I now have comprehensive source material from Harrison's 22e, Barash Clinical Anesthesia 9e, and Schwartz's Surgery 11e, plus recent PubMed evidence. Let me compile the complete ICU sepsis management protocol.
ICU Sepsis Management Protocol
Based on: Surviving Sepsis Campaign 2021 Guidelines | Harrison's Principles of Internal Medicine 22e (2025) | Barash Clinical Anesthesia 9e | Schwartz's Principles of Surgery 11e
1. Definitions (Sepsis-3)
| Term | Criteria |
|---|
| Sepsis | Suspected/confirmed infection + acute SOFA increase ≥2 from baseline |
| Septic Shock | Sepsis + vasopressor requirement to maintain MAP >65 mmHg + lactate >2 mmol/L despite adequate fluid resuscitation |
Rapid Screening — qSOFA (≥2 = high risk):
- Respiratory rate ≥22 breaths/min
- GCS <15
- Systolic BP ≤100 mmHg
qSOFA is more specific but less sensitive than SIRS; it does not replace SOFA for formal diagnosis.
2. Immediate Actions (First Hour — "Hour-1 Bundle")
| Priority | Action |
|---|
| Access | IV access (peripheral ± central venous catheter); arterial line if vasopressors required |
| Cultures | Blood cultures ×2 sets (aerobic + anaerobic) before antibiotics, without delaying therapy |
| Imaging | Rapid imaging to confirm and identify source of infection |
| Antibiotics | Empiric IV broad-spectrum antibiotics within 1 hour of septic shock recognition |
| Lactate | Measure serum lactate; repeat every 2–4 h to guide resuscitation |
| Fluid bolus | IV crystalloid 30 mL/kg within the first 3 hours |
| ICU admission | Target ICU admission within 6 hours of diagnosis |
3. Antimicrobial Therapy
Timing
- Septic shock: empiric antibiotics within 1 hour — each hour of delay is associated with ~7–8% increase in mortality
- Sepsis without shock: empiric antibiotics within 3 hours if no alternative diagnosis found
- Do not use procalcitonin to decide when to start antibiotics; procalcitonin may guide de-escalation
Empiric Selection (by scenario)
| Scenario | Regimen |
|---|
| Undifferentiated sepsis, low Pseudomonas risk | Ceftriaxone or cefotaxime (gram-negative coverage) ± vancomycin if MRSA risk |
| High Pseudomonas risk | Cefepime, piperacillin-tazobactam, or carbapenem (imipenem/meropenem) |
| Highly resistant gram-negatives (prior XDR/MDR) | Two-drug gram-negative coverage |
| Suspected fungal infection | Add antifungal (echinocandin preferred for Candida) |
| SSTI/skin source | Add coverage for Group A Strep and Staph (vancomycin if MRSA concern) |
De-escalation
- Reassess regimen daily upon availability of culture/sensitivity results
- Narrow spectrum as soon as pathogen identified
- Total duration: 7–10 days for most infections
- Stop antibiotics promptly if non-infectious etiology confirmed
Recent Evidence (2024): A JAMA meta-analysis (PMID: 38864162) found that prolonged/extended infusions of β-lactams are associated with improved clinical outcomes in sepsis vs. intermittent bolus dosing — consider for appropriate agents (e.g., meropenem, piperacillin-tazobactam extended infusion).
4. Source Control
- Identify anatomic infection source as rapidly as possible
- Implement drainage, debridement, or removal of infected device as soon as possible after initial resuscitation
- Remove all potentially infected intravascular devices and replace at new sites
5. Hemodynamic Resuscitation
Resuscitation Goals
| Parameter | Target |
|---|
| MAP | ≥65 mmHg |
| Urine output | ≥0.5 mL/kg/h |
| Lactate normalization | Serial measurement; target clearance ≥10–20% per hour |
| Mixed venous O₂ saturation | >65% |
Fluids
- First-line: Balanced crystalloids (e.g., lactated Ringer's, Plasmalyte) — preferred over normal saline to reduce hyperchloremic acidosis
- Volume: 30 mL/kg IV in first 3 hours; ongoing challenges guided by dynamic assessments (pulse pressure variation, stroke volume variation, fluid challenge response, point-of-care ultrasound — IVC collapsibility)
- Albumin: Consider as adjunct when large volumes of crystalloid are required
- Avoid: Hydroxyethyl starch (hetastarch) — associated with AKI and increased mortality; gelatin-based fluids not recommended
- After initial resuscitation, reassess fluid responsiveness before each bolus — avoid fluid overload
Vasopressors
| Drug | Role | Notes |
|---|
| Norepinephrine | First-line | Central administration preferred; titrate to MAP ≥65 mmHg |
| Vasopressin | Add-on when NE dose reaches 0.25–0.5 μg/kg/min | Fixed dose (0.03–0.04 units/min); do not use as monotherapy |
| Epinephrine | Second-line adjunct | Add if hypotension persists despite NE + vasopressin; risk of splanchnic hypoperfusion |
| Dobutamine | Cardiogenic component | Add to NE for low cardiac output states (persistent hypoperfusion + adequate filling) |
| Dopamine | Avoid in most cases | Only in selected patients with bradycardia and low arrhythmia risk |
| Phenylephrine | Not recommended | — |
| Levosimendan / Terlipressin | Not recommended | — |
Arterial line: Insert for all patients requiring vasopressors for continuous BP monitoring.
Recent Evidence (2025): A systematic review and meta-analysis (PMID: 39965613) supports adjunctive vasopressors (vasopressin, angiotensin II) alongside norepinephrine as reducing short-term mortality in septic shock.
6. Corticosteroids
- Indication: Refractory septic shock — hypotension persisting despite adequate volume resuscitation and vasopressor therapy
- Regimen: IV hydrocortisone 200 mg/day (as continuous infusion or 50 mg q6h) — dose <300 mg/day
- Do not use corticosteroids in sepsis patients who are not in shock
- Taper when vasopressors are weaned
A 2025 RCT (PMID: 40423970) found that hydrocortisone in septic shock may reduce the need for kidney replacement therapy.
7. Respiratory Support
Oxygen & Intubation
- Supplemental O₂ for SpO₂ <94%; escalate to high-flow nasal oxygen or non-invasive ventilation as needed
- Intubate for refractory hypoxemia, airway compromise, or hemodynamic instability precluding spontaneous breathing
Mechanical Ventilation (Sepsis-Induced ARDS)
| Parameter | Target |
|---|
| Tidal volume | 6 mL/kg predicted body weight (lung-protective) |
| Plateau pressure | ≤30 cmH₂O |
| PEEP | Titrate to avoid derecruitment; higher PEEP for moderate-severe ARDS |
| FiO₂ | Minimum to maintain SpO₂ 88–95% |
| Prone positioning | ≥12 hours/day for moderate-to-severe ARDS (PaO₂/FiO₂ <150) |
| NMB | Intermittent bolus dosing preferred over continuous infusion; use to facilitate proning in severe ARDS |
| Head of bed | Elevate 30–45° (semi-recumbent) unless contraindicated |
- Pulmonary artery catheter — not recommended for routine monitoring
- ECMO (VV): Consider in severe ARDS failing mechanical ventilation if team has expertise and resources
8. Other Organ Support & Preventive Measures
Transfusion
| Trigger | Threshold |
|---|
| RBC transfusion | Hgb <7.0 g/dL (restrictive strategy) — unless active ischemia, hemorrhage, or significant CAD |
| Platelets (prophylactic) | <10,000/μL without bleeding; <20,000/μL with bleeding risk; <50,000/μL for procedures or active bleeding |
| FFP | Do not give to correct INR without active bleeding |
Glucose Control
- Protocolized glycemic management
- Target blood glucose: 140–180 mg/dL (upper limit ≤180 mg/dL)
- Avoid hypoglycemia; check glucose every 1–2 hours if on insulin infusion
Renal Support
- Optimize fluid balance; use diuretics for fluid overload once hemodynamically stable
- Renal replacement therapy (RRT) for severe AKI with metabolic derangements
- Bicarbonate may be used to correct pH <7.2 in the setting of AKI
DVT & Stress Ulcer Prophylaxis
- DVT: Low-dose unfractionated heparin or LMWH unless contraindicated; mechanical prophylaxis if anticoagulants contraindicated
- Stress ulcer: PPI or H₂-blocker for patients with significant risk factors (mechanical ventilation, coagulopathy)
Sedation & Analgesia (PADIS Bundle)
- Use validated pain/sedation scoring (NRS/CPOT for pain; RASS for sedation)
- Target light sedation (RASS –1 to 0) unless clinical need for deeper sedation
- Analgesia-first approach (opioid-based); add sedatives only when needed
- Daily spontaneous awakening trials (SAT) + spontaneous breathing trials (SBT)
- Delirium screening: CAM-ICU or CPOT; minimize deliriogenic medications
9. Antimicrobial Stewardship & De-escalation
| Timepoint | Action |
|---|
| 24–48 h | Review culture results; de-escalate to narrowest effective regimen |
| 48–72 h | Reassess clinical response; consider procalcitonin trend for antibiotic duration guidance |
| Day 7–10 | Discontinue antibiotics for most infections; shorter courses acceptable for uncomplicated infections |
| Ongoing | Remove unnecessary catheters (central lines, urinary catheters) when no longer needed |
10. Goals of Care & Long-Term Considerations
- Discuss prognosis and goals of care with patient/family early; integrate palliative care principles
- Set advance directives and realistic expectations about likely outcomes
- Limitation of support decisions should be incorporated into daily care planning
Discharge Planning for Sepsis Survivors:
- 50% of initial survivors are rehospitalized within 1 year; 1 in 6 die within the first year
- Screen for economic and social support needs
- Establish follow-up to address Post-Intensive Care Syndrome (PICS): physical deconditioning, cognitive decline, depression, PTSD
- DVT and cardiovascular event risk remains elevated — counsel accordingly
11. Quick Reference Summary Card
HOUR 1:
□ Blood cultures ×2 (before abx)
□ Broad-spectrum IV antibiotics (within 1 h for shock; 3 h for sepsis)
□ Crystalloid 30 mL/kg IV
□ Measure lactate
□ Start norepinephrine if MAP <65 despite initial fluid
HEMODYNAMIC TARGETS:
□ MAP ≥65 mmHg
□ Urine output ≥0.5 mL/kg/h
□ Lactate clearance (serial monitoring)
VASOPRESSOR LADDER:
1. Norepinephrine (first-line)
2. Add vasopressin at 0.25–0.5 μg/kg/min NE
3. Add epinephrine if still refractory
4. Hydrocortisone 200 mg/day if vasopressor-refractory
VENTILATION (ARDS):
□ TV 6 mL/kg PBW, Pplat ≤30 cmH₂O
□ Prone ≥12 h/day for P/F <150
DAILY CHECKLIST:
□ De-escalate antibiotics (culture review)
□ SAT + SBT (ventilated patients)
□ DVT + stress ulcer prophylaxis
□ Glucose 140–180 mg/dL
□ Remove unnecessary catheters
□ Goals-of-care discussion
Key Sources:
- Harrison's Principles of Internal Medicine 22e (2025), Chapter 315
- Barash, Cullen & Stoelting's Clinical Anesthesia 9e, Table 57-4
- Schwartz's Principles of Surgery 11e, Chapter 6
- Evans L et al. Surviving Sepsis Campaign: International Guidelines 2021. Crit Care Med 49:e1063, 2021
- Abdul-Aziz MH et al. β-Lactam extended infusion in sepsis. JAMA 2024 (PMID: 38864162)
- Bauer SR et al. Adjunctive vasopressors in septic shock. Shock 2025 (PMID: 39965613)
- Donaldson LH et al. Hydrocortisone and RRT risk in septic shock. JAMA Netw Open 2025 (PMID: 40423970)