Multisystem Processes & Disorders

Antibiotic stewardship: principles and inpatient applications

Clinical Overview and When to Suspect Inappropriate Antibiotic Use

— Mandated by Joint Commission and CMS Conditions of Participation for all US acute care hospitals since 2017

— Built on CDC Core Elements: leadership commitment, accountability, pharmacy expertise, action, tracking, reporting, education

— ~30% of inpatient antibiotic days are unnecessary or suboptimal

— Antibiotic exposure drives C. difficile, MDR gram-negatives (CRE, ESBL), VRE, candidemia, AKI, QT prolongation, and drug interactions

— CDC: >2.8 million resistant infections and ~35,000 deaths annually in the US

— Empiric vancomycin + piperacillin-tazobactam continued >48–72h without culture data or de-escalation

— Duplicate anaerobic coverage (e.g., metronidazole + pip-tazo)

— Treating asymptomatic bacteriuria outside pregnancy or pre-urologic procedure

— Treating positive respiratory cultures in a clinically improving patient (colonization vs infection)

— Fluoroquinolones for uncomplicated cystitis or acute bronchitis

— IV antibiotics when oral bioavailability is equivalent (fluoroquinolones, linezolid, metronidazole, fluconazole, TMP-SMX, doxycycline)

— Antibiotics continued past guideline-recommended duration (e.g., >5 days for uncomplicated CAP)

Antibiotic stewardship = coordinated interventions to optimize antimicrobial use, improve outcomes, reduce resistance, minimize toxicity, and lower cost

Why it matters on Step 3:

When to suspect inappropriate use (trigger stewardship review):

Board pearl: The two most impactful inpatient stewardship interventions are prospective audit with feedback and formulary restriction/preauthorization — both Class A evidence from IDSA/SHEA 2016 guidelines. Antibiotic "time-outs" at 48–72 hours are the bedside expression of audit-and-feedback.

Step 3 frames stewardship as a systems-based competency: the right answer often involves reassessing cultures, narrowing therapy, or switching IV→PO rather than escalating.

Presentation Patterns and Key History

— Day 3 of pip-tazo + vancomycin for "sepsis" — now afebrile, normotensive, WBC trending down, blood cultures negative at 48h → de-escalate or stop

— CAP treated with ceftriaxone + azithromycin, improving on day 3, tolerating PO → switch to oral, complete 5-day course

— Cellulitis on vancomycin, MSSA grows from blood → narrow to cefazolin or nafcillin (MSSA is better treated with a beta-lactam than vancomycin)

— Urine culture in asymptomatic catheterized patient grows E. coli → do not treat (asymptomatic bacteriuria)

— Sputum culture grows Pseudomonas in an improving COPD patient never intubated → likely colonization

— Duration of current therapy and clinical trajectory (fever curve, WBC, vasopressor needs, oxygenation)

— Microbiology data available at 48–72h: blood cultures, urine, respiratory, wound

— Allergy history — most "penicillin allergies" are not true IgE-mediated; mislabeled allergy drives broader, costlier, less effective therapy

— Prior antibiotic exposure in last 90 days (predicts resistance)

— Healthcare exposure: recent hospitalization, dialysis, SNF, indwelling devices

— Renal/hepatic function for dose optimization

— Drug interactions: warfarin, statins, QT-prolonging agents, serotonergic drugs (linezolid)

Stewardship "cases" on Step 3 are typically a hospitalized patient on broad empiric antibiotics where the vignette signals an opportunity to narrow, stop, or switch

Pattern recognition for the stem:

Key historical anchors stewardship questions hinge on:

Step 3 management: When a stem provides culture data at 48–72h with a clinically improving patient, the correct next step is almost always de-escalation, IV-to-PO conversion, or discontinuation — not "continue current regimen" and not "broaden coverage."

Key distinction: Colonization (organism present without inflammation/symptoms) vs infection (clinical syndrome + organism). Treating colonization is a stewardship failure and a frequent distractor answer.

Physical Exam Findings and Clinical Reassessment

— Defervescence sustained ≥24–48h

— Hemodynamic stability off vasopressors

— Resolving leukocytosis or bandemia

— Improving oxygenation, decreasing FiO₂/PEEP

— Resolution of localizing signs (erythema margin regressing, decreased purulence, improving mental status in meningitis)

— Tolerating enteral intake → candidate for IV-to-PO switch

— Persistent fever despite appropriate-spectrum therapy >72h

— Recurrent bacteremia with same organism

— New or expanding fluid collection, abscess, empyema

— Retained hardware, infected line, undrained obstruction (cholangitis, pyelonephritis with stone)

— CCS pearl: Persistent fever on appropriate antibiotics = look for a source to drain, a line to pull, or a stone/obstruction to relieve, not a broader antibiotic. Order imaging (CT, ultrasound, TEE), consult IR/surgery, remove central lines.

— New rash, eosinophilia, facial swelling → drug reaction (consider DRESS, SJS/TEN)

— New diarrhea → consider C. difficile testing

— Tendon pain (fluoroquinolones), hearing changes (aminoglycosides, vancomycin), red man syndrome (rapid vancomycin infusion)

— Confusion, serotonin syndrome features (linezolid + SSRI)

— Falling platelets (linezolid >14d, vancomycin)

— Afebrile ≥24h, hemodynamically stable, improving signs, tolerating PO, functional GI tract, no malabsorption

Stewardship depends on daily structured reassessment, not exam in the traditional diagnostic sense — but exam findings drive the decision to continue, narrow, or stop

Signs supporting clinical improvement (favor de-escalation/stop):

Signs suggesting inadequate source control (do NOT just escalate antibiotics):

Findings of antibiotic toxicity (trigger to stop or change):

Vital sign criteria for IV-to-PO conversion ("Hospital-in-Home" / step-down):

Board pearl: The bedside "antibiotic time-out" at 48–72 hours is the most testable stewardship intervention — review cultures, reassess diagnosis, narrow spectrum, set a stop date.

Diagnostic Workup — Cultures, Biomarkers, and Initial Data

— Two sets of blood cultures from separate sites before empiric therapy

— Urine culture with urinalysis (pyuria supports infection)

— Respiratory cultures: sputum Gram stain/culture, tracheal aspirate, or BAL in vented patients

— Wound/abscess cultures from deep tissue, not superficial swab

— MRSA nasal PCR: negative predictive value ~95–99% for MRSA pneumonia → allows vancomycin discontinuation in pneumonia

— Procalcitonin: low or falling levels in lower respiratory infections support shorter courses or stopping antibiotics; not validated for all infections (poor performance in endocarditis, abscess)

— Multiplex PCR panels (BioFire, Verigene) for blood, respiratory, GI, meningitis/encephalitis — speed pathogen ID and resistance gene detection (mecA, vanA, KPC)

— MALDI-TOF for rapid organism identification from positive blood cultures

— Two-step: GDH + toxin EIA, with PCR for discordant results — or NAAT alone in symptomatic patients

— Only test diarrheal stool (≥3 loose stools/24h); do not test asymptomatic patients or formed stool — testing colonized patients drives inappropriate treatment

— CBC with differential, CMP (renal/hepatic dosing), lactate, CRP

— Vancomycin AUC₂₄/MIC monitoring (preferred over trough alone per 2020 IDSA/ASHP) — target AUC 400–600 mg·h/L

— Aminoglycoside levels, β-lactam therapeutic drug monitoring in critically ill (emerging)

Cultures before antibiotics whenever feasible — but never delay antibiotics in septic shock beyond 1 hour

Rapid diagnostics (high-yield stewardship tools):

C. difficile testing:

Supportive labs:

Step 3 management: A vignette showing MRSA PCR negative + improving pneumonia = stop vancomycin. Procalcitonin <0.25 in LRTI with clinical improvement = stop or shorten antibiotics. These are textbook stewardship triggers.

Key distinction: A positive culture ≠ infection. Always correlate with the clinical picture — pyuria without symptoms in a catheterized patient is not a UTI.

Diagnostic Workup — Advanced and Source Identification Studies

— CT abdomen/pelvis with contrast: intra-abdominal abscess, perforation, cholangitis, complicated diverticulitis

— CT chest: empyema, necrotizing pneumonia, lung abscess

— MRI spine: vertebral osteomyelitis, epidural abscess, discitis (especially in S. aureus bacteremia with back pain)

— Ultrasound: cholecystitis, soft tissue abscess, DVT (mimicker of cellulitis)

— TEE: persistent S. aureus bacteremia, suspected endocarditis, prosthetic valve

— Mandatory in S. aureus bacteremia — document clearance every 48–72h

— Persistent bacteremia >72h on appropriate therapy → endocarditis, endovascular infection, retained hardware

— Once organism + susceptibilities return, narrow to the most targeted, narrowest-spectrum, least toxic agent

— MSSA bacteremia: cefazolin or nafcillin (NOT vancomycin — inferior outcomes)

— Enterococcus faecalis (susceptible): ampicillin over vancomycin

— ESBL E. coli: ertapenem or meropenem; piperacillin-tazobactam is inferior for ESBL bacteremia (MERINO trial)

— Streptococcus pneumoniae (PCN-susceptible): penicillin or amoxicillin

— Local hospital antibiogram guides empiric therapy

— Unit-level antibiograms (ICU vs ward) may differ substantially

— Detailed history; most rashes from childhood are not true allergies

— Penicillin skin testing or direct amoxicillin oral challenge for low-risk histories

— Removes a major driver of inappropriate broad-spectrum use

When empiric therapy fails or source is unclear, escalate diagnostics before broadening antibiotics

Imaging for source control:

Repeat blood cultures:

Susceptibility-driven narrowing:

Antibiogram review:

Beta-lactam allergy delabeling:

CCS pearl: In S. aureus bacteremia, the diagnostic bundle = repeat blood cultures q48h until clear, TTE (and TEE if high risk), ID consult, source identification, ≥14 days IV therapy from first negative culture. ID consultation independently improves mortality and is a stewardship best practice.

Risk Stratification and Empiric Therapy Logic

— Severity: septic shock → broadest reasonable empiric coverage; stable patient → narrower

— Site: skin/soft tissue vs intra-abdominal vs CNS vs bloodstream

— Healthcare exposure: hospitalization in last 90d, dialysis, SNF, recent antibiotics → cover MDR organisms

— Colonization history: known MRSA, ESBL, CRE, VRE colonization mandates coverage

— Immune status: neutropenia, transplant, biologics → broader and earlier

— Known MRSA colonization, recent IV antibiotics, prior MRSA infection, hemodialysis, severe illness with cavitary/necrotizing pneumonia

— Structural lung disease (bronchiectasis, severe COPD), recent broad-spectrum antibiotics, prolonged hospitalization, neutropenia, prior Pseudomonas

— Septic shock from gram-negative source: dual coverage only until susceptibilities known, then monotherapy

— No routine benefit to dual gram-negative coverage in non-shock patients

— Cultures negative or growing susceptible organism → narrow

— Alternative non-infectious diagnosis emerging → stop antibiotics

— Clinical improvement supports shorter course

Empiric therapy is selected by syndrome + host risk + local epidemiology, then narrowed when data return

Framework for empiric breadth:

MRSA risk factors (when to add vancomycin or linezolid):

Pseudomonas risk factors (when to add antipseudomonal β-lactam):

Avoid double coverage unless justified:

Empiric de-escalation triggers (the 48–72h time-out):

Board pearl: Shorter is better — current evidence supports 5 days for uncomplicated CAP, 7 days for HAP/VAP, 7 days for gram-negative bacteremia from a controlled source, 5–7 days for pyelonephritis (fluoroquinolone), 5 days for cellulitis without complications, 4 days for intra-abdominal infection after source control (STOP-IT trial).

Step 3 management: When the stem says "patient is clinically improved on day 3," the answer is usually complete a short course and stop, not "continue 14 days."

Pharmacotherapy — Optimizing Drug, Dose, Duration

— Narrow when possible: MSSA → cefazolin/nafcillin; strep → penicillin; E. coli (susceptible) → ceftriaxone or ampicillin

— Avoid fluoroquinolones for uncomplicated infections (FDA boxed warnings: tendinopathy, aortic dissection, dysglycemia, neuropathy, CNS effects)

— Time-dependent killing (β-lactams): extended or continuous infusion of piperacillin-tazobactam, cefepime, meropenem in critically ill improves time above MIC

— Concentration-dependent (aminoglycosides, fluoroquinolones): once-daily high-dose aminoglycoside dosing

— AUC-dependent (vancomycin): AUC₂₄ 400–600 target

— Adjust for renal function, obesity, augmented renal clearance (young, septic, hyperdynamic patients clear drugs faster)

— CAP: 5 days (afebrile 48h, ≤1 instability sign)

— HAP/VAP: 7 days

— Pyelonephritis: 5–7d fluoroquinolone, 7d β-lactam, 14d TMP-SMX

— Uncomplicated cystitis: nitrofurantoin 5d, TMP-SMX 3d, fosfomycin 1 dose

— Cellulitis: 5–6 days if improving

— Intra-abdominal infection (source controlled): 4 days (STOP-IT)

— Gram-negative bacteremia (source controlled): 7 days

— S. aureus bacteremia: ≥14 days IV from first negative culture (uncomplicated), 4–6 weeks (complicated/endocarditis)

— Fluoroquinolones, linezolid, metronidazole, fluconazole, TMP-SMX, doxycycline, clindamycin, rifampin

— Criteria: hemodynamically stable, afebrile, tolerating PO, functional gut, no deep-seated infection requiring IV

The 4 D's of stewardship: right Drug, right Dose, right Duration, De-escalation

Right drug — match spectrum to pathogen:

Right dose — pharmacokinetic/pharmacodynamic optimization:

Right duration — shortest effective course:

IV-to-PO conversion candidates with excellent oral bioavailability:

Board pearl: MSSA bacteremia treated with vancomycin has higher mortality than with cefazolin or nafcillin — narrowing isn't optional, it's life-saving stewardship.

Stewardship Interventions and Program Structure

— Leadership commitment: dedicated resources, salary support

— Accountability: physician leader (often ID)

— Pharmacy expertise: ID-trained pharmacist co-leader

— Action: implement interventions

— Tracking: monitor antibiotic use (days of therapy per 1000 patient-days) and outcomes (C. diff rates, resistance)

— Reporting: feedback to prescribers

— Education: clinicians and patients

— Prospective audit and feedback: ID/pharmacy reviews active antibiotic orders and recommends changes

— Formulary restriction and preauthorization: requires approval for restricted agents (carbapenems, linezolid, daptomycin, echinocandins)

— Both are more effective than passive education alone

— Antibiotic time-outs at 48–72h (bedside)

— Facility-specific guidelines for common syndromes

— Computerized decision support in EHR

— Allergy delabeling programs

— Rapid diagnostics with stewardship-team-driven response

— Procalcitonin protocols for LRTI duration

— IV-to-PO conversion protocols (often pharmacist-driven)

— Automatic stop dates on empiric orders

— Days of therapy (DOT) per 1000 patient-days — preferred over defined daily doses

— Standardized antimicrobial administration ratio (SAAR) — NHSN-reported, risk-adjusted

— C. difficile incidence, resistance trends, cost

— ~60% of outpatient antibiotic prescriptions for respiratory infections are unnecessary

— Discharge antibiotic prescriptions are a major transition-of-care risk — review duration, indication, and necessity at discharge

CDC Core Elements of Hospital Antibiotic Stewardship (memorize for Step 3 systems-based questions):

Core interventions (IDSA/SHEA 2016, Class A evidence):

Supplemental interventions:

Metrics for tracking:

Outpatient and transitions:

CCS pearl: When a Step 3 case asks about reducing hospital C. difficile rates, the highest-yield system intervention is a formal stewardship program with audit-and-feedback plus restriction of high-risk agents (fluoroquinolones, clindamycin, cephalosporins, carbapenems) — not just hand hygiene alone.

Special Populations — Elderly and Renal/Hepatic Impairment

— Higher baseline rates of asymptomatic bacteriuria (up to 50% in nursing home women) — do NOT treat unless symptomatic or pre-urologic procedure

— Atypical infection presentations (delirium, falls, anorexia) tempt overtreatment of incidental positive urine cultures — resist

— Polypharmacy increases interaction risk (warfarin + TMP-SMX, statins + macrolides/clarithromycin → rhabdomyolysis, QT-prolonging combinations)

— Higher C. difficile risk and worse outcomes

— Fluoroquinolones: increased risk of tendinopathy, aortic dissection, delirium, hypoglycemia (with sulfonylureas), QT prolongation

— Dose-adjust: vancomycin, aminoglycosides, β-lactams (most), fluconazole, TMP-SMX, daptomycin (less frequent dosing), levofloxacin

— Nitrofurantoin: avoid if CrCl <30 (formerly <60) — inadequate urinary concentration and toxicity risk

— TMP-SMX: hyperkalemia, AKI (pseudo-creatinine elevation via tubular secretion inhibition vs true AKI)

— Vancomycin + piperacillin-tazobactam: synergistic nephrotoxicity — use cefepime instead when feasible

— Aminoglycosides: nephro- and ototoxicity; avoid prolonged courses

— Contrast + nephrotoxic antibiotics: stagger or substitute

— Caution/dose-adjust: metronidazole, clindamycin, tigecycline, rifampin, isoniazid, antifungals (voriconazole, itraconazole)

— Monitor LFTs on prolonged therapy

— Vancomycin: loading dose then post-HD dosing guided by levels

— Cefazolin: post-HD dosing for MSSA bacteremia is well-established and preferred over vancomycin

— Avoid nitrofurantoin and most aminoglycosides

Elderly patients are stewardship-sensitive on multiple axes:

Renal impairment:

Hepatic impairment:

Hemodialysis:

Step 3 management: An 82-year-old nursing home resident with a positive urine culture, no urinary symptoms, and confusion attributable to another cause (dehydration, new medication) → do NOT treat the urine — pursue the actual cause of delirium. Treating asymptomatic bacteriuria is a classic stewardship trap.

Special Populations — Pregnancy, Pediatrics, Immunocompromised

— Screen at 12–16 weeks with urine culture

— Safe options: nitrofurantoin (avoid at term and in G6PD), amoxicillin, cephalexin, fosfomycin

— Avoid: fluoroquinolones (cartilage), tetracyclines (teeth, bone), TMP-SMX (first trimester — folate antagonism; third trimester — kernicterus), aminoglycosides (ototoxicity)

— Penicillins, cephalosporins, azithromycin, erythromycin (not estolate), clindamycin, metronidazole (after first trimester) generally acceptable

— Avoid fluoroquinolones as first-line (cartilage concerns) — reserved for specific indications

— Avoid tetracyclines <8 years (tooth discoloration); doxycycline short courses now considered acceptable for serious infections (RMSF, ehrlichiosis)

— Weight-based dosing critical; verify renal maturation in neonates

— Otitis media: watchful waiting in select children ≥2y with non-severe symptoms — a pediatric stewardship cornerstone

— Febrile neutropenia (ANC <500): empiric antipseudomonal β-lactam (cefepime, piperacillin-tazobactam, meropenem) within 1 hour

— Add vancomycin only for specific indications: hemodynamic instability, suspected catheter infection, severe mucositis, known MRSA colonization, skin/soft tissue infection

— De-escalate once stable and afebrile per IDSA guidelines

— Solid organ transplant, stem cell transplant: complex prophylaxis (PJP with TMP-SMX, CMV with valganciclovir, antifungal prophylaxis) — stewardship still applies to treatment courses

— Desensitize and give penicillin rather than substitute inferior agent — required for syphilis in pregnancy

Pregnancy is one of the few clear indications to treat asymptomatic bacteriuria (risk of pyelonephritis, preterm labor, low birth weight)

Pediatrics:

Immunocompromised:

Penicillin allergy in pregnancy with syphilis or GBS:

Board pearl: Pregnancy flips the asymptomatic bacteriuria rule — screen and treat. Outside pregnancy and pre-urologic instrumentation, asymptomatic bacteriuria should not be treated regardless of age, catheter, or diabetes status.

Complications and Adverse Outcomes of Antibiotic Misuse

— Highest-risk agents: clindamycin, fluoroquinolones, cephalosporins (especially 3rd/4th gen), carbapenems

— Treatment: fidaxomicin (preferred) or oral vancomycin for initial episode; fidaxomicin or vancomycin pulse/taper for recurrence; fecal microbiota transplant for multiple recurrences

— Severe/fulminant: oral vancomycin + IV metronidazole; surgical consult for toxic megacolon

— ESBLs, CRE, MDR Pseudomonas, MRSA, VRE — frequency rises with broad-spectrum exposure

— Patient-level and community-level harm

— Aminoglycosides: nephrotoxicity, ototoxicity (often irreversible), neuromuscular blockade

— Vancomycin: AKI (especially with pip-tazo), red man syndrome (infusion reaction — slow rate, antihistamines), DRESS, neutropenia

— Fluoroquinolones: tendon rupture (Achilles), aortic dissection/aneurysm, QT prolongation, peripheral neuropathy, CNS effects, hypoglycemia, C. diff, dysglycemia

— β-lactams: rash, anaphylaxis, interstitial nephritis, cytopenias, seizures (cefepime in renal failure, imipenem), encephalopathy

— Linezolid: thrombocytopenia (>14d), serotonin syndrome with SSRIs/MAOIs, peripheral and optic neuropathy, lactic acidosis

— Daptomycin: myopathy (monitor CK weekly), eosinophilic pneumonia; inactivated by surfactant — do not use for pneumonia

— TMP-SMX: hyperkalemia, hyponatremia, AKI, SJS/TEN, bone marrow suppression, hemolysis in G6PD

— Macrolides: QT prolongation, hepatotoxicity, drug interactions via CYP3A4 (especially clarithromycin)

— Anaphylaxis, DRESS, SJS/TEN, AGEP, serum sickness

C. difficile infection:

Antimicrobial resistance emergence:

Direct drug toxicities:

Allergic/hypersensitivity:

Microbiome disruption: candidiasis, opportunistic infections, long-term metabolic effects

Key distinction: Vancomycin red man syndrome = histamine-release infusion reaction (slow infusion, premedicate) — not an IgE allergy. True vancomycin anaphylaxis is rare. Mislabeling this as allergy drives inappropriate daptomycin/linezolid use.

Escalation — ICU Transfer, Consultation, and Triage

— S. aureus bacteremia (mandatory at most institutions — reduces mortality)

— Endocarditis, prosthetic device infection, osteomyelitis, CNS infection

— Persistent fever or bacteremia despite appropriate therapy

— Multidrug-resistant organisms requiring complex regimens

— Immunocompromised host with infection

— Penicillin allergy with limited alternatives (consider delabeling)

— OPAT (outpatient parenteral antimicrobial therapy) planning

— Dose optimization in renal replacement therapy, ECMO, augmented renal clearance

— Suspected drug interactions (linezolid + serotonergic, rifampin + multiple CYP substrates)

— TDM (vancomycin AUC, aminoglycosides, β-lactams in critically ill)

— Septic shock requiring vasopressors

— Respiratory failure requiring intubation

— Multi-organ dysfunction (lactate >4, AKI, encephalopathy, coagulopathy)

— Need for source control with hemodynamic instability

— Necrotizing fasciitis — emergent surgical debridement (antibiotics adjunctive only)

— Intra-abdominal abscess — percutaneous drainage

— Empyema — chest tube or VATS

— Infected hardware — removal

— Cholangitis — ERCP within 24–48h

— Obstructive pyelonephritis with stone — urgent decompression (stent or nephrostomy)

— Within 1 hour of recognition: lactate, blood cultures, broad-spectrum antibiotics, 30 mL/kg crystalloid for hypotension/lactate ≥4, vasopressors for MAP <65 after fluids

— Reassess and narrow within 48–72h based on culture data — stewardship integrates here

When to involve Infectious Diseases consultation (improves outcomes):

When to involve stewardship/ID pharmacy:

ICU escalation triggers:

Surgical/IR consultation for source control:

Sepsis bundle (Surviving Sepsis Campaign):

CCS pearl: In septic shock, administer broad-spectrum antibiotics within 1 hour even before cultures finalize if needed — every hour delay increases mortality. Stewardship is achieved by rapid de-escalation later, not by withholding initial empiric therapy. The two principles complement, not conflict.

Key Differentials — Infectious Mimics and Same-Category Pitfalls

— Stasis dermatitis (bilateral, chronic, no fever) — most commonly misdiagnosed as cellulitis

— DVT, superficial thrombophlebitis

— Contact dermatitis, lipodermatosclerosis

— Gout, erysipeloid, herpes zoster

— Necrotizing fasciitis (pain out of proportion, crepitus, systemic toxicity) — surgical emergency, not just antibiotics

— Heart failure (volume overload, BNP elevation, bilateral infiltrates)

— Pulmonary embolism

— Atelectasis, aspiration pneumonitis (chemical, not always infectious)

— Organizing pneumonia, eosinophilic pneumonia

— Lung cancer with post-obstructive pneumonia (treat infection but address mass)

— Asymptomatic bacteriuria + delirium from other cause (dehydration, drug, hypoxia)

— Vaginitis, urethritis (STIs)

— Prostatitis (requires longer courses, tissue-penetrating agents)

— Interstitial cystitis

— Inadequate source control (abscess, line, hardware, obstruction)

— Wrong organism (fungal, viral, atypical, mycobacterial)

— Wrong drug or inadequate dose

— Drug fever (β-lactams classic — typically 7–10 days into therapy)

— Endocarditis, endovascular focus

— VTE, hematoma

— Underlying inflammatory or malignant process

— Contaminant vs true pathogen — coag-negative staph from one of two bottles, no source, no device → likely contaminant

— Skin flora (Bacillus, diphtheroids, Propionibacterium) — usually contaminants unless prosthetic material

Many "treatment failures" are misdiagnoses — broadening antibiotics is wrong; reconsider the diagnosis

Cellulitis mimics:

Pneumonia mimics:

UTI mimics:

Persistent fever differential despite appropriate antibiotics:

Bacteremia same-category considerations:

Key distinction: Bilateral lower extremity erythema is almost never bilateral cellulitis — it's stasis dermatitis. Treating it as cellulitis is one of the most common antibiotic overuse scenarios on stewardship audits and on Step 3.

Key Differentials — Non-Infectious Causes of "Sepsis-Like" Presentations

— Venous thromboembolism — PE classically presents with low-grade fever, tachycardia, hypoxemia

— Drug fever and DRESS — eosinophilia, rash, organ involvement, 2–8 weeks after starting culprit (anticonvulsants, allopurinol, sulfonamides, vancomycin, β-lactams)

— Pancreatitis (sterile inflammation — antibiotics not indicated in mild/moderate acute pancreatitis even with fever and SIRS)

— Adrenal insufficiency — hypotension, hyponatremia, hyperkalemia, fever

— Thyroid storm — fever, tachycardia, agitation

— Neuroleptic malignant syndrome / serotonin syndrome / malignant hyperthermia

— Transfusion reactions — febrile non-hemolytic, hemolytic, TRALI

— Postoperative atelectasis (controversial as fever cause but classic differential)

— Hematoma resorption, recent surgery

— Acute MI, dissection, mesenteric ischemia — can present with leukocytosis and lactate

— Tumor fever (lymphoma, renal cell, hepatocellular)

— Autoimmune flare (lupus, vasculitis, Still's disease, gout/pseudogout)

— DKA, alcohol withdrawal — tachycardia, leukocytosis, mimics sepsis

— Hemophagocytic lymphohistiocytosis — fever, cytopenias, ferritin >10,000

— Eosinophilia (drug reaction, parasites, adrenal insufficiency)

— Markedly elevated ferritin (>10,000 — HLH, Still's)

— Procalcitonin low despite high CRP — favors non-bacterial inflammation

Stewardship excellence requires recognizing when fever and SIRS are not infection so antibiotics can be stopped

Non-infectious causes of fever/SIRS in hospitalized patients:

Lab clues against infection:

Step 3 management: When a patient has been on broad-spectrum antibiotics for 5 days without clinical improvement, cultures negative, no source on imaging — the answer is often to stop antibiotics and pursue alternative diagnoses, not to add more antibiotics. Antibiotic perseverance in the absence of evidence is a stewardship failure.

Transitions, Discharge Antibiotics, and Long-Term Plan

— Up to half of discharge antibiotic prescriptions have excess duration, wrong drug, or no indication

— Mandatory review at discharge: indication documented, duration specified, IV vs PO appropriate, dose adjusted for outpatient renal function

— Most patients meeting clinical stability criteria should leave on oral therapy

— OPAT reserved for infections requiring IV (endocarditis, osteomyelitis, certain bacteremias, deep abscesses with limited oral options)

— Requires ID involvement, weekly labs (CBC, BMP, drug levels), home health or infusion center, vascular access (PICC) plan

— Indications: endocarditis, vertebral osteomyelitis, prosthetic joint infection, deep abscess, MDR organisms

— Risks: line complications (DVT, infection), drug toxicity not caught between visits, missed clinical deterioration

— Asplenia: vaccinate (pneumococcal, meningococcal, Hib); consider daily prophylaxis in children and immediately post-splenectomy

— Recurrent UTI: behavioral measures, topical estrogen in postmenopausal women, postcoital or low-dose prophylaxis only if frequent

— Rheumatic heart disease: secondary prophylaxis per AHA

— Endocarditis prophylaxis: only high-risk cardiac conditions (prosthetic valve, prior IE, certain CHD, transplant valvulopathy) for dental procedures involving gingival manipulation — narrowed dramatically from older guidelines

— Surgical prophylaxis: single dose within 60 minutes of incision (120 min for vancomycin/fluoroquinolones), redose for long procedures or major blood loss, discontinue within 24 hours post-op (48h for cardiac historically, now also 24h)

Discharge is a high-risk transition for inappropriate antibiotic continuation

IV-to-PO conversion at discharge:

OPAT (Outpatient Parenteral Antimicrobial Therapy):

Antibiotic prophylaxis (long-term stewardship):

Board pearl: Surgical antibiotic prophylaxis beyond 24 hours post-op confers no benefit and increases C. difficile, resistance, and AKI. The Surgical Care Improvement Project (SCIP) measures this specifically.

Step 3 management: At discharge, always specify stop date rather than open-ended "continue antibiotics" — a CCS-style order line.

Follow-Up, Monitoring Parameters, and Counseling

— Clinical: temperature, vital signs, mental status, exam findings, oxygen requirement

— Labs: CBC, BMP, LFTs (frequency depends on agent); CK weekly on daptomycin; weekly CBC on linezolid >14d; vancomycin AUC monitoring

— Repeat cultures: blood cultures q48h in S. aureus bacteremia until clear; not routinely needed for most other infections

— Uncomplicated infections (CAP, cellulitis, UTI): PCP follow-up in 1–2 weeks, confirm resolution, address vaccinations

— Pneumonia: consider chest imaging at 6–12 weeks in smokers >50 to exclude underlying malignancy if not resolved

— S. aureus bacteremia, endocarditis, osteomyelitis: ID follow-up at 1–2 weeks post-discharge, weekly OPAT labs, end-of-therapy assessment

— C. difficile: monitor for recurrence (15–25% after first episode); do NOT retest stool to confirm cure (PCR/toxin can remain positive)

— Pneumococcal vaccines (PCV20 or PCV15+PPSV23) per ACIP for adults ≥65 and high-risk younger adults

— Influenza annually, COVID per current recommendations, RSV for ≥75 and high-risk 60–74

— Tdap, zoster (≥50), HPV as indicated

— Vaccination reduces infections and therefore antibiotic use

— Complete prescribed course as directed (but shorter evidence-based courses are increasingly standard)

— Do not save or share antibiotics

— Signs to return: persistent fever, worsening symptoms, new rash, severe diarrhea

— C. difficile precautions if recent antibiotics + diarrhea

— Probiotics: weak evidence; not routinely recommended for prevention of antibiotic-associated diarrhea or C. difficile in most patient populations

During therapy monitoring:

Post-discharge follow-up cadence:

Vaccination as stewardship:

Patient counseling:

CCS pearl: Order discharge follow-up appointment, medication reconciliation, and patient education as explicit orders in the CCS interface — these are scored systems-based competencies and reduce readmission for treatment failure or adverse drug events.

Ethical, Legal, and Patient Safety Considerations

— Joint Commission and CMS now require formal ASPs as a Condition of Participation

— Failure to implement is a regulatory and accreditation risk

— De-escalation or stopping antibiotics in an improving patient does not require explicit consent — it is standard care; explaining the rationale to patient/family builds trust and prevents pushback

— Patient or family demands for antibiotics ("just to be safe") for viral illness or asymptomatic bacteriuria require respectful refusal with explanation — clinician autonomy and beneficence outweigh autonomy-based demand for non-indicated therapy

— Beta-lactam allergy delabeling via skin test or oral challenge requires informed consent and supervised setting

— NHSN reporting of antimicrobial use (AU) and antimicrobial resistance (AR) modules is required for CMS reimbursement

— Reportable infections vary by state (e.g., CRE, certain MRSA bloodstream infections)

— C. difficile rates publicly reported and tied to hospital value-based purchasing

— Discharge with continued IV antibiotics without OPAT plan → line complications, missed monitoring

— Open-ended prescriptions ("continue antibiotics") without a stop date → overtreatment

— Failure to communicate de-escalation rationale to outpatient provider → restart of broad-spectrum at follow-up

— Medication reconciliation must capture all antibiotic changes

— Antibiotic costs vary widely; preferring narrow generic options aids adherence

— OPAT requires resources (home health, infusion access, transportation) — assess social determinants before discharge

— Adverse drug event from inappropriate antibiotic (C. difficile from unnecessary fluoroquinolone, AKI from vanco/pip-tazo) → ethical obligation to disclose, document, and report through institutional patient safety event reporting

— Apology and disclosure are linked to reduced litigation, not increased

Stewardship is a patient safety imperative, not merely cost containment

Informed consent edge cases:

Mandatory reporting and surveillance:

Transition-of-care risks (Step 3 favorite):

Equity and access considerations:

Disclosure of error:

Board pearl: A Step 3 stem describing a patient who developed C. difficile after unnecessary antibiotics for asymptomatic bacteriuria tests both stewardship principles and error disclosure — the correct response includes honest communication with the patient and a patient safety event report.

High-Yield Associations and Rapid-Fire Clinical Facts

— CAP 5d, HAP/VAP 7d, pyelonephritis 5–7d FQ, cystitis 1–5d, cellulitis 5d, intra-abdominal (source-controlled) 4d, GNR bacteremia 7d

CDC Core Elements: Leadership, Accountability, Pharmacy Expertise, Action, Tracking, Reporting, Education

Two highest-evidence interventions: prospective audit/feedback and formulary restriction/preauthorization

Antibiotic time-out: 48–72 hours — review cultures, narrow, set stop date

MSSA bacteremia: cefazolin or nafcillin > vancomycin (lower mortality)

ESBL bacteremia: carbapenem > piperacillin-tazobactam (MERINO trial)

Vancomycin + pip-tazo: synergistic nephrotoxicity → use cefepime when feasible

Daptomycin: do NOT use for pneumonia (inactivated by surfactant)

MRSA nasal PCR negative: stop empiric vancomycin in pneumonia

Procalcitonin: useful for stopping antibiotics in lower respiratory infections

Short-course durations:

Asymptomatic bacteriuria — treat ONLY in: pregnancy, pre-urologic procedure with mucosal bleeding

Surgical prophylaxis: single dose within 60 min of incision; stop within 24h post-op

Endocarditis prophylaxis: only high-risk cardiac + dental gingival manipulation

Highest C. difficile-risk antibiotics: clindamycin, fluoroquinolones, cephalosporins, carbapenems

C. difficile first-line: fidaxomicin (preferred) or oral vancomycin; do NOT retest stool to confirm cure

Highly oral-bioavailable: fluoroquinolones, linezolid, metronidazole, fluconazole, TMP-SMX, doxycycline

ID consult improves mortality in: S. aureus bacteremia, endocarditis, candidemia

STOP-IT trial: 4 days antibiotics after source control for intra-abdominal infection

Sepsis bundle: antibiotics within 1 hour of recognition

Penicillin "allergy" history: ~95% are not true allergies — delabel when possible

DOT/1000 patient-days: standard inpatient stewardship metric

SAAR: NHSN risk-adjusted comparator

CCS pearl: When stewardship-related questions appear, default decision-making is: culture before antibiotics, treat broadly when sick, narrow when data return, stop early when better.

Board Question Stem Patterns

Day 3 of pip-tazo/vanco for sepsis, blood cultures grow pan-susceptible E. coli, patient improved → Answer: narrow to ceftriaxone (or ampicillin if susceptible), discontinue vancomycin

80-year-old nursing home resident, mild confusion attributable to UTI vs other causes, urine culture positive, no urinary symptoms → Answer: do not treat; evaluate other causes of delirium

Bacteremia treated with vancomycin, cultures grow MSSA → Answer: switch to cefazolin or nafcillin

Patient with childhood rash from amoxicillin, severe infection needing β-lactam → Answer: allergy assessment, skin testing or graded challenge rather than alternative inferior agent

CAP day 3, afebrile, tolerating PO, improving → Answer: switch to oral and complete 5-day course

Post-op day 3, still on cefazolin for wound prophylaxis → Answer: discontinue (should have stopped within 24h)

Day 5 of appropriate antibiotics, still febrile, cultures clear → Answer: imaging for source/abscess, reassess diagnosis, consider drug fever — NOT broaden antibiotics

Hospitalized patient, recent fluoroquinolone, new diarrhea, positive testing → Answer: stop offending antibiotic if possible, start fidaxomicin or oral vancomycin

Healthy patient with mitral valve prolapse, no regurgitation, dental cleaning → Answer: no prophylaxis needed

"Most effective intervention to reduce hospital antibiotic use" → Answer: prospective audit with feedback and formulary restriction, not passive education

Patient ready for discharge, IV antibiotics being continued without clear indication → Answer: convert to oral and define stop date

Pattern 1 — De-escalation opportunity:

Pattern 2 — Asymptomatic bacteriuria:

Pattern 3 — MSSA bacteremia narrowing:

Pattern 4 — Penicillin allergy delabeling:

Pattern 5 — IV-to-PO conversion:

Pattern 6 — Surgical prophylaxis duration:

Pattern 7 — Persistent fever:

Pattern 8 — C. difficile management:

Pattern 9 — Endocarditis prophylaxis:

Pattern 10 — Stewardship system question:

Pattern 11 — Discharge transition:

Board pearl: When in doubt on Step 3 antibiotic questions, the conservative, narrower, shorter, oral, stewardship-aligned answer wins — escalation/broadening/longer duration is rarely correct unless the patient is clinically worsening or has objective new findings.

One-Line Recap

Antibiotic stewardship is the disciplined practice of giving the right drug, at the right dose, for the right duration, to the right patient — narrowing and stopping based on culture data and clinical trajectory, supported by structured hospital programs (audit-and-feedback, restriction, time-outs) that reduce resistance, C. difficile, toxicity, and cost while improving outcomes.

High-yield recap bullets:

Start broad when sick, narrow fast when data return, stop early when better — the operational mantra of inpatient stewardship; 48–72h time-outs operationalize it at the bedside

Shorter courses are usually correct: CAP 5d, HAP/VAP 7d, intra-abdominal (source controlled) 4d, gram-negative bacteremia 7d, cellulitis 5d, cystitis 1–5d — long courses are legacy, not evidence

Asymptomatic bacteriuria is treated ONLY in pregnancy and pre-urologic procedures; treating positive cultures without symptoms drives C. difficile, resistance, and AEs — and is a Step 3 trap in elderly patients with delirium

The two highest-evidence stewardship interventions are prospective audit-with-feedback and formulary restriction/preauthorization; both are required components of CMS- and Joint Commission-mandated hospital ASPs, supported by rapid diagnostics (MRSA PCR, procalcitonin, multiplex panels), allergy delabeling, IV-to-PO conversion, and ID consultation for high-risk syndromes (S. aureus bacteremia, endocarditis, candidemia) where ID involvement independently reduces mortality

CCS pearl: Always document indication, drug, dose, route, and stop date in your antibiotic orders — and reassess every order daily; stewardship excellence is built one order at a time.