---

ICU Management Protocol for Sepsis and Septic Shock

---

1. DEFINITIONS (Sepsis-3)

• SOFA (Sequential Organ Failure Assessment) - preferred for organ dysfunction quantification
• qSOFA (Respiratory rate >=22/min + GCS <15 + SBP <=100 mmHg): score >=2 signals high risk, but the SSC 2021/2026 guidelines recommend against using qSOFA alone as the primary screening tool due to low sensitivity
• National Early Warning Score (NEWS) and AI-based tools (e.g., TREWS) are alternatives; none is preferentially endorsed over others

---

2. INITIAL ASSESSMENT (Hour 0-1)

Workup (obtain before or simultaneously with antibiotics)

• Blood cultures x2 (aerobic + anaerobic) from separate sites - do NOT delay antibiotics >45 minutes to obtain cultures
• Serum lactate (repeat if initial >2 mmol/L)
• CBC, CMP, LFTs, coagulation (PT/INR, fibrinogen), procalcitonin
• ABG, chest X-ray, point-of-care ultrasound (cardiac, lung, IVC)
• Urinalysis + urine culture
• Additional cultures directed by suspected source (CSF, wound, sputum)
• Imaging (CT, ultrasound) to identify infection source - perform promptly

Hemodynamic monitoring

• Continuous pulse oximetry, cardiac monitoring
• Arterial line for continuous BP monitoring (preferred over NIBP in shock)
• Central venous access - not required before starting vasopressors; initiate peripherally first
• Dynamic measures of fluid responsiveness preferred over static CVP (passive leg raise, pulse pressure variation)

---

3. THE HOUR-1 BUNDLE

1. Measure lactate; remeasure if initial lactate >2 mmol/L
2. Obtain blood cultures before antibiotics
3. Administer broad-spectrum antibiotics
4. Begin IV crystalloid (30 mL/kg) for hypotension or lactate >=4 mmol/L
5. Apply vasopressors for MAP <65 mmHg that does not respond to initial fluids

---

4. ANTIMICROBIAL THERAPY

Timing

• Septic shock: Administer empiric antibiotics immediately, ideally within 1 hour of recognition. Each hour of delay increases mortality ~7-8%.
• Sepsis without shock: Initiate within 3 hours; if an alternative diagnosis is not identified within 3 hours of presentation, administer empiric antibiotics.

Empiric antibiotic selection by scenario

De-escalation

• Reassess antibiotic appropriateness as soon as culture results return
• Use procalcitonin to guide when to stop therapy (not when to start)
• Standard duration: 7-10 days for most infections; shorter courses adequate for many infections including uncomplicated bacteremia
• Procalcitonin-guided de-escalation reduces antibiotic exposure without increasing mortality

Beta-lactam infusion strategy

• Prolonged (extended or continuous) infusion of beta-lactams is recommended over intermittent bolus in adults with sepsis or septic shock (JAMA 2024 meta-analysis, PMID: 38864162), associated with improved clinical cure rates.

Source control

• Identify the infectious source promptly
• Implement source control as soon as feasible (drainage of abscess, removal of infected device, debridement, surgical intervention)
• Remove intravascular access devices that may be the source

---

5. FLUID RESUSCITATION

Initial phase (first 3 hours)

• 30 mL/kg IV balanced crystalloid (lactated Ringer's preferred over normal saline to reduce risk of hyperchloremic acidosis) for hypotension OR lactate >=4 mmol/L
• Reassess hemodynamics and perfusion after each 500-1000 mL bolus

Fluid type

• Balanced crystalloids (LR, PlasmaLyte) preferred over 0.9% NaCl
• Albumin (4-5%) can be considered when large volumes of crystalloid have already been given; not superior to crystalloid in initial resuscitation but may reduce total fluid load
• Avoid: Hetastarch and other synthetic colloids (HES) - associated with increased AKI and mortality

Resuscitation endpoints

Fluid management strategy - later phases

• SSC 2026 and CLOVERS/CLASSIC trial data: a conservative/restrictive fluid strategy (prioritize vasopressors early, limit IV fluids) is comparable or superior to a liberal strategy and avoids fluid overload
• Target net-zero or negative fluid balance once the patient is stabilized (de-resuscitation phase)
• Active fluid removal (diuresis or ultrafiltration) may be used in the de-resuscitation phase

---

6. VASOPRESSORS AND INOTROPES

Target

• MAP >=65 mmHg (may use 60-65 mmHg in patients >=65 years old per SSC 2026; higher targets do not improve survival)
• Can initiate via peripheral IV while awaiting central line - do not delay

Vasopressor ladder

• Dopamine: Avoid as first-line vasopressor; use only in highly selected patients (e.g., bradycardia with low cardiac output)
• Levosimendan: Not recommended
• Terlipressin: Not routinely recommended
• Pulmonary artery catheter: Not routine; consider in mixed septic + cardiogenic shock

---

7. CORTICOSTEROIDS

• Indication: Septic shock refractory to adequate fluid resuscitation and vasopressor therapy
• Drug: Hydrocortisone 200 mg/day IV (either 50 mg IV every 6 hours or 200 mg as continuous infusion)
• Evidence: Reduces duration of shock and vasopressor dependence; mortality benefit uncertain in non-refractory shock
• Do not use corticosteroids to prevent septic shock in patients with sepsis without shock
• Can add fludrocortisone 50 mcg oral/NG daily to hydrocortisone (based on APROCCHSS trial data)
• Taper when vasopressors are weaned

---

8. RESPIRATORY SUPPORT

Oxygenation targets

• SpO2 target: 90-96% (avoid hyperoxia)
• SpO2/FiO2 (SF ratio) <=315 suggests ARDS

Non-invasive support

• High-flow nasal cannula (HFNC): Preferred first step for sepsis-induced hypoxic respiratory failure; reduces need for intubation vs. conventional oxygen and standard NIV
• NIV (BiPAP/CPAP): Consider for mild-moderate ARDS (P/F ratio 200-300), especially in immunocompromised patients

Mechanical ventilation (invasive)

Moderate-severe ARDS (P/F <150)

• Prone positioning: >=12 consecutive hours/day - reduces mortality (PROSEVA trial); use neuromuscular blockade (NMBA) to facilitate prone; intermittent bolus NMBA preferred over continuous infusion when prone
• Neuromuscular blockade: Use to facilitate prone positioning; not routinely recommended for 48-hour continuous infusion
• Rescue therapies for refractory hypoxia:
  • Recruitment maneuvers (sustained inflation or incremental PEEP)
  • Inhaled nitric oxide or prostacyclin (bridge)
  • VV-ECMO: Suggest for severe ARDS (P/F <80) failing mechanical ventilation if team has experience and resources (EOLIA trial data)

Sedation and analgesia (PADIS guidelines, updated 2025)

• Analgesia-first approach (treat pain before sedation)
• Target light sedation (RASS -1 to 0) unless clinical indication for deeper sedation (refractory ICP, severe ARDS, NMBA use)
• Preferred sedatives: Propofol or dexmedetomidine over benzodiazepines in most mechanically ventilated ICU patients
• Daily sedation interruption (SAT) + spontaneous breathing trials (SBT)
• Delirium monitoring (CAM-ICU or ICDSC); avoid antipsychotics routinely; treat reversible causes

---

9. ORGAN-SPECIFIC MANAGEMENT

Acute Kidney Injury (AKI) - affects ~67% of septic patients

• Maintain adequate renal perfusion (MAP >=65)
• Avoid nephrotoxic agents (NSAIDs, aminoglycosides, contrast when avoidable, ACEi/ARBs)
• Renal replacement therapy (CRRT preferred in hemodynamically unstable patients) for:
  • Progressive AKI
  • Refractory hyperkalemia
  • Uremia (BUN >100 or uremic symptoms)
  • Volume overload not responsive to diuretics
  • Metabolic acidosis with pH <7.2 + AKI stage 2-3
• IV sodium bicarbonate: Use when pH <7.2 with AKI (stage 2-3) per BICAR-ICU data

Hematologic / DIC - affects ~35% with septic shock

• Restrictive transfusion strategy strongly recommended (SSC 2026); liberal transfusion does not improve outcomes

Hepatic dysfunction - affects ~50% with septic shock

• Avoid hepatotoxic agents
• Stress ulcer prophylaxis for high-risk patients (mechanical ventilation, coagulopathy)

Neurologic / Septic encephalopathy - affects ~54%

• EEG if seizures suspected; brain MRI/CT if focal deficits
• Minimize sedatives and deliriogenic agents
• Correct metabolic abnormalities (glucose, electrolytes, ammonia)
• Seizures treated with standard antiepileptics

---

10. METABOLIC AND NUTRITIONAL SUPPORT

Glucose control

• Initiate insulin infusion when blood glucose >=180 mg/dL (10 mmol/L)
• Target glucose 144-180 mg/dL (8-10 mmol/L) - tighter targets increase hypoglycemia risk without mortality benefit
• Point-of-care glucose monitoring every 1-2 hours during insulin infusion

Nutrition

• Enteral nutrition (EN): Start early (within 48-72 hours) once shock is controlled; EN preferred over parenteral nutrition
• Parenteral nutrition: Use if EN goals not met by day 7 or if EN contraindicated
• Avoid overfeeding (start at 15-20 kcal/kg/day, titrate to 25-30 kcal/kg/day by day 3-4)

Other adjuncts

• Vitamin C: No mortality benefit in sepsis; not routinely recommended
• Thiamine: Consider in patients with suspected deficiency or refractory vasopressor-dependent shock
• Bicarbonate: Use for severe metabolic acidosis (pH <7.2) only in the setting of AKI stage 2-3; no benefit in lactic acidosis from septic shock alone

---

11. VTE AND STRESS ULCER PROPHYLAXIS

• VTE prophylaxis: Unfractionated heparin or LMWH unless contraindicated (active bleeding, thrombocytopenia)
• Stress ulcer prophylaxis: Recommended for patients with mechanical ventilation >48 hours, coagulopathy, or prior GI ulceration; use PPI or H2 blocker; not indicated in all ICU patients

---

12. MONITORING AND REASSESSMENT

---

13. ICU ADMISSION AND TRANSITIONS OF CARE

• Target ICU admission within 6 hours of recognition (mortality increases with ED stays >6 hours)
• ABCDEF Bundle implementation (Assess/prevent/manage pain, Both SAT+SBT, Choice of sedation, Delirium monitoring, Early mobility, Family engagement)
• Upon transfer out of ICU: Use a structured handoff process and a critical care transition program to reduce post-ICU complications and readmission
• Goals of care discussions: Engage patient and family early; integrate palliative care principles as appropriate; routine formal palliative care consultation not recommended for all patients

---

14. QUICK REFERENCE SUMMARY

HOUR 0-1:
  ✔ Blood cultures x2
  ✔ Serum lactate
  ✔ Broad-spectrum IV antibiotics (within 1h for shock, 3h for sepsis)
  ✔ IV balanced crystalloid 30 mL/kg if hypotension or lactate ≥4
  ✔ Vasopressors if MAP <65 despite fluids → Norepinephrine first

ONGOING ICU:
  ✔ MAP target ≥65 mmHg (60-65 if age ≥65)
  ✔ Vasopressor ladder: NE → add vasopressin at NE ≥0.25 → add epi
  ✔ Hydrocortisone 200 mg/d if vasopressor-refractory shock
  ✔ Lung-protective ventilation: Vt 6 mL/kg PBW, Pplat ≤30
  ✔ Prone ≥12h/d for moderate-severe ARDS (P/F <150)
  ✔ Insulin at glucose ≥180; target 144-180 mg/dL
  ✔ Enteral nutrition within 48-72h
  ✔ Restrictive transfusion: Hgb <7.0 g/dL trigger
  ✔ Daily SAT + SBT; light sedation target
  ✔ VTE and stress ulcer prophylaxis
  ✔ Procalcitonin-guided de-escalation
  ✔ Source control as soon as feasible

---

Sources

• Harrison's Principles of Internal Medicine 22nd Edition (2025), Chapter on Sepsis - Sections: Initial Sepsis Management, Cardiovascular Dysfunction, Organ-Specific Complications
• Barash Clinical Anesthesia 9th Edition - Table 57-4: Management of Sepsis and Septic Shock
• Goldman-Cecil Medicine - Sections: Vasopressors, Circulatory Therapy, Corticosteroids
• Surviving Sepsis Campaign International Guidelines 2026 - Prescott H, Antonelli M, et al. Crit Care Med. 2026. doi:10.1097/CCM.0000000000007075 (129 statements, 46 new vs. 2021 version)
• Abdul-Aziz MH et al. Prolonged vs Intermittent Beta-Lactam Infusions in Sepsis. JAMA. 2024 Aug 27. [PMID: 38864162]
• Lewis K et al. Focused Update to PADIS Clinical Practice Guidelines. Crit Care Med. 2025. [PMID: 39982143]

Disclaimer: This protocol is an educational reference based on published guidelines. All clinical decisions must be individualized to the patient and reviewed by a qualified clinician. Local antimicrobial resistance patterns and institutional formularies should guide antibiotic selection.