Provide demographic and biochemical data to compute the probability of C. difficile cure.
Guidelines & Evidence
Clinical Details
Section 1
When to Use
When to Use
Prognostic assessment of Clostridioides difficile infection (CDI) at the time of diagnosis
To predict the probability of treatment failure, 30-day mortality, and cure rates
To guide the choice of initial therapy (e.g., Fidaxomicin vs. Vancomycin) based on risk factors
Evaluated within the first 48 hours of positive toxin/PCR identification
Patient Population
Derived from adult clinical trial cohorts (including thousands of patients from the original Fidaxomicin trials). Validated for both inpatient and outpatient settings.
When Not to Rely on it Solely
Fulminant CDI (megacolon, shock) — these patients require immediate surgical/ICU evaluation regardless of ATLAS score
Recurrent CDI — the score was primarily derived for initial episodes, though it retains some utility in recurrence
Pregnancy — not validated in the obstetric population
Section 2
Formula & Logic
The ATLAS Acronym
01
A (Age): 1 point if > 60 years.
02
T (Temperature): 1 point if > 37.5°C.
03
L (Leukocytes): 1 point if WBC > 15,000 cells/µL.
04
A (Albumin): 1 point if < 3.5 g/dL (35 g/L).
05
S (Serum Creatinine): 1 point if > 1.2 mg/dL (106 µmol/L). Point is doubled (2 pts) if > 1.8 mg/dL.
Scoring Logic
Total score ranges from 0 to 6 (Note: Initial versions used different weighting, but the simplified version is common for clinical prediction). Note: Some validated models weight Creatinine and Age more heavily.
Probability of Cure
0–1 Points
95–100% cure
2 Points
85–90% cure
3 Points
75–80% cure
4 Points
60–65% cure
5–6 Points
35–45% cure
Section 3
Pearls/Pitfalls
Why ATLAS is Superior to IDSA Severity Alone
The IDSA severity criteria are binary (Severe/Non-severe). ATLAS provides a more granular probability spectrum. An ATLAS score of 4–6 identifies a very high-risk population where standard Vancomycin often fails, suggesting that these patients may benefit significantly from early Fidaxomicin or Bezlotoxumab.
Predicting Mortality
Mortality risk increases exponentially with scores > 3. A score of 6 is associated with a 30-day mortality rate exceeding 15% in multiple validation cohorts.
Clinical Pearls
Albumin is the most predictive single lab variable in the ATLAS model
The score serves as an excellent trigger for early infectious disease (ID) consultation
Fidaxomicin has shown superior cure rates over Vancomycin specifically in the high-ATLAS score subgroups
Section 4
Next Steps
Treatment Strategies
01
Low ATLAS (0-2): standard oral Fidaxomicin (or Vancomycin) is likely sufficient.
02
High ATLAS (3-6): strongly prefer Fidaxomicin; consider Bezlotoxumab if risk factors for recurrence are present; ensure early surgical surgical aware if score rises.
Complementary Tools
CDI Severity (IDSA/SHEA)
Zar Score (CDI Predictor)
Albumin-Leukocyte Ratio
Section 5
Evidence Appraisal
Original Index Development
Derivation and validation of a simple clinical prediction rule (ATLAS) for Clostridium difficile infection outcomes.
Miller MA et al. • BMC Infectious Diseases. 2013;13:372. The foundational paper using fidaxomicin trial data.
Developed by Dr. Mark Miller and the OPT-80 research team (Fidaxomicin developers). They recognized that while new antibiotics were expensive, identifying patients at high risk for Vancomycin failure could make these therapies more cost-effective.
