Severity of Illness Scoring

First Things First

• Combines clinical observations and statistical theory to create a single metric of the patient’s condition and associated outcome (e.g., mortality, length of hospital stay, etc.)

History and Physical (assess for the following)

• Scoring systems are calibrated based on a logistic regression analysis of variables from large observational studies.
  • Regression analysis results must be considered in light of the facilities and inclusion criteria of the original study and may not apply to all institutions.
• Epidemiological tools that cannot be used to determine an individual patient’s outcome
• Can be subdivided to particular condition and population studied, such as:
  • Trauma vs. non-trauma surgical vs. medical vs. cardiac populations, etc.
  • Can evaluate multiple aspects of the patient’s condition (e.g., APACHE II) or can be organ-specific (e.g., Glasgow Coma Scale [GCS])
  • Performed at the time of admission or at intervals throughout hospital stay

Diagnostic Tests

• Have to be individualized to the particular scoring system employed

General Management Principles

• Scoring systems typically incorporate elements of:
  • Anatomic injury (mainly in trauma-specific scoring systems)
  • Physiological derangement
  • Effects of comorbidities
• Allow for between-facility comparison in the quality and efficacy of intensive care
• Increasingly used to determine risk/benefit of applying particular therapies (e.g., drotrecogin alpha)

Specific Treatment

General severity of illness scoring methods for adult ICU admissions

Acute Physiology and Chronic Health Evaluation (APACHE II; 1985)

• Incorporates 2/3 scoring elements of physiologic derangement, and comorbid diseases
• Score is based on the initial data collection at the time of admission and includes:
  • 12 physiological measurements
  • Age
  • Previous health status
• Score ranges from 0-71 points
• Benefits
  • Drotrecogin alpha may be considered for the treatment of sepsis based on the presence of both:
    • Sepsis-induced multiple organ failure, septic shock, or sepsis-induced acute respiratory distress syndrome [ARDS]) and,
    • An APACHE II score >25
  • Allows for adjustment in scores based upon a multitude of comorbidity modifiers
• Limitations
  • Lacks anatomic injury component and therefore less useful in trauma patients
  • APACHE III (1992) is intended to rectify trauma limitation but the system is proprietary
  • More complex than other commonly available ICU admission scoring systems

Simplified Acute Physiology Score (SAPS II; 1993)

• Similar to APACHE II
  • 12 routine physiological measurements during the first 24 hours, age, and prior health status
  • Score ranges from 0-24
• Benefits
  • Most commonly used severity of illness scoring system
• Limitations
  • Also lacks an anatomic injury component
  • Uses only 3 chronic health conditions in scoring (metastatic or hematologic malignancy, AIDS)

Sequential Organ Failure Assessment (SOFA; 1996)

• Uses only 6 physiologic measurements and no chronic health or age-related modifiers
• Score ranges from 0-24
• Benefits
  • Ease of use
  • Uses common and readily available variables
  • Can be performed on a daily basis

• Limitations:
  • Also lacks an anatomic injury component
  • Fails to take into account preexisting comorbidity modifiers

MODS (Multiple Organ Dysfunction Score)

• Similar criteria and scoring system as SAPS II, with the same benefits/limitations

Severity of illness scoring methods for adult trauma admissions

Abbreviated Injury Scale (AIS)

• Measures anatomic injury only
  • Originally intended for documentation of motor vehicle accidents
  • Comprises a lexicon of six-digit coded injuries (>2,000 listed) based on:
    • Region
    • Type of anatomic structure
    • Specific structure
    • Level of injury
  • Injuries are assigned a severity value of 1 (minor) to 6 (fatal)
  • Uses maximum AIS score based on the highest AIS severity among a set of injuries
• Benefits
  • Most advanced and comprehensive trauma-specific coding system
• Limitations
  • Uses only the single worst injury for a single area to provide the maximum AIS
  • Ineffective in the multiply injured patient
  • Proprietary system that requires specially trained coding personnel
  • Assigned values are subjective
  • Does not include comorbidities or physiological variables

Injury Severity Score (ISS; 1974)

• Summation of AIS scores from the squared values of the three most severely injured body areas
• Scores range from 0-75
• Benefits: includes more injuries in its score derivation
• Limitations
  • Same as AIS scoring, with the exception of number of injuries scored
  • Recognizes only the 3 most severe injuries
  • Includes only 1 injury from each area

Trauma-Related Injury Scoring System (TRISS; 1987)

• Overcomes the lack of comorbid conditions and physiologic variables missing from AIS/ISS
  • Incorporates ISS, age, and 3 physiological variables
  • Final modifying equation is applied depending on whether the trauma was blunt or penetrating
• Benefits
  • Has become the most accepted means of estimating trauma-related survival probabilities
  • Overcomes the ISS limitations of lack of physiological and comorbid disease data
• Limitations
  • Same as AIS scoring
  • Only 3 physiological variables included

Revised Trauma Score (RTS)

• Measures physiologic derangement
• Combines coded values for systolic BP (SBP), respiratory rate (RR) & GCS
• Coded values are given for each GCS, SBP, RR
• GCS
  • 13-15 = 4
  • 9-12 = 3
  • 6-8 = 2
  • 4-5 = 1
  • 3 = 0
• SBP
  • >89mmHg=4
  • 76-89 mmHg = 3
  • 50-75 mmHg = 2
  • 1-49 mmHg = 1
  • 0 mmHg = 0
• RR
  • 10-29 bpm = 4
  • >29bpm=3
  • 6-9 bpm = 2
  • 1-5 bpm = 1
  • 0 bpm = 0
• Coded values can be added for score of 0-12 for use in pre-hospital triage.
  • RTS value < 11 suggests need for transfer to trauma center.
• Values are weighted for in-hospital use as follows:
  • RTS = 0.9368 (GCScoded) + 0.7326 (SBPcoded) + 0.2908 (RRcoded)
  • Range of values 0-7.84
  • RTS >5, >90% survival
  • RTS < 3, < 20 % survival
  • Used primarily as research tool for outcome assessment & quality assurance

Severity of illness scoring methods for pediatric ICU admissions

Pediatric Risk of Mortality (PRISM; 1988)

• Uses 12 physiologic measurements, papillary response and GCS assessment – similar to APACHE II
• Benefit: comprehensive scoring tool
• Limitation: does not include comorbid disease assessments

Severity of illness scoring methods for pediatric trauma admissions

Pediatric Trauma Score (PTS)

• Combines aspects of anatomic injury, physiologic derangement & physiologic reserve
• Components include pt size, airway, consciousness, SBP, fracture, cutaneous injury
• Used for pre-hospital triage
• Each category is scored -1, +1 or +2.
• Size
  • Child >20 kg = +2
  • Toddler 11-20 kg = +1
  • Infant < 10 kg = -1
• Airway
  • Normal = +2
  • Assisted w/ oxygen = +1
  • Intubated = -1
• Consciousness
  • Awake = +2
  • Obtunded/lost consciousness = +1
  • Unresponsive = -1
• SBP
  • >90mmHg=+2
  • 51-90 mmHg = +1
  • < 50 mmHg = -1
• Fracture
  • No fracture = +2
  • Single closed fracture = +1
  • Multiple or open fracture = -1
• Cutaneous
  • No visible injury = +2
  • Contusion/abrasion or laceration not through fascia = +1
  • Tissue loss, gunshot or stab through fascia = -1
• Total score ranges from -6 to +12
• PTS >8, 0% mortality; PTS < 0, 100% mortality
• PTS < 9 indicates need for transfer to a pediatric trauma center.

Ongoing Assessment

Commonly used organ specific scoring methods

• GCS (1974)
  • Easily applied and reproducible assessment of level of consciousness
  • Based on combined scores for visual, motor, and verbal interaction
  • Eye Opening
    • Spontaneous = 4
    • To Voice = 3
    • To Pain = 2
    • None = 1
  • Verbal Response
    • Oriented = 5
    • Confused = 4
    • Inappropriate words = 3
    • Incomprehensible sounds = 2
    • None = 1
  • Motor Response
    • Obeys commands = 6
    • Localizes pain = 5
    • Withdraws from pain = 4
    • Flexor posturing to pain = 3
    • Extensor posturing to pain = 2
    • None = 1
  • Benefits:
    • GCS < 9 suggests inability to protect airway
    • Can be performed throughout the patient’s hospital course
• Killip classification (1967) of heart failure after acute myocardial infarction
  • Stratified according to physiological variables and thoracic auscultation
    • Killip class I - no clinical signs of heart failure; 6% mortality
    • Killip class II - rales or crackles in the lungs, S3 gallop, and elevated jugular venous pressure; 17% mortality
    • Killip class III describes individuals with frank acute pulmonary edema; 38% mortality
    • Killip class IV - cardiogenic shock evidence of peripheral vasoconstriction; 81% mortality
  • Predictor of all-cause mortality for the following:
    • ST and non-ST segment elevation myocardial infarction
    • Unstable angina
    • Those undergoing percutaneous coronary intervention
• Risk of acute renal failure, Injury to the kidney, Failure of kidney function,Loss of kidney function and End-stage kidney disease (RIFLE; 2007)
  • Uses the duration and extent of decline in urine output and creatinine levels or glomerular filtration rate to define the extent of acute kidney injury
  • Stratified injury among:
    • 3 grades of increasing severity of kidney injury: Risk, Injury, and Failure and
    • 2 outcome classes: Loss and End-stage
    • Particular stratification predicts mortality across multiple patient populations
    • Easily applied and reproducible scoring system

Complications

• Avoid extrapolation of mortality prediction model results to individual patients.
• Scoring systems are largely used in the research and quality assurance settings.
• SOFA score and GCS are easily applied and reproducible and can be repeated daily.

Author

• Written by James B. Sampson, MD, and Jeffrey D. Kerby, MD, PhD
  
• Revised by Patrick F. Allan, MD