Diagnosis and Classification of Delirium in the ICU
Learning Objective
Apply diagnostic and classification criteria to assess delirium in critically ill patients and guide initial management.
1. Introduction and Clinical Imperative
Delirium is an acute, fluctuating disturbance of attention and cognition that represents a form of organ dysfunction of the brain. It affects up to 80% of mechanically ventilated patients in the intensive care unit (ICU) and is independently associated with increased mortality, prolonged mechanical ventilation, extended hospital stays, and significant long-term cognitive impairment resembling dementia.
Despite its prevalence and severe consequences, delirium is frequently missed. Unstructured bedside assessments fail to detect up to 60% of cases, particularly the non-agitated, hypoactive subtype. The pathogenesis is complex, involving neurotransmitter imbalances (e.g., acetylcholine, dopamine), systemic inflammation crossing the blood-brain barrier, and profound disruption of the normal sleep-wake cycle. Therefore, integrating validated screening tools and a structured diagnostic work-up into daily ICU workflow is not merely best practice—it is a critical patient safety imperative.
The “Quiet” Danger
Hypoactive delirium, characterized by lethargy and inattention, accounts for approximately 50% of all delirium cases in the ICU. Without formal screening, it is often misclassified as appropriate sedation, depression, or fatigue, leading to missed opportunities for intervention and worse patient outcomes.
2. Confusion Assessment Method for the ICU (CAM-ICU)
The CAM-ICU is a rapid, validated, and widely used bedside tool for delirium screening in critically ill adults. It has demonstrated excellent performance, with a sensitivity of 93–100% and a specificity of 89–100% in diverse ICU populations. The assessment is designed to be performed by nurses or other trained clinicians in under two minutes.
2.1 Diagnostic Criteria and Administration
The CAM-ICU evaluates four key features. A patient is considered CAM-ICU positive (i.e., delirious) if they exhibit Feature 1 AND Feature 2, plus EITHER Feature 3 OR Feature 4.
| Feature | Assessment Criteria |
|---|---|
| 1. Acute Onset or Fluctuating Course | Is there evidence of an acute change in mental status from baseline, or has the patient’s mental status fluctuated during the past 24 hours? (Source: family, chart, nursing report) |
| 2. Inattention | Does the patient have difficulty focusing attention? (Assessed via standardized letter or picture recognition tasks. More than 2 errors indicates a positive feature.) |
| 3. Altered Level of Consciousness | Is the patient’s current level of consciousness anything other than “alert and calm”? (Assessed using the Richmond Agitation-Sedation Scale [RASS]; any score other than 0 is positive for this feature). |
| 4. Disorganized Thinking | Is the patient’s thinking disorganized? (Assessed by asking simple yes/no questions and testing for a logical command. More than 1 error indicates a positive feature.) |
2.2 Common Pitfalls and Training
Accurate use of the CAM-ICU requires structured training and consistent application. Common pitfalls include attempting to assess deeply sedated or comatose patients (RASS –4 or –5, who are “unable to be assessed”), language barriers interfering with the inattention task, and inconsistent delivery of the standardized questions. To ensure high fidelity, institutions should implement formal training with didactics, video examples, and supervised practice, followed by annual competency checks. High inter-rater reliability (κ ≥ 0.8) is achievable with this approach.
Embedding CAM-ICU in the Workflow
The most effective way to ensure consistent delirium screening is to integrate it directly into the electronic health record (EHR). Building mandatory CAM-ICU assessment prompts into nursing flowsheets, linked to RASS documentation, dramatically increases compliance and improves the rate of delirium detection across an institution.
3. Intensive Care Delirium Screening Checklist (ICDSC)
The ICDSC is an alternative validated tool that uses an 8-item checklist based on observations over an entire nursing shift. It provides an ordinal score from 0 to 8, with a score of 4 or greater indicating the presence of delirium. While the CAM-ICU provides a binary (yes/no) result, the ICDSC can also be used to track the severity of delirium over time.
3.1 Scoring Items and Thresholds
Each of the following eight items is scored as 1 point if present during the shift:
- Altered level of consciousness
- Inattention
- Disorientation
- Hallucinations, delusions, or psychosis
- Psychomotor agitation or retardation
- Inappropriate speech or mood
- Sleep-wake cycle disturbance
- Symptom fluctuation
A total score of 4 or more is diagnostic for delirium. A score of 1-3 suggests subsyndromal delirium.
3.2 Validation and Integration
The ICDSC has a reported sensitivity of approximately 74% and a specificity of 82% in mixed ICU populations. While its diagnostic accuracy is slightly lower than the CAM-ICU, its strength lies in its ability to quantify delirium severity. Worsening scores can trigger specific interventions, and improving scores can demonstrate response to therapy. The ICDSC should be completed at least every 12 hours and can be easily integrated into routine nursing assessments, with EHR calculators and alerts to flag high-scoring patients.
4. Additional Tools and Emerging Scales
While the CAM-ICU and ICDSC are the most robustly validated tools for the ICU setting, several other instruments exist. These are primarily used in non-ICU settings or are still under investigation for widespread ICU use.
- 3D-CAM: A 3-minute, structured adaptation of the CAM algorithm, well-validated for emergency department and hospital ward settings but less studied in ventilated patients.
- 4AT: A rapid (<2 minute) tool that combines alertness, orientation, attention, and acute change assessment. It is highly effective in older, non-ICU patients, but its validity in critically ill populations remains to be established.
- Cornell Scale for Depression in Dementia: This is a specialized tool for assessing mood in patients with pre-existing cognitive deficits and is not designed for acute delirium diagnosis.
Editor’s Note: Detailed comparative accuracy data for tools like the 3D-CAM and 4AT specifically within ventilated ICU subpopulations are limited. A comprehensive evaluation would require dedicated studies assessing their performance metrics against the CAM-ICU across different age groups, sedation levels, and primary diagnoses, along with data on training requirements and inter-rater reliability in the critical care environment.
5. Laboratory and Imaging Assessment
Once delirium is identified, a targeted work-up is essential to uncover and reverse contributing factors. This should be a systematic but judicious process, avoiding a “shotgun” approach.
5.1 Foundational Metabolic and Infectious Work-Up
The initial investigation should focus on common, reversible causes:
- Metabolic Panel: Check for electrolyte abnormalities (sodium, calcium, magnesium), renal failure (BUN, creatinine), and glucose dysregulation.
- Liver Function: Assess for hepatic encephalopathy with LFTs and an ammonia level if suspected.
- Infection Markers: A CBC with differential, CRP, or procalcitonin can signal infection. Obtain blood, urine, and other relevant cultures as clinically indicated.
- Oxygenation: An arterial blood gas can reveal hypoxemia or hypercarbia.
- Medication Review: Conduct a thorough review of all medications, especially sedatives (benzodiazepines), analgesics (opioids), and anticholinergic agents, which are common culprits. A toxicology screen may be warranted.
5.2 Escalated Neurologic Evaluation
Advanced neuroimaging and EEG are not routine for all delirium cases. They should be reserved for patients with specific “red flag” findings:
- CT or MRI of the Head: Indicated for new focal neurologic deficits (e.g., unilateral weakness), recent head trauma, or delirium that persists for days despite correction of all metabolic and infectious causes.
- Electroencephalogram (EEG): Indicated for suspected nonconvulsive seizures, particularly in patients with profound and unexplained altered mental status. In non-seizure encephalopathy, EEG typically shows generalized slowing or triphasic waves.
Image Wisely
Avoid routine head CT scans for every patient with delirium. The diagnostic yield is very low in the absence of focal neurologic signs or a history of trauma. A targeted approach based on red flags optimizes resource utilization and minimizes unnecessary radiation exposure and patient transport.
6. Subtype Differentiation and Clinical Implications
Delirium is not a monolithic entity. It presents with distinct motoric subtypes—hyperactive, hypoactive, and mixed—each with different clinical features, prognoses, and management priorities.
Hyperactive
Characterized by agitation, restlessness, and emotional lability. Patients may attempt to remove lines and tubes, posing a safety risk. This is the most recognized but least common subtype.
Hypoactive
Characterized by lethargy, apathy, and reduced motor activity. Patients are inattentive and withdrawn. This is the most common subtype and is associated with the highest mortality.
Mixed
Features fluctuations between hyperactive and hypoactive states throughout the day. This subtype is also very common and requires dynamic assessment and management.
Management Implications
Recognizing the subtype guides initial management. Hyperactive delirium requires immediate focus on patient safety, including environmental modifications and, if necessary, judicious use of pharmacotherapy per sedation guidelines. In contrast, hypoactive delirium management centers on reviewing and reducing sedative medications, promoting early mobility and physical therapy, optimizing the sleep-wake cycle, and engaging family for reorientation.
Unmasking Hypoactive Delirium
Daily sedation interruption trials (spontaneous awakening trials) are a powerful diagnostic and therapeutic tool. By temporarily stopping sedatives, clinicians can “unmask” an underlying hypoactive delirium that was being obscured by medication, allowing for a true assessment of neurologic function and preventing sedative accumulation.
7. Diagnostic Pathway Algorithm
A standardized pathway ensures consistent screening, evaluation, and management. It facilitates clear communication and timely, multidisciplinary interventions.
References
- Ely EW, Inouye SK, Bernard GR, et al. Delirium in mechanically ventilated patients: validity and reliability of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). JAMA. 2001;286(21):2703–2710.
- Devlin JW, Skrobik Y, Gélinas C, et al. Clinical practice guidelines for pain, agitation/sedation, delirium, immobility, and sleep disruption in adult ICU patients. Crit Care Med. 2018;46(9):e825–e873.
- Ankravs MJ, Fitzgerald JC, Ely EW, Pandharipande PP. Precision-based approaches to delirium in critical illness: a narrative review. Pharmacotherapy. 2023;43(11):1139–1153.
