ICU Sepsis Management Protocol

Important: This protocol is a clinical reference tool. All management decisions must be individualized based on patient-specific factors, local resistance patterns, and institutional resources. The SSC 2026 guidelines contain 129 statements (46 new) from a 69-member international panel representing 23 countries.

1. Definitions (Sepsis-3)

2. Recognition and Risk Stratification

Screening Tools

• qSOFA (bedside, rapid): sensitivity limited, but high specificity for poor outcomes
• SOFA score: preferred for formal diagnosis - use at baseline and serially
• Other tools (NEWS, MEWS, AI-based TREWS): no single tool is preferentially endorsed by SSC 2026

Immediate Workup (within 1 hour of suspected sepsis)

• Cultures: Blood cultures x2 (aerobic + anaerobic) from separate sites before antibiotics; urine, sputum, wound cultures as indicated
• Labs: CBC with differential, CMP, coagulation panel (PT/INR, fibrinogen), procalcitonin (baseline), lactate (arterial or venous), LFTs, lipase (if abdominal source)
• Imaging: CXR immediately; CT chest/abdomen/pelvis if source unclear
• ABG: assess oxygenation and acid-base status
• ECG + troponin: if cardiac involvement suspected

Procalcitonin: do NOT use to decide whether to START antibiotics; it can guide STOPPING antibiotics (de-escalation).

3. The Hour-1 Bundle (SSC 2026)

1. Measure lactate - remeasure if initial lactate >2 mmol/L
2. Obtain blood cultures before administering antibiotics
3. Administer broad-spectrum antibiotics
4. Begin IV crystalloid (30 mL/kg for hypotension or lactate >=4 mmol/L)
5. Apply vasopressors if hypotension persists during or after resuscitation - target MAP >=65 mmHg

4. Antimicrobial Therapy

Timing

• Septic shock: Empiric antimicrobials within 1 hour of shock recognition. Each hour of delay in appropriate antibiotics increases mortality by ~7-8% (Harrison's 22E).
• Sepsis without shock: Initiate within 3 hours if clinical evaluation does not identify an alternative diagnosis.

Empiric Antibiotic Selection

Optimization

• Consider prolonged infusion of beta-lactams (extended or continuous infusion) to maximize pharmacodynamic target attainment - a 2024 JAMA meta-analysis (PMID 38864162) supports mortality benefit with prolonged vs. intermittent beta-lactam infusions in adults with sepsis.
• Consult pharmacy/ID for PK/PD optimization
• De-escalate based on culture results and clinical improvement - target the shortest effective duration

Source Control

• Identify and control source as rapidly as possible - target within 6-12 hours for most sources
• Includes: abscess drainage, infected device removal (IV catheters, urinary catheters, prosthetic material), debridement of necrotizing fasciitis, biliary decompression, bowel repair
• Remove indwelling catheters if catheter-related infection is suspected or no alternative source identified
• Source control drives antibiotic duration - post-source control courses can often be shortened

5. Hemodynamic Resuscitation

Fluid Therapy

Resuscitation Endpoints

• MAP >=65 mmHg
• Urine output >=0.5 mL/kg/hr
• Lactate clearance >=10% per 2 hours; target normalization <2 mmol/L
• Capillary refill time - SSC 2026 recommends using capillary refill time as an adjunct resuscitation guide
• CVP and ScvO2 are no longer primary targets (PROCESS/ARISE/ProMISe trials)

Vasopressor Therapy

Central venous access is preferred for vasopressor infusion, though peripheral vasopressors for limited durations are acceptable in resource-constrained settings.

6. Corticosteroids

• Do NOT use high-dose corticosteroids
• Consider adding fludrocortisone 50 mcg/day enteral (based on APROCCHSS trial data)
• Taper when vasopressors are no longer required
• Routine use in sepsis without shock is NOT recommended
• Evidence: decreases duration of shock but mortality benefit remains uncertain (Rosen's EM)
• Do NOT use etomidate for intubation in septic patients - it causes adrenal suppression

7. Respiratory Support and Mechanical Ventilation

Oxygen Therapy

• Target SpO2 92-96% (avoid hyperoxia)
• Deliver supplemental O2 via nasal cannula, high-flow nasal oxygen (HFNO), or non-invasive ventilation as appropriate
• HFNO preferred over conventional O2 for hypoxemic respiratory failure

Intubation

• Avoid etomidate (adrenal suppression in sepsis)
• Preferred agents: ketamine (hemodynamically stable), or reduced-dose propofol/midazolam with vasopressor support
• RSI with video laryngoscopy when feasible

Mechanical Ventilation (Sepsis-Induced ARDS)

• Tidal volume: 6 mL/kg ideal body weight (low stretch ventilation)
• Plateau pressure: <=30 cmH2O
• PEEP: Titrate to optimize oxygenation/compliance; use higher PEEP for moderate-severe ARDS
• FiO2/PEEP table: Use ARDSnet or similar protocol
• Prone positioning: >=12 hours/day for moderate-severe ARDS (PaO2/FiO2 <150); strong recommendation
• Neuromuscular blockade (NMB): Use to facilitate proning; intermittent bolus strategy preferred over continuous infusion
• Semi-recumbent positioning: HOB 30-45 degrees unless contraindicated
• Avoid: Routine pulmonary artery catheters in ALI/ARDS
• Rescue therapy for VVECMO criteria: Consider VV-ECMO for severe ARDS failing conventional MV if team has expertise and resources

Weaning

• Use standardized weaning protocols + sedation protocols
• Daily sedation interruption (SAT) if on continuous infusion sedation
• Target lightest effective sedation (RASS -1 to 0 if possible)
• Integrated ABCDEF bundle (Assess, Breathe, Coordinate, Delirium, Early mobility, Family)

8. Organ Support and Adjunctive Therapies

Renal

• Target urine output >=0.5 mL/kg/hr
• Avoid nephrotoxins (NSAIDs, contrast when possible, unnecessary aminoglycosides)
• Renal replacement therapy (CRRT or IHD): initiate for severe AKI with fluid overload, hyperkalemia, severe acidosis (pH <7.2), or uremia unresponsive to conservative management
• Bicarbonate: consider for arterial pH <7.2 in the setting of AKI

Hematologic / Transfusion

• RBC transfusion threshold: Hemoglobin 7 g/dL in most patients (7-9 g/dL target); higher threshold (8-9 g/dL) for active ischemia, coronary artery disease, or acute hemorrhage
• Platelets: Transfuse if <10,000/mcL (prophylactic); <50,000/mcL if bleeding or procedure planned
• FFP: Only for documented coagulopathy with active bleeding or before invasive procedures

Glycemic Control

• Target blood glucose 140-180 mg/dL (7.8-10 mmol/L) using insulin protocols
• Avoid hypoglycemia (glucose <70 mg/dL) - check BG q1-2h until stable, then q4h
• Very tight control (<110 mg/dL) increases hypoglycemia risk with no mortality benefit (NICE-SUGAR)

GI / Nutrition

• Early enteral nutrition within 24-48 hours if hemodynamically stable
• Do NOT initiate EN during active resuscitation or refractory shock
• Parenteral nutrition: use if EN not tolerated for >3-5 days
• Stress ulcer prophylaxis (PPI or H2 blocker) for patients on MV or with coagulopathy
• DVT prophylaxis (LMWH preferred; UFH if CrCl <30; mechanical compression if anticoagulation contraindicated)

Sedation and Analgesia (PADIS 2025 Update - PMID 39982143)

• Analgesia-first approach: treat pain before sedation
• Preferred analgesics: opioids (fentanyl, hydromorphone, morphine); consider adjuncts (ketamine, acetaminophen)
• Preferred sedative: propofol or dexmedetomidine (lighter sedation, lower delirium incidence vs. benzodiazepines)
• Target: lightest sedation that achieves goals; RASS -1 to 0 for most, deeper for ARDS/proning
• Assess for delirium daily (CAM-ICU or ICDSC)
• Early mobilization and family engagement

9. Monitoring and Targets Summary

10. Goals of Care

• Discuss prognosis and goals of care with patient and family early - initiate within 72 hours of ICU admission
• Integrate palliative care principles
• Identify code status and document clearly
• Consider ethics consultation for complex cases

11. ICU Admission Criteria

• Refractory hypotension or vasopressor requirement
• Respiratory failure or anticipated intubation
• Altered mental status
• Multi-organ dysfunction (SOFA >=2 organs)
• Lactate >=4 mmol/L

Key References

• Surviving Sepsis Campaign 2026 Adult Guidelines - SCCM/ESICM (129 recommendations, 46 new statements; available at sccm.org)
• Harrison's Principles of Internal Medicine 22E (2025), McGraw Hill - Initial Sepsis Management
• Barash, Cullen & Stoelting's Clinical Anesthesia 9e - Table 57-4 Management of Sepsis and Septic Shock
• Abdul-Aziz MH et al. Prolonged vs Intermittent Infusions of Beta-Lactams in Sepsis/Septic Shock. JAMA 2024 Aug 27. [PMID 38864162]
• Lewis K et al. PADIS Guidelines Update 2025. Crit Care Med 2025. [PMID 39982143]
• NICE-SUGAR Study Investigators. NEJM 2009 - Tight glucose control vs conventional in ICU
• PROCESS/ARISE/ProMISe Trials - EGDT reassessment

Note: The SSC 2026 guidelines are the most current major update (released 2026), replacing the 2021 guidelines. Key changes include: more nuanced antibiotic timing framework, updated vasopressor recommendations (selepressin now explicitly not recommended as NE alternative), explicit recommendation for active fluid removal after resuscitation, and capillary refill time as a resuscitation adjunct. Always verify against your institution's formulary and local antibiogram.