| Letter | Element | Action |
|---|---|---|
| A | Assess/prevent/manage pain | CPOT or BPS q4h. Analgesia-first approach. |
| B | Both SAT + SBT | Daily sedation awakening trial + spontaneous breathing trial. |
| C | Choice of sedation | Dexmedetomidine or propofol. Avoid benzos. Target RASS 0 to −2. |
| D | Delirium assess/prevent/manage | CAM-ICU or ICDSC q8-12h. Non-pharmacologic prevention first. |
| E | Early mobilization | PT/OT within 48h of ICU admission when safe. |
| F | Family engagement | Open visitation, shared decision-making, emotional support. |
| Domain | Tool | Target |
|---|---|---|
| Pain | CPOT (Critical-Care Pain Observation Tool) or BPS | CPOT < 3, BPS < 5 |
| Sedation | RASS (Richmond Agitation-Sedation Scale) | RASS 0 to −2 (light sedation) |
| Delirium | CAM-ICU or ICDSC | CAM-ICU negative |
| Anxiety | No validated ICU-specific tool yet (2025 gap) | Assess clinically; consider nonpharmacologic interventions |
| Recommendation | 2025 Status | vs 2018 | Clinical Notes |
|---|---|---|---|
| Dexmedetomidine over propofol when light sedation or delirium reduction are priorities | Suggested Conditional, low certainty | CHANGED 2018: no preference between dex and propofol | Dexmedetomidine may reduce delirium incidence and improve time at target RASS. Higher risk of bradycardia and hypotension. NOT appropriate for deep sedation (RASS −4/−5) or patients needing NMBAs, dex cannot achieve that depth. |
| Propofol remains appropriate for rapid on/off sedation or deep sedation | Niche preserved Not displaced, just not preferred when light sedation is the goal | UNCHANGED Still first-line when deep sedation or fast neuro-check turnaround is needed | Faster wake-up for neuro checks and SAT/SBT. Monitor triglycerides q48h, hypertriglyceridemia is dose-related. Propofol infusion syndrome (PRIS) risk if > 80 mcg/kg/min for > 48h, watch for unexplained metabolic acidosis, rhabdo, bradycardia. |
| Avoid benzodiazepines for routine ICU sedation | Avoid (routine) Conditional | UNCHANGED 2018 already against routine use | Associated with increased delirium (OR ~2–3x), longer MV duration, longer ICU stay, worse cognitive outcomes. Reserve for alcohol withdrawal (CIWA-guided), seizures, or benzodiazepine-dependent patients. |
| Analgesia-first approach | Recommended Conditional | UNCHANGED 2018 core principle, carried forward | Treat pain before escalating sedation. Fentanyl infusion as first-line. Many patients labeled "agitated" are actually in untreated pain, sedation alone treats the wrong problem. |
| Clinical Scenario | Preferred Agent | Rationale |
|---|---|---|
| Light sedation target (RASS 0 to −2) | Dexmedetomidine | Less delirium, preserves respiratory drive, cooperative sedation |
| Deep sedation needed (NMBAs, proning) | Propofol ± fentanyl | Titratable, rapid on/off. Dex inadequate for RASS −4/−5 |
| Neuro checks needed frequently | Propofol | Fastest wake-up time for neurological assessment |
| Hemodynamically unstable / bradycardia risk | Propofol or fentanyl | Dex causes bradycardia and hypotension; avoid in shock |
| Alcohol withdrawal / seizure risk | Benzodiazepine (CIWA-guided) | Only appropriate indication for benzo ICU sedation |
| Post-extubation agitation / anxiety | Dexmedetomidine low-dose | Preserves airway reflexes; anxiolytic without respiratory depression |
| Feature | Assessment | Positive If |
|---|---|---|
| 1. Acute onset / fluctuating course | Change from baseline OR fluctuating mental status in past 24h? | Yes |
| 2. Inattention | Squeeze my hand when I say "A", S-A-V-E-A-H-A-A-R-T | < 8/10 correct (errors > 2) |
| 3. Altered consciousness | RASS score at time of assessment | RASS โ 0 (any score other than alert/calm) |
| 4. Disorganized thinking | "Will a stone float on water?" "Are there fish in the sea?" + command | < 4/5 correct |
| Drug | Dose | Evidence | 2025 vs 2018 |
|---|---|---|---|
| Haloperidol (Haldol) | 0.5–2 mg IV/IM q4-6h PRN | HOPE-ICU, 2013 and MIND-USA, 2018: no benefit on delirium duration, coma-free days, or mortality. AID-ICU, 2022 reinforced no mortality benefit. | CLARIFIED 2025: no recommendation for or against. 2018: weak conditional against routine use for prevention. Clinical bottom line unchanged, only for safety-threatening agitation, not as a delirium "treatment." |
| Quetiapine (Seroquel) | 25–200 mg PO BID | Limited evidence, may reduce delirium duration in small trials. Preferred when haldol contraindicated (Parkinson's, Lewy body dementia). | CLARIFIED 2025: no recommendation for or against, same as 2018. Use only when antipsychotic is genuinely needed. |
| Dexmedetomidine | 0.2–1.5 mcg/kg/hr IV | MENDS, 2007: more delirium-free and coma-free days vs lorazepam. SEDCOM, 2009: less delirium (54% vs 76.6%) and shorter time to extubation vs midazolam. DahLIA, 2016: faster resolution of agitated delirium. | ELEVATED 2025: suggested over propofol when light sedation or delirium reduction prioritized. 2018: dex and propofol considered equivalent. |
| Melatonin | 3–5 mg PO at bedtime | Low-risk intervention. May improve sleep quality and reduce delirium incidence. Note: not FDA-regulated, brand-to-brand variability in bioavailability. | NEW 2025 2018: no recommendation, off-label use only. Now formally suggested. |
| Level | Activity | When |
|---|---|---|
| 0 | Passive ROM in bed | Hemodynamically unstable, on NMBAs, high FiOโ |
| 1 | Active-assisted ROM in bed, bed exercises | On vasopressors but stable, mechanical ventilation |
| 2 | Sitting on edge of bed (dangle) | RASS ≥ −2, FiOโ ≤ 0.6, vasopressors weaning |
| 3 | Standing, transfer to chair | Cooperative, stable vitals, able to follow commands |
| 4 | Ambulation (with or without assist) | Off vasopressors, tolerating lower vent support or extubated |
| Drug | Dose | Onset | Key Points |
|---|---|---|---|
| Fentanyl (Sublimaze) | 25–200 mcg/hr IV infusion | 1–2 min | First-line analgesia. Less hypotension than morphine. Accumulates in renal/hepatic failure. |
| Dexmedetomidine (Precedex) | 0.2–1.5 mcg/kg/hr IV | 5–10 min | PADIS 2025 preferred for light sedation. Bradycardia risk. No respiratory depression. Anxiolytic. |
| Propofol (Diprivan) | 5–50 mcg/kg/min IV | 30–60 sec | Fastest wake-up. Monitor TGs q48h. Propofol infusion syndrome (PRIS) risk if > 80 mcg/kg/min > 48h. Best for neuro checks. |
| Midazolam (Versed) | 0.02–0.1 mg/kg/hr IV | 2–3 min | Avoid for routine sedation (โ delirium). Reserve for alcohol withdrawal/seizures. Accumulates. |
| Ketamine | 0.1–0.5 mg/kg/hr IV | 1–2 min | Opioid-sparing analgesic. Useful in refractory pain or bronchospasm. Emergence reactions. |
| Melatonin | 3–5 mg PO at bedtime | 30–60 min | PADIS 2025 new recommendation. Sleep promotion. May reduce delirium. Not FDA-regulated. |
| Haloperidol (Haldol) | 0.5–2 mg IV/IM q4-6h PRN | 10–20 min | No PADIS recommendation for delirium. QTc monitoring. Risk of EPS. Use only for acute safety threats. |
| Drug | Dose | Key Points |
|---|---|---|
| Glycopyrrolate (Robinul) | 0.1–0.2 mg IV q4–6h PRN, or 1–2 mg PO/NG TID | First-line in ICU. Does not cross the blood-brain barrier, so no delirium risk (unlike IV atropine or scopolamine). Workhorse for ketamine infusions, clozapine drooling, and post-extubation secretions. |
| Atropine 1% ophthalmic drops | 1–2 drops sublingual q4–6h | Off-label but cheap and effective. Minimal systemic absorption. Useful when the patient cannot swallow pills. |
| Scopolamine patch (Transderm Scōp) | 1.5 mg transdermal q72h | Convenient sustained effect but 6–12 h to onset. Avoid in elderly or already delirious patients (CNS anticholinergic effects). |
| Hyoscyamine (Levsin) | 0.125 mg SL q4h PRN | Alternative when glycopyrrolate is unavailable. |
| Ipratropium | 500 mcg nebulized q6h, or SL | Reserve for refractory cases or when other routes are limited. |
| Botulinum toxin (parotid / submandibular injection) | Ultrasound-guided by ENT or neurology | Chronic or refractory sialorrhea only (bulbar ALS, Parkinson, cerebral palsy). Effect lasts ~3 months. |