What is the purpose and interpretation of the Apache (Acute Physiology and Chronic Health Evaluation) score in intensive care units?

APACHE Score: Purpose and Interpretation in the ICU

Primary Purpose

The APACHE II score is a severity of disease classification system designed to predict mortality risk and stratify disease severity in critically ill ICU patients, incorporating 12 physiologic variables, age, and chronic health status to generate a score ranging from 0 to 71. 1

Core Components

The APACHE II score consists of three distinct parts 1, 2:

• 12 acute physiologic variables measured within the first 24 hours of ICU admission, including vital signs, laboratory values, and oxygenation parameters 1
• Age points reflecting the impact of premorbid factors on outcomes 1
• Chronic health evaluation accounting for pre-existing comorbidities that influence prognosis 1

The total score ranges from 0 to 71, with higher scores indicating greater disease severity and increased mortality risk 1.

Clinical Interpretation and Risk Stratification

For risk stratification in complicated intra-abdominal infections, APACHE II is the preferred severity of illness score when calculated within 24 hours of hospitalization or ICU admission. 3

Mortality Prediction Thresholds

• APACHE II score ≥8 serves as an optimal cut-off with 83.3% sensitivity and 91% specificity for mortality prediction, triggering enhanced monitoring protocols 4
• APACHE II score of 15 provides the best diagnostic accuracy for predicting mortality with 85.3% sensitivity and 77.4% specificity 5
• APACHE II score ≥15-17 should prompt consideration of continuous or extended infusion beta-lactam antibiotics for severe infections, which improves clinical cure rates and reduces mortality 4
• APACHE II score ≥20 indicates patients who benefit most from continuous beta-lactam administration, with reduced mortality (RR 0.73) compared to intermittent dosing 4
• APACHE II score ≥29.5 correlates with significantly lower mortality when using prolonged piperacillin/tazobactam infusions (12.9% vs. 40.5%) 4

Predictive Performance

APACHE II demonstrates superior discriminative power for mortality prediction with a pooled AUC of 0.81, outperforming other scoring systems including SOFA (AUC 0.75). 4, 6

The score shows strong correlation with hospital mortality across diverse ICU populations 1. In validation studies, the area under the ROC curve ranges from 0.847 to 0.899, confirming excellent prognostic value 7.

Advantages Over Alternative Scoring Systems

APACHE II offers distinct advantages compared to other ICU scoring tools 4, 6:

• Incorporates age and chronic comorbidities, which SOFA does not include in its assessment 4, 6
• Higher accuracy for predicting severe acute conditions compared to disease-specific scores like Ranson, BISAP, and CTSI 4
• Comprehensive assessment reflecting both premorbid factors and acute physiologic derangements 1

Clinical Applications

Serial Monitoring

Regular recalculation of APACHE II score provides critical information about disease progression and treatment response. 4

• Daily scoring tracks disease trajectory, with pattern changes indicating response to treatment or onset of complications like sepsis 4
• Particularly valuable in severe acute pancreatitis management for monitoring evolving severity 4

Research and Quality Assessment

APACHE II serves multiple purposes beyond individual patient care 1, 2:

• Risk stratification to account for case mix in clinical studies 2
• Quality comparison among different ICUs 2
• Resource utilization evaluation and efficacy assessment over time 1

Delirium Risk Prediction

APACHE II score is incorporated into validated delirium prediction models 3:

• The PRE-DELIRIC model uses APACHE II as one of 10 predictors with an AUROC of 0.77 3
• Higher APACHE scores correlate with increased delirium risk in ICU patients 3

Important Limitations and Caveats

Calculation Complexity

• The APACHE II calculation is cumbersome and not all required parameters are routinely collected in every ICU 4
• This complexity may limit real-time clinical application at the bedside 4

Regional Calibration Requirements

The original US APACHE II model shows variable accuracy when applied to different populations, requiring local recalibration for optimal performance. 4

• Different pre-ICU care patterns may necessitate regional adjustments to the model 4
• Pre-ICU care quality can significantly impact physiological measurements, creating potential "lead time bias" 4
• UK recalibration was necessary when applying the US model to British patients 4

Clinical Context Dependency

• APACHE II should augment, not replace, clinical judgment 6
• The score predicts group outcomes but individual patient trajectories may vary 2
• Observed mortality rates may differ from predicted rates depending on local ICU quality and resources 5

Practical Implementation Algorithm

For optimal clinical utility, follow this structured approach:

1. Calculate APACHE II within 24 hours of ICU admission using the worst physiologic values during that period 1

2. Interpret the score using validated thresholds:

   • Score <8: Lower risk, standard ICU monitoring 4
   • Score 8-14: Moderate risk, enhanced monitoring 4
   • Score 15-20: High risk, consider treatment intensification 4, 5
   • Score >20: Very high risk, optimize antibiotic delivery and consider goals of care discussion 4

3. Recalculate serially (daily or every 48 hours) to track disease trajectory and treatment response 4

4. Combine with disease-specific information for most accurate prognostication, as APACHE II alone provides severity assessment but disease category adds critical context 1

5. Use for treatment decisions such as antibiotic dosing strategies in patients with scores ≥15-17 4