17.2 PICU Pain, Sedation, Delirium and Iatrogenic Withdrawal
Key Takeaways
Use a pain tool matched to developmental and communication ability. Self-report is preferred when reliable. Use FLACC or COMFORT-B for children who cannot self-report.
Set and reassess an individualized sedation target during mechanical ventilation. Use a validated sedation tool and minimize unnecessary sedation and benzodiazepine exposure.
A CAPD score of 9 or higher prompts delirium assessment. Compare with baseline, treat contributing illness and medication effects, and support sleep, mobility and family involvement.
Use WAT-1 or the Sophia scale to assess possible opioid or benzodiazepine withdrawal. The dose history, symptoms and competing causes determine the clinical interpretation.
Adjust a symptomatic taper and use class-specific replacement when indicated. Verify enteral conversion and overlap with the local protocol and pediatric pharmacy.
Pain, Sedation, Delirium and Iatrogenic Withdrawal in the PICU
Quick Answer: Assess pain, sedation depth, delirium and withdrawal separately. Select a tool the child can use or one matched to their developmental and communication needs. Set a daily sedation target for mechanical ventilation. A positive delirium or withdrawal screen should lead to a focused assessment, including review of baseline behavior, acute illness and recent medication changes.
Measure Pain and Set the Sedation Target
Self-report is preferred when the child can report pain reliably. SCCM's 2022 pediatric PANDEM guideline suggests routine self-report for communicative critically ill children aged six years or older. Choose a scale the child understands. The Wong-Baker FACES tool is intended for self-report from age three when the child understands the task. For noncommunicative children, use a validated observational tool with caregiver input about baseline behavior. Heart rate and blood pressure alone cannot separate pain from fear, hypoxia or other stress.
| Tool | Bedside use |
|---|---|
| Faces, Legs, Activity, Cry, Consolability (FLACC) | Five observed behaviors, each scored 0–2, give a total of 0–10. The original validation studied postoperative children aged two months to seven years. A revised version incorporates individualized behaviors in children with cognitive impairment. Interpret changes in light of ventilation, paralysis, sedation and usual behavior. |
| COMFORT-Behavior (COMFORT-B) | Six behavioral items yield a score from 6 to 30. Respiratory response is scored in ventilated children and crying in spontaneously breathing children. It can assess distress and is an option for tracking sedation in ventilated children. A high score can reflect pain or agitation and requires bedside assessment. |
| Faces or numeric scale | Have the child select a face or number that describes their own pain. Use a numeric or visual analog scale when the child understands its anchors and can distinguish different intensities. The clinician should avoid scoring a facial self-report scale from the child's expression. |
| State Behavioral Scale (SBS) | A sedation assessment option for mechanically ventilated children. The team sets the depth target based on illness and treatment needs, then uses observed scores to titrate and reassess. |
PANDEM guideline · Original FLACC validation · Revised FLACC validation · COMFORT-B study · Wong-Baker instructions
Treat the cause of pain and reassess after each intervention. Use titrated IV opioids for moderate to severe pain. Choose an agent and dose based on opioid exposure, ventilation, perfusion and organ function, and monitor for respiratory depression and ileus. Combine with suitable adjuncts such as acetaminophen, selected NSAIDs or regional analgesia for postoperative pain. Check hepatic risk and total acetaminophen from all products. A U.S. IV acetaminophen label for pain in children aged 2–12 years lists 15 mg/kg every six hours or 12.5 mg/kg every four hours, with no more than 75 mg/kg/day across routes. Assess bleeding, volume status and renal risk before an NSAID. Pediatric ketorolac use may be off-label and needs a local protocol. Consider nonpharmacologic support, including caregiver comfort and nonnutritive sucking with sucrose for brief invasive procedures in neonates and young infants. PANDEM guideline · IV acetaminophen label · Ketorolac label
Revisit the individualized sedation target at least daily and when clinical needs change. PANDEM supports protocolized titration and does not suggest routinely adding daily sedation interruption. It conditionally supports alpha-2 agonists as a primary sedative class for mechanically ventilated children. Dexmedetomidine is also a consideration after pediatric cardiac surgery when early extubation is expected. Monitor hemodynamics, and minimize benzodiazepine exposure when feasible. Adequate analgesia and sedation must continue during neuromuscular blockade, when behavior cannot be scored reliably. Prolonged high-dose propofol increases the risk of propofol infusion syndrome, and U.S. labeling does not indicate propofol for pediatric ICU sedation. PANDEM describes a conditional, short off-label ICU strategy in selected children. PANDEM guideline · RESTORE trial · Diprivan label
Recognize and Evaluate Delirium
Delirium is an acute, fluctuating disturbance of attention and awareness with a change in cognition or behavior from baseline. Hyperactive delirium can involve agitation and device pulling. Hypoactive delirium can present as reduced interaction, staring or slow responses. Mixed delirium fluctuates between these patterns. The chart shows the proportions in one prospective PICU cohort. Hypoactive and mixed presentations predominated, and the rounded values do not add to 100%. The authors reported an association with longer stay and in-hospital mortality. A separate international study found that 25% of assessable children screened positive at a single point in time. Neither study supplies a universal rate for every PICU population. Prospective PICU cohort · International point-prevalence study
Use a validated tool routinely from admission through PICU discharge or transfer. Follow the unit schedule and reassess changes from baseline. Establish whether the child is arousable enough to assess. Consider excessive drug effect or coma in a child who cannot be aroused.
- Cornell Assessment of Pediatric Delirium (CAPD): Eight observational items across a shift give a score of 0–32. A score of 9 or higher is a positive screen. The original study reported 94.1% sensitivity and 79.2% specificity at that cutoff. Specificity was lower with developmental delay. Use developmental anchors through age two and caregiver input on the child's usual communication and responses.
- pCAM-ICU and psCAM-ICU: Assess acute change or fluctuation and inattention, plus altered level of consciousness or disorganized thinking. The original pCAM-ICU validation included children at least five years old developmentally and chronologically. The preschool adaptation was originally validated in children six months through five years.
- SOS-PD: Combines distinct withdrawal and delirium components. Overlapping symptoms still require clinical assessment.
CAPD validation · Developmental anchors · pCAM-ICU validation · psCAM-ICU validation · SOS-PD validation
Address Causes and Reduce Distress
Evaluate oxygenation and ventilation, infection, glucose and electrolytes, organ dysfunction, seizures, pain and the medication timeline. Review sedative accumulation, withdrawal and other deliriogenic exposure. In a child with head injury, reassess acute neurologic deterioration urgently. Establish day–night cues, cluster care when safe, mobilize when ready, provide hearing or vision aids, and involve familiar caregivers in soothing and reorientation. The ICU Liberation ABCDEF bundle connects pain assessment, awakening and breathing readiness, choice of sedation, delirium care, early mobility and family engagement. Apply the pediatric sedation and extubation protocols to decide when and how to lighten sedation. SCCM ICU Liberation
PANDEM discourages routine antipsychotics for prevention or shortening delirium. It allows consideration of haloperidol or an atypical antipsychotic for severe refractory manifestations, alongside treatment of causes and specialist review. Check a baseline ECG, review QT-prolonging drugs, correct electrolyte abnormalities and monitor QTc and electrolytes during treatment. U.S. haloperidol lactate labeling does not approve IV use and calls for ECG monitoring if it is used intravenously. A QTc over 500 ms is a high-risk marker in the label. A change or markedly prolonged baseline warrants review of the treatment and local action thresholds. Dexmedetomidine can help meet selected sedation goals but is not a routine treatment for a positive delirium screen. PANDEM guideline · Haloperidol lactate label
Recognize and Manage Iatrogenic Withdrawal
Withdrawal can follow a rapid dose reduction, interruption or cessation after repeated opioid, benzodiazepine or alpha-2 agonist exposure. Risk generally increases with duration and cumulative exposure. PANDEM suggests screening as early as three to five days with higher opioid or benzodiazepine doses. Follow the drug timeline and assess for tremor, agitation, sweating, vomiting, diarrhea or sleep disruption. Tachycardia and hypertension after reducing an alpha-2 agonist also raise concern. Exclude competing causes such as hypoxemia, fever, sepsis and untreated pain.
WAT-1 and the Sophia Observation Withdrawal Symptoms scale help assess opioid or benzodiazepine withdrawal. The WAT-1 uses the prior 12 hours, quiet observation, a standard stimulus and recovery. Eleven items give a total from 0 to 12. The original instructions call for assessment at least once per 12-hour shift. A score of 3 or more flags possible withdrawal in context. In the original validation that cutoff had sensitivity 0.87 and specificity 0.88 against a nurse-rated withdrawal intensity of at least 4 on a 0–10 scale. The tool cannot distinguish which drug class is responsible. For suspected alpha-2 agonist withdrawal, PANDEM advises using the clinical findings with an adjunct opioid or benzodiazepine tool while a specific validated screen is unavailable. WAT-1 original study · Multicenter WAT-1 study · PANDEM guideline
Pause or slow the taper and reassess when the pattern suggests clinically important withdrawal. Use class-specific replacement when indicated. An opioid treats opioid withdrawal, a benzodiazepine treats benzodiazepine withdrawal, and an alpha-2 agonist treats withdrawal from that class. Adjust the subsequent taper under the unit protocol with serial scores, respiratory and hemodynamic checks, and pediatric pharmacy input. Oral methadone and lorazepam are options for selected transitions. Check enteral absorption, drug interactions, overlap and the risk of accumulation. Methadone also warrants attention to QT risk. Comparative pediatric studies have used different starting doses and taper schedules.
Worked exposure: If fentanyl ran at 4 mcg/kg/hour for an uninterrupted day, the basal exposure is 96 mcg/kg/day, or 0.096 mg/kg/day. Add any boluses and account for rate changes when reconstructing actual exposure. This calculation describes the amount of fentanyl received. Use the local pediatric conversion protocol and pharmacy verification to select a different drug's starting dose. Document the taper and recurrence plan at transfer. PANDEM guideline · Methadone taper trial · Pediatric conversion cohort
A 14-month-old is extubated after Glenn palliation. During the next shift she alternates brief periods of usual interaction with staring, diminished eye contact and slow responses. Her parent confirms a change from baseline. Vital signs are stable and a developmentally anchored CAPD score is 14. Which initial interpretation and response fit?
Hyperactive delirium. Begin scheduled haloperidol and use restraints as first-line treatment.
Expected post-extubation fatigue. Stop delirium assessments until transfer.
Opioid withdrawal is confirmed. Give a fentanyl bolus without reviewing the medication course.
Possible hypoactive delirium. Evaluate medical and medication causes, support sleep and reorientation, and involve the family.
An 8-month-old received fentanyl at 4 mcg/kg/hour and midazolam at 0.2 mg/kg/hour for ten days during respiratory failure. During a taper the infant develops tremor, sweating, yawning and watery stools. WAT-1 is 6. What is the best next action?
Suspect iatrogenic withdrawal. Pause or slow the taper, review the drug timeline and competing causes, and use class-specific replacement if clinically indicated before resuming an individualized taper.
Assume sepsis and accelerate the benzodiazepine taper before evaluating other causes.
Diagnose neuropathic pain from the score and double fentanyl without reassessment.
Continue the same taper because the score is below the positive screen threshold.
A 15-year-old with traumatic brain injury has severe, dangerous hyperactive delirium after evaluation and treatment of reversible causes. Nonpharmacologic measures have been insufficient. The team is considering off-label IV haloperidol. Which safety assessment is relevant before and during therapy?
Check lipase at baseline and daily as the main measure of acute haloperidol toxicity.
Obtain a baseline ECG, review QT-prolonging drugs, correct electrolyte abnormalities and monitor ECG for QTc prolongation and arrhythmias if IV haloperidol is used.
Follow PaCO₂ as the primary toxicity marker and use 30 mmHg as the stopping threshold.
Use EEG alone to decide whether to continue and omit ECG if baseline vital signs are stable.
Sections you finish are checked off in the contents.