

◆Duration & stopping

Secondary peritonitis / abscess with adequate source control: ~4 days (STOP-IT). Extend only if source control is inadequate or signs persist.
24 hours for gastroduodenal/proximal small-bowel perforation operated early, and for traumatic bowel injury repaired within 12–24 h.
Post-cholecystectomy for acute cholecystitis: stop within 24 h unless infection extends beyond the gallbladder wall. Cholangitis/biliary sepsis: 4–7 days after drainage.
Uncomplicated diverticulitis: 4–5 days if treated — or no antibiotics at all in selected immunocompetent patients (see below).
Gram-negative bacteremia of abdominal source: 7 days with source control and clinical response. Infected pancreatic necrosis: 14 days after source control (our institution).

◆Key decisions a fellow owns

Source control is the intervention — drain the abscess, divert/resect the perforation, decompress the obstructed biliary tree. The antibiotic clock starts at adequate source control; without it, duration is open-ended.
De-escalate to culture and step down to oral once the patient is improving and tolerating PO — route doesn't matter if the drug reaches the site.
Reserve carbapenems and novel agents for ESBL/AmpC/CRE/DTR-Pseudomonas risk; don't empiric-carbapenem every cIAI.
Don't over-test: no blood cultures in the well-appearing non-septic patient; no routine cultures in uncomplicated appendicitis.

What's changed since 2010 (last full IDSA cIAI guideline)

Reviewer synthesis of newer evidence — not the guideline. Each claim cited; flags where guidance lags practice.

IDSA's 2024 update is Part 1 of a multi-part series — risk, imaging, and microbiology ONLY. Antimicrobial therapy and source control are still forthcoming as of mid-2026, so the operative therapy references remain the 2010 IDSA/SIS guideline, the SIS 2024 update, and STOP-IT. Recency-sensitive — re-check whether the IDSA therapy part has published. Bonomo, CID 2024;79(S3):S81–7, doi:10.1093/cid/ciae346 · SIS update Huston, Surg Infect 2024;25:419–435 (PMID 38990709)
Source control + ~4 fixed days of antibiotics ≈ a longer course. STOP-IT: composite SSI/recurrent IAI/death 21.8% (4±1 d) vs 22.3% (longer, max 10 d), P=0.92. This is the backbone of every "4 days after source control" line in the local CPGs. Sawyer, NEJM 2015;372:1996–2005 (PMID 25992746)
Short course holds even in the critically ill. DURAPOP: ICU postoperative IAI, 8 vs 15 days — more antibiotic-free days (15 vs 12), equivalent 45-day mortality. Extends STOP-IT (which under-enrolled the sick) to the ICU. Montravers, Intensive Care Med 2018;44:300–310 (PMID 29484469)
Imaging is now algorithmic and modality-specific (CT-first adults; US-first children/pregnant) — the core of the 2024 update and a real change from the imaging-agnostic 2010 text. Bonomo, CID 2024;79(S3) (doi:10.1093/cid/ciae346)
Non-operative management of appendicitis is a legitimate option. CODA: antibiotics non-inferior to appendectomy on 30-day health status, but 29% had appendectomy by 90 days (41% with an appendicolith vs 25% without) and ~half by 4 years. Appendicolith predicts antibiotic failure. Flum/CODA, NEJM 2020;383:1907–1919 (PMID 33017106) · APPAC, Salminen JAMA 2015
Uncomplicated diverticulitis often needs NO antibiotics. AVOD and DIABOLO: omitting antibiotics in immunocompetent CT-confirmed uncomplicated disease did not increase complications, recurrence, or sigmoid resection (DIABOLO 24-mo: recurrence 15.4% vs 14.9%). Both local CPGs now allow observation. AVOD, Br J Surg 2012;99:532–9 · DIABOLO long-term, van Dijk Am J Gastroenterol 2018;113:1045–52 (PMID 29700480)
New β-lactam/β-lactamase-inhibitors cover resistant-GNB cIAI — ceftolozane-tazobactam, ceftazidime-avibactam, imipenem-relebactam (all + metronidazole) and eravacycline (monotherapy) are FDA-approved for cIAI and all post-date 2010 (meropenem-vaborbactam is cUTI-labeled, used off-label for resistant cIAI). Reserve for ESBL/AmpC/CRE/DTR-Pseudomonas; selection per IDSA's AMR guidance. IDSA AMR guidance, Tamma, CID 2024 (PMID 39108079)

◆Anki cards minted this run

IDSA 2024 imaging branch — CT-first adults vs US-first children/pregnant.
IDSA 2024 culture stewardship — when to draw blood cultures; intra-abdominal cultures at source control.
Uncomplicated diverticulitis — observation without antibiotics (AVOD/DIABOLO).
DURAPOP — 8 vs 15 days in critically ill postoperative IAI.
STOP-IT 4-day duration — already in deck (×2), held as duplicate. CODA appendicolith→surgery — already carded, held.

Sources: IDSA cIAI 2024 Part 1 (Bonomo, CID 2024;79(S3):S81–7, doi:10.1093/cid/ciae346; component manuscripts S88–126); Michigan Medicine Intra-abdominal Infections CPG (rev 9/2023); our institution Intra-abdominal Infections CPG (rev 5/2025); STOP-IT (Sawyer, NEJM 2015;372:1996–2005, PMID 25992746); DURAPOP (Montravers, Intensive Care Med 2018;44:300–310, PMID 29484469); SIS 2024 update (Huston, Surg Infect 2024;25:419–435, PMID 38990709); CODA (Flum, NEJM 2020;383:1907–1919, PMID 33017106); AVOD (Chabok, Br J Surg 2012;99:532–9); DIABOLO long-term (van Dijk, Am J Gastroenterol 2018;113:1045–52, PMID 29700480); IDSA AMR guidance (Tamma, CID 2024, PMID 39108079). PubMed cited for primary literature.